Potent and selective inhibitors of Helicobacter pylori glutamate racemase (MurI): Pyridodiazepine amines

Article abstract of DOI:10.1016/j.bmcl.2008.11.113

An SAR study of an HTS screening hit generated a series of pyridodiazepine amines as potent inhibitors of Helicobacter pylori glutamate racemase (MurI) showing highly selective anti-H. pylori activity, marked improved solubility, and reduced plasma protei

Full text of DOI:10.1016/j.bmcl.2008.11.113

Bioorganic & Medicinal Chemistry Letters 19 (2009) 930–936
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Potent and selective inhibitors of Helicobacter pylori glutamate racemase
(MurI): Pyridodiazepine amines
a
a
a
a
Bolin Geng a, , Gregory Basarab , Janelle Comita-Prevoir , Madhusudhan Gowravaram , Pamela Hill ,
Andrew Kiely a, James Loch a, Lawrence MacPherson a, Marshall Morningstar a, George Mullen a,
Ekundayo Osimboni a, Alexander Satz a, Charles Eyermann a, Tomas Lundqvist b
*
a AstraZeneca R&D Boston, Infection Discovery, 35 Gatehouse Drive, Waltham, MA 02451, USA
b AstraZeneca R&D Mölndal, Global Structural Chemistry, Mölndal, SE-43183, Sweden
a r t i c l e i n f o
a b s t r a c t
Article history:
An SAR study of an HTS screening hit generated a series of pyridodiazepine amines as potent inhibitors of
Helicobacter pylori glutamate racemase (MurI) showing highly selective anti-H. pylori activity, marked
improved solubility, and reduced plasma protein binding. X-ray co-crystal E–I structures were obtained.
These uncompetitive inhibitors bind at the MurI dimer interface.
Received 26 August 2008
Revised 26 November 2008
Accepted 26 November 2008
Available online 6 December 2008
Ó 2008 Elsevier Ltd. All rights reserved.
Keywords:
Glutamate racemase
Helicobacter pylori
Pyridodiazepine
MurI inhibitor
It is well documented that Helicobacter pylori infection can
cause chronic peptic ulceration and can increase the risk of gastric
adenocarcinoma.1 Novel anti-H. pylori agents with a new mode-of-
action (MOA) would be highly beneficial for the patient commu-
nity, as current therapies combining two antibiotics with a proton
pump inhibitor (Triple Therapy) have numerous drawbacks, most
notably broad-spectrum antibacterial activity leading to the dis-
ruption of commensal bacteria in the gastrointestinal (GI) tract
and poor patient compliance.2 Additionally, ever emerging resis-
tant H. pylori strains have diminished the effectiveness of the com-
bination therapies. Toward meeting this medical need, we chose a
novel enzyme target for the design of an anti-H. pylori drug,
namely that of glutamate racemase (MurI).3 MurI is a bacterial
the chances of disturbing off target commensal GI flora and dimin-
ish resistance development in other bacterial species as has been
seen with the current therapies.
Previously, we reported potent and selective pyrazolopyrimid-
inedione H. pylori MurI inhibitors that were originally identified
via high-throughput screening (HTS) of the former Astra com-
pound collection.3,7 The inhibitors additionally prevented H. pylori
growth, but were not suitable for progression into animal efficacy
experiments, as high plasma protein binding (PPB) compromised
their capability to control disease in vivo.3a
Herein we report a second scaffold represented by the novel
benzodiazepine amine 2 (Fig. 1) discovered by an HTS campaign
cytoplasmic enzyme that catalyzes the conversion of
L-glutamate
to -glutamate, one of the essential amino acids in peptidoglycan
D
H
H
synthesis.4 The disruption of peptidoglycan biosynthesis is lethal
to bacteria and therefore inhibitors of glutamate racemase should
be useful as antibacterials. The murI gene is conserved in all bacte-
rial species5,6 that synthesize peptidoglycan and its essentiality has
been well-demonstrated in a number of bacteria.7 The unique bio-
physical and biochemical properties of H. pylori MurI relative to the
MurI of other bacteria could allow for the discovery and develop-
ment of specific MurI inhibitors.7 Such specificity would reduce
9
2
N
CH3
N
CH3
N
N
N
8
7
3
S
S
N
N
6
5
O
2
15
Lead
Hit
* Corresponding author. Tel.: +1 781 839 4297; fax: +1 781 839 4230.
E-mail address: bolin.geng@astrazeneca.com (B. Geng).
Figure 1. HTS hit 2 and lead 15.
0960-894X/$ - see front matter Ó 2008 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2008.11.113

B. Geng et al. / Bioorg. Med. Chem. Lett. 19 (2009) 930–936
931
O
NH2
H
N
O
HO
S
a
O
O
NHBoc
+
S
Boc NH
H
O
H
N
N CH3
c, d
b
N
S
N
S
N
2
Scheme 1. Reagents and conditions: (a) DCC, CH2Cl2; (b) HBr, CH3CO2H and NH4OAc; (c) Lawesson’s reagent, dioxane; (d) MeNH2, HgCl2, THF.
of the former Zeneca compound collection. Compound 2 is highly
selective against MurI of H. pylori, with an IC50 of 1.7 M, versus
isozymes from three other bacteria species: Enterococcus faecalis,
Escherichia coli and Staphylococcus aureus, with IC50s > 400 M. This
in this letter. Furthermore, we also observed corresponding micro-
biological selectivity. Compound 2 showed an MIC of 0.5 g/mL
against H. pylori, but demonstrated MICs > 64 g/mL against a
l
l
l
l
broad panel of bacteria that included the three bacteria species
selectivity was maintained throughout the analog series described
above from which the MurI isozymes had been isolated.7
Table 1
Effect of R2 variations (Core = benzene, R3 = 20-thienyl, R5 = phenyl)
R2
N
S
N
a
Compound
R2
IC50
400
1.7
1.1
4.0
1.1
(lM)
MICb
32
(l
g/mL)
Soluc
0.6
0.5
4.0
NT
(
lM)
PPBd
98.0
99.7
99.7
NT
Cl inte
Human
(l
L/mL/kg)
Mouse
NH2
1
2
3
4
5
6
7
>100
>100
97
Me
Et
0.5
2
43
HN
HN
HN
46
>100
>100
74
n-Pr
16
30
8
NT
97.0
NT
50
HN
HN
O
Me
18.9
0.5
32
NT
80
>100
>100
OH
0.25
0.3
99.0
49
N
N
8
4.0
8
NT
NT
43
87
a
MurI protein was expressed from murI gene of chromosomal DNA of wild-type Helicobacter pylori strain J 99 (human isolate, Tenn. Genome Therapeutic Corporation, now
Oscient Pharmaceuticals, Inc.) IC50 is the mean of duplicate determinations.
b
Helicobacter pylori strain SS1, original Sydney strain from U.S.W. Australia; MICs were measured according to NCCLS guidelines and repeat MICs were generally within a
twofold dilution range.
c
Equilibrium solubility; NT, not tested.
Experimental PPB value; NT, not tested.
Liver microsomal clearance.
d
e

932
B. Geng et al. / Bioorg. Med. Chem. Lett. 19 (2009) 930–936
Substitution of the aromatic groups at C-3 of the benzodiazepine
the corresponding ketone. Subsequent amination, amide coupling,
Boc-deprotection, and cyclization were followed by conversion to
the amidine. The conversion to the amidine from pyridodiazepinone
intermediate can also be performed in high yield with TiCl4 and
methylamine in a microwave reactor.9,10 The Boc-protected amino
acids used in this program are generally known in the literature
and are easily prepared according to literature procedures.
Structures of all final products were established by both NMR
and mass spectroscopy techniques. The structures of a number of
compounds were also confirmed by small molecule X-ray crystal-
lography (data not shown).
scaffold represents a novel chemotype relative to the plethora of
benzodiazepines emanating substantially from CNS research.15
Another novel feature is the embedded amidine functionality. In
an effort to optimize this hit to a lead, we set out to improve solubil-
ity (<0.6 lM for 2) and reduce plasma protein binding (PPB > 99% for
2) deemed necessary for developing a viable drug for oral adminis-
tration. Within this context we would need to retain the good
enzyme inhibitory potency and antibacterial activity of 2.
Here, we describe the structure–activity relationships (SAR)
that led to the discovery and elaboration of a novel 9-aza lead ser-
ies of potent and selective anti-H. pylori compounds as exemplified
by Compound 15 (Fig. 1). The incorporation of a nitrogen atom into
the benzodiazepine scaffold, forming a pyridodiazepine, adds
another distinguishing feature relative to the vast knowledge base
in the literature for this scaffold.
An X-ray structure of H. pylori MurI in a ternary complex with 2
and D
-glutamate was obtained16 demonstrating a distinctive non-
competitive binding interaction at the enzyme dimer interface
(Picture 1 left). Previously, the X-ray structure of a ternary MurI,
D-glutamate, pyrazolopyrimidinedione inhibitor complex was
The synthesis of HTS hit 2 is shown in Scheme 1. The DCC amide
coupling product of amino benzophenone with the Boc-protected
thiophene amino acid was deprotected and cyclized to yield a ben-
zodiazepinone intermediate that was converted to corresponding
thioamide using Lawesson’s reagent. Conversion to the amidine
was achieved with mercury chloride and methylamine in THF. This
sequence was applied analogously for the synthesis of the other
benzodiazepine amines 1 and 3–9 (Tables 1 and 2).
Thesynthesisof pyridodiazepineanalogsis exemplifiedby thatof
14 in Scheme 2. Lithiation of 2-fluoropyridine at the 3-position with
LDA was followed by reaction with 2-thiophene carboxaldehyde.
The resulting alcohol was oxidized with manganese dioxide8 to yield
described.7 However, though MurI requires glutamate for binding
both the benzodiazepine and pyrazolopyrimidinedione classes of
inhibitors, the binding sites for the two are distinct from one
another. Unlike other Gram-negative MurI, H. pylori MurI exists
in a homodimer, more similar to Gram-positive MurI isozymes.
However, the dimer interface in the Gram-positive isozymes is sit-
uated backside to the glutamate binding site while for H. pylori
MurI it abuts the substrate glutamate.7 The rather unusual inhibi-
tor-binding site might not have been recognized a priori, but rather
required the identification of an inhibitor through HTS. To our
knowledge, these are the first of such unique binding interactions
of MurI/inhibitor. This result once again demonstrated the power
Table 2
Effect of core variations (R2 = methylamino, R3 = 20-thienyl, R5 = phenyl)
NNMe
S
N
Core
N
a
Compound
Core
IC50
(
lM)
MICb
(lg/mL)
Soluc
NT
(lM)
PPBd
NT
Cl inte
Human
(l
L/mL/kg)
Mouse
9
13.0
2.2
8
52
>100
82
Cl
N
10
11
0.17
98
95.9
64.7
40
O
N+
67.0
16
NT
<14
41
N
Me
12
1.7
50
0.25
70
96.1
84.2
73
36
>100
26
N
13
64
NT
N
a
MurI protein was expressed from murI gene of chromosomal DNA of wild-type Helicobacter pylori strain J 99 (human isolate, Tenn. Genome Therapeutic Corporation, now
Oscient Pharmaceuticals, Inc.) IC50 is the mean of duplicate determinations.
b
Helicobacter pylori strain SS1, original Sydney strain from U.S.W. Australia; MICs were measured according to NCCLS guidelines and repeat MICs were generally within a
twofold dilution range.
c
Equilibrium solubility; NT, not tested.
Experimental PPB value; NT, not tested.
Liver microsomal clearance.
d
e

B. Geng et al. / Bioorg. Med. Chem. Lett. 19 (2009) 930–936
933
N
NH2
O
N
F
N
F
c
a
O
S
b
OH
S
N
F
S
H
N CH3
H
O
N
N
N
N
d, e
f, g
S
S
N
N
or h
S
S
14
Scheme 2. Reagents and conditions: (a) LDA, THF, ꢀ78 °C, 2-thiophene carboxaldehyde; (b) MnO2, THF reflux; (c) 7 M NH3, methanol, in sealed tube; d, e, f and g are same as
a, b, c and d in Scheme 1; (h) TiCl4/MeNH2 in CH2Cl2.
of HTS in discovering inhibitors that otherwise would be difficult
to design.
at room temperature for compound 2 vs 11% for compound 10 over
same period), and the racemization may well be faster under phys-
iological conditions. Introduction of a second substitution at C3
was attempted, to block the racemization,14 without success.
Addition of a single chlorine atom to the 8-position of 2
(compound 9) (Table 2) is not well tolerated, in contrast to the
established SAR15 trends for CNS-active benzodiazepines where
8-substitution offers improvement. By way of contrast, a single
methyl substitution in the analogous 7-position (exemplified in
pyridodiazepine 12 relative to 10) is well tolerated and gives an
improvement in MurI activity. Selected benzodiazepines and pyr-
idodiazepines herein did not inhibit GABA receptors.
To improve solubility, a nitrogen atom was incorporated into
the core scaffold affording a pyridodiazepine amine (compound
10) with a reduced LogD value (2.7 vs 4.1 of compound 2). MurI
inhibitory potency and antibacterial activity were retained. More
importantly, the solubility and PPB characteristics are markedly
improved (compare 2 with 10). An X-ray co-crystal structure of
10 with MurI shows a water molecule bridging the pyridine
The SAR around 2 was first explored through selective variation
of substituents around the diazepine scaffold and benzene core as
allowed by the constraints of the inhibitor-binding pocket delin-
eated in the X-ray structure. As shown in Table 1 (compounds 1–
8), variations of the 2-position amine (R2) did not produce any
compounds that markedly improved solubility or decreased PPB.
For inhibitory potency, only small alkylamines were tolerated, with
methylamino remaining the best. Inhibitor potency and antibacte-
rial activity were retained with the hydroxy azetidine substituent
in compound 7 in an attempt to build an H-bonding interaction.
An X-ray co-crystal structure of 7 with MurI revealed H-bonding
interactions between the azetine hydroxyl and both Ser143 and
Glu151 (Picture 1, right).16 Initial SAR activity at R2 aiming to
address microsomal stability issues of the series, was unsuccessful
(data not shown).
It should be noted that compounds 1–25 are all racemic. The
R- and S-isomers of 2 were separated by chiral chromatography
with the greater MurI enzyme inhibitory potency residing in the
one enantiomer (sevenfold difference). However, the two optical
antipodes showed identical MICs. For compound 10, the same
trend was observed with a 42-fold difference in IC50 and only four-
fold difference in MIC. We speculated this observation might be the
result of racemization and it was shown that racemization did
occur slowly (28% racemization over a 3-week period in methanol
nitrogen to bound
D-Glu product in neighboring active site.
Hence, the N-oxide analog (11) was considerably less potent as
an inhibitor, disrupting the water-to-glutamate bridge. The 8-
Me substituted analog (12), in which the methyl may increase
the hydrogen bonding capability of 9-aza nitrogen lone pair,
was slightly more active compared with 10. Interestingly, intro-
duction of two nitrogen atoms (13), which was expected to in-
Picture 1. (Left) Co-crystal structure shows that compound 2 (purple) binds at MurI enzyme (green) dimer interface. Two D-Glu products are shown at the active sites (Red,
white, blue balls). (right): Weak H-bonds (yellow dotted lines) between the hydroxyl group of compound 7 and both Serine-143 and Glutamate-151 of MurI enzyme. The
discrete red balls are water molecules.

934
B. Geng et al. / Bioorg. Med. Chem. Lett. 19 (2009) 930–936
Table 3
Effect of R5 variations (Core = pyridine, R2 = methylamino, R3 = 20-thienyl)
NHMe
S
N
N
N
R5
Compound
R5
IC50
(lM)
MICb
(
l
g/mL)
Soluc
(
lM)
PPBd
Cl inte
Human
(l
L/mL/kg)
Mouse
a
14
15
16
17
18
19
0.7
2.0
1.7
7.7
0.6
4.1
1.1
0.25
0.19
1.0
227
1365
39
97.0
81.8
83.0
74.3
75.4
95.2
75.9
30
13
<14
<5
19
59
18
>100
30
S
O
N
33
S
N
1.0
211
1730
16
19
O
HN
0.40
8
36
>100
37
20
0.32
176
N
S
a
MurI protein was expressed from murI gene of chromosomal DNA of wild-type Helicobacter pylori strain J 99 (human isolate, Tenn. Genome Therapeutic Corporation, now
Oscient Pharmaceuticals, Inc.) IC50 is the mean of duplicate determinations.
b
Helicobacter pylori strain SS1, original Sydney strain from U.S.W. Australia; MICs were measured according to NCCLS guidelines and repeat MICs were generally within a
twofold dilution range.
c
Equilibrium solubility; NT, not tested.
Experimental PPB value; NT, not tested.
Liver microsomal clearance.
d
e
crease solubility and reduce PPB, diminished both MurI enzy-
matic and antibacterial activity.
antibacterial potency losses. This was mirrored by the early SAR
studies at this position in the original hit series where core = ben-
zene, R5 = phenyl, R2 = MeNH, R3 = alkyls and aromatic substituted
alkyls (data not shown).
Variation of the substituent in the 5-position (R5) with 5- and 6-
membered aromatic rings generally retains inhibitory potency and
antibacterial activity (see Table 3, compounds 14–20). Both cylic
(19) and acyclic (data not shown) alkyl substituents at R5 dimin-
ished inhibitory potency considerably. This may suggest that this
In general, a good correlation between MurI enzyme inhibition
(IC50) and whole-cell growth inhibition (MIC) was observed. In
contrast to the previously reported pyrazolopyrimidinedione ser-
ies,3 the current series was not affected by efflux pumps (such as
hefBꢀ), therefore an excellent whole cell activity was observed in
spite of moderate MurI enzyme potency. Studies to evaluate emer-
gence of resistance demonstrated the series has a moderate ten-
dency of spontaneous resistance generation (resistance rate
<10ꢀ7). Mode-of-action (MOA) studies using both a MurI-over-
expressing strain and strains with point mutations in MurI con-
firmed the whole cell activity was the result of inhibition of MurI
enzyme in the cell. The cidal killing kinetic was comparable to
the b-lactam Amoxicillin.17
In summary, we discovered a novel series of uncompetitive
MurI inhibitors with potent and selective anti-H. pylori activities
and a unique binding interaction. A privileged pyridodiazepine
scaffold provided a drugable lead that could be further developed
into a viable mono-therapeutic agent. Combination of the pyridine
core with 5-membered aromatic R5 heterocycles improved physi-
cal properties while retaining enzymatic and antibacterial potency
and selectivity. A set of compounds with reasonably low protein
binding and improved solubility (15, 18 and 20) were identified.
site favors a flat electron-rich
p
system.11,12 Five-membered aro-
matic heterocycles oftentimes improved solubility and increased
the free fraction (compare 10 with 15).12 There was also a general
improvement in microsomal stability in moving from benzene to
heterocyclic R5 subsituents.
The substitution at 3-position (R3) (see Table 4, compounds 21–
25) tolerated a variety of 5-membered aromatic rings. Phenyl and
pyridyl rings are less active. This region of the R3 binding site is
quite constrained and can only tolerate small substituents except
where there is a 1,3-relationship to the azepine ring bond connec-
tion, such as 40-position substituted thiophene. The region here
leads to a hydrophilic water channel that further leads to solvent.
Larger substituents occupying this region (as in 22) are tolerated
but less active. Overall, considering MICs, solubility and PPB, the
unsubstituted 2-thiophene group at R3 was viewed to be preferred.
However, thiophenes can often be metabolic and safety liabili-
ties.13 An extensive effort of searching for an alternative to the
direct C3-linked 20-thiophene group was conducted, unfortunately,
in most cases, the modification resulted in both enzymatic and

B. Geng et al. / Bioorg. Med. Chem. Lett. 19 (2009) 930–936
935
Table 4
Effect of R3 variations (Core = pyridine, R2 = methylamino, R5 = 20-furyl)
NHMe
N
N
R3
N
O
R3
IC50
(lM)
MICb
0.5
(lg/mL)
Soluc
191
(lM)
PPBd
98.7
Cl inte
Human
(lL/mL/kg)
a
Compound
Mouse
Cl
21
2.8
74
>150
S
S
OH
22
4.9
4.0
201
196
90.9
7
17
N
23
24
8.6
8.7
4.0
2.0
8.0
60.9
73.6
80.7
22
26
9
32
27
19
S
>1000
192
N
S
25
18.5
O
a
MurI protein was expressed from murI gene of chromosomal DNA of wild-type Helicobacter pylori strain J 99 (human isolate, Tenn. Genome Therapeutic Corporation, now
Oscient Pharmaceuticals, Inc.) IC50 is the mean of duplicate determinations.
b
Helicobacter pylori strain SS1, original Sydney strain from U.S.W. Australia; MICs were measured according to NCCLS guidelines and repeat MICs were generally within a
twofold dilution range.
c
Equilibrium solubility; NT, not tested.
Experimental PPB value; NT, not tested.
Liver microsomal clearance.
d
e
Parr, T. R.; Smith, M. C. J. Med. Chem. 2002, 45, 4559; (g) Kim, W.-C.; Rhee, H.-I.;
Park, B.-K.; Suk, K.-H.; Cha, S.-H. J. Biomol. Screen. 2000, 5, 435; For most recent
allosteric inhibitors, see: (h) Geng, B.; Breault, G.; Comita-Prevoir, J.; Petrichko,
R.; Eyermann, C.; Lundqvist, T.; Doig, P.; Gorseth, E.; Noonan, B. Bioorg. Med.
Improvement in both in vitro and in vivo pharmacokinetic profiles
is necessary to optimize the lead toward the design of a viable drug
for human utility.18
Chem. Lett. 2008, 18, 4368; (i) Breault, G.; Comita-Prevoir, J.; Eyermann, C.;
Geng, B.; Petrichko, R.; Doig, P.; Gorseth, E.; Noonan, B. Bioorg. Med. Chem. Lett.
2008, 18, 6100.
Acknowledgments
4. Van Heijenoort, J. J. Nat. Prod. Rep. 2001, 18, 503.
The authors thank Nancy DeGrace, Sharon Tentarelli and Mark
Zambrowski, AstraZeneca R&D Boston, for chiral separation
support. Dr. Birgitta Stensland, AstraZeneca R&D Södertälje,
Sweden for small molecule X-ray analyses. We also wish to thank
Department of Biochemistry for IC50 determinations, Department
of Microbiology for MIC determinations and Department of DMPK
for solubility, PPB and Clearance determinations.
5. (a) Doublet, P.; van Heijenoort, J.; Mengin-Lecreulx, D. J. Bacteriol. 1992, 174,
5772; (b) Glavas, S.; Tanner, M. E. Biochemistry 2001, 40, 6199.
6. Doublet, P.; van Heijenoort, J.; Mengin-Lecreulx, D. Biochemistry 1994, 33, 5285.
7. Lundqvist, T.; Fisher, S. L.; Kern, G.; Folmer, R. H. A.; Xue, Y.; Newton, D. T.;
Keating, T. A.; Alm, R.; de Jonge, B. L. M. Nature 2007, 447, 817.
8. Gungor, T.; Marsais, F.; Queguiner, G. J. Organometal. Chem. 1981, 215, 139.
9. Fryer, R. I.; Early, J. V.; Field, G. F.; Zally, W.; Sternbach, L. H. J. Org. Chem. 1969, 34, 1143.
10. Typical synthesis of pyridodiazepine Amine (14).Step a: (2-Fluoropyridin-3-
yl)(2-thienyl)methanol To a solution of 2-fluoropyridine (2.0 g, 20.6 mmol) in
anhydrous THF (5 mL) was added a solution of 2 M LDA (heptane/THF, 12 mL,
24 mmol) at ꢀ78 °C under nitrogen. An orange suspension was formed. To the
above suspension was added a solution of 2-thiophene carboxaldehyde in THF
(5 mL). The reaction was stirred at ꢀ78 °C for 2 h and was allowed to warm to
room temperature overnight. The reaction was worked up as follows: the
reaction mixture was poured into an ice water mixture and was extracted with
ethyl acetate (3 ꢁ 100 mL). The combined organic layers were dried over
magnesium sulfate and purified using Flash Master silica-gel chromatography
with ethyl acetate/hexane as eluent. The product was dried under reduced
pressure to yield a yellow oil (2.6 g, yield, 63%). MS (ES+) 210 (MH+) for
C10H8FN OS.Step b: (2-Fluoropyridin-3-yl)(2-thienyl)methanone S(2-
fluoropyridin-3-yl)(2-thienyl)methanol (2.6 g, 12.3 mmol) was refluxed with
manganese (IV) oxide (3.0 g, 33.2 mmol) in THF (14 mL) for 48 h. The reaction
mixture was filtered and the product was purified by silica-gel
chromatography using ethyl acetate/hexane as eluant. The desired fractions
were collected, concentrated and dried under reduced pressure to give
(2-fluoropyridin-3-yl)(2-thienyl)methanone as a yellow solid (2.25 g, yield,
References and notes
1. Uemura, N.; Okamoto, S.; Yamamoto, S.; Matsumura, N.; Yamaguchi, S.;
Yamakido, M.; Taniyama, K.; Sasaki, N.; Schlemper, R. J. N Engl. J. Med. 2001,
345, 784.
2. Suebaum, S.; Michetti, P. N. Engl. J. Med. 2002, 347, 1175.
3. (a) Basarab, G.; Hill, P. J.; Rastagar, A.; Webborn, P. J. Bioorg. Med. Chem. Lett.
2008, 18, 4716; (b) Gowravaram, M.; Basarab, G.; Eyermann, C.; Green, O.;
MacPherson, L.; Osimoni, E. 45th ICAAC 2005. Abstracts F-1877; (c) Basarab, G.;
Eyermann, C.; Gowravaram, M.; Green, O.; Hill, P.; Macpherson, L.;
Morningstar, M.; Osimoni, E.; Rastagar, A. 45th ICAAC 2005. Abstracts F-1879;
For competitive inhibitors, see: (d) Glavas, S.; Tanner, M. E. Bioorg. Med. Chem.
Lett. 1997, 7, 2265; (e) Tanner, M. E.; Miao, S. Tetrahedron Lett. 1994, 35, 4073;
(f) De Dios, A.; Priesto, L.; Martin, J. A.; Rubio, A.; Ezquerra, J.; Tebbe, M.; Lopez
de Uralde, B.; Martin, J.; Sanchez, A.; LeTourneau, D.; McGee, J. E.; Boylan, C.;

936
B. Geng et al. / Bioorg. Med. Chem. Lett. 19 (2009) 930–936
87%). MS (ES+) 208 (MH+) for C10H6FN OS. Step c: (2-aminopyridin-3-yl)(2-
thienyl)methanone (2-fluoropyridin-3-yl)(2-thienyl)methanone (2.25 g,
were dried under reduced pressure to give a yellow solid (0.19 g, yield, 38%).
MS (ES+) 342 (MH+) for C16H11N3S3. Step g: N-Methyl-3,5-di-2-thienyl-3H-
pyrido[2,3-e][1,4]diazepin-2-amine (compound 14) 3,5-di-2-thienyl-1,
3-dihydro-2H-pyrido[2,3-e][1,4]diazepine-2-thione (0.19 g, 0.557 mmol) was
dissolved in a 2 M methylamine solution (10 mL, 20 mmol) in THF. The mixture
was heated to 60 °C for 15 min. The reaction mixture was concentrated and
purified on Flash Master silica-gel chromatography using ethyl acetate/hexane
as eluant and dried under reduced pressure to give a white solid (55 mg, yield,
31%) 1H NMR (300 MHz, CHLOROFORM-d) d (ppm): 2.94 (d, J = 4.58 Hz, 3H);
4.92(s, 1H); 5.13 (s, br, 1H); 7.0–7.2 (m, 5H); 7.47 (dd, J = 13.6, 4.54 Hz, 2H);
8.05 (d, J = 7.59 Hz, 1H); 8.71 (d, J = 3.70 Hz, 1H). MS (ES+) 339 (MH+) for
C17H14N4S2.For more experimental details of other compounds described in
this letter, see: Basarab, G.; Eyermann, C.; Geng, B.; Gowravaram, M.; Loch, J.;
MacPherson, L.; Morningstar, M.; Mullen, G.; Satz, A.; Kiely, A. PCT Int. Appl.
WO200511101, 2005.
10.87 mmol) was treated with ammonia methanol solution (7 N) with
moderate heating (or conducted in a bomb reaction at 40 °C). After heating
overnight, the reaction mixture was concentrated and purified on Flash Master
silica-gel chromatography using ethyl acetate/hexane as eluant. The desired
fractions were collected, concentrated and dried under reduced pressure to
give (2-aminopyridin-3-yl)(2-thienyl)methanone as
yield, 78%). MS (ES+) 205 (MH+) for C10H8N2 OS. Step d and e: 3,5-Di-2-thienyl-
1, 3-dihydro-2H-pyrido[2,3-e][1,4]diazepin-2-one to solution of (2-
aminopyridin-3-yl)(2-thienyl)methanone (0.88 g, 4.31 mmol) and N-Boc
amino (2-thienyl)acetic acid (2.37 g, 9.22 mmol) in anhydrous
a yellow solid (1.73 g,
a
dichloromethane (15 mL) was added a solution of DCC (2.97 g, 8 mmol) in
anhydrous dichloromethane (10 mL) at 0 °C. The reaction mixture was stirred
at 0 °C for 3 h and was allowed to warm to room temperature overnight. The
suspension was filtered and the filtrate was concentrated and purified on Flash
master silica-gel chromatography using ethyl acetate/hexane as eluant
yielding a white solid.The coupling product was then treated with a TFA
solution (30 mL, 30%) in dichloromethane at 0 °C for 1 h and continued stirring
for 1 h at room temperature. After deprotection was complete (monitored by
LC-MS) the reaction mixture was concentrated to remove all solvent and TFA.
The oil like deprotected product was then dissolved with 30 mL of acetic acid
and mixed with ammonium acetate solid. The suspension was heated to 60 °C
for 12 h. After removing the acetic acid, product 3,5-di-2-thienyl-1, 3-dihydro-
2H-pyrido[2,3-e][1,4]diazepin-2-one was purified on Flash master silica-gel
chromatography using ethyl acetate/hexane as eluant and dried under reduced
pressure to give a brown solid (0.48 g, yield, 69%) 1H NMR (300 MHz, DMSO-
d6) d (ppm): 5.55 (d, 1 H); 7.15–7.70 (m, 4H); 7.37 (m, 1H); 7.52 (d, 1H); 7.80
(d, 1H); 8.20 (d, 1H); 8.60 (d, 1H); 11.19 (s, 1H). MS (ES+) 326 (MH+) for
C16H11N3 OS2. Step f: 3,5-Di-2-thienyl-1,3-dihydro-2H-pyrido[2,3-e][1,4]
11. Zhang, W.; Liu, R.; Huang, Q.; Zhang, P.; Koehler, K. F.; Harries, B.; Skolnick, P.;
Cook, J. M. Eur. J. Med. Chem. 1995, 3, 483.
12. Patani, G. A.; LaVoie, E. J. Chem. Rev. 1996, 96, 3147.
13. Dalvie, D. K.; Kalhutkar, A. S.; Khojasteh-Bakht, S. C.; Obach, R. S.; O’Donnell, J.
P. Chem. Res. Toxicol. 2002, 15, 269.
14. Carlier, P. R.; Zhao, H.; DeGuzman, J.; Lam, P. C.-H. J. Am. Chem. Soc. 2003, 125,
11482.
15. Burger’s Medicinal Chemistry Drug Discovery, 6th ed.; Abraham, D. J., Eds.; ISBN
0-471-27401-1, John Wiley & Sons, Inc.: Hoboken, NJ, 2003; Vol. 6, Chapter 9,
pp 525–597.
16. (a) Fisher, S. L. Microbiol. Biotechnol. 2008, 1, 345; (b) Geng, B.; MacPherson, L.;
Mullen, G.; Loch, J.; Comita-Prevoir, J.; Morningstar, M.; Satz, A.; Kiely, A.;
Basarab, G.; Hill, P.; Eyermann, C.; Gowravaram, M.; Osimboni, E. 46th ICAAC
2006, 5, 20. Abstracts F2-1173; c Keating, T.; Fisher, S. L.; Lundqvist, T.; 46th
ICAAC, 2006, Abstracts F2-1172. X-ray co-crystal structure PDB code of
compound 7: 2w4i.
diazepine-2-thione
3,5-di-2-thienyl-1,3-dihydro-2H-pyrido[2,3-e][1,4]-
diazepin-2-one (0.48 g, 1.476 mmol) was refluxed with Lawesson’s reagent
(1.0 g, 2.47 mmol) in dioxane for 12 h. The reaction mixture was then
concentrated and the product was purified on Flash master silica-gel
chromatography using ethyl acetate/hexane as eluant. The desired fractions
17. (a) De Jonge, B. L. M.; Kutschke, A. Antimicrob. Agents Chemother.
2005, 49, 3009; (b) De Jonge, B. L. M.; Kutschke, A. 46th ICAAC
2006, 5, 20.
18. Newman, J. V.; Rooney, M. T.; Cederberg, C. 46th ICAAC 2006, 5, 20.