Design, synthesis and evaluation of a series of 5-methoxy-2,3-naphthalimide derivatives as AcrB inhibitors for the reversal of bacterial resistance

Article abstract of DOI:10.1016/j.bmcl.2019.02.003

A series of novel 5-methoxy-2,3-naphthalimide derivatives were designed, synthesized and evaluated for their biological activities. In particular, the ability of the compounds to synergize with antimicrobials, to inhibit Nile Red efflux, and to target AcrB was assayed. The results showed that the most of the tested compounds more sensitized the Escherichia coli BW25113 to the antibiotics than the parent compounds 7c and 15, which were able to inhibit Nile Red efflux. Significantly, compound A5 possessed the most potent antibacterial synergizing activity in combination with levofloxacin by 4 times and 16 times at the concentration of 8 and 16 μg/mL, respectively, whilst A5 could effectively abolish Nile Red efflux at 100 μM. Additionally, target effect of A5 was confirmed in the outer- or inner membrane permeabilization assays. Therefore, A5 is an excellent lead compound for further structural optimization.

Full text of DOI:10.1016/j.bmcl.2019.02.003

Accepted Manuscript
Design, synthesis and evaluation of a series of 5-methoxy-2,3-naphthalimide
derivatives as AcrB inhibitors for the reversal of bacterial resistance
Chaobin Jin, Rawaf Alenazy, Yinhu Wang, Rumana Mowla, Yinhui Qin, Jin
Quan Eugene Tan, Natansh Deepak Modi, Xinjie Gu, Steven W Polyak,
Henrietta Venter, Shutao Ma
PII:
S0960-894X(19)30067-8
https://doi.org/10.1016/j.bmcl.2019.02.003
BMCL 26283
DOI:
Reference:
Bioorganic & Medicinal Chemistry Letters
To appear in:
Received Date:
Revised Date:
Accepted Date:
10 December 2018
29 January 2019
2 February 2019
Please cite this article as: Jin, C., Alenazy, R., Wang, Y., Mowla, R., Qin, Y., Quan Eugene Tan, J., Deepak Modi,
N., Gu, X., Polyak, S.W., Venter, H., Ma, S., Design, synthesis and evaluation of a series of 5-methoxy-2,3-
naphthalimide derivatives as AcrB inhibitors for the reversal of bacterial resistance, Bioorganic & Medicinal
Chemistry Letters (2019), doi: https://doi.org/10.1016/j.bmcl.2019.02.003
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.

Design, synthesis and evaluation of a series of 5-methoxy-2,3-naphthalimide derivatives
as AcrB inhibitors for the reversal of bacterial resistance
a
b
a
b
Chaobin Jin a,1, Rawaf Alenazyb,1, Yinhu Wang , Rumana Mowla , Yinhui Qin , Jin Quan Eugene Tan ,
Natansh Deepak Modi b, Xinjie Gu a, Steven W Polyak b, Henrietta Venter b,*, Shutao Ma a,*
a Department of Medicinal Chemistry, Key Laboratory of Chemical Biology (Ministry of Education), School of
Pharmaceutical Sciences, Shandong University, 44 West Culture Road, Jinan 250012, China
b School of Pharmacy and Medical Sciences, University of South Australia, SA 5000, Australia
Running title: 5-methoxy-2,3-naphthalimides as efflux pump inhibitors
* Corresponding authors.
E-mail addresses: rietie.venter@unisa.edu.au (H. Venter), mashutao@sdu.edu.cn (S. Ma).
1 These authors contributed equally.
1

ABSTRACT
A series of novel 5-methoxy-2,3-naphthalimide derivatives were designed, synthesized and evaluated for their
biological activities. In particular, the ability of the compounds to synergize with antimicrobials, to inhibit Nile
Red efflux, and to target AcrB was assayed. The results showed that the most of the tested compounds more
sensitized the Escherichia coli BW25113 to the antibiotics than the parent compounds 7c and 15, which were
able to inhibit Nile Red efflux. Significantly, compound A5 possessed the most potent antibacterial synergizing
activity in combination with levofloxacin by 4 times and 16 times at the concentration of 8 and 16 µg/mL,
respectively, whilst A5 could effectively abolish Nile Red efflux at 100 M. Additionally, target effect of A5
was confirmed in the outer- or inner membrane permeabilization assays. Therefore, A5 is an excellent lead
compound for further structural optimization.
Keywords: 5-Methoxy-2,3-naphthalimide, Antimicrobial resistance, Efflux pump inhibitor, AcrB, Synergism,
Inhibition of efflux
2

The discovery of antibiotics has greatly guaranteed the survival of human beings and improved human life.
However, with the widespread use and abuse of antibiotics, multi-drug resistant bacteria have emerged across the
world, which has threatened human life and health seriously.1,2 Up to date a lot of new antibacterial drugs2-5 and
novel therapeutic strategies1,6-10 have been developed to solve the problem of the bacterial resistance. In
particular, over-expression of efflux pump proteins on the membrane of bacteria is one of the major reasons for
bacterial multidrug resistance.6 Among them, the AcrAB-TolC efflux pump widely exists in Gram-negative
bacteria, which are characterized by strong conservation and wide array of efflux substrates.11 Therefore, an
effective way to solve the multidrug resistance of Gram-negative bacteria is through the discovery of the efflux
pump inhibitors targeting the AcrB transporter.12
In the past decades, multiple classes of AcrB efflux pump inhibitors have been reported. For example, the
peptidomimetic phenylalanylarginine-β-naphthylamide (PaβN)13 is the first discovered efflux pump inhibitor
with broad-spectrum activity, but its toxicity prevents it from entering clinical trials.14 Naphthylpiperazines
(NMP) can increase the antibacterial activity of several antibiotics and has been one of the most potent efflux
pump inhibitors until now. However, its “serotonin agonist’’ property seems to be too toxic to be used in clinical
practice.15,16 Another class of the efflux pump inhibitors such as the quinoline derivatives show full activity only
at high concentrations but their toxicity is unknown. Another promising class of efflux pump inhibitors are the 5-
arylidenehydantoin derivatives.17 Among them, compound 1517 is the best representative, which increases the
antibacterial activity of sparfloxacin by 20 fold at 23 µg/mL.
Our group has reported that 4-substituted-2-napthamide derivatives17,18 inhibit AcrB efflux activity, which
are designed and synthesized based on the structure of natural product shikonin.20 Among them, the
representative compound 7c19 can increase the antibacterial activity of erythromycin by 8 times at 128 µg/mL.
Structural studies upon the efflux pump of Escherichia coli have revealed a binding site in AcrB subunit that can
accommodate structurally unrelated compounds.21,22 This is achieved primarily through a hydrophobic pocket
that is lined with five phenylalanine residues at the positions 136, 178, 610, 615 and 628, which provide
opportunities for multiple hydrophobic and π-π stacking interactions. Molecular docking studies upon the 2-
naphthamide inhibitor with the crystal structure of AcrB highlighted a key π-π stacking interaction between the
napthamide aromatic ring and the benzyl side chain of Phe626.18 These structural data, in combination with our
previous findings, suggest that a novel 2,3-naphthalimide skeleton is designed by fusing the imide ring of 15 to
the strong aromatic bicycles from 4-substituted-2-naphthamide 7c (Fig. 1). This new scaffold can facilitate
hydrophobic alkyl and arylalkyl groups to be appended onto the imide heterocycle, which may exploit the
hydrophobic binding sites of AcrB and yield novel and potent inhibitors. In this study, a series of the 5-methoxy-
2,3-naphthalimide derivatives were designed and synthesised as efflux pump inhibitors against antibiotic-
resistant Gram-negative E. coli. Those compounds were tested for their antibacterial activity, synergism with
antimicrobials and inhibition of Nile red efflux. Off-target effects such as membrane permeabilization were also
investigated.
3

O
OH O OH
N
H
O
OH O
Compound 7c
Shikonin
4-fold MIC-decrease of erythromycin at 128 g/mL
(Eur J Med Chem. 2018;143:699-709)
O
O
N R
O
OCH3
O
N
H3CO
OH
H
N
NH
O
Compound 15
4-fold MIC-decrease of chloramphenicol and nalidixic acid
at 23 mg/mL (Bioorg Med Chem. 2013;21:135-145)
Fig. 1. The design of the 5-methoxy-2,3-naphthalimide derivatives.
The 5-methoxy-2,3-naphthalimide derivatives (A1~A42) were synthesized from the commercially available
2,3-dimethylphenol (1) as outlined in Scheme 1. Methylation of 2,3-dimethylphenol (1) in acetone using methyl
iodide (CH3I) at 50 °C gave 1-methoxy-2,3-dimethylbenzene (2) in an almost quantitative yield. Bromonation of
2 with N-bromosuccinimide (NBS) in tetrachloromethane (C4Cl4) in the presence of 2,2-azobis(2-
methylpropionitrile) (AIBN) under reflux conditions produced 1,2-bis(dibromomethyl)-3-methoxybenzene (3) in
a good yield. Cyclization of 3 with maleic anhydride in the presence of potassium iodide (KI) (4.83g, 30 mmol)
in N,N-dimethylformamide (DMF) underwent the Diels-Alder reaction at 70 °C to produce 5-
methoxynaphthalene-2,3-dicarboxylic acid (4) in a good yield. Finally, dehydration of 4 in acetic anhydride
(Ac2O) under reflux generated the 5-methoxynaphtho[2,3-c]furan-1,3-dione (5), which was followed by reaction
with the corresponding amines in pyridine to provide forty-two 5-methoxy-2,3-naphthalimide derivatives
1
(A1~A42). The newly synthesized compounds were characterized by MS, H NMR and 13C NMR. All the
spectral data were in agreement with the proposed structures.
OH
O
2
O
O
O
O
O
O
CHBr2
CHBr2
COOH
COOH
a
c
d
e
b
N
R
O
O
O
5
1
3
4
A1~A42
A1 R = benzyl
A15 R = 2-methoxyphenyl
A16 R = 3-methoxyphenyl
A17 R = 4-methoxyphenyl
A18 R = 4-(tert-butyl)phenyl
A19 R = 2-chlorophenyl
A20 R = 2-isopropylphenyl
A21 R = 2-ethylphenyl
A29 R = 2-phenethyl
A30 R = 4-fluorophenethyl
A2 R = 2-chlorobenzyl
A3 R = 4-chlorobenzyl
A4 R = 3-fluorobenzyl
A5 R = 2-methoxybenzyl
A6 R = 4-methoxybenzyl
A7 R = 4-methylbenzyl
A8 R = 4-fluorobenzyl
A9 R = phenyl
A31 R = 3-methoxyphenethyl
A32 R = 2-chlorophenethyl
A33 R = 4-methoxyphenethyl
A34 R = cyclopentyl
A35 R = cyclohexyl
A36 R = hexyl
A22 R = 4-isopropylphenyl
A23 R = 4-butylphenyl
A37 R = pentyl
A10 R = 2-fluorophenyl
A11 R = 4-fluorophenyl
A12 R = 3-chlorophenyl
A13 R = o-tolyl
A24 R = 2-(trifluoromethyl)phenyl
A25 R = 4-(trifluoromethyl)phenyl
A26 R = 2-(trifluoromethoxy)phenyl
A27 R = 4-(trifluoromethoxy)phenyl
A28 R = 4-ethylphenyl
A38 R = tert-butyl
A39 R = butyl
A40 R = isopropyl
A41 R = ethyl
A14 R = m-tolyl
A42 R = methyl
Scheme 1. Reagents and conditions: (a) CH3I, acetone, 50°C, 24h (crude product); (b) NBS, AIBN, CCl4, reflux,
24h, 72% for two-step yeild; (c) Maleic anhydride, KI, DMF, 70°C, 8h, 61%; (d) Ac2O, reflux, 2h; (e)
Corresponding amine, pyridine, reflux, 12h.
4

The ability of the target compounds to bind to AcrB and reverse antimicrobial resistance was investigated by
assessing reversal of resistance, inhibition of efflux and target specific activity.23 However, only thirty of the
target compounds were characterised because compounds A3, A7, A8, A14~A16, A22~A24, A28, A30 and A32
had limited solubility in dimethyl sulphoxide (DMSO).
The antibacterial activity of the solubilized compounds was assessed by determining their minimum
inhibitory concentrations (MICs) against the wild-type resistant strain of E. coli BW25113. As a result, none of
the tested compounds showed any antimicrobial activity at the highest test concentration (128 mg/L) against E.
coli BW2113 nor an isotypic strain lacking the AcrB efflux pump (results not shown).
The target compounds were then tested for their ability to reverse AcrB-mediated resistance. This was
performed using an antimicrobial susceptibility assay in the presence of varying concentrations of target
compounds together with a panel of antimicrobials using the checkerboard titration assay. The antimicrobials
tested here were chloramphenicol, erythromycin, tetraphenylphosphonium and levofloxacin.24 The data
presented in Table 1 shows the activity of the twelve compounds that enhance the sensitivity of E. coli BW25113
to erythromycin, chloramphenicol, tetraphenyl phosphonium chloride (TPP) and/or levofloxacin with an
antibacterial sensitizing activity equivalent to or better than that of the lead compounds 7c19 and 15.17 The most
potent activity was observed with A5 and levofloxacin, which lowered the levofloxacin MIC by 16-fold (Table
1). It was noteworthy that benzyl substituted compound A21 and alkyl substituted compounds A36 and A37
displayed broad synergistic activity with all the four antibacterial agents.
5

Table 1. The synergetic effect of 5-methoxy-2,3-naphthalimide derivatives with different antibiotics and TPP against the wild-type drug-resistant strain of E. coli BW25113,
overexpressing AcrB indicated at (+AcrB). The MIC values of antimicrobials against the same E. coli strain with a deletion of the acrB gene are presented for comparison
and are indicated as (-AcrB).
O
O
N
R
O
MIC (µg/mL)
ERY TPP
64
MIC (µg/mL)
concentration
concentration
compound
compound
(µg/mL)
0
(µg/mL)
0
CAM
16
2
LEV
0.063
0.0039
0.063
0.063
0.063
0.063
0.063
0.063
0.063
0.063
0.063
0.063
0.016
0.0039
0.0039
0.0019
0.0019
CAM
16
2
ERY
TPP
1024
32
LEV
0.063
0.0039
0.063
0.063
0.063
0.063
0.063
0.063
0.063
0.063
0.063
0.0313
0.063
0.063
0.063
0.0313
0.0313
NONE (+AcrB)
NONE (-AcrB)
1024
32
NONE (+AcrB)
NONE (-AcrB)
64
4
0
4
0
8
16
16
16
16
8
32
32
32
32
16
64
64
64
64
32
64
64
64
64
64
1024
1024
1024
1024
512
8
16
16
16
16
16
16
16
16
8
64
64
32
16
8
1024
1024
1024
1024
1024
1024
1024
512
16
32
64
128
8
16
32
64
128
8
A1
A2
A5
A34
A36
A37
16
16
16
16
16
16
16
16
16
16
1024
1024
1024
1024
512
64
32
32
16
8
16
32
64
128
8
16
32
64
128
8
Cl
512
8
128
1024
1024
1024
1024
512
16
16
8
64
64
32
32
8
1024
1024
1024
1024
512
16
32
64
128
16
32
64
128
H3CO
8
4
6

8
16
32
64
128
8
16
16
16
16
16
16
16
16
16
16
16
16
16
16
8
64
32
32
32
32
64
64
64
64
16
64
64
64
64
32
1024
1024
1024
1024
1024
1024
1024
1024
1024
512
0.063
0.063
0.063
0.063
0.063
0.063
0.063
0.063
0.063
0.063
0.063
0.063
0.0313
0.0313
0.0313
8
16
32
64
128
8
16
16
16
16
16
8
64
64
64
32
32
32
32
32
32
32
64
64
64
32
32
1024
1024
1024
1024
512
0.063
0.063
0.063
0.063
0.063
0.063
0.063
0.063
0.0313
0.0313
0.063
0.063
0.0313
0.0313
0.0313
CH3
CH3
CH3
OCH3
A6
A19
A21
A38
A40
A42
1024
1024
1024
1024
512
16
32
64
128
8
16
32
64
128
8
8
Cl
CH3
CH3
8
8
8
1024
1024
1024
1024
512
16
16
16
16
16
1024
1024
1024
1024
1024
16
32
64
128
16
32
64
128
H3C
CH3
ERY = Erythromycin; CAM = Chloramphenicol; TPP = Tetraphenyl phosphonium chloride; LEV= levofloxacin
7

The ability of the target compounds identified above to inhibit pump-mediated efflux was determined by
measuring the inhibition of the efflux of the fluorescent substrate Nile Red.25,26 Nile Red is strongly fluorescent
in non-polar environments such as the cell membrane, but undergoes a significant decrease in fluorescent
quantum yield when in aqueous solutions.24,25 In the Nile Red efflux assay, A19, A21 and A36 were found to
abolish Nile Red efflux to the level of the pump-deleted (-)AcrB E. coli strain at 50 µM, while inhibition of Nile
Red efflux was observed at 100 M for A1, A5, A6 and A37. However, no efflux inhibition was observed for
A34 at 100 and 200 M was sufficient to completely abolish Nile Red efflux. A38, A40 and A42 seemed to have
a low affinity for AcrB as only partial inhibition of Nile Red efflux could be seen at 200 M, while 500 M of
those compounds were needed to completely abolish AcrB-mediated efflux activity (Fig. 2).
A1
A5
A6
A19
A36
A40
A21
A34
A37
A38
A42
8

Fig. 2. Inhibition of Nile Red efflux. Wild-type-resistant strains with active pump (thick black line, (+)AcrB),
the strains with a deletion of AcrB (grey line, (-)AcrB) or wild-type strains in the presence of the compounds
(dotted lines) were preloaded with Nile Red before the start of fluorescence measurements. Efflux was triggered
at 100 sec by the addition of 0.2% glucose (indicated by arrow).9 Representative fluorescent traces are shown for
experiments with different batches of cells done on three different days.
The outer membrane (OM) of Gram-negative organisms acts as a permeability barrier.27 Hence, the
compounds that permeabilize the OM, also synergize with antibiotics as well, and are independent of any
inhibition of efflux.28 The effect on the OM was determined by measuring the rate of nitrocefin hydrolysis when
incubated with E. coli that expresses β-lactamase.26 Nitrocefin is a chromogenic β-lactam antibiotic. Hydrolysis
of nitrocefin by the bacterial β-lactamase releases a colored compound that can be measured spectroscopically at
490 nm. The rate of nitrocefin hydrolysis is limited by the rate of diffusion across the outer membrane. Hence,
an increased rate of hydrolysis of nitrocefin is indicative of outer membrane permeabilization.29 The rate of
nitrocefin hydrolysis in the presence of the tested compounds was comparable to that of the untreated cells (Fig.
3) and much slower than the rate of nitrocefin hydrolysis of cells treated with polymixin B nonapeptide (PMBN),
with a well-described ability to permeabilise the Gram-negative OM.30,31 Thus, none of the compounds
compromised the OM in any way, which illustrated that the synergism of the above compounds with the
antimicrobials observed was not due to off-target effects such as outer membrane permeabilisation.
A1
A5
A6
A19
A21
A34
9

A36
A37
A38
A40
A42
Fig. 3. None of the test compounds permeabilize the outer membrane of E. coli. Nitrocefin was added to the
bacterial cells that received no compound (blue circles) or were treated with the outer membrane permeabilizer
polymyxin B nonapeptide (PMBN, red squares) as positive control or the test compounds (green triangles).
Nitrocefin hydrolysis by the periplasmic β-lactamase was observed as an increase in A490. Compounds that
permeabilize the outer membrane, such as PMBN, shows a fast initial rate of nitrocefin hydrolysis in comparison
with the untreated the bacterial cells.9 Representative traces are shown from experiments done in different days.
AcrB is an RND-type transporter that is powered by the proton motive force (pmf) across the bacterial inner
membrane.32 The effect of the active compounds on the integrity of the inner membrane was assessed to
distinguish between a direct inhibition of AcrB and the disruption of the proton motive force by permeabilisation
of the inner membrane. We previously used the potentiometric probe 3,3-diethyloxacarbocyanine iodide
(DiOC2(3)) to measure the magnitude and stability of  in bacterial cells and proteoliposomes33 and to
determine the effect of compounds on the integrity of the bacterial inner-membrane.18,19 Bacterial cells were
energized by the addition of glucose to establish a proton motive force (negative and basic inside the cell), which
would lead to an increase in fluorescence associated with aggregation of the DiOC2(3). Upon addition of the
ionophore CCCP, the pmf was dissipated and the fluorescence intensity dropped to the level before addition of
glucose. Of all the compounds tested, only A38 showed a marked decrease in the magnitude of the Δψ, while
A36 and A37 somehow interfered with the ability of CCCP to disrupt the pmf (Fig. 4). These results indicated
10

that, apart from A38, none of the compounds caused disruption of the bacterial inner membrane and hence the
synergism with antimicrobials observed was not due to an indirect effect of disrupting the pmf.
A1
A5
A6
A19
A36
A40
A21
A34
A37
A38
A42
11

Fig. 4. Compounds that effected a reduction in MIC of one or more antimicrobial compounds were subjected to
a bacterial inner membrane stability assay to exclude off-target effect such as disruption of the energy source
over the inner membrane. Bacterial suspensions were either left untreated (solid blue line) or exposed to
appropriate concentration of compounds (broken red line) for 10 min after which the potentiometric probe,
DiOC2(3) was added and the fluorescence monitored until it plateaued. Bacterial cells were then re-energized
with 0.5% glucose and the establishment of a membrane potential (inside negative) was measured as an increase
in fluorescence until it plateaued. The membrane potential was subsequently disrupted by the addition of the
proton ionophore CCCP (observed as a sharp drop in fluorescence intensity). After pre-treatment of the bacterial
cells with compounds, ∆Ψ was maintained for in the presence of all compounds apart from JCB-38, indicating
that these compounds do not act on the bacterial inner membrane.
In a 48 h tetrazole (MTT)-based assay, A5, A21 and A37 were found to show a very low cytotoxicity on
human cervical cancer (Hela) cells, with the IC50 values of >64 µg/mL. Obviously, their IC50 values were much
higher than the MIC value of A5 in synergism of antibacterial agents against the wild-type drug-resistant strain
of E. coli BW25113.
Docking analysis was performed to explore the binding mode of A5 to the AcrB (PDB code 5ENO). This
molecule was found to bind to the distal binding pocket of AcrB by multiple hydrophobic interactions (Fig. 5).
The methoxy group at the 8-position of the naphthalene ring produced hydrophobic interactions with Phe178 and
Val 612. The naphthalene ring interacted with Phe136 and Phe615 by π-π stacking, which is critical for the tight
binding. Additionally, the ortho-methoxybenzyl on the N atom of the imide ring was anchored to the
hydrophobic trap to stabilize their interactions (Fig. 6).
Fig. 5. Molecular docking of compound A5.
12

Fig. 6. Interactions of compound A5 with the amino acid residues of AcrB.
Preliminary structure-activity relationship (SAR) study indicated that appending both alkyl (A36, A37, A38,
A40 and A42) arylalkyl (A1, A2, A5, A6, A19 and A21) side chains into the N atom of the imide ring of the 2,3-
naphthalimide skeleton was an effective strategy to inhibit AcrB mediating efflux of antibacterials. The length of
the side chains on the N atom of the imide ring appeared to be important for the antibacterial sensitizing activity.
For instance, A30, A31 and A33 with longer linkers were all inactive. Besides, the position and the type of the
substituents on the terminal benzene of the side chains were critical to the synergistic activity. For example, the
ortho-substitution was better than the para-substitution (A5 vs A6), and meta-substitution (A19 vs A12 with no
activity) in the synergistic activity. In particular, the methoxy group was the optimal substituent for the
synergistic activity. And other electron-donating groups were also promising. Interesting, all fluorine-containing
compounds such as A4, A11, A24, A25, A26 and A27 were inactive in the antibacterial sensitizing activity,
probably due to the electronegativity of fluorine atom damaging the π-π stacking interaction. We also observed
activity with a variety of appended alkyl groups including cyclohexyl (A34), hexyl (A36), pentyl (A37), tertiary
butyl (A38), isopropyl (A40) and methyl (A42) groups. Together those findings highlight that AcrB can bind a
variety of structurally diverse structures, and the 5-methoxy-2,3-naphthalimide pharmacophore represents a
valuable scaffold for the chemical development of new efflux inhibitors to target multi-drug resistance in Gram
negative bacteria.
In conclusion, a series of novel 5-methoxy-2,3-naphthalimide derivatives were designed, synthesized and
evaluated for their inherent antibacterial activity, synergism with antimicrobials and inhibition of efflux-pump
driven substrate transport while off-target effects such as inner- or outer membrane permeabilization were also
ruled out. This study identified five 5-methoxy-2,3-naphthalimide derivatives that displayed synergism to at least
one antibacterial with potency equivalent to or better than the 4-fold synergism reported for parent compounds
7C (erythromycin at 128 µg/mL) and 15 (chloramphenicol and nalidixic acid at 23 µg/mL), namely A1, A5,
A34, A36 and A37. Eleven compounds were identified to be able to inhibit substrate efflux, among which, A19,
A21 and A36 exerted strong activity at 50 M, and A1, A5, A6 and A37 also showed activity at 100 M.
However, complete inhibition of substrate transport by A40 and A42 could only be achieved at relatively high
concentrations (500 M) indicating a lower affinity for the AcrB transporter compared to the other test
compounds. Hence, the modest effects on the MIC values of antimicrobials observed in the presence of the two
compounds were most probably due to competition. A36 was able to lower the MIC of all substrates tested at
128 g/mL, with the most potent activity against for erythromycin at 16 g/mL and were able to completely
abolish Nile Red efflux even at 50 M. A5 possessed the most potent antibacterial sensitizing activity for
levofloxacin by 4 and 16 times at 8 and 16 µg/mL, respectively. Only a handful of AcrB inhibitors reported so
13

far had the same synergistic antibacterial effect at concentrations as low as 8 and 16 µg/mL. Therefore, A5, A21,
A36 and A37 are excellent lead compounds for further structural optimization.
Conflict of interest
The authors declare that this study was carried out only with public funding. There is no funding or no
agreement with commercial for profit firms.
Acknowledgements
This research was supported financially by the National Natural Science Foundation of China (81673284),
Key research and development project of Shandong Province (2017CXGC1401), National Health and Medical
Research Council of Australia (GN1147538) and China–Australia Centre for Health Sciences Research. RA is
the recipient of a PhD scholarship from the Government of Saudi Arabia.
A. Supplementary material
Supplementary data associated with this article can be found, in the online version, at http://
References
1. Rakesh KP, Marichannegowda MH, Srivastava S, et al. Combating a master manipulator: Staphylococcus
aureus immunomodulatory molecules as targets for combinatorial drug discovery. ACS Comb Sci.
2018;20:681–693.
2. Zha G-F, Leng J, Darshini N, et al. Synthesis, SAR and molecular docking studies of benzo[d]thiazole-
hydrazones as potential antibacterial and antifungal agents. Bioorg Med Chem Lett. 2017;27:3148–3155.
3. Ravindar L, Bukhari SNA, Rakesh KP, et al. Aryl fluorosulfate analogues as potent antimicrobial agents:
SAR, cytotoxicity and docking studies. Bioorg Chem. 2018;81:107–118.
4. Chen X, Leng J, Rakesh KP, et al. Synthesis and molecular docking studies of xanthone attached amino
acids as potential antimicrobial and anti-inflammatory agents. MedChemComm. 2017;8:1706–1719.
5. Zha G-F, Wang S-M, Rakesh KP, et al. Discovery of novel arylethenesulfonyl fluorides as potential
candidates against methicillin-resistant of Staphylococcus aureus (MRSA) for overcoming multidrug
resistance of bacterial infections. Eur J Med Chem. 2019;162:364–377.
6. Rakesh KP, Vivek HK, Manukumar HM, et al. Promising bactericidal approach of dihydrazone analogues
against bio-film forming Gram-negative bacteria and molecular mechanistic studies. RSC Advances.
2018;8:5473–5483.
7. Mohammed YHE, Manukumar HM, Rakesh KP, et al. Vision for medicine: Staphylococcus aureus biofilm
war and unlocking key's for anti-biofilm drug development. Microb Pathog. 2018;123:339–347.
8. Zhang X, Manukumar HM, Rakesh KP, et al. Role of BP*C@AgNPs in Bap-dependent multicellular
behavior of clinically important methicillin-resistant Staphylococcus aureus (MRSA) biofilm adherence: A
key virulence study. Microb Pathog. 2018;123:275–284.
9. Zhang X, Marichannegowda MH, Rakesh KP, et al. Master mechanisms of Staphylococcus aureus:
consider its excellent protective mechanisms hindering vaccine development! Microbiol Res. 2018;212-
213:59–66.
10. Manukumar HM, Chandrasekhar B, Rakesh KP, et al. Novel T-C@AgNPs mediated biocidal mechanism
against biofilm associated methicillin-resistant Staphylococcus aureus (Bap-MRSA) 090, cytotoxicity and
its molecular docking studies. MedChemComm. 2017;8:2181–2194.
14

11. Schuster S, Kohler S, Buck A, et al. Random mutagenesis of the multidrug transporter AcrB from
Escherichia coli for identification of putative target residues of efflux pump inhibitors. Antimicrob Agents
Chemother. 2014;58:6870–6880.
12. Higgins CF. Multiple molecular mechanisms for multidrug resistance transporters. Nature 2007:446:749–
757.
13. Renau TE, Leger R, Flamme EM, et al. Inhibitors of efflux pumps in Pseudomonas aeruginosa potentiate
the activity of the fluoroquinolone antibacterial levofloxacin. J Med Chem. 1999;42:4928–4931.
14. Lomovskaya O, Zgurskaya HI, Totrov M, et al. Waltzing transporters and 'the dance macabre' between
humans and bacteria. Nat Rev Drug Discov. 2007;6:56–65.
15. Schumacher A, Steinke P, Bohnert JA, et al. Effect of 1-(1-naphthylmethyl)-piperazine, a novel putative
efflux pump inhibitor, on antimicrobial drug susceptibility in clinical isolates of Enterobacteriaceae other
than Escherichia coli. J Antimicrob Chemother. 2006, 57:344–348.
16. Pages JM, Amaral L. Mechanisms of drug efflux and strategies to combat them: challenging the efflux
pump of Gram-negative bacteria. Biochim Biophys Acta. 2009;1794:826–833.
17. Handzlik J, Szymanska E, Alibert S, et al. Search for new tools to combat Gram-negative resistant bacteria
among amine derivatives of 5-arylidenehydantoin. Bioorg Med Chem. 2013;21:135–145.
18. Wang Y, Mowla R, Guo L, et al. Evaluation of a series of 2-napthamide derivatives as inhibitors of the
drug efflux pump AcrB for the reversal of antimicrobial resistance. Bioorg Med Chem Lett. 2017;27:733–
739.
19. Wang Y, Mowla R, Ji S, et al. Design, synthesis and biological activity evaluation of novel 4-subtituted 2-
naphthamide derivatives as AcrB inhibitors. Eur J Med Chem. 2018;143:699–709.
20. Ohene-Ageyei T, Mowla R, Rahman T, et al. Phytochemicals increase the antibacterial activity of
antibiotics by acting on a drug efflux pump. MicrobiologyOpen 2014; 3:885–896.
21. Vargiu AV, Ramaswamy VK, Malloci G, et al. Computer simulations of the activity of RND efflux pumps.
Res Microbiol. 2018;169:384–392.
22. Ramaswamy VK, Cacciotto P, Malloci G, et al. Computational modelling of efflux pumps and their
inhibitors. Essays Biochem. 2017;61:141–156.
23. Venter H, Mowla R, Ohene-Agyei T, et al. RND-type drug effux pumps from Gram-negative bacteria:
molecular mechanism and inhibition. Front. Microbiol. 2015;6:337.
24. Zloh M, Kaatz GW, Gibbons S. Inhibitors of multidrug resistance (MDR) have affinity for MDR
substrates. Bioorg Med Chem Lett. 2004;14, 881–885.
25. Bohnert JA, Karamian B, Nikaido H. Optimized Nile Red efflux assay of AcrAB-TolC multidrug efflux
system shows competition between substrates. Antimicrob Agents Chemother. 2010;54:3770–3775.
26. Kinana AD, Vargiu AV, May T, et al. Aminoacyl β-naphthylamides as substrates and modulators of AcrB
multidrug efflux pump. Proc Natl Acad Sci U S A. 2016;113:1405–1410.
27. Arzanlou M, Chai WC, Venter H. Intrinsic, adaptive and acquired antimicrobial resistance in Gram-
negative bacteria. Essays Biochem. 2017; 61:49–59.
28. Opperman TJ, Nguyen ST. Recent advances toward a molecular mechanism of efflux pump inhibition.
Front microbiol. 2015;6:421.
29. Lomovskaya O, Warren MS, Lee A, et al. Identification and characterization of inhibitors of multidrug
resistance efflux pumps in Pseudomonas aeruginosa: novel agents for combination therapy. Antimicrob
Agents Chemother. 2001;45:105–116.
30. Mamelli L, Petit S, Chevalier J, et al. New antibiotic molecules: bypassing the membrane barrier of gram
15

negative bacteria increases the activity of peptide deformylase inhibitors. PLoS One. 2009;4:e6443.
31. Allam A, Maigre L, Vergalli J, et al. Microspectrofluorimetry to dissect the permeation of ceftazidime in
Gram-negative bacteria. Sci. Rep. 2017;7:986.
32. Amaral L, Martins A, Spengler G, et al. Efflux pumps of Gram-negative bacteria: what they do, how they
do it, with what and how to deal with them. Front. pharmacol. 2014;4:168.
33. Venter H, Shilling RA, Velamakanni S, et al. Investigating the discernible and distinct effects of platinum-
based chemotherapy regimens for metastatic triple-negative breast cancer on time to progression. Nature
2003;426:866-870.
Graphical Abstract:
Design, synthesis and evaluation of a series of 5-methoxy-2,3-naphthalimide derivatives as AcrB inhibitors
for the reversal of bacterial resistance
Chaobin Jin a,1, Rawaf Alenazyb,1, Yinhu Wang a, Rumana Mowla b, Yinhui Qin a, Jin Quan Eugene Tan b,
Natansh Deepak Modi b, Xinjie Gu a, Steven W Polyak b, Henrietta Venter b,*, Shutao Ma a,*
a Department of Medicinal Chemistry, Key Laboratory of Chemical Biology (Ministry of Education), School of
Pharmaceutical Sciences, Shandong University, 44 West Culture Road, Jinan 250012, China
b School of Pharmacy and Medical Sciences, University of South Australia, SA 5000, Australia
Research Highlights
16

˃ 5-Methoxy-2,3-naphthalimide derivatives were designed and synthesized. ˃ They were evaluated for their
AcrB inhibitory activity and target specific activity. ˃ Twelve derivatives could enhance the sensitivity of
antibiotics to strains. ˃ A5 showed the most potent antibacterial sensitizing activity.
17