Nicotinamide phosphoribosyltransferase inhibitors, design, preparation, and structure-activity relationship

Article abstract of DOI:10.1021/jm4009949

Existing pharmacological inhibitors for nicotinamide phosphoribosyltransferase (NAMPT) are promising therapeutics for treating cancer. By using medicinal and computational chemistry methods, the structure-activity relationship for novel classes of NAMPT inhibitors is described, and the compounds are optimized. Compounds are designed inspired by the NAMPT inhibitor APO866 and cyanoguanidine inhibitor scaffolds. In comparison with recently published derivatives, the new analogues exhibit an equally potent antiproliferative activity in vitro and comparable activity in vivo. The best performing compounds from these series showed subnanomolar antiproliferative activity toward a series of cancer cell lines (compound 15: IC₅₀ 0.025 and 0.33 nM, in A2780 (ovarian carcinoma) and MCF-7 (breast), respectively) and potent antitumor in vivo activity in well-tolerated doses in a xenograft model. In an A2780 xenograft mouse model with large tumors (500 mm³), compound 15 reduced the tumor volume to one-fifth of the starting volume at a dose of 3 mg/kg administered ip, bid, days 1-9. Thus, compounds found in this study compared favorably with compounds already in the clinic and warrant further investigation as promising lead molecules for the inhibition of NAMPT.

Full text of DOI:10.1021/jm4009949

Article
pubs.acs.org/jmc
Nicotinamide Phosphoribosyltransferase Inhibitors, Design,
Preparation, and Structure−Activity Relationship
Mette K. Christensen,†,∇ Kamille D. Erichsen,†,▲ Uffe H. Olesen,†,§ Jette Tjørnelund,† Peter Fristrup,⊥
Annemette Thougaard,†,○ Søren Jensby Nielsen,†,◆ Maxwell Sehested,†,§ Peter B. Jensen,†,¶ Einars Loza,#
Ivars Kalvinsh,# Antje Garten,∥ Wieland Kiess,∥ and Fredrik Bjorkling*,†,‡
̈
†Topotarget A/S, Symbion, Fruebjergvej 3, Copenhagen DK-2100, Denmark
‡Department of Drug Design and Pharmacology, Faculty of Health and Medical Sciences, University of Copenhagen,
Universitetsparken 2, Copenhagen DK-2100, Denmark
§Experimental Pathology Unit, National University Hospital, Biocentre, Ole Maaloes Vej 5, Copenhagen DK-2200, Denmark
∥Center for Pediatric Research, Hospital for Children and Adolescents, University of Leipzig, Liebigstr. 21, Leipzig 04301, Germany
⊥Department of Chemistry, Technical University of Denmark, Kemitorvet 207, Lyngby DK-2800, Denmark
#Latvian Institute of Organic Synthesis, Aizkraukles 21, Riga LV-1006, Latvia
S
* Supporting Information
ABSTRACT: Existing pharmacological inhibitors for nicotinamide phosphor-
ibosyltransferase (NAMPT) are promising therapeutics for treating cancer. By
using medicinal and computational chemistry methods, the structure−activity
relationship for novel classes of NAMPT inhibitors is described, and the
compounds are optimized. Compounds are designed inspired by the NAMPT
inhibitor APO866 and cyanoguanidine inhibitor scaffolds. In comparison with
recently published derivatives, the new analogues exhibit an equally potent
antiproliferative activity in vitro and comparable activity in vivo. The best
performing compounds from these series showed subnanomolar antiproliferative
activity toward a series of cancer cell lines (compound 15: IC50 0.025 and 0.33 nM, in A2780 (ovarian carcinoma) and MCF-7
(breast), respectively) and potent antitumor in vivo activity in well-tolerated doses in a xenograft model. In an A2780 xenograft
mouse model with large tumors (500 mm3), compound 15 reduced the tumor volume to one-fifth of the starting volume at a
dose of 3 mg/kg administered ip, bid, days 1−9. Thus, compounds found in this study compared favorably with compounds
already in the clinic and warrant further investigation as promising lead molecules for the inhibition of NAMPT.
sized from nicotinamide, a sequence where nicotinamide
phosphoribosyltransferase (NAMPT) catalyzes the rate-limiting
step.12−14 Thus, inhibition of NAMPT enzyme activity causes a
direct inhibition of NAD production. Importantly, normal cells
can use an alternative pathway for NAD synthesis from
nicotinic acid (NA), catalyzed by nicotinic acid phosphor-
ibosyltransferase (NAPRT).15−17 Unlike most normal tissues,
many cancer cell lines, and primary tumors, are deficient in
NAPRT activity.18 Furthermore, an increased concentration of
NAMPT in colorectal, ovarian, and prostate cancer cells has
been reported.10,19,20 Altogether, inhibition of NAD production
via NAMPT inhibition is suggested to be a valid principle for
selective inhibition of cancer cell growth and as such represents
an attractive target for drug discovery.
INTRODUCTION
■
Inhibition of nicotinamide adenine dinucleotide (NAD)
production has recently been suggested as a principle for
inducing death of cells with high demand for this dinucleotide.1
This is particularly true for cancer cells due to increased
metabolism and high activity of NAD-consuming enzymes.
NAD is an essential cofactor in redox reactions and as such is
involved in cellular energy production and metabolism without
being substantially consumed. However, besides being a
cofactor, NAD serves as the substrate for mono-ADP-
ribosyltransferases,2 poly-ADP-ribose polymerases (PARPs),3
and sirtuins,4 all of these converting NAD to nicotinamide.
Also, NAD is consumed as the precursor for a number of Ca2+-
releasing second messengers (e.g., cADPR and NAADP).5,6
Thus, the increased dependence on glycolysis and elevated
expression or activity of PARPs7−9 and sirtuins,4 characteristic
for malignant cells, make these more sensitive to NAD
availability as compared with normal cells.10,11
A few classes of NAMPT inhibitors have been reported such
as compounds APO866 (1)21 and CHS828 (2),22 which have
entered the clinic, and more recent structures such as TRON-
823 and CB3086524 for which biological data have been
In the organism, several pathways for the synthesis of NAD
are known. Besides the de novo pathway using tryptophan as a
precursor, NAD can alternatively be synthesized or resynthe-
Received: July 2, 2013
Published: October 28, 2013
© 2013 American Chemical Society
9071
dx.doi.org/10.1021/jm4009949 | J. Med. Chem. 2013, 56, 9071−9088

Journal of Medicinal Chemistry
Article
Figure 1. NAMPT inhibitors. 1, APO866 (for activity, see Table 1); 2, CHS-828 (for activity, see Table 1); 3, EB 1627 (prodrug of CHS828);
TRON-8 (IC50 (SH-SY5Y) 3.8 nM);23 and CB30865 (IC50 (W1L2) 2.8 nM).37
Scheme 1. Convergent Synthesis of 2-Cyanoguanidine Derivatives 11−25
reported (Figure 1). Biological data for NAMPT inhibitor
MCP-8640 is also reported, but no chemical structure is
reported.25 These NAMPT inhibitors show potent antiprolifer-
ative activity in a spectrum of cancer cell lines and in vivo
efficacy in both solid tumors and leukemia in preclinical
studies.26,27 A comprehensive literature review on NAMPT
inhibitors including patents was recently published.28 Origi-
nally, 1 was developed as an inhibitor of NAMPT and displays
antiproliferative activities comparable to compound 2 for which
the molecular target was unknown until recently.29−31 Safety,
pharmacokinetics, and biological effect of compounds 1 and 2
(and a prodrug thereof, 3, Figure 1) have been reported from
five phase I clinical trials performed in patients with advanced
disease.32−34 In these trials, not surprisingly in phase I, no
objective response was found, and for both compounds, the
dose-limiting toxicity was found to be thrombocytopenia.
9072
dx.doi.org/10.1021/jm4009949 | J. Med. Chem. 2013, 56, 9071−9088

Journal of Medicinal Chemistry
Article
Scheme 2. Synthesis of 2-Cyanoguanidine Derivatives 4−10 via Carboxylic Acid Intermediates VIIa−e
Scheme 3. Convergent Synthesis of 1,2-Diaminocyclobutene-3,4-dione Derivatives 26−36
Compound 2 was tested in an oral formulation but was later
transformed to a prodrug, compound 3, to obtain improved
pharmacokinetics.35,36 To obtain good exposure, both com-
pounds 1 and 3 were administered iv using long-term
continued infusion.33,34 Both compounds have entered phase
II clinical trials; however, results from these trials are not yet
available.
9073
dx.doi.org/10.1021/jm4009949 | J. Med. Chem. 2013, 56, 9071−9088

Journal of Medicinal Chemistry
Article
Scheme 4. Synthesis of N,N′-Disubstituted Urea Derivatives 37, 38, and 40−50
Scheme 5. Synthesis of N,N′-Disubstituted Thiourea Derivatives 39 and 51
Inspired by these results and with the aim of finding
compounds with improved toxicological and activity properties,
we here report the structure−activity relationship (SAR) for a
new series of NAMPT inhibitors based on structural
modifications of compounds 1 and 2. The results from the
SAR study were rationalized, and new compounds were
designed using computational chemistry methods.
RESULTS AND DISCUSSION
■
Chemistry. A series of 2-cyanoguanidines 11−25 was
synthesized using a general strategy (Scheme 1). Treatment of
dimethyl cyanocarbonimidodithioate (I) with 4-pyridylamine
(IIa), 3-pyridylamine (IIb), or aniline (IIc) produced methyl
N′-cyano-N-arylcarbamimidothioates IIIa−c, which were con-
densed with preassembled hydroxamate amines IVa−c or
9074
dx.doi.org/10.1021/jm4009949 | J. Med. Chem. 2013, 56, 9071−9088

Journal of Medicinal Chemistry
Article
Scheme 6. Preparation of Hydroxylamines VIIIa,b,d,e
Scheme 7. Preparation of Hydroxylamines VIIIc,f,h and Amines VIIIg,i
sulfonamide amines Va−k to give the corresponding 2-
cyanoguanidines 11−25.
appropriate preassembled amines IVb and Va−f,k,n (Scheme
4).
Although this convergent approach was successful in most
cases, a stepwise formation of the side chain was used for some
2-cyanoguanidine derivatives such as 4−10 (Scheme 2).
Thus, the condensation of methyl N′-cyano-N-(pyridin-4-
yl)carbamimidothioate (IIIa) with 7-aminoheptanoic acid
(VIa) or optically active 2-methyl or 2-benzyl-7-amino-
heptanoic acids VIb−e38 produced 7-guanidinoheptanoic acid
derivatives VIIa−e. These intermediate acids VIIa−e afforded
the target N-hydroxycarboxamides 4−10 when treated with
hydroxylamines VIIIa−c in the presence of 1-ethyl-3-(3-
dimethylaminopropyl)carbodiimide (EDC) or O-(7-azabenzo-
triazol-1-yl)-N,N,N′,N′-tetramethyluronium hexafluorophos-
phate (HATU).
The N,N′-disubstituted thiourea derivative 39 was synthe-
sized by the condensation of 3-picolylamine IId with di(2-
pyridyl) thionocarbonate (DPT) followed by treatment of in
situ formed intermediate isothiocyanate XIIb with amine IVb.
The N,N′-disubstituted thiourea derivative 51 was prepared
from amine Vd and isothiocyanate XIIa, which in turn was
obtained from 4-pyridilamine IIa and carbon disulfide (Scheme
5).
The hydroxylamine intermediates VIIIa,b,d,e of carboxamide
amines IV and sulfonamide amines V were synthesized starting
from 2-hydroxyisoindoline-1,3-dione that was alkylated with
cyclohexyl alcohol XIIIa upon Mitsunobu conditions, with
cyclohexylmethyl bromide XIIIb in the presence of K2CO3 in
DMSO or with 2-(4-morpholinyl)ethyl chloride (XIIIc) in the
presence of 1,8-diazabicyclo[5.4.0]undec-7-ene (DBU) to give
the O-substituted derivatives XIVa−c. Removal of the
phthalimide protective group of compounds XIVa−c gave O-
hydroxylamine derivatives VIIIa, VIIId, and VIIIe, respectively.
Compound VIIIa was further converted into N-monobenzyl
derivative VIIIb via 2-nitrobenzosulfonamide intermediates XV
and XVI (Scheme 6).
A similar convergent approach to Scheme 1 was used for the
preparation of a series of 1,2-diaminocyclobutene-3,4-diones
26−36 (Scheme 3).39 Reaction of 3,4-diethoxy-3-cyclobutene-
1,2-dione (IX) with 4-pyridylamine (IIa) afforded intermediate
amidoester X, which was treated with amines IVa,b,d and Va−
e,j,l,m to obtain target 1,2-diaminocyclobutene-3,4-diones 26−
36.
N,N′-Disubstituted urea derivatives 37, 38, and 40−50 were
prepared by reacting of such carbonic acid derivatives as N,N′-
carbonyldiimidazole (CDI) or 4-nitrophenyl chloroformate
with 4-pyridylamine (IIa) or 3-picolylamine (IId) followed by
Another series of N,O-disubstituted hydroxylamines VIIIc,f−
i were prepared from cyclic ketones XVIIa−c and O-(2-
morpholinoethyl)hydroxylamine (VIIIe) or 3-morpholinopro-
9075
dx.doi.org/10.1021/jm4009949 | J. Med. Chem. 2013, 56, 9071−9088

Journal of Medicinal Chemistry
Article
pan-1-amine (XVIII) by reductive amination (Scheme 7).
Condensation of the ketones XVIIa−c with the amino
compounds VIIIe or XVIII gave the corresponding inter-
mediate oximes or imines XIXa−e, which were reduced to
saturated structures VIIIc,f−i.
By a similar one-pot reductive amination protocol, O-(2-
morpholinoethyl)hydroxylamine (VIIIe) and benzaldehyde
afforded N-benzyl-O-(2-morpholinoethyl)hydroxylamine
(VIIIj) (Scheme 8).
are the pyridyl head group, a functional group with hydrogen-
binding capabilities, a linker chain, and an aromatic group in the
end of the linker. With the aim of discovering new compounds
with improved activity and toxicological properties, the SAR of
these compounds was explored by modification of the
mentioned key structural elements (Figure 2).
The primary determination of activity was performed using a
WST-1 cell viability and proliferation assay in two cell lines, a
breast cancer, MCF-7, and an ovarian carcinoma, A2780. The
WST-1 assay determines the metabolic activity of the cells, in a
process dependent on NADH as coenzyme; thus, this assay will
have a strong functional connection to the NAMPT inhibition.
Modification of Aromatic End Group. Initially, the
aromatic end group of compound 2 was modified by preparing
novel hydroxamic acid esters that immediately gave potent
analogues in the in vitro test system, represented by the first hit
compound 4 with nanomolar activity (Table 1 and Figure 3).
In the crystal structure of NAMPT cocrystallized with
compound 1, a rather large binding region near the surface of
the protein can be observed where the side chain end group of
the ligand is placed.41 As in earlier work, the published crystal
structure of NAMPT cocrystallized with 1 was used as starting
point for docking analysis42 (PDB ID 2GVJ), which was carried
out in Glide (further details in the Experimental Section).
Previously, we found that different ligands for NAMPT had
such similar geometric features that compounds 2 and 5 could
even be docked into the crystal structure of compound 1
without removing the crystallographic water molecules in the
active site.42 However, due to the much larger and more diverse
set of ligands, we chose to completely remove the crystallo-
graphic waters in the current study. In the Supporting
Information, we have included additional pictures showing
the crystal structure with the crystallographic water molecules
present, and it is evident that only a few water molecules
penetrate the active site when the ligand is present.
Scheme 8. Synthesis of N-Benzyl-O-(2-
morpholinoethyl)hydroxylamine (VIIIj)
The hydroxamate amines IVa−d were prepared from N-Boc
protected ω-amino hexanoic, heptanoic, and octanoic acids
XXa−c that were condensed with hydroxylamine VIIIj in the
presence of EDC or with hydroxylamine VIIIc in the presence
of HATU to afford the corresponding hydroxamates XXIa−d
that were deprotected to obtain compounds IVa−d (Scheme
9).
The synthesis of sulfonamide amines Va−n was based on the
condensation of phthalyl-protected ω-aminopentane, amino-
hexane, and aminoheptane sulfonyl chlorides XXVa−c with
appropriate amine or hydroxylamine derivatives to produce the
corresponding sulfonamides XXVIa−j,l−n (Scheme 10). In the
case of tertiary sulfonamides XXVIh,i,l,m, the synthesis
included an extra alkylation step of initially obtained secondary
sulfonamides XXVIIa−c. The following treatment of the
sulfonamides XXVIa−j,l−n and XXVIIc with hydrazine
produced the sulfonamide amines Va−n. The intermediate
sulfonyl chlorides XXVa,b were obtained by a short synthetic
sequence from potassium phthalimide XXII, including
monoalkylation of the latter with 1,5-dibromopentane or 1,6-
dibromohexane, treatment of the obtained bromo derivatives
XXIIIa,b with sodium sulfite, and conversion of the resulting
sodium sulfonates XXIVa,b into the corresponding sulfonyl
chlorides XXVa,b with PCl5. The sulfonyl chloride XXVc was
obtained using the literature procedure.40
When 1 was redocked into the empty active, it achieved the
crucial hydrogen bonding to Ser275, albeit with a slightly
longer distance (2.3 Å) compared to the 1.8 Å observed in the
X-ray structure (Figure 4).
Docking of compound 4 suggested a similar binding mode as
compared to 1 (Figure 5) where the cyanoguanidine is capable
of achieving a similar hydrogen bond to Ser275. Furthermore,
the cyanoguanidine established two new hydrogen bonds to
Asp219, which may serve as part of the explanation for the
efficiency of this chemical motif.
SAR. Compounds 1 and 2 (Figure 1) represent two distinct
compound classes, which are bioisosters, both targeting the
NAMPT enzyme. The common features in the two compounds
Scheme 9. Synthesis of Hydroxamate Amines IVa−d
9076
dx.doi.org/10.1021/jm4009949 | J. Med. Chem. 2013, 56, 9071−9088

Journal of Medicinal Chemistry
Article
Scheme 10. Synthesis of Sulfonamide Amines Va−n
To investigate the preferred binding orientation of the side
chain in compound 4, the close α-methyl and α-benzyl
analogues were prepared in an enantiomerically pure form. A
stereochemical preference for the (S)-methyl derivative (6) was
found, with a 40 times higher activity, as compared with the
(R)-methyl derivative (7) (Table 2). However, this was not
reflected in the calculated docking scores for the interaction
that showed a reversed preference. The reasons for this
discrepancy were not clear from visual inspection of the docked
structures, since they were virtually superimposable in spite of
the difference in stereochemistry. Therefore, we speculate that
the observed difference in activity could be due to one of
several factors that are not accounted for in simple docking
calculations such as protein flexibility or different interactions
with the hydrogen-bond network between enzyme and water
molecules. The similarity in physicochemical properties leads us
to believe that the observed difference is not due to differences
in pharmacokinetics. The activity difference was very small for
the corresponding benzyl enantiomers (8 and 9), probably due
to the similar size of the binding end groups in the molecule
(Table 2). Thus, the docking analysis of these structures did
not reveal any obvious differences, or a preferred fit, of the
enantiomers with the protein structure. Even though the chiral
derivatives were very potent with subnanomolar activity, this
compound group was not further explored due to a generally
low stability and rapid hydrolysis in mouse plasma (data not
shown).
An improvement of both stability and activity was found for
the hydroxamic acid esters bearing an additional substituent on
the hydroxamic acid nitrogen, compounds 5, 10, 11 and 12 all
9077
dx.doi.org/10.1021/jm4009949 | J. Med. Chem. 2013, 56, 9071−9088

Journal of Medicinal Chemistry
Article
Figure 2. Overview of the sequence of the described SAR work: (1)
Modification of aromatic end group D to hydroxamic acid esters (4−
12); (2) SAR for pyridyl head group A (10 and 13); (3) SAR of
hydrogen-binding group B, replacement of cyanoguanidine with
squaric acid and urea (26, 28, 29, 38, and 39); (4) Change of the
hydroxamic acid ester for a preferred alkoxy sulphonamide or
sulphonamide in D (e.g., 10 vs 15 and 26 vs 31); (5) Optimization
of linker length C and end group for squaric acids (31−34 and 36)
and urea derivatives (40−42 and 44); (6) Final optimization of the
cyanoguanidine series with a pyridyl head group and an alkoxy
sulphoneamide (17 and 20−25).
exhibiting subnanomolar activities (Table 1). From the docked
structures, it is clear that these ligands are capable of spanning
the wide entrance of the cleft of the protein active site thereby
increasing their binding affinities. At the present time, it is
unclear whether this increase is due to additional (nonspecific)
binding interactions with the protein surface or that the
increased binding is caused by a decreased number of accessible
conformations of the free ligands.
SAR for Pyridyl Head Group. In earlier studies, the pyridyl
head group in the parent compounds has been found essential
for high activity.22 To investigate the importance of the head
group in this series, a phenyl derivative, compound 13, was
prepared and found approximately 1000−2000 times less
potent as compared with the analogous compound 10 (Table
Figure 3. Structures of molecular fragment B in Tables 1, 3, and 4.
1). This large difference in activity is difficult to explain since
the two head groups (phenyl and pyridine) are both capable of
sandwiching between Tyr18 and Phe193 (Figure 6).
To investigate this difference in more detail, we carried out
an analysis of the binding energy using density functional
theory (DFT) in combination with the B3LYP functional with
added dispersion corrections (DFT-d3).
Due to the computational demands of DFT-d3, the enzyme
was reduced to an active site model consisting of Arg196, B-
Table 1. NAMPT Inhibitors with a Cyanoguanidine Binding Group
cell line
a
a
A2780
IC50, (mean, nM)
MCF-7
IC50, (mean, nM)
b
compd
R1
n
A
B
structure
SD
1.5
SD
1, APO866
−
−
−
−
6
6
6
7
5
6
6
6
6
6
6
7
5
6
6
6
5
6
−
−
−
−
1
1.6
7.4
1.6
940
1
0.68
1.3
2, CHS828
0.56
0.19
13
4
4-pyridyl
4-pyridyl
4-pyridyl
4-pyridyl
4-pyridyl
phenyl
CO
CO
CO
CO
CO
CO
SO2
SO2
SO2
SO2
SO2
SO2
SO2
SO2
SO2
SO2
SO2
SO2
35
377
0.80
0.08
5
2
0.055
0.1
0.036
0.11
0.014
0.10
34
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
4
0.42
c
6
0.01
ND
0.021
1016
3.4
6
0.081
98
0.002
737
4
4-pyridyl
4-pyridyl
3-pyridyl
4-pyridyl
4-pyridyl
4-pyridyl
4-pyridyl
4-pyridyl
4-pyridyl
4-pyridyl
4-pyridyl
4-pyridyl
3
0.16
0.18
0.021
0.31
0.060
0.35
0.091
0.017
0.0032
0.031
0.11
0.03
0.07
2.3
4
0.025
0.51
0.33
2.9
0.63
1.10
0.19
0.24
0.42
0.015
4
8
0.052
0.16
0.1
9
0.11
0.29
0.036
4
0.089
0.011
0.0041
0.049
0.091
0.022
0.053
4
c
10
11
12
10
13
ND
0.007
0.1
0.008
0.01
0.01
0.29
0.041
0.63
a
b
c
Activities were determined in a WST-1 assay. See Figure 3. ND, not determined.
9078
dx.doi.org/10.1021/jm4009949 | J. Med. Chem. 2013, 56, 9071−9088

Journal of Medicinal Chemistry
Article
Figure 4. (left) Illustration of 1 docked in the NAMPT X-ray structure (PDB ID 2GVJ). Notice the lack of crystallographic waters in the binding
cleft (additional orientations are included in the Supporting Information). (center) Pose of 1 observed in the X-ray with hydrogens added and waters
removed. (right) Pose obtained by docking 1 into the “dry” active site of NAMPT.
Figure 5. (left) Conformation of 4 docked into the NAMPT active
site (PDB ID 2GVJ). (right) Close-up of the pyridine sandwich with
the three hydrogen bonds made by the cyanoguanidine moiety.
Figure 6. Docking of 10 and 13 in the NAMPT X-ray structure (PDB
ID 2GVJ), where it is clear that the head groups are virtually
superimposable.
Table 2. NAMPT Inhibitors with a Chiral End Group
DFT calculations, since 4-pyridyl and 3-pyridyl head groups
had a similar stabilization energy (within 1 kJ/mol). At present,
it is unclear whether more extended conformational sampling
combined with, for example, advanced mixed quantum
mechanics/molecular mechanics (QM/MM) calculations
could delineate this interesting experimental difference in
activity.
cell line
a
a
A2780
IC50, (nM)
MCF-7
IC50, (nM)
compd
R
SD
b
SD
6
7
8
9
(S)-Me
(R)-Me
(S)-Bn
(R)-Bn
0.08
3.9
ND
0.64
1
0.8
SAR of Hydrogen-Binding Group. The next structural
modification made was a replacement of the cyanoguanidine
with other hydrogen-binding groups. There have been previous
attempts to substitute the cyanoguanidine or amide group of
compounds 1 and 2 with other groups retaining activity (Figure
1).28 In this investigation, the most promising substitutions
were found to be the squaric acid and urea analogues. The
squaric acid compounds 26, 28, and 29 (Table 3) and the urea
and thiourea derivatives with a 3-picolyl head group,
compounds 38 and 39, respectively (Table 4), both series
with a hydroxamic acid ester end group, all were potent
inhibitors of proliferation. As expected, docking of these
structures showed that they too were capable of obtaining the
crucial hydrogen-bonding interactions in the active site of
NAMPT (Figure 7).
Squaric acids with similar side chains as the parent
compounds 1 and 2 have been reported in the patent literature
to have antiproliferative activity.44 This finding suggests a
further exploration of the squaric acid bioisosters with novel
side chains.
Change of the Hydroxamic Acid Ester. With three
structural core elements in hand, the cyanoguanidine, the
38.3
5.3
1.7
27.8
3.4
0.32
0.22
0.03
0.28
0.6
a
b
Activities were determined in a WST-1 assay. ND, not determined.
chain Tyr18, B-chain Asp16, Asp219, Phe193, Arg311, and
Ser275. The amino acids were truncated at the Cα in line with
earlier work.43 Since the aim of the study was primarily to
delineate the effect of the head group, it was also decided to
truncate the ligands in the spacer region so it consisted of just a
single methyl group. A single-point energy calculation shows
that the pyridine has an interaction energy that is 22 kJ/mol
larger than that for the corresponding structure with a phenyl as
the head group, which is in qualitative agreement with the
experimental results. In the cyanoguanidine series, the 4-pyridyl
head group was compared with the analogues 3-pyridyl
derivative, compound 15 versus 16 (Table 1). This comparison
showed a greater than 10-fold activity preference for the 4-
pyridyl group. A similar activity preference was found for other
4-pyridyl versus 3-pyridyl compound pairs (data not shown),
but in this case, the observation could not be explained by the
9079
dx.doi.org/10.1021/jm4009949 | J. Med. Chem. 2013, 56, 9071−9088

Journal of Medicinal Chemistry
Article
Table 3. NAMPT Inhibitors with a Squaric Acid Binding Group
cell line
a
a
A2780
IC50, (mean, nM)
MCF-7
b
compd
R1
n
A
B
structure
SD
IC50, mean, (nM)
SD
c
26
27
28
29
30
31
32
33
34
35
36
4-pyridyl
4-pyridyl
4-pyridyl
4-pyridyl
4-pyridyl
4-pyridyl
4-pyridyl
4-pyridyl
4-pyridyl
4-pyridyl
4-pyridyl
6
6
6
5
6
6
7
5
6
6
5
CO
SO2
CO
CO
SO2
SO2
SO2
SO2
SO2
SO2
4
0.03
4.1
0.02
2.2
ND
5
21.6
12.3
23
4.5
5.0
1.1
2.3
2.9
5.2
8.4
2.4
6.4
6.3
6
0.29
0.68
0.73
0.49
0.49
0.57
0.59
0.2
0.32
0.47
0.03
0.22
0.17
0.12
0.35
0.14
0.05
6
7
10.8
9
4
4
8
4
35.8
13.7
29.1
46.3
10
11
SO2
b
10
c
0.38
a
Activities were determined in a WST-1 assay. See Figure 3. ND, not determined.
Table 4. NAMPT Inhibitors with a Urea (Thiourea) Binding Group
cell line
a
a
A2780
IC50, (mean, nM)
MCF-7
IC50, (mean, nM)
b
compd
R1
n
A
B
structure
SD
SD
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
4-pyridyl
3-picolyl
6
6
6
6
6
5
7
5
7
6
6
6
5
5
6
SO2
CO
CO
SO2
SO2
SO2
SO2
SO2
SO2
SO2
SO2
SO2
SO2
SO2
4
0.25
0.27
0.91
0.56
0.17
3.4
0.13
0.13
0.31
0.01
0.06
2.2
0.05
0.37
5.5
1.4
0.93
28
0.01
0.51
2.2
4
c
3-picolyl
4-pyridyl
3-picolyl
4-pyridyl
4-pyridyl
3-picolyl
3-picolyl
3-picolyl
3-picolyl
3-picolyl
3-picolyl
3-picolyl
4-pyridyl
4
7
0.05
0.11
11.2
4.6
7
4
4
1.8
0.07
4.5
7.4
6.5
3.6
5.8
11
4
4.6
5.7
4
0.31
0.58
2.2
0.11
0.18
0.12
0.21
6.4
2.3
10
11
12
11
10
0.15
5.0
1.7
8.7
49
7.2
13
33.7
1.2
2.7
1.9
7
c
SO2
b
10
c
0.16
0.15
0.31
0.40
a
Activities were determined in a WST-1 assay. See Figure 3. The compound is a thiourea.
Figure 7. Compounds docked in the NAMPT X-ray structure (PDB ID 2GVJ). (left) 26, a representative of the squaric acid series with dotted lines
indicating the hydrogen bonds made to Ser275 and Asp219. (center) 38, the urea linker also allows hydrogen bonds to be made to Asp219. (right)
39, also the larger thiourea can be accommodated in the active site.
squaric acid, and the urea, all producing highly active
compounds, a further optimization was structured. Changing
the hydroxamic acid ester for an alkoxy sulphonamide or
sulphonamide gave a more robust series of compounds with
high activity, and this substitution was used in the further SAR
work (see e.g., 10 vs 15, Table 1 and 26 vs 31, Table 3).
9080
dx.doi.org/10.1021/jm4009949 | J. Med. Chem. 2013, 56, 9071−9088

Journal of Medicinal Chemistry
Article
Squaric Acid Series. In the squaric acid series, compounds
31, 34, and 35 were prepared to explore the effect of ring size in
the end group; however, no significant difference in binding
affinity/antiproliferative activity was found (Table 3). This is
understandable also from the docking results, since the head
groups dock in a similar position in the active site, as expected
(Figure 8, left), whereas the flexible end of the chain can freely
position itself in the wide entrance region of the catalytic cleft.
does not reveal any differences in the π−π stacking ability of
the two head groups, which is in line with the observation for
the urea series.
Similar to the squaric acid analogues, a linker of five to six
carbons gave highly active compounds both with a pyridyl head
group (37, 42, 43) and a picolyl head group (46, 47, 49, 50,
Table 4).
Also in the urea series, many of the analogues showed
subnanomolar activities either with a pyridyl or picolyl head
group, a urea or thio-urea binding group, and a series of
different end groupings.
Cyanoguanidine Series. Returning to the cyanoguanidine
series with a pyridyl head group and an alkoxy sulphone amide
connecting the linker and end groups, the most potent
compounds with low picomolar antiproliferative activity, for
example, compounds 17 and 20−25, were found (Table 1).
Similar to the other series, a linker of six carbons was optimal.
In summary, using a functional cell proliferation assay, a
rather broad SAR was determined. Highly active compounds
were found in all three series, the cyanoguanidine, the squaric
acid, and the urea series, suggesting an analogous interaction
with the target NAMPT enzyme. Also, a similar substitution
pattern was used to obtain the most potent compounds in the
three series suggesting a similar binding to the target.
Figure 8. Compounds docked in the NAMPT X-ray structure (PDB
ID 2GVJ). (left) Compounds 31, 34, and 35 all dock into the
sandwich region of the enzyme. (right) Wide entrance region allows
the ligands to obtain different positions that do not offer any
differences in activity between compounds having a five-membered
(green), six-membered (red), or four-membered rings (blue) as
substituent on nitrogen.
IN VITRO ACTIVITY AND MODE OF ACTION
■
To further characterize the key compounds, the IC50 values of a
number of them were established in the colon cancer cell line
HCT-116 and a derived compound 1 resistant cell line HCT-
116/APO866 (Table 5). Determination of the actual sensitivity
Likewise, the linker length was varied from five to seven
carbons in compounds 31−33 and 34 versus 36 that gave
compounds with similar activity, however with a slight activity
preference for compounds with six carbons in the linker (Table
3). Shorter and longer linker chains gave compounds with
lower antiproliferative activity (data not shown).
For compounds 31−33, the docking analysis shows a
preference for a hydrogen-bond interaction between the
sulphonamide moiety and His191. This enforces a constraint
on the length between the sulphonamide and the pyridine head
group, which is suitable for a six-carbon linker (31). The
docking results showed that the shorter chain introduces strain
in the structure (33) and the longer seven-carbon linker has to
coil up to satisfy the distance requirement (32). However, we
should stress that the similarity in the biological activity shows
that such accommodation is possible and indeed takes place
without significant loss in activity.
Table 5. IC50 Values for Key Compounds for
Antiproliferative Effects in HCT-116 and Compound 1
Resistant HCT-116/APO Cells
cell line
a
a
HCT-116
IC50 (nM)
HCT-116/APO866
IC50 (nM)
compd
SD
6.1
1
10.9
1,9
3,6
39
946
15
17
31
37
0,2
2,4
13
>5000
>5000
>5000
>5000
5,3
2,9
The squaric acid 27, a sulphone amide, was less active;
however, this compound was not strictly analogous to any
alkoxy sulphone amide, since it has an additional pyridine group
in the end group of the molecule. In all three series, the
sulphonamide end groups were found equipotent or slightly
less potent as compared with similar alkoxy sulphonamide (see
compounds 18, 23, 27, and 48, Tables 1−4).
Urea Series. The urea derivatives were designed as
analogues to compound 1 where the unsaturated amide was
replaced by a urea group. In this series, compounds with a 4-
pyridyl or a 3-picolyl head group were compared, compounds
40 versus 41 and 42 versus 44, and found equipotent despite
the structural difference in the head group (Table 4).
Computational modeling of these interactions showed that,
despite the difference in the location of the pyridine nitrogen,
the aromatic group positions itself perfectly aligned in the
pocket between Tyr18 and Phe193. This introduces a slight
difference in the orientation of the urea moiety but does not
hinder the formation of hydrogen bonds to Asp219. As
mentioned earlier, our simple DFT model of the binding site
a
Activities were determined in a WST-1 assay.
of the cell line with acquired resistance was not possible.
However, the resistance toward the compounds was at least
128−2632-fold higher in HCT-116/APO866 compared to the
parental cell line. The mechanism of resistance in HCT-116/
APO866 has been determined to be specifically due to a
mutation, H191R, in the active site of NAMPT that results in
highly specific resistance.29,42 Thus, the high level of cross-
resistance observed for the compounds in this study strongly
suggests that their mechanism of action likewise is through
inhibiting NAMPT and thus similar to that of compound 1.
In order to confirm the on-target mechanism of action for
the new series of compounds, the enzyme inhibitory activity
was determined using a NAMPT enzymatic assay with HepG2
lysates as the source of NAMPT enzyme (Table 6). Even
though the correlation with the antiproliferative activity data
was not perfect, all tested compounds showed high potency
with low nanomolar activity, which is a strong support for a
9081
dx.doi.org/10.1021/jm4009949 | J. Med. Chem. 2013, 56, 9071−9088

Journal of Medicinal Chemistry
Article
Treatment with compound 17 (15 mg/kg bid ip for 10
consecutive days (small tumors) or two 5 day cycles (large
tumors) had good therapeutic effect in both small and large
A2780 tumors (Figure 9). A clear decrease was seen in the
Table 6. NAMPT Enzyme Assay (IC50, nM)
compd
1
2
4
5
10
11
14
15
17
IC50
(nM)
2.2
18.3
1.1
38.5
0.2
4.4
2.4
0.3
3.2
compd
18
27
32
33
49.6
37
39
40
133.2
41
IC50 (nM)
2.4
22.8
3.6
49
8.2
11.8
direct enzyme interaction. The discrepancy between enzyme
inhibition and antiproliferative activity is suggested to be due to
the different properties of the compounds such as cell
penetration that is important in the cellular assay. Compounds
were selected for further testing primarily based on in vitro
activity but also to cover the chemical classes. Key compounds
were also characterized in a clonogenic assay in several cancer
cell lines (Table 7). Compounds 15 and 17 showed similar or
increased potency in these assays as compared to reference
compound 1 and were thus selected for further in vivo
characterization.
Figure 9. Tumor growth curves of compound 17 in an A2780
xenograft mouse model. 15 mg/kg bid ip on days 0−4 and 7−11 (large
tumors; starting volume 500 mm3) or days 0−9 (small tumors; starting
volume 100 mm3). Inserted graph: body weight change during
treatment.
Table 7. Clonogenic Assay for Compounds 15, 17, and 1 in a
Selection of Cancer Cell Lines
tumor volumes during the treatment period, and some mice
even showed a transient cure and eradication of the tumor.
After treatment, the tumors resumed growth at various time
points and grew to the maximum allowed size (1000 mm3).
Treatment with compound 17, 15 mg/kg, using the schedules
above was well tolerated and did not affect the body weight as
compared with vehicle-treated animals (Figure 9, insert).
Similarly, compound 15 was tested in the A2780 xenograft
model with large tumors (3 mg/kg, ip, bid, 10 consecutive
days) (Figure 10). Compound 15 showed good efficacy, and on
average, treatment reduced the tumor volume significantly to
one-fifth of the volume from start of treatment (Figure
10A).This dose was found well tolerated in the mice (Figure
10C). Compound 15 was tested at higher doses (5 and 10 mg/
kg) with very clear therapeutic effects; however, already at 5
mg/kg, four of nine mice showed toxic signs in the form of
body weight loss and reduced activity level as compared to the
control that was further accentuated at 10 mg/kg (data not
shown). Thus, it was concluded that a 3 mg/kg bid dose was at
the MTD level for compound 15, in this model.
compd
cell line
15, IC50 (nM)
17, IC50 (nM)
1, IC50 (nM)
A2780
A431
0.55 (24 h)
>50
0.43 (16 h)
5.7
6.1
a
ND
a
DU145
MCF-7
>50
6.39
ND
>50 (24 h)
1.9 (96 h)
0.3
>50 (24 h)
0.98 (96 h)
0.05
8.4
a
ND
NYH
1.5
3.8
211
PC-3
0.53
0.35
SK-OV-3
9.9
3.4
a
ND, not determined.
In Vivo Antitumor Activity. Compounds with potent
(nM) activity from all series were taken forward to tests of
pharmacokinetics and preliminary toxicology effects in mouse
as a selection filter for in vivo test in an A2780 xenograft mouse
model (Table 8). Several of the new compounds compared well
with the reference compound 1, and on the basis of these data,
compounds were selected. Most compounds in these series
exhibited a relatively short half-life, approximately 20 min, but
with an adequate systemic exposure (AUC).Therefore, in this
early screen for pharmacological activity in vivo, the
compounds were administered ip to secure a sufficient exposure
of the compound in the animal.
In a comparative study with these two compounds (15 and
17) and the reference compound 1, compound 15 was found
approximately 12 times more potent than 1 in the A2780
xenograft mouse model (schedule: ip bid day 0−4, starting
tumor volume 100 mm3) as measured by the tumor volume at
day 14. Significant reduction of A2780 tumor volume at day 14
was observed after treatment with compound 15, 1.25 and 2.5
Table 8. Toxicological and Pharmacokinetic Data for Selected Derivatives as Determined in Mouse
a
toxicology
pharmacokinetics
1
compd
MTD (mg/kg)
dose for PK (mg/kg) route t / (h) tmax (h) Cmax (ng/mL) Vx (mL/kg) CL (mL/(h·kg)) AUC ((h·ng)/mL)
2
1
10 < MTD < 50
MTD < 10
20
50
50
50
50
iv
iv
iv
iv
iv
0.4
0.25
0.08
0.08
0.083
0.08
14 563
14 102
78 519
33 535
14 827
998
1620
4906
1052
3995
7105
12 337
10 184
47 514
12 499
7018
15
17
31
37
0.18
0.36
0.21
0.45
1275
547
10 < MTD < 50
10 < MTD < 50
MTD < 10
1200
4600
a
The compound toxicity (MTD) was estimated in NMRI mice dosing the compounds bid ip for five consecutive days, determining weight loss and
blood cell counts.
9082
dx.doi.org/10.1021/jm4009949 | J. Med. Chem. 2013, 56, 9071−9088

Journal of Medicinal Chemistry
Article
Figure 11. Tumor volume at day 14 after treatment with compounds
15 (0.63, 1.25, 2.5 mg/kg), 17 (15 mg/kg), and 1 (15 mg/kg) bid day
0−4 as compared with vehicle, in a A2780 xenograft mouse model.
The structure-based method used for the optimization of these
structures gave a rationale for the obtained activities and
knowledge about the scope and limitation in the design of new
NAMPT inhibitors. The most active compounds in these new
series compared favorably, in respect to toxicology and in vivo
activity, with compounds already in the clinic and warrant
further investigation as promising lead molecules for the
inhibition of NAMPT.
EXPERIMENTAL SECTION
■
1
Reaction conditions and yields were not optimized. H and 13C NMR
spectra were recorded on a Bruker Avance 300 spectrometer (300
MHz) or Varian 400 OXFORD NMR spectrometer (400 MHz).
Chemical shifts are reported in parts per million (δ) and referenced to
hexamethyldisiloxane (HMDSO) as an internal standard or using the
1
signal according to deuterated solvent for H spectra (CDCl3, 7.26;
CD3OD, 3.31; (CD3)2SO, 2.50) and 13C spectra (CDCl3, 77.23;
CD3OD, 49.00; (CD3)2SO, 39.52). The value of a multiplet, either
defined (doublet (d), triplet (t), double doublet (dd), double triplet
(dt), quartet (q)) or not (m) at the approximate midpoint is given
unless a range is quoted. A broad singlet is indicated by “bs”. Mass
spectrometry (MS) was performed using an LC-MS using a Bruker
Esquire 3000+ ESI ion-trap with an Agilent 1200 HPLC-system or on
an Acquity UPLC system (Waters) connected to the Micromass Q-
TOF micro hybrid quadrupole time-of-flight mass spectrometer
operating in the electrospray ionization (ESI) positive ion mode and
using a reverse-phase Acquity UPLC BEH C18 column (1.7 μm, 2.1
mm × 50 mm) on a gradient of 5−98% acetonitrile−water 0.1%
formic acid. All tested compounds were of sufficient purity (>95%) as
determined by high-performance liquid chromatography (HPLC),
using an Agilent 1200 HPLC system. High-resolution MS (HRMS)
was carried out on a Micromass Q-Tof micro mass spectrometer.
Elemental analyses were performed on Carlo Erba CHNS-O EA-1108
apparatus. Melting points were measured on a ‘‘Boetius’’ or
Gallenkamp melting point apparatus and are uncorrected. Silica gel,
0.035 e 0.070 mm, (Acros) was employed for column chromatog-
raphy.
Preparation of Key Compounds 15 and 17. For a full
description of the preparation and spectroscopic data of compounds
reported in this paper please see the Supporting Information.
6-(2-Cyano-3-(pyridin-4-yl)guanidino)-N-cyclohexyl-N-(2-
morpholinoethoxy)hexane-1-sulfonamide (15). 2-(2-
Morpholinoethoxy)isoindoline-1,3-dione (XIVc). Synthesis of com-
pound XIVc by a modified version of published procedures45−47 was
performed as follows: To a mixture of 2-hydroxyisoindoline-1,3-dione
(33.92 g, 208 mmol) and 4-(2-chloroethyl)morpholine hydrochloride
(50.24 g, 270 mmol) in 1-methyl-2-pyrrolidinone (160 mL) slowly
was added DBU (80 mL, 535 mmol), and the resulting mixture was
Figure 10. Treatment with compound 15 in an A2780 xenograft
mouse model (large tumors, 3 mg/kg, ip, bid). (A) Mean tumor
volume; (B) tumor volume in individual mice; (C) body weight during
treatment in individual mice.
mg/kg, and compound 1, 15 mg/kg. A dose−response effect of
15, 0.63, 1.25, and 2.5 mg/kg, was observed with percent
volume of treated tumors versus control T/C% of 82%, 40%,
and 24%, respectively. Also, some effect (T/C = 52%), however
not significant, was determined after treatment with compound
17 and compound 1 (Figure 11). Critical body weight losses
were not observed in any treatment groups.
In summary, in vivo the selected compounds showed an
equal or increased potency as compared with reference
compound 1, with no critical signs of toxicology using effective
doses. These highly promising results suggest a continuation of
studies of this compound class.
CONCLUSION
■
The present study describes the successful discovery of novel
NAMPT inhibitors with promising biological activities as
antiproliferative agents in cancer cell lines in vitro and potent
tumor reduction efficacy in vivo in a xenograft mouse model.
9083
dx.doi.org/10.1021/jm4009949 | J. Med. Chem. 2013, 56, 9071−9088

Journal of Medicinal Chemistry
Article
stirred at 45 °C for 6 h. The mixture was poured into water (500 mL)
and extracted with ethyl acetate (3 × 250 mL), and the combined
extract was washed with brine (200 mL) and dried (Na2SO4). The
solvent was evaporated, and the residue was dried in vacuo to give
compound XIVc (39.0 g, 68%) as an oil that solidified on standing. 1H
NMR (200 MHz, CDCl3) δ 2.50 (m, 4H), 2.79 (t, J = 5.5 Hz, 2H),
3.59 (m, 4H), 4.37 (t, J = 5.5 Hz, 2H), 7.70−7.89 (m, 4H).
3.56 (t, 2H, J = 7.0 Hz), 7.77−7.92 (m, 4H). LCMS (ESI): m/z 312
[Msulfonic acid + H]+.
6-(1,3-Dioxo-1,3-dihydro-isoindol-2-yl)-hexanesulfonyl chloride
(XXVb). A mixture of sodium 6-(1,3-dioxo-1,3-dihydro-2H-isoindol-
2-yl)-1-hexanesulfonate (XXIVb) (13.99 g, 42.0 mmol) and
phosphorus pentachloride (28.0 g, 134.5 mmol) was carefully ground
in a mortar (Caution: a good hood has to be used!). An evaluation of
some amount of gas was observed, and gradually (in 2−3 min), the
solid mixture turned into an oily liquid. The obtained liquid was mixed
with toluene (280 mL), the precipitated solid material was filtered off
and washed with toluene, and the filtrates were combined. The solvent
was evaporated, and the residue was azeotropically dried several times
with toluene. The obtained white solid was dissolved in ethyl acetate
(300 mL), washed successively with water (100 mL), saturated sodium
bicarbonate (2 × 100 mL), and brine (2 × 100 mL), and dried
(Na2SO4). The solvent was evaporated, and the residue was dried in
vacuo over P2O5 to afford compound XXVb (10.4 g, 75%) as white
crystals. mp 73−75 °C. 1H NMR (400 MHz, CDCl3) δ 1.41 (qui, 2H,
J = 7.6 Hz), 1.55 (qui, 2H, J = 7.6 Hz), 1.72 (qui, 2H, J = 7.4 Hz), 2.04
(m, 2H); 3.65 (m, 2H); 3.69 (t, 2H, J = 7.1 Hz); 7.67−7.75 (m, 2H);
7.80−7.87 (m, 2H). Anal. calcd. for C14H16ClNO4S: C, 50.99; H, 4.89;
N, 4.25; S, 9.72. Found: C, 51.03; H, 5.01; N, 4.17; S, 9.68.
O-(2-Morpholinoethyl)hydroxylamine (VIIIe). To a solution of 2-
(2-morpholinoethoxy)isoindoline-1,3-dione (XIVc) (39.0 g, 141
mmol) in a mixture of methanol (200 mL) and dichloromethane
(100 mL) was added hydrazine hydrate (20 mL, 411 mmol), and the
obtained mixture was stirred at room temperature overnight. The
resulting precipitate was filtered off, and the filtrate was concentrated
in vacuo. The residue (22.9 g) was mixed with water (200 mL), to this
mixture was added conc. HCl (30 mL), and the solid material was
filtered off. The filtrate was washed with EtOAc (200 mL), and the pH
of the medium was raised to 10 by adding 5 N aqueous NaOH. The
mixture was extracted with chloroform (3 × 300 mL), and the extract
was washed with brine (100 mL) and dried (Na2SO4). The solvent
was evaporated, and the residue was dried in vacuo to give O-(2-
morpholinoethyl)hydroxylamine (VIIIe) (20.7 g, quantitative yield).
1H NMR (200 MHz, CDCl3) δ 2.44−2.55 (m, 4H), 2.59 (t, J = 5.4
Hz, 2H), 3.69−3.77 (m, 4H), 3.81 (t, J = 5.4 Hz, 2H), 5.50 (b s, 2H).
Cyclohexanone O-(2-morpholin-4-ylethyl)oxime (XIXd). To a
stirred solution of cyclohexanone oxime (XVIIc) (22.64 g, 0.2 mol)
in dimethylformamide (200 mL) at ice-bath temperature, 60% sodium
hydride in mineral oil (16.0 g, 0.4 mol) was added portion-wise, and
the resulting mixture was stirred at this temperature for 1 h. To the
reaction mixture was added a suspension of 4-(2-chloroethyl)-
morpholine (37.22 g, 0.2 mol) in dimethylformamide (100 mL).
The ice-bath was removed, and the reaction mixture was stirred at
room temperature for 16 h and at 60 °C for 3 h. The mixture was
allowed to cool to room temperature and then filtered, and the filtrate
was evaporated. The residue was mixed with a saturated ammonium
chloride solution in water (300 mL) and extracted with diethyl ether
(3 × 200 mL). The combined organic extracts were washed
successively with 0.5 N sodium hydroxide (200 mL), brine (200
mL), and dried (Na2SO4). The solvent was evaporated, and the
residue (36.47 g) was chromatographed on silica gel (250 g) with
chloroform−methanol (40:1) as eluent to give compound XIXd (35.1
N-Cyclohexyl-6-(1,3-dioxo-1,3-dihydro-2H-isoindol-2-yl)-N-(2-
morpholinoethoxy)-1-hexanesulfonamide (XXVIe). A solution of N-
cyclohexyl-O-(2-morpholin-4-yl-ethyl)-hydroxylamine (VIIIc) (12.56
g, 55 mmol) and triethylamine (14.0 mL, 100 mmol) in dry
dichloromethane (100 mL) under argon atmosphere was cooled to
−20 °C. To the stirred solution slowly during 1 h was added a solution
of 6-(1,3-dioxo-1,3-dihydro-2H-isoindol-2-yl)hexane-1-sulfonyl chlor-
ide (XXVb) (16.49 g, 50 mmol) in dichloromethane (50 mL), and the
resulting mixture was stirred at −20 °C for 20 h. The reaction mixture
was concentrated to a small volume and filtered, and the precipitated
solid material was washed with dichloromethane. The filtrate was
evaporated, and the residue (32.25 g) was dissolved in a small volume
of chloroform and chromatographed on silica gel (450 g) with
hexane−isopropanol (gradient from 7:3 to 6:4) as eluent. The eluate,
containing pure product by thin layer chromatography (TLC), was
separated, and the impure material was rechromatographed using the
same eluent. The eluates with TLC pure material were combined, the
solvent was evaporated, and the residue was dried in vacuo to afford
compound XXVIe (16.4 g, 62.8%) as a crystalline solid. 1H NMR (400
MHz, CDCl3) δ 1.10 (tq, 1H, J = 3.5, 12.9 Hz), 1.20−1.44 (m, 4H),
1.44−1.76 (m, 7H), 1.76−1.95 (m, 6H), 2.49 (m, 4H), 2.58 (t, 2H, J =
5.6 Hz), 3.08 (bs, 2H); 3.58 (tt, 1H, J = 3.6, 11.7 Hz), 3.68 (t, 2H, J =
7.0 Hz), 3.69 (m, 4H), 4.12 (bs, 2H), 7.71 (m, 2H), 7.83 (m, 2H).
6-Amino-N-cyclohexyl-N-(2-morpholinoethoxy)-1-hexanesulfo-
namide (Ve). N-Cyclohexyl-6-(1,3-dioxo-1,3-dihydro-2H-isoindol-2-
yl)-N-(2-morpholinoethoxy)-1-hexanesulfonamide (XXVIe) (17.608
g, 33.75 mmol) was dissolved in a mixture of chloroform (150 mL)
and absolute ethanol (150 mL), and hydrazine hydrate (4.2 mL, 86.54
mmol) was added. The obtained mixture was refluxed for 5 h and then
stirred at room temperature overnight. The mixture was filtered, the
precipitate was washed with dichloromethane, and the combined
filtrates were evaporated. The residue was dissolved in dichloro-
methane (50 mL), and the mixture was kept in a refrigerator (∼5 °C)
for 1 h and then filtered again. The filtrate was evaporated, and the
residue (14.392 g) was chromatographed on silica gel (200 g) with
methanol−30% ammonium hydroxide aqueous solution (gradient
from 25:1 to 20:1) to give 8.38 g of an oil. The oil was dissolved in
dichloromethane (100 mL), washed successively with water (2 × 20
mL) and brine (20 mL), and dried (Na2SO4). The solvent was
evaporated, and the residue was dried in vacuo at 50 °C to give
1
g, 77.5%) as a yellow oil. H NMR (400 MHz, CDCl3) δ 1.54−1.71
(m, 6H), 2.18 (m, 2H), 2.43 (m, 2H), 2.52 (m, 4H), 2.66 (t, 2H, J =
5.8 Hz), 3.71 (m, 4H), 4.15 (t, 2H, J = 5.8 Hz).
4-{2-[(Cyclohexylamino)oxy]ethyl}morpholine (VIIIc). To a sol-
ution of cyclohexanone O-(2-morpholin-4-ylethyl)oxime (XIXd) (13.9
g, 61.4 mmol) in methanol (100 mL) at ice-bath temperature were
added sodium cyanoborohydride (7.72 g, 122.8 mmol) and trace of
Methyl Orange. To the obtained slightly yellow solution, slowly 2 N
HCl solution in methanol was added until the color of the reaction
mixture changed from yellow to pink (in about 15 min). The reaction
mixture was stirred at room temperature for 5 h, and the solvent was
evaporated. To the residue was added water (30 mL), and the pH of
the obtained solution was raised to a pH value greater than 9 with 6 N
KOH, saturated with sodium chloride. The obtained mixture was
extracted with chloroform (3 × 150 mL), and the combined organic
extract was washed with brine (100 mL) and dried (Na2SO4). The
solvent was removed, and the residue was dried in vacuo to afford title
1
compound VIIIc (13.0 g, 92%) as a yellow oil. H NMR (400 MHz,
CDCl3) δ 1.00−1.33 (m, 5H), 1.62 (m, 1H), 1.73 (m, 2H), 1.84 (m,
2H), 2.48 (m, 4H), 2.56 (t, 2H, J = 5.7 Hz), 2.84 (tt, 1H, J = 3.7, 10.5
Hz), 3.71 (m, 4H), 3.81 (t, 2H, J = 5.7 Hz), 5.43 (br s, 1H). LCMS
(ESI) m/z: 299 [M + H]+.
Sodium 6-(1,3-Dioxo-1,3-dihydro-2H-isoindol-2-yl)-1-hexanesul-
fonate (XXIVb). To a hot solution of 2-(6-bromohexyl)-1H-isoindole-
1,3(2H)-dione (XXIIIb)48 (12.67 g, 40.8 mmol) in ethanol (82 mL)
was added a solution of sodium sulfite (10.3 g, 81.7 mmol) in water
(80 mL), and the resulting mixture was refluxed overnight. The hot
mixture was filtrated, crystallized from ethanol, and dried in vacuo over
1
compound Ve (7.67 g, 63.8%) as a yellow oil. H NMR (400 MHz,
CDCl3) δ 1.10 (tq, 1H, J = 3.4, 12.9 Hz), 1.20−1.52 (m, 8H), 1.52−
1.69 (m, 5H), 1.75−1.98 (m, 6H), 2.49 (m, 4H), 2.59 (t, 2H, J = 5.6
Hz), 2.70 (t, 2H, J = 6.8 Hz), 3.09 (bs, 2H), 3.59 (tt, 1H, J = 3.6, 11.7
Hz), 3.69 (m, 4H), 4.12 (b s, 2H). LCMS (ESI) m/z: 392 [M + H]+.
Anal. calcd. for C18H37N3O4S·0.15 H2O: C, 54.83; H, 9.54; N, 10.66;
S, 8.13. Found: C, 54.86; H, 9.66; N, 10.63; S, 8.14.
1
P2O5 to give compound XXIVb (8.43 g, 62%). H NMR (200 MHz,
(CD3)2SO) δ 1.16−1.41 (m, 4H), 1.41−1.68 (m, 4H), 2.37 (m, 2H),
9084
dx.doi.org/10.1021/jm4009949 | J. Med. Chem. 2013, 56, 9071−9088

Journal of Medicinal Chemistry
Article
6-(2-Cyano-3-(pyridin-4-yl)guanidino)-N-cyclohexyl-N-(2-
morpholinoethoxy)hexane-1-sulfonamide (15). A mixture of 6-
amino-N-cyclohexyl-N-(2-morpholinoethoxy)-1-hexanesulfonamide
(Ve) (2.49 g, 6.4 mmol), 4-[(cyanoimino)(methylsulfanyl)methyl]-
aminopyridine (IIIa) (1.22 g, 6.3 mmol), triethylamine (3 mL, 21.6
mmol), and 4-dimethylaminopyridine (0.1 g, 0.8 mmol) in dry
pyridine (4 mL) was stirred at 75−80 °C for 5 h. The solvent was
evaporated to dryness, and the residue was chromatographed on silica
gel (150 g) with acetonitrile−water (10:1) as eluent to give compound
15 (1.8 g, 53%) as a foam together with a less pure material (1.0 g,
29%) that can be purified repeatedly by column chromatography to
2H, J = 6.9 Hz), 3.08 (m, 2H), 3.54 (td, 2H, J = 5.0, 23.6 Hz), 3.87 (d,
2H, J = 6.5), 4.62 (td, 2H, J = 5.0, 47.0 Hz).
6-(2-Cyano-3-(pyridin-4-yl)guanidino)-N-(cyclohexylmethoxy)-N-
(2-fluoroethyl)hexane-1-sulfonamide (17). A mixture of 6-amino-N-
(cyclohexylmethoxy)-N-(2-fluoroethyl)-1-hexanesulfonamide (Vi)
(1.95 g, 5.76 mmol), 4-[(cyanoimino)(methylsulfanyl)methyl]-
aminopyridine (IIIa) (1.1 g, 5.76 mmol), triethylamine (0.93 mL,
6.68 mmol), and 4-dimethylaminopyridine (0.15 g, 1.22 mmol) in dry
pyridine (20 mL) was stirred at 85 °C for 20 h. The solvent was
evaporated to dryness; the residue was azeotropically dried with
toluene (2 × 5 mL) and then vigorously stirred with ether (30 mL)
until the precipitation occurred (∼2 h). The obtained suspension was
filtered, and the solid material (2.7 g) was chromatographed on silica
gel with chloroform−methanol−30% ammonium hydroxide aqueous
solution (6:1:0.015) as eluent to give compound 17 (2.48 g, 89%) as
white crystals. mp 100−102 °C. 1H NMR (400 MHz, CDCl3) δ 0.90−
1.04 (m, 2H), 1.11−1.29 (m, 3H), 1.41 (qui, 2H, J = 7.3 Hz), 1.48−
1.75 (m, 10H), 1.91 (qui, 2H, J = 7.6 Hz), 3.09 (t, 2H, J = 7.5 Hz),
3.36 (q, 2H, J = 6.6 Hz), 3.52 (dt, 2H, J = 4.9, 24.0 Hz), 3.86 (d, 2H, J
= 6.5 Hz), 4.61 (dt, 2H, J = 4.9, 47.0 Hz), 5.51 (bs, 1H), 7.20 (d, 2H, J
= 4.9 Hz), 7.68 (bs, 1H), 8.55 (d, 2H, J = 4.9 Hz). Anal. calcd. for
C22H35FN6O3S: C, 54.75; H, 7.31; N, 17.41. Found: C, 54.85; H, 7.42;
N, 17.50. HRMS m/z calcd. for C22H36FN6O3S [M + H]+, 483.2554;
found, 483.2526.
Cell Culture. Human breast carcinoma MCF-7 and ovarian
carcinoma A2780 were grown according to American Type Culture
Collection guidelines. Cell culture media were from Invitrogen unless
otherwise stated. MCF-7 was maintained in DMEM and A2780 in
RPMI 1640 with GlutaMax. Media was supplemented with 10%(v/v)
FCS (Perbio, Thermo Fischer Scientific), penicillin (100 U/mL), and
streptomycin (0.1 mg/mL), and cells were incubated at 37 °C in an
atmosphere containing 5% CO2.
WST-1 Proliferation Assay. Cells were seeded in 96-well plates (3
× 103 cells/well) in culture medium (100 μL). The following day
compounds were serially diluted in culture medium, and 100 μL of
each dilution were added per well in triplicate to the cell culture plates.
Plates were incubated (72 h, 37 °C, 5% CO2 atmosphere), and the
number of viable cells were assessed using cell proliferation reagent
WST-1 (Roche, Mannheim, Germany). Reagent (10 μL) was added to
each well, and after a 1 h incubation period, absorbance was measured
at 450 nm subtracting absorbance at 690 nm as a reference. Data were
analyzed using GraphPad Prism (GraphPad Software, CA, USA) and
Calcusyn (Biosoft, Cambridge, UK) as appropriate.
1
increase the yield of the process. H NMR (200 MHz, (CD3)2SO) δ
0.79−1.66 (m, 13H), 1.66−1.96 (m, 5H), 2.38−2.47 (m, 4H), 2.47−
2.60 (m, 2H, overlapped with DMSO), 3.12−3.33 (m, 4H), 3.39−3.54
(m, 1H), 3.52−3.63 (m, 4H), 4.02 (t, J = 5.6 Hz, 2H), 7.21 (d, J = 5.3
Hz, 2H), 7.87 (t, J = 5.5 Hz, 1H), 8.38 (d, J = 5.6 Hz, 2H), 9.41 (bs,
1H). HRMS m/z calcd. for C25H42N7O4S [M + H]+, 536.3019; found,
536.2976.
6-(2-Cyano-3-(pyridin-4-yl)guanidino)-N-(cyclohexylmethoxy)-N-
(2-fluoroethyl)hexane-1-sulfonamide (17). N-(Cyclohexylmethoxy)-
6-(1,3-dioxo-1,3-dihydro-2H-isoindol-2-yl)-1-hexanesulfonamide
(XXVIIc). To a solution of O-(cyclohexylmethyl)hydroxylamine
(VIIId) (1.1 g, 8.51 mmol) and triethylamine (2.3 mL, 16.55
mmol) in dry dichloromethane (40 mL) at ice-bath temperature
slowly for 2 h was added 6-(1,3-dioxo-1,3-dihydro-2H-isoindol-2-
yl)hexane-1-sulfonyl chloride (XXVb) (3.08 g, 9.34 mmol) portion-
wise. The reaction mixture was allowed gradually to warm up to room
temperature (for 1 h) and evaporated. The residue was dissolved in
ethyl acetate (100 mL), washed successively with water (2 × 15 mL)
and brine (30 mL), and dried (Na2SO4). The solvent was evaporated,
and the residue was dried in vacuo over P2O5 to afford compound
XXVIIc (3.2 g, 89%) as crystalline solid. 1H NMR (400 MHz, CDCl3)
δ 0.86−1.00 (m, 2H), 1.08−1.30 (m, 3H), 1.33−1.44 (m, 2H), 1.46−
1.57 (m, 3H), 1.61−1.77 (m, 7H), 1.75−1.85 (m, 2H), 3.18 (m, 2H),
3.68 (t, 2H, J = 7.2 Hz), 3.80 (d, 2H, J = 6.2 Hz), 6.99 (s, 1H), 7.71
(m, 2H), 7.84 (m, 2H).
N-(Cyclohexylmethoxy)-6-(1,3-dioxo-1,3-dihydro-2H-isoindol-2-
yl)-N-(2-fluoroethyl)-1-hexanesulfonamide (XXVIi). To a solution of
N-(cyclohexylmethoxy)-6-(1,3-dioxo-1,3-dihydro-2H-isoindol-2-yl)-1-
hexanesulfonamide (XXVIIc) (2.11 g, 5.0 mmol), 2-fluoroethanol (0.4
g, 6.2 mmol), and triphenylphosphine (2.16 g, 8.2 mmol) in
dichloromethane (40 mL) at ice-bath temperature slowly in 10 min
was added a solution of diethyl azodicarboxylate (1.43 g, 8.2 mmol) in
dichloromethane (1.5 mL), and the resulting mixture was stirred at
this temperature for 20 min. The ice bath was removed, and then, the
reaction mixture was stirred at room temperature for 3 h and
evaporated. The residue was mixed with petroleum ether−ethyl
acetate (4:1, 25 mL), and the obtained suspension was stirred for 30
min at ice-bath temperature and filtered. The filtrate was evaporated,
and the residue was chromatographed on silica gel with toluene−ethyl
acetate (9:1) as eluent to afford compound XXVIi (1.57 g, 67%) as
Clonogenic Assays. HCT-116/APO866 resistant cell line was
obtained as described previously.42
In vitro colony forming assays were performed essentially as
previously published.49 Briefly, HCT116 cells were cultured with
compounds for the indicated times and seeded onto 35 mm dishes in
agar (3% w/v) containing a sheep erythrocyte feeder layer. Agar plates
were cultured for 14−21 days at 37 °C, and colonies were counted
using a digital colony counter and Sorcerer image analysis software
(Perceptive Instruments Ltd., SuVolk, UK). Data were analyzed using
GraphPad Prism and Calcusyn as appropriate.
NAMPT Enzyme Assay. NAMPT enzyme activity was measured
as described previously with minor modifications.50,51 In this
procedure, the NAMPT-catalyzed formation of 14C-nicotinamide
mononucleotide (NMN) was determined, using 14C-nicotinamide
and 5-phosphoribosyl-1-pyrophosphate (PRPP) as substrates.
For preparation of lysates, confluent HepG2 cells were washed
twice with phosphate-buffered saline (PBS) (4 °C, Ca2+, Mg2+ free),
once with NaHPO4 buffer, (0.01 N, pH 7.4), and scraped in NaHPO4
buffer. After centrifugation (10 min, 500g, 4 °C), the pelleted cells
were resuspended by pipetting in NaHPO4 buffer to a concentration of
approximately 107 cells/100 μL for HepG2 and aliquoted (200 μL
aliquots in 2 mL tubes). Cells were then broken up by sonography on
ice (Bandelin Sonopuls, 3 × 10 s, approximately 30% power). Cell
debris was removed by centrifugation (23 000g, 90 min, 0 °C).
Protamine sulfate solution (1% in NaHPO4 buffer) was added to the
supernatant (70 μL/mL supernatant) to precipitate DNA by
1
white crystals. H NMR (400 MHz, CDCl3) δ 0.90−1.06 (m, 2H),
1.09−1.33 (m, 4H), 1.33−1.45 (m, 2H), 1.45−1.79 (m, 9H), 1.83−
1.95 (m, 2H), 3.07 (m, 2H), 3.54 (td, 2H, J = 5.0, 23.8 Hz), 3.68 (t,
2H, J = 7.1 Hz), 3.86 (d, 2H, J = 6.4), 4.62 (td, 2H, J = 5.0, 47.1 Hz),
7.71 (m, 2H), 7.83 (m, 2H).
6-Amino-N-(cyclohexylmethoxy)-N-(2-fluoroethyl)-1-hexanesul-
fonamide (Vi). N-(Cyclohexylmethoxy)-6-(1,3-dioxo-1,3-dihydro-2H-
isoindol-2-yl)-N-(2-fluoroethyl)-1-hexanesulfonamide (XXVIi) (1.53
g, 3.26 mmol) was dissolved in a mixture of ethanol (20 mL) and
chloroform (10 mL), and hydrazine hydrate (0.5 mL, 103 mmol) was
added. The reaction mixture was stirred at 60 °C for 2 h, left overnight
at room temperature, and cooled in the refrigerator (5 °C). The
precipitated solid was filtered off, and the filtrate was evaporated. The
residue was chromatographed on silica gel with chloroform−
methanol−30% ammonium hydroxide aqueous solution (5:1:0.15)
1
as eluent to give compound Vi (0.93 g, 84%) as white crystals. H
NMR (400 MHz, CDCl3) δ 0.90−1.04 (m, 2H), 1.09−1.30 (m, 3H),
1.33−1.54 (m, 6H), 1.54−1.78 (m, 6H), 1.82−2.02 (m, 4H), 2.72 (t,
9085
dx.doi.org/10.1021/jm4009949 | J. Med. Chem. 2013, 56, 9071−9088

Journal of Medicinal Chemistry
Article
incubation on ice for 15 min. After centrifugation (23 000g, 30 min, 0
°C), aliquots of the supernatant were stored at −80 °C.
AUTHOR INFORMATION
Corresponding Author
*Phone: +45 35 33 62 31; fax: +45 35 33 60 41; e-mail: fb@
sund.ku.dk.
■
Various concentrations of inhibitor or adequate concentrations of
DMSO as solvent control and cell lysates (10 μL) were added to a
total of 50 μL of reaction mixture (50 mmol/L Tris−HCl pH 7.4; 2
mmol/L ATP; 5 mmol/L MgCl2; 0.5 mmol/L PRPP; 6.2 μmol/L 14C-
nicotinamide; American Radiolabeled Chemicals, St. Louis; MO,
USA) and incubated (37 °C, 1 h). The reaction was terminated by
transfer into tubes containing acetone (2 mL). The whole mixture was
then pipetted onto acetone-presoaked glass microfiber filters (GF/A Ø
24 mm; Whatman, Maidstone, UK). After rinsing with acetone (2 × 1
mL), filters were dried and transferred into vials with scintillation
cocktail (6 mL, Betaplate Scint, PerkinElmer, Waltham, MA, USA),
and radioactivity of 14C-NMN was quantified in a liquid scintillation
counter (Wallac 1409 DSA, Perkin−Elmer). After subtraction of blank
values, NAMPT activity was normalized to total protein as measured
by BCA assay (Pierce).
Present Addresses
∇Mette K. Christensen: Novo Nordisk A/S, 2880 Bagsværd,
Denmark.
○Annemette Thougaard: H. Lundbeck A/S, 2500 Valby,
Denmark.
◆Søren Jensby Nielsen: Nuevolotion A/S 2100 Copenhagen,
Denmark.
¶Peter B. Jensen: Medical Prognosis Institute, 2970 Hørsholm,
Denmark.
▲Kamille D. Erichsen: BioAdvice A/S, 2950 Vedbæk, Den-
mark.
Xenograft Studies. The antitumor effect in vivo was tested in an
A2780 (ovarian cancer) subcutaneous (sc) xenograft model in nude
mice (female, NMRI/nude, Harlan or Taconic). Cancer cells were
grown in RPMI with 10% FBS, washed once with PBS, and suspended
in 100 μL of PBS and 100 μL of matrigel (BD) and injected sc.
Treatment started at tumor volumes around 100 mm3 (small tumors)
or 500 mm3 (large tumor). The compounds were formulated in 2%
DMSO, 20% HP-β-CD, and isotonic saline at 10 mL/kg ip injection.
Tumor diameters were measured during tumor growth and tumor
volumes (Tv) estimated according to the following formula: Tv =
(width2 × length)/2. Mice were observed for tumor regression after 1
week or else sacrificed. The experiments were conducted at
Topotarget A/S, Copenhagen and approved by the Experimental
Animal Inspectorate, Danish Ministry of Justice.
Pharmacokinetic Analysis. Mouse plasma samples were prepared
for analysis by protein precipitation on Sirocco plates (Waters,
Milford, MA, USA). A Waters Acquity UPLC system with a Quattro
Premier MS-MS system was used for separation and detection.
Acetonitrile containing 1 μg/mL internal standard was used in the
ratio 3:1 (v/v) for precipitation. Separation was performed with an
acetonitrile/0.05% formic acid gradient on a Acquity UPLC BEH C18,
2.1 mm × 50 mm, 1.7 μm reversed-phase column (Waters A/S)
operated at 40 °C. Detection was performed using electrospray MRM
in the positive mode. Pharmacokinetic parameters were calculated
using noncompartmental analysis methods as included in WinNonlin
ver 5.02 (Pharsight, CA, USA).
Docking Analysis. The structure was downloaded from the
protein data bank (PDB ID 2GVJ) and prepared for docking using the
built-in protein preparation wizard in Maestro version 9.3. During this
process, bond orders were assigned, and hydrogens were added to the
crystal structure. Furthermore, the four seleno-methionines that had
been incorporated to allow for better X-ray diffraction were changed to
cysteines (chain A: residues 368 and 372; chain B: residues 368 and
372). The docking was carried out using Glide version 5.8 in extra
precision (XP) mode. The ligands were docked flexibly, and nitrogen
inversions and ring flips were allowed. The van der Waals radii of the
nonpolar ligand atoms (partial charge < 0.15) were scaled by a factor
of 0.8 to accommodate slightly inaccurate initial dockings. A
postdocking minimization was carried out for the best 25 poses for
each ligand, and finally, the 10 best poses were reported.
Notes
The authors declare the following competing financial
interest(s): Author J.T. is an employee of Topotarget A/S.
Authors K.D.E, M.K.C., U.H.O., A.T., S.J.N., M.S., P.B.J., and
F.B. are previous employees of Topotarget A/S.
ACKNOWLEDGMENTS
■
We thank Anja Barnikol-Oettler for expert technical assistance
with the NAMPT enzymatic assay and support from the
German Research Council (DFG, KFO152-TP7) and the
Leipzig LIFE (LIFE Child Health and Child Obesity) program
to W.K. P.F. was supported by a Sapare Aude grant from the
Danish Council for Independent Research no. 11-105487. We
thank Nicolaj Høj and Søren Ryborg for organic synthesis
assistance.
ABBREVIATIONS USED
■
cADPR, cyclic ADP ribose; NAADP, nicotinic acid adenine
dinucleotide phosphate; AUC, area under the curve; bid, twice
daily; DMEM, Dulbecco’s modified eagle medium; HRMS,
high-resolution mass spectrometry; IC50, concentration of a test
compound that produces half maximal inhibition; LC-MS,
liquid chromatography−mass spectrometry; MTD, maximum
tolerated dose; tmax, time of maximum drug concentration; Cmax
,
maximum drug concentration; Vz, volume of distribution; CL,
drug clearance; NMR, nuclear magnetic resonance; SAR,
structure−activity relationship; SD, standard deviation
REFERENCES
■
(1) Khan, J. A.; Forouhar, F.; Tao, X.; Tong, L. Nicotinamide adenine
dinucleotide metabolism as an attractive target for drug discovery.
Expert Opin. Ther. Targets 2007, 11 (5), 695−705.
(2) Diefenbach, J.; Burkle, A. Introduction to poly(ADP-ribose)
metabolism. Cell. Mol. Life Sci. 2005, 62 (7−8), 721−730.
(3) Ziegler, M. New functions of a long-known molecule. Emerging
roles of NAD in cellular signaling. Eur. J. Biochem. 2000, 267 (6),
1550−1564.
DFT Analysis. This study was carried out using Jaguar version
8.0.52 DFT used the B3LYP functional53−55 with added d3
corrections56,57 to account for dispersion interactions. We used the
6-31G** basis set58 throughout.
(4) Saunders, L. R.; Verdin, E. Sirtuins: critical regulators at the
crossroads between cancer and aging. Oncogene 2007, 26 (37), 5489−
5504.
(5) Genazzani, A. A.; Billington, R. A. NAADP: An atypical Ca2+-
release messenger. Trends Pharmacol. Sci. 2002, 23, 165−167.
(6) Malavasi, F.; Deaglio, S.; Funaro, A.; Ferrero, E.; Horenstein, A.
L.; Ortolan, E.; Vaisitti, T.; Aydin, S. Evolution and function of the
ADP ribosyl cyclase/CD38 gene family in physiology and pathology.
Physiol. Rev. 2008, 88, 841−886.
ASSOCIATED CONTENT
■
S
* Supporting Information
Experimental procedures; analytical and spectral data for all
intermediate and final compounds; computation chemistry
docking scores and associated docking poses. This material is
available free of charge via the Internet at http://pubs.acs.org.
(7) Hageman, G. J.; Stierum, R. H. Niacin, poly(ADP-ribose)
polymerase-1 and genomic stability. Mutat. Res. 2001, 475 (1−2), 45−
56.
9086
dx.doi.org/10.1021/jm4009949 | J. Med. Chem. 2013, 56, 9071−9088

Journal of Medicinal Chemistry
Article
(8) Lengauer, C.; Kinzler, K. W.; Vogelstein, B. Genetic instabilities
in human cancers. Nature 1998, 396 (6712), 643−649.
(9) Nomura, F.; Yaguchi, M.; Togawa, A.; Miyazaki, M.; Isobe, K.;
Miyake, M.; Noda, M.; Nakai, T. Enhancement of poly-adenosine
diphosphate-ribosylation in human hepatocellular carcinoma. J.
Gastroenterol. Hepatol. 2000, 15 (5), 529−535.
(10) Hufton, S. E.; Moerkerk, P. T.; Brandwijk, R.; de Bruïne, A. P.;
Arends, J.-W.; Hoogenboom, H. R. A profile of differentially expressed
genes in primary colorectal cancer using suppression substractive
hybridization. FEBS Lett. 1999, 463, 77−82.
(11) van Beijnum, J. R.; Moerker, P. T.; Gerbers, A. J.; de Bruïne, A.
P.; Arends, J.-W.; Hoogenboom, H. R.; Hufton, S. E. Target validation
for genomics using peptide-specific phage antibodies: a study of five
gene products overexpressed in colorectal cancer. Int. J. Cancer 2002,
101, 118−127.
Patrone, F.; Nencioni, A. APO866 activity in hematologic
malignancies: A preclinical in vitro study. Blood 2009, 113, 6035−
6037.
(27) Nahimana, A.; Attinger, A.; Aubry, D.; Greaney, P.; Ireson, C.;
Thougaard, A. V.; Tjørnelund, J.; Dawson, K. M.; Dupuis, M.;
Duchosal, M. A. The NAD biosynthesis inhibitor APO866 has potent
antitumor activity against hematologic malignancies. Blood 2009, 113,
3276−3286.
(28) Galli, U.; Travelli, C.; Massarotti, A.; Fakhfouri, G.; Rahimian,
R.; Tron, G. C.; Genazzani, A. A. Medicinal chemistry of nicotinamide
phosphoribosyltransferase (NAMPT) Inhibitors. J. Med. Chem. 2013,
56 (16), 6279−6296.
(29) Olesen, U. H.; Christensen, M. K.; Bjorkling, F.; Jaattela, M.;
̈
̈
̈
̈
Jensen, P. B.; Sehested, M.; Nielsen, S. J. Anticancer agent CHS-828
inhibits cellular synthesis of NAD. Biochem. Biophys. Res. Commun.
2008, 367 (4), 799−804.
(12) Foster, J. W.; Moat, A. G. Nicotinamide adenine dinucleotide
biosynthesis and pyridine nucleotide cycle metabolism in microbial
systems. Microbiol. Rev. 1980, 44 (1), 83−105.
(30) Watson, M.; Roulston, A.; Belec, L.; Billot, X.; Marcellus, R.;
́
Bedard, D.; Bernier, C.; Branchaud, S.; Chan, H.; Dairi, K.; Gilbert, K.;
(13) Magni, G.; Amici, A.; Emanuelli, M.; Orsomando, B.; Raffaelli,
N.; Ruggieri, S. Enzymology of NAD+ homeostasis in man. Cell. Mol.
Life Sci. 2004, 61 (1), 19−34.
Goulet, D.; Gratton, M.-O.; Isakau, H.; Jang, A.; Khadir, A.; Koch, E.;
Lavoie, M.; Lawless, M.; Nguyen, M.; Paquette, D.; Turcotte, E.;
Berger, A.; Mitchell, M.; Shore, G. C.; Beauparlant, P. The small
molecule GMX1778 is a potent inhibitor of NAD+ biosynthesis:
Strategy for enhanced therapy in nicotinic acid phosphoribosyltrans-
ferase 1-deficient tumors. Mol. Cell. Biol. 2009, 5872−5888.
(31) Khan, J. A.; Tao, X.; Tong, L. Molecular basis for the inhibition
of human NMPRTase, a novel target for anticancer agents. Nat. Struct.
Mol. Biol. 2006, 13, 582−588.
(14) Preiss, J.; Handler, P. Enzymatic synthesis of nicotinamide
mononucleotide. J. Biol. Chem. 1957, 225, 759−770.
(15) Preiss, J.; Handler, P. Biosynthesis of diphosphopyridine
nucleotide. I. Identification of intermediates. J. Biol. Chem. 1958,
233, 488−492.
(16) Rongvaux, A.; Andris, F.; van Gool, F.; Leo, O. Reconstructing
eukaryotic NAD metabolism. Bioessays 2003, 25, 683−690.
(17) Gross, J. W.; Rajavel, M.; Grubmeyer, C. Kinetic mechanism of
nicotinic acid phosphoribosyltransferase: Implications for energy
coupling. Biochemistry 1998, 37, 4189−4199.
(32) Holen, K.; Saltz, L. B.; Hollywood, E.; Burk, K.; Hanauske, A. R.
The pharmacokinetics, toxicities, and biologic effects of FK866, a
nicotinamide adenine dinucleotide biosynthesis inhibitor. Invest. New
Drugs 2008, 26, 45−51.
(18) Olesen, U. H.; Hastrup, N.; Sehested, M. Expression patterns of
nicotinamide phosphoribosyltransferase and nicotinic acid phosphor-
ibosyltransferase in human malignant lymphomas. APMIS. 2011, 119
(4−5), 296−303.
(19) Shackelford, R. E.; Bui, M. M.; Coppola, D.; Hakam, A. Over-
expression of nicotinamide phosphoribosyltransferase in ovarian
cancers. Int. J. Clin. Exp. Pathol. 2010, 3 (5), 522−527.
(20) Wang, B.; Hasan, M. K.; Alvarado, E.; Huan, H.; Wu, H.; Chen,
W. Y. NAMPT overexpression in prostate cancer and its contribution
to tumor cell survival and stress response. Oncogene 2011, 30 (8),
907−21.
(33) Hovstadius, P.; Larsson, R.; Jonsson, E.; Skov, T.; Kissmeyer, A.
M.; Krasilniko, V. K.; Bergh, J.; Karlsson, M. O.; Lonnebo, A.; Ahlgren,
J. A phase I study of CHS 828 in patients with solid tumour
malignancy. Clin. Cancer Res. 2002, 8, 2843−2850.
(34) Pishvaian, M.; Marshall, J.; Hwang, J.; Malik, S.; He, A.; Deeken,
J.; Kelso, C.; Cotaria, I.; Berger, M. A. Phase I trial of GMX1777, an
inhibitor of nicotinamide phosphoribosyl transferase (NAMPRT),
given as a 24-h infusion. In J. Clin. Oncol., ASCO Annual Meeting
Proceedings (Post-Meeting Edition), Orlando, FL, May 29 to June 2,
2009, ASCO: Alexandria, VA, 2009; Vol. 26, No 15S (May 20
Supplement).
(21) Hasmann, M.; Schemainda, I. FK866, a highly specific
noncompetitive inhibitor of nicotinamide phosphoribosyltransferase,
represents a novel mechanism for induction of tumor cell apoptosis.
Cancer Res. 2003, 63, 7436−7442.
(35) Binderup, E.; Bjorkling, F.; Hjarnaa, P. V.; Latini, S.; Baltzer, B.;
̈
Carlsen, M.; Binderup, L. EB1627: A soluble prodrug of the potent
anticancer cyanoguanidine CHS828. Bioorg. Med. Chem. Lett. 2005, 15,
2491−2494.
(22) Schou, C.; Ottosen, E. R.; Petersen, H. J.; Bjorkling, F.; Latini,
̈
́
(36) Beauparlant, P.; Bedard, D.; Bernier, C.; Chan, H.; Gilbert, K.;
S.; Hjarnaa, P. V.; Bramm, E.; Binderup, L. Novel cyanoguanidines
with potent oral antitumour activity. Bioorg. Med. Chem. Lett. 1997, 7
(24), 3095−3100.
Goulet, D.; Gratton, M.-O.; Lavoie, M.; Roulston, A.; Turcotte, E.;
Watson, M. Preclinical development of the nicotinamide phosphor-
ibosyl transferase inhibitor prodrug GMX1777. Anti-Cancer Drugs
2009, 20, 346−354.
(23) Colombano, G.; Travelli, T.; Galli, U.; Caldarelli, A.; Chini, M.
G.; Canonico, P. L.; Sorba, G.; Bifulco, G.; Tron, G. C.; Genazzani, A.
A. A novel potent nicotinamide phosphoribosyltransferase inhibitor
synthesized via click chemistry. J. Med. Chem. 2010, 53, 616−623.
(24) Fleischer, T. C.; Murphy, B. R.; Flick, J. S.; Terry-Lorenzo, R.
T.; Gao, Z.-H.; Davis, T.; McKinnon, R.; Ostanin, K.; Willardsen, J. A.;
Boniface, J. J. Chemical proteomics identifies Nampt as the target of
CB30865, an orphan cytotoxic compound. Chem. Biol. 2010, 17, 659−
664.
(37) Skelton, L. A.; Ormerod, M. G.; Titley, J.; Kimbell, R.; Brunton,
L. A.; Jackman, A. L. A novel class of lipophilic quinazoline-based folic
acid analogues: Cytotoxic agents with a folate-independent locus. Br. J.
Cancer 1999, 79 (11/12), 1692−1701.
(38) Orwig, K. S.; Dix, T. A. Synthesis of Cα methylated carboxylic
acids: isosteres of arginine and lysine for use as N-terminal capping
residues in polypeptides. Tetrahedron Lett. 2005, 46 (41), 7007−7009.
(39) Butera, J. A.; Antane, M. M.; Antane, S. A.; Argentieri, T. M.;
Freeden, C.; Graceffa, R. F.; Hirth, B. H.; Jenkins, D.; Lennox, J. R.;
Matelan, E.; Norton, N. W.; Quagliato, D.; Sheldon, J. H.; Spinell, W.;
Warga, D.; Wojdan, A.; Woods, M. Design and SAR of novel
potassium channel openers targeted for urge urinary incontinence. 1.
N-cyanoguanidine bioisosteres possessing in vivo bladder selectivity. J.
Med. Chem. 2000, 43, 1187−1202.
(25) Carlson, R. O.; Willardsen, J. A.; Lockman, J. W.; Bradford, C.
L.; Patton, J. S.; Papac, D. I.; Boniface, J. J.; Yager, K.; Baichwal, V. R.
Pharmacokinetics, anti-tumor activity and therapeutic index of Nampt
inhibitor MPC-8640 in mice. In Mol. Cancer Ther., Proceedings of the
AACR-NCI-EORTC International Conference: Molecular Targets and
Cancer Therapeutics, San Francisco, CA, November 12−16, 2011;
AACR: Philadelphia, PA, 2011; Vol. 10, (11 Suppl.), Abstract nr B137.
(26) Cea, M.; Zoppoli, G.; Bruzzone, S.; Fruscione, F.; Moran, E.;
Garuti, A.; Rocco, I.; Cirmena, G.; Casciaro, S.; Olcese, F.; Pierri, I.;
Cagnetta, A.; Ferrando, F.; Ghio, R.; Gobbi, M.; Ballestrero, A.;
(40) Humljan, J.; Gobec, S. Synthesis of N-phthalimido β-
aminoethanesulfonyl chlorides: The use of thionyl chloride for a
simple and efficient synthesis of new peptidosulfonamide building
blocks. Tetrahedron Lett. 2005, 43, 4069−4072.
9087
dx.doi.org/10.1021/jm4009949 | J. Med. Chem. 2013, 56, 9071−9088

Journal of Medicinal Chemistry
Article
(41) Kim, M. K.; Lee, J. H.; Kim, H.; Park, S. J.; Kim, S. H.; Kang, G.
B.; Lee, Y. S.; Kim, J. B.; Kim, K. K.; Suh, S. W.; Eom, S. H. Crystal
structure of visfatin/pre-B cell colony-enhancing factor 1/nicotinamide
phosphorybosyltransferase, free and in complex with the anti-cancer
agent FK-866. J. Mol. Biol. 2006, 362, 66−77.
(42) Olesen, U. H.; Petersen, J. G.; Garten, A.; Kiess, W.; Yoshino, J.;
Imai, S.; Christensen, M. K.; Fristrup, P.; Thougaard, A. V.; Bjorkling,
̈
F.; Jensen, P. B.; Nielsen, S. J.; Sehested, M. Target enzyme mutations
are the molecular basis for resistance towards pharmacological
inhibition of nicotinamide phosphoribosyltransferase. BMC Cancer.
2010, 10, 677.
(43) Murthi, K. K.; Kostler, R.; Smith, C.; Brandstetter, T.; Kluge, A.
̈
F. Derivatives of squaric acid with anti-proliferative activity. EU Pat. Appl.
EP 1 674 457 A1, 2004.
(44) Linderoth, L.; Fristrup, P.; Hansen, M.; Melander, F.; Madsen,
R.; Andresen, T. L.; Peters, G. H. Mechanistic study of the sPLA2-
mediated hydrolysis of a thio-ester pro anticancer ether lipid. J. Am.
Chem. Soc. 2009, 131, 12193−12200.
(45) Kornet, M. J. Potential antifibrillatory agents. N-(ω-amino-
alkoxy) phthalimides. J. Med. Chem. 1966, 9, 269.
(46) Deraeve, C.; Bon, R. S.; Stigter, E. A.; Wetzel, S.; Waldmann, H.;
Guo, Z.; Blankenfeldt, W.; Alexandrov, K.; Goody, R. S.; Wu, Y. W.;
DiLucrezia, R.; Wolf, A.; Menninger, S.; Choidas, A. Psoromic acid is a
selective and covalent rab-prenylation inhibitor targeting autoinhibited
RabGGTase. J. Am. Chem. Soc. 2012, 134, 7384−7391.
(47) Boehringer Ingelheim International, GMBH, Patent: WO2009/
112565 A1, 2009.
(48) Zhao, X.; Jiang, X. K.; Shi, M.; Yu, Y. H.; Xia, W.; Li, Z. T. Self-
assembly of novel [3]- and [2]-rotaxanes with two different ring
components: Donor−acceptor and hydrogen bonding interactions and
molecular-shuttling behavior. J. Org. Chem. 2001, 66 (21), 7035−7043.
(49) Natarajan, A.; Fan, Y.-H.; Chen, H.; Guo, Y.; Iyasere, J.;
Harbinski, F.; Christ, W. J.; Aktas, H.; Halperin, J. A. 3,3-Diaryl-1,3-
dihydroindol-2-ones as antiproliferatives mediated by translation
initiation inhibition. J. Med. Chem. 2004, 47, 1882−1885.
(50) Elliott, G. C.; Ajioka, J.; Okada, C. Y. A rapid procedure for
assaying nicotinamide phosphoribosyl-transferase. Anal. Biochem.
1980, 107, 199−205.
(51) Garten, A.; Petzold, S.; Barnikol-Oettler, A.; Korner, A.; Thasler,
̈
W. E.; Kratzsch, J.; Kiess, W.; Gebhardt, R. Nicotinamide
phosphoribosyltransferase (NAMPT/PBEF/visfatin) is constitutively
released from human hepatocytes. Biochem. Biophys. Res. Commun.
2010, 391, 376−381.
(52) Jaguar, version 8.0 release 47; Schrodinger, LLC: New York,
2013.
(53) Becke, A. D. Density-functional thermochemistry. III. The role
of exact exchange. J. Chem. Phys. 1993, 98, 5648−5652.
(54) Becke, A. D. A new mixing of Hartree−Fock and local density-
functional theories. J. Chem. Phys. 1993, 98, 1372−1377.
(55) Lee, C.; Yang, W.; Parr, R. G. Development of the Colle-Salvetti
correlation-energy formula into a functional of the electron density.
Phys. Rev. B 1988, 37, 785−789.
(56) Elstner, M.; Hobza, P.; Frauenheim, T.; Suhai, S.; Kaxiras, E.
Hydrogen bonding and stacking interactions of nucleic acid base pairs:
A density-functional-theory based treatment. J. Chem. Phys. 2001, 114,
5149−5155.
(57) Grimme, S. Accurate description of van der Waals complexes by
density functional theory including empirical corrections. J. Comput.
Chem. 2004, 25, 1463−1473.
(58) Ditchfield, R.; Hehre, W. J.; Pople, J. A. Self-consistent
molecular-orbital methods. IX. An extended Gaussian-type basis for
molecular-orbital studies of organic molecules. J. Chem. Phys. 1971, 54,
724−728.
9088
dx.doi.org/10.1021/jm4009949 | J. Med. Chem. 2013, 56, 9071−9088