Synthesis and bioactivity investigation of quinone-based dimeric cationic triazolium amphiphiles selective against resistant fungal and bacterial pathogens

Article abstract of DOI:10.1016/j.ejmech.2016.12.008

A series of synthetic dimeric cationic anthraquinone analogs (CAAs) with potent antimicrobial activities against a broad range of fungi and bacteria were developed. These compounds were prepared in 2–3 steps with high overall yield and possess alkyl chain, azole, quinone, and quaternary ammonium complexes (QACs). In vitro biological evaluations reveal prominent inhibitory activities of lead compounds against several drug-susceptible and drug-resistant fungal and bacterial strains, including MRSA, VRE, Candida albicans and Aspergillus flavus. Mode of action investigation reveals that the synthesized dimeric CAA's can disrupt the membrane integrity of fungi. Computational studies reveal possible designs that can revive the activity of QACs against drug-resistant bacteria. Cytotoxicity assays in SKOV-3, a cancer cell line, show that the lead compounds are selectively toxic to fungi and bacteria over human cells.

Full text of DOI:10.1016/j.ejmech.2016.12.008

European Journal of Medicinal Chemistry 126 (2017) 696e704
Contents lists available at ScienceDirect
European Journal of Medicinal Chemistry
journal homepage: http://www.elsevier.com/locate/ejmech
Research paper
Synthesis and bioactivity investigation of quinone-based dimeric
cationic triazolium amphiphiles selective against resistant fungal and
bacterial pathogens
Jaya P. Shrestha a, Coleman Baker a, Yukie Kawasaki b, Yagya P. Subedi a,
,
*
Nzuwah Nziko Vincent de Paul a, Jon Y. Takemoto b, Cheng-Wei Tom Chang a
a Department of Chemistry and Biochemistry, Utah State University, 0300 Old Main Hill, Logan, UT 84322-0300, USA
b Department of Biology, Utah State University, 5305 Old Main Hill, Logan, UT 84322-5305, USA
a r t i c l e i n f o
a b s t r a c t
Article history:
A series of synthetic dimeric cationic anthraquinone analogs (CAAs) with potent antimicrobial activities
against a broad range of fungi and bacteria were developed. These compounds were prepared in 2e3
steps with high overall yield and possess alkyl chain, azole, quinone, and quaternary ammonium com-
plexes (QACs). In vitro biological evaluations reveal prominent inhibitory activities of lead compounds
against several drug-susceptible and drug-resistant fungal and bacterial strains, including MRSA, VRE,
Candida albicans and Aspergillus flavus. Mode of action investigation reveals that the synthesized dimeric
CAA's can disrupt the membrane integrity of fungi. Computational studies reveal possible designs that
can revive the activity of QACs against drug-resistant bacteria. Cytotoxicity assays in SKOV-3, a cancer cell
line, show that the lead compounds are selectively toxic to fungi and bacteria over human cells.
© 2016 Published by Elsevier Masson SAS.
Received 1 September 2016
Received in revised form
19 November 2016
Accepted 2 December 2016
Available online 5 December 2016
Keywords:
Cationic anthraquinone analogs
Triazole antibiotics
1,4-Naphthoquinone analogs
MRSA
Multi-cationic
Quaternary ammonium compound
1. Introduction
bilayer to compromise cell membrane has been attributed as the
primary mode of action [8].
Amphiphilic molecules have attracted great interest as anti-
fungal or antibacterial agents for many years. Recently, many
studies on aminoglycoside-based amphiphilic molecules have
been published (Fig. 1) [1e3]. The protonated amines on an ami-
noglycoside form a cationic head and together with a long alkyl
chain form an amphiphilic molecule. Similarly, amphiphilic mol-
ecules based on quaternary ammonium complexes (QAC) and
hydrophobic alkyl chain have been known for many years, and it is
one of the most common antimicrobial agents used as disinfec-
tants (Fig. 1) [4,5]. They are widely used in domestic, industrial,
agricultural and clinical applications [6,7]. Several different types
of QACs either possessing mono- or multi-cationic functional
groups can be obtained in the market. Due to the strong positively
charged cationic head and long lipophilic tail, these compounds
act as amphiphilic molecules and are similar to phospholipids
found in the cell membrane. Disruption of the phospholipid
Our laboratory is dedicated to synthesizing cationic amphiphiles
based on aminoglycoside and 1,4-naphthoquionone moieties that
resemble QACs [9,10]. These compounds, called cationic anthra-
quinone analogs (CAAs), are antimicrobial when substituted with
an alkyl chain [10], but anticancer when substituted with the aro-
matic ring [11]. The antimicrobial CAAs, which are much more
active against Gram positive (Gþ) bacteria than Gram negative(G-)
bacteria, act as redox inhibitors at minimum inhibitory concen-
trations (MICs) and act as membrane disrupters at higher concen-
trations [12]. The anticancer CAAs act as ROS-generating molecules
[13]. Our studies have shown that the biological activities of CAAs
are tunable depending on the attached substituents. Dequalinium
chloride which contains two cationic motifs has been noted
recently for its new antimicrobial applications different from
mono-cationic QACs [14]. Inspired by the design of dequalinium
chloride, we decided to explore the activities of the dimeric form of
CAAs (Fig. 2). In this article, we are describing novel small multi-
cationic QACs which possess two redox active cationic anthraqui-
none heads and two alkyl lipophilic tails.
* Corresponding author.
E-mail address: tom.chang@usu.edu (C.-W.T. Chang).
http://dx.doi.org/10.1016/j.ejmech.2016.12.008
0223-5234/© 2016 Published by Elsevier Masson SAS.

J.P. Shrestha et al. / European Journal of Medicinal Chemistry 126 (2017) 696e704
697
Fig. 1. Cationic amphiphiles.
Fig. 2. Structural comparison of CAAs.
2. Results and discussion
2.2. Antibacterial activity
2.1. Design and chemical synthesis
The newly synthesized compounds were tested for antibacterial
and antifungal activities. The antibacterial activities of the dimeric
CAAs against Staphylococcus aureus (ATCC25923 Gþ), Methicillin-
resistant S. aureus (MRSA) (ATCC33591, Gþ) and Escherichia coli
(ATCC25922, G-) using vancomycin and kanamycin A as controls
are summarized in Table 1. No antibacterial activity was observed
for intermediate compound 3 at 5 mg/mL, which emphasizes the
importance of the cationic property and alkylation. The final
compounds (3a-3f) exhibited strong activities against drug sus-
ceptible S. aureus (ATCC25923) with compounds 3a-3c being the
The redox-active quinone heads are connected with an alkyl
linker via triazole moiety. Alkylation at the triazole ring generates
the cationic property, and the positive charge generated by
alkylation is delocalized around the triazole ring. The molecule is
hybrid in nature as it is the combination of several popular anti-
microbial scaffolds, such as quinone [15e18], triazole [19e21], and
QAC [22].
The synthesis began with cyclization/oxidation using 1,4-
naphthoquinone and 1,6-diazidohexane, 2 following the reported
procedure to form 3 (Scheme 1). The linker, 2 can be prepared by a
substitution reaction using 1,6-dibromohexane or 1,6-
ditosylhexane and sodium azide. Dialkylation of 3 using alkyl tri-
flates with various chain lengths, including methyl, ethyl, butyl,
hexyl, octyl and decyl, provided the desired dimeric forms of CAAs.
The alkyl triflates are commercially available or can be synthesized
from the triflation of the corresponding alcohols. Because we
observed solubility issues for the dimeric CAAs with chain length
longer than decyl group, no derivatives with carbon chain longer
than C10 were synthesized. All the compounds (3a-f) were char-
acterized by 1H NMR, 13C NMR, and high-resolution mass spec-
trometry. HPLC analysis determined the purity level of all the final
compounds that were found to be >95% pure. It is worth
mentioning that all of the compounds can be synthesized from a
low-cost starting material, 1,4-naphthoquinone, and do not require
cumbersome column chromatography purification. Therefore, large
quantity of lead compounds can be synthesized easily for future
applications.
most active (MIC 0.125
MIC value (0.032e0.064
m
g/mL). The MIC value is higher than the
mg/mL) for mono-cationic compound
(NQM108, Fig. 2); the drop-off in MIC value is most likely due to
increase in molecular weight of di-cationic compounds. The MICs
decreased by 32e64 fold as the alkyl chain length increased to
hexyl (C6) or longer. Interestingly, the trend was observed in
reverse order for S. aureus (MRSA) (ATCC33591) and E. coli
(ATCC25922). In both of the latter cases, the MICs for 3a-3c (C1, C2
and C4) were >250
mg/mL while 3d-3f (C6, C8 and C10) were more
active with MICs ranging from 4 to 16
mg/mL.
We have reported that antibacterial CAAs may exert two types of
antibacterial mode of actions: 1) interference with bacterial redox
processes at MIC levels, and 2) membrane disruption at higher
concentration [12]. Since increasing alkyl length will enhance the
lipophilicity of the dimeric CAAs, the tendency to cause membrane
disruption by these compounds is predicted to increase. Therefore,
we conclude that 3a-3c manifest antibacterial activities by inhib-
iting redox processes in S. aureus. However, this mode of action
does not appear to apply to resistant S. aureus (MRSA) and G- E. coli

698
J.P. Shrestha et al. / European Journal of Medicinal Chemistry 126 (2017) 696e704
Scheme 1. Synthesis of the dimeric CAAs.
Table 1
Antibacterial activity (MICs) of the dimeric CAAs (mg/mL).
Compound
MIC(mg/mL)
S. aureusa
MRSAb
E. faecalisc
E. colid
P. aeruginosae
K. pneumoniaef
3a
3b
3c
3d
3e
3f
0.125
0.125
0.125
4
4
8
1
2e4
0.5
>250
>250
>250
8e16
4e8
4e8
2e4
125e250
16
16
32
16
4
1
1
125
250
>250
>250
>250
>250
8e16
2e4
4e8
64e125
4
250
250
32
16
4
250
250
250
250
16
32
>250
1
4
Vancomycin
Kanamycin A
Amikacin
>250
125
1
1
1
a
Staphylococcus aureus (ATCC25923).
Staphylococcus aureus (ATCC 33591) (MRSA).
E. faecalis (ATCC51299) (VRE).
Escherichia coli (ATCC 25922).
Pseudomonas aeruginosa (ATCC 27853).
Klebsiella pneumoniae (ATCC 13883).
b
c
d
e
f
(more details will be discussed later). Only when the chain length
and the associated lipophilicity of the dimeric CAAs increase do
these compounds (3e-3f) become bactericidal through membrane
disruption. In other word, compounds 3e-3f become broad spec-
trum active against a wide range of bacteria via, likely, membrane
disruption mode of action. To further exemplify our observation,
more bacteria, including Enterococcus faecalis (ATCC51299) (van-
comycin resistant), Pseudomonas aeruginosa (ATCC 27853) and
Klebsiella pneumonia (ATCC 13883), were examined and com-
pounds 3e-3f still possess significant activity against these strains
of pathogens.
candidiasis caused by strains of the genus Candida [25]. The dimeric
CAAs possess both azole moieties and QAC properties, which
prompted us to explore their in vitro antifungal activity. The com-
pounds were tested against major human pathogenic fungi,
including Aspergillus flavus, Candida albicans 64124 (azole-resis-
tant), Candida albicans MYA2876 (azole-susceptible), Cryptococcus
neoformans and Rhodotorula pilimanae using K20, and fluconazole
as the controls. The MIC values of 3a-3f, K20 [1], and fluconazole
against various fungi are summarized in Table 2. Compounds 3a-3d
did not exhibit antifungal activity against A. flavus, C. albicans
64124, and C. albicans MYA2876. However, they exhibited mild to
moderate antifungal activities against C. neoformans and R. pilima-
nae with MIC values ranging from 125 mg/mL and 62.5e15.6 mg/mL,
respectively. Both 3e and 3f exhibited excellent antifungal activities
2.3. Antifungal activity
against all tested fungal species with the MIC values ranging from
Azoles and QAC's are also known to have potent antifungal ac-
tivities [23,24]. Azole-based drugs, such as fluconazole, pos-
aconazole, and itraconazole are considered first line treatments for
15.63 to 7.81
mg/mL, 3.91 mg/mL, 3.91 mg/mL, 3.91e2.0 mg/mL, and
2.0 g/mL for A. flavus, C. albicans 64124, C. albicans MYA2876,
m

J.P. Shrestha et al. / European Journal of Medicinal Chemistry 126 (2017) 696e704
699
Table 2
Antifungal activity of the dimeric CAAs (
m
g/mL).
Candida albicans 64124
Compound
Aspergillus flavus
Candida albicans MYA2876
Cryptococcus neoformans
Rhodotorula pilimanae
3a
3b
3c
3d
3e
3f
>250
>250
>250
>250
15.63
7.81
>250
>250
250
>250
>250
>250
>250
3.91
3.91
15.6
1.56
125
e
62.5
e
e
e
250
125
3.9
2.0
7.8
e
15.6
2.0
2.0
e
3.91
3.91
31.3
>250
K20
Fluconazole
>250
e
e
C. neoformans, and R. pilimanae, respectively. In general, compound
3f is the most active compound followed by 3e, and then by 3d. The
results showed that antifungal activity depended closely on the
alkyl chain length. Aspergillus species cause invasive aspergillosis in
immunocompromised individuals [26], and A. flavus is one of the
leading causes of the reduction in crop yield. The latter is know to
produce aflatoxin, which is highly potent carcinogen and regulated
in many countries [27]. Similarly, C. albicans is the most prevalent of
fungal pathogens. Therefore, the present results are very significant
as both 3e and 3f are active against Aspergillus species and azole-
resistant C. albicans.
red to which compound 3a is docked is shown Fig. 5. The docking
predicted that a similar binding mode to the dequalinium chloride-
QacR binding with the ligands completely in the electron rich site of
the active site as shown in the superposition picture in Fig. 5 (right).
In addition to the electrostatic interactions between compound 3a
and the active site, the
pꢀp interactions also exist between the
aromatic moieties of 3a: one interacts with the aromatic residues of
Y103, and the other interacts with W61 in the active site. Similar
interactions were observed with dequalinium chloride suggesting
that compound 3a will induce a similar conformational change as
dequalinium chloride does and, hence, trigger the production of
QacA efflux pump, which leads to the inactivation of 3a.
From the docking experiment, a drastic difference was noted
between the binding of compound 3f toward QacR (Fig. 6). First of
all, the arrangements of two cationic anthraquinone rings are
completely different from that of the compound 3a and dequali-
2.4. Fungal mechanism of action
Disruption of the cell membrane is the primary mode of action
for antimicrobial amphiphiles including aminoglycoside amphi-
philes and QAC amphiphiles. The effect of our most active anti-
fungal candidate, 3f, was used to examine its effect on the fungal
membrane. Sytox Green [28], a dye becomes fluorescent upon
penetrating membrane-compromised cells and binds to nucleic
acid, was used to investigate the cell membrane disruption. The
study was performed on C. albicans ATCC 64124 and filamentous
fungus, Fusarium graminearum B4-A5. The fungi were exposed to 3f
at MIC levels for 1 h. As expected, the compound functioned as a
membrane disrupter in both fungi (Figs. 3 and 4). The green fluo-
rescence found inside the fungal cells resulting from the treatment
of compound 3f clearly showed that compound 3f exert its anti-
fungal activity via a pore formation mechanism.
nium chloride. One ring may have significant
p p interaction with
ꢀ
Y103 while the other is interacts with Y123 leaving W61 out of the
range of interaction with 3f. The presence of alkyl groups on 3f
(decyl groups) clearly intrude into the binding pocket of QacR and,
thus, may explain the dramatic difference in the binding of 3f vs.
the binding of 3a and dequalinium chloride.
Based on the results from the docking experiment, proposed are
several reasons that can be attributed toward the activities of our
dimeric CAAs against MRSA and other bacteria. Firstly, the signifi-
cant changes in binding conformation as a result of the extended
alkyl chains when compound 3f is bound may not trigger the
release of QacR and, thus, the expression of efflux QacA. In our
previous studies, we have also noticed that NQM108 of about 8e16
folds more active than the one with only methyl group (NQM111)
against MRSA (ATCC33591) (Fig. 7) [32]. Hence, the second reason
for the activity of compound 3f may be due to the presence of decyl
chains that immobilize the molecules in the bacterial membrane,
and exert the antibacterial effect via inhibition of redox processes
or pore formation. Thirdly, the presence of redox active cationic
anthraquinone core may augment the damage toward the bacterial
membrane and lead to the cell death, especially for the G-bacteria.
Finally, when there is no advantage of lipophilicity (shorter alkyl
chains), these CAAs are more active against susceptible Gþ bacteria
via the inhibition of bacterial redox processes as reported [12].
Although an exact antibacterial mode of action of our dimeric
CAAs cannot be pinpointed, we can conclude that the attachment of
longer alkyl chains will change the behavior of QACs toward the
binding of QacR and may revive the activity of traditional QACs.
With the long alkyl group, the redox active CAAs core could exert
oxidative damage toward various microbe pathogens. Many G-
pathogens, such as Pseudomonas and Burkholderia sp., are equip-
ped with diverse efflux pumps that are known to be MDR. The
revealed insight in this work may offer new strategies for battling
these drug resistant pathogens. It is likely that a combination of
redox property and lipophilicity is responsible for the antifungal
activity of the dimeric CAAs as well.
2.5. Computational binding study with QAC and QacR
The overall structure of the dimeric CAAs is very similar to
dequalinium chloride (Fig. 1), a commercial multi-cationic QAC.
Dequalinium chloride is known to trigger bacterial resistance
involved in the qacA/qacR regulation system [29]. QacR is a nega-
tive transcriptional regulator (repressor) of qacA/qacR genes that
are coded for QacA, a membrane-affiliated efflux pump responsible
for pumping of structurally diverse QACs. Binding of QACs to QacR
induces a conformational change in the protein which eventually
lead to the dissociation of QacR from binding toward the DNA and
thereby enabling the synthesis of QacA, which pumps QACs out of
the cells and renders the bacteria become resistant toward various
QACs, including dequalinium chloride.
MRSA is equipped with various drug resistant genes/mecha-
nisms, including QacR, which is known to confer multi-drug
resistant (MDR) to many agents including QACs [30]. Neverthe-
less, several of our dimeric CAAs are active against MRSA. There-
fore, we were interested in revealing the factors that enable our
dimer CAAs to be active against MRSA where traditional QACs
remain inactive. We employed the reported X-ray structure of QacR
for docking experiments [31].
The electrostatic map of QacR with an electron rich active site in

700
J.P. Shrestha et al. / European Journal of Medicinal Chemistry 126 (2017) 696e704
Fig. 3. Candida albicans, single hypha experiment (Top) Control, no compound added. (Bottom) Incubated with 3f for 1 h.
Fig. 4. Fusarium graminerium single hypha experiment (Top) Control, no compound added. (Bottom) Incubated with 3f for 1 h.
2.6. Cytotoxicity
At a concentration of 10
other derivatives displayed no toxicity. All the MIC values for
antifungal activity are lower than 10 g/mL except the 3e MIC for
A. flavus. The most active compounds 3c and 3d showed the least
toxicity with IC50 value of 115.25 13.45 and 108.74 7.24 g/mL,
respectively. The IC50 value calculated for the most active antifungal
compounds 3e and 3f are 74.03 5.55 and 68.42 9.35 g/mL,
mg/mL, 3a displayed mild toxicity while
A cytotoxicity assay was carried out against a human ovary
cancer cell, SKOV3 to evaluate the cytotoxicity of our new com-
pounds. This assay is critical as it determines the activity of these
compounds against desired pathogens over human cells. The re-
sults are shown in Fig. 8 and all IC50 values are calculated in Table 3.
m
m
m

J.P. Shrestha et al. / European Journal of Medicinal Chemistry 126 (2017) 696e704
701
Fig. 5. (Left) QacR cavity with bound 3a. (Right)
interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)
p
ꢀ
p interaction of aromatic rings of 3a (Pink) and dequalinium chloride (Purple) with QacR aromatic amino acid residue. (For
Fig. 6. (Left) QacR cavity with bound 3f. (Right)
pꢀp interaction of aromatic rings of 3f.
shorter alkyl chain were very active against S. aureus while com-
pounds substituted with longer alkyl chain were active against
E. coli, MRSA, VRE and various strains of fungi. The cytotoxicity
assay showed that these dimeric CAAS manifest little to no cyto-
toxicity against a human cell line. Mode of action investigation
reveals that the active dimeric CAAs target fungal membranes as at
least one mode of action. Computational studies suggest that the
increase in alkyl chain length on dimeric CAAs can avoid the in-
duction of efflux pumps, which are major causes of bacterial
resistance against commonly used QACs. These results can pave the
way to the development of new QAC-based antibacterial and
antifungal agents to help overcome the problem of microbial
resistance.
Fig. 7. Structure of NQM111.
respectively. These values are at least 17 times higher than their
antifungal activity for C. albicans.
4. Experimental section
Proton nuclear magnetic resonance spectra were recorded using
JOEL 300 or Bruker ascend 500 spectrometers. Chemical shifts were
reported in parts per million (ppm) downfield from tetrame-
3. Conclusions
In conclusion, we successfully synthesized a new series of QAC
type of antimicrobial agents, dimeric CAAs. The developed syn-
thesis is concise and has high overall yield, which makes it suitable
for the scale-up production of the lead compounds. Our previously
reported CAAs are active against Gþ bacteria and cancer cells but
not against G-bacteria and fungi. Nevertheless, the dimeric CAAs
display strong activities against G-bacteria and fungi which offer
complementary bioactivities. These dimeric CAAs substituted with
thylsilane in d unit and coupling constants were given in cycles per
seconds (Hz). Signal multiplicities were indicated by s (singlet),
d (doublet), t (triplet), q (quartet), and m (multiplet). 13C NMR
spectra were obtained using JOEL 300 at 75 MHz or Bruker ascend
500 at 125 MHz. Routine 13C NMR spectra were fully decoupled by
broadband WALTZ decoupling. All NMR spectra were at ambient
temperature. Atmospheric pressure chemical ionization (APCI) or
electron spray ionization (ESI) was provided by the mass

702
J.P. Shrestha et al. / European Journal of Medicinal Chemistry 126 (2017) 696e704
Fig. 8. Cytotoxicity assay for all dimeric CAAs analogs.
Table 3
IC50 for the dimeric CAAs.
4.3. General procedure for N-3 alkylation: synthesis of 3b-3f
The alcohol (4 equiv) and pyridine (8 equiv) were dissolved in
anhydrous toluene (10 mL) and cooled in an iceꢀwater bath before
Tf2O (8 equiv) was slowly added. The mixture was stirred at 0 ꢁC for
2 h. The reaction mixture was passed through a sodium sulfate
filter. The sodium sulfate filter cake was further washed with
another 10 mL of anhydrous toluene. The toluene fractions were
combined, and 3 (0.11 g, 1 equiv) was then added. This mixture was
then refluxed at 110 ꢁC for overnight. After removal of solvent,
diethyl ether (25 mL) was added and the reaction mixture was
stirred for 30 min. The precipitated products were collected with
Hirsh funnel, washed with more diethyl ether (25 mL), and dried
under the vacuum.
Compound
IC50 (mg/mL)
3a
3c
3d
3e
3f
46.14 13.25
115.25 13.45
108.74 7.24
74.03 5.55
68.42 9.35
spectrometry facilities, University of California, Riverside.
Chemical reagents and chromatography solvents were pur-
chased from Aldrich Chemical Co. or Acros Chemical Co. and were
used without further purification unless otherwise noted. Sytox
green was purchased from Invitrogen.
4.4. 1,1'-(hexane-1,6-diyl)bis(1H-naphtho[2,3-d][1,2,3]triazole-4,9-
dione) (3)
4.1. General procedure for [3þ2] cycloaddition/oxidation: synthesis
of compound 3
This compound was synthesized according to the general pro-
To a solution of alkyl azides (estimated 3.2 mmol) in DMF
(10 mL), 1,4-naphthoquinone (2.0 g, 12.8 mmol) was added, and the
resulting solution was heated at 80 ꢁC for two days. After removal of
solvent, diethyl ether (25 mL) was added and the reaction mixture
was stirred until the products precipitated. The precipitated prod-
ucts were collected with Hirsh funnel, washed with more diethyl
ether (25 mL), and dried under vacuum. The yields were calculated
based on the estimated amount of alkyl azides used.
cedures with 84% yield. Yield ¼ 84%, color ¼ brownish grey, Melting
Point ¼ 208e210 ꢁC. 1H NMR (300 MHz, CDCl3)
d 8.3e8.4 (m, 2H),
8.2e8.3 (m, 2H), 7.8e7.9 (m, 4H), 4.8 (t, J ¼ 7.2 Hz, 4H), 2.0e2.1 (m,
4H), 1.5e1.6 (m, 4H).; 13C NMR (75 MHz, CDCl3)
d 176.9 (2C), 175.6
(2C), 145.7 (2C), 135.3 (2C), 134.4 (2C), 133.5 (2C), 133.4 (2C), 127.9
(2C), 127.5 (2C), 127.2 (2C), 50.4 (2C), 29.9 (2C), 25.9 (2C).; ESI/APCI
calcd for C26H20N6Oþ4 m/z ([MþH]þ) 481.1619; measured ([MþH]þ)
481.1639.
4.2. Synthesis of 3a
4.5. 1,1'-(hexane-1,6-diyl)bis(3-methyl-4,9-dioxo-4,9-dihydro-1H-
To a solution of compound 3 (0.35 mmol) in anhydrous toluene
(10 mL), methyl trifluoromethanesulfonate (0.30 mL, 2.80 mmol)
was added, and the resulting solution was heated at 100 ꢁC over-
night. After removal of solvent, diethyl ether (25 mL) was added
and the reaction mixture was stirred for 30 min. The precipitated
products were collected with Hirsh funnel, washed with more
diethyl ether (25 mL), and dried under the vacuum.
naphtho[2,3-d][1,2,3]triazol-3-ium) triflate (3a)
This compound was synthesized according to the general pro-
cedures. Yield
¼
96%, color
¼
brownish grey, Melting
Point ¼ 222e224 ꢁC. 1H NMR (300 MHz, CD3OD)
d 8.4e8.3 (m, 4H),
8.1e8.0 (m, 4H), 5.1 (t, J ¼ 6.9 Hz, 4H), 4.7 (s, 6H), 2.2e2.1 (m, 4H),
1.5e1.6 (m, 4H); 13C NMR (125 MHz, DMSO - D6)
d 173.2 (2C), 173.0

J.P. Shrestha et al. / European Journal of Medicinal Chemistry 126 (2017) 696e704
703
(2C), 136.9 (2C), 136.4 (2C), 135.9 (2C), 132.9 (2C), 132.6 (2C), 128.3
(4C), 40.9 (2C), 28.6 (2C), 25.2 (2C); ESI/APCI calcd for C28H26N6O4þþ
([M]þþ) m/z 510.2016; measured m/z 255.1002 (Mþþ/2).
136.8 (4C), 136.1 (4C), 132.7 (4C), 128.3 (4C), 128.1 (2C), 54.2 (2C),
54.1 (2C), 31.7 (2C), 29.4 (2C), 29.1 (2C), 28.7 (4C), 28.5 (2C), 25.8
(2C), 25.1 (2C), 22.6 (2C), 14.4 (2C).; ESI/APCI calcd for C46H62N6O4þþ
([M]þþ) m/z 762.4822; measured m/z 381.2395 (Mþþ/2).
4.6. 1,1'-(hexane-1,6-diyl)bis(3-ethyl-4,9-dioxo-4,9-dihydro-1H-
naphtho[2,3d][1,2,3]triazol-3-ium) triflate (3b)
4.11. General procedure for cell culture
This compound was synthesized according to the general pro-
SKOV-3 cells were grown in commercial DMEM 1X (Gibco) with
10% fetal bovine serum (FBS, Gibco), 100 U/mL penicillin, and
cedures. Yield
¼
90%, color
¼
brownish grey, Melting
Point ¼ 204e206 ꢁC. 1H NMR (300 MHz, DMSO - D6)
d
8.4e8.2 (m,
100 m
g/mL streptomycin at 37 ꢁC and 5% CO2. The cells were
4H), 8.1e8.0 (m, 4H), 5.0e5.1 (m, 8H), 2.1e1.9 (m, 4H), 1.60 (t,
J ¼ 7.2 Hz, 6H), 1.5e1.4 (m, 4H); 13C NMR (125 MHz, DMSO - D6)
allowed to adhere for 48 h before drug treatment.
d
173.9 (4H), 137.7 (2C), 137.6 (2C), 136.9 (2C), 136.8 (2C), 133.5 (4C),
4.12. General procedure for cell viability assay
129.1 (2C), 129.0 (2C), 54.9 (2C), 50.1 (2C), 29.4 (2C), 25.9 (2C), 15.1
(2C); ESI/APCI calcd for C30H30N6Oþ4 þ ([M]þþ) m/z 538.2329;
measured m/z 269.1146 (Mþþ/2).
The cells were seeded in 96-well microtiter plates (5000/
200
mL). After 48 h of incubation in the corresponding media, cells
were treated with various concentrations of test compounds (0.0,
4.7. 1,1'-(hexane-1,6-diyl)bis(3-butyl-4,9-dioxo-4,9-dihydro-1H-
naphtho[2,3-d][1,2,3]triazol-3-ium) triflate (3c)
0.10, 1.0, 10.0, and 100.0 mM) for 24 h. 20 mL of MTT stock solution
(5 mg/mL) was added to each well and incubated for 4 h at 37 ꢁC.
Upon completion of incubation, the media was carefully removed
This compound was synthesized according to the general pro-
and washed twice with 100 mL of PBS buffer. Then, 100 mL of DMSO
cedures. Yield
¼
73%, color
¼
black/grey, Melting
8.4e8.2 (m, 4H),
was added to each well; agitated on an orbital shaker for 15 min,
and the absorbance at 570 nm with 630 nm filter was determined
with a microplate reader. The results were expressed as viability
compared with that of control. The experiment was carried out in
triplicate in three independent experiments.
Point ¼ 160e161 ꢁC. 1H NMR (300 MHz, DMSO)
d
8.1e8.0 (m, 4H), 5.0e5.1 (m, 8H), 2.1e1.9 (m, 8H), 1.5e1.3 (m, 8H),
0.90 (t, J ¼ 7.6 Hz, 6H); 13C NMR (125 MHz, DMSO)
d 172.6 (2C),
172.5 (2C), 135.9 (2C), 135.8 (2C), 135.7 (4C), 132.6 (2C), 132.5 (2C),
127.7 (2C), 127.6 (2C), 54. (2C), 53.9 (2C), 30.4 (2C), 28.2 (2C), 25.1
(2C), 19.1 (2C), 12.2 (2C); ESI/APCI calcd for C34H38N6Oþ4 þ ([M]þþ
)
4.13. General procedure for sytox dye experiment
m/z 594.2955; measured m/z 297.1462 (Mþþ/2).
C. albicans (5 ꢂ 105 CFU/mL) or F. graminearum B4-5A (5 ꢂ 105
4.8. 1,1'-(hexane-1,6-diyl)bis(3-hexyl-4,9-dioxo-4,9-dihydro-1H-
conidia/mL) were grown for 18 h in PDB with continuous agitation.
naphtho[2,3-d][1,2,3]triazol-3-ium) triflate (3d)
Aliquots (500
10,000ꢂg. The fungal pellet was suspended in 10 mM HEPES, pH
7.4, centrifuged again, and suspended in 500 L distilled water. The
fungal cells were exposed to 3.91
g/mL 3f for 1 h at 28 ꢁC. The
treated fungi were assessed 10 min after addition of 0.5 L Sytox
Green (5 mM, Invitrogen). Glass slides were prepared with 10 L of
each mixture and observed in dark-field and fluorescence (excita-
tion and emission wavelength 470/40e535/40 nm) modes with an
Olympus IX70 fluorescence microscope (Olympus, Center Valley,
PA, USA).
mL) were taken and centrifuged for 2 min at
This compound was synthesized according to the general pro-
m
cedures. Yield
¼
88%, color
¼
brownish grey, Melting
m
Point ¼ 142e143 ꢁC. 1H NMR (300 MHz, DMSO)
d
8.4e8.2 (m, 4H),
m
8.1e8.0 (m, 4H), 5.0e5.1 (m, 8H), 2.1e1.9 (m, 8H), 1.5e1.2 (m, 16H),
m
0.84 (t, J ¼ 6.8 Hz, 6H); 13C NMR (125 MHz, DMSO)
d
173.1 (4C),
136.8 (4C), 136.1 (4C), 132.7 (4C) 128.3 (2C), 128.1 (2C), 54.2 (2C),
54.1 (2C), 30.9 (2C), 28.7 (2C), 28.5 (2C), 25.4 (2C), 25.2 (2C), 22.3
(2C), 14.3 (2C); ESI/I calcd for C38H46N6Oþ4 ([M]þþ) m/z 650.3581;
measured m/z 325.1773 (Mþþ/2).
4.14. Molecular docking study
4.9. 1,1'-(hexane-1,6-diyl)bis(3-octyl-4,9-dioxo-4,9-dihydro-1H-
naphtho[2,3-d][1,2,3]triazol-3-ium) triflate (3e)
Docking simulations were performed using AuotDock Vina [33]
while AutoDock Tools (ADT) [34,35] Version 1.5.6 was used for
parametrization. The protein crystal structure (QacR) used in the
docking simulation was obtained from the protein Data Bank with
Identification code 3BT9 (PDB ID: 3BT9) [31]. This protein was
prepared to be used as the receptor. The Kollman's charges were
added, water molecules were removed and the polar hydrogen
atoms were added. Compound 3a and 3f were used as the ligand.
Their structures were generated with the used of Gaussview 5.0
and optimized using Gaussian 09 [36] with the AM1 [37] methods.
The grid box size was set to cover 80% of the receptor with box size
set at 24.0 Å ꢂ 52.0 Å ꢂ 28.0 Å along the x, y and z axis. Ligand
flexible docking simulations were performed with the default value
of docking exhaustiveness parameter of 8 used. Binding modes
were ranked with the standard AutoDock Vina scoring function
weight settings for each energy term. Docking simulations were
repeated three times for each system to check the consistency of
the results. The result with the lowest docking energy analysis
(strong binding) rank 1 was used for further analysis. For visuali-
zation of the docked conformations, pymol [38] software package
was used.
This compound was synthesized according to the general pro-
cedures. Yield
¼
81%, color
¼
brownish grey, Melting
Point ¼ 126e128 ꢁC. 1H NMR (300 MHz, DMSO)
d 8.3e8.2 (m, 4H),
8.1e8.0 (m, 4H), 5.0e5.1 (m, 8H), 2.1e1.9 (m, 8H), 1.5e1.1 (m, 24H),
0.81 (t, J ¼ 6.9 Hz, 6H); 13C NMR (125 MHz, DMSO)
d 173.1 (4C),
136.9 (2C), 136.8 (2C), 136.1 (4C), 132.7 (4C), 128.3 (2C), 128.2 (2C),
54.2 (2C), 54.1 (2C), 31.6 (2C), 28.9 (2C), 28.8 (2C), 28.7 (2C), 28.5
(2C), 25.8 (2C), 25.1 (2C), 22.5 (2C), 14.4 (2C); ESI/APCI calcd for
C
42H54N6O4þþ ([M]þþ) m/z 706.9309; measured m/z 353.2083
(Mþþ/2).
4.10. 1,1'-(hexane-1,6-diyl)bis(3-decyl-4,9-dioxo-4,9-dihydro-1H-
naphtho[2,3-d][1,2,3]triazol-3-ium triflate (3f)
This compound was synthesized according to the general pro-
cedures. Yield
¼
77%, color
¼
brownish grey, Melting
Point ¼ 119e121 ꢁC. 1H NMR (500 MHz, DMSO)
d 8.4e8.3 (m, 4H),
8.2e8.1 (m, 4H), 5.0e5.1 (m, 8H), 2.0e1.9 (m, 8H), 1.5e1.3 (m, 32H),
0.81 (t, J ¼ 6.9 Hz, 6H); 13C NMR (125 MHz, DMSO)
d 173.1 (4C),

704
J.P. Shrestha et al. / European Journal of Medicinal Chemistry 126 (2017) 696e704
G. Meinardi, M. Di Somma, G. Tosolini, E. Vecchi, Studies on the antibacterial
Acknowledgement
and antifungal properties of 1, 4-naphthoquinones, Br. J. Pharmacol. 40 (1970)
871e880.
[17] M.d.P.S.B.C. Ferreira, M.F.d.C. Cardoso, F.d.C. da Silva, V.F. Ferreira, E.S. Lima,
J.V.B. Souza, Antifungal activity of synthetic naphthoquinones against der-
matophytes and opportunistic fungi: preliminary mechanism-of-action tests,
Ann. Clin. Microbiol. Antimicrob. 13 (2014) 1e6.
We acknowledge the support from NSF Award CHE-1429195 for
500 MHz Bruker NMR. We thank the support from the Department
of Chemistry and Biochemistry, Utah State University.
ꢀ
ꢀ
ꢀ
[18] J.M. Sanchez-Calvo, G.R. Barbero, G. Guerrero-Vasquez, A.G. Duran, M. Macías,
M.A. Rodríguez-Iglesias, J.M.G. Molinillo, F.A. Macías, Synthesis, antibacterial
and antifungal activities of naphthoquinone derivatives: a structureeactivity
relationship study, Med. Chem. Res. 25 (2016) 1274e1285.
Appendix A. Supplementary data
Supplementary data related to this article can be found at http://
dx.doi.org/10.1016/j.ejmech.2016.12.008.
[19] D.S. Schiller, H.B. Fung, Posaconazole: an extended-spectrum triazole anti-
fungal agent, Clin. Ther. 29 (2007) 1862e1886.
[20] L.R. Peyton, S. Gallagher, M. Hashemzadeh, Triazole antifungals: a review,
Drugs Today 51 (2015) 705e718.
References
[21] C. Lass-Florl, Triazole antifungal agents in invasive fungal infections:
a
comparative review, Drugs 71 (2011) 2405e2419.
[1] S.K. Shrestha, C.-W.T. Chang, N. Meissner, J. Oblad, J.P. Shrestha, K.N. Sorensen,
M.M. Grilley, J.Y. Takemoto, Antifungal amphiphilic aminoglycoside K20:
bioactivities and mechanism of action, Front. Microbiol. 5 (2014).
[2] C.W.T. Chang, J.Y. Takemoto, Antifungal amphiphilic aminoglycosides, Med-
ChemComm 5 (2014) 1048e1057.
[3] M. Fosso, M.N. AlFindee, Q. Zhang, V.d.P.N. Nziko, Y. Kawasaki, S.K. Shrestha,
J. Bearss, R. Gregory, J.Y. Takemoto, C.-W.T. Chang, Structureeactivity re-
lationships for antibacterial to antifungal conversion of kanamycin to
amphiphilic analogues, J. Org. Chem. 80 (2015) 4398e4411.
[4] J.J. Merianos, Quaternary ammonium antimicrobial compounds, Disinfect.
Steriliz. Preserv. (1991) 225e255.
[5] M. Frier, Derivatives of 4-amino-quinaldinium and 8-hydroxyquinoline, In-
hibition and Destruction of the Microbial Cell, Academic Press, Ltd., London,
England, 1971, pp. 107e120.
[6] D.L. Fredell, Biological properties and applications of cationic surfactants, in:
J. Cross, E.J. Springer (Eds.), Cationic Surfactants: Analytical and Biological
Evaluations, Marcel Dekker, New York, 1994, p. 31.
[7] M.C. Jennings, K.P.C. Minbiole, W.M. Wuest, Quaternary ammonium com-
pounds: an antimicrobial mainstay and platform for innovation to address
bacterial resistance, ACS Infect. Dis. 1 (2015) 288e303.
[8] S.P. Denyer, Mechanisms of action of antibacterial biocides, Int. Biodeterior.
Biodegrad. 36 (1995) 227e245.
[9] J.P. Shrestha, C.W. Chang, Safe and easy route for the synthesis of 1,3-
dimethyl-1,2,3-triazolium salt and investigation of its anticancer activities,
Bioorg Med. Chem. Lett. 23 (2013) 5909e5911.
ꢀ
[22] E. Obła˛k, A. Piecuch, A. Krasowska, J. Łuczynski, Antifungal activity of gemini
quaternary ammonium salts, Microbiol. Res. 168 (2013) 630e638.
[23] D.J. Sheehan, C.A. Hitchcock, C.M. Sibley, Current and emerging azole anti-
fungal agents, Clin. Microbiol. Rev. 12 (1999) 40e79.
[24] G.P. Bodey, Azole antifungal agents, Clin. Infect. Dis. 14 (1992) S161eS169.
[25] J.A. Vazquez, Optimal management of oropharyngeal and esophageal candi-
diasis in patients living with HIV infection, HIV AIDS 2 (2010) 89e101.
[26] V.R. Barrs, T.M. van Doorn, J. Houbraken, S.E. Kidd, P. Martin, M.D. Pinheiro,
M. Richardson, J. Varga, R.A. Samson, Aspergillus felis sp. nov., an emerging
agent of invasive aspergillosis in humans, cats, and dogs, PLoS One 8 (2013).
[27] M.A. Klich, Aspergillus flavus: the major producer of aflatoxin, Mol. Plant
Pathol. 8 (2007) 713e722.
[28] B.L. Roth, M. Poot, S.T. Yue, P.J. Millard, Bacterial viability and antibiotic sus-
ceptibility testing with SYTOX green nucleic acid stain, Appl. Environ.
Microbiol. 63 (1997) 2421e2431.
[29] B.A. Mitchell, M.H. Brown, R.A. Skurray, QacA multidrug efflux pump from
Staphylococcus aureus: comparative analysis of resistance to diamidines,
biguanidines, and guanylhydrazones, Antimicrob. Agents Chemother. 42
(1998) 475e477.
[30] S.S. Costa, M. Viveiros, L. Amaral, I. Couto, Multidrug efflux pumps in Staph-
ylococcus aureus: an update, Open Microbiol. J. 7 (2013) 59e71.
[31] K.M. Peters, J.T. Schuman, R.A. Skurray, M.H. Brown, R.G. Brennan,
M.A. Schumacher, QacR-cation recognition is mediated by a redundancy of
residues capable of charge neutralization, Biochemistry 47 (2008)
8122e8129.
[10] M.Y. Fosso, K.Y. Chan, R. Gregory, C.-W.T. Chang, Library synthesis and anti-
bacterial investigation of cationic anthraquinone analogs, ACS Comb. Sci. 14
(2012) 231e235.
[32] J. Zhang, N. Redman, A.P. Litke, J. Zeng, J. Zhan, K.Y. Chan, C.W. Chang, Syn-
thesis and antibacterial activity study of a novel class of cationic anthraqui-
none analogs, Bioorg Med. Chem. 19 (2011) 498e503.
[11] J.P. Shrestha, M.Y. Fosso, J. Bearss, C.W. Chang, Synthesis and anticancer
structure activity relationship investigation of cationic anthraquinone ana-
logs, Eur. J. Med. Chem. 77 (2014) 96e102.
[33] O. Trott, A.J. Olson, AutoDock Vina: improving the speed and accuracy of
docking with
a new scoring function, efficient optimization, and multi-
threading, J. Comput. Chem. 31 (2010) 455e461.
[12] K.Y. Chan, J. Zhang, C.W. Chang, Mode of action investigation for the anti-
bacterial cationic anthraquinone analogs, Bioorg. Med. Chem. Lett. 21 (2011)
6353e6356.
[34] G.M. Morris, R. Huey, W. Lindstrom, M.F. Sanner, R.K. Belew, D.S. Goodsell,
A.J. Olson, AutoDock4 and AutoDockTools4: automated docking with selective
receptor flexibility, J. Comput. Chem. 30 (2009) 2785e2791.
[13] J.P. Shrestha, Y.P. Subedi, L. Chen, C.-W.T. Chang, A mode of action study of
cationic anthraquinone analogs: a new class of highly potent anticancer
agents, MedChemComm 6 (2015) 2012e2022.
[14] W. Mendling, E.R. Weissenbacher, S. Gerber, V. Prasauskas, P. Grob, Use of
locally delivered dequalinium chloride in the treatment of vaginal infections:
a review, Arch. Gynecol. Obstet. 469 (2016) 469e484.
[15] A.M. Amani, Synthesis, characterization and antibacterial and antifungal
evaluation of some para-quinone derivatives, Drug Res. 64 (2014) 420e423.
[16] V. Ambrogi, D. Artini, I. De Carneri, S. Castellino, E. Dradi, W. Logemann,
[35] M.F. Sanner, Python: a programming language for software integration and
development, J. Mol. Graph Model 17 (1999) 57e61.
[36] M.J.T. Frisch, G.W. Schlegel, H.B. Scuseria, G.E. Robb, M.A. Cheeseman,
J.R. Scalmani, G. Barone, V. Mennucci, B. Petersson, G. A, Gaussian 09, Revision
C.01, in, Gaussian, Inc, Wallingford, CT, 2010.
[37] G.B. Rocha, R.O. Freire, A.M. Simas, J.J. Stewart, RM1: a reparameterization of
AM1 for H, C, N, O, P, S, F, Cl, Br, and I, J. Comput. Chem. 27 (2006) 1101e1111.
€
[38] The PyMOL Molecular Graphics System, in, Schrodinger, LLC.