Novel non-nucleobase inhibitors of Staphylococcus aureus DNA polymerase IIIC

Article abstract of DOI:10.1016/j.bmcl.2005.11.009

The preparation and biological evaluation of 5-substituted-6-hydroxy-2- (anilino)pyrimidinones as a new class of DNA polymerase IIIC inhibitors, required for the replication of chromosomal DNA in Gram-positive bacteria, are described. These new dGTP compe

Full text of DOI:10.1016/j.bmcl.2005.11.009

Bioorganic & Medicinal Chemistry Letters 16 (2006) 891–896
Novel non-nucleobase inhibitors of Staphylococcus aureus
DNA polymerase IIIC
*
*
´
Yannick Rose, Stephane Ciblat, Ranga Reddy, Adam C. Belley, Evelyne Dietrich,
Dario Lehoux, Geoffrey A. McKay, Hugo Poirier, Adel Rafai Far and Daniel Delorme
Targanta Therapeutics Inc., 7170 Frederick-Banting, 2nd Floor, Montre´al, Que´bec, Canada H4S 2A1
Received 4 October 2005; revised 1 November 2005; accepted 1 November 2005
Available online 17 November 2005
Abstract—The preparation and biological evaluation of 5-substituted-6-hydroxy-2-(anilino)pyrimidinones as a new class of DNA
polymerase IIIC inhibitors, required for the replication of chromosomal DNA in Gram-positive bacteria, are described. These
new dGTP competitive inhibitors displayed good levels of in vitro inhibition and antibacterial activity against Staphylococcus aure-
us. A new class of dATP competitive inhibitors, 6-substituted-2-amino-5-alkyl-pyrimidin-4-ones, whose antibacterial activity was
unaffected by serum, were identified.
ꢀ 2005 Elsevier Ltd. All rights reserved.
The emergence of bacterial resistance to antibiotics is an
alarming problem in the treatment of infectious diseases.
Numerous multi-resistant strains of Gram-positive
(Gr+) pathogens have been identified, most notably
vancomycin1,2 and methicillin resistant3 forms of Staph-
ylococcus aureus. With this prospect, new antibiotics
with novel modes of action have become a priority.
Our current efforts to establish new targets based on
bacteriophage inhibitory mechanisms have identified
several components of the DNA replication machinery
as attractive possibilities in this respect.4 Within this
machinery, DNA polymerase IIIC (DNA Pol IIIC),
encoded by the gene polC, has long been recognized as
essential for the replication of the host chromosome of
the low G+C content Gr+ bacteria.5,6 When the enzyme
is inhibited, the chromosomal DNA fails to replicate
resulting in bacterial death.7,8 Furthermore, it has little
homology with its mammalian counterpart.
enzyme, as demonstrated by Cozzarelli8 and Wright
and co-workers.11 The molecular recognition of these
inhibitors is believed to result from hydrogen-bonding
with a corresponding cytosine (dGTP analogs) on the
DNA template and the hydrophobic interaction of 6-
aminoaryl of the nucleobase with a pocket near the en-
zyme active site. From this rationalization, a new series
of inhibitors based on 6-hydroxy-2-(anilino)pyrimidi-
none scaffold 4 were designed. Much like the 6-anilino-
uracils, a single position in these pyrimidinediones
(position 5) is the primary position to be optimized in
terms of substitution. In this letter, efforts toward this
achievement are detailed.
O
O
HN
HN
H
N
O
N
H
N
H
O
N
H
N
H
1
2
To date, most inhibitors of DNA Pol IIIC were uracil
derivatives 1–3.9–13 These molecules inhibit Pol IIIC of
Bacillus subtilis by promoting the formation of a catalyt-
ically inactive ternary complex between DNA and the
O
OH
R
N
HN
O
N
H
N
H
O
N
H
N
H
Keywords: Antibacterial; Antimicrobial; DNA polymerase IIIC;
In vivo; Protein binding.
3
4
*
Corresponding authors. Tel.: +1 514 4967408; fax: +1 514
4964778; e-mail
targanta.com
A small library of 5-substituted-6-hydroxy-2-(arylami-
no)-pyrimidin-4(3H)-ones was initially prepared by con-
densation of an arylguanine hydrochloride 6 with a
substituted malonate 5 to yield the hydroxypyrimidinone
addresses:
yrose@targanta.com;
sciblat@
Present address: Neurochem Inc., 275 Armand-Frappier Blvd, Laval,
Que´bec., Canada H7V 4A7.
0960-894X/$ - see front matter ꢀ 2005 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2005.11.009

892
Y. Rose et al. / Bioorg. Med. Chem. Lett. 16 (2006) 891–896
7 (Scheme 1).14 The results for percent inhibition of S.
aureus DNA Pol IIIC are shown in Table 1.15 Taken from
this small library, compounds 8 and 9 having an allyl or a
benzyl substituents, respectively, displayed IC50s from 31
to 37 lM and exhibited reasonable antimicrobial activity
with MIC values lower than 16 lg/mL.
activity with an MIC of 8 lg/mL. In contrast, the
phenylbutyl 12 showed a drastic decrease in its enzy-
matic activity compared to the phenylpropyl 11 and
its antimicrobial potency was eliminated. The maxi-
mum length of the carbon chain was determined to
be three carbons between the phenyl group and the
pharmacophore. As we demonstrated, the length of
the side chain of the pyrimidinone is very important.
In order to determine the optimal length in the series
derived from 8, several compounds possessing different
alkyl chain lengths were evaluated as shown in Table
1. It was observed that as the chain length increases,
up to a 5-carbon chain (15), the DNA Pol IIIC inhibi-
tion also increased.
A systematic study of the elongation of the alkyl linker
between the phenyl group and the pharmacophore was
undertaken and the results are listed in Table 1. Elon-
gation of the alkyl chain from the benzyl group of 9 to
the phenylethyl 10 and phenylpropyl 11 showed an in-
crease in activity of 2- and 10-fold, respectively,
against DNA Pol IIIC, and a concurrent antibacterial
Unfortunately, the insolubility of the compounds in
aqueous media mirrored the increase in chain length.
To circumvent this problem, an oxygen atom was
introduced in the chain. The phenyl in the previous
set of molecules was replaced for a phenoxy group
(Table 1). The three modified compounds 17, 18, and
19 retained both their in vitro and antimicrobial activ-
ities. The optimum number of carbon atoms tolerated
with the phenoxy was again determined to be three.
The phenoxypropyl 18 was the most potent compound
in vitro with an IC50 of 4 lM and an MIC of 8 lg/mL.
The same avenue explored by Wright and co-workers
was attempted to increase the aqueous solubility of
the compounds by introducing amino or hydroxyl
groups on their side chains16,17 as shown with com-
pounds 20, 21, and 22. Their affinity toward the en-
zyme was not affected by the introduction of these
polar functionalities, but we observed total loss of
their antibacterial activity, unlike that reported for
the 6-anilinouracil series.16,17
OH
R2
R1
O
R2
R3
CO2Et
NH
NH2 HCl
N
R1
CO2Et
R3
N
H
N
N
H
H
5
6
7
Scheme 1. Reagents and condition: MeONa, MeOH, 125 ꢁC.
Table 1. In vitro DNA Pol IIIC inhibition and Staphylococcus aureus
MIC of 5-alkyl-6-hydroxy-2-(2,3-dihydro-1H-inden-6-ylamino)pyrim-
idin-4(3H)-ones
OH
R
N
O
N
N
H
H
Compound
R
% inhibitiona
65 (31)
MICb
16
8
Having these results in hand, we decided to focus on
derivatives from compound 18. Electron-donating
groups were first introduced on the phenol ether.
Although compounds 23 and 24 were slightly more
active than 18 (respectively, 83%, 81%, and 80% inhi-
bition at 100 lM compared to 74% for 2), the pres-
ence of an electron-donating group (at least in the
para position) was detrimental in terms of antibacteri-
al activity. On the other hand, the presence of a sim-
ple electron-withdrawing group appeared to be
promising (27 inhibits DNA Pol IIIC to 91% at
100 lM (IC50 = 5.6 lM)). Other derivatives were then
prepared as shown in Table 2. We were pleased to
see that a close analog of our lead structure showed
more potent inhibition but unfortunately it resulted
in less antibacterial activity (Table 2, compound 28,
IC50 = 3 lM, MIC = 16 lg/ml). Early in the study, it
was found that this series of compounds was difficult
to solubilize in aqueous media. To solve this problem
and to maintain the antibacterial activity, we prepared
compounds possessing a more polar electron with-
drawing group.
9
90 (37)
8
10
11
12
13
14
15
16
17
18
19
20
21
22
Phenylethyl
Phenylpropyl
Phenylbutyl
68 (18)
75 (3)
35
8
8
>128
8–16
8
CH2@CH–CH2–
CH2@CH(CH2)2–
CH2@CH(CH2)3–
CH2@CH(CH2)4–
Phenoxyethyl
65 (30.9)
87 (17.9)
89
16
81
16
74
16
Phenoxypropyl
Phenoxybutyl
74 (4)
67 (14)
40
8
8
3-Aminopropyl
3-Hydroxypropyl
N-Propyl-piperidine
>128
128
128
74
69
The inhibition of DNA Pol IIIC was retained in com-
pounds 29 and 31, but unfortunately, the introduction
of these functional groups led to a complete loss of anti-
bacterial activity (>128 lg/mL).
a % inhibition against Pol IIIC of S. aureus at 100 lM concentration of
inhibitor. Values in parentheses correspond to the IC50
b Minimum inhibitory concentration (lg/mL) in S. aureus RN4220.
.

Y. Rose et al. / Bioorg. Med. Chem. Lett. 16 (2006) 891–896
893
dine), and protein synthesis ([35S]-methionine).18 Panels
A and B demonstrate that compounds 8 and 9 inhibit
DNA synthesis of S. aureus without affecting the produc-
tion of either of the other macromolecules investigated,
which is consistent with DNA Pol IIIC inhibition.
Table 2. Effect of substitution on the 2-amino-5-alkene-6-hydroxy-
pyrimidine side chain
O
O
NH
R
HO
N
N
H
To ensure the exclusion of grossly toxic compounds, they
were also tested for their ability to inhibit the growth of
mammalian cells so as to ascertain potential cytotoxicity
in the host. This was monitored through both ATP and
MTS assays performed in 96-well microtiter plates.19 In
the ATP assay, the production of ATP was used as the
indicator of cell viability and in the MTS assay the pro-
duction of reducing equivalents asthe indicator of cell via-
bility. None of the tested compounds showed any
cytotoxicity in the MTS assay (not shown here).
Compound
R
H
% inhibitiona
MICb
18
23
24
25
26
27
28
29
30
31
74 (4.1)
83 (9)
81 (11.6)
70 (2.6–7.7)
72
8
p-NH2
p-OMe
p-NO2
>64
16
4–8
8
m-OMe, p-NO2
p-CN
91 (5.6)
95 (3)
90
8
16
o-OMe, p-CN
p-COOH
>128
64
>128
p-COOMe
p-Tetrazole
74
93
As was determined by ATP production (Fig. 2), most of
the pyrimidinones did not exhibit excessive toxicity (a
few selected compounds appear in this figure).
O2N
32
85 (5.1)
64
N
a % inhibition against Pol IIIC of S. aureus at 100 lM concentration of
In the case of 8, 28, and 33, the ATP production is unaf-
fected even at 100 lM. For compounds 11 and 34 the
ATP production decreases only at a concentration of
100 lM, whereas compound 18 appears more cytotoxic
as the ATP production is affected at 50 lM. The graph
in Figure 2 illustrates clearly that most of these com-
pounds are not cytotoxic at the levels used for the in vivo
test. Finally, before completing the in vivo test, we per-
formed an acute toxicity tolerability (ATT) test on three
mice (two doses of 50 mg of compound at a 4 h interval
were injected intraperitoneally in 40% HPCD or 10%
DMSO, 90% peanut oil). In all cases, we observed a
100% mouse survival.
inhibitor. Values in parentheses correspond to the IC50
.
b Minimum inhibitory concentration (lg/mL) in S. aureus RN4220.
In the case of 6-anilinouracils, Wright and co-work-
ers11b,11c had determined the 3-ethyl-4-methylphenyl
group to be the optimum substituent, closely followed
by the indanyl, on the 6-amino nitrogen. To confirm if
this tendency could be extrapolated to this series, the
most active compounds were synthesized as their 3-eth-
yl-4-methylphenyl derivatives (Table 3). Compounds 33,
34, and 35 with the 2-(3-ethyl-4-methylphenylamino)
substituent displayed a 4- to 9-fold increase in affinity
in vitro with respect to their counterparts 8, 9, and 28,
respectively. Unfortunately this increase in affinity did
not improve the antibacterial activity.
Given the potency of their antibacterial effects against
the Smith strain of S. aureus, we selected compounds
8, 28, and 33 for testing in an in vivo infection model
employing the same organism. Specifically, the in vivo
mouse peritonitis model was performed according to
the protocol described by Tarantino (S. aureus
ATCC13709 in PBS was used for the infection).16 The
results, expressed as the number of mice surviving at 3
days after infection, are summarized in Table 4. As
expected, all the animals treated with vancomycin sur-
vived through the 3 day period, while the 10 mice in
the control group (treated with vehicle alone) died.
In order to determine the mode of action of this newly syn-
thesized class of inhibitors, the cellular levels of DNA syn-
thesis in Gr+ bacteria were investigated. The results are
summarized in Figure 1. The compounds were tested
against three different metabolism pathways: DNA syn-
thesis (methyl-[3H]thymidine), RNA synthesis (5-[3H]uri-
Table 3. Effect of the 2-(3-ethyl-4-methylanilino) substituent of differ-
ent inhibitors
Concerning compound 8, 5 out of 10 mice survived the
infection, after two injections at 50 mg/kg, which is
promising given its relatively high IC50 (31 lM). When
the dose was decreased to a single injection at 50 mg/
kg, 4 out of 10 mice survived the infection. We were
pleased to see that all animals treated with 28 at a dose
of 2 · 50 mg/kg survived. Surprisingly, compound 33,
which showed good in vitro potency, was not effective.
Even though 28 and 33 had similar IC50s and MICs,
compound 28,which was more water soluble than 33,
due to the substituents on its aromatic moiety, displayed
better in vivo efficacy against S. aureus than 33. These
encouraging results suggest that the 2-amino-6-hydroxy-
pyrimidin-4-one possesses potential as a new class of
anti-infective compounds.
O
R
NH
HO
N
N
H
Compound
R
% inhibitiona
MICb
33
34
35
Benzyl
Allyl
94 (4)
90 (7.7)
81
8
8
(4-Cyano-2-methoxy
phenyl)propyl
16
a % inhibition against Pol IIIC of S. aureus at 100 lM concentration of
inhibitor. Values in parentheses correspond to the IC50
b Minimum inhibitory concentration (lg/mL) in S. aureus RN4220.
.

894
Y. Rose et al. / Bioorg. Med. Chem. Lett. 16 (2006) 891–896
A
B
160
140
120
100
80
160
140
120
100
80
DNA
60
40
20
0
DNA
60
RNA
RNA
40
Protein
Protein
20
0
0
5
2
5
1
2
4
8
0
5
1
2
4
8
32
64
16
64
25
0.
16
32
0.
0.
125
128
125
0.
128
0.
0.
Conc. (µg/ml)
Conc. (µg/ml)
Figure 1. Mode of action study by whole-cell dose–response labeling: (A) 2-(2,3-dihydro-1H-inden-6-ylamino)-5-allyl-6-hydroxypyrimidin-4(3H)-
one, 8; (B) 2-(2,3-dihydro-1H-inden-6-ylamino)-5-benzyl-6-hydroxypyrimidin-4(3H)-one, 9.
12.5µM
25µM
50µM
be seen as a vinylogous acid and, that would favor
binding to albumin. This issue needed to be addressed
by the selection of a proper bioisostere, which would
play the same role in the mechanism of inhibition
(hydrogen donor or acceptor) but would lack the acid
character which is probably the origin of the strong
serum binding.
120
100
80
60
40
20
0
100µM
We then explored a series of molecules where the 6-posi-
tion was substituted by different isosteres. For the first
set of modifications, the 6-hydroxyl group of 8 was re-
placed by functionalities that lacked acidic hydrogen.
Compound 36, having methyl ether at position 6, was
prepared by simple Fisher esterification in methanol cat-
alyzed by sulfuric acid of the corresponding 8. At this
position, the introduction of a hydrogen bond acceptor
group, for instance a methoxy group (36), was ineffec-
tive. The ether group might be sterically too large to
fit into the active site of the enzyme.
8
11
18
28
33
34
Figure 2. ATP production in the presence of increasing concentration
of the compound.
The so-called Ôfurry test tubeÕ serves as a less stringent
infection model in antibiotic testing. In this assay, each
animal is infected and treated at the same peritoneal
compartment. This model allows bypassing the issues
related with intravenous doses. Indeed, the problem
associated with the 2-amino-6-hydroxypyrimidinone
class is a strong in vitro tendency to bind to serum
albumin that severely decreased in antibacterial activi-
ty. This resulted in MIC values P128 lg/mL for most
of the compounds when measured in the presence of
50% human or mouse serum. A potential reason is
the presence of the 6-hydroxy functionality which can
We then designed another series of inhibitors where po-
sition 6 was substituted by a chloride. Compounds 37
and 38 were prepared from the corresponding 6-
hydroxypyrimidinones by refluxing them overnight in
the presence of phosphorus oxychloride and diethylani-
line in acetonitrile.20 The resulting 4,6-dichloropyrimi-
dines were then heated in a sealed tube with a mixture
of 50% aqueous NaOH and dioxane to furnish the de-
sired 6-chloropyrimidinones.21 The 6-chloroderivatives
37 and 38, substituted with 5-allyl and 5-phenylpropyl
groups, respectively, displayed fair in vitro inhibitions
and antimicrobial activities as shown in Table 5. Com-
pound 37 was very promising (IC50 of 2.3 lM and a
MIC of 1 lg/mL). The 6-chloroderivatives provided a
16-fold increase in potency as compared with 8 (IC50
of 31 lM and a MIC of 16 lg/mL). Unfortunately, in
the presence of 50% human serum, these inhibitors lost
all antimicrobial activity, a result suggestive of protein
binding (Table 5).
Table 4. Effect of some compounds on the lethal infection of mice by
Staphyloccocus aureusa
Compound
No. of survivors at
3 days postinfection
(total no. treated)
Vehicleb (40%HPCD)
Vancomycin (20 mg/kg)
0 (10)
10 (10)
5 (10)
4 (10)
10 (10)
1 (10)
8c
8d
28c
33d
Compound 39, being substituted by a 6-hydrogen and a
5-propyl, was obtained by simple catalytic reduction of
the compounds 37 with 10% palladium on charcoal un-
der a hydrogen atmosphere in a mixture of ethanol and
concentrated ammonium hydroxide.22 This compound
showed no activity. The presence of a substituent other
than hydrogen at position 5 seemed mandatory to main-
tain some DNA Pol IIIC inhibition.
aIntraperitoneal infection followed by a single or double injection of
drug or drug vehicle intraperitoneally.
b Saline for vancomycin.
c Injection of 2 · 50 mg/kg of drug at 15 and 60 min postinfection.
d Injection of 1 · 50 mg/kg of drug at 15 min postinfection.

Y. Rose et al. / Bioorg. Med. Chem. Lett. 16 (2006) 891–896
895
Table 5. In vitro DNA Pol IIIC inhibition and Staphylococcus aureus MIC of 2-(2,3-dihydro-1H-inden-6-ylamino)pyrimidin-4(3H)-ones
R2
R1
O
R3
R4
N
N
N
H
H
R1
R2
R3
R4
% dGTP inha
dATP IC50
MICc
MIC HSd
b
Compound
8
36
37
38
39
40
41
42
43
44
45
46
Allyl
Allyl
OH
MeO
Cl
–CH2–CH2–CH2
–CH2–CH2–CH2
–CH2–CH2–CH2
–CH2–CH2–CH2
–CH2–CH2–CH2
65 (31)
33
—
—
16
128
1
>128
>128
>128
>128
>128
>128
32
Allyl
53 (2.3)
46
—
—
Phenylpropyl
Propyl
H
Cl
H
16
7
—
—
>128
>128
16
NH2
NH2
NH2
NH2
NH2
NH2
NH2
–CH2–CH2–CH2–
Methyl
Methyl
5
36 (104)
H
Allyl
Ethyl
Ethyl
28
32 (164)
0
7.7
>100
7.9
>100
9.7
8
32
Phenylpropyl
Phenylpropyl
Benzyl
Benzyl
–CH2–CH2–CH2–
Methyl Ethyl
–CH2–CH2–CH2–
Methyl Ethyl
>128
>128
>128
>128
>128
>128
>128
>128
12
17
3
a % inhibition against Pol IIIC of S. aureus at 100 lM concentration of inhibitor. Values in parentheses correspond to the IC50
b IC50 (lM) in the Ôlow dATP assay.Õ
.
c Minimum inhibitory concentration (lg/mL) against S. aureus RN4220.
d Minimum inhibitory concentration (lg/mL) against S. aureus RN4220 in the presence of 50% human serum.
We then explored a third series of molecules where the 6-
position was substituted by a basic amino group pre-
sumably less likely to favor binding to serum proteins.
Synthesis of the 6-aminopyrimidinones 43 was achieved
by cyclocondensation of the appropriately substituted
ethyl cyanoacetate 47 on the arylguanine 6 in acetoni-
trile at 125 ꢁC as illustrated in Scheme 2. Unexpectedly,
compound 40 substituted at position 2 by a 2,3-dihydro-
1H-inden-5-amine did not display any in vitro DNA Pol
IIIC inhibition or antibacterial activity like the other
series. However, the 3-ethyl-4-methylaniline analog 41
exhibited an IC50 of 104 lM and an MIC of 16 lg/mL
(Table 5). Introduction of an allyl side chain at position
5 of this inhibitor (compound 39) increased the antibac-
terial activity (MIC 8 lg/mL) by 2-fold but decreased
the inhibitory potency toward the enzyme (164 lM).
Other 6-amino derivatives were prepared (43, 44, 45,
and 46) without improvement in the inhibitory or anti-
bacterial activities (Table 5). Interestingly, the 2,6-dia-
minopyrimidinone derivatives 41 and 42 were tested in
the presence of 50% human serum and retained their
antibacterial activities (MIC of 32 lg/mL). Further-
more, they did not show cytotoxicity in ATP and
MTS assays at 100 lM. The basic amino group at posi-
tion 6 is believed to be partially positively charged at
physiological pH and therefore resulted in reduced affin-
ity for serum proteins.
Pol IIIC enzyme led us to the hypothesis that these
inhibitors are dATP competitors and interact preferably
with thymidine residues as shown in Figure 3.
The IC50s of DNA Pol IIIC inhibitors were determined
in an in vitro DNA replication assay utilizing S. aureus
replication proteins. The results are shown in Table 5. In
this assay, duplication of singly primed circular ssM13
template by polC was dependent on the presence of
the processivity clamp b, the clamp loading complex
proteins dd0 and s, and single-stranded binding protein
(SSB). The assay conditions were adapted from those
described by Bruck et al.23 for use in Streptavidin-coated
FlashPlatesTM with [3H]dTTP as the marker. The assay
conditions were sensitized for dATP and dGTP analogs
by using reduced dATP or dGTP concentrations,
respectively.24 Inhibitor 41 and its parent 42 resulted
in IC50s of 28 and 7.7 lM, respectively. Compounds
43 and 45 substituted with phenylpropyl and benzyl,
respectively, retained their in vitro activities (IC50s of
7.9 and 9.7 lM) but demonstrated no antibacterial
activity. Conversely, parent compounds 44 and 46,
substituted with 2,3-dihydro-1H-inden-5-amine, com-
pletely lost their activities. In this series, a clear trend
can be noted; the affinity and selectivity conferred by
NH2
N
O
N
Structural relationships between the 2,6-diaminopyri-
midinone analogs and the active site model for DNA
(I)
(II)
R
O
R
NH
Ar
H
Ar
N
H
N
H
N
H
N
NH2
N
CO2Et
CN
H
H
N
H
N
6
N
N
O
O
O
H
O
N
H
N
H
N
H3C
47
43
Figure 3. (I) Hydrogen bonding of aminopyrimidinone with cytosine;
(II) hydrogen bonding of aminopyrimidinone with thymidine.
Scheme 2. Reagents and condition: CH3CN, 125 ꢁC.

896
Y. Rose et al. / Bioorg. Med. Chem. Lett. 16 (2006) 891–896
the 30-ethyl-40-methyl-aniline versus the 2,3-dihydro-1H-
inden-5-amine are essential to produce in vitro DNA Pol
IIIC inhibition as well as antimicrobial activity.
12. Ali, A.; Aster, S. D.; Grham, D. W.; Patel, G. F.; Taylor,
G. E.; Tolman, R. L.; Painter, R. E.; Silver, L. L.; Young,
K.; Ellsworth, K.; Geissler, W.; Harris, G. S. Bioorg. Med.
Chem. Lett. 2001, 11, 2185.
13. Ali, A.; Taylor, G. E.; Ellworth, K.; Harris, G.; Painter, R.;
Silver, L. L.; Young, K. J. Med. Chem. 2003, 46, 1824.
14. All new compounds showed purity >95% by HPLC at two
wavelengths (254 nm and 220 nm) and are characterized
In summary, 6-hydroxy-2-(anilino)pyrimidinones as a
novel non-nucleobase class of S. aureus DNA Pol IIIC
inhibitors were developed. From the SAR study, it was
determined that the activities within this series of com-
pounds were sensitive to the variation in the chain length
of the alkyl linker at position 5. These compounds were
found to have good in vitro and in vivo antibacterial
activities. Further structural modifications of the pyri-
midinone core may afford improved in vivo properties
leading to a compound suitable for the treatment of
infections. We have also developed novel 2,6-diamino-
pyrimidinone inhibitors of S. aureus DNA Pol IIIC.
From the SAR study, it was determined that the intro-
duction of an amino group at position 6 of the 2,6-diami-
1
by H NMR and HPLC–MS spectroscopy.
15. For the DNA Pol IIIC inhibition assay, a double-stranded
DNA oligonucleotide was used as the substrate. Assays
were completed in triplicate in a 25 lL volume (20 mM
Tris-HCl, pH 7.5, 4% glycerol, 0.1 mM EDTA, 5 mM
DTT, 40 lg/ml IgG, 8 mM MgCl2, 5 lM dATP, 5 lM
dCTP, 0.2 lM dGTP, 2.5 lM [3H]dTTP (1 Ci/mmol), and
0.4 lM PolC oligonucleotide substrate. Reactions were
initiated by the addition of 125 ng of purified PolC and
incubated at 37 ꢁC for 30 min. The reactions were
quenched by the addition of 20 mM EDTA (final concen-
tration) followed by spotting of the reaction mixtures on
DE81 filter paper. The filter papers were dried then
washed 4· with ꢀ200 ml 0.15 mM ammonium formate/
0.01 mM sodium phosphate solution, followed by one
wash in ꢀ200 ml of H2O. Filters were dried and then
radioactivity was determined by liquid scintillation count-
ing. The compounds were screened in triplicate at three
initial concentrations; 10, 50, and 100 lM. Compounds
with 70% inhibition or greater at 100 lM and/or MICs of
16 lg/mL or lower in S. aureus RN4220 were further
analyzed by a 12-point serial dilution, spanning the IC50
by approximately 2-fold.
nopyrimidinone
was
beneficial
for
retaining
antimicrobial activity in the presence of 50% human
serum. This novel series of inhibitors does not exhibit
cytotoxicity and selectively inhibits S. aureus DNA Pol
IIIC by competing with dATP. Further SAR studies on
this new series of inhibitors are currently under
investigation.
Acknowledgments
16. Tarantino, P. M.; Zhi, C.; Wright, G. E. Antimicrob.
Agents Chemother. 1999, 43, 1982.
17. Daly, J. S.; Giehl, T. J.; Brown, N. C.; Zhi, C.; Wright, G.
E., ; Ellison, R. T., III Antimicrob. Agents Chemother.
2000, 44, 2217.
The authors thank Francis Arhin for the MOA assay,
and Ramakrishnan Srikumar and Jothi Latha Krishna-
moorthy for the MIC determination.
18. Norfloxacin, which selectively inhibits DNA synthesis and
Rifampicin, which inhibits both protein and RNA syn-
thesis of bacteria, were used as control compounds.
19. Cryopreserved primary human hepatocytes were used in
the ATP assay and cultured HeLa cells were used in the
MTS assay. For the ATP assay, compounds at 100, 50, 25,
and 12.5 lM were incubated with 1 · 104 primary human
hepatocytes per well in Krebs–Henseleit buffer for 2 h at
37 ꢁC under 5% CO2. At the end of the incubation, the
ATP content was determined by the addition of luciferin
and luciferase, and monitored by luminescence. For the
MTS assay, compounds at 100, 50, 25, and 12.5 lM were
incubated with 2 · 104 HeLa cells per well in DulbeccoÕs
modified EagleÕs Medium containing 1% bovine growth
serum for 18 h at 37 ꢁC under 5% CO2. At the end of the
incubation, the amount of reducing equivalents was
determined by the reduction of MTS reagent to a
formazan product as revealed by absorbance at 490 nm.
20. Siddiqi, S. M.; Jacobson, K. A.; Esker, J. L.; Olah, M. E.;
Ji, X.; Melman, N.; Tiwari, K. N.; Secrist, J. A., III;
Schneller, S. W.; Cristalli, G.; Stiles, G. L.; Johnson, C.
R.; IJzerman, A. P. J. Med. Chem. 1995, 38, 1174.
21. Burgdorf, L. T.; Carell, T. Chem. Eur. J. 1998, 8, 293.
22. Chou, S.; Yang, P.; Wang, C.; Lu, H.; Chen, Y.; Kao, J. J.
Chin. Chem. Soc. 1999, 46, 53.
References and notes
1. (a) Hiramatsu, K.; Hanaki, H.; Ino, Y.; Yabuta, K.;
Oguri, T.; Tenover, F. C. J. Antimicrob. Chemother. 1997,
40, 135; (b) Hiramatsu, K.; Aritaka, N.; Hanaki, H.;
Kawasaki, S.; Hosada, Y.; Hori, S.; Fukushi, Y.; Kobay-
ashi, I. Lancet 1997, 1670.
2. Tomasz, A. N. Engl. J. Med. 1994, 330, 1247.
3. Michel, M.; Gutmann, L. Lancet 1997, 349, 1901.
4. Liu, J.; Dehbi, M.; Moeck, G.; Arhin, F.; Bauda, P.;
Bergeron, D.; Callejo, M.; Ferretti, V.; Ha, N.; Kwan, T.;
McCarty, J.; Srikumar, R.; Williams, D.; Wu, J. J.; Gros,
P.; Pelletier, J.; DuBow, M. Nat. Biotechnol. 2004, 22, 185.
5. Amjad, A.; Taylor, G. E. Expert Opin. Ther. Patent 2005,
15, 947.
6. (a) Kornberg, A.; Baker, T. A. DNA Replication; W.H.
Freeman: New York, 1992; (b) Barnes, M. H.; Miller, S.
D.; Brown, N. C. J. Bacteriol. 2002, 184, 1824.
7. Brown, N. C.; Handschumacher, R. E. J. Biol. Chem.
1966, 241, 3083.
8. Cozzarelli, N. R. Annu. Rev. Biochem. 1977, 46, 641.
9. Wright, G. E.; Brown, N. C. J. Med. Chem. 1977, 20,
1181.
23. Bruck, I.; Georgescu, R. E.; OÕDonnell, M. J. Biol. Chem.
2005, 280, 18152.
10. Brown, N. C.; Gambino, J.; Wright, G. E. J. Med. Chem.
1977, 20, 1186.
24. To validate our Ôlow dATP assay,Õ some of our previous
inhibitors were tested in Ôlow dGTP assayÕ to determine their
IC50. The obtained values were coherent with the ones
reported.
11. (a) Wright, G. E.; Brown, N. C. J. Med. Chem. 1980, 23,
34; (b) Wright, G. E.; Gambino, J. J. J. Med. Chem. 1984,
27, 181; (c) Trantolo, D. J.; Wright, G. E.; Brown, N. C. J.
Med. Chem. 1986, 29, 676.