Novel purine benzimidazoles as antimicrobial agents by regulating ROS generation and targeting clinically resistant Staphylococcus aureus DNA groove

Article abstract of DOI:10.1016/j.bmcl.2018.03.046

A novel series of purine benzimidazole hybrids were designed and synthesized for the first time with the aim to circumvent the increasing antibiotic resistance. Hexyl appended hybrid 3c gave potent activities against most of the tested bacteria and fungi especially against multidrug-resistant strains Staphylococcus aureus (MIC = 4 μg/mL). Structure-activity relationships revealed that the benzimidazole fragment at the 9-position of purine played an important role in exerting potentially antibacterial activity. Both cell toxicity and ROS generation assays indicated that the purine derivative 3c showed low cytotoxicity and could be used as a safe agent. Molecular modeling suggested that hybrid 3c could bind with the residues of Topo IA through hydrogen bonds and electrostatic interactions. Quantum chemical studies were also performed on the target compound 3c to understand the structural features essential for activity. The active molecule 3c could effectively interact with S. aureus DNA to form 3c–DNA complex through groove binding mode, which might block DNA replication to display their powerful antimicrobial activity.

Full text of DOI:10.1016/j.bmcl.2018.03.046

Accepted Manuscript
Novel purine benzimidazoles as antimicrobial agents by regulating ROS gen-
eration and targeting clinically resistant Staphylococcus aureus DNA groove
Ya-Nan Wang, Rammohan R. Yadav Bheemanaboina, Gui-Xin Cai, Cheng-He
Zhou
PII:
S0960-894X(18)30239-7
https://doi.org/10.1016/j.bmcl.2018.03.046
BMCL 25700
DOI:
Reference:
Bioorganic & Medicinal Chemistry Letters
To appear in:
Received Date:
Revised Date:
Accepted Date:
31 December 2017
28 February 2018
17 March 2018
Please cite this article as: Wang, Y-N., Yadav Bheemanaboina, R.R., Cai, G-X., Zhou, C-H., Novel purine
benzimidazoles as antimicrobial agents by regulating ROS generation and targeting clinically resistant
Staphylococcus aureus DNA groove, Bioorganic & Medicinal Chemistry Letters (2018), doi: https://doi.org/
10.1016/j.bmcl.2018.03.046
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers
we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and
review of the resulting proof before it is published in its final form. Please note that during the production process
errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

Bioorganic & Medicinal Chemistry Letters
journal homepage: www.els evier.com
Novel purine benzimidazoles as antimicrobial agents by regulating ROS
generation and targeting clinically resistant Staphylococcus aureus DNA groove
Ya-Nan Wang, Rammohan R. Yadav Bheemanaboina,a Gui-Xin Cai* and Cheng-He Zhou*
Institute of Bioorganic & Medicinal Chemistry, Key Laboratory of Applied Chemistry of Chongqing Municipality, School of Chemistry and Chemical
Engineering, Southwest University, Chongqing 400715, China.
* E-mail: gxcai@swu.edu.cn, zhouch@swu.edu.cn (C. H. Zhou); Fax: +86-23-68254967; Tel: +86-23-68254967.
a Postdoctoral fellow from CSIR-Indian Institute of Integrative Medicine (IIIM), India.
ARTICLE INFO
ABSTRACT
Article history:
Received
Revised
Accepted
Available online
A novel series of purine benzimidazole hybrids were designed and synthesized for the first time
with the aim to circumvent the increasing antibiotic resistance. Hexyl appended hybrid 3c gave
potent activities against most of the tested bacteria and fungi especially against multidrug-
resistant strains Staphylococcus aureus (MIC = 4 µg/mL). Structure-activity relationships
revealed that the benzimidazole fragment at the 9-position of purine played an important role in
exerting potentially antibacterial activity. Both cell toxicity and ROS generation assays indicated
that the purine derivative 3c showed low cytotoxicity and could be used as a safe agent.
Molecular modeling suggested that hybrid 3c could bind with the residues of Topo IA through
hydrogen bonds and electrostatic interactions. Quantum chemical studies were also performed
on the target compound 3c to understand the structural features essential for activity. The active
molecule 3c could effectively interact with S. aureus DNA to form 3c–DNA complex through
groove binding mode, which might block DNA replication to display their powerful
antimicrobial activity.
Keywords:
Purine
Benzimidazole
Antibacterial
Antifungal
ROS generation
DNA groove
2018 Elsevier Ltd. All rights reserved.
Microbial infection is a common and frequently occurring
infectious disease worldwide. A large number of antibiotics and
synthetic drugs have been widely used in clinic and have played
an important role in the treatment of microbial infections.1
However, the high morbidity and mortality caused by bacteria
and fungi are still increasing serious threat due to the appearance
of drug-resistant strains in recent decades.2 Multiple strategies
have been adopted to prevent and control antimicrobial
resistance, one useful strategy is to develop the new drug hybrids
which are generally considered a practical promise to tackle the
growing problem of drug resistance.3
processes, it arouses our immense interest to modify the purine
skeleton using isosteric ring systems to develop novel structural
molecules and investigate their possibility as new antimicrobial
agents.
Azoles as nitrogen-containing five-membered aromatic
heterocycles with desirable electron rich characteristic can
readily bind with a variety of enzymes and receptors in biological
systems through diverse weak noncovalent interactions, thereby
exhibiting broad bioactivities.7 Thus, the combination of azoles
and the purine nucleus provides a great prospect for medicinal
and biological application. They have been developed as
antimycobacterial and antiviral drugs and also as adenosine
receptor agonists and antagonists. Especially, benzimidazole with
special benzene-fused imidazole ring as a key component for
various biological activities like antiparasitic, anticancer, anti-
inflammatory and antiulcer agents etc. has attracted considerable
concern.8 Recently, extensive biochemical and pharmacological
studies revealed that benzimidazoles possessed large potentiality
to inhibit the growth of bacterial and fungal strains.9 Further
research disclosed that benzimidazole nucleus structurally
resembling with purine scaffold in nucleotides was sometimes
addressed as 1,3-dideaza-purine and had the capability of
competing with purines, distinctly inhibiting the synthesis of
nucleic acids and proteins, thereby killing bacterial strains or
inhibiting their growth.10,11 Clearly, benzimidazole-based
Purines can be regarded as one of the most ubiquitous and
functional N-heterocyclic compounds found in nature. Purine
scaffold is the core structural fragment of adenine and guanine in
RNA and DNA.4 Purine nucleotides such as ATP, GTP, cAMP,
cGMP, NAD and FAD act as co-factors, substrates or mediators
in the function of many proteins.5 These proteins are estimated to
include half of the most pharmaceutically available targets,
primarily enzymes and receptors. Some naturally existing and
synthetic purines have been applied as antitumor, antiviral and
antiparasitic agents and adenosine receptor ligands while the
application in antimicrobial aspect has rarely been reported.6
Therefore, the development of purine-based novel antimicrobial
agents is of significant value in medicinal chemistry. In view of
the pivotal role of purines in the regulation of many biological

derivatives possess great potential as new antibacterial and
antifungal agents. Reasonably, this structural similarity prompts
us for the first time to combine benzimidazole moiety and purine
backbone to generate a novel type of hybrids as potentially
antimicrobial agents (Figure 1) and evaluate their antimicrobial
potency.
different lengths and substituted phenyl moieties including chloro
and fluoro groups were introduced at N-1 position on
benzimidazole moiety to investigate their effect on
bioactivities;13
(4) The substituted groups at C-2, 5, 6 positions on
benzimidazole nucleus had the capacity to modulate the
pharmacokinetic properties and enhance the antimicrobial
efficacy. The substituents on benzimidazole nucleus were
changed to explore their impacts on biological activities;
Design idea for target compounds was mainly from three
aspects: (I) Structural modification on the benzimidazole
nucleus; (II) Structural change on the purine scaffold; (III) The
different linkers between benzimidazole and purine rings. The
related considerations were given as follows:
(5) It is well known that saturated nitrogen heterocycles such
as morpholine, piperidine and pyrrolidine are beneficial building
blocks in drug design and prevalently present in many approved
drugs.14 The above mentioned alicyclic amines were introduced
to C-6 position on purine nucleus for investigating the influence
of the substituents in the purine nucleus on structure-activity
relationship and bioactive profiles.
(1) The C-6 and N-9 positions in the purine scaffold are the
most widely exploited anchoring points and the substituted
purines are commercially available.12 Therefore the purine
nucleus through the N-9 position with benzimidazoles was
hybridized;
(2) The incorporated flexible –CH2– aliphatic linker between
benzimidazole and purine nucleus is helpful for regulating the
molecular conformation to interact with targets. The N-alkyl
benzimidazole moiety was reported to have good performance in
exerting bioactivities, the–CH2CH2– aliphatic linker was also
introduced into target compounds in order to evaluate its
contribution towards antimicrobial efficacy;
The designed structures of this series of novel purine
benzimidazoles are shown in Schemes 1–2. All the newly
synthesized compounds were characterized by spectral analyses
and evaluated for their antimicrobial activities in vitro against
five Gram-positive bacteria, six Gram-negative bacteria and five
fungi according to the Clinical and Laboratory Standards Institute
(CLSI).15 Structure-activity relationship, cytotoxicity, reactive
oxygen species, molecular docking and quantum chemical
studies as well as the preliminary antimicrobial mechanism with
DNA to highly active compound were also discussed and
investigated.
(3) A lot of work revealed that substituents at N-containing
heterocycle could significantly influence the pharmacological
properties by adjusting lipid-water partition coefficient and
binding affinity. Rationally, various aliphatic chains with
Figure 1. Design of novel purine benzimidazole hybrids.
The target purine benzimidazole hybrids were prepared
starting from commercial 6-chloro-9H-purine. Their synthetic
routes were outlined in Schemes 1–2 and Schemes S1–S2. The
reaction of 6-chloro-9H-purine with morpholine at 80 oC
produced 4-(9H-purin-6-yl)morpholine 1 and then was further
treated by benzimidazoles 8, 10 and 12 in dimethylformamide to
give the derivatives 2, 3 and 4 respectively in yields of 48.8–
71.5% (Scheme 1). The commercial o-phenylenediamines were
reacted with various carboxylic acids through the cyclization to
form benzimidazoles 7a–f, which were subsequently proceeded
N-alkylation with 1,2-dibromoethane in acetonitrile using
potassium carbonate at 70 oC to afford N-substituted
benzimidazoles 8a–f in 46.3%–54.7% yields (Scheme S1).
Intermediates 10 and 12 were prepared from the cyclization of N-
mono substituted o-phenylenediamine with chloroacetic acid in
hydrochloric acid at reflux in 74.2–83.6% yields. It was noticed
that N-monosubstituted o-phenylenediamines 9 and 11 were
provided in yields of 62.4–69.7% by the nucleophilic substitution
reaction of o-phenylenediamine with a series of halogenated
aliphatic chains at ratio of 1.2 : 1 (Scheme S2). It was reported
that the presence of aliphatic nitrogen heterocycles had some
influence on the bioactivity, and thus target compounds 6a and
6b were further prepared to explore their effect on the
bioactivities (Scheme 2). The starting material 6-chloro-9H-
purine was also reacted with piperidine and pyrrolidine in ethanol
at reflux using triethylamine as base to produce intermediates 5a
and 5b with 92.7% and 91.4% yields, respectively. Further
coupling with N-hexyl benzimidazole 10c in dimethylformamide
at 50 oC was carried out to generate the final products 6a and 6b
in 67.4% and 69.3% yields.

Scheme 1. Synthetic routes of purine benzimidazoles 2–4. Reagents and conditions: (i) morpholine, 80 oC; (ii) benzimidazoles 8a–f, potassium carbonate,
dimethylformamide, 50 oC; (iii) benzimidazoles 10a–i, potassium carbonate, dimethylformamide, r.t.; (iv) benzimidazoles 12a–g, potassium carbonate,
dimethylformamide, r.t..
Scheme 2. Synthetic routes of purine benzimidazoles 6a–b. Reagents and conditions: (i) piperidine/pyrrolidine, triethylamine, ethanol, reflux; (ii)) compound
10c, dimethylformamide, potassium carbonate, 50 oC.
The in vitro antimicrobial activities for all the target
compounds were evaluated for five Gram-positive bacteria
(Methicillin-Resistant Staphylococcus aureus N315 (MRSA),
Enterococcus faecalis, Staphylococcus aureus, Staphylococcus
aureus ATCC 25923 and Staphylococcus aureus ATCC 29213),
six Gram-negative bacteria (Klebsiella pneumonia, Escherichia
coli, Pseudomonas aeruginosa, Acinetobacter baumanii,
Pseudomonas aeruginosa ATCC 27853, Escherichia coli ATCC
25922) and five fungi (Candida albicans, Candida tropicalis,
Aspergillus fumigates, Candida albicans ATCC 90023, Candida
parapsilosis ATCC 22019) using two folds serial dilution
technique in 96–well micro-test plates recommended by CLSI
with the positive control of clinically antimicrobial drugs
norfloxacin (Table 1) and fluconazole (Table S1). As shown in
Table 1, most of the prepared compounds gave weak to moderate
antibacterial activity against the tested strains in vitro.
Preliminary active screening revealed that morpholine derivative
3c showed better antibacterial efficacy than other alicyclic amine
ones 6a and 6b, and more importantly hexyl-appended hybrid 3c
exhibited stronger inhibition ability against most of the tested
bacterial strains among the all target compounds.
carbons) should be not beneficial to regulate the hydrophilicity of
this purine benzimidazole system for their antibacterial activity.
The hexyl and decyl modified derivatives 3c and 3e exhibited
relatively stronger antibacterial activities than reference
norfloxacin against P. aeruginosa ATCC27853. It was
noteworthy that hybrid 3c, with prominent MIC value of 4
µg/mL against S. aureus, 16-fold more potent than clinical drug
norfloxacin, had immense potentiality to be developed as anti S.
aureus agent. However, the allyl and propargyl derivatives 3h–i
did not display obvious effects on suppressing the growth of
bacteria, which indicated that the unsaturated aliphatic chain was
not propitious to exert inhibitory potency.
Among phenyl ones, fluorobenzyl derivatives 4b, 4d and 4f
also gave equivalent or even lower inhibitory concentrations than
chlorobenzyl ones 4c, 4e and 4g towards S. aureus ATCC29213,
P. aeruginosa and P. aeruginosa ATCC27853, which implied
that fluorine atom on the benzyl moieties exerted positive effect
on microbial inhibition. Probably because fluorine atom could
easily and efficiently form non-covalent forces thereby being
helpful for the biological transportation and distribution in
organism. In addition, it was also found that the substituted
position also exerted some influence on the antibacterial activity
with the order of the o-fluoro/chlorobenzyl derivatives > m-
fluoro/chloro ones > p-fluoro/chloro ones against most of the
tested strains.
Further studies focused on the influence of alkyl chains with
different lengths (series 3) and aromatic frameworks with
different halogen moieties (series 4) on antimicrobial activity
(Table 1). It could be observed that the antibacterial efficacies of
series 3 and 4 were comparable or even better than compound 1
towards some tested strains like S. aureus and P. aeruginosa
ATCC27853 manifesting the importance of benzimidazole
moieties. It was indicated that long alkyl chains (over ten
The introduction of benzimidazoles at the 9-position of purine
scaffolds using –CH2CH2– as a bridge yielded compound series
2. None of the tested strains was sensitive to compound 2a.

Table 1. Antibacterial data as MIC (µg/mL) for target compounds 1–6.a,b,c
Gram-positive bacteria
Gram-negative bacteria
P. A.
pneumoniae coli aeruginosa baumanii
P.
Compds
E.
S.
S. aureus
S. aureus
ATCC29213
K.
E.
E. coli
ATCC25922
MRSA
aeruginosa
ATCC27853
faecalis aureus ATCC25923
1
512
256
128
128
256
256
512
256
512
256
128
256
256
256
256
128
256
256
256
256
256
128
2a
2b
2c
2d
2e
2f
256
512
256
256
256
256
512
16
128
256
64
256
256
32
256
32
128
128
128
64
64
256
64
256
512
32
32
256
128
64
256
256
64
128
256
64
256
256
128
16
64
128
64
16
128
64
128
32
128
128
128
64
32
64
32
128
256
128
512
256
256
512
256
128
256
256
16
8
64
32
128
128
128
8
3a
3b
3c
3d
3e
3f
128
256
128
256
128
128
32
256
256
4
256
256
64
256
256
64
128
256
64
256
256
64
256
128
16
128
256
32
512
256
512
512
512
512
256
256
512
512
512
256
512
512
64
512
256
16
512
256
512
256
256
128
512
128
256
32
256
128
256
128
128
128
128
64
256
256
128
512
256
256
256
32
256
256
512
256
128
256
256
128
256
256
256
256
512
128
64
128
16
256
128
256
256
256
512
256
256
256
256
256
256
256
256
16
32
128
256
128
128
128
32
3g
3h
3i
64
512
256
128
256
64
128
256
128
64
256
128
512
256
256
256
256
512
512
32
4a
4b
4c
4d
4e
4f
256
128
128
256
256
256
128
512
256
256
32
64
128
256
256
128
256
256
256
256
256
512
128
256
512
256
512
256
128
256
128
4
64
64
16
128
256
64
128
64
128
256
256
512
512
512
512
8
128
128
256
256
64
4g
5a
6a
5b
6b
A
128
512
64
128
64
16
128
128
256
512
512
64
64
512
64
512
128
>512
256
64
256
16
32
>512
64
32
32
1
a Minimum inhibitory concentrations were determined by micro broth dilution method for microdilution plates and MICs was completely inhibited concentration.
b
MRSA, Methicillin-Resistant Staphylococcus aureus N315; E. faecalis, Enterococcus faecalis; S. aureus, Staphylococcus aureus; S. aureus ATCC25923,
Staphylococcus aureus ATCC25923; S. aureus ATCC29213, Staphylococcus aureus ATCC29213; K. pneumonia, Klebsiella pneumonia; E. coli, Escherichia
coli; P. aeruginosa, Pseudomonas aeruginosa; A. baumanii, Acinetobacter baumanii; P. aeruginosa ATCC27853, Pseudomonas aeruginosa ATCC27853; E.
coli ATCC25922, Escherichia coli ATCC25922.
c A = Norfloxacin.
However, the further modifications at the benzimidazole nucleus
showed large effect on antibacterial activity. The electron-
withdrawing trifluoromethyl, nitro and acetyl groups and
electron-donating methyl groups were introduced at the 2-, 5- and
6-positions of the benzimidazole moiety. It was found that the
electron-withdrawing groups were effective for enhancing anti-
Gram-negative bacteria activities. Specifically, 2-acetyl
derivative 2f showed good MIC values (8 µg/mL) against P.
aeruginosa and stronger antibacterial potency than reference
drug against K. pneumoniae (32 µg/mL) and P. aeruginosa
ATCC27853 (32 µg/mL). In comparison to electron-withdrawing
groups, the electron-donating methyl purine benzimidazoles 2b–c
showed much poorer inhibitory potency.
antifungal potency which suggested that alicyclic amines and
benzimidazole moieties were positive to antimicrobial activities.
In the series of purine benzimidazoles 3a−i, the lengths of
aliphatic chains exhibited obvious effects on antifungal activity.
The target molecule 3c exerted the relatively good activities in
inhibiting the growth of all the tested fungal strains, especially
possessed lower MIC values (32−128 µg/mL) than the reference
−256 µg/mL) against clinical
fluconazole (64
ly drug-resistant
−g
strains. Among benzyl series of purine benzimidazoles 4a , 2-
and 3-fluorobenzyl ones resulted in relatively good activities
against C. albicans ATCC90023 with the MIC values of 64 and
32 µg/mL, respectively. The replacement of fluorobenzyl moiety
by chlorobenzyl fragment did not give obvious effects on
antifungal activities which displayed that fluorine atom was
advantageous in inhibiting the growth of microorganisms. The
The antifungal evaluation in vitro displayed that some
prepared purine benzimidazoles exhibited moderate bioactivities
against the tested fungal strains (Table S1). Compound 1 showed
poor antifungal activities with MIC values ranging from 256 to
512 µg/mL, while compounds 3c, 6a and 6b exerted stronger
−CH
−CH
−f
change of linker
2− to
2CH2− yielded compounds 2a
,
exhibited low activities against all test strains. Notably,

compound 2e showed much stronger activities (64 µg/mL)
against C. tropicalis than fluconazole (MIC = 256 µg/mL).
The most noticeable difference came from the substituents at the
1-position of the benzimidazole nucleus. In the optimized
structures of series 3 showing folded shape, oxygen and nitrogen
atoms with electron negatively properties had a strong ability to
Computational methods can be used to predict important
pharmacokinetic properties of drug candidates and have been
employed to rationalize biological activity. The molecular
electrostatic potential (MEP) surface gives an indication of the
charged surface area and hydrophilicity of compounds.16
Therefore, MEP maps of the target compounds were produced to
investigate key structural features such as steric and electrostatic
−π conjugative effect
interact with the enzyme. Because of the π
and steric hindrance, series 4 with unfolded shape possessed
weaker negatively charged surface area. Compared to the MEP
surfaces of target compounds 3, the electron negatively charged
zones of the nitrogen atoms were slightly decreased as a result of
the replacement of morpholine by the pyrroline at the 6-position
of the purine ring. When using –CH2CH2– group as bridge, the
electron positively charged zones greatly enhanced. On the other
hand, the introduction of the electro-withdrawing groups like
nitro group had much influence on the electron negatively
properties of the molecules.
interactions, hydrogen
donor/acceptor properties, and
lipophilicity. As shown in Table 2, the selection of compounds at
the neutral state resulted in a comparison of MEP surfaces by
DFT-B3LYP theoretical calculations.
In the comparison to the MEP maps of the two modes of
purine benzimidazoles, series 3 and 4 were taken as examples.
Table 2. Comparison of the MEP surfaces of target compounds 2e, 3c, 4b and 6b.
Compds
MEPs
Compds
MEPs
-8.000e-2
8.000e-2
120
100
80
60
40
20
0
to be one of the most important properties for active drug
candidates.
Compound 3c
Evading persistent drug resistance in bacteria is essential for
the recovery of human health and impels intervention strategies.
E. coli and other life threatening infectious pathogens such as S.
aureus and M. tuberculosis are significantly sensitive to changes
in the intracellular oxidative environment. ROS are metabolic
byproducts of aerobic respiration and are responsible for
maintaining redox homeostasis in cells.17 Therefore, small
molecules that enhance intracellular ROS can interfere with the
cellular oxidative environment and induce cell death, and hence
can serve as a new standard for determining cytotoxicity. ROS
level was investigated using DCFH-DA as a fluorescence probe
and imaged by fluorescence microscope. As shown in Figure 3,
target compound 3c showed much lower ROS generation than
H2O2 at concentration of 6.25, 12.5, 25, 50 µM, respectively.
Therefore, the newly synthesized compound 3c could possess an
excellent safety profile by regulating ROS generation.
0
100
200
300
400
500
Concentration of compound 3c (mg/mL)
Figure 2. Cytotoxicity assay with compound 3c in the A549 cell line were
tested by MTT methodology. Each data bar is an average of five replicates.
The active molecule 3c was further evaluated for its toxicity in
vitro against human lung adenocarcinoma cells (A549) by the
colorimetric cell proliferation MTT assay. Experimental results
revealed that the cell viability of A549 cell against compound 3c
was more than 85%, which suggested that this molecule showed
low toxicity to the cells at concentrations < 500 µg/mL and the
cell viability slowly decreased with the increasing concentration
of compound (Figure 2). Compound 3c did not affect the cell
viability of A549 within the MIC value. All demonstrated that
compound 3c possessed low cell toxicity which was considered
To rationalize the observed antimicrobial activity and to
understand the possible mechanism of the hybrids, molecular
docking of active compound 3c with DNA topoisomerase IA
(Topo IA) was undertaken. The crystal structure data was
obtained from the Protein Data Bank. The interaction of
compound 3c with Topo IA receptor was shown in Figure 4 and
Figure S5. The docking result of interaction between compound
3c and Topo IA might rationalize the possibly antibacterial
mechanism. The nitrogen atoms of benzimidazole ring at 3-

position and purine ring at 3-position were in close vicinity to the
residue GLN 197 of Topo IA through hydrogen bonds with
distance of 2.8 and 2.0 Å, respectively, which revealed that the
nitrogen atom played important role in biological activity.
Besides, electrostatic interactions existed between the aromatic
fragment of compound 3c and residues THR 496, VAL 196 and
ILE 500 in Topo IA. Specifically, morpholine ring could also
form electrostatic interactions with residue TYR 177. These
hydrogen bonds and electrostatic interactions might be
responsible for the good inhibitory efficacy of compound 3c
against the tested strains. Furthermore, quantum chemical studies
were also carried out to validate the capability of hydrogen bond
formation through the nitrogen atoms of purine skeleton and
benzimidazole nucleus (Table 2 and S3).
absorption spectra were detected with the increasing
concentration of compound 3c. As shown in Figure S9, UV–vis
spectra displayed that the maximum absorption peak of DNA at
260 nm was gradually enhanced, simultaneously along with a
slightly red shift during titration with the proportional addition of
compound 3c which was different from compound 3c alone
(Figure S8). Meanwhile, spectral data displayed that the
absorption value of simply sum of free DNA and free compound
3c was a little greater than the measured value of 3c–DNA
complex which revealed a weak hypochromic effect existed
between DNA and compound 3c. This hypochromic effect on the
spectra of 3c–DNA complex and the slight red spectral shift
might be resulted from the charge attraction of prepared molecule
3c to DNA base eventually leading to formation of binary
complexes. Moreover, the intercalation of the aromatic fragment
of compound 3c into the helix and the strong overlap of π-π*
states in the large π-conjugated system with the electronic states
of DNA bases were in accordance with the observed spectral
changes.
To learn more about the interaction between compound 3c and
DNA, the absorption spectra of competitive interaction of
compound 3c was investigated. Neutral Red (NR) is a planar
phenazine dye and has been widely employed as probe due to its
lower toxicity, higher stability and more convenient application
than other common probes. Furthermore, spectrophotometric and
electrochemical
techniques
have
already
sufficiently
demonstrated that NR can bind with DNA through an
intercalative mode. Therefore, NR was taken as a spectral probe
to study the binding mode of 3c with DNA in our work. The
absorption spectra of the NR dye upon the addition of DNA are
shown in (Figure S11). Experimental results displayed that the
absorption peak of the NR around 460 nm gradually decreased
with increasing concentration of DNA and a new band at around
530 nm developed, which suggested the formation of new DNA–
NR complex. An isosbestic point at 504 nm provided evidence of
DNA–NR complex formation. Figure S12 displayed the
absorption spectra of a competitive binding between NR and 3c
with DNA. As shown, with the increasing concentration of 3c, an
apparent intensity change was not observed in the developing
band around 460 nm. Compared with the absorption around 460
nm of free NR in the presence of the increasing concentrations of
DNA (Figure S11), the absorbance at the same wavelength did
not exhibit the reverse process (inset of Figure S12). The results
disclosed that compound 3c failed to intercalate into the double
helix of DNA by substituting for NR in the DNA–NR complex.
Figure 3. Compound 3c induced ROS formation in Hela cells and H2O2 as
The fluorescence quenching spectra of compound 3c with
increasing amounts of DNA at the excitation wavelength of 278
nm was shown in Figure S13. The fluorescence intensity of
compound 3c at around 316 nm regularly decreased, but the
maximum emission wavelength of compound 3c did not
obviously change with the increase of DNA concentration. The
results indicated that DNA could quench the intrinsic
fluorescence of compound 3c and the binding of compound 3c
with DNA indeed existed.
positive control.
Dynamic quenching and static quenching are the two well-
known quenching processes. Dynamic quenching is a process
that the fluorophore contact with the quencher during the
transient presence of the excited state, whereas static quenching
results in the formation of fluorophore and quencher complex.
The method to distinguish static and dynamic quenching is to
calculate their different binding constants based on temperature
or viscosity or to measure their lifetimes. In this work, the
calculation of binding constant dependent on the temperature was
used to elucidate the quenching mechanism.
Figure 4. Three-dimensional conformation of compound 3c docked in Topo
IA (PDB code: 3PWT).
To investigate the preliminary mechanism of antimicrobial
action, the binding studies in vitro between the active molecule
3c and DNA isolated from S. aureus bacterial strain (Figure S6
and Figure S7) on molecular level were carried out by UV–vis
spectroscopy. With a fixed concentration of DNA, UV–vis

The fluorescence quenching data can be analyzed by the well-
known Stern-Volmer equation:
Table S5 listed the thermodynamic parameters for the
interaction of compound 3c with DNA. The negative values of
∆G indicated that the binding process was spontaneous. The
negative values of ∆H suggested that the binding process was
primarily driven by the enthalpy and through the hydrogen
binding interactions. The negative ∆S value is often used as an
evidence for vander Waals force. Therefore, ∆H < 0 and ∆S < 0
obtained in this case manifested that both hydrogen bonds and
vander Waals force exerted an important role in the binding of
compound 3c to DNA and was conducive to the stability of the
complex.
F0
(1)
= 1 + KSV[Q]
F
where F0 and F represent fluorescence intensity of compound
3c in the absence and presence of DNA, respectively. KSV
(L/mol) is the Stern-Volmer quenching constant, and [Q] is the
concentration of DNA. The Stern-Volmer equation was applied
to determine KSV by linear regression of a plot of F0/F against
[Q]. Hence, the value of KSV at different temperatures could be
calculated and was shown in Figure S14.
It is well known that the highly negatively charged quencher
is expected to be repelled by the negatively charged phosphate
backbone of DNA. Therefore, if a molecule is protected from
being quenched by anionic quencher, the binding pattern of the
small molecule with DNA is intercalation. While different from
intercalation, groove binding provides much less protection for
the bound molecule. The value of quenching constants (KSV) for
the groove binding should be higher than that of the molecule
bounding to DNA by intercalative pattern. Negatively charged I−
ion was chosen for this purpose. The values of KSV of compound
3c by I− ion in the absence and presence of DNA were calculated
to be 28.74 (R = 0.994) and 78.24 (R = 0.998) L/mol,
respectively, by the Stern-Volmer equation (Figure S17). The
results showed that iodide quenching effect was enhanced when
compound 3c was bound to DNA, suggesting that the binding
mode of compound 3c with DNA might be of groove binding
nature.
The results showed that in dynamic quenching, the quenching
constants were expected to increase with a gradually increasing
temperature due to larger diffusion coefficients. While the
increase of temperature is likely to lead to a smaller KSV in static
quenching due to the dissociation of weakly bound complexes.
Figure S14 showed that the Stern-Volmer quenching constant,
KSV, decreased linearly with increasing temperatures. These
calculated KSV were 1.87 × 104 (R=0.9999), 1.52 × 104
(R=0.9999) and 1.21 × 104 (R=0.9999) L/mol at 294, 304 and
314 K, respectively, at pH 7.4. Thus, the fluorescence quenching
mechanism of compound 3c interacted with DNA was verified to
be resulted from static quenching. Therefore, the fluorescence
quenching of compound 3c by DNA could be described by the
modified Stern-Volmer equation:
F0
1
1
1
+
(2)
=
∆F
f Ka [Q] fa
a
In conclusion, we have successfully developed a series of
novel purine benzimidazole hybrids as safely antimicrobial
agents via convenient and efficient synthetic procedures. All new
compounds were confirmed by 1H NMR, 13C NMR, IR and
HRMS spectra. Some of the synthesized compounds exhibited
good or even superior bioactivities to reference drugs against the
tested bacterial strains especially against the clinically resistant S.
aureus. The structure-activity relationship found that the
benzimidazole moiety and various substituents played important
roles in the biological activity and hexyl derivative 3c was the
most prominent one (MIC = 4 µg/mL against S. aureus).
Moreover, the active molecule 3c also displayed low cell toxicity
and could be used a safe agent. Molecular docking revealed that
compound 3c could bind with the residues of Topo IA through
hydrogen bonds and electrostatic interactions. Further quantum
chemical studies validated the capability of hydrogen bond
formation through the nitrogen atoms of purine skeleton and
benzimidazole nucleus which was consistent with the binding
mode obtained from the docking result. The preliminary action
mechanism investigation revealed that hybrid 3c could interact
with DNA isolated from S. aureus through groove binding driven
by hydrogen bonds and van der Waals force to form 3c–DNA
complex, thus exerting its powerful antibacterial and antifungal
activities. Hence, compound 3c could open new prospect for
being developed as therapeutic agent for the purpose of coping
multidrug resistant bacterial infections. Further researches,
including the in vivo bioactive evaluation, resistance study, time-
kill kinetic assay and introduction of various functional
substituents into the hybrid backbone to investigate the effects on
antimicrobial activities are now in progress in our group.
where ∆F is the difference in fluorescence intensity of
compound 3c in the absence and presence of DNA at
concentration [Q], fa is the fraction of accessible fluorescence,
and Ka is the effective quenching constant for the accessible
fluorophores, which are analogous to associative binding
constants for the quencher-acceptor system. The dependence of
F0/∆F on the reciprocal value of quencher concentration 1/[Q] is
linear with the slope equaling to the value of 1/( faKa). The value
of 1/fa is fixed on the ordinate. The constant Ka is a quotient of
the ordinate 1/fa and the slope 1/(faKa). The modified Stern-
Volmer plots were showed in Figure S15 and the calculated
results were depicted in Table S4. The decreasing trend of Ka
with increasing temperature was in accordance with KSV’s as
discussed above, which accorded with static quenching.
Generally, the noncovalent interaction such as hydrogen
bonds, van der Waals force, hydrophobic force and electrostatic
interactions are the principal force for binding between small
molecules and biomacromolecules. The main evidence to
confirm the binding force is the thermodynamic parameters of
binding reaction. If the enthalpy change (∆H) does not vary
significantly over the studied temperatures range, then its value
and that of entropy change (∆S) can be determined from the van’t
Hoff equation:
∆H
∆S
+
(3)
lnK = −
RT
R
where K is analogous to the associative binding constants at
the corresponding temperature and R is the gas constant. The
temperatures used were 294, 304 and 314 K, respectively. The
values of ∆H and ∆S were obtained from the slope and intercept
of the linear van’t Hoff plot based on lnKa versus 1/T (Figure
S16). The free energy change (∆G) was then calculated from the
following equation:
Acknowledgments
This work was partially supported by National Natural
Science Foundation of China (Nos. 21672173 and 21372186) and
Chongqing Special Foundation for Postdoctoral Research
Proposal (No. Xm2017184).
(4)
∆G = ∆H − T∆S

Jeyakkumar, P.; Gopala, L.; Li, S.; Geng, R. X.; Zhou, C. H. Med.
Chem. Commun. 2017, 8, 907; (d) Zhang, Y.; Damu, G. L. V.; Cui,
S. F.; Mi, J. L.; Tangadanchu, V. K. R.; Zhou, C. H. Med. Chem.
Commun. 2017, 8, 1631.
References and notes
1. (a) Nübel, U. Curr. Top. Microbiol. 2016, 398, 35; (b) Cui, S. F.;
Addla, D.; Zhou, C. H. J. Med. Chem. 2016, 59, 4488; (c) Konai,
M. M.; Ghosh, C.; Yarlagadda, V.; Samaddar, S.; Haldar, J. J.
Med. Chem. 2014, 57, 9409; (d) Zhou, C. H.; Wang, Y. Curr. Med.
Chem. 2012, 19, 239.
14. (a) Bolleddula, J.; DeMent, K.; Driscoll, J. P.; Worboys, P.;
Brassil, P. J.; Bourdet, D. L. Biotransformation and bioactivation
reactions of alicyclic amines in drug molecules, Drug Metab. Rev.
2014, 46, 379; (b) Cavender, F. L. Patty's Toxicology (6th
Edition) 2012; Vol. 2, pp 433–516.
15. (a) Clinical and Laboratory Standards Institute/National
Committee for Clinical Laboratory Standards: Reference Method
for Broth Dilution Antifungal Susceptibility Testing of Yeast, 3d
ed. Wayne, PA: Clinical and Laboratory Standards Institute; 2009.
Approved Standard, Document M27–a3. (b) Clinical and
Laboratory Standards Institute/National Committee for Clinical
Laboratory Standards: Reference Method for Broth Dilution
Antifungal Susceptibility Testing of Yeast, 3rd ed. Wayne, PA:
Clinical and Laboratory Standards Institute; 2008. Approved
Standard, Document M38–A2.
16. (a) Li, Z. Z.; Gopala, L.; Tangadanchu, V. K. R.; Gao, W. W.;
Zhou, C. H. Bioorg. Med. Chem. 2017, 25, 6511; (b) Kang, J.;
Tangadanchu, V. K. R.; Gopala, L.; Gao, W. W.; Cheng, Y.; Liu,
H. B.; Geng, R. X.; Li, S.; Zhou, C. H. Chin. Chem. Lett. 2017, 28,
1369; (c) Cao, Q. C.; Wang, H.; Tangadanchu, V. K. R.; Gopala,
L.; Cai, G. X.; Zhou, C. H. Sci. Sin. Chim. 2017, 47, 844; (d) Gao,
W. W.; Gopala, L.; Bheemanaboina, R. R. Y.; Zhang, G. B.; Li, S.;
Zhou, C. H. Eur. J. Med. Chem. 2018, 146, 15.
2. (a) Schillaci, D.; Spanò, V.; Parrino, B.; Carbone, A.; Montalbano,
A.; Barraja, P.; Diana, P.; Cirrincione, G.; Cascioferro, S. J. Med.
Chem. 2017, 60, 8268; (b) Song, H.; Ahmad Nor, Y.; Yu, M.;
Yang, Y. N.; Zhang, J.; Zhang, H. W.; Xu, C.; Mitter, N.; Yu, C.
Z.; J. Am. Chem. Soc. 2016, 138, 6455; (c) Pomastowski, P.;
Sprynskyy, M.; Žuvela, P.; Rafińska, K.; Milanowski, M.; Liu, J.
J.; Yi, M.; Buszewski, B. J. Am. Chem. Soc. 2016, 138, 7899; (d)
Gao, W. W.; Zhou, C. H. Future Med. Chem. 2017, 9, 1461.
3. (a) Petchiappan, A.; Chatterji, D. ACS Omega. 2017, 2, 7400; (b)
Piddock, L. J. V. Nat. Rev. Microbiol. 2017, 15, 639; (c) Cheng,
Y.; Avula, S. R.; Gao, W. W.; Addla, D.; Tangadanchu, V. K. R.;
Zhang, L.; Lin, J. M.; Zhou, C. H. Eur. J. Med. Chem. 2016, 124,
935; (d) Zhang, H. Z.; Damu, G. L. V.; Cai, G. X.; Zhou, C. H.
Eur. J. Med. Chem. 2013, 64, 329.
4. (a) Yan, L. H.; Roux, A. L.; Boyapelly, K.; Lamontagne, A. M.;
Archambault, M. A.; Frédéric, P. J.; David, L. S.; SaintPierre E.;
Najmanovich, R.; Fortier, L. C.; Lafontaine, D.; Éric, M. Eur. J.
Med. Chem. 2018, 143, 755; (b) Reddy, P. L.; Khan, S. I.; Ponnan,
P.; Tripathi, M.; Rawat, D. S. Eur. J. Med. Chem. 2017, 126, 675.
5. (a) Chen, L.; Chen, D.; Tang, L.; Ren, J.; Chen, J. J.; Zhen, X. C.;
Liu, Y. C.; Zhang, C. H.; Luo, H. B.; Shen, J. K.; Xiong, B.
Bioorg. Med. Chem. 2017, 25, 3315; (b) Sunagar, M. G.; Gaonkar,
S.; Sunagar, S. G.; Deshapande, N.; Belavagi, N. S.; M. Khazi, I.
A. RSC Adv. 2016, 6, 15286; (c) Luo, Y. L.; Baathulaa, K.;
Kannekanti, V. K.; Zhou, C. H.; Cai, G. X. Sci. China Chem. 2015,
58, 483.
17. Dharmaraja, A. T. Role of Reactive Oxygen Species (ROS) in
Therapeutics and Drug Resistance in Cancer and Bacteria, J. Med.
Chem. 2017, 60, 3221.
6. (a) Yu, J.; Zhao, L. X.; Park, J.; Lee, H. W.; Sahu, P. K.; Cui, M.;
Moss, S. M.; Hammes, E.; Warnick, E.; Gao, Z. G.; Noh, M.;
Choi, Sun.; Ahn, H. C.; Choi, J.; Jacobson, K. A.; Jeong. L. S. J.
Med. Chem. 2017, 60, 3422; (b) Peng, X. M.; Damu, G. L. V.;
Zhou, C. H. Curr. Pharm. Des. 2013, 19, 3884; (c) Chen, Y. Y.;
Gopala, L.; Bheemanaboina, R. R. Y.; Liu, H. B.; Cheng, Y.;
Geng, R. X.; Zhou, C. H. ACS Med. Chem. Lett. 2017, 8, 1331.
7. (a) Novosjolova, I.; Bizdēna, Ē.; Turks, M. Eur. J. Org. Chem.
2015, 17, 3629; (b) Fang, X. F.; Li, D.; Tangadanchu, V. K. R.;
Gopala, L.; Gao, W. W.; Zhou, C. H. Bioorg. Med. Chem. Lett.
2017, 27, 4964; (c) Zhang, H. Z.; Jeyakkumar, P.; Kannekanti, V.
K.; Zhou, C. H. New J. Chem. 2015, 39, 5776; (d) Zhang, L.; Peng,
X. M.; Damu, G. L. V.; Geng, R. X.; Zhou, C. H. Med. Res. Rev.
2014, 34, 340.
8. (a) Keri, R. S.; Hiremathad, A.; Budagumpi, S.; Nagaraja, B. M.
Chem. Biol. Drug Des. 2015, 86, 19; (b) Moreiraa, J. B.; Manna, J.;
Neidlea, S.; McHughb, T. D.; Taylora, P. W. Int. J. Antimicrob.
Ag. 2013, 42, 361; (c) Ajani1, O. O.; Aderohunmu1, D. V.; Ikpo,
C. O.; Adedapo1, A. E.; Olanrewaju1, I. O. Arch. Pharm. Chem.
Life Sci. 2016, 349, 475.
9. (a) Marinescu, M.; Tudorache, D. G.; Marton, G. I.; Zalaru, C. M.;
Popa, M.; Chifiriuc, M. C.; Stavarache, C. E.; Constantinescu, C. J.
Mol. Struct. 2017, 1130, 463; (b) Zhang, L.; Addla, D.;
Jeyakkumar, P.; Wang, A.; Xie, D.; Wang, Y. N.; Zhang, S. L.;
Geng, R. X.; Cai, G. X.; Li, S.; Zhou, C. H. Eur. J. Med. Chem.
2016, 111, 160; (c) Liu, H. B.; Gao, W. W.; Tangadanchu, V. K.
R.; Zhou, C. H.; Geng, R. X. Eur. J. Med. Chem. 2018, 143, 66.
10. (a) AlNeyadi, S. S.; Salem, A. A.; Ghattas, M. A.; Atatreh, N.;
Abdou, I. M. Eur. J. Med. Chem. 2017, 136, 270; (b) Fang, X. J.;
Jeyakkumar, P.; Avula, S. R.; Zhou, Q.; Zhou, C. H. Bioorg. Med.
Chem. Lett. 2016, 26, 2584; (c) Jeyakkumar, P.; Liu, H. B.;
Gopala, L.; Cheng, Y.; Peng, X. M.; Geng, R. X.; Zhou, C. H.
Bioorg. Med. Chem. Lett. 2017, 27, 1737.
11. (a) Saadeh, H. A.; Mubarak, M. S. Curr. Top. Med. Chem. 2017,
17, 895; (b) Zhang, H. Z.; He, S. C.; Peng, Y. J.; Zhang, H. J.;
Gopala, L.; Tangadanchu, V. K. R.; Gan, L. L.; Zhou, C. H. Eur. J.
Med. Chem. 2017, 136, 165; (c) Jeyakkumar, P.; Zhang, L.; Avula,
S. R.; Zhou, C. H. Eur. J. Med. Chem. 2016, 122, 205.
12. (a) Shi, Y.; Park, J.; Lagisetti, C.; Zhou, W.; Sambucetti, L. C.;
Webb, T. R. Bioorg. Med. Chem. Lett. 2017, 27, 406; (b) Huang,
K. X.; Xie, M. S.; Zhao, G. F.; Qu, G. R.; Guo, H. M. Adv. Synth.
Catal. 2016, 358, 1.
13. (a) Gong, H. H.; Addla, D.; Lv, J. S.; Zhou, C. H. Curr. Top. Med.
Chem. 2016, 16, 3303; (b) Peng, X. M.; Peng, L. P.; Li, S.; Avula,
S. R.; Kannekanti, V. K.; Zhang, S. L.; Tam, K. Y.; Zhou, C. H.
Future Med. Chem. 2016, 8, 1927; (c) Duan, J. R.; Liu, H. B.;

Graphical Abstract
Novel purine benzimidazoles as safely
antimicrobial agents by regulating ROS
generation and targeting clinically resistant
Staphylococcus aureus DNA groove
Ya-Nan Wang, Rammohan R. Yadav Bheemanaboina,a
Gui-Xin Cai* and Cheng-He Zhou*