Design and synthesis of novel antibacterial agents with inhibitory activity against DNA polymerase III

Article abstract of DOI:10.1016/S0960-894X(01)00407-3

4-Substituted 2-amino-6-(anilino)pyrimidines have been found to be selective inhibitors of DNA polymerase III, a replicative enzyme known to be essential in the DNA synthesis of Gram-positive bacteria. Among the analogues, 18 displayed an IC₅₀ of 10 μM against DNA polymerase III from Staphylococcus aureus.

Full text of DOI:10.1016/S0960-894X(01)00407-3

Bioorganic & Medicinal Chemistry Letters 11 (2001) 2185–2188
Design and Synthesis of Novel Antibacterial Agents with
Inhibitory Activity against DNA Polymerase III
Amjad Ali,a,* Susan D. Aster,a Donald W. Graham,a Gool F. Patel,a Gayle E. Taylor,a
Richard L. Tolman,a,y Ronald E. Painter,b Lynn L. Silver,b Katherine Young,b
Kenneth Ellsworth,c Wayne Geisslerc and Georgianna S. Harrisc
aDepartment of Medicinal Chemistry, Merck Research Laboratories, PO Box 2000, Rahway, NJ 07065-0900, USA
bDepartment of Infectious Diseases, Merck Research Laboratories, PO Box 2000, Rahway, NJ 07065-0900, USA
cDepartment of Endocrinology and Chemical Biology, Merck Research Laboratories,
PO Box 2000, Rahway, NJ 07065-0900, USA
Received 8 March 2001; accepted 4 June 2001
Abstract—4-Substituted 2-amino-6-(anilino)pyrimidines have been found to be selective inhibitors of DNA polymerase III, a
replicative enzyme known to be essential in the DNA synthesis of Gram-positive bacteria. Among the analogues, 18 displayed an
IC50 of 10 mM against DNA polymerase III from Staphylococcus aureus. # 2001 Elsevier Science Ltd. All rights reserved.
The last decade of the 20th century has seen an alarming
rise in the incidence of multi-antibiotic resistant strains
of pathogenic Gram-positive (Gr+) bacteria, including
vancomycin resistant Staphylococcus aureus.1,2 Curtail-
ing this rise will depend largely on the development of
chemotherapeutic agents capable of selectively attacking
novel bacterial targets. One such target that has received
considerable attention over recent years is DNA poly-
merase III (Pol III), an enzyme known to be essential in
the replicative DNA synthesis of Gr+ bacteria.3,4 We
have recently synthesized and characterized a series of
4-substituted 2-amino-6-(anilino)pyrimidine based inhi-
bitors of Pol III with the goal of identifying novel,
potent and selective antibacterial agents. In this Letter,
we detail our initial efforts in this area.
tial features of the mechanism of complex formation
have been worked out by Cozzarelli9 and by Wright10
and involve: (1) hydrogen bonding of the uracil and its
6-NH moiety with cytosine in the DNA template; and
(2) binding of the phenyl group with a hydrophobic site
on the enzyme. These early prototype inhibitors were
unattractive candidates for further development since
they were insoluble, simple uracil derivatives, many of
which contained objectionable functionality such as the
hydrazino group. In an attempt to overcome some of
these limitations, we decided to replace the 6-anilinour-
acil moiety with a 4-substituted 2-amino-6-(anilino)pyr-
imidine subunit, a hitherto unexplored area. We
postulated that introduction of a lipophilic 4-substituent
such as a substituted phenoxy group would aid solubi-
lity whilst maintaining a degree of activity towards the
Pol III target. Our design strategy for selecting the 2-
amino-6-(anilino) pyrimidine platform was based on the
pioneering work of Wright and co-workers. In his stud-
ies, Wright8 has shown that 6-anilinouracil derivatives
such as 3 hydrogen bond with cytosine residues in the
DNA template and that their action is competitive with
20-deoxyguanosine-50-triphosphate (dGTP) (Fig. 1, panel
A). We reasoned that 4-substituted 2-amino-6-(anilino)
pyrimidines (e.g., 4) should also retain the capacity to
form hydrogen bonds, albeit with thymine bases, and
would therefore be competitive with 20-deoxyadenosine-
50-triphosphate (dATP) (Fig. 1, panel B).
The potential utility of Pol III inhibitors as antibacterial
agents was first described by Langley in 1962.5 The
prototype inhibitors [e.g., 6-(phenylhydrazino)uracils
1,6 6-(benzylamino)uracils 2,7 and 6-anilinouracils 38]
were shown to inhibit Pol III of Bacillus subtilis by
promoting the formation of a catalytically inactive
ternary complex with DNA and the enzyme. The essen-
*Corresponding author. Fax: +1-732-594-9545; e-mail: amjad_ali@
merck.com
yCurrent address: Geron Corporation, 555 Constitution Drive, Menlo
Park, CA 56980, USA.
0960-894X/01/$ - see front matter # 2001 Elsevier Science Ltd. All rights reserved.
PII: S0960-894X(01)00407-3

2186
A. Ali et al. / Bioorg. Med. Chem. Lett. 11 (2001) 2185–2188
Scheme 1. (a) 3,4-Disubstituted anilines, EtOH, 60 ꢀC; (b) NaSH,
(CH2OH)2, 100 ꢀC; (c) substituted phenols, Cs2CO3, DMF, 120 ꢀC.
Our initial studies focused on determining whether the
2-amino-6-(anilino)pyrimidine platform would be viable
for inhibitor design. Although we were confident that
the 2-amino-6-(anilino) pyrimidine platform would
impart inhibitory activity against Pol III according to
our working model depicted in B, the extent of this
activity was unclear. Thus, a series of 2-amino-4-chloro-
6-(anilino)pyrimidines was initially prepared by reaction
of 2-amino-4,6-dichloropyrimidine with the appro-
priately substituted aniline, according to Scheme 1 (path
a).11 The resulting 2-amino-4-chloro-6-(anilino)pyr-
imidine analogues were shown to be weakly active Pol
III inhibitors with IC50 values in the 50–120 mM range.12
A number of these analogues also displayed encoura-
ging levels of antibacterial activity against several spe-
cies of relevant Gr+ bacteria (Table 1) and were
uniformly inactive against an outer-membrane perme-
able strain of Escherichia coli, a Gram-negative Pol III
deficient organism.13
Figure 1. (A) Hydrogen bonding of a 6-aminouracil with cytosine; (B)
hydrogen bonding of a 2-amino-6-(anilino)pyrimidine with thymine.
Each compound was tested versus five different labeled
precursors to investigate the effect on DNA synthesis
(6-3H thymidine), RNA synthesis (5,6-3H uracil), pro-
tein synthesis (4,5-3H leucine), phospholipid synthesis
(2-3H glycerol) and cell wall synthesis (2,3-3H alanine). As
shown in Figure 2, panel D, the 2-amino-4-chloro-6-(30-
ethyl-40-methyl)anilinopyrimidine derivative 11 selectively
inhibits DNA synthesis of S. aureus without affecting any
of the other macromolecules investigated. Panels E and F
confirm these findings for 2-amino-4-chloro-6-(30-
bromo-40-methyl)anilinopyrimidine 6 and 2-amino-4-
chloro-6-(30,40-dimethyl)anilinopyrimidine 9, respectively.
The ability of these compounds to selectively inhibit the
DNA synthesis of Gr+ bacteria was also investigated
by whole-cell, dose–response labeling.14 Vancomycin,
rifampicin and ciprofloxacin, which selectively inhibit
bacterial cell wall, RNA synthesis and DNA synthesis,
respectively, were run as control compounds. The
results are summarized in Figure 2.
Variation of substituents in the 4-position of the 2-
amino-6-(anilino)pyrimidines (paths b and c, Scheme 1)
gave mixed results.15 Data for inhibition of Pol III by a
series of 2-amino-4-substituted-6-(anilino)pyrimidines
are presented in Table 2.
Table 1. In vitro anti-Pol III and antibacterial activity of 2-amino-4-chloro-6-(anilino)pyrimidines
MIC (mg/mL)a
Compd
R1
R2
IC50 (mM)
B. subtilis ATCC 6633
E. faecium RLAId
S. aureus MB2985e
S. pneumonia CL5350e
5
6
7
8
I
Br
Cl
Cl
CH3
CH3
CH3
Cl
51
94
126
98
115
119
53
5
8
13
7
6
3
36
31
27
>36b
31
27
29
>25b
>26b
>26b
7
18
31
27
>29c
>25c
>26c
26
29
9
CH3
CH3
>25b
>26b
13
10
11
12
30,40-(CH2)3–
CH2CH3
H
CH3
tBu
3
7
89
14
14
aMinimum inhibitory concentration (mg/mL).
bGrowth unaffected at the highest level tested.
cGrowth affected but incomplete inhibition at the highest level tested.
dRLAI (ampicillin and imipenem-resistant) was obtained from Dr. Barbara Murray.
eMB2985 (the Staph Smith strain) and CL5350 are from Merck collections.

A. Ali et al. / Bioorg. Med. Chem. Lett. 11 (2001) 2185–2188
2187
Figure 2. Mode of action study by whole-cell dose–response labeling. (A) vancomycin; (B) rifampicin; (C) ciprofloxacin; (D) 2-amino-4-chloro-6-(30-
ethyl-40methyl)anilinopyrimidine 11; (E) 2-amino-4-chloro-6-(30-bromo-40-methyl)anilinopyrimidine 6; (F) 2-amino-4-chloro-6-(30,40-dimethyl)anili-
nopyrimidine 9.
Table 2. In vitro anti-Pol II activity of 2-amino-4-substituted-6-(ani-
lino)pyrimidines
The tautomeric 4-thio derivative 1316 was found to be 2-
fold less active than the prototype 4-chloro analogue 5.
Etherification of the 4-thio compounds decreased activ-
ity in general (e.g., 14 and 15), as did replacement of the
4-chloro moiety by electron-releasing 4-methyl and 4-
N,N-dimethylamino groups; all these derivatives were
Compd
X
30
40
% Inhibitiona
essentially inactive. The substituted 4-phenoxy analo-
gues (path c, Scheme 1) gave interesting results. In gen-
eral, the phenoxy derivatives were weaker inhibitors
than the parent 4-chloro compounds. However, 3- or 4-
substituted phenoxy analogues (e.g., 18 and 21) often
exhibited improved activity. We found that small
hydrophobic groups in the 3- and 4-positions of the
phenoxy ring gave compounds equivalent to or better
than the parent 4-chloro inhibitors. Indeed, the p-
bromo analogue 18 was 5-fold more active than the
prototype 4-chloro compound. An IC50 determination
of 18 produced a value of 10 mM, which represented our
13
14
15
16
17
18
19
20
21
SH
SCH3
SC6H5
I
I
I
I
CH3
CH3
CH3
CH3
CH3
CH3
CH3
CH3
CH3
54
20
2
21
CH3
N(CH3)2
4-Br-C6H4O
4-Cl-C6H4O
4-F-C6H4O
3-CH3-C6H4O
CH3
C2H5
C2H5
C2H5
C2H5
27
97 (10 mM)b
13
33
88 (54 mM)b
a% Inhibition against Pol III of S. aureus at 500 mM concentration of
inhibitor.
bIC50 value.

2188
A. Ali et al. / Bioorg. Med. Chem. Lett. 11 (2001) 2185–2188
best Pol III inhibitor. Interestingly, close analogues of
18 (e.g., 19 and 20) were surprisingly poor inhibitors of
Pol III. With respect to antibacterial activity, the 4-
phenoxy derivatives generally demonstrated dis-
appointing levels of potency against our panel of gram-
positive organisms (MICs were in the ꢁ20 to >40 mg/
mL range). The reasons behind this apparent dis-
crepancy between affinity for the Pol III target in vitro
and growth of its host bacterial cell in vivo are not
known, but could be due to permeability barriers or to
differences between the strain used for whole-cell work
versus that used for cloning the enzyme. Such dis-
crepancies between in vitro versus whole-cell activity
have been noted by other researchers for different Pol
III inhibitors.3b
3. (a) Tarantino, P. M.; Zhi, C.; Wright, G. E.; Brown, N. C.
Antimicrob. Agents Chemother. 1999, 43, 1982. (b) Taratino,
P. M.; Zhi, C.; Gambino, J. J.; Wright, G. E.; Brown, N. C. J.
Med. Chem. 1999, 42, 2035.
4. (a) Cozzarelli, N. R.; Law, R. L. Biochem. Biophys. Res.
Comm. 1973, 51, 151. (b) Gass, K. B.; Low, R. L.; Cozzarelli,
N. R. Proc. Natl. Acad. Sci. U.S.A. 1973, 70, 103.
5. Langley, B. W. Imperial Chemical Industries Ltd. British
patent 890,076, 1962.
6. Wright, G. E.; Brown, N. C. J. Med. Chem. 1977, 20, 1181.
7. Brown, N. C.; Gambino, J.; Wright, G. E. J. Med. Chem.
1977, 20, 1186.
8. Wright, G. E.; Brown, N. C. J. Med. Chem. 1980, 23, 34.
9. Cozzarelli, N. R. Annu. Rev. Biochem. 1977, 46, 641.
10. Brown, N. C.; Wright, G. E. Pharmac. Ther. 1977, A1,
437.
11. All new compounds showed satisfactory NMR spectro-
scopic and mass spectrometry data.
In summary, we have discovered a novel series of Pol III
inhibitors based on the 2-amino-6-(anilino)pyrimidine
platform. We have shown that small hydrophobic sub-
stituents on the 3- and 4-positions of the anilino group
combined with 4-halosubstituted phenoxy substituents
(e.g., 18) lead to increases in the anti-Pol III activities of
these Gram-positive selective inhibitors.
12. The enzyme inhibition assay was performed at 30 ꢀC for
20 min in 0.1 mL reaction mix (5% final DMSO concentration
of inhibitor solutions) containing 30 mM Tris HCl, pH 7.5,
20% glycerol, 4 mM dithiothreitol, 10 mM magnesium acetate,
0.003 mM dATP, 0.003 mM dGTP, 0.001 mM dCTP,
0.001 mM 3H-dTTP, and 0.35 mg/mL activated calf thymus
DNA purchased from Worthington Enzymes. S. aureus DNA
polymerase was cloned from a conveniently available source
of chromosomal DNA, strain R27.
13. Wright, G. E.; Brown, N. C. Pharmac. Ther. 1990, 47, 447.
14. Compounds which were active in the MIC assay were
tested for the mode of action (MOA) by whole-cell dose–
response labeling. All labeled precursors were purchased either
from Amersham or New England Nuclear.
15. For SARs in a related series see: Trantolo, D. J.; Wright,
G. E.; Brown, N. C. J. Med. Chem. 1986, 29, 676.
16. Noell, C. W.; Robins, R. K. J. Med. Pharm. Chem. 1962,
5, 558.
References and Notes
1. (a) Hiramatsu, K.; Hanaki, H.; Ino, T.; Yabuta, K.; Oguri,
T.; Tenover, F. C. J. Antimicrob. Chemother. 1997, 40, 135. (b)
Hiramatsu, K.; Aritaka, N.; Hanaki, H.; Kawasaki, S.; Hos-
oda, Y.; Hori, S.; Fukuchi, Y.; Kobayashi, I. Lancet 1997,
1670.
2. Tomasz, A. N. Engl. J. Med. 1994, 330, 1247.