Negishi cross-coupling enabled synthesis of novel NAD+-dependent DNA ligase inhibitors and SAR development

Article abstract of DOI:10.1016/j.bmcl.2015.09.075

Two novel compounds, pyridopyrimidines (1) and naphthyridines (2) were identified as potent inhibitors of bacterial NAD⁺-dependent DNA ligase (Lig) A in a fragment screening. SAR was guided by molecular modeling and X-ray crystallography. It was observed that the diaminonitrile pharmacophore made a key interaction with the ligase enzyme, specifically residues Glu114, Lys291, and Leu117. Synthetic challenges limited opportunities for diversification of the naphthyridine core, therefore most of the SAR was focused on a pyridopyrimidine scaffold. The initial diversification at R¹ improved both enzyme and cell potency. Further SAR developed at the R² position using the Negishi cross-coupling reaction provided several compounds, among these compounds 22g showed good enzyme potency and cellular potency.

Full text of DOI:10.1016/j.bmcl.2015.09.075

Bioorganic & Medicinal Chemistry Letters xxx (2015) xxx–xxx
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Negishi cross-coupling enabled synthesis of novel NAD+-dependent
DNA ligase inhibitors and SAR development
a
Kerry E. Murphy-Benenato a, Lakshmaiah Gingipalli b, , P. Ann Boriack-Sjodin ,
⇑
Gabriel Martinez-Botella a, Dan Carcanague a, Charles J. Eyermann a, Madhu Gowravaram a, Jenna Harang a,
Michael R. Hale b, Georgine Ioannidis a, Harris Jahic a, Michele Johnstone a, Amy Kutschke a,
Valerie A. Laganas a, James T. Loch III a, Matthew D. Miller a, Herbert Oguto a, Sahil Joe Patel c
a Infection Innovative Medicines Unit, AstraZeneca R&D Boston, 35 Gatehouse Drive, Waltham, MA 02451, USA
b Oncology Innovative Medicines Unit, AstraZeneca R&D Boston, 35 Gatehouse Drive, Waltham, MA 02451, USA
c Discovery Sciences, AstraZeneca R&D Boston, 35 Gatehouse Drive, Waltham, MA 02451, USA
a r t i c l e i n f o
a b s t r a c t
Article history:
Two novel compounds, pyridopyrimidines (1) and naphthyridines (2) were identified as potent inhibitors
of bacterial NAD+-dependent DNA ligase (Lig) A in a fragment screening. SAR was guided by molecular
modeling and X-ray crystallography. It was observed that the diaminonitrile pharmacophore made a
key interaction with the ligase enzyme, specifically residues Glu114, Lys291, and Leu117. Synthetic chal-
lenges limited opportunities for diversification of the naphthyridine core, therefore most of the SAR was
focused on a pyridopyrimidine scaffold. The initial diversification at R1 improved both enzyme and cell
potency. Further SAR developed at the R2 position using the Negishi cross-coupling reaction provided
several compounds, among these compounds 22g showed good enzyme potency and cellular potency.
Ó 2015 Elsevier Ltd. All rights reserved.
Received 28 June 2015
Revised 28 September 2015
Accepted 30 September 2015
Available online xxxx
Keywords:
Cross-coupling
Diaminonitrile
Pyridopyrimidine
NAD+-dependent ligase resistance
Antibacterial
As bacterial pathogens are developing resistance to existing
drugs from the overuse and over prescription of current therapies,
the need for new and novel antibiotics has become more urgent.1
In addition, this has been hindered through reduction of focus
and resources directed towards discovery and development of
new agents in the pharmaceutical industry. There are fewer
options for the treatment of Gram-negative pathogens in compar-
ison to Gram-positive pathogens and the discovery and optimiza-
tion against these targets is typically more challenging due to
increased difficulty in accessing the cellular targets.2 One approach
for overcoming resistance to existing antibacterial drugs is through
the development of drugs that inhibit new targets. Bacterial
NAD+-dependent DNA ligase (LigA) was selected as a target for
these reasons. LigA is essential for the growth of a broad spectrum
of Gram-positive and Gram-negative pathogens,3 and has limited
homology to the human DNA ligases. DNA ligases are critical in
DNA repair, replication, and recombination. Approaches to LigA
inhibitors have been recently reported. Bayer reported a series of
potent pyridochromanone inhibitors.4 AstraZeneca recently
of classes include: aryl amino acids,7 tetracyclic indoles,8
glycosyl ureides and glycosylamines,9 pyrimidopyrimidines,10
aminoalkoxypyrimidines,11 thienopyridines12 and anilines.13
Our group recently identified two scaffolds as potent inhibitors
of LigA through a fragment based screening effort pyridopyrimidi-
nes (1) and naphthyridines (2) (Fig. 1). The compounds demon-
strated excellent in vitro activity and good physical properties.
NH2
H2N
N
O
O
N
N
N
O
N
O
NH2
O
OH
F
Bayer Analog
AZ Adenosine
OH
H2N
N
N
N
H2N
NC
N
N
H2N
NC
N
N
O
N
reported on
a
series of adenosine based inhibitors5
NH2
NH2
NH2
HN
NH
and napthyridines.6 Additional reports of inhibitors from a number
2
3
1
⇑
Corresponding author.
Figure 1. Reference and lead series for ligase inhibition.
http://dx.doi.org/10.1016/j.bmcl.2015.09.075
0960-894X/Ó 2015 Elsevier Ltd. All rights reserved.
Please cite this article in press as: Murphy-Benenato, K. E. ; et al. Bioorg. Med. Chem. Lett. (2015), http://dx.doi.org/10.1016/j.
bmcl.2015.09.075

2
K. E. Murphy-Benenato et al. / Bioorg. Med. Chem. Lett. xxx (2015) xxx–xxx
Table 1
Antibacterial activity of initial compounds
a
#
R2
Core
Ligase IC50
(l
M)
MIC (lM)
Soln (
lM)
AZ logD14
Spn
0.037
<0.02
0.15
Sau
Eco
Hin
Spn ARC 548
Sau ARC 516
Hin ARC 158
1
2
3
2-Pyrrole
H
2-Pyrrole
Pyridopyrimidine
Napthyridine
Pyridopyrimidine
0.077
<0.04
0.5
0.077
0.23
NT
0.03
2.24
2.09
50
3.13
50
25
3.13
50
12.5
50
100
130
14.2
50
2.5
2.1
2.5
a
Spn = Streptococcus pneumoniae, Sau = Staphylococcus aureus, Eco = Escherichia coli, Hin = Haemophilus influenzae, NT = not tested. Pathogens; Streptococcus pneumoniae
ARC548 (ATCC), Staphylococcus aureus ARC516 (CLSI QC MRSA reference strain), Haemophilus influenzae ARM158 (acrB gene deletion based on ARC446), Escherichia coli
ARC524. Soln is aqueous solubility measured at pH 7.4 from dried DMSO stock as determined in a 0.1 M phosphate buffer at 20 °C after 18 h.14
H2N
NC
N
N
R1
H2N
NC
N
N
R1
H2N
NC
N
Br
a, b
N
X
N
X = C, N
NH2 R2
NH2
9
4a-k
Figure 2. Generic structure of diaminonitrile scaffold.
H2N
NC
N
N
NH
CN
CN
c, b
EtO
+
N
H2N
Compound 1, a pyridopyrimidine, demonstrated broad spectrum
biochemical potency, and compound 2, a diaminonitrile naph-
thyridine, demonstrated improved cellular activity (Table 1). When
we combined aspects of both series in pyridopyrimidine core 3,
this resulted in a drop of both enzymatic and cellular activity.
Through modeling and crystallography, it was observed the
diaminonitrile pharmacophore made key interactions with the
ligase enzyme and was therefore not modified, these interactions
include residues Glu114, Lys291, and Leu117. For improving inter-
action with the enzyme, the key variables were the R1 and R2 sub-
stituents (Fig. 2). R1 reaches into a hydrophobic tunnel which is
conserved among ligase enzymes. R2 affords access to the ribose
pocket, a region previously shown to afford excellent biochemical
potency in a series of adenosine-based inhibitors as shown in
Figure 1, this series also contained a t-Butyl in a position analogous
to R1.5,14
NH2
4l
10
11
R1:
H
Me
4b#
4a#
4e (17%)
4f (61%)
4g (86%)
4c (98% 4d (37%)
4k (91%)
4h (38%)
4i (28%)
4j (8%)
Scheme 2. Synthesis of R1 substituted ligase inhibitors. Reagents and conditions:
(a) R1ZnBr, 5 mol % Pd(PPh3)4, THF, 60 °C, 18–98%; (b) CNCH2CN, NaOEt, EtOH,
75 °C, 3–12%; (c) Na, EtOH, 45%. #Commercially available.
The initial chemistry to access the diaminonitrile lead did not
allow for facile diversification at R1 and R2. As can be seen in
Scheme 1, the synthesis of diaminonitrile 2 begins by condensation
of the sodium salt of (E)-4,4-dimethyl-3-oxopent-1-en-1-olate (5)
and ethyl 3-amino-3-iminopropanoate (6) to provide pyridine 7.
Subsequent saponification of the ester, cyclization with phosgene
and condensation with malononitrile, affords the target diaminon-
itrile 2. In order to achieve our optimization goals, we required a
synthesis that would allow for late stage diversification of the R1
and R2 substituents. The current naphthyridine scaffold was not
amenable to this functionalization, so we modified the core to a
pyridopyrimidine and developed this methodology.
enabled a variety of alkyl based R1 substituents to be introduced.
We focused mainly on alkyl groups since this portion of the bind-
ing pocket is made up of mainly hydrophobic amino acids. As can
be seen in Scheme 2, bromopyrimidine 9 undergoes Negishi cross-
coupling reaction with a variety of alkylzinc bromides, yields are
moderate to excellent and are unoptimized. Subsequent condensa-
tion with malononitrile afforded the desired targets. Synthesis of
the tert-butyl derivative 4l required an alternative synthesis since
reaction of pyrimidine 9 with tert-butyl zinc bromide afforded only
trace amounts of desired product, and instead provided iso-butyl
derivative 4i. Instead, we found condensation of tert-butyl amidine
10 and acrylonitrile 11 and subsequent cyclization provided the
desired inhibitor, 4l (R1 = tert-Bu).
We first sought to explore the SAR of the R1 position in order to
confirm that tert-butyl was the optimal substituent. Towards this
end we developed a two-step route to final compounds which
H
O
NH
O
H2N
EtO
N
O
N
N
b, c
a
NaO
+
H2N
OEt
O
O
O
5
6
8
7
H2N
NC
N
N
d
NH2
2
Scheme 1. Synthesis of diaminonitrile 2. Reagents and conditions: (a) pyridine, dioxane, 100 °C, 69%; (b) NaOH, EtOH, 80 °C, 69%; (c) phosgene, Na2CO3, H2O, 72%;
(d) malononitrile, Et3N, DMF, 110 °C, 40%.
Please cite this article in press as: Murphy-Benenato, K. E. ; et al. Bioorg. Med. Chem. Lett. (2015), http://dx.doi.org/10.1016/j.
bmcl.2015.09.075

K. E. Murphy-Benenato et al. / Bioorg. Med. Chem. Lett. xxx (2015) xxx–xxx
3
As can be seen in Table 2, compound 4a indicates substitution
at R1 is required for activity. As the size increased (4b, 4c) so did
the activity, with compound 4c showing moderate cellular activity
and excellent solubility. Cyclic derivatives 4d–4g show comparable
biochemical activity to 4c. Extending the alkyl substituent, as in 4h
and 4i caused a marked decrease in biochemical activity. Aryl
derivatives 4j and 4k were also less potent in biochemical assays
when compared to the alkyl derivatives. As seen with other
series,15 tert-butyl derivative 4l demonstrated the best biochemical
and cellular activity as compared to the other substituents. How-
ever, when comparing the matched pair to 4l, compound 2, there
is an overall loss in all biochemical activity with the pyridopyrim-
idine core in contrast to the napthyridine.
Based on this data, we planned to move forward to explore the
R2 position with a tert-butyl group at R1. We chose a substrate with
a chloro substituent at the R2 position to enable access to a diverse
set of compounds. Condensation of diethyl malonate and amidine
10 (Scheme 3), followed by Vilsmeier reaction, provided
dichloroaldehyde 13. Oxime formation, elimination to the nitrile,
followed by mono displacement with ammonia, provided aminon-
itrile 14. Due to concern about the stability of chloride 14 to the
basic cyclization conditions, we performed a Stille cross-coupling
at this point to provide 15. Direct cyclization with malononitrile,
analogous to the previous highlighted routes, did not afford any
desired product so a two-step conversion was explored. However,
after formation of amide 16, no cyclization conditions could be
found to form the desired product (see Scheme 4).
With the inability of the chloride to survive the cyclization con-
ditions and our inability to cyclize with the R2 substituent in place,
we required a more inert cross-coupling partner. Thioethers offer a
similar reactivity profile in cross-coupling reactions as aryl chlo-
rides and should be more chemically stable.16,17 Recent reports
have demonstrated that thioethers undergo efficient transition
metal catalyzed cross-coupling reactions with a variety of nucle-
ophiles. Therefore, we sought an efficient synthesis of a thioether
cross coupling partner. Beginning with dinitrile 17, cyclization pro-
vided thiomethyl substituted pyrimidine 18. Alkylation with
cyanoacetate, followed by cyclization afforded 19. Chlorination
and two-step introduction of the amine provided the key substrate
for diversification, 21. The antibacterial activity of compound 21
(Table 3) was improved as compared to the unsubstituted deriva-
tive 4l. The biochemical and cellular activity was comparable to the
original naphthyridine 2, indicating substitution at this position is
favorable.
Intermediate 21 enables the exploration of
a variety of
cross-coupling reactions to access C-linked derivatives. Initial
Table 2
SAR of R1 substituted pyridopyrimidines
H2N
NC
N
N
R1
N
NH2
#
R1
Ligase IC50
(l
M)a
MIC (
lM)
Soln (
lM)
AZ logD
Spn
Sau
Hin
Spn ARC 548
Sau ARC 516
Hin ARC 158
4a
4b
4c
4d
4e
4f
4g
4h
4i
H
Me
iso-Pr
Cyclopropyl
Cyclobutyl
Cyclopentyl
Cyclohexyl
CH2-cyclohexyl
iso-Bu
44
13
1.5
>200
>10
>10
>10
13
>10
>10
>10
>10
>10
>10
>200
NT
NT
25
100
50
50
200
NT
NT
>200
NT
NT
NT
NT
NT
25
6
783
>1000
23
217
89
81
8
268
NT
13
0.1
0.3
1.3
1
1.5
2
2.5
2.9
1.7
2.1
1.7
1.9
>10
0.34
0.53
0.46
0.52
0.59
5
0.41
0.67
0.69
0.47
0.81
3.7
>200
200
200
200
>200
NT
NT
>200
NT
200
>200
200
>200
NT
NT
>200
NT
1.8
1.1
4j
4k
4l
Ph
Bn
tert-Butyl
0.72
>10
0.12
0.97
>10
0.22
12.5
25
200
37.2
a
Spn = Streptococcus pneumoniae, Sau = Staphylococcus aureus, Hin = Haemophilus influenzae, NT = not tested. Pathogens; Streptococcus pneumoniae ARC548, Staphylococcus
aureus ARC516 (CLSI QC MRSA reference strain), Haemophilus influenzae ARM158 (acrB gene deletion based on ARC446). Soln is aqueous solubility measured at pH 7.4 from
dried DMSO stock as determined in a 0.1 M phosphate buffer at 20 °C after 18 h.
H2N
NC
N
Cl
N
O
O
a, b
c, d, e
O
N
N
EtO
OEt
Cl
Cl
12
14
13
H
N
H2N
NC
N
N
NC
g
f
N
S
O
N
S
NC
N
N
16
15
Scheme 3. Attempts at synthesis of R2 substituted analogues. Reagents and conditions: (a) 10, Na, EtOH, 90 °C, 79%; (b) POCl3, DMF, 110 °C, 12 h, 48%; (c) HONH2ÁHCl, H2O,
AcOH, 30 min, 64%; (d) SOCl2, 3 h, 97%; (e) NH3, EtOH, 75 °C, 30 min, 31%; (f) 2-tributylstannylthiazole, Pd(PPh3)2Cl2, Ag2O, DMF, 100 °C, 24 h, 53%; (g) 2-cyanoacetic acid,
Ac2O, 80 °C, 30 min, 39%.
Please cite this article in press as: Murphy-Benenato, K. E. ; et al. Bioorg. Med. Chem. Lett. (2015), http://dx.doi.org/10.1016/j.
bmcl.2015.09.075

4
K. E. Murphy-Benenato et al. / Bioorg. Med. Chem. Lett. xxx (2015) xxx–xxx
utilized in a cross-coupling with thioethers as the electrophile,23
H
N
O
N
H2N
NC
N
and to date there are limited reports of cross-coupling with
heteroarylzinc bromides containing multiple heteroatoms. How-
ever we were pleased to observe the desired product formation
(22g–22q) when thioether 21 was treated with catalytic Pd
(PPh3)4 and a heteroarylzinc bromide, yields are low to moderate
and are unoptimized.
NC
NC
SMe
SMe
a
b,c
N
N
NC
NH2 SMe
SMe
18
17
19
An evaluation of the SAR at the R2-position of the pyridopyrim-
idine 21 was carried out next (Table 3). Substitution at R2 with dif-
ferent alkyl groups (22a–f) showed moderate biochemical activity
as compared to compound 2, however the higher logD analogs
afforded reduced MICs. Next we made a series of five membered
heterocycles (22g–i and 26o–q), activity in Haemophilus influenzae,
a representative Gram-negative pathogen is generally more diffi-
cult to achieve with this substitution. The thiazole, 22g, is the mod-
erately potent in Gram + positive MICs, but is weaker in
Haemophilus. Simple five membered heterocycles offer the best
overall profile based on cellular potency across multiple pathogens
for compounds made in this campaign. In the six membered series,
pyridine 22k–m at R2 position showed only modest differences in
biochemical potency compared to phenyl 22j, however 3-pyridine
22l showed improved Gram positive MICs. The matched pair
pyrimidine 22n, was not an improvement when compared to the
other six membered aryl analogs. We next focused on further
improving enzyme and cell potency of compound 22g through
the exploration of further substitution.
Cl
N
N
H2N
NC
N
N
N
d
e,f
N
NC
NH2 SMe
NH2 SMe
21
20
Scheme 4. Synthesis of substrate for diversification. Reagents and conditions: (a)
10, Na, EtOH, 42%; (b) CNCH2CO2H, Ac2O, CH2Cl2, 40 °C, 75%; (c) LiHMDS, THF, 70 °C,
39–93%; (d) POCl3, >98%; (e) para-methoxybenzylamine, DMF, 50 °C, 70%; (f) TFA,
95%.
experimentation with compound 21 indicated Pd-catalyzed
Cu-mediated Stille reactions were effective,18 however the analo-
gous Suzuki cross-couplings19 did not afford any desired products.
Early attempts at Negishi cross-couplings showed promise so we
decided to pursue this further as alkyl and aryl zinc reagents are
readily available20 and we desired to avoid toxic tin reagents.21
Towards this end, treatment of 21 with a variety of commer-
cially available alkylzinc bromides directly afforded C-linked ana-
logues 22a–22q (Scheme 5).22 Based on our previous results with
aryl substituted pyridopyrimidine 1 (Fig. 1) we were also inter-
ested in investigating the Pd-catalyzed cross-coupling with arylz-
inc bromides, especially heteroaryl zinc reagents. At the time of
this work there were no reports of heteroaryl zinc bromides being
Though we were able to access the desired aryl substituted ana-
logues 22g–22q, the yields were often poor and attempts at intro-
duction of N-containing heterocycles via the same methods
provided the desired products in <5% yield. We hypothesized the
Table 3
SAR of R2 substituted pyridopyrimidines (R1 = tert-Butyl)
H2N
NC
N
N
N
NH2 R2
#
R2
Spn
Sau
Hin
Spn ARC 548
Sau ARC 516
Hin ARC 158
Soln (lM)
AZ logD
21
-S-methyl
-Methyl
-n-Butyl
-(CH2)4-CN
-(CH2)5-CN
Cyclopentyl
-(CH2)5-cyclohexyl
-2-Thiazole
-2-Thiophene
-3-Thiophene
-Ph
-2-Pyridine
-3-Pyridine
-4-Pyridine
-2-Pyrimidine
-4-Carboethoxy-2-thiazole
-5-Carboethoxy-2-thiazole
-2-Thiophene-5-carboxylate
-2-Imidazole
-2-(4-Methyl imidazole)
-3-Pyrrazole
-4-1,2,3 triazole
-3-1,2,4 triazole
-2-Oxazole
0.032
0.067
0.043
0.044
0.12
0.051
0.3
<0.02
0.044
0.31
0.059
0.22
0.11
0.26
0.35
0.12
1.6
0.057
0.16
0.74
0.47
1.01
0.43
0.4
1.44
0.53
>200
0.55
0.036
0.068
0.1
0.042
0.27
0.11
0.084
0.17
0.72
1.7
1.1
0.65
1.8
2.1
>200
0.49
0.52
2.19
1.8
6.25
6.25
50
12.5
50
12.5
25
>200
50
>200
>200
>200
1.56
6.25
50
>200
100
50
>200
200
>200
>200
>200
6.25
>200
6.25
25
12.5
12.5
>200
>200
>200
25
>200
50
>200
50
>200
>200
25
14
37
6
16
6
<1
6
<1.5
56
40
4
87
>200
44
86
NT
NT
>1000
<1
5
3.4
2.1
3
22a
22b
22c
22d
22e
22f
22g
22h
22i
1.9
2.2
3.4
4.2
2.7
3.1
2.8
2.8
2.3
1.9
1.9
1.3
3
25
>200
0.78
1.56
25
50
50
100
22j
22k
22l
0.15
0.42
0.11
0.16
>200
>100
50
6.6
1.33
2.75
13.98
>10
>200
0.62
0.31
NT
0.78
0.64
1.4
1.65
0.5
3.13
50
22m
22n
22o
22p
22q
26a
26b
26c
26d
26e
26f
30
>200
>200
>200
>200
>200
>200
>200
25
0.55
100
>200
>200
>200
3.13
12.5
1.56
0.78
3.13
6.25
3.13
>200
<0.02
<0.39
0.034
0.021
<0.02
0.031
0.013
0.12
3
À0.4
2.2
2.4
2.4
1.9
1.9
2
13.5
30
64.5
50
<1
<1
100
12.5
>200
>200
>200
0.026
0.007
<0.02
-Thiazole alcohol
-Thiazole methyl ether
2.1
2.6
32
0.79
a Spn = Streptococcus pneumoniae, Sau = Staphylococcus aureus, Hin = Haemophilus influenzae, NT = not tested. Pathogens; Streptococcus pneumoniae ARC548, Staphylococcus
aureus ARC516 (CLSI QC MRSA reference strain), Haemophilus influenzae ARM158 (acrB gene deletion based on ARC446). Soln is aqueous solubility measured at pH 7.4 from
dried DMSO stock as determined in a 0.1 M phosphate buffer at 20 °C after 18 h.
Please cite this article in press as: Murphy-Benenato, K. E. ; et al. Bioorg. Med. Chem. Lett. (2015), http://dx.doi.org/10.1016/j.
bmcl.2015.09.075

K. E. Murphy-Benenato et al. / Bioorg. Med. Chem. Lett. xxx (2015) xxx–xxx
5
H2N
NC
N
N
a
21
N
NH2 R2
22a-r
R2:
Me
N
N
S
S
N
22a (4%) 22b (11%)
22c
22d (6%)
22e
22f
(6%)
22g (23%) 22h (
50%)
(9%)
(6%)
N
S
N
S
N
N
S
N
N
S
N
CO2H
22q
CO2Et
(3%)
EtO2C
)
2p (31%
2
22i
22j
22k
22l
22m
22n
22o
(66%)
(31%)
(19%)
(2%)
(4%)
(3%)
(98%
ester)
Scheme 5. Synthesis of C-linked analogues via Negishi cross-coupling. Reagents and conditions: (a) R2ZnBr (22a–q), 10 mol % Pd(PPh3)4, THF, 65 °C, 3–98%.
PMB(H)N
NSEM
N
N
H2N
NC
N
N
X
X
a
X
X
b, c, d
NH
N
N
NC
X
X
NH2 SMe
NH2
X
X
NH
X
24a f
-
23a-f
26a-f
25
N
N
N
N
N
NH
NH
NH
NH
NH
O
N
N
N
26a (44%)
26b (78%)
26c (13%)
26d (12%)
26e (53%)
26f (5%)
Scheme 6. Synthesis of azole substituted analogues. Reagents and conditions: (a) SEMCl, NaH, THF, rt, 16–60%; (b) 24a–f, n-BuLi, ZnBr2, THF, À78 °C, 1 h; Pd(PPh3)4, THF,
65 °C; (c) HCl, 65–98%; (d) TFA, 15–92%.
efficiency of the reaction suffered due to the poor solubility of the
starting material. To address this, we examined the same cross
coupling with PMB protected analogue 25 (see Scheme 6). To
access the required arylzinc bromides we protected the desired
azole (23a–e) with a SEM group to provide 24a–e. Deprotonation
with n-BuLi at À78 °C followed by treatment with dry ZnBr2
formed the desired arylzinc bromide in situ. Subsequent addition
of substrate 25 and Pd-catalyst afforded the desired cross coupling
in good yields. Two step removal of protecting groups gave the
desired final products 26a–f.
aureus. Pyrazole 26c and triazoles 26d and 26e had comparable
MICs to thiazole 22g in Streptococcus pneumoniae and had
improved solubility. However, triazoles 26d and 26e were less
potent than 22g against S. aureus.
In an attempt to improve the physical properties of the most
potent analogue 22g, we sought to introduce substitution off the
thiazole. Beginning with ester 27, since it proved to have better
enzymatic activity than its isomer 22o, LiAlH4 reduction provided
alcohol 28 (Scheme 7). The intermediate alcohol could be silylated
or methylated to provide 29 and 31, respectively. Treatment of
bromides 29 and 31 with activated zinc dust produced the arylzinc
bromides in situ. Pd-catalyzed cross coupling with methyl
thioether 25 and subsequent removal of protecting groups afforded
the final products 30 and 32. Unfortunately, the introduction of the
As can be seen in Table 3, the azole analogues 26a–26f demon-
strated good biochemical potency against bacterial ligase and pro-
vided Gram-positive MICs. Methyl substituted imidazole 26b is the
only analogue that did not have any activity against Staphylococcus
Br
Br
Br
H2N
NC
N
N
a
b
c-e
N
S
N
N
S
S
N
S
HO
TBSO
EtO2C
NH2
N
30
27
28
29
HO
f
Br
H2N
NC
N
N
c-e
N
S
N
S
MeO
NH2
N
32
31
MeO
Scheme 7. Synthesis of functionalized thiazoles. Reagents and conditions: (a) LiAlH4, THF, 0 °C; (b) TBSCl, imidazole, DMF, 78%; (c) Zn dust, Br(CH2)2Br, TMSCl, THF; (d) Pd
(PPh3)4, THF 65 °C, 43%; (e) TFA, 25–52%; (f) MeI, NaH, DMF, 60%.
Please cite this article in press as: Murphy-Benenato, K. E. ; et al. Bioorg. Med. Chem. Lett. (2015), http://dx.doi.org/10.1016/j.
bmcl.2015.09.075

6
K. E. Murphy-Benenato et al. / Bioorg. Med. Chem. Lett. xxx (2015) xxx–xxx
Figure 3. Structure of compound 22g in H. influenzae DNA ligase (PDB Code: 4UFZ).
solubilizing groups did not increase the solubility and was detri-
mental to the antibacterial potency (Table 3).
Crystallographic analysis of 22g in H. influenzae DNA Ligase con-
firmed the binding mode (Fig. 3). The inhibitor maintains multiple
interactions with the active site: Leu117, Glu114, Lys291. In addi-
References and notes
1. Mills, S.; Eakin, A.; Buurman, E.; Newman, J.; Gao, N.; Huynh, H.; Johnson, K.;
Lahiri, S.; Shapiro, A.; Walkup, K.; Yang, W.; Stokes, S. S. Antimicrob. Agents
Chemother. 2011, 1088.
2. Hale, R. M.; Brassington, C.; Carcanague, D.; Embrey, K.; Eyermann, J. C.;
Gingipalli, L.; Gowravaram, M.; Harang, J.; Howard, T.; Ioannidis, G.; Jahic, H.;
Kutschke, A.; Laganas, A. V.; Loch, J.; Miller, D. M.; Murphy-Benenato, E. K.;
Oguto, H.; Otterbein, L.; Patel, S. J.; Shapiro, B. A.; Boriack-Sjodin, P. A.
Tetrahedron Lett. 2015, 56, 3108.
3. Ciarrocchi, G.; MacPhee, D. G.; Deady, L. W.; Tilley, L. Antimicrob. Agents
Chemother. 1999, 43, 2766.
4. Srivastava, S. K.; Dube, D.; Vandna, K.; Jha, A. K.; Hajela, K.; Ramachandran, R.
Proteins 2007, 69, 97.
5. (a) Srivastava, S. K.; Dube, D.; Tewari, N.; Dwivedi, N.; Tripathi, R. P.;
Ramachandran, R. Nucleic Acids Res. 2005, 33, 7090; (b) Srivastava, S. K.;
Tripathi, R. P.; Ramachandran, R. J. Biol. Chem. 2005, 280, 30273.
6. Shuman, S. J. Biol. Chem. 2009, 284, 17365.
7. Wang, T.; Duncan, L.; Gu, W.; O’Dowd, H.; Wei, Y.; Perola, E.; Parsons, J.; Gross,
C. H.; Moody, C. S.; Ryan, A. S. J.; Charifson, P. S. Bioorg. Med. Chem. Lett. 2012,
22, 3699.
8. Murphy-Benenato, K.; Wang, H.; McGuire, H. M.; Davis, H. E.; Gao, N.; Prince, D.
B.; Jahic, H.; Stokes, S.; Boriack-Sjodin, P. A. Bioorg. Med. Chem. Lett. 2014, 24,
360.
tion, the aromatic core
p-stacks with the Tyr226 residue. The t-
butyl in this case efficiently fills the flexible lipophilic pocket.
The electron density of the thiazole was not well defined indicating
it does not pick up any particular interactions. The reduction in
potency of 22g in H. influenzae compared to the carboxamide com-
pound 1 may be rationalized because in compound 1, the carbox-
amide is making direct H-bonds to both the backbone carbonyl and
the backbone NH of Leu 117. Overall, replacing the carboxamide
with the aminonitrile may reduce the overall strength of the H-
bonding interactions with Leu117. In general a carboxamide NH2
is a much stronger H-bond donor than an aromatic NH2 moiety
and the nitrile nitrogen is a weaker H-bond acceptor than an amide
carbonyl.
In conclusion, we developed and utilized two Negishi cross-
coupling reactions to enable evaluation of SAR in NAD+-Dependent
DNA Ligase inhibitors, our design was guided by modeling and
crystallography. We utilized the earlier structures of LigA to design
the novel analogs focusing on evaluation of diversity at two posi-
tions as well as incorporating a range of properties. Because of
the synthetic difficulty for the diversification of R1 and R2 function-
alization in the napthyridine core, we evaluated these positions in
the pyridopyrimidine series. Initial SAR for the hydrophobic pocket
was explored via Negishi cross-coupling chemistry. Evaluation of
R1 was done with R2 = H, and it was found tert-butyl 4l gave the
best biochemical and cellular activity compared to other sub-
stituents, however the compound still required improvements in
cellular potency. We then explored the ribose pocket, R2. Function-
alization at this position was challenging and we found Negishi
9. (a) Gu, W.; Wang, T.; Maltais, F.; Ledford, B.; Kennedy, J.; Wei, Y.; Gross, C. H.;
Parsons, J.; Duncan, L.; Arends, S. J.; Moody, C.; Perola, E.; Green, J.; Charifson, P.
S. Bioorg. Med. Chem. Lett. 2012, 22, 3693; (b) Meier, T. I.; Yan, D.; Peery, R. B.;
McAllister, K. A.; Zook, C.; Peng, S. B.; Zhao, G. FEBS J. 2008, 275, 5258.
10. Stokes, S. S.; Gowravaram, M.; Huynh, H.; Lu, M.; Mullen, G. B.; Chen, B.; Albert,
R.; O’Shea, T. J.; Rooney, M. T.; Hu, H.; Newman, J. V.; Mills, S. D. Bioorg. Med.
Chem. Lett. 2012, 22, 85.
11. Liebiskind, S. L.; Srogl, J. Org. Lett. 2002, 4, 979.
12. Dubbak, R. S.; Vogel, P. Angew. Chem., Int. Ed. 2005, 44, 7674.
13. Egi, M.; Liebiskind, S. L. Org. Lett. 2003, 5, 801.
14. Miyaura, N.; Suzuki, A. Chem. Rev. 1995, 95, 2457.
15. All of the zinc bromides were synthesized by refluxing of zinc and the
corresponding alkyl bromides in THF.
16. Utas, E. J.; Olofsson, B.; Akermark, B. Synlett 2006, 1965.
17. Angiolelli, E. M.; Casalnuovo, L. A.; Selby, P. T. Synlett 2000, 905.
18. Guillaume, D.; Knochel, P. Synlett 2006, 407.
19. (a) Melzig, L.; Metzger, A.; Knochel, P. J. Org. Chem. 2010, 75, 2131; (b) Melzig,
Laurin Dissertation, 2011.
20. (a) Stokes, S. S.; Huynh, H.; Gowravaram, M.; Albert, R.; Cavero-Tomas, M.;
Chen, B.; Harang, J.; Loch, J. T.; Lu, M.; Mullen, G. B.; Zhao, S.; Liu, C. F.; Mills, S.
Bioorg. Med. Chem. Lett. 2011, 21, 4556; (b) Brötz-Oesterhelt, H.; Knezevic, I.;
Bartel, S.; Lampe, T.; Warnecke-Eberz, U.; Ziegelbauer, K.; Häbich, D. J. Biol.
Chem. 2003, 278, 39435.
cross-coupling with
a variety of alkylzinc bromides readily
afforded pyridopyrimidines 22a–f. Compound 22g, a diaminoni-
trile based inhibitor, was found to be more potent than compound
2. Additional efforts are required to further improve the biochem-
ical potency of this series including synthesis and profiling the cor-
responding napthyridine based aminocarboxamides.
21. Shuman, S. Structure 2004, 1335.
22. (a) Denyer, S. P.; Maillard, J. Y. J. Appl. Microbiol. 2002, 92, 35S; (b) Talbot, G. H.;
Bradley, J. S.; Edwards, J. E.; Gilbert, D.; Rice, L. B.; Scheld, M.; Spellberg, B.;
Bartlett, J. Clin. Infect. Dis. 2009, 48, 1.
23. Giamarellou, H.; Poulakou, G. Drugs 2009, 69, 1879.
Acknowledgment
The authors would like to thank Melissa Vasbinder and Hoan
Huynh for the critical review and the helpful discussions.
Please cite this article in press as: Murphy-Benenato, K. E. ; et al. Bioorg. Med. Chem. Lett. (2015), http://dx.doi.org/10.1016/j.
bmcl.2015.09.075