Synthesis and biological evaluation of novel 1,3,5-triazine derivatives as antimicrobial agents

Article abstract of DOI:10.1016/j.bmcl.2008.01.031

Numerous studies have contributed to the development of natural and synthetic antimicrobial peptides as a prospective source of antibiotic agents. Based on the concept that cationic charge, bulk, and lipophilicity are major factors determining antibacterial activity in these peptides, we designed and screened several combinatorial libraries based on 1,3,5-triazine as a template. A set of compounds were identified to show potent antimicrobial activity together with low hemolytic activity.

Full text of DOI:10.1016/j.bmcl.2008.01.031

Available online at www.sciencedirect.com
Bioorganic & Medicinal Chemistry Letters 18 (2008) 1308–1311
Synthesis and biological evaluation of novel
1,3,5-triazine derivatives as antimicrobial agents
Chunhui Zhou, Jaeki Min, Zhigang Liu, Anne Young, Heather Deshazer,
Tian Gao, Young-Tae Changà and Neville R. Kallenbach*
Department of Chemistry, New York University, 100 Washington Square East, New York, NY 10003, USA
Received 21 November 2007; revised 3 January 2008; accepted 8 January 2008
Available online 11 January 2008
Abstract—Numerous studies have contributed to the development of natural and synthetic antimicrobial peptides as a prospective
source of antibiotic agents. Based on the concept that cationic charge, bulk, and lipophilicity are major factors determining anti-
bacterial activity in these peptides, we designed and screened several combinatorial libraries based on 1,3,5-triazine as a template.
A set of compounds were identified to show potent antimicrobial activity together with low hemolytic activity.
ꢀ 2008 Elsevier Ltd. All rights reserved.
As multidrug-resistant bacterial strains proliferate, the
need for new kinds of antibiotics is growing. In the past
two decades it has been found that a wide range of anti-
microbial peptides (AMPs) are secreted by innate im-
mune system of multicellular organisms in response to
infection by foreign bacteria, viruses, or fungi.1–3 These
host defense peptides have been considered as prospec-
tive antibiotic agents because their effect is rapid, broad
spectrum, and indifferent to resistance to standard anti-
biotics such as penicillin.4 AMPs differ dramatically in
size, sequence, and structure, apparently sharing only
amphiphilicity and positive charge.1,2 The proposed
mechanism of action of AMPs focuses on the interaction
between the peptides and the negatively charged cell wall
and/or plasma membranes of bacterial cells, although
certain antimicrobial peptides can also employ more
sophisticated mechanisms.5 Recently, a pharmacophore
for short cationic antimicrobial peptides of the cathelic-
idin family including indolicidin and tritrpticin has been
identified.6,7 A range of very short cationic peptides con-
taining only positively charged Arginine (R) and lipo-
philic Tryptophan (W) residues have been found to
serve as effective antimicrobial agents with MIC values
in the low micromolar range.8,9
Despite some potential advantages over conventional
antibiotics, practical applications of antimicrobial pep-
tides, however, are limited by several factors. Peptides
tend to be expensive to synthesize in bulk, are vulnerable
to protease degradation, and have relatively high cyto-
toxicity to red blood cells. A number of mimetics con-
structed from unnatural b-amino acids, cyclization of
peptides, peptoid mimics, or other peptide analogs have
been investigated in efforts to overcome these prob-
lems.10–13 Here we have designed and screened several
combinatorial libraries of small compounds to mimic
the hydrophobic and charge pattern detected in the
pharmacophore of short cationic antimicrobial peptides.
The present work focuses on stepwise design and opti-
mization of functional groups selected to reproduce
the RW pharmaceutical properties based on 1,3,5-tri-
azine as a template. The results reveal potential new
antimicrobial leads, as well as provide some insight into
the structure–function relationship of these agents.
1,3,5-Triazine possessing threefold symmetry was cho-
sen in our libraries as the scaffold because they allow
for versatile modifications uncomplicated by regiochem-
ical concerns and have proven themselves to be useful
biological targets.14–16 The first two libraries were cre-
ated in order to screen various functional groups for
their antimicrobial activity. A total of 79 compounds
in library 1 and 72 compounds in library 2 (Scheme 1)
were synthesized utilizing an orthogonal synthetic ap-
proach based on the triazine scaffold as documented
previously.17,18 Two mono-substituted triazine tem-
plates were used for these libraries, in which amantadine
Keywords: Cationic antimicrobial peptide mimetic; 1,3,5-Triazine;
Membranolytic.
*
Corresponding author. Tel.: +1 212 998 8757; fax: +1 212 260
7905; e-mail: nrk1@nyu.edu
These authors contributed equally to this work.
à
Present address: Department of Chemistry, National University of
Singapore, 3 Science Drive 3, Singapore 117543, Singapore.
0960-894X/$ - see front matter ꢀ 2008 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2008.01.031

C. Zhou et al. / Bioorg. Med. Chem. Lett. 18 (2008) 1308–1311
1309
Cl
N
Cl
N
OMe
a or b
N
N
N
Y
Cl
N
Cl
X
N
+
Cl
OMe
N
X
d
N
N
Cl
OMe
OMe
CHO
c
e
Y
N
OMe
OMe
O
H
OMe
O
NH2
NH
HN
N
N
f
or
N
N
N
N
Y
R1
N
R2
R2'
library 2
R1'
OMe
N
X
N
library 1
R2 (or R2' )
X = 1-adamantylamine for library 1 or R1' for library 2
Y = R1 for library 1 or TG-Boc for library 2
N
R1
A
B
C
D
E
F
G
H
HN
NH
HN
NH
NH
NH
NH
NH
MeO
HO
R2
1
2
3
4
5
6
7
8
9
10
NH
NH
NH
HN
HN
HN
HN
NH
O
Bz
NH
O
NH
N
O
H
O
O
H
O
H
OH
HO
OH
OH
OH
HO
R1'
I
J
K
L
M
N
O
P
Cl
R2'
11
12
13
14
15
16
17
18
19
HN
HN
N
NH
N
N
HN
HN
NH
N
N
HN
N
N
N
N
N
N
HO
O
Scheme 1. General synthetic scheme for libraries 1 and 2. Reagents and conditions: (a) 1-Adamantylamine, DIEA, THF, rt, 1 h, for library 1; (b)
R1 MgX, THF, 0 ꢁC, 2 h, for library 2; (c) R1 (for library 1) or TG-Boc (for library 2), 2% acetic acid in THF, rt, 1 h, followed by NaB(OAc) 3H, rt,
0
0
12 h; (d) THF, DIEA, 60 ꢁC, 1 h; (e) R2 or R2 , DIEA, NMP:n-BuOH = 1:1, 120 ꢁC, 3 h; (f) 10% TFA in dichloromethane, rt, 1 h.
or 2,20-(Ethylenedioxy)bis(ethylamine) (TG-Boc) were
prefixed onto the triazine scaffold to represent one lipo-
R1: B, C, F(3), H, 11, 13, 15, 17
NH
R2: 11,15
philic/bulky group and one charged group, respectively.
In addition to the prefixed groups, series of compounds
were synthesized to introduce two more groups (hydro-
phobic, bulky, positively charged, etc.) to positions R1
and R2. Antimicrobial activities were tested against the
Gram-positive bacteria, Bacillus subtilis, which showed
library 1 with amantadine to be superior to library 2
in general (data not shown).
N
N
R1
N
R2
Figure 1. Design of library 3. B, C, F(3), H, 11, 13, 15, and 17 are
functional groups shown in Scheme 1.
Some of these show the highest antimicrobial activity
we have obtained so far, for example TZ-4 and TZ-12.
The antimicrobial test data for these compounds re-
vealed some interesting trends and various degrees of
activities against B. subtilis (Fig. 2). In general, increased
bulk (ring size) on one side chain of the triazine com-
pounds enhanced antimicrobial activity, with ring sizes
7 and 8 being optimal. But if the ring size was increased
to 12 (TZ-5), activity decreased almost eightfold relative
to rings of size 7 or 8. On the other hand, increasing the
amine chain length did not show any obvious trend.
The subsequent library 3 with 2 · 8 compounds (Fig. 1)
was designed based on eight functional groups with
highest frequency of effectiveness screened in libraries
1 and 2.19 C11 was identified to be the most active anti-
microbial with IC50 = 2.5 lM in library 3, see Ref. 20 for
assay results. A further set of 22 triazine compounds was
synthesized with optimization of R1 and R2 groups
based on C11, then purified and confirmed by NMR.21

1310
C. Zhou et al. / Bioorg. Med. Chem. Lett. 18 (2008) 1308–1311
HN
HN
HN
HN
HN
HN
N
N
N
N
N
N
N
N
N
N
N
N
Bu
Bu
Bu
Bu
Bu
Bu
N
Bu
N
H
N
N
H
N
Bu
N
H
N
N
H
N
Bu
N
H
N
N
H
N
Bu
N
H
N
N
H
N
Bu
N
H
N
N
H
N
Bu
N
H
N
N
H
TZ-1: 16.4 uM
TZ-2: 7.9 uM
TZ-3: 4.6 uM
Lead C11: 2.5 uM
TZ-4: 2.0 uM
TZ-5: 15.9 uM
HN
HN
HN
HN
HN
HN
N
N
N
N
N
N
N
N
N
N
N
N
Bu
Et
Bu
Bu
Bu
Bu
N
Bu
N
H
N
N
Et
N
H
N
N
Bu
N
H
N
N
H
N
Bu
N
H
N
Bu
N
Bu
N
H
N
N
H
N
N
Bu
N
H
N
N
H
N
N
H
Bu
TZ-8: 39 uM
TZ-6: 7.6 uM
TZ-7: 5.5 uM
TZ-9: 4.4 uM
TZ-11: 10.6 uM
TZ-10: 5.0 uM
HN
HN
HN
HN
HN
HN
N
N
N
N
N
N
Et
N
N
N
N
N
N
Et
N
Et
Et
Et
N
N
N
N
H
N
N
Et
N
H
N
N
Et
N
H
N
N
H
N
N
H
N
N
H
N
N
H
N
H
N
H
Et
N
H
N
H
N
H
O
TZ-17: 5.6 uM
TZ-16: 7 uM
TZ-15: 5.3 uM
O
TZ-13: 3.7 uM
TZ-12: 2.7 uM
TZ-14: 3.5 uM
HN
HN
HN
HN
HN
N
N
N
N
N
N
N
N
N
N
H N
N
H
N
N
H
N
N
N
H
N
N
N
N
N
H
N
H N
N
H
N
H
N
H
N
H
N
H
TZ-19: 5.9 uM
TZ-21: 5.9 uM
TZ-22: >>40 uM
TZ-18: 5.7 uM
TZ-20: 4.7 uM
Figure 2. Optimization of R1 and R2 groups based on lead compound C11 from library 3. The activity data are IC50 which are the means of three
independent experiments performed in parallel against B. subtilis.
To test spectrum of antibacterial activity, the three most
active compounds C11, TZ-4, and TZ-12 were further
tested against both Gram-positive and Gram-negative
bacteria, including ampicillin- and streptomycin-resis-
tant E. coli (D31) and multidrug-resistant S. aureus.
Hemolytic activity was assayed against red blood cells
(Table 1).22 The results showed that these compounds
were generally more selective against Gram-positive
bacteria, with IC50 value in the low micromolar range
and that none was hemolytic even at high concentration.
Finally, to test whether the active triazine compounds
kill bacteria via disrupting membrane integrity, a calcein
dye leakage experiment was carried out on TZ-12 with
ampicillin and one of the triazine compounds with little
antimicrobial activity (TZ-22) as control (Fig. 3).23 TZ-
Figure 3. Calcein leakage induced by 1% Triton, TZ-12 (50 lM), TZ-
12 causes 100% dye leakage at 50 lM. The results con-
22 (90 lM), and ampicillin (70 lM). Fraction of leakage was calculated
from the fluorescence intensity at 515 nm, with 100% leakage
calibrated by addition of 1 % Triton X-100.
firmed that the antimicrobial active triazine compound
could cause dye leakage from negatively charged 1-pal-
mitoyl-2-oleoyl-sn-glycero-3-[phospho-rac-(1-glycerol)]
(POPG) lipid vesicle, while neither ampicillin nor com-
pound TZ-22 caused dye leakage.
Interestingly, triazines with two amantadine groups had
much lower antimicrobial activity relative to com-
pounds having only one amantadine group.
In summary, we have designed and screened privileged
libraries consisting of tri-substituted triazine compounds
Table 1. Spectrum of antibacterial activity for triazine compounds C11, TZ-4, and TZ-12
a
b
Compounds
IC50 (lM)
HD50 (lM)
Escherichia coli
Acinetobacter baumannii
Bacillus anthracis
Staphylococcus aureus
C11
TZ-4
TZ-12
620
550
200
53
84
22
41
86
21
68
70
17
>3500
>3500
>3500
a The results are means of three independent experiments performed in parallel.
b HD50 determined from dose–response curve is peptide concentrations corresponding to 50% hemolysis.

C. Zhou et al. / Bioorg. Med. Chem. Lett. 18 (2008) 1308–1311
1311
15. Giacomelli, G.; Porcheddu, A.; De Luca, L. Curr. Org.
Chem. 2004, 8, 1497.
as antimicrobials. Several lead compounds possessing
potent antimicrobial and low hemolytic activity were
identified. Further, these compounds preferentially kill
Gram-positive bacteria relative to gram-negative ones
and probably kill them via disrupting membrane integ-
rity as shown by a dye leakage assay. These responses
are similar to those of most natural or synthetic AMPs
reported in the literature. These data open a possible
route to the elaboration of novel antibiotics derived
from AMP mimetics that afford more drug-like pharma-
ceutical properties than peptides. Further investigation
on the mechanism of these triazine compounds will be
needed to distinguish their mode of action from that
of other triazine compounds.
16. Silen, J. L.; Lu, A. T.; Solas, D. W.; Gore, M. A.;
Maclean, D.; Shah, N. H.; Coffin, J. M.; Bhinderwala, N.
S.; Wang, Y.-W.; Tsutsui, K. T.; Look, G. C.; Campbell,
D. A.; Hale, R. L.; Navre, M.; Deluca-Flaherty, C.
Antimicrob. Agents Chemother. 1998, 42, 1447.
17. Min, J.; Kim, Y. K.; Cipriani, P. G.; Kang, M.;
Khersonsky, S. M.; Walsh, D. P.; Lee, J. Y.; Niessen, S.;
Yates, J. R.; Gunsalus, K.; Piano, F.; Chang, Y. T. Nat.
Chem. Biol. 2007, 3, 55.
18. Uttamchandani, M.; Walsh, D. P.; Khersonsky, S. M.;
Huang, X.; Yao, S. Q.; Chang, Y. T. J. Comb. Chem.
2004, 6, 862.
19. Frequency of functional group effectiveness from libraries
1 and 2. The number of occurrences of the top eight side
chain groups in triazines at 100 lM versus B. subtilis
(>90% inhibition): functional group B(four times), C(3),
F(4), H(3), 11(6), 13(3), 15(3), and 17(4). These functional
groups were used to design library 3.
Acknowledgments
20. Assay results of screening library 3. Triazine compounds:
B11(IC50
This work was supported by a grant from ONR
(N00014-03-1-0129). We acknowledge the NCRR/NIH
for a Research Facilities Improvement Grant (C06
RR-16572) at NYU. The authors thank Brian Rasimick
for helpful discussions.
7.5 lm),
C11(2.5 lm),
C15(7.5 lm),
11–
H11(7.5 lm),
H15(3.0 lm),
11–11(7.5 lm),
15(7.5 lm), and 11–17(7.5 lm). All these compounds
cause less than 10% hemolysis at 100 lM.
21. The purity and identity of all the products were monitored
by LC–MS at 250 nm (Agilent 1100) and more than 90%
of the compounds demonstrated >90% purity. NMR and
1
References and notes
ESI-MS characterization data for example: C11 H NMR
(400 MHz, CDCl3): d 4.81 (br, 2H), 3.96 (brs, 1H), 3.42
(m, 2H), 2.67 (m, 1H), 2.57 (m, 4H), 2.10 (m, 8H), 1.99 (m,
2H), 1.83 (m, 3H), 1.69–1.45 (m, 22H), 1.34 (m, 4H), 0.93
(t, 6H, J = 7.5 Hz). ESI-MS (M+H)+ calcd, 526.4; Found,
526.2. Data for TZ-4: 1H NMR (400 MHz, CDCl3): d 4.79
(br, 2H), 4.02 (brs, 1H), 3.41 (m, 2H), 2.61 (m, 1H), 2.50
(t, 4H, J = 6.8 Hz), 2.10 (m, 8H), 1.88 (m, 2H), 1.78 (m,
3H), 1.68–1.46 (m, 24H), 1.32 (m, 4H), 0.92 (t, 6H,
J = 7.3 Hz). ESI-MS (M+H)+ calcd, 540.4; Found, 540.2.
Data for TZ-12: 1H NMR (400 MHz, CDCl3): d 7.29–7.13
(m, 6H), 4.70 (br, 2H), 3.52 (m, 2H), 3.33 (m, 2H), 2.78 (m,
2H), 2.54 (m, 6H), 2.02 (m, 8H), 1.70 (m, 3H), 1.60 (m,
6H), 1.02 (t, 6H, J = 6.8 Hz). ESI-MS (M+H)+ calcd,
478.3; Found, 478.1.
1. Zasloff, M. Nature 2002, 415, 389.
2. Boman, H. G. J. Intern. Med. 2003, 254, 197.
3. Hancock, R. E. W. Nat. Rev. Drug Discov. 2007, 6, 28.
4. Hancock, R. E. W.; Sahl, H. G. Nat. Biotechnol. 2006, 24,
1551.
5. Yeaman, M. R.; Yount, N. Y. Pharmacol. Rev. 2003, 55,
27.
6. Staubitz, P.; Peschel, A.; Nieuwenhuizen, W. F.; Otto, M.;
Gotz, F.; Jung, G.; Jack, R. W. J. Pept. Sci. 2001, 7, 552.
7. Strom, M. B.; Haug, B. E.; Skar, M. L.; Stensen, W.;
Stiberg, T.; Svendsen, J. S. J. Med. Chem. 2003, 46, 1567.
8. Haug, B. E.; Stensen, W.; Stiberg, T.; Svendsen, J. S. J.
Med. Chem. 2004, 47, 4159.
22. Growth inhibition and hemolysis assays here and men-
tioned above were carried out using a microdilution
method as previously described in Ref. 9. Bacterial strain:
A. baumannii (ATCC BAA-747; Rockville, MD); B.
subtilis (ATCC 6633; Rockville, MD); Ampicillin- and
streptomycin-resistant strain E. coli (D31; E. coli Genetic
Resource Center, Yale University, New Haven, CT);
multi-drug resistant strain S. aureus (ATCC BAA-44); B.
anthracis (32F2 Sterne) (gift from Dr. Martin Blaser,
NYU medical school).
23. The method of preparation of dye-encapsulated vesicles
has been reported in detail: Lasch, V.; Weissig, M. In Bran
in Liposomes: a Practical Approach; Taylor, K. M. G.,
Craig, D. Q. M., Eds.; Oxford University Press: Oxford,
2003; p 10.
9. Liu, Z.; Brady, A.; Young, A.; Rasimick, B.; Chen, K.;
Zhou, C.; Kallenbach, N. R. Antimicrob. Agents Chemo-
ther. 2007, 51, 597.
10. Porter, E. A.; Wang, X.; Lee, H. S.; Weisblum, B.;
Gellman, S. H. Nature 2000, 404, 565.
11. Fernandez-Lopez, S.; Kim, H. S.; Choi, E. C.; Delgado,
M.; Granja, J. R.; Khasanov, A.; Kraehenbuehl, K.;
Long, G.; Weinberger, D. A.; Wilcoxen, K. M.; Ghadiri,
M. R. Nature 2001, 412, 452.
12. Patch, J. A.; Barron, A. E. J. Am. Chem. Soc. 2003, 125,
12092.
13. Liu, Z.; Young, A. W.; Hu, P.; Rice, A. J.; Zhou, C.;
Zhang, Y.; Kallenbach, N. R. Chembiochem 2007.
doi:10.1002/cbic.200700502.
14. Walsh, D. P.; Chang, Y. T. Chem. Rev. 2006, 106, 2476.