Triaminotriazine DNA helicase inhibitors with antibacterial activity

Article abstract of DOI:10.1016/j.bmcl.2005.11.076

Screening of a chemical library in a DNA helicase assay involving the Pseudomonas aeruginosa DnaB helicase provided a triaminotriazine inhibitor with good antibacterial activity but associated cytotoxicity toward mammalian cells. Synthesis of analogs provided a few inhibitors that retained antibacterial activity and demonstrated a significant reduction in cytotoxicity. The impact of serum and initial investigations toward a mode of action highlight several features of this class of compounds as antibacterials.

Full text of DOI:10.1016/j.bmcl.2005.11.076

Bioorganic & Medicinal Chemistry Letters 16 (2006) 1286–1290
Triaminotriazine DNA helicase inhibitors with antibacterial activity
Geoffrey A. McKay,a Ranga Reddy,a Francis Arhin,a Adam Belley,a Dario Lehoux,a
Greg Moeck,a Ingrid Sarmiento,a Thomas R. Parr,a Philippe Gros,b
Jerry Pelletierb and Adel Rafai Fara,*
aTarganta Therapeutics Inc., 7170 Frederick Banting, 2nd Floor, St. Laurent, Que´bec, Canada H4S 2A1
bDepartment of Biochemistry, 3655 Promenade Sir William Osler, McIntyre Medical Sciences Building,
McGill University, Montreal, Que´bec, Canada, H3G 1Y6
Received 24 October 2005; accepted 18 November 2005
Available online 15 December 2005
Abstract—Screening of a chemical library in a DNA helicase assay involving the Pseudomonas aeruginosa DnaB helicase provided a
triaminotriazine inhibitor with good antibacterial activity but associated cytotoxicity toward mammalian cells. Synthesis of analogs
provided a few inhibitors that retained antibacterial activity and demonstrated a significant reduction in cytotoxicity. The impact of
serum and initial investigations toward a mode of action highlight several features of this class of compounds as antibacterials.
Ó 2005 Elsevier Ltd. All rights reserved.
Resistance to antibacterials is a public health issue of
increasing concern that is compounded by the inability
of the pharmaceutical industry to generate new classes
of antibiotics to combat infections.1 To tackle this prob-
lem, researchers have in recent years increased their fo-
cus on the identification of new targets and on the
exploitation of targets beyond those against which mar-
keted antibiotics have been rendered ineffective by the
development of resistance.2 From this perspective,
DNA helicases, enzymes involved in the nucleoside tri-
phosphate dependent unwinding of duplex DNA into
single strands,3 an essential activity for replication,4
appear to be attractive targets. Whereas replicative heli-
cases require assistance from helicase loaders to assem-
ble at chromosomal origins of replication,5 some
helicases6 including the replicative helicase DnaB of
Pseudomonas aeruginosa7 are able to unwind model oli-
gonucleotide substrates in their absence, providing a
convenient in vitro assay to study helicase mechanism
and inhibition.
antiinfectives against this bacterium, a library of
230,000 commercially available compounds was
screened against the DnaB helicase in a functional assay
measuring the unwinding of a short DNA partial du-
plex.10 Amongst the resulting active compounds, the
triaminotriazine 1 (Fig. 1) was identified as a low
micromolar inhibitor of the enzyme.
Compound 1 does not display antibacterial activity
against Gram-negative organisms (P. aeruginosa or
Escherichia coli), but has moderate activity against
Gram-positive organisms (Staphylococcus aureus, Strep-
tococcus pneumoniae) (Table 1) and even Candida albi-
cans (data not shown). This could be a result of the
high degree of homology between Gram-positive and
Gram-negative helicases (41% identity and 62% similar-
ity between the P. aeruginosa and the S. aureus heli-
cases). It does however also result in significant
cytotoxicity against HeLa cells as determined by an
MTS assay (Table 2).11
The increased recognition of P. aeruginosa as a key fac-
tor in nosocomial infections8 as well as a major patho-
gen in cystic fibrosis patients9 is compounded by the
very robust nature of this microbe. To discover new
NH
Cl
N
N
O
N
N
N
N
H
H
Keywords: Antibacterial; Cytotoxic; Helicase.
*
1
Corresponding author. Tel.: +1 514 4966667; fax: +1 514 3326033;
e-mail: afar@targanta.com
Figure 1.
0960-894X/$ - see front matter Ó 2005 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2005.11.076

G. A. McKay et al. / Bioorg. Med. Chem. Lett. 16 (2006) 1286–1290
Table 1. Inhibitory and antibacterial activities of triaminotriazines 1–15
1287
NH
N
Y
N
R1
N
N
N
H
R2
Compound
X-
Y-
IC50 (lM)
MIC (lg/mL)
1
N
R2
S. aureus RN4220
E. coli LBB925
P. aeruginosa Z61
O
O
O
O
O
O
O
N
N
N
N
N
N
N
1
2
4-NO2
H-
4-Cl
5
>100
20
4
>128
nd
>128
nd
N
H
H-
nd
N
H
3
4-Cl
4-Cl
4
>128
16
>128
>128
nd
N
H
4
4-Cl
H-
65
64
N
H
5
4-OMe
4-F
4-OMe
4-F
>100
>100
19
nd
nd
N
H
6
nd
nd
nd
N
H
7
3,4-Di F
4-F
3,4-Di F
4-F
16
16
>128
>128
>128
>128
>128
>128
>128
>128
>64
N
H
N
8
22
32
16
N
N
N
9
4-F
4-F
23
64
32
OH
N
N
10
11
12
13
14
15
3,4-Di-F
3,4-Di-F
3,4-Di-F
3,4-Di-F
3-Cl
3,4-Di-F
3,4-Di-F
3,4-Di-F
3,4-Di-F
3-Cl
18
16
16
N
N
OH
5
>128
>128
4
>128
>128
>128
>128
>64
OPh
N
N
6
N
N
Ph
5
31
>128
>64
N
H
Cl
4-Br
4-Br
6
N
H
MIC: minimum inhibitory concentration.
nd: not determined.
There is no inhibition of mammalian DNA replication
in the presence of up to 100 lM of 1. Given the impact
of DNA binders on the activity of helicases,12 binding to
DNA was measured in a dye displacement assay.13
DNA binding by triaminotriazine 1 was found to be
weak relative to distamycin A (minor groove binder)
and m-AMSA (intercalator) controls, suggesting an
alternate mechanism for its inhibitory activity on DnaB.
bacterial activity, but with reduced cytotoxicity, a num-
ber of analogs were prepared. This can be done
from cyanuryl chloride by the selective stepwise dis-
placement of each of the three chlorine atoms by an
amine (Scheme 1).
The inhibitory activities of these compounds toward the
P. aeruginosa DnaB helicase were measured and com-
pared to their antibacterial activities14 against a wild-
type strain of S. aureus (RN4220) as well as against
hypersusceptible strains of E. coli (LBB925, tolC
In an effort to establish whether the scaffold presented in
1 can be used to generate helicase inhibitors with anti-

1288
G. A. McKay et al. / Bioorg. Med. Chem. Lett. 16 (2006) 1286–1290
Table 2. ClogD (calculated logP value adjusted to pH 7.40),
antibacterial activities, and cytotoxicities (measured by the MTS
method over three concentrations) for selected compounds
in compounds 14 and 15 which lack the aliphatic side
chains of 1–13, yet maintain respectable levels of inhibi-
tion. They do not display antibacterial activity, a mea-
sure of the role of an aliphatic chain with a tertiary
amino group for cellular activity.
Compound
ClogD
(pH 7.40)
MIC (lg/mL)
S. aureus
Cytotoxicity, lM
(% survival)
12.5
25
50
A comparison of the MICs against S. aureus, the cyto-
toxicities as established through the MTS assay, and
the ClogD (calculated logP adjusted for pH)15 is pre-
sented in Table 2. MICs and cytotoxicities mirror each
other fairly consistently, with compounds resulting in
both antibacterial activity and cytotoxicity, or not dis-
playing either. The exceptions are compounds 7, 10,
and 13 all of which bear the 3,4-difluorophenyl moieties
as the two aryls on the triazine core and whose antibac-
terial activities are accompanied by a noticeable
decrease in cytotoxicity.
1
3
3.36
4.46
3.25
3.53
4.66
2.88
2.88
5.58
5.65
3.95
6.76
4
4
78
103
98
19
10
6
90
74
4
64
16
14
47
11
51
47
99
97
86
64
7
101
100
101
80
91
64
8
64
64
9
104
91
10
11
12
13
14
16
>128
>128
4
97
96
100
99
99
74
99
83
>128
It is clear that the four fluorine substituents impart selec-
tive activity toward the bacteria. The more hydrophobic
compounds 11 and 12 are neither antibacterial nor cyto-
toxic, a matter that could result from their inability to
penetrate the cells efficiently either as a consequence of
reduced diffusion across the membrane and/or cell wall,
increased efflux or the inability to sustain a sufficient
intracellular concentration.16 This is mirrored by com-
pound 14 both in terms of hydrophobicity and bioactiv-
ities. The role of the hydrophobic character of the
molecules must not however be overstated, as the com-
parison between compounds 9 and 10 reveals that two
compounds of similar levels of inhibition and hydro-
phobicities result in different antibacterial activities. This
also points to the favorable role of the 3,4-difluoro-
phenyl substituents.
NH
Cl
N
a, b, c
N
N
N
R2
H
R3
R4
N
N
N
Cl
N
Cl
1-15
Scheme 1. Reagents and conditions: (a) R1NH2, DIPEA, THF, 60 °C;
(b) R2NH2, DIPEA, THF, 60 °C; (c) R3R4NH, acetone, room
temperature, for aliphatic amines, R3R4NH DIPEA, THF, 60 °C,
for aromatic amines.
deficient) and P. aeruginosa (Z61, permeability barrier
deficient) (Table 1).
These data establish the correlation between the elec-
tron-withdrawing substituents on the aryl rings of 1
and its inhibitory activity. Indeed, compounds 2 (lacking
substituents) and 5 (bearing electron-rich substituents)
are completely inactive, whereas the replacement of the
nitro group in 1 by a chlorine atom (compound 3)
results in a reasonable inhibitory activity. The replace-
ment of the same nitro group with a hydrogen atom
(compound 4) brings about an intermediate activity in
a consistent manner. Surprisingly, substitution of each
of the aryl groups with a fluorine atom (compound 6) re-
sults in a completely inactive compound, whereas the
parent molecule bearing four fluorine atoms (compound
7) remains capable of inhibition. Substituting the (2-
morpholinoethyl)amino group in 6 by a piperazine re-
stores the inhibitory activity (compound 8). The use of
four fluorine atoms and substituted piperazines results
in respectable inhibitory activities and the more hydro-
phobic substituents on the piperazine result in a slightly
improved inhibition. In terms of antibacterial activities,
the MIC values against S. aureus seem to match the
inhibitory activity against the helicase with the exception
of the more hydrophobic compounds 11 and 12. The
activity against E. coli responds positively to the intro-
duction of the fluorine atoms and the piperazine moiety.
Yet, no activity is detected against P. aeruginosa.
The antibacterial activities of these compounds were
further characterized through the ability of compounds
1 and 3 to inhibit macromolecular biosynthesis in
S. aureus (Fig. 2).17 The expected inhibition of DNA
synthesis, given the role of the DnaB helicase in repli-
cation, is accompanied by a decrease in transcription
and translation as well. This contrasts with other
inhibitors of DNA replication such as DNA gyrase
inhibitors like the fluoroquinolone antibiotics, which
selectively impact the biosynthesis of DNA. Given
the lack of precedence of well-defined helicase inhibi-
tors,18 the meaning of this unselective behavior is
difficult to gauge. The lack of selectivity toward the
inhibition of DNA synthesis may suggest an impact
on the uptake or retention of the labeled macromole-
cule precursors, as has been noted for inhibitors of
histidine protein kinases that were found to act via
membrane disruption.19 Alternatively, since the shal-
low slope of the concentration-dependent decrease in
synthesis of DNA, RNA, and protein by compounds
1 and 3 contrasts somewhat from the sigmoidal curves
generally associated with well-defined inhibitors,
antibacterial activity could be the result of the
conjunction of weak inhibitory activities over more
than one cellular target. Such a phenomenon has been
reported to complicate the predictions of mechanisms
of action of novel antibacterial agents by gene
expression profiling.20
The role of the electron withdrawing substituents in
bringing about inhibitory activity is further highlighted

G. A. McKay et al. / Bioorg. Med. Chem. Lett. 16 (2006) 1286–1290
1289
140
120
100
80
120
100
80
60
40
20
0
DNA
RNA
60
Protein
40
20
0
0.01
0.1
1
10
100
0.01
0.1
1
10
100
Concentration (ug/mL)
Concentration (ug/mL)
Figure 2. Inhibition of macromolecular biosynthesis by compounds (A) 1 and (B) 3.
The promiscuity of these inhibitors is further supported
by the impact of serum on the MIC against S. aureus
References and notes
1. (a) Overbye, K. M.; Barrett, J. F. Drug Discovery Today
2005, 10, 45; (b) Levy, S. B. Adv. Drug Delivery Rev. 2005,
57, 1446.
2. (a) Black, M. T.; Hodgson, J. Adv. Drug Delivery Rev.
2005, 57, 1528; (b) Brown, E. D.; Wright, G. D. Chem.
Rev. 2005, 105, 759.
3. (a) Egelman, E. H. J. Struct. Biol. 1998, 124, 123; (b)
Tuteja, N.; Tuteja, R. Eur. J. Biochem. 2004, 271, 1835; (c)
Caruthers, J. M.; McKay, D. B. Curr. Opin. Struct. Biol.
2002, 12, 123.
4. For an example of the essentiality of the helicase in S.
aureus: Liu, J.; Dehbi, M.; Moeck, G.; Arhin, F.; Bauda,
P.; Bergeron, D.; Callejo, M.; Ferretti, V.; Ha, N.; Kwan,
T.; McCarthy, J.; Srikumar, R.; Williams, D.; Wu, J. J.;
Gros, P.; Pelletier, J.; DuBow, M. Nat. Biotechnol. 2004,
22, 185.
5. (a) Van Der Ende, A.; Baker, T. A.; Ogawa, T.; Kornberg,
A. Proc. Natl. Acad. Sci. U.S.A. 1985, 82, 3954; (b) Davey,
M. J.; OÕDonnell, M. Curr. Biol. 2003, 13, R594.
6. (a) Houston, P.; Kodadek, T. Proc. Natl. Acad. Sci.
U.S.A. 1994, 91, 5471; (b) Soni, R. K.; Mehra, P.;
Choudhury, N. R.; Mukhopadhyay, G.; Dhar, S. K.
Nucleic Acids Res. 2003, 31, 6828.
7. (a) Zhong, Z.; Helinski, D.; Toukdarian, A. J. Biol. Chem.
2003, 278, 45305; (b) Caspi, R.; Pacek, M.; Consiglieri, G.;
Helinski, D. R.; Toukdarian, A.; Konieczny, I. EMBO J.
2001, 20, 3262.
which shifts to >128 lg/mL in the presence of 50% hu-
man or mouse serum in the growth medium.21,22 A sim-
ilar shift is obtained with 4% human serum albumin
(HSA). A gradual increase in albumin concentration re-
sults in a gradual linear shift in MIC, the slope of which
points to a stoichiometry of 0.47:1 for the binding of 1
to HSA and 0.23:1 for that of compound 3 and
HSA.21 Although these low stoichiometries of serum
protein binding by compounds 1 and 3 may leave some
room for attempts to retain target-specific activity while
reducing serum binding, the steeper slope of the MIC:se-
rum protein curve when using whole serum in place of
HSA is evidence that these compounds bind several
components of serum. This renders the possibility of
avoiding serum shift by design very small.
Because of this somewhat unselective character, it is not
surprising that when evaluated for potential as topical
antibacterial agents in vivo in the mouse wound model
of S. aureus infection,23 compounds 1 and 3 (0.2 mL
of compound formulated as 2% by weight in PEG400,
applied over the wound three times per day) were found
to be completely inactive (data not shown).
The present study has allowed the identification of 1,3,5-
triaminotriazines as inhibitors of the replicative helicase
DnaB from P. aeruginosa. The compounds lack antibac-
terial activity against this organism, but are active
against S. aureus and sporadically against a tolC mutant
of E. coli. Additional attributes of a lack of detectable
inhibition of mammalian DNA replication and little
affinity for DNA in vitro must be countered with the
demonstrable HeLa cell cytotoxicity of the original hit.
The development of brief structure–activity relation-
ships around the initial structure has allowed the discov-
ery of compounds with comparable antibacterial activity
but with a highly reduced cytotoxicity. Further investi-
gations into the activity associated with this class how-
ever show an unselective mode of action. This is
additionally supported by the detectable binding to a
number of serum components. While this sort of behav-
ior leaves little room to develop these compounds into
useable therapeutics, it serves to highlight the necessity
of stringent assessment of specificity and selectivity as
early as possible in the characterization of inhibitors
of in vitro enzyme assays.
8. Trautmann, M.; Lepper, P. M.; Haller, M. Am. J. Infect.
Control 2005, 33, S41.
9. Goldberg, J. B.; Pier, G. B. Trends Microbiol. 2000, 8, 514.
10. Zhang, Y.; Yang, F.; Kao, Y.-C.; Kurilla, M. G.;
Pompliano, D. L.; Dicker, I. B. Anal. Biochem. 2002,
304, 174.
11. Malich, G.; Markovic, B.; Winder, C. Toxicology 1997,
124, 179.
12. George, J. W.; Ghate, S.; Matson, S. W.; Besterman, J. M.
J. Biol. Chem. 1992, 261, 10683.
13. (a) Boger, D. L.; Fink, B. E.; Brunette, S. R.; Tse, W. C.;
Hedrick, M. P. J. Am. Chem. Soc. 2001, 123, 5878; (b)
Boger, D. L.; Fink, B. E.; Hedrick, M. P. J. Am. Chem.
Soc. 2000, 122, 6382.
14. National Committee for Clinical Laboratory Standards:
Methods for Dilution Antimicrobial Susceptibility Tests
for Bacteria that Grow Aerobically, 4th ed.: Approved
Standard M7-A4. Wayne (PA): NCCLS.
15. Calculated with Pallas PrologD, CompuDrug Internation-
al, Inc.; Sedona, AZ, USA, 2003.
16. The negative impact of higher logP values on antibacte-
rials has already been noted. See: Cronin, M. T. D.;
Aptula, A. O.; Dearden, J. C.; Duffy, J. C.; Netzeva, T. I.;
Patel, H.; Rowe, P. H.; Schultz, T. W.; Worth, A. P.;

1290
G. A. McKay et al. / Bioorg. Med. Chem. Lett. 16 (2006) 1286–1290
Voutzoulidis, K.; Schuurmann, G. J. Chem. Inf. Comput.
¨ ¨
Sci. 2002, 42, 869.
20. Brazas, M. D.; Hancock, R. E. W. Drug Discovery Today
2005, 10, 1245.
17. Kaito, C.; Kurokawa, K.; Hossain, M. S.; Akimitsu, N.;
Sekimizu, K. FEMS Microbiol. Lett. 2002, 210, 157.
18. Although a number of inhibitors of mammalian and viral
helicases exist, the only report of inhibitors of bacterial
helicases is: Xu, H.; Zieglin, G.; Schro¨der, W.; Frank, J.;
Ayora, S.; Alonso, J. C.; Lanka, E.; Saenger, W. Nucleic
Acids Res. 2001, 29, 5058.
19. Hilliard, J. J.; Goldschmidt, R. M.; Licata, L.; Baum, E.
Z.; Bush, K. Antimicrob. Agents Chemother. 1999, 43,
1693.
21. For the use of serum to detect promiscuity in antibacterial
assays: Roychoudhury, S.; Brill, J. L.; Lu, W.-P.; White,
R. E.; Chen, Z.; Demuth, T. P., Jr. J. Biomol. Screening
2003, 8, 555.
22. For the use of serum albumin to detect promiscuity in
in vitro enzymatic assays: McGovern, S. L.; Caselli, E.;
Grigorieff, N.; Shoichet, B. K. J. Med. Chem. 2002, 45,
1712.
23. Gisby, J.; Bryant, J. Antimicrob. Agents Chemother. 2000,
44, 255.