Mode of action investigation for the antibacterial cationic anthraquinone analogs

Article abstract of DOI:10.1016/j.bmcl.2011.08.107

Reported previously by our group, we have developed a novel class of antibacterial cationic anthraquinone analogs with superb potency (MIC <1 μg/mL) against Gram positive (G+) pathogens including Methicillin-resistant Staphylococcus aureus (MRSA). However, most of these compounds only manifest modest antibacterial activity against Gram negative (G-) bacteria. Further investigation on the antibacterial mode of action using fluorogenic dyes reveals that these compounds exert two different modes of action that account for the difference in their antibacterial profile. It was found that most of the compounds exert their antibacterial activity by disrupting the redox processes of bacteria. At high concentration, these compounds can also act as membrane disrupting agents. This information can help to design new therapeutics against various bacteria.

Full text of DOI:10.1016/j.bmcl.2011.08.107

Bioorganic & Medicinal Chemistry Letters 21 (2011) 6353–6356
Contents lists available at SciVerse ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Mode of action investigation for the antibacterial cationic
anthraquinone analogs
⇑
Ka Yee Chan, Jianjun Zhang, Cheng-Wei Tom Chang
Department of Chemistry and Biochemistry, Utah State University, 0300 Old Main Hill, Logan, UT 84322 0300, USA
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 4 July 2011
Revised 20 August 2011
Accepted 25 August 2011
Available online 31 August 2011
Reported previously by our group, we have developed a novel class of antibacterial cationic anthraqui-
none analogs with superb potency (MIC <1 lg/mL) against Gram positive (G+) pathogens including Meth-
icillin-resistant Staphylococcus aureus (MRSA). However, most of these compounds only manifest modest
antibacterial activity against Gram negative (GÀ) bacteria. Further investigation on the antibacterial
mode of action using fluorogenic dyes reveals that these compounds exert two different modes of action
that account for the difference in their antibacterial profile. It was found that most of the compounds
exert their antibacterial activity by disrupting the redox processes of bacteria. At high concentration,
these compounds can also act as membrane disrupting agents. This information can help to design
new therapeutics against various bacteria.
Keywords:
Anthraquinone analogs
Antibacterial
Ó 2011 Elsevier Ltd. All rights reserved.
In light of the epidemic of antibiotic resistant bacteria,1 our
group has been devoting effort in the development of new antibac-
terial agents.2 We have recently reported the discovery of a new
class of antibacterial cationic anthraquinone analogs derived from
1-alkyl-1H-naphtho[2,3-d]triazole-4,9-diones (Fig. 1).3 Our studies
show that the compound attached with an octyl group (3a) dis-
plays significant antibacterial activity. Further shortening the alkyl
chain length leads to the decrease of antibacterial activity. Interest-
ingly, increasing the alkyl chain length to dodecyl (C12) and hexa-
decyl (C16) groups, such as in compounds 3e and 3f, similar
activities against both G+ and GÀ bacteria were noted. Although
compound 3a has highly activity toward G+ bacteria, it is only
modestly active against GÀ bacteria. Therefore, we were interested
in studying more details on the possible mode of antibacterial ac-
tion and structure activity relationship (SAR) of these compounds.
For this purpose, four additional cationic anthraquinone analogs
were synthesized (Scheme 1).
Compounds 2b, 2c and 2d were prepared from the cycloaddi-
tion/oxidation approach using naphthoquinone and the corre-
sponding alkyl bromides in a one-pot protocol.4 Methylation of
the N-3 nitrogen provided the desired products 3b, 3c and 3d with
nonyl, decyl, and undecyl groups, respectively. In order to investi-
gate the effect of the chain length on N-3 nitrogen, ethylation of 2a
was conducted using ethyl triflate to yield 3g. Ion-exchange using
Dowex resin offered the final products with chloride as the counter
ion.
All of the newly synthesized compounds along with 3a, 3d and
3e, which were prepared previously, were tested against
Escherichia coli (GÀ, ATCC 25922) and Staphylococcus aureus (G+,
ATCC 25923). The minimum inhibitory concentrations (MIC’s)
were determined (Table 1). From the antibacterial testing results,
it is apparent that these cationic compounds are much more active
against G+ bacteria than GÀ bacteria which is consistent with the
antibacterial profile of cationic antiseptic agents, such as hexade-
cyltrimethylammonium bromide (HTB) and cetrimonium bro-
mide.5 However, unlike most of the antiseptic agents that often
contain much longer linear alkyl chain (C12-C18), compounds we
synthesized that incorporated with octyl, nonyl and decyl groups
provide the highest activity, especially, against S. aureus. An
increase in the chain length greater than C12 results in a decrease
in the compounds’ antibacterial activity. To our surprise, the MIC’s
of 3e and 3f were higher than what we previously reported.3 We
noted some induced resistance in E. coli when the experiments
were observed for 10–18 h. This could be accounted for the dis-
crepancies in MIC’s. Interestingly, the incorporation of ethyl group
1) MeOTf, toluene
O
O
Me
Cl-
N
100 °C
3
2) Dowex 1X-8
N
N
(Cl-form)
N
N1
N
R
R
O
O
Cationic anthraquinone analogs
1-alkyl-1H-naphthotriazole-4,9-
dione
⇑
Corresponding author. Tel.: +1 4357973545; fax: +1 4357973390.
E-mail address: tom.chang@usu.edu (C.-W.T. Chang).
Figure 1. Cationic anthraquinone analogs.
0960-894X/$ - see front matter Ó 2011 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2011.08.107

6354
K. Y. Chan et al. / Bioorg. Med. Chem. Lett. 21 (2011) 6353–6356
1) MeOTf, toluene
O
O
O
O
O
Me
Cl-
N
100 °C
2) Dowex 1X-8
(Cl-form)
NaN3
N
N
R
Br
DMF
120
N
N
R
N
°
C
R
O
1
2a, R = C8H17
3a, R = C8H17
2b, R = C9H19, 38%
2c, R = C10H21, 47%
, 51%
3b, R = C9H19, 87%
3c, R = C10H21, 99%
, 88%
2d, R = C11 23
H
3d, R = C11 23
H
2e, R = C12 25
H
3e, R = C12 25
H
2f, R = C16 33
H
3f, R = C16 33
H
1) EtOTf, toluene
O
O
Et
100 °C
2) Dowex 1X-8
(Cl-form)
N
Cl-
2a
N
51%
N
C8H17
3g
Scheme 1.
Table 1
S. aureus
E. coli
MIC’s of cationic anthraquinone analogs (unit:
l
g/mL)
Compound
E. coli
S. aureus
10
3a
3b
3c
3d
3e
3f
16–32
64
32
32–64
16–32
P250
P250
0.032–0.064
0.032–0.064
0.032
0.25–0.5
0.064–0.125
0.25–0.5
min.
3g
0.25–0.5
2 hrs
instead of methyl group (3a vs 3g) results in an eight-fold decrease
in the antibacterial activity. It seems that the antibacterial activity
is sensitive to the lipophilicity of the incorporated alkyl groups.
Further investigation using compounds with various chain lengths
at both nitrogen atoms and quantity structure activity relationship
(QSAR) analysis has been carried out.
Figure 2. Bacteria treated with HTB (2 mg/mL) and Sytox dye.
and the other was below MIC to evaluate the effect of compounds
with some live bacteria. To simplify the task, we selected MIC’s of
The cationic anthraquinone compounds have two distinct struc-
tural motifs: the cationic anthraquinone core and the lipophilic
alky chain(s). Molecules bearing naphthoquinone or anthraqui-
none cores are known to uncouple mitochondria oxidative phos-
phorylation via redox chemistry.6 Antiseptic agents with
lipophilic alkyl chain(s) are noted for their ability to disrupt the
bacterial membrane.7 Therefore, we postulate that these cationic
anthraquinone compounds exert their antibacterial activity
through one mode or a combination of these modes of action. To
evaluate the hypothesis of membrane disruption, a fluorescence
dye, Sytox was used.8 Sytox dye cannot penetrate the intact bacte-
rial membrane unless the membrane is damaged by the membrane
disrupting agents. Once inside the cells, Sytox dye can bind to nu-
cleic acid and emit strong fluorescence. Therefore, the observation
of fluorescence is an evidence of membrane disruption. To investi-
gate the other hypothesis involving disruption of the redox
processes inside the cell, a fluorogenic redox indicator, 5-cyano-
2,3-ditolyltetrazolium chloride (CTC) dye was employed.9 CTC
dye can penetrate the bacterial membrane and, once inside the
cells, CTC can be readily reduced by non-specific electron carriers.
The reduced form of CTC dye will emit strong fluorescence indicat-
ing the cell redox processes are operating normally. Therefore, an
observation of no fluorescence is an indication of cell death as a re-
sult of inhibited redox processes.
2 mg/mL and 4 lg/mL for E. coli, and 8 and 0.064 lg/mL for S. aur-
eus. For the Sytox dye experiment, the results were observed fol-
lowing 10 min and 2 h of incubation of bacteria with the
compounds. A known membrane disrupting agent HTB was used
as the control. For the CTC dye experiment, the results were ob-
served following 2 and 6 h of incubation of bacteria with the com-
pounds. Neomycin was used as the control. Neomycin interferes
with the bacterial protein synthesis, which is a slower than the
process of redox inhibition. Thus, we expect to observe fluores-
cence from neomycin treated cells during the designed time
frames even at a concentration above the MIC of neomycin. For
all the experiments, two types of photos were taken for compari-
son: fluorescence and phase modes.
From the Sytox dye experiments, cells that were treated with
HTB (2 mg/mL) produced green fluorescence after 10 min of incu-
bation and more cells with fluorescence appeared following 2 h
of incubation (Fig. 2). When E. coli was treated with compound
3a at a concentration below MIC, very few cells emitted green fluo-
rescence (Fig. 3). In contrast, when E. coli was treated with com-
pound 3a at a concentration above MIC, many cells emitted
green fluorescence. The results suggest that the bacterial mem-
brane can only be disrupted by cationic anthraquinone analogs at
high concentration but remain largely unaffected with compounds
at lower concentration. When S. aureus was treated with cationic
anthraquinone analogs, very few cells emitted fluorescence at con-
centrations both above and below the MIC’s (Fig. 4). The results
Two concentrations of cationic anthraquinone analogs were
used: one was above MIC to provide the bactericidal condition

K. Y. Chan et al. / Bioorg. Med. Chem. Lett. 21 (2011) 6353–6356
6355
2 mg/mL
2 mg/mL
4 µg/mL
4 µg/mL
10
2
min.
hrs
2
6
hrs
hrs
Figure 3. E. coli treated with 3a and Sytox dye.
Figure 6. E. coli treated with 3a and CTC dye.
0.064 µg/mL
8 µg/mL
these cationic anthraquinone compounds kill bacteria by mem-
brane disruption which occurs much faster than the antibacterial
mode of neomycin. For the assay against S. aureus, some fluores-
cence was observed with the concentration below the compounds’
MIC suggesting that the redox processes were hampered even while
cells were still alive (Fig. 7). Finally, almost no fluorescence was seen
when cells were treated with concentration above the compounds’
MIC since the redox processes had been inhibited. Therefore, it is
clear that the redox process(s) can be disrupted by these cationic
anthraquinone analogs. Again, similar trends occurred using other
compounds.11
10
min
.
2
hrs
GÀ bacteria differ from G+ bacteria by the presence of outer
membrane and the lipopolysaccharides (LPS), which contain vari-
ous negatively charged functional groups.10 The cationic anthra-
quinone analogs exert their antibacterial activity predominately
by inhibiting or disrupting the redox processes inside the cells.
To inhibit the redox processes, these cationic compounds need to
diffuse across the bacterial membrane. The outer membrane and
LPS of GÀ bacteria can form the additional barrier that account
for the relatively lower antibacterial activities of these cationic
compounds.5 However, at high concentration, these cationic com-
pounds can still function as membrane disrupting agents like HTB.
Nevertheless, increasing the chain length (lipophilicity) will make
it more difficult for the compounds to pass through the bacterial
membrane, and thus results in lower antibacterial activity.
In conclusion, we have identified the optimal chain length for
antibacterial activity of the cationic anthraquinone analogs. We
have also shown that these compounds can inhibit the redox pro-
cesses inside the cells which could be the cause of bactericidal ef-
fect of these compounds. These compounds can also disrupt the
bacterial membrane at a higher concentration. The unique combi-
nation of these two antibacterial modes of action can help to de-
sign better antibiotics that can be either selective or broad
spectrum.
Figure 4. S. aureus treated with 3a and Sytox dye.
imply that the bacterial membrane is largely unaffected at concen-
trations both above and below the MIC’s. This suggests that differ-
ent target(s) inside the G+ bacteria are responsible for the
antibacterial activity of these cationic anthraquinone analogs. Sim-
ilar trends occurred using other compounds.11
From the CTC dye experiments, red fluorescence appeared for
both E. coli and S. aureus that were treated with neomycin at con-
centrations both above and below its MIC’s following the designed
incubation time (Fig. 5). The bactericidal effect of neomycin occurs
in a relatively long time frame (10–18 h). Therefore, this observa-
tion of fluorescence indicated that the redox processes remained
active and most of the cells were still alive even in the presence
of bactericidal neomycin after 6 h.
For the assay using CTC against E. coli, significant fluorescence
was observed after a 2 and 6 hrs of incubation with the concentra-
tion below the compounds’ MIC indicating that the majority of the
bacteria were still alive (Fig. 6). In contrast, very little or no fluoresce
was noted when the concentration was above the compounds’ MIC
indicating cell death as a result of the inhibited redox processes.
Combining the results from Sytox dye experiments, it confirms that
S. aureus
E. coli
0.064 µg/mL
8 µg/mL
2
2
hrs
hrs
6
6
hrs
hrs
Figure 5. Bacteria treated with neomycin (2 mg/mL) and CTC dye.
Figure 7. S. aureus treated with 3a (NQM108) and CTC dye.

6356
K. Y. Chan et al. / Bioorg. Med. Chem. Lett. 21 (2011) 6353–6356
2. (a) Elchert, B.; Li, J.; Wang, J.; Hui, Y.; Rai, R.; Ptak, R.; Ward, P.; Takemoto, J. Y.;
Acknowledgments
Bensaci, M.; Chang, C.-W. T. J. Org. Chem. 2004, 69, 1513; (b) Wang, J.; Li, J.;
Chen, H.-N.; Chang, H.; Tanifum, C. T.; Liu, H.-H.; Czyryca, P. G.; Chang, C.-W. T.
J. Med. Chem. 2005, 48, 6271; (c) Rai, R.; Chen, H.; Czyryca, P. G.; Li, J.; Chang, C.-
W. T. Org. Lett. 2006, 8, 887; (d) Zhang, J.; Chiang, F.-I.; Wu, L.; Czyryca, G. P.; Li,
D.; Chang, C.-W. T. J. Med. Chem. 2008, 51, 7563.
We thank the financial support from USTAR TCG grant and
URCO at USU. We also thank Professor Takemoto from Department
of Biology at USU for his valuable suggestion and discussion in
using the fluorescence dyes. We thank the technical support from
Mr. Sanjib Shrestha at Biology Department, and Professor Dong
Chen at the USTAR Synthetic Bio Manufacturing Center (SBC),
USU for allowing the access of fluorescence microscope.
3. Zhang, J.; Redman, N.; Litke, P.; Zeng, J.; Zhan, J.; Chan, K. Y.; Chang, C.-W. T.
Bioorg. Med. Chem. 2011, 19, 498.
4. Zhang, J.; Chang, C.-W. T. J. Org. Chem. 2009, 74, 4414.
5. For examples: (a) Thorsteinsson, T.; Másson, M.; Kristinsson, K. G.;
Hjálmarsdóttir, H. H.; Loftsson, T. J. Med. Chem. 2003, 46, 4173; (b) Maeda, S.;
Kita, T.; Meguro, K. J. Med. Chem. 2009, 52, 597; (c) Kanazawa, A.; Ikeda, T.;
Endo, T. Antimicrob. Agents Chemother. 1994, 38, 945.
6. (a) Wallace, K. B.; Starkov, A. A. Annu. Rev. Pharmacol. Toxicol. 2000, 40, 353; (b)
Phelpsf, D. C.; Crane, F. L. Biochemistry 1975, 14, 116.
Supplementary data
7. Ioannou, C. J.; Hanlon, G. W.; Denyer, S. P. Antimicrob. Agents Chemother. 2007,
51, 296.
Supplementary data (experimental procedures, spectroscopic
information for the synthesized compounds and additional photos
from the antibacterial experiments) associated with this article can
be found, in the online version, at doi:10.1016/j.bmcl.2011.08.107.
8. For examples of using Sytox dye: (a) Makovitzki, A.; Avrahami, D.; Shai, Y. Proc.
Natl. Acad. Sci. U.S.A. 2006, 103, 15997; (b) Roth, B. L.; Poot, M.; Yue, S. T.;
Millard, P. J. Appl. Environ. Microbiol. 1997, 63, 2421; (c) Chang, C.-W. T.; Fosso,
M.; Kawasaki, Y.; Shrestha, S.; Bensaci, M. F.; Wang, J.; Evans, C. K.; Takemoto, J.
Y. J. Antibiot. 2010, 63, 667.
9. For examples of using CTC dye: (a) Bovill, R. A.; Shallcross, J. A.; Mackey, B. M. J.
Appl. Bacteriol. 1994, 77, 353; (b) Rodriguez, G. G.; Phipps, D.; Ishiguro, K.;
Ridgway, H. F. Appl. Environ. Microbiol. 1992, 58, 1801; (c) Kalyuzhnaya, M. G.;
Lidstrom1, M. E.; Chistoserdova, L. ISME J. 2008, 2, 696.
10. Lee, M.; Hesek, D.; Mobashery, S. ACS Symp. Ser. 2008, 990, 54.
11. These can be found in Supplementary data.
References and notes
1. For reviewing: (a) Dan, I.; Andersson, D. I.; Hughes, D. Nat. Rev. Microbiol. 2010,
8, 260; (b) Davies, J.; Davies, D. Microbiol. Mol. Biol. Rev. 2010, 74, 417; (c)
Hellinger, W. C. South. Med. J. 2000, 93, 848; (d) Siegel, R. E. Respir. Care 2008,
53, 471.