Synthesis of nucleobase-neomycin conjugates and evaluation of their DNA binding, cytotoxicities, and antibacterial properties

Article abstract of DOI:10.1007/s00044-018-2169-x

Neomycin is known to preferentially bind to A-form nucleic acid structures including triplex DNA, DNA and RNA hybrid, and duplex RNA. Tethering a DNA intercalator moiety to the C5” position of the ring III of neomycin is a practical approach to develop po

Full text of DOI:10.1007/s00044-018-2169-x

MEDICINAL
CHEMISTRY
RESEARCH
Medicinal Chemistry Research
https://doi.org/10.1007/s00044-018-2169-x
ORIGINAL RESEARCH
Synthesis of nucleobase-neomycin conjugates and evaluation of
their DNA binding, cytotoxicities, and antibacterial properties
Siwen Wang1 Mandeep Singh Mingheng Ling Danni Li Krege M. Christison Joan Lin-Cereghino
1 ●
1 ●
1 ●
1 ●
2 ●
●
2 ●
1
Geoff P. Lin-Cereghino Liang Xue
Received: 28 November 2017 / Accepted: 16 March 2018
Springer Science+Business Media, LLC, part of Springer Nature 2018
©
Abstract
Neomycin is known to preferentially bind to A-form nucleic acid structures including triplex DNA, DNA and RNA hybrid,
and duplex RNA. Tethering a DNA intercalator moiety to the C5” position of the ring III of neomycin is a practical approach
to develop potent binders targeting various nucleic acid secondary structures via a synergistic effect; however, the minimal
stacking surface of the intercalating moiety required to exhibit the effect remains unclear. In the present work, we
synthesized four nucleobase and neomycin conjugates via click chemistry. All four conjugates stabilized a DNA
oligonucleotide triplex in the thermal denaturation experiments monitored by UV. The guanine-neomycin conjugate (6b)
showed a better triplex stabilization effect than neomycin. All the conjugates, as well as neomycin, exhibited no thermal
stabilization effect on a human telomeric DNA G-quadruplex. These results suggest that the synergistic effect of binding is
vastly dependent on the surface area of the stacking moiety of the conjugates. In addition, tethering a nucleobase to the C5”
position of neomycin enhanced the cytotoxicity of neomycin toward MCF-7 and HeLa cancer cells but decreased the
antibacterial effect of neomycin against several Gram-negative and Gram-positive bacterial species.
●
●
●
Keywords Neomycin DNA binding ligands Cell viability Antibacterial effect
Abbreviations
NMR
nuclear magnetic resonance
Neo
neomycin
UHPLC ultra-high-performance liquid chromatography
rRNA
TAR
AMEs
BQQ
DMSO
MTT
ribosomal RNA
IR
ESI
HRMS
TOF
FBS
DMEM
KB test
OD
infrared
trans-activation response element
aminoglycoside-modifying enzymes
benzo[f]quino[3,4-b]quinoxaline
dimethyl sulfoxide
3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetra-
zolium bromide
electrospray ionization
high-resolution mass spectrometry
time of flight
fetal bovine serum
Dulbecco’s modified eagle’s medium
Kirby–Bauer test
IC50
MIC
half maximal inhibitory concentration
minimum inhibitory concentration
optical density
These authors contributed equally: Siwen Wang, Mandeep Singh.
Electronic supplementary material The online version of this article
Introduction
(https://doi.org/10.1007/s00044-018-2169-x) contains supplementary
material, which is available to authorized users.
Neomycin, an aminoglycoside antibiotic, exerts its function
by binding tightly to the A-site decoding region of bacterial
ribosomal RNA (rRNA) and thus disrupting protein
synthesis (Fourmy et al. 1998; Wong et al. 1998). Recent
studies have revealed that neomycin also preferentially
binds to A-form nucleic acid structures including triplex
DNA, DNA and RNA hybrid, and duplex RNA (Arya et al.
*
Liang Xue
lxue@pacific.edu
1
2
Department of Chemistry, University of the Pacific, Stockton, CA
9
5211, USA
Department of Biological Sciences, University of the Pacific,
Stockton, CA 95211, USA

Medicinal Chemistry Research
Scheme 1 Synthesis of nucleobase-neomycin conjugates (6a–6d). Reagents and conditions: i) (Boc)2O, DMF-H2O, 60 °C; ii) 2,4,6-triisopro-
pylbenznenesulfonyl chloride, pyridine, r.t.; iii) NaN3, DMF, 90 °C; iv) 11–14, sodium ascorbate, CuSO4, DMF, r.t.; v) 4MHCl in Dioxane, r.t
2
003c). The binding specificity to nucleic acids results from
affinities of these neomycin conjugates have been attributed
to the synergistic effect (dual recognition mode), in which
neomycin binds into a DNA grove while the other binding
moiety stacks with DNA triplets or G-quartets.
its flexible oligosaccharide structure and multiple positive
charges present under physiological conditions (Arya et al.
2
003a). Over the years, many neomycin based conjugates
have been synthesized to target various nucleic acid struc-
tures. One of the major approaches for conjugation is to
tether a moiety at the C5” position of the ring III of neo-
mycin (Scheme 1). Modification of this site has the least
interference with the binding of neomycin to its target. An
acridine-neomycin conjugate synthesized by Tor’s group
binds tightly to the HIV-1 Rev-Response Element (Kirk
et al. 2000). Several nucleobase-aminoglycoside conjugates
exhibit increased binding affinities to the HIV-1 TAR and
decreased affinities to the A-site of rRNA compared with
the parent antibiotic (Blount and Tor 2006). Arya and his
coworkers have developed a series of neomycin based
conjugates by coupling a DNA binding moiety such as
intercalators with neomycin (Arya et al. 2003b; Xue et al.
Aminoglycosides are known for their inherent toxicity,
usually nephrotoxicity (Oliveira et al. 2009) or ototoxicity
(Guthrie 2008), due to their ability to sequester and generate
redox active metal complexes that induce cell damage.
However, in general, they are considered to be less cyto-
toxic toward tumor cells (e.g., MCF-7 cells, human breast
adenocarcinoma) and nontumorigenic cells. Tethering a
moiety to the C5” of neomycin could result in a conjugate
with unique cytotoxicity. Marchan and coworkers reported
that a Ru(II) arene-neomycin complex displays moderate
antiproliferative activity in MCF-7 cells and enhanced
cytotoxicity in normal cells (Grau-Campistany et al. 2013).
In recent years, the emergence of resistant bacteria,
especially those containing aminoglycoside-modifying
enzymes (AMEs) (Ramirez and Tolmasky 2010), has
become a global concern to public health. Modification of
aminoglycosides is a useful approach to discover novel
antibiotics (Green et al. 2010; Houghton et al. 2010). Ide-
ally, the resulting aminoglycoside derivatives should still
bind to their RNA receptor but cannot interact with AMEs.
2
002). Most of the intercalator-neomycin conjugates show a
better DNA triplex stabilization effect than both parent
compounds (neomycin and intercalator). A perylene-
neomycin conjugate developed by the same group shows
a high binding affinity to human telomeric G-quadruplex
DNA (Xue et al. 2011). The enhanced DNA binding

Medicinal Chemistry Research
The C5” position has been used to design novel neomycin-
based antibiotics to overcome the bacterial resistance to
aminoglycosides that results from intensive clinical use.
Chang and his coworkers developed a series of C5”-acy-
lated neomycin derivatives and found two leads that
exhibited prominent activity against both methicillin-
resistant and vancomycin-resistant bacterial species
2,4,6- triisopropylbenzenesulfonyl chloride. Use of this
bulky reagent enabled us to selectively react the 5” primary
hydroxyl group of neomycin. Replacement of 2,4,6-triiso-
propylbenzenesulfonylate in 3 with an azido group yielded
compound 4. Nucleobase derivatives 11–14 were synthe-
sized based on previously published procedures (Edward
Lindsell et al. 2000; Lu et al. 2007). The click reaction
between propargyl nucleobases (11–14) and 4 afforded
compounds 5a–5d in the presence of catalytic amount of Cu
(I), a catalyst that warrants the formation of 1,4-regioisomer
of 1,2,3-triazole (Liu et al. 2011). Deprotection of the Boc
groups from 5a–5d using a dioxane solution saturated with
HCl yielded the desired nucleobase-neomycin conjugates
(6a: adenine-neomycin; 6b: guanine-neomycin; 6c: cyto-
sine-neomycin; 6d: thymine-neomycin) with a yield ranging
from 77.0–93.5%. All the compounds were characterized by
(
Zhang et al. 2009). Furthermore, bisubstrate derivatives
consisting of adenosine and neamine (a truncated neomy-
cin) using methylene groups as linkers were used as com-
petitive
inhibitors
of
aminoglycoside
3’-O-
phosphotransferases (Liu et al. 2000).
In this report, we present the synthesis of four conjugates
by coupling DNA nucleobases (adenine, guanine, thymine,
and cytosine) to the C5” position of neomycin via click
chemistry. One goal of this research is to explore the size
limitation of the aromatic surface in neomycin conjugates
on their synergistic binding ability to DNA triplex and G-
quadruplex. The previously reported conjugates contain a
canonical DNA intercalator with a sizeable aromatic sur-
face, so they are not suitable for this task. Native nucleo-
bases are used here because they have a much smaller
aromatic surface than DNA intercalators. In addition, the
cytotoxicity against HeLa and MCF-7 cancer cells and
antibacterial effect of these novel nucleobase-neomycin
conjugates are of interest.
1
IR, H NMR, and high-resolution ESI mass spectrometry.
According to the 1H NNR integration analysis, the purity of
all final compounds (6a–6d) is no less than 95%, which is
sufficient for the biochemical and biophysical studies con-
ducted in the following session.
The surface area of the aromatic moiety at the C5”
position of neomycin has a significant effect on the
thermal stability of triplex DNA and G-quadruplex
DNA
Thermal denaturation of triplex and G-quadruplex DNA in
the presence of nucleobase-neomycin conjugates was car-
ried out using UV spectroscopy. The triplex used was a
short homooligomer dA •2dT . The G-quadruplex was
Results and discussion
Synthesis of nucleobase-neomycin conjugates
2
2
22
formed from a human telomeric DNA fragment 5′-AGGG
Nucleobases containing a propargyl group (11–14 con-
taining adenine, guanine, cytosine, and thymine, respec-
tively) and C5”-azido neomycin (4) were prepared
separately for the click reaction. The C5” position of the
ring III of neomycin (Scheme 1) was chosen for conjuga-
tion. The modification of this site has a minimal effect on
the structure of neomycin and does not alter its charge
distribution (amino groups). Tor’s group has reported the
direct attachment of three DNA bases (adenine, cytosine,
and thymine) to neomycin for studying the selectivity
toward two RNA targets (Blount and Tor 2006). Our con-
jugates are different from theirs by having a linker (1,2,3-
triazole) between the nucleobase and neomycin. The pre-
sence of the linker may give additional flexibility to the
nucleobase for an optimized intercalating position during
the synergistic binding event.
(TTAGGG) T for this study. The UV wavelengths to
3
monitor the melting of triplex and G-quadruplex DNA as a
function of increasing temperature were 260 and 295 nm,
respectively.
All the melting profiles of the triplex DNA are shown in
Fig. 1a. In the absence of any ligand (control), the melting
curve of dA •2dT shows a clear biphasic transition
2
2
22
(hypochromic effect). The hyperchromicity observed at
8.7 °C represents a dissociation of triplex DNA into duplex
DNA (dA •dT ) and a single strand (dT ). This transition
2
2
22
22
was readily observed at a temperature less than 10 °C using
a recently published protocol (Xue et al. 2017). The dis-
sociation of the remaining dA •dT into random coils is
2
2
22
reflected by another hyperchromic effect observed at 54.1 °C.
In order to compare the effect of various ligands on triplex
stabilization, we chose a fixed concentration, 10 µM, to
conduct the experiments. As shown in Table 1 and Fig. 1b,
in the presence of neomycin, the dissociation temperature of
triplex DNA (Tm3→2) increases significantly to 39.3 °C,
The modification of neomycin was based on a previously
published procedure (Kirk et al. 2000). All six amino
groups of neomycin (1) were protected using di-tert-butyl
dicarbonate to yield compound 2. The 5” primary hydroxyl
group of 2 was converted into a good leaving group using
while the dissociation temperature of duplex DNA (Tm2→1
)
is not affected. This observation is consistent with previous

Medicinal Chemistry Research
Fig. 1 a UV melting profiles of a 22 mer T•A•T triplex (1 μM) in the
presence of neomycin (neo) and compound 6a–6d (10 μM). b Bar
graphs showing the melting temperatures of triplex to duplex transition
(Tm3→2) and duplex to random coil transition (Tm2→1). Buffer condi-
tion: 10 mM sodium cacodylate, 200 mM NaCl, 1 mM EDTA, pH 7.0
Table 1 Temperatures of triplex to duplex transition (Tm3→2) and
duplex to random coil transition (Tm2→1). The A22•2T22 triplex was
denaturated in the presence of neomycin (neo) and compound 6a–6d
we observed a 4.1 °C increment of Tm3→2 for 6b as com-
pared with that of neomycin, indicating a synergistic
binding effect. In this case, the destabilization effect arising
from the modification of the C5” position was rectified. The
results reported here are consistent with the order for the
(
10 μM)
Compound
Tm3→2
ΔTm3→2
Tm2→1
ΔTm2→1
2
6
6
6
6
a
b
c
36.9 ± 1.0
43.4 ± 2.4
35.1 ± 1.7
28.4 ± 3.2
39.3 ± 1.8
8.7 ± 0.5
28.2
34.7
26.4
19.7
30.6
N/A
54.1 ± 0.2
54.0 ± 0.5
53.9 ± 0.3
53.8 ± 0.8
54.1 ± 0.5
54.1 ± 0
0
stacking area (Å ) of native nucleobases, which is guanine
−0.1
−0.2
−0.3
0
(G) > adenine (A) > cytosine (C) > thymine (T) (Guckian
et al. 2000). The relatively larger stacking surface of gua-
nine enables 6b to bind to triplex DNA in a dual recognition
mode. The guanine moiety sufficiently stacks with DNA
triplets while the neomycin moiety binds into the groove of
triplex DNA to form a stable complex. In addition, the
exocyclic amino group of guanine may interact with T•A•T
triplets and/or the backbone of DNA (Wellenzohn et al.
d
Neo
No Ligand
N/A
The ΔTm values were calculated based on the difference in the melting
temperatures of the corresponding DNA in the presence and absence
of a ligand
2001). However, 6b cannot stabilize triplex DNA to the
same extent as other conjugates with a much larger aromatic
surface. Arya and coworkers developed a series of neo-
mycin conjugates containing intercalators such as pyrene
and BQQ (Arya et al. 2003b; Xue et al. 2002). They found
that all of them exhibit significantly stronger triplex stabi-
lization effect than neomycin. For instance, at a 4 μM
concentration, pyrene-neomycin showed an impressive 26 °
C more increment of Tm3→2 for poly(dA)•2poly(dT) than
neomycin. BQQ-neomycin bound to triplex DNA so tightly
that only the direct dissociation of DNA triplex into random
coils was observed. Their results indicate the importance of
the aromatic surface for the dual recognition mode, which
aligns well with our data. Further evidence for the role of
the aromatic surface came from the thermal denaturation
experiments of telomeric G-quadruplex DNA. Arya and
coworkers reported a perylene diimide conjugated with two
neomycin moieties that binds to G-quadruplex DNA with
high affinity (Xue et al. 2011). The sizable aromatic surface
of perylene facilitates the stacking of the molecule with G-
quartets that are larger than DNA triplets. In our experi-
ments, no increments in the Tm values of G-quadruplex
studies that neomycin stabilizes triplex DNA but not duplex
DNA (Arya et al. 2003a). Neomycin snugly fits into the
Watson-Hoogsteen groove of DNA triplex based on a
computational study (Arya et al. 2003a). In the presence of
a nucleobase-neomycin conjugate, a similar trend was
observed. All four conjugates increase the Tm3→2 values but
do not affect duplex melting temperatures as the ΔTm2→1
values are negligible (Table 1). Based on the ΔTm3→2
values, the order of triplex stabilization effect of these
ligands is guanine-neomycin (6b) > neomycin (neo) > ade-
nine-neomycin (6a) > cytosine-neomycin (6c) > thymine-
neomycin (6d). Interestingly, the modification of the C5”
position on the ring III of neomycin decreases neomycin’s
ability to stabilize DNA triplex because the Tm3→2 values in
the presence of 6a, 6c, and 6d are slightly less than that in
the presence of neomycin. Adding a moiety at this position
could sterically affect the binding of neomycin to the
groove of triplex DNA. It is known that the binding pocket
of neomycin is greatly affected by the surrounding struc-
tural geometry (Degtyareva et al. 2017). On the contrary,

Medicinal Chemistry Research
DNA were observed (Data not shown). Based on the CD
spectrum (shown in the supporting information Fig. S9), the
G-quadruplex DNA used in our study (Liu et al. 2017)
could be a mixture of hybrid-type and basket-type con-
+
formations under the experimental conditions (50 mM K )
(
Ambrus et al. 2006). G-quartets have a much larger surface
than DNA triplets so relatively small nucleobases could not
sufficiently stack with them. Furthermore, neomycin does
not thermally stabilize G-quadruplex DNA. These data
indicate that a synergistic binding requires at least one
moiety with a strong binding affinity to the DNA target. The
first binding moiety could serve as an anchor to facilitate the
binding of the second moiety. The size of the aromatic
surface of intercalator-neomycin conjugates affects the
binding specificity. Taken together, we conclude that the
size of guanine could be a minimum requirement for
developing intercalator-neomycin conjugates to target tri-
plex DNA in a synergistic manner. Intercalator-neomycin
conjugates that bind to G-quadruplex DNA require a much
larger intercalator moiety to stack with G-quartets
efficiently.
Fig. 2 Cytotoxic effect of neomycin and compounds 6a–6d (500 μM)
in the MCF-7 and HeLa cancer cell lines. Cell viability was deter-
mined by the MTT assay
viability values of the four conjugates for MCF-7 cells range
from 47.6 to 61.9%. The difference in values is much smaller
than that for HeLa cells (15.4–68.3%). These data indicate
that all nucleobase-neomycin conjugates have differential
cytotoxicity toward MCF-7 and HeLa cancer cells, which
could result from the cell uptake efficiency. The conjugate 6b
with the largest stacking moiety gave the most prominent
effect on reduction in cell viability, but the correlation
between the stacking moiety and cytotoxicity could not be
drawn. The observed cytotoxicity could result from binding
of these conjugates to other cellular targets such as RNA
(Degtyareva et al. 2017; Kirk et al. 2000). No further
concentration-dependent studies were carried out as the IC50
values for these conjugates would be higher than 250 μM
(considered to be inactive) (Grau-Campistany et al. 2013).
Conjugation of a nucleobase moiety to the C5” of neomycin
leads to the enhanced cytotoxicity toward two cancer cell
lines used here. The increase in the cytotoxicity of neomycin
by tethering a Ru (II) moiety at the same position has been
previously reported (Grau–Campistany et al. 2013). How-
ever, the factors attributed to the enhanced cytotoxicity
reported in the paper were more complicated because the Ru
(II) moiety itself could induce cell damage via redox reac-
tions. Because nucleobases are not cytotoxic, the increased
cytotoxicity observed by us could merely result from the
structural change at the C5” position of neomycin. Hence,
tethering a moiety to the C5” of neomycin could be a useful
approach to design aminoglycoside based cytotoxic reagents.
Nucleobase-neomycin conjugates are more
cytotoxic than neomycin
The cytotoxicity of nucleobase-neomycin conjugates against
two cancer cell lines (MCF-7 and HeLa) was measured using
the MTT assay. In this assay, the absorbance of a reduction
product (formazan) from a tetrazolium salt (MTT) is deter-
mined spectroscopically. Only live cells can release active
reductases that catalyze the reduction reaction; therefore, the
absorbance of formazan is proportional to the number of live
cells in culture. Because neomycin was reported to have
minimal cytotoxicity toward MCF-7 cells (IC > 250 μM)
50
(
Grau-Campistany et al. 2013), we used a higher con-
centration (500 μM) for the initial test. The results from the
MTT assay are shown in Fig. 2. Neomycin shows little
cytotoxicity toward both cells lines as expected. The viability
values of MCF-7 and HeLa cells in the presence of 500 μM
neomycin are (94.6 ± 12.5) and (96.4 ± 1.5)%, respectively.
In general, all nucleobase-neomycin derivatives are more
cytotoxic toward both cancer cell lines than neomycin in a
statistically significant manner (p < 0.05). All nucleobase-
neomycin derivatives except for 6d are more cytotoxic
toward HeLa cells than MCF-7 cells. The orders of
increasing cytotoxicity are neo < 6a < 6d < 6c < 6b for HeLa
cells and neo < 6d < 6a < 6c < 6b for MCF-7 cells. In both
cases, 6b (guanine–neomycin) is the most cytotoxic, and it is
more effective in reducing the HeLa cell viability. The HeLa
cell viability (%) in the presence of 500 μM 6b is (15.4 ±
Nucleobase-neomycin conjugates inhibit bacterial
growth less efficiently than neomycin
0
.8)%, at least 2-fold less than those values of the other three
We investigated the antibacterial effect of nucleobase-
neomycin conjugates on aminoglycoside resistant and non-
resistant bacterial species.
conjugates. By contrast, the MCF-7 cell viability (%) in the
presence of 500 μM 6b is only (47.6 ± 6.7)%. The cell

Medicinal Chemistry Research
Fig. 3 a Petri dish images (Kirby–Bauer Test) showing the zone of
inhibition by compound 6a–6d (50 nmol) against various bacterial
species (S. marcescens (Gram negative), E. coli (Gram negative), B.
subtilis (Gram positive), and P. vulgaris (Gram negative)). Neomycin
(50 nmol) was used as a control. b Diameter values of the zone of
inhibition for each compound
All nucleobase-neomycin conjugates (50 nmol) were first
assayed (Fig. 3a) against four antibiotic-susceptible species
including Serratia marcescens (S. marcescens, Gram
negative), Escherichia coli (E. coli, Gram negative),
Bacillus subtilis (B. subtilis, Gram positive), and Proteus
vulgaris (P. vulgaris, Gram negative) using the
Kirby–Bauer (agar disk-diffusion) test (Balouiri et al.
a different inhibitory mechanism or a different mechanism
of cellular penetration through the outer membranes
between Gram-negative and Gram-positive bacteria. Con-
jugation of nucleobases to neomycin significantly reduces
the antibacterial activity of neomycin against E. coli and
P. vulgaris. We observed the greatest reduction effect for
the latter. Amongst all the nucleobase-neomycin deriva-
tives, 6b is more active against all four bacterial species,
and the compound 6a is the least active compound against
the three Gram-negative bacterial species. It is noteworthy
that adenine-neomycin (6a) does not show a detectable
inhibition zone for P. vulgaris at the concentration we used.
The minimum inhibitory concentrations (MICs) of
nucleobase-neomycin conjugates against four bacterial
species were further determined on 96-well microtiter
plates. The OD600 value of the well in the absence of bac-
teria was used as a control, representing a complete inhi-
bition effect of bacterial growth. The results are shown in
Fig. 4 and Table S1 of the supporting information. The MIC
values of neomycin toward E. coli and B. subtilis are
comparable to previously reported values (Fridman et al.
2003). The order of bacterial species regarding decreasing
inhibitory effect by the compounds is B. subtilis > E. coli >
S. marcescens > P. vulgaris. All compounds inhibit the
growth of the Gram-positive bacterium, B. subtilis, more
efficiently than that of any Gram-negative bacteria. This
data is consistent with the Kirby–Bauer test mentioned
above. While none of the conjugates exhibits a better
inhibitory effect than the parent neomycin for all the spe-
cies, the compound 6b inhibits the bacterial growth more
efficiently than the other three conjugates. The reduction in
the inhibition of bacterial growth is more significant for
Gram-negative than Gram-positive bacteria. For instance,
2
016). Neomycin with the same amount was used as a
control. The inhibition of bacterial growth is positively
related to the inhibition-zone size on agar, in which no
bacterial growth occurs. The diameters of inhibition zones
were directly measured using a ruler (Fig. 3b). The values
(
cm) of the inhibition zone of neomycin are 1.5 for
S. marcescens, 2.2 for E. coli, 1.9 for B. subtilis, and 2.0 for
P. vulgaris. We calculated the median values of inhibition
zones of four conjugates for each bacterial species. The
values (cm) are 1.2 for S. marcescens, 1.3 E. coli, 1.6 for B.
subtilis, and 0.7 for P. vulgaris. By subtracting these
median values from the values of the inhibition zone of
neomycin for each bacterial species (0.3 cm for S. marces-
cens, 0.9 cm for E. coli, 0.3 cm for B. subtilis, and 1.3 cm
for P. vulgaris), we elucidated the effect of conjugation on
the antibacterial activity of neomycin. Neomycin gives the
best inhibitory effect against all four bacterial species
compared with nucleobase–neomycin conjugates. Con-
jugation of nucleobases reduces the antibacterial activity of
neomycin. Modification of the C5”-position could decrease
the binding of neomycin to its target rRNA. The reduction
effect is bacterial species dependent. Conjugation of
nucleobases to neomycin has less effect on reduction in the
antibacterial activity of neomycin against S. marcescens and
B. subtilis. B. subtilis is a Gram-positive bacterium so the
decreased reduction effect on this species could result from

Medicinal Chemistry Research
these conjugates. It is reasonable that the nucleobase
moiety may not be large enough to interfere with the
binding.
Conclusion
We have completed the synthesis of four novel nucleobase-
neomycin conjugates. The results from the thermal dena-
turation of triplex DNA, together with the reports by others,
suggest that the stacking size of guanine could be the
minimal limit for neomycin conjugates in the same category
to exert the synergistic binding effect. Judging from the
melting data, we conclude that the synergistic effect
requires at least one moiety to have a strong binding affinity
to the target. Modification of the C5” position of neomycin
is an effective approach to enhance the cytotoxicity of an
aminoglycoside (neomycin) against two cancer cell lines
used in this research. Based on the antibacterial profile,
tethering a nucleobase to the C5” position of neomycin
could affect the binding of neomycin to its rRNA target but
this modification does not block the function of aminogly-
coside 3’-O-phosphotransferase. Our results presented here
provide useful information for developing aminoglycoside-
based compounds with better binding affinities, enhanced
antibacterial activities, and altered toxicities.
Fig. 4 Minimal inhibition concentrations (MIC) of neomycin and
compound 6a–6d against various bacterial species (S. marcescens
(
Gram negative), E. coli (Gram negative), B. subtilis (Gram positive),
and P. vulgaris (Gram negative))
the MIC value of 6b against B. subtilis is 2 nmol/mL, only
two-fold more than that of neomycin. The MIC value of 6b
against S. marcescens is 32 nmol/mL, eight-fold more than
that of neomycin. With limited data, no relationships
between the stacking surface of nucleobases and anti-
bacterial effect could be ascertained at this point.
The inhibition of bacterial growth by these conjugates
was also investigated using an E. coli strain that was
transfected with an aminoglycoside 3’-O-phosphotransfer-
ase (APHs) Ia gene. This strain is known to have a resis-
tance profile including kanamycin, neomycin, and a few
other aminoglycosides (Vakulenko and Mobashery 2003).
The resistance to aminoglycosides results from the transfer
of a phosphate group to the 3’ or 5” hydroxyl group of the
aminoglycoside molecule and thus affecting its binding to
the target rRNA (Vakulenko and Mobashery 2003).
Nucleobase-neomycin conjugates have modified structures
that could block the binding to APHs and thus inhibit its
function for phosphorylation. We were only able to obtain
two MICs values for neomycin and 6b, respectively, even
when the concentration was used at as high as 512 μM. The
MIC value (64 μM) of neomycin for this strain is 8-fold
larger than that for un-transfected E. coli, indicating the
presence of drug resistance. The effect of resistance to
kanamycin is much stronger as no MIC value was observed
Experimental section
Materials and methods
Chemicals for synthesis were purchased from Fisher
Scientific (Pittsburgh, PA) and used without further pur-
ification. Reactions were carried out under Argon using
dry solvent unless otherwise noted. IR spectra were col-
lected on a Shimadzu (Pleasanton, CA) IR Prestige-21
1
FT-IR spectrophotometer. H NMR spectra were collected
on a JEOL (Peabody, MA) ECA 600 MHz FT-NMR
spectrometer. HRMS (ESI) spectra were collected on an
Agilent (Santa Clara, CA) 1290 UHPLC connected to an
Agilent 6230 TOF mass spectrometer through an elec-
trospray ionization source. UV spectra were collected on a
Varian Cary 100 Bio UV-Vis spectrophotometer equipped
with a thermoelectrically controlled 6 × 6 cell holder
(Walnut Creek, CA). The MTT read-outs were measured
by a TriStar LB 941 multimode microplate reader. Dul-
becco’s modified eagle’s medium (DMEM) was pur-
chased from Sigma (St. Louis, MO). Advanced MEM,
glutaMAX, penicillin, and streptomycin were purchased
from Invitrogen (Carlsbad, CA). Fetal bovine serum
(FBS) was purchased from Tissue Culture Biologicals
(Seal Beach, CA).
(
6
Table S1 of the supporting information). The compound
b displays a MIC value of 512 µM toward the transfected
strain, which is 32-fold more than that (16 µM) for the un-
transfected species. These results suggest that none of the
nucleobase-neomycin conjugates can overcome the bac-
terial resistance as their MIC values are all worse than that
of neomycin for this species. Tethering a nucleobase
moiety to the C5” position of neomycin does not alter the
binding of aminoglycoside 3’-O-phosphotransferase to

Medicinal Chemistry Research
Synthesis of 5a
(2.0 mg, 0.008 mmol), and compound
4
(64.0 mg,
0.054 mmol). The reaction mixture was stirred at room
Adenine-9-prop-2ynyl (11) (12.5 mg, 0.072 mmol) was
added to an anhydrous DMF solution (6 mL) of sodium
ascorbate (13.7 mg, 0.069 mmol), copper (II) sulfate
temperature overnight. The completion of the reaction was
monitored by TLC (silica gel 60 F254, EMD Millipore)
under
UV
(R = 0.435,
f
CH OH:CH Cl :NH OH
3 2 2 4
(
0
2.0 mg, 0.008 mmol), and compound
.058 mmol). The reaction mixture was stirred at room
4
(71.0 mg,
1.5:8.5:0.05 v/v/v). The reaction mixture was concentrated
under vacuum. Flash chromatography of the residue (silica
gel, CH OH:CH Cl 1.25:8.75 v/v) yielded compound 5c
temperature overnight. The completion of the reaction was
monitored by TLC (silica gel 60 F254, EMD Millipore)
under UV (R = 0.23, CH OH:CH Cl 0.8:9.2 v/v). The
reaction mixture was concentrated under vacuum. Flash
chromatography of the residue (silica gel, CH OH:CH Cl
2
3
2
2
−
1
(30.0 mg, 42.0%) as a pale white solid. IR (KBr) (cm )
3369, 2978, 2932, 1701, 1647, 1524, 1394, 1367, 1280,
f
3
2
2
1
1251, 1169, 1043, 865. H NMR (CD OD, 600 MHz) δ
3
8.07 (s, 1H), 7.73 (d, J = 7.26 Hz, 1H), 5.85 (d, J =
7.26 Hz, 1H), 5.47 (s, 1H), 5.41 (s, 1H), 5.00–5.13 (m, 3H),
4.92 (d, J = 1.56 Hz, 1H), 4.66 (dd, J = 13.02 Hz, 4.38 Hz,
1H), 4.58 (s, 1H), 4.07–4.36 (m, 3H), 3.95 (m, 1H), 3.88 (s,
1H), 3.69–3.79 (m, 2H), 3.57–3.67 (m, 1H), 3.39–3.55 (m,
4H), 3.31–3.38 (m, 2H), 3.09 (m, 1H), 1.86–1.97 (m, 1H),
1.25–1.49 (m, 54H). HRMS (ESI) calcd. for C H N O
3
2
1
:9 v/v) yielded compound 5a (67.5 mg, 82.0%) as a pale
−
1
white solid. IR (KBr) (cm ) 3343, 2976, 2938, 1644, 1649,
1
1
6
2
1
4
3
2
631, 1367, 1248, 1165, 1043. H NMR (CD OD,
3
00 MHz) δ 8.32 (s, 1H), 8.22–8.29 (m, 2H), 6.47–6.83 (m,
H), 5.58–5.70 (m, 1H). 5.33–5.44 (bs, 1H), 5.01–5.11 (m,
H), 4.97 (s, 1H), 4.78–4.86 (m, 1H), 4.66–4.76 (m, 1H),
.07–4.41 (m, 3H), 3.98 (t, 1H, J = 5.8 Hz), 3.90 (s, 1H),
.66–3.83 (m, 3H), 3.50–3.65 (m, 2H), 3.41–3.49 (m, 2H),
.98–3.08 (m, 1H), 1.90–2.01 (m, 1H), 1.15–1.56 (m, 54H).
6
1
100 14 25
+
[M + H] 1389.7001, found 1389.6990.
Synthesis of 5d
HRMS (ESI) calcd. for C H N O
100 14 24
[M + H]+
61
1
413.7113, found 1413.7090.
Thymine-1-prop-2ynyl (14) (13.7 mg, 0.0836 mmol) was
added to an anhydrous DMF solution (6 mL) of sodium
ascorbate (15.8 mg, 0.079 mmol), copper (II) sulfate
Synthesis of 5b
(2.0 mg, 0.008 mmol), and compound
4
(55.0 mg,
Guanine-9-prop-2ynyl (11) (14.0 mg, 0.074 mmol) was
added to an anhydrous DMF solution (6 mL) of sodium
ascorbate (15.0 mg, 0.075 mmol), copper (II) sulfate
0.0442 mmol). The reaction mixture was stirred at room
temperature overnight. The completion of the reaction was
monitored by TLC (silica gel 60 F254, EMD Millipore)
under UV (R = 0.2125, CH OH:CH Cl 0.75:9.25 v/v).
(
3.0 mg, 0.012 mmol), and compound
4
(46.0 mg,
f
3
2
2
0
.037 mmol). The reaction mixture was stirred at room
The reaction mixture was concentrated under vacuum. Flash
chromatography of the residue (silica gel, CH OH:CH Cl
2
temperature overnight. The completion of the reaction was
monitored by TLC (silica gel 60 F254, EMD Millipore)
under
3
2
0.75:9.25 v/v) yielded compound 5d (60.0 mg, 96.7%) as a
−
1
UV
(R = 0.275,
f
CH OH:CH Cl :NH OH
pale white solid. IR (KBr) (cm ) 3371, 2978, 2934, 1690,
3
2
2
4
1
.25:8.75:0.1 v/v/v). The reaction mixture was concentrated
1514, 1458, 1392, 1367, 1267, 1250, 1167, 1043, 862, 781.
1
under vacuum. Flash chromatography of the residue (silica
gel, CH OH:CH Cl :NH OH 1.25:8.75:0.1 v/v/v)) yielded
compound 5b (25.0 mg, 47.3%) as a pale white solid. IR
H NMR (CD OD, 600 MHz) δ 8.10 (s, 1H), 7.56 (s, 1H),
3
6.55–6.67 (m, 1H), 6.37–6.53 (m, 1H), 5.47 (s, 1H), 5.41 (s,
1H), 4.89–5.14 (m, 3H), 4.54–4.68 (m, 1H), 4.04–4.47 (m,
3H), 3.95 (t, J = 6.6 Hz, 1H), 3.70–3.80 (m, 2H), 3.60–3.69
(m, 1H), 3.36–3.59 (m, 3H), 3.08 (t, J = 8.4 Hz 1H),
1.87–1.95 (m, 1H), 1.85 (d, J = 1.3 Hz, 3H), 1.25–1.48 (m,
54H). HRMS (ESI) calcd. for C H N O [M + H]+
3
2
2
4
−
1
(
KBr) (cm ) 3408, 1689, 1633, 1516, 13676, 1251, 1167,
1
1
7
5
1
3
3
043, 856. H NMR (CD OD, 600 MHz) δ 8.36 (s, 1H),
3
.83 (s, 1H), 6.62–6.84 (m, 1H), 5.56 (d, J = 14.4 Hz, 1H),
.36 (s, 1H), 5.28 (d, J = 15.6 Hz, 1H), 5.01 (s, 1H), 4.96 (s,
H), 4.61–4.73 (m, 1H), 3.90–4.15 (m, 4H), 3.87 (s, 1H),
.68–3.79 (m, 2H), 3.38–3.59 (m, 7H), 3.28–3.37 (m, 1H),
.04 (br s, 1H), 1.84–1.96 (m, 1H), 1.25–1.49 (m, 54H).
6
1
101 11 26
1404.6997, found 1404.6971.
General procedure for the synthesis of 6a–6d
HRMS (ESI) calcd. for C H N O
100 14 25
[M + H]+
61
1
429.7062, found 1429.7042.
Hydrochloric acid in dioxane (4 M, 1 mL) and ethanedithiol
(
3 μL) were added into a dioxane solution (1 mL) of com-
pound 5a (17.0 mg), 5b (24.0 mg), 5c (12.0 mg), or 5d
24.0 mg) in a 15-mL conical vial. The reaction mixture was
Synthesis of 5c
(
Cytosine-1-prop-2ynyl (13) (24.0 mg, 0.161 mmol) was
added to an anhydrous DMF solution (6 mL) of sodium
ascorbate (14.6 mg, 0.0738 mmol), copper (II) sulfate
swirled for 5 min and became turbid. The deprotected pro-
duct was further precipitated by adding ether and hexane
(1 mL each). The solid was collected by centrifugation

Medicinal Chemistry Research
(
3000 r.p.m.), decanted, washed with ether (3 × 1 mL) and
(m, 3H), 3.74–3.83 (m, 2H), 3.69–3.73 (m, 1H), 3.54 (t, J
= 10.2 Hz, 1H), 3.47–3.51 (m, 1H), 3.20–3.46 (m, 8H),
2.32 (m, 1H), 1.70–1.82 (m, 4H). HRMS (ESI) calcd. for
hexane (3 × 1 mL), and dried overnight under vacuum. The
residue was re-dissolved in milli-Q water and lyophilized to
give a pale-white fluffy solid.
+
+
C H N O
54 11 14
[M + H] 804.3846, found 804.3850.
3
1
Compound 6a
Measurements of high-resolution mass spectra
−
1
Yield: 11.6 mg, 93.5%. IR (KBr) (cm ) 3396, 3058, 2948,
1
Samples were introduced by flow injection analysis (FIA) in
1
685, 1601, 1491, 1144, 1044, 1018, 769, 638. H NMR
positive-ion mode. The mobile phase was 90% H O and
2
(
D O, 600 MHz) δ 8.32 (s, 1H), 8.30 (s, 1H), 8.17 (s, 1H),
10% MeOH with 0.1% formic acid. The nebulizer pressure
was 50 psi with 12 L/min of nitrogen (drying gas) at 325 °C
for desolvation. The capillary voltage was set at 3500 V, the
fragmentor voltage was set at 150 V, the skimmer voltage
was set at 65 V, and the octopole voltage was set at 750 V.
Spectra were collected from 300–1500 Da at a rate of
2 spectra/sec with 4962 transients/spectrum.
2
5
1
5
.94 (d, J = 4.2 Hz, 1H), 5.56 (s, 2H), 5.31 (d, J = 3.6 Hz,
H), 5.21 (d, J = 1.8 Hz, 1H), 4.76 (dd, J = 15.0 Hz,
.0 Hz, 1H), 4.68 (m, 1H), 4.45 (m, 2H), 4.23 (m, 1H), 4.12
(
t, J = 6.0 Hz, 1H), 4.05 (t, J = 9.6 Hz, 1H), 3.82–3.93 (m,
4
9
H), 3.72 (s, 1H), 3.62 (t, J = 9.6 Hz, 1H), 3.21–3.51 (m,
H), 2.38 (m, 1H), 1.82 (m, 1H). HRMS (ESI) calcd. for
+
+
C H N O
53 14 12
[M + H] 813.3962, found 813.3945.
3
1
Preparation of triplex DNA for thermal denaturation
Compound 6b
The oligonucleotides dA22 (1 μM) and dT22 (2 μM) were
mixed with a solution (1 mL, pH 7.0) of sodium cacodylate
(10 mM), EDTA (1 mM), and sodium chloride (200 mM) in
the absence or presence of a ligand (10 μM). The mixtures
were heated at 90 °C for 5 min, slowly cooled at 25 °C, and
incubated at 4 °C overnight before use.
−
1
Yield: 14.7 mg, 84.0% IR (KBr) (cm ) 3383, 2922, 1701,
637, 1604, 1500, 1490, 1369, 1143, 1047, 1024, 781, 605.
1
1
H NMR (D O, 600 MHz) δ 8.16 (s, 1H), 8.05 (s, 1H), 5.52
d, J = 4.2 Hz, 1H), 5.34 (s, 2H), 5.23 (d, J = 4.2 Hz, 1H),
.19 (d, J = 1.2 Hz, 1H), 4.71–4.75 (m, 1H), 4.65 (s, 1H),
.43 (q, J = 4.2 Hz, 1H), 4.38 (t, J = 4.8 Hz, 1H), 4.20 (t, J
2
(
5
4
=
4.8 Hz, 1H), 4.09 (t, J = 3 Hz, 1H), 3.74–3.90 (m, 4H),
Preparation of G-quadruplex DNA for thermal
denaturation
3
4
.70 (s, 1H), 3.17–3.51 (m, 11H), 2.35 (dt, J = 8.4 Hz,
.2 Hz, 1H), 1.75–1.86 (m, 1H). HRMS (ESI) calcd. for
+
+
C H N O
3
[M + H] 829.3911, found 829.3909.
The G-quadruplex sequence 5′-AGG(TTAGGG) T (1 μM)
1
53 14 13
3
was mixed with a solution (1 mL, pH 7.0) of lithium
cacodylate (10 mM), EDTA (1 mM), and potassium chlor-
ide (50 mM) in the absence or presence of a ligand (10 μM).
The mixtures were heated at 90 °C for 5 min, slowly cooled
at 25 °C, and incubated at 4 °C overnight before use.
Compound 6c
−
1
Yield: 6.7 mg, 77.0%. IR (KBr) (cm ) 3392, 1676, 1497,
1
1
1
144, 1053, 1018, 797. H NMR (D O, 600 MHz) δ 8.08 (s,
H), 7.83 (d, J = 7.8 Hz, 1H), 6.04 (d, J = 7.2 Hz, 1H), 5.93
2
(
d, J = 3.0 Hz, 1H), 5.30 (d, J = 2.4 Hz, 1H), 5.22 (s, 1H),
Thermal denaturation of triplex DNA or G-
quadruplex DNA
5
3
3
3
.02 (s, 2H), 4.55–4.84 (m, 2H), 4.45 (s, 2H), 4.24 (t, J =
.6 Hz, 1H), 4.12 (s, 1H), 4.05 (t, J = 9.6 Hz, 1H),
.77–3.99 (m, 4H), 3.73 (s, 1H), 3.60 (t, J = 10.2 Hz, 1H),
UV melting spectra were recorded at 260 nm for triplex
DNA and 295 nm for G-quadruplex DNA as a function of
increasing temperature (4–80 °C, heating rate: 0.5 °C/min).
.19–3.55 (m, 9H), 2.38 (m, 1H), 1.83 (q, J = 12.0 Hz, 1H).
+
[M + H]+
HRMS (ESI) calcd. for C H N O
53 14 12
30
7
89.3850, found 789.3859.
Compound 6d
Yield: 14.7 mg, 84.0%. IR (KBr) (cm ) 3387, 2918, 1701,
The melting temperatures (T ) were determined using the
first derivative method.
m
The MTT assay
−
1
1
1
6
601, 1244, 1220, 1143, 1049, 1026, 773. H NMR (D O,
All the experiments were carried out in triplicate. MCF-7
(human breast adenocarcinoma) and HeLa cells were
maintained in advanced DMEM medium supplemented
with 5% fetal bovine serum (FBS), L-glutamine and anti-
mycotic (antibiotic) at 37 °C in a humid atmosphere con-
2
00 MHz) δ 8.03 (s, 1H), 7.51 (d, J = 1.2 Hz, 1H), 5.82 (d,
J = 3.6 Hz, 1H), 5.28 (d, J = 3.6 Hz, 1H), 5.21 (d, J =
1
1
2
.2 Hz, 1H), 4.94 (d, J = 2.4 Hz, 2H), 4.75 (dd, J =
5.0 Hz, 3.0 Hz, 1H), 4.56–4.72 (m, 1H), 4.39–4.48 (m,
H), 4.20–4.25 (m, 1H), 4.11 (t, J = 3.0 Hz, 1H), 3.84–3.96
taining 5% CO . Neomycin and nucleobase-neomycin
2

Medicinal Chemistry Research
conjugates (6a–6b) were pre-dissolved in sterilized water.
The cells (3000 cells/well, 200 μL) were seeded in a 96-well
microtiter plate and incubated for 24 h. The supernatant in
each well was removed followed by adding neomycin or
conjugates in FBS free media (200 μL) to reach a final
concentration of 500 µM. After incubation at 37 °C for 72 h,
an MTT solution (20 μL, 0.5 mg/mL) was added to each
well and incubated for 4 h. The formazan product was
dissolved in 150 μL of DMSO, and the absorbance of each
well was measured at 590 nm using a plate reader.
Untreated cells and media without cells were used as
controls.
inhibition of bacterial growth, and the OD600 value of the
bacterial solution without any reagent was used as a positive
control.
Supplementary data
Spectroscopic information for the synthesized compounds
and data for MIC determination are presented.
Acknowledgements This work was supported by the University of the
Pacific. We also thank Dr. William K. Chan for providing MCF-7 and
HeLa cells.
To calculate the viability (%), the following equation was
used.
Compliance with ethical standards
Conflict of interest The authors declare that they have no conflict of
interest.
ðAtreatedÞ À ðAmediaÞ
Viability ð%Þ ¼
 100%
ðAuntreatedÞ À ðAmediaÞ
References
Atreated: The absorbance of the solution containing treated
cells Amedia: The absorbance of the media Auntreated: The
absorbance of the solution containing untreated cells
Ambrus A, Chen D, Dai J, Bialis T, Jones RA, Yang D (2006) Human
telomeric sequence forms
a hybrid-type intramolecular g-
quadruplex structure with mixed parallel/antiparallel strands in
potassium solution. Nucleic Acids Res 34:2723–2735
Arya DP, Micovic L, Charles I, Coffee Jr. RL, Willis B, Xue L
Statistical analysis
(2003a) Neomycin binding to Watson-Hoogsteen (W-H) DNA
triplex groove: a model. J Am Chem Soc 125:3733–3744
Arya DP, Xue L, Tennant P (2003b) Combining the best in triplex
recognition: synthesis and nucleic acid binding of a BQQ-
neomycin conjugate. J Am Chem Soc 125:8070–8071
Minitab software was used to determine the statistical sig-
nificance. Two-sample student’s t-test was performed to
show statistically significant (p < 0.05) and insignificant (p
Arya DP, Xue L, Willis B (2003c) Aminoglycoside (neomycin) pre-
ference is for a-form nucleic acids, not just RNA: results for a
competition dialysis study. J Am Chem Soc 125:10148–10149
Balouiri M, Sadiki M, Ibnsouda SK (2016) Methods for in vitro
evaluating antimicrobial activity: a review. J Pharm Anal 6:71–79
Blount KF, Tor Y (2006) A tale of two targets: differential RNA
selectivity of nucleobase-aminoglycoside conjugates. ChemBio-
Chem 7:1612–1621
Degtyareva NN, Gong C, Story S, Levinson NS, Oyelere AK, Green
KD, Garneau-Tsodikova S, Arya DP (2017) Antimicrobial
activity, AME resistance, and A-site binding studies of
anthraquinone-neomycin conjugates. ACS Infect Dis 3:206–215
Edward Lindsell W, Murray C, Preston PN, Woodman TAJ (2000)
Synthesis of 1,3-diynes in the purine, pyrimidine, 1,3,5-triazine
and acridine series. Tetrahedron 56:1233–1245
>
0.05) data.
Kirby–Bauer test protocol (Disk diffusion zone
determination)
Neomycin or conjugate (50 nmol) impregnated paper discs
were placed onto the surface of Müller Hinton agar plates
(
70191 Fluka) that were pre-seeded with bacteria (Bacillus
subtilis, Escherichia coli, Proteus vulgaris, and Serratia
marcescens) and incubated aerobically at 37 °C for 24 h.
The diameters of inhibition zones were measured using a
ruler to the nearest 1 mm.
Fourmy D, Recht MI, Puglisi JD (1998) Binding of neomycin-class
aminoglycoside antibiotics to the a-site of 16 s rrna. J Mol Biol
Determination of minimum inhibition
concentrations (MICs)
2
77:347–362
Fridman M, Belakhov V, Yaron S, Baasov T (2003) A new class of
branched aminoglycosides: pseudo-pentasaccharide derivatives
of neomycin b. Org Lett 5:3575–3578
Grau-Campistany A, Massaguer A, Carrion-Salip D, Barragan F,
Artigas G, Lopez-Senin P, Moreno V, Marchan V (2013) Con-
jugation of a ru(ii) arene complex to neomycin or to guanidino-
neomycin leads to compounds with differential cytotoxicities and
accumulation between cancer and normal cells. Mol Pharm
Bacterial species were grown overnight in the Müller Hin-
ton broth at 37 °C. A dilution of the overnight culture was
made in broth to an approximate concentration of 10 cfu/
mL. A Müller Hinton broth (100 µL) solution containing the
desired antibacterial reagent (neomycin or compound
6
1
0:1964–1976
6
a–6d) was inoculated with a freshly diluted bacterial
Green KD, Chen W, Houghton JL, Fridman M, Garneau-Tsodikova S
(2010) Exploring the substrate promiscuity of drug-modifying
enzymes for the chemoenzymatic generation of n-acylated ami-
noglycosides. ChemBioChem 11:119–126
suspension (5 µL) in a 96-well plate and incubated at 37 °C
overnight. The OD600 value in each well was recorded to
determine the bacterial growth. The OD600 value of the
broth was used as a reference to determine the complete

Medicinal Chemistry Research
Guckian KM, Schweitzer BA, Ren RX, Sheils CJ, Tahmassebi DC,
Kool ET (2000) Factors contributing to aromatic stacking in
water: evaluation in the context of DNA. J Am Chem Soc
Oliveira JF, Silva CA, Barbieri CD, Oliveira GM, Zanetta DM,
Burdmann EA (2009) Prevalence and risk factors for aminogly-
coside nephrotoxicity in intensive care units. Antimicrob Agents
Chemother 53:2887–2891
Ramirez MS, Tolmasky ME (2010) Aminoglycoside modifying
enzymes. Drug Resist Updat 13:151–171
1
22:2213–2222
Guthrie OW (2008) Aminoglycoside induced ototoxicity. Toxicology
49:91–96
2
Houghton JL, Green KD, Chen W, Garneau-Tsodikova S (2010) The
future of aminoglycosides: the end or renaissance? ChemBio-
Chem 11:880–902
Kirk SR, Luedtke NW, Tor Y (2000) Neomycin−acridine conjugate: a
potent inhibitor of rev-rre binding. J Am Chem Soc 122:980–981
Liu M, Haddad J, Azucena E, Kotra LP, Kirzhner M, Mobashery S
Vakulenko SB, Mobashery S (2003) Versatility of aminoglycosides
and prospects for their future. Clin Microbiol Rev 16:430–450
Wellenzohn B, Flader W, Winger RH, Hallbrucker A, Mayer E, Liedl
KR (2001) Exocyclic groups in the minor groove influence the
backbone conformation of DNA. Nucleic Acids Res
29:5036–5043
(
2000) Tethered bisubstrate derivatives as probes for mechanism
Wong C-H, Hendrix M, Scott Priestley E, Greenberg WA (1998)
Specificity of aminoglycoside antibiotics for the A-site of the
decoding region of ribosomal RNA. Chem Biol 5:397–406
Xue L, Charles I, Arya DP (2002) Pyrene–neomycin conjugate: dual
recognition of a DNA triple helix. Chem Commun 70–71
Xue L, Ling M, Wang S (2017) A reliable method to measure the
and as inhibitors of aminoglycoside 3‘-phosphotransferases. J
Org Chem 65:7422–7431
Liu W, Minier MA, Franz AH, Curtis M, Xue L (2011) Synthesis of
nucleobase-calix[4]arenes via click chemistry and evaluation of
their complexation with alkali metal ions and molecular assem-
bly. Supramol Chem 23:806–818
melting temperatures of DNA oligonucleotide triplexes.
J
Liu W, Wang S, Dotsenko IA, Samoshin VV, Xue L (2017) Aryl-
sulfanyl groups-suitable side chains for 5-substituted 1,10-phe-
nanthroline and nickel complexes as g4 ligands and telomerase
inhibitors. J Inorg Biochem 173:12–20
Undergrad Chem Res 16:19–23
Xue L, Ranjan N, Arya DP (2011) Synthesis and spectroscopic studies
of the aminoglycoside (neomycin)--perylene conjugate binding to
human telomeric DNA. Biochemistry 50:2838–2849
Lu W, Sengupta S, Petersen JL, Akhmedov NG, Shi X (2007) Mit-
sunobu coupling of nucleobases and alcohols: An efficient,
practical synthesis for novel nonsugar carbon nucleosides. J Org
Chem 72:5012–5015
Zhang J, Keller K, Takemoto JY, Bensaci M, Litke A, Czyryca PG,
Chang CW (2009) Synthesis and combinational antibacterial
study of 5”-modified neomycin. J Antibiot 62:539–544