Amphiphilic nebramine-based hybrids Rescue legacy antibiotics from intrinsic resistance in multidrug-resistant Gram-negative bacilli

Article abstract of DOI:10.1016/j.ejmech.2019.05.003

The inability to discover novel class of antibacterial agents, especially against Gram-negative bacteria (GNB), compel us to consider a broader non-conventional approach to treat infections caused by multidrug-resistant (MDR)bacteria. One such approach is

Full text of DOI:10.1016/j.ejmech.2019.05.003

Accepted Manuscript
Amphiphilic nebramine-based hybrids Rescue legacy antibiotics from intrinsic
resistance in multidrug-resistant Gram-negative bacilli
Xuan Yang, Derek Ammeter, Temilolu Idowu, Ronald Domalaon, Marc Brizuela,
Oreofe Okunnu, Liting Bi, Yanelis Acebo Guerrero, George G. Zhanel, Ayush Kumar,
Frank Schweizer
PII:
S0223-5234(19)30406-4
DOI:
https://doi.org/10.1016/j.ejmech.2019.05.003
EJMECH 11313
Reference:
To appear in: European Journal of Medicinal Chemistry
Received Date: 12 November 2018
Revised Date: 24 April 2019
Accepted Date: 1 May 2019
Please cite this article as: X. Yang, D. Ammeter, T. Idowu, R. Domalaon, M. Brizuela, O. Okunnu, L. Bi,
Y.A. Guerrero, G.G. Zhanel, A. Kumar, F. Schweizer, Amphiphilic nebramine-based hybrids Rescue
legacy antibiotics from intrinsic resistance in multidrug-resistant Gram-negative bacilli, European Journal
of Medicinal Chemistry (2019), doi: https://doi.org/10.1016/j.ejmech.2019.05.003.
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.

ACCEPTED MANUSCRIPT

ACCEPTED MANUSCRIPT
1
Amphiphilic Nebramine-based Hybrids Rescue Legacy Antibiotics
from Intrinsic Resistance in Multidrug-resistant Gram-negative
Bacilli
a
a
a
a
a
Xuan Yang , Derek Ammeter , Temilolu Idowu , Ronald Domalaon , Marc Brizuela , Oreofe
a
a
b
c
b
Okunnu , Liting Bi , Yanelis Acebo Guerrero , George G. Zhanel , Ayush Kumar , Frank
a, b, *
Schweizer
a
Department of Chemistry, University of Manitoba, Winnipeg, MB, R3T 2N2, Canada.
b
Department of Microbiology, University of Manitoba, Winnipeg, MB R3T 2N2, Canada.
c
Department of Medical Microbiology, University of Manitoba, Winnipeg, MB R3T 1R9,
Canada.

ACCEPTED MANUSCRIPT
2
Abstract
The inability to discover novel class of antibacterial agents, especially against Gram-
negative bacteria (GNB), compel us to consider a broader non-conventional approach to treat
infections caused by multidrug-resistant (MDR) bacteria. One such approach is the use of
adjuvants capable of revitalizing the activity of current existing antibiotics from resistant
pathogens. Recently, our group reported a series of tobramycin (TOB)-based hybrid adjuvants
that were able to potentiate multiple classes of legacy antibiotics against various MDR GNB.
Herein, we report the modification of TOB-based hybrid adjuvants by replacing TOB domain by
the pseudo-disaccharide nebramine (NEB) through selective cleavage of the α-D-glucopyranosyl
linkage of TOB. Potent synergism was found for combinations of NEB-based hybrid adjuvants
with multiple classes of legacy antibiotics including fluoroquinolones (moxifloxacin and
ciprofloxacin), tetracyclines (minocycline), or rifamycin (rifampicin) against both wild-type and
MDR P. aeruginosa clinical isolates. We also demonstrated that a combination of the optimized
NEB-CIP hybrid 1b and rifampicin protects Galleria mellonella larvae from the lethal effects of
extensively drug-resistant (XDR) P. aeruginosa. Mechanistic evaluation of NEB-based hybrid
adjuvants revealed that the hybrids affect the outer- and inner membranes of wild-type P.
aeruginosa PAO1. This study describes an approach to optimize aminoglycoside-based hybrids
to yield lead adjuvant candidates that are able to resuscitate the activity of partner antibiotics
against MDR GNB.
Keywords: Nebramine; Hybrids; Adjuvants; Synergy; Antibacterial; Multi-drug resistant
Gram-negative baccilli

ACCEPTED MANUSCRIPT
3
1
. Introduction
Starting from Fleming’s discovery of penicillin in 1929 [1], a large number of antibiotics
have been discovered, developed, and marketed. Antibiotics have saved countless lives and
played a key role in the advancement of medical science for the past 70 years [2,3]. However,
rampant and indiscriminate use of antibiotics has escalated the prevalence of multidrug-resistant
(MDR) bacterial infections, especially those that are caused by Gram-negative pathogens. Worse
still, there is a steady decline in the discovery of novel drug classes able to eradicate MDR
Gram-negative pathogens which is largely due to the lack of understanding of the
physicochemical properties necessary for antibacterial agents to efficiently traverse and
accumulate inside Gram-negative bacterial cell [4,5]. There is an urgent need to find novel and
perhaps unconventional approaches to address bacterial infection. Co-administration of helper
molecules called adjuvants capable of enhancing the activity of currently used antibiotics and
extend the life of legacy antibiotics is a viable strategy to overcome antimicrobial resistance [6,7].
Our group recently has demonstrated that amphiphilic TOB-based conjugates were able
to revive the antibacterial activity of multiple classes of antibiotics against MDR Gram-negative
bacilli (GNB), especially against P. aeruginosa [7–14]. For example, we first reported TOB-
ciprofloxacin hybrids with poor intrinsic antibacterial activity were able to restore the activity of
fluoroquinolone antibiotics against ciprofloxacin-resistant MDR or XDR (extensively drug-
resistant) Pseudomonas aeruginosa in combination therapy [8]. Structure-activity studies
revealed that the presence of both TOB and ciprofloxacin pharmacophores tethered by a 12-
carbon-long (C ) aliphatic linker is critical to the potentiation of fluoroquinolone antibiotics [8].
1
2
Encouraged by these results, an unconventional structure-activity relationship study was pursued
by replacing the ciprofloxacin fragment of TOB-ciprofloxacin hybrid by other pharmacophores.

ACCEPTED MANUSCRIPT
4
Since then, we have developed a series of TOB-moxifloxacin hybrids [9], TOB-efflux pump
inhibitors conjugates [10,11], TOB-lysine peptoid conjugates [12,13], as well as TOB-polymyxin
B hybrids [14]. Biological evaluations revealed that these TOB-based conjugates retained the
3
adjuvant properties of TOB-ciprofloxacin hybrids to a variable extent. These results suggest that
the TOB fragment linked to C tether is the core scaffold that is responsible for the adjuvant
12
properties. Mechanistic studies revealed that these TOB-based conjugates permeabilize the outer
membrane and dissipate the proton motive force (PMF) located in the cytoplasmic membrane of
P. aeruginosa [9,10,12].
TOB is known to eradicate Gram-negative bacteria by disruption of the outer membrane
at higher concentrations (≥ 8 µg/mL) [15]. However, at lower concentrations (< 4 µg/mL), TOB,
selectively interacts with the 16S rRNA, thereby causing inhibition of bacterial protein
translation [15,16]. The TOB’s pseudo-disaccharide segment (ring I and ring II), namely NEB
(NEB), is the essential pharmacophore responsible for most of the specificity of the interactions
with the ribosome [17–21]. In addition, it was recently reported that an amphiphilic NEB
derivative displayed potent activity against certain TOB-resistant Gram-negative bacteria
suggesting that amphiphilic NEB analogs possess a different mode of action than TOB [22]. In
addition, neamine-based and neosamine-based amphiphiles have been reported to possess potent
antipseudomonal properties by interacting with the outer membrane of P. aeruginosa [23–25].
To understand the effect of TOB on the overall adjuvant activity of previously reported
TOB-based conjugates, we decided to replace TOB by NEB and evaluate its microbiological
activity. We questioned whether the modification of the TOB domain would retain the adjuvant
properties of our reported conjugates. To accomplish this aim, we selected TOB-moxifloxacin
and TOB-ciprofloxacin as lead compounds and replaced TOB by NEB, while keeping the

ACCEPTED MANUSCRIPT
5
moxifloxacin (MOX) and ciprofloxacin (CIP) fragments and the C12 hydrocarbon tether as
shown in hybrid NEB-MOX (1a) and NEB-CIP (1b) (Fig. 1). In addition, we also prepared a
NEB-NMP (1-(1-napthylmethyl)-piperazine) hybrid 2 containing a slightly reduced C10
hydrocarbon tether to potentially reduce non-specific protein binding and investigated its
adjuvant properties. NMP is a well-known efflux pump inhibitor (EPI) of various efflux pumps
in Gram-negative bacteria except P. aeruginosa [26].
2
2
. Results and Discussion
.1 Chemistry
The preparation of NEB-based hybrids (1a, 1b, and 2) were done by selective
degradation of TOB-based hybrids as outlined in Scheme 1. Commercially available tobramycin
was transformed into the N-Boc-, and O-TBDMS-protected tobramycin 3 with the exception of
the sterically hindered C-5 alcohol, following previously reported procedures (Scheme 1)
[9,10,12,27]. Alkylation of 3 with 1,12-dibromododecane or 1,10-dibromodecane in the presence
of a phase transfer catalyst, tetrabutylammonium hydrogen sulfate (TBAHS), afforded
bromoalkylated tobramycin (4a and 4b). This bromide (4a) was then converted to primary
alcohol 5 followed by oxidation reaction using pyridinium chlorochromate (PCC) to generate
aldehyde 6 in good yield. Protected hybrid 7a and 7b were synthesized via reductive amination
between moxifloxacin methyl ester or ciprofloxacin methyl ester and aldehyde 6. Heating
hybrids 7a and 7b in aqueous HCl solution resulted in regioselective hydrolysis of the α- -
D
glucopyranosyl bond [21,22] along with the simultaneous removal of Boc and TBDMS
protecting groups. To avoid a laborious separation and purification of the obtained pseudo-

ACCEPTED MANUSCRIPT
6
disaccharides, NEB-MOX and NEB-CIP hybrids, the four free amino groups of NEB were
protected by (Boc) O to afford the corresponding N-Boc-protected NEB-MOX hybrid 8a and N-
2
Boc-protected NEB-CIP hybrid 8b that could easily be purified by flash chromatography. De-
esterification and subsequent global deprotection of the amino groups finally resulted in desired
the NEB-MOX (1a) and NEB-CIP (1b) compounds (Scheme 1). A related strategy was used to
synthesize NEB-NMP (2) (Scheme 1).
2
.2 Combination Study of Hybrids with Antibiotics
To determine whether the NEB-MOX hybrid 1a retain the adjuvant properties of
previously reported TOB-moxifloxacin hybrids, checkerboard studies were performed. Initially,
we assessed the combination of hybrid 1a with three different classes of clinically-used
antibiotics including the fluoroquinolone antibiotic moxifloxacin, the tetracycline antibiotic
minocycline, and the rifamycin antibiotic rifampicin against wild-type P. aeruginosa PAO1
(Table 1) by using the fractional inhibitory concentration index (FICI) as a measure of the
interaction between two agents. FICI of ≤ 0.5, > 0.5 to ≤ 4, and > 4 indicate synergy, no
interaction, and antagonism, respectively [28]. In accordance with previous findings against
wild-type P. aeruginosa PAO1 [8,10,12], NEB-MOX hybrid 1a displayed weak antibacterial
activity (MIC = 32 µg/mL) as a stand-alone agent. However, it was found to be synergistic (FICI
of 0.25) with the fluoroquinolone moxifloxacin (Table 1). Synergism was also observed with
minocycline (FICI of 0.38) as well as the outer membrane-impermeable antibiotic rifampicin
(
FICI of 0.07) (Table 1). The absolute MICs [the MIC of antibiotics in the presence of 8 µg/mL
7.5 µM) hybrid 1a] of three tested antibiotics, moxifloxacin, minocycline, or rifampicin, in
(
combination therapy with hybrid 1a were significantly reduced compared to monotherapy,

ACCEPTED MANUSCRIPT
7
especially for rifampicin (≥ 256-fold potentiation). It should be noted that the clinically-approved
β-lactamase inhibitor avibactam is typically administered at 15 µM concentration to potentiate
the cephalosporin ceftazidime in in vitro studies [29,30].
To validate our findings in wild-type P. aeruginosa strain, we performed the same
checkerboard study using a panel of eight MDR or XDR P. aeruginosa clinical isolates
(Supplementary Information Table S1) as previously studied for TOB-based hybrids [8,10–12].
Notably, among this panel of clinical isolates, two strains (P. aeruginosa 91433 and 101243) are
non-susceptible or resistant to colistin that is considered to be the antibiotic of last resort for the
treatment of carbapenem-resistant Gram-negative bacterial infections [31]. We determined the
FIC index of hybrid 1a in combination with moxifloxacin, ciprofloxacin, minocycline, or
rifampicin across the eight clinical isolates panel. Strong potentiation was seen with
moxifloxacin, ciprofloxacin, minocycline, or rifampicin (FIC indices of 0.004 to 0.28) against
these pathogens, with the exception of ciprofloxacin against P. aeruginosa 100036 and 101885
strains (FICI > 0.5) (Table 2, Table 3, Table 4).
Next, we evaluated the adjuvant potencies of hybrid 1a by comparing the absolute MICs
[in the presence of 8 µg/mL (7.5 µM) hybrid 1a] of the four antibiotics to their established
susceptibility breakpoints. According to the Clinical and Laboratory Standards Institute (CLSI),
the susceptible breakpoint, a chosen concentration (µg/mL) of an antibiotic which defines a
strain of bacteria whether it is susceptible to this antibiotic, of ciprofloxacin for P. aeruginosa is
1
µg/mL [32]. However, the established susceptibility breakpoints for the other three tested
antibiotics against P. aeruginosa are not available since they are unconventional antibiotics for
the treatment of P. aeruginosa infections. Therefore, the susceptibility breakpoints of
minocycline for Acinetobacter spp. (≤ 4 µg/mL) and rifampicin for Enterococcus spp. (≤1

ACCEPTED MANUSCRIPT
8
µg/mL) reported by CLSI were used as interpretative guidelines [32]. It is noteworthy that the
French Society for Microbiology has established a rifampicin breakpoint for Acinetobacter
baumannii based on MIC distributions (susceptible, ≤ 4 µg/mL; intermediate, 8–16 µg/mL; and
resistant, ≥16 µg/mL) [33]. In addition, we conservatively considered the susceptibility
breakpoint of moxifloxacin for P. aeruginosa to be similar to that of ciprofloxacin, as both
belong to the fluoroquinolone class of antibiotics.
For the two fluoroquinolones, combinations of NEB-MOX hybrid 1a with moxifloxacin
yielded stronger potentiation than ciprofloxacin against the panel of MDR/XDR P. aeruginosa
clinical isolates (Table 2). The adjuvant potency of hybrid 1a in combination with moxifloxacin
is comparable to previously reported TOB-ciprofloxacin hybrid [8]. In 37.5% of cases, both
hybrid 1a and TOB-ciprofloxacin hybrid 1b, at concentrations of ≤ 8 µg/mL (6.8 – 7.5 µM),
were able to bring down the MIC of moxifloxacin below its interpretative susceptibility
breakpoint (≤ 1 µg/mL) against moxifloxacin-resistant MDR/XDR P. aeruginosa isolates (Table
2
) [8]. However, the same susceptibility breakpoint was not reached for ciprofloxacin in
combination with hybrid 1a at concentration of 8 µg/mL (7.5 µM) against all the tested
ciprofloxacin-resistant MDR/XDR P. aeruginosa isolates (Table 2). In contrast, the MICs of
minocycline (8/8 minocycline-resistant MDR/XDR P. aeruginosa isolates) and rifampicin (7/8
rifampicin-resistant MDR/XDR P. aeruginosa isolates) were strongly reduced below their
susceptibility breakpoints in the presence of 8 µg/mL (7.5 µM) hybrid 1a, an effect that is
consistent with previously reported TOB-efflux pump inhibitor conjugates and TOB-lysine
peptoid conjugates (Table 3, Table 4) [10,12].
A summarized result of antibacterial activity of minocycline (MIN) and rifampicin (RIF)
alone or in combination with a fixed concentration of 8 µg/mL (7.5 µM) hybrid 1a against the

ACCEPTED MANUSCRIPT
9
panel of eight MDR/XDR P. aeruginosa clinical isolates is shown in Table 5. The MIC80 of
minocycline and rifampicin in combination with 8 µg/mL (7.5 µM) hybrid 1a were significantly
lower in comparison to the MIC of the antibiotic alone. Moreover, the absolute MIC of
8
0
80
minocycline (1 µg/mL) and rifampicin (0.13 µg/mL) were found to be less than their respective
CLSI susceptibility breakpoints. Similarly, we demonstrated strong synergy of NEB-CIP (1b)
with minocycline or rifampicin against wild-type and MDR P. aeruginosa strains (Table 6). For
instance, in presence of only 4 µg/mL (4.1 µM) of hybrid 1b, minocycline showed a 32-fold
potentiation while rifampicin resulted in a 128 − ≥256-fold potentiation in wild-type and MDR P.
aeruginosa strains (Table 6).
The observed potentiation of rifampicin by hybrid 1a and 1b may be explained by our
previous findings that demonstrated that amphiphilic TOB-based hybrid adjuvants perturb the
outer membrane of P. aeruginosa in a dose-dependent manner, thus facilitating the entry of
antibiotics that are unable to cross the outer membrane of Gram-negative bacteria, such as
rifampicin [8–10,12]. NEB-based hybrids seem to have a similar membrane effect as that of
TOB-based hybrids. Since rifampicin is a poor substrate for P. aeruginosa RND efflux pumps
[10,12], outer membrane perturbation is most likely to be the reason to explain the observed
strong synergistic effects of hybrid 1a and 1b in combination with rifampicin against P.
aeruginosa.
We also investigated synergy of NEB-NMP hybrid 2 with various antibiotic classes
including fluoroquinolones (moxifloxacin, ciprofloxacin), tetracyclines (minocycline), or
rifamycin (rifampicin) against wild-type and MDR P. aeruginosa strains (Table 7). More
importantly, NEB-NMP (2) reduced the MIC of minocycline below its CLSI susceptibility

ACCEPTED MANUSCRIPT
1
0
breakpoint (≤ 4 µg/mL) against all tested P. aeruginosa strains. The observed adjuvant property
of NEB-NMP (2) is consistent with that of reported TOB-NMP conjugate [10].
Besides P. aeruginosa, we also explored the synergistic effects of NEB-based hybrids (1a
and 1b) with minocycline or rifampicin against other MDR GNBs such as Acinetobacter
baumannii, Escherichia coli, Klebsiella pneumoniae, and Enterobacter cloacae (Tables 8 and 9).
Again, both NEB-MOX 1a and NEB-CIP 1b displayed poor antibacterial activity by themselves
against these pathogens (MICs of ≥ 8 µg/mL). In the case of A. baumannii, minocycline was not
potentiated against the four tested isolates while NEB-MOX 1a was able to synergize rifampicin,
leading to 32- to 64-fold reductions in MICs at a concentration of 8 µg/mL (7.5 µM) of the
adjuvant. Similarly, against A. baumannii, we observed an additive relationship of NEB-CIP (1b)
with minocycline while the combination of NEB-CIP (1b) with rifampicin remained synergistic
(
Table 9). With respect to K. pneumonia, E. cloacae, or E. coli, NEB-MOX (1a) and NEB-CIP
1b) displayed strong synergism with rifampicin against all isolates tested while synergism of
(
NEB-MOX (1a) and NEB-CIP (1b) with minocycline was only observed in few isolates (Tables
8
and 9).
2
.3 Time-kill Curve
To confirm the synergistic activity between NEB-based hybrids and minocycline or
rifampicin, time-kill assays were performed. We first studied the time killing kinetics of
minocycline at 4 µg/mL (½ × MIC) in combination with NEB-MOX 1a at sub-inhibitory
concentration (½ × MIC = 16 µg/mL or ¼ × MIC = 8 µg/mL) against P. aeruginosa wild-type
PAO1 (Fig. 2A). We set a fixed concentration of 4 µg/mL minocycline for the kinetic study since
the CLSI susceptibility breakpoint of minocycline is ≤ 4 µg/mL. It was demonstrated that

ACCEPTED MANUSCRIPT
11
combination of bacteriostatic minocycline (4 µg/mL) with ½ × MIC of hybrid 1a became
bactericidal and resulted in complete eradication of P. aeruginosa PAO1 over a 24 h time period.
This enhanced killing efficiency of minocycline in combination with hybrid 1a is consistent with
our previous findings for TOB-based hybrids [10,12] and are likely the results of the membrane
effects induced by hybrid 1a. Furthermore, we also studied the killing kinetics of NEB-CIP 1b in
combination with rifampicin against XDR P. aeruginosa PA259. A combination of sub-MIC of
1
b (1 µg/mL, MIC of 1b is >128 µg/mL) and rifampicin (1/16 × MIC = 1 µg/mL) yielded a 3-
order magnitude decrease in viable bacterial counts over 8 h time period (Fig. 2B). Complete
eradications were observed at a higher concentration of 1b (4 µg/mL) in combination with
rifampicin at 1 µg/mL or 4 µg/mL for only 4 h of antimicrobial exposure (Fig. 2B).
2
.4 Hemolytic Activity and In Vivo Efficacy Study
To investigate whether the adjuvant properties of the NEB-based hybrids translates into a
measurable in vivo effect, we selected the established in vivo Galleria mellonella larvae infection
model to study the efficacy of hybrid 1b-rifampicin combination therapies against P. aeruginosa
[
34,35]. Initially, we demonstrated that NEB-CIP 1b was non-hemolytic to pig erythrocytes
<10% at 512 µg/mL) (Fig. 3). We also examined the tolerability of 1b on G. mellonella and
(
found that the larvae survived beyond 96 h when administered with 100 mg/kg dosage of 1b (Fig.
). However, colistin resulted in 70% and 90% larvae deaths at the dosage of 75 mg/kg and 100
4
mg/kg, respectively, after 96 h, consistent with known toxicity of colistin to eukaryotic cells. We
also established that 5 CFU of XDR P. aeruginosa PA264 alone resulted in 100% lethality of the
larvae after 18 h. To assess the ability of combination therapy of 1b and rifampicin to protect
against XDR P. aeruginosa PA264-challenge larvae, single treatment doses of rifampicin +

ACCEPTED MANUSCRIPT
1
2
compound 1b (25 + 25 mg/kg, 50 + 50 mg/kg, and 75+75 mg/kg) were administered 2 h post
inoculation with 5 CFU XDR P. aeruginosa PA264 (bacterial isolate was only susceptible to
colistin). The results showed that monotherapy with a single dose of rifampicin (75 mg/kg) or 1b
(100 mg/kg) resulted in 10% and 0% survival of the larvae after 18 h, respectively (Fig. 5). In
contrast, combination of rifampicin and 1b improved the survival of the wax moth larvae in a
dose-dependent manner (Fig. 5). For instance, a single dose combination of rifampicin with 1b
(75 + 75 mg/kg) resulted in 87% and 53% survival after 18 h and 24 h respectively. A 50 + 50
mg/kg single dose combination of rifampicin and 1b resulted in 60% and 33% survival after 18 h
and 24 h, respectively, while a 25 + 25 mg/kg single dose combination of rifampicin and 1b
resulted in a 33% and 13% survival after 18 h and 24 h, respectively. This clearly demonstrates a
dose-dependent survivability of the infected larvae when treated with a combination of
rifampicin and compound 1b. It is interesting to note that a 25 + 25 mg/kg rifampicin + 1b was
more effective (13% survival) than 25 mg/kg of colistin (0% survival) after 24 h, although it is
unclear whether the inability of colistin to protect the larvae is due to lack of effectiveness or a
colistin-induced toxicity. Overall, these results demonstrate the therapeutic potential of NEB-
based hybrid 1b + rifampicin to treat MDR/XDR infections in vivo. We also assessed the
toxicity of adjuvants 1a or 1b against the HepG2 and HEK293 cell lines alone and in
combination with rifampicin. These results confirmed that adjuvants 1a and 1b do not possess
elevated toxicity at their synergistic concentration alone and in combination with Rifampicin.
(
Figure S1).

ACCEPTED MANUSCRIPT
1
3
2
.5 Tetracycline Uptake Assay
To gain insight to the synergistic mechanism of NEB-MOX hybrid 1a with minocycline,
fluorescence-based tetracycline uptake assay for Gram-negative bacteria [36] was performed to
investigate the effect of hybrid 1a on the uptake of tetracycline (Fig. 6). Our results indicate that,
similar to TOB-based hybrids [10], hybrid 1a enhances the uptake of tetracycline in P.
aeruginosa PAO1 in a concentration-dependent manner. Comparable enhancements in
tetracycline uptake were also observed with membrane-targeting antibiotic colistin (Fig. 6). We
previously reported that TOB-based hybrids not only permeabilize the Gram-negative bacterial
outer membrane but also depolarize the cytoplasmic membrane [8–10,12]. TOB-based hybrids
specifically dissipate the electrical component (∆Ψ) of the proton motive force (PMF) resulting
in a compensatory increase in transmembrane chemical component (∆pH) in order to counter this
effect and maintain ATP synthesis level. The effect of TOB-based hybrids on ∆Ψ is likely
retained in NEB-MOX hybrid 1 which is consistent with the observed synergy of hybrid 1a with
minocycline, as tetracycline uptake is ꢀpH-dependent [36]. Disruption of ∆Ψ by hybrid 1a is
compensated by an increase in ∆pH that in turn enhances the uptake of tetracycline antibiotics.
This was further corroborated by the observation that CCCP (carbonyl cyanide m-chlorophenyl
hydrazone) inhibits tetracycline accumulation (Fig. 6). CCCP is an uncoupler of oxidative
phosphorylation that disrupts the proton gradient (∆pH) of bacterial membranes [37]. Moreover,
minocycline is known to inhibit preferentially the biosynthesis of envelop proteins [38] which,
perhaps, elicits further compromise of the intrinsic resistance barrier (the Gram-negative
bacterial outer membrane), thereby augmenting the effects of NEB-based hybrid 1a.

ACCEPTED MANUSCRIPT
1
4
3
Conclusions
In this study, we demonstrated that NEB-based hybrids (1a, 1b, and 2) are capable of
potentiating multiple classes of antibiotics including fluoroquinolones (moxifloxacin,
ciprofloxacin), tetracycline (minocycline), and rifamycin (rifampicin) against wild-type and
MDR/XDR GNBs including P. aeruginosa, A. baumannii, K. pneumonia, and E. cloacae strains.
The adjuvant potencies of NEB-based hybrids (1a, 1b, and 2) are comparable to that of TOB-
based hybrids as studied before [7–14], suggesting that cleavage of the 3-deoxy-3-amino α-D-
glucosidic linkage in TOB to yield NEB did not significantly alter the adjuvant properties of this
scaffold. Mechanistic study of NEB-MOX 1a confirm that it also retains the membrane effects of
TOB-based hybrid adjuvants including TOB-fluoroquinolone hybrids. Modification of the
tobramycin domain of TOB-based hybrid suggests that the pseudo-disaccharide NEB linked to
C tether is the essential membrane active core responsible for the adjuvant properties. This
12
study provides further insight into structural optimization of previously investigated TOB-based
hybrid adjuvants. Moreover, the reduced number of basic functions in NEB when compared to
TOB may result in reduced aminoglycoside-induced cytotoxicity.
4
4
4
Experimental section
.1 Synthetic Chemistry
.1.1 General Comments
Reactions were monitored by thin-layer chromatography (TLC) on silica gel 60 F254
0.25 mm, Merck) and the compounds were visualized using ultraviolet light and/or stain with
(

ACCEPTED MANUSCRIPT
1
5
1
13
ninhydrin solution (ninhydrin and acetic acid in ethanol). 1D and 2D ( H, C, DEPT, COSY,
HSQC, HMBC) nuclear magnetic resonance (NMR) characterization experiments were
performed on either Bruker AMX-500 or Bruker AMX-300 spectrophotometer in the noted
deuterated solvents. Chemical shifts (δ) are reported in parts per million with CHCl (7.26 ppm),
3
DHO (4.79 ppm), and CD HOH (3.31 ppm) used as internal standards. NMR spectra were
2
analyzed using Mnova Software 8.0 version. Electrospray ionization (ESI) mass spectrometry
(MS) experiments were carried out on a Varian 500 MS ion trap mass spectrometer. Matrix-
assisted laser desorption ionization (MALDI) MS experiments were performed on a Bruker
Daltonics Ultraflex MALDI TOF/TOF mass spectrometer. Analytical high-performance liquid
chromatography (HPLC) was carried out on Breeze HPLC Waters with 2998 PDA detector (1.2
nm resolution) connected to a Synergi 4 µm Polar-RP 80 Å LC column (50 mm × 4.6 mm,
Phenomenex). Yields are given following purification, unless otherwise stated. All of the tested
compounds are at least 95% pure as estimated by HPLC.
4
.1.2 Synthetic Procedures and Characterizations
Detailed experimental procedures of compounds 3, 4a, 4b, 5, 6, 9, moxifloxacin methyl
ester and ciprofloxacin methyl ester were described in the supporting information.
General synthetic procedure A: final deprotection of compounds 8a and 8b. Compound
8
a or 8b (0.014 mmol) was dissolved in MeOH (2 mL). 2 N lithium hydroxide solution (1 mL)
was then added to the solution while stirring. The reaction mixture was stirred at room
temperature for 30 min. In an ice bath, careful acidification of the solution to pH = 6 was done
by slow addition of 1 N HCl (a.q.). The solvent was removed in vacuo. The residue was purified
by flash chromatography (elution with a gradient of DCM/MeOH from 9:1 to 1:1, v/v) to afford a

ACCEPTED MANUSCRIPT
1
6
2
white solid which was dissolved in TFA (2 mL) and H O (1 mL) solution and stirred for 2 h at
room temperature. The solvent was removed under reduced pressure to give a yellow residue. 20
mL of methanol and ether solution (1:20, v/v) was added to this residue in several portions and
then the solvent was decanted to get a yellow solid as nebramine-based hybrid TFA salt.
Stoichiometric amount of HCl aqueous solution was added into it before lyophilizing the
solution to afford a yellow solid as the final product as the HCl salt.
General synthetic procedure B: synthesis of 7a and 7b via reductive amination.
Moxifloxacin methyl ester (0.49 mmol) or ciprofloxacin methyl ester was mixed with aldehyde 6
(0.41 mmol), followed by the addition of dry DCE (25 mL) and AcOH (2.3 µL, 0.041 mmol)
under N gas. The reaction mixture was stirred at room temperature for 7 h before NaBH(OAc)
2
3
(1.64 mmol) was added at 0 ºC. The solution was gradually warmed to room temperature while
stirring overnight. The reaction mixture was cooled to 0 ºC and quenched carefully by the drop-
wise addition of saturated NaHCO solution (10 mL). The solution was then extracted with DCM
3
(3 × 15 mL). The organic layer was washed with brine, dried over anhydrous sodium sulfate and
concentrated in vacuo. The crude product was purified by flash chromatography (elution with a
gradient of DCM/MeOH from 50:1 to 10:1, v/v) to afford the desired product as a white solid.
General synthetic procedure C: synthesis of Boc-proteted NEB-based hybrids 8a, 8b and
1
0 [21,22]. 40% HCl (3 mL) and MeOH (5 mL) were added to 7a, 7b or 9 (0.048 mmol) slowly.
º
The reaction was heated to 70 C and stirred 24 h. The solution was cooled down to room
temperature and neutralized with solid sodium bicarbonate before concentrated to dryness. The
residue was taken up in 30 mL of MeOH, filtered and concentrated to give a crude 5-O-(dodecyl-
moxifloxacin methyl ester)-nebramine or 5-O-(dodecyl-ciprofloxacin methyl ester)-nebramine
HCl salt. The above crude product was dissolved in MeOH (15 mL) at room temperature.

ACCEPTED MANUSCRIPT
1
7
Triethylamine (69 µL, 0.48 mmol) and (Boc) O (0.38 mmol) were added into the solution and
2
stirred overnight at 55 ºC. Upon completion, the reaction mixture was concentrated in vacuo.
Purification by flash column chromatography (elution with a gradient of DCM/MeOH from 30:1
to 10:1 for 8a and 8b, elution with a gradient of DCM/MeOH from 100:1 to 40:1 for compound
1
0) to afford a white solid as desired product.
5
-O-(dodecyl-moxifloxacin)-nebramine 5·HCl (1a). Synthesized following general
1
procedure A. Yield: 8 mg (53%). H NMR (500 MHz, deuterium oxide) δ 8.96 (s, 1H, N-CH of
aromatic ring), 7.58 (d, J = 69.9 Hz, 1H, F-C-CH of aromatic ring), 5.67 – 5.53 (m, 1H, CH of
H-1’), 4.39 – 4.02 (m, 5H, CH of cyclopropyl, 2×N-CHH of moxifloxacin, CH of H-5’, CH of
H-4), 4.02 – 3.56 (m, 13H, OCH of linker, CH of H-4’, CH of moxifloxacin, N-CHH of
2
moxifloxacin, CH of H-2’, N-CHH of linker, OCH of moxifloxacin, CH of H-5, CH of H-6, CH
3
of H-1), 3.56 – 3.29 (m, 5H, 2×N-CHH of moxifloxacin, CH
2
of H-6’, CH of H-3), 3.28 – 3.09
(m, 2H, N-CHH of moxifloxacin, N-CHH of linker), 3.09 – 2.95 (m, 1H, CH of moxifloxacin),
2
.58 – 2.50 (m, 1H, CHH of H-2), 2.39 – 2.28 (m, 1H, CHH of H-3’), 2.23 – 2.14 (m, 1H, CHH
of H-3’), 2.14 – 0.82 (m, 29H, CHH of H-2, 10×CH of linker, 2×C-CH of moxifloxacin,
2
2
1
3
2
×CH of cyclopropyl). C NMR (125 MHz, deuterium oxide, some carbons are doubling due to
2
fluorine atom) δ 176.37 (CO of quinoline), 169.71 (CO of carboxylic acid), 153.23 (CF of
quinoline), 152.25 (CF of quinoline), 151.04 (CH, C-2 of quinoline), 141.77 (C-8 of quinoline),
1
35.02 (C-7 of quinoline), 134.85 (C-8a of quinoline), 117.57 (C-4a of quinoline), 106.84 (CH,
C-5 of quinoline), 106.67 (C-3 of quinoline), 92.08 (C-1’), 82.73 (C-6), 75.07 (C-4), 73.48 (O-
CH of linker), 73.22 (C-5’), 72.69 (C-5), 63.83 (C-4’), 62.24 (O-CH ), 56.96 (N-CH of linker),
4.21 (CH of pyrolidine), 52.41 (CH of pyrolidine), 50.90 (CH of pyrolidine), 49.87 (C-1),
2
3
2
5
4
2
2
8.94 (C-3), 47.53 (C-2’), 41.36 (N-CH of piperidine), 39.30 (C-6’), 36.00 (CH of cyclopropyl),
2

ACCEPTED MANUSCRIPT
1
8
3
2 2 2 2
5.56 (CH of pyrolidine), 29.64 (O-CH -CH of linker), 29.27 (CH of linker), 29.12 (CH of
linker), 28.99 (C-3’), 28.82 (CH of linker), 28.70 (CH of linker), 28.21 (CH of linker), 28.06
2
2
2
(
C-2), 27.92 (CH of linker), 26.04 (CH of linker), 25.30 (CH of linker), 20.15 (CH of
2 2 2 2
piperidine), 17.95 (CH of piperidine), 9.90 (CH of cyclopropyl), 8.17 (CH of cyclopropyl).
2
2
2
+
MALDI-TOF-MS m/e calcd for C H FN O Na [M+Na] : 896.5273, found: 896.5290.
4
5
72
7
9
5
-O-(dodecyl-ciprofloxacin)-nebramine 5·HCl (1b). Synthesized following general
1
procedure A. Yield: 13.4 mg (32%). H NMR (500 MHz, deuterium oxide) δ 8.67 (s, 1H, N-CH
of aromatic ring), 7.62 – 7.44 (m, 2H, C-CH of aromatic ring), 5.58 (d, J = 3.2 Hz, 1H, anomeric
H-1’), 4.14 – 4.06 (m, 2H, H of piperazine), 4.05 – 3.94 (m, 3H, NC-H of cyclopropyl, C-H of
C-O of linker), 3.89 – 3.72 (m, 7H, H-6, H-2’, H-4’, 2C-H of C-O of linker, 2C-H of C-N of
linker), 3.69 – 3.58 (m, 2H, H-1, H-5’), 3.48 – 3.26 (m, 9H, H-3, 2H of H-6’, 4H of piperazine,
2
1
C-H of linker), 2.53 (m, 1H, H-2), 2.32 (m, 1H, H-3’), 2.16 (m, 1H, H-3’), 1.96 (m, 1H, H-2),
.88 – 1.78 (m, 2H, C-H of linker), 1.69 – 1.58 (m, 2H. C-H of linker), 1.50 – 1.28 (m, 18H, CH2
1
3
of cyclopropyl and CH of linker), 1.23 (m, 2H, CH of cyclopropyl). C NMR (125 MHz,
2
2
deuterium oxide) δ 176.11 (CO of quinoline), 169.10 (CO of carboxylic acid), 154.48 (CF of
quinoline), 152.48 (CF of quinoline), 148.48 (CH, C-2 of quinoline), 144.32 (C-7 of quinoline),
1
44.24 (C-7 of quinoline), 139.10 (C-8a of quinoline), 118.99 (C-4a of quinoline), 110.97 (C-5
of quinoline), 110.78 (C-5 of quinoline), 106.85 (C-8 of quinoline), 105.83 (C-3 of quinoline),
2.05 (C-1’), 82.69 (C-5), 74.99 (C-4), 73.59 (O-CH of linker), 73.19 (C-5’), 72.68 (C-6), 63.86
9
2
(
C-4’), 57.17(N-CH of linker), 51.46 (N-CH of piperazine), 49.87 (C-1), 48.96 (C-3), 47.54 (C-
2 2
2
’), 46.61(N-CH of piperazine), 39.33 (C-6’), 36.28 (CH of cyclopropyl), 29.55 (O-CH -CH of
2 2 2
linker), 28.93 (CH
2
of linker), 28.89 (CH
2
of linker), 28.82 (C-3’), 28.81 (CH
2
of linker), 28.68
(
CH of linker), 28.34 (CH
2
2
of linker), 27.90 (CH
2
of linker), 25.85 (CH of linker), 25.33 (CH
2
2

ACCEPTED MANUSCRIPT
1
9
of linker), 23.43 (CH
2
2
of linker), 7.60 (CH of cyclopropyl). MALDI-TOF-MS m/e calcd for
+
C H FN O Na [M+Na] : 826.4851, found: 826.4820.
41
66
7
8
5
-O-((10-(4-(naphthalen-1-ylmethyl)piperazin-1-yl)docyl)-nebramine 6·HCl (2).
Compound 10 (9 mg, 0.008 mmol) was dissolved in TFA (0.5 mL) and H O (0.25 mL) solution
2
and then stirred for 2 h at room temperature. The solvent was removed under reduced pressure to
give a white residue. Amount of 5 mL of methanol and ether solution (1:20, v/v) was added to
this residue in several portions and then the solvent was decanted to get the final product 2 as
TFA salt. Stoichiometric amount of HCl aqueous solution was added into it before lyophilizing
1
the solution to afford a white solid as the final product 2 as HCl salt. Yield: 5.4 mg (72%). H
NMR (500 MHz, deuterium oxide) δ 8.27 – 8.21 (m, 1H), 8.07 – 7.97 (m, 2H), 7.72 – 7.61 (m,
2
H), 7.61 – 7.54 (m, 2H), 5.56 (d, J = 3.1 Hz, 1H, anomeric CH of H-1’), 4.22 (s, 2H, CH of
2
naphthylmethyl), 4.08 – 3.93 (m, 3H), 3.88 – 3.48 (m, 8H), 3.43 – 3.29 (m, 4H), 3.28 – 2.78 (m,
’
7
H), 2.56 – 2.45 (m, 1H, CHH of H-2), 2.34 – 2.27 (m, 1H, CHH of H-3 ), 2.18 – 2.10 (m, 1H,
’
CHH of H-3 ), 1.92 – 1.82 (m, 1H, CHH of H-2), 1.74 – 1.66 (m, 2H, N-CH -CH of linker),
2
2
1
3
1
2 2 2
.66 – 1.58 (m, 2H, O-CH -CH of linker), 1.41 – 1.24 (m, 12H, 6×CH of linker). C NMR (125
MHz, deuterium oxide) δ 133.69 (C of naphthyl), 131.88 (C of naphthyl), 129.30 (CH of
naphthyl), 129.13 (CH of naphthyl), 128.82 (CH of naphthyl), 126.77 (CH of naphthyl), 126.33
(CH of naphthyl), 125.53 (CH of naphthyl), 123.90 (CH of naphthyl), 92.10 (C-1’), 82.70 (C-6),
7
5.26 (C-4), 73.47 (O-CH of linker), 73.11 (C-5’), 72.67 (C-5), 63.73 (C-4’), 58.24 (CH of
2
2
naphthylmethyl), 56.82 (4×CH of piperazine, N-CH of linker), 49.82 (C-1), 48.85 (C-3), 47.47
2
2
(
C-2’), 39.18 (C-6’), 29.48 (O-CH -CH of linker), 28.80 (C-3’), 28.70 (2×CH of linker), 28.55
2
2
2
(
CH
2
of linker), 28.23 (CH
2
of linker), 28.03 (C-2), 25.77 (CH
2
2
of linker), 25.26 (CH of linker),

ACCEPTED MANUSCRIPT
2
0
+
2
3.46 (N-CH
found: 693.469.
-O-(dodecyl-moxifloxacin methyl ester)-1,3,2′,6′,3′′-penta-N-(tert-butoxycarbonyl)-
2 2 62 6 5
-CH of linker). MALDI-TOF-MS m/e calcd for C37H N O Na [M+Na] : 693.468,
5
4
′,2′′,4′′,6′′-tetra-O-TBDMS−tobramycin (7a). Synthesized following general procedure B.
1
Yield: 764 mg (93%). H NMR (500 MHz, methanol-d ) δ 8.71 (s, 1H), 7.68 (d, J = 14.1 Hz,
4
1
3
1
1
2
1
1
8
6
3
2
1
H), 5.51 – 5.39 (m, 2H, anomeric H), 4.26 – 4.17 (m, 1H), 4.16 – 4.08 (m, 1H), 4.07 – 3.89 (m,
H), 3.87 (s, 3H), 3.82 – 3.52 (m, 16H), 3.53 – 3.35 (m, 5H), 3.31 – 3.25 (m, 1H), 2.87 – 2.78 (m,
H), 2.66 – 2.55 (m, 1H), 2.52 – 2.35 (m, 3H), 2.15 – 2.00 (m, 1H), 1.99 – 1.90 (m, 1H), 1.90 –
.81 (m, 1H), 1.80 – 1.72 (m, 1H), 1.72 – 1.16 (m, 72H), 1.03 – 0.90 (m, 37H), 0.26 – 0.06 (m,
1
3
4H). C NMR (125 MHz, methanol-d some carbons are doubling due to fluorine atom) δ
4
,
74.88, 174.86, 166.74, 158.19, 158.03, 157.49, 157.44, 157.00, 155.93, 153.95, 152.25, 142.63,
42.57, 137.98, 137.89, 135.13, 121.96, 121.90, 109.81, 108.78, 108.59, 96.80 (2C, anomeric C),
6.59, 80.66, 80.57, 80.32, 80.16, 79.52, 78.28, 75.22, 74.83, 73.57, 72.62, 68.81, 65.12, 62.90,
1.65, 57.67, 56.98, 55.53, 55.48, 52.95, 52.08, 51.78, 50.17, 49.93, 42.14, 41.20, 38.54, 36.69,
1.90, 31.24, 30.84, 30.76, 30.73, 30.70, 29.29, 29.24, 29.12, 28.96, 28.91, 28.90, 28.71, 28.10,
7.69, 27.02, 26.93, 26.73, 26.68, 26.64, 26.54, 26.21, 25.18, 23.85, 19.52, 19.11, 18.97, 18.93,
0.00, 9.96, -3.26, -3.82, -3.93, -4.16, -4.21, -4.42, -4.62, -4.87. MALDI-TOF-MS m/e calcd for
+
C H FN O Si Na [M+Na] : 2028.2198, found: 2028.2174.
1
01 181
8
23
4
5
-O-(dodecyl-ciprofloxacin methyl ester)-1,3,2′,6′,3′′-penta-N-(tert-butoxycarbonyl)-
4
′,2′′,4′′,6′′-tetra-O-TBDMS−tobramycin (7b). Synthesized following general procedure B.
1
Yield: 724 mg (69%). H NMR (300 MHz, CDCl ) δ 8.52 (s, 1H), 8.00 (d, J = 13.2 Hz, 1H),
3
7
4
.26 (d, J = 7.1 Hz, 1H), 5.29 – 5.00 (m, 4H), 4.88 – 4.70 (m, 1H), 4.64 – 4.50 (m, 1H), 4.34 –
.00 (m, 3H), 3.90 (s, 3H), 3.84 – 3.11 (m, 22H), 2.75 – 2.65 (m, 3H), 2.50 – 2.39 (m, 2H), 2.06

ACCEPTED MANUSCRIPT
2
1
1
3
–
1.94 (m, 1H), 1.60 – 1.08 (m, 76H), 0.97 – 0.81 (m, 36H), 0.19 – -0.01 (m, 24H). C NMR (75
MHz, CDCl , some carbons are doubling due to fluorine atom) δ 173.11, 166.49, 155.08, 148.38,
3
1
4
1
44.56, 138.00, 123.02, 113.39, 113.08, 109.98, 104.79, 79.44, 79.25, 58.67, 52.89, 52.09, 49.83,
9.77, 48.33, 34.53, 30.08, 29.74, 29.67, 28.65, 28.52, 28.42, 27.59, 26.72, 26.16, 26.02, 25.80,
8.53, 18.36, 18.12, 17.93, 8.15, -3.77, -4.17, -4.87, -4.95, -5.05, -5.20. MALDI-TOF-MS m/e
+
calcd for C H FN O Si Na [M+Na] : 1958.178, found: 1958.125.
9
7
175
8
22
4
5
-O-(dodecyl-moxifloxacin methyl ester)-1,3,2′,6′-tetra-N-(tert-butoxycarbonyl)-
nebramine (8a). Synthesized following general procedure C. Yield: 40 mg (65%, two steps). H
NMR (300 MHz, methanol-d ) δ 8.71 (s, 1H), 7.68 (d, J = 14.5 Hz, 1H), 5.31 – 5.25 (m, 1H,
anomeric H-1’), 4.16 – 4.06 (m, 1H), 3.99 – 3.88 (m, 1H), 3.87 (s, 3H), 3.76 – 3.51 (m, 12H),
.50 – 3.36 (m, 4H), 3.31 – 3.23 (m, 3H), 2.92 – 2.81 (m, 1H), 2.68 – 2.58 (m, 1H), 2.50 – 2.37
m, 3H), 2.07 – 1.90 (m, 2H), 1.89 – 1.40 (m, 46H), 1.39 – 1.14 (m, 19H), 0.99 (dd, J = 4.7, 2.4
1
4
3
(
1
3
Hz, 1H). C NMR (75 MHz, methanol-d , some carbons are doubling due to fluorine atom) δ
4
1
1
7
3
2
1
74.97, 166.77, 158.24, 157.95, 157.49, 156.56, 153.28, 152.30, 142.59, 142.49, 138.06, 137.91,
35.19, 121.93, 121.83, 109.81, 108.84, 108.52, 97.46 (anomeric C), 87.23, 80.81, 80.42, 80.25,
8.93, 77.23, 74.00, 73.26, 67.14, 63.05, 61.65, 56.74, 55.35, 52.29, 52.12, 49.00, 41.20, 38.59,
5.93, 34.84, 31.55, 30.80, 30.69, 30.51, 28.90, 28.88, 28.81, 28.76, 28.49, 27.85, 27.31, 24.98,
+
3.59, 21.01, 13.95, 10.01, 9.91. MS (ESI) m/e calcd for C H FN O [M+H] : 1289.6, found:
6
6
107
7
17
289.4.
5
-O-(dodecyl-ciprofloxacin methyl ester)-1,3,2′,6′-tetra-N-(tert-butoxycarbonyl)-
nebramine (8b). Synthesized following general procedure C. Yield: 52 mg (31%, two steps). H
NMR (300 MHz, CDCl ) δ 8.54 (s, 1H), 7.99 (d, J = 13.2 Hz, 1H), 7.28 (d, J = 7.3 Hz, 1H), 5.36
5.12 (m, 2H), 5.09 – 4.88 (m, 3H), 3.96 – 3.21 (m, 25H), 3.18 – 3.08 (m, 2H), 2.78 – 2.68 (m,
1
3
–

ACCEPTED MANUSCRIPT
2
2
1
3
4
H), 2.52 – 2.43 (m, 2H), 2.37 – 2.25 (m, 1H), 2.21 – 2.08 (m, 1H), 1.72 – 1.03 (m, 76H). C
NMR (75 MHz, CDCl ) δ 173.28, 166.28, 157.97, 155.12, 148.41, 144.56, 138.06, 122.91,
3
1
2
13.37, 113.07, 109.80, 104.90, 80.26, 79.40, 58.62, 52.85, 52.09, 49.65, 46.06, 34.66, 30.16,
9.60, 29.54, 29.50, 28.45, 28.42, 28.35, 28.33, 27.46, 26.52, 26.09, 8.18. MALDI-TOF-MS m/e
+
calcd for C H FN O Na [M+Na] : 1240.711, found: 1240.658.
6
2
100
7
16
5
-O-((10-(4-(naphthalen-1-ylmethyl)piperazin-1-yl)docyl)-1,3,2′,6′-tetra-N-(tert-
butoxycarbonyl)-nebramine (10). Synthesized following general procedure C. Yield: 27 mg
1
(
53%, two steps). H NMR (500 MHz, methanol-d
4
) δ 8.29 – 8.25 (m, 1H), 7.87 – 7.83 (m, 1H),
7
.81 – 7.77 (m, 1H), 7.52 – 7.44 (m, 2H), 7.44 – 7.37 (m, 2H), 5.25 (d, J = 3.8 Hz, 1H, anomeric
CH of H-1’), 3.94 (s, 2H, CH of naphthylmethyl), 3.93 – 3.88 (m, 1H), 3.70 – 3.47 (m, 5H),
2
3
.47 – 3.31 (m, 5H), 3.28 – 3.19 (m, 3H), 2.83 – 2.44 (m, 8H), 2.44 – 2.37 (m, 1H), 2.01 – 1.92
(m, 1H), 1.92 –1.85 (m, 1H), 1.65 – 1.58 (m, 1H), 1.55 – 1.48 (m, 4H), 1.48 – 1.38 (m, 36H, 4 ×
1
3
t-Bu of Boc), 1.33 – 1.22 (m, 12H, 6 × CH of linker). C NMR (125 MHz, methanol-d ) δ
2
4
1
1
5
2
59.35, 158.27, 157.98, 157.49, 135.47, 134.63, 133.92, 129.42, 129.29, 128.92, 126.80, 126.70,
26.08, 125.78, 97.43, 87.27, 80.83, 80.43, 80.29, 78.92, 77.28, 74.04, 73.25, 67.14, 67.13, 61.63,
9.62, 54.06, 53.50, 52.85, 50.84, 50.29, 42.28, 35.95, 34.82, 31.54, 30.77, 30.61, 30.59, 30.56,
8.89, 28.86, 28.80, 28.75, 28.57, 27.30, 27.14. MALDI-TOF-MS m/e calcd for C H N O
5
7
95
6
13
+
[
M+H] : 1071.696, found: 1071.716.

ACCEPTED MANUSCRIPT
2
3
4
.2 Microbiology
4
.2.1 Clinical Isolates
Clinically-relevant bacterial strains were collected from the Canadian National Intensive
Care Unit (CAN-ICU) study [39] and Canadian Ward Surveillance (CANWARD) studies
40,41]. All pathogens obtained from CAN-ICU and CANWARD studies have received ethics
[
approval from the University of Manitoba Ethics Committee. In addition, participating Canadian
health centers have obtained appropriate ethics approval to submit clinical specimens.
4
.2.2 Antimicrobial Susceptibility Testing
The antimicrobial activity of the compounds against a panel of bacteria was evaluated by
broth microdilution assay in accordance with the Clinical and Laboratory Standards
Institute (CLSI) guidelines [32]. was performed to assess the in vitro antibacterial activity.
Bacterial cultures grown overnight were diluted in saline to achieve a 0.5 McFarland turbidity,
followed by 1:50 dilution in Mueller-Hinton broth (MHB) for inoculation to a final concentration
5
of 5×10 CFU/mL. The minimum inhibitory concentrations (MICs) of the antimicrobial agents
were determined using 96-well plates containing 2-fold serial dilutions with MHB and incubated
with equal volumes of inoculum for 18 h at 37 ºC. MIC was determined as the lowest
concentration to inhibit visible bacterial growth in the form of turbidity, which was confirmed
using an EMax Plus microplate reader (Molecular Devices, San Jose, CA, USA) at a wavelength
of 590 nm. The wells containing MHB broth with or without bacterial cells were used as positive
or negative controls, respectively.

ACCEPTED MANUSCRIPT
2
4
4
.2.3 Checkerboard Assay
The checkerboard method [42] was used to assess synergism in all tested combinations.
Fractional inhibitory concentrations (FICs) were calculated as follows: FICantibiotic
=
MICcombo/MICantibiotic alone; FICadjuvant = MICcombo/MICadjuvant alone, where MICcombo is the lowest
inhibitory concentration of drug in the presence of the adjuvant. The FIC index was calculated by
adding the FIC values. FIC indices (FICI) were interpreted as follows: ≤0.5, synergy, 0.5< FICI
≤4.0, no interaction, and ≥4.0, antagonism [28].
4
.2.4 Time-kill Curve Assay
The kinetics of bacterial killing was measured using P. aeruginosa PAO1 and PA259 as
previously described [12]. Overnight bacterial culture was diluted in saline to 0.5 McFarland
turbidity and then 1:50 diluted in Luria-Bertani broth (LB). The cell suspension (P. aeruginosa
PAO1) was incubated with the combination of ¼ × MIC (8 µg/mL) or ½ × MIC (16 µg/mL) of
hybrid 1a with 4 µg/mL of minocycline. Untreated cells in media and cells treated with 4 × MIC
(4 µg/mL) of colistin were used as negative and positive controls respectively (Fig. 2A). Respect
to the time-kill curve of 1b (Fig. 2B), the cell suspension (P. aeruginosa PA259) was incubated
with rifampicin (4 µg/mL) and 1b (4 µg/mL) alone or the combination of 1b with rifampicin at
various concentrations as shown in Fig 2B. Samples were incubated at 37 °C for 24 h. At
specific intervals (Fig. 2), aliquots (100 µL) were removed from the samples, serially diluted in
PBS, and plated on LB agar plates. Bacterial colonies were formed and counted after 20 h of
incubation at 37 °C.

ACCEPTED MANUSCRIPT
2
5
4
.2.5 Hemolytic Assay
The hemolytic activities of the newly synthesized compounds were determined and
quantified as the amount of hemoglobin released by lysing porcine erythrocytes. Fresh blood
drawn from the antecubital vein of a pig (Animal Care and Use Program, University of
Manitoba) was centrifuged at 1000 g at 4 °C for 10 mins, washed with PBS thrice and
8
resuspended in the same buffer. The final cell concentration used was 3 × 10 cells/mL.
Compounds were serially diluted with PBS and added to wells in a 96-well plate at twice the
desired concentrations. Equal volumes of erythrocyte solution were then added to each well and
incubated at 37 °C for 1 h. Intact erythrocytes were subsequently pelleted by centrifuging at 1000
g at 4 °C for 10 mins, and the supernatants were transferred to a new 96-well plate. Hemoglobin
®
release was determined by measuring the absorbance on EMax Plus microplate reader
(Molecular Devices, Sunnyvale, CA, USA) at 570 nm. Blood cells in PBS (0% hemolysis) and
0
.1 % Triton X-100 (100% hemolysis) were used as negative and positive controls, respectively.
Percent hemolysis was calculated as [% hemolysis = (X – 0%) / (100% – 0%)], where X is the
optical density values of the compounds at different concentrations.
4
.2.6 Galleria mellonella Model of P. aeruginosa Infection
In vivo synergistic effects were determined using Galleria mellonella infection model, as

previously described [8]. Briefly, worms were purchased from The Worm Lady Live Feeder
(
ON, Canada), stored in their natural habitat at 16 ºC, and used within 10 days of delivery. The
worms (average weight of 250 mg) were used for tolerability and efficacy studies. Tolerability
study was performed by injecting 10 L of antimicrobial agents only at concentrations
µ

ACCEPTED MANUSCRIPT
2
6
equivalent to 100 mg/kg or 200 mg/kg. The worms (ten in each group) were incubated at 37 °C
and monitored for 96 h. For efficacy studies, the virulence and bacterial load required to kill 100 %
of the worms within 12 – 18 h was first determined, which is approximately 5 CFU. Overnight
grown culture of MDR P. aeruginosa PA264 isolate was standardized to 0.5 McFarland standard
2
and diluted in PBS to a final concentration of 5 × 10 CFU/mL. 10
µ
L of this solution (∼ 5 CFU)
was injected into each worm and incubated for 2 h at 37 °C. After the 2 h challenge, worms in
monotherapy experimental groups (fifteen worms per group) were treated with 10 µL injection
of rifampicin, compound 1b, or PBS alone. The worms in combination therapy groups were
treated with rifampicin + compound 1b (25 + 25 mg/kg, 50 + 50 mg/kg, or 75 + 75 mg/kg).
Worms treated with 10 µL PBS negative control. The worms were incubated at 37 °C in Petri
dishes lined with filter paper and scored for survivability every 6 h for up to 24 h. This
experiment was repeated to give a total of thirty worms (n = 30) in each case. Survival data
curves were plotted using Kaplan-Meier survival analysis. Worms were considered dead if they
do not respond to touch.
4
.2.7 Tetracycline Uptake Assay
Fluorescence-based tetracycline uptake assay in bacterial cells was performed following
previously reported method [36]. Culture of P. aeruginosa PAO1 was grown to OD600 = 0.6
followed by washing and re-suspending it in ¼ volume of 10 mM HEPES, pH 7.2, 100 µL/well
cell suspension was treated with varying concentrations of test compounds in the presence of 128
µg/mL of tetracycline. Fluorescence was recorded at a continuous interval of 1 min for 30 min at
room temperature on a FlexStation 3 (Molecular Devices, Sunnyvale, USA) microplate reader at

ACCEPTED MANUSCRIPT
2
7
the excitation wavelength of 405 nm and emission wavelength of 535 nm. Experiments were
performed in triplicates. Averages of triplicate experiments are shown in Fig. 6.
Conflicts of Interest
There are no conflicts of interest to declare.
Acknowledgments
The work was supported by NSERC-DG 321252 and MHRC.
References
[1]
American Chemical Society International Historic Chemical Landmarks. Discovery and
Development of Penicillin.
http://www.acs.org/content/acs/en/education/whatischemistry/landmarks/flemingpenicillin
.html (accessed on October 11, 2018), (n.d.).
[2]
[3]
[4]
A. Coates, Y. Hu, R. Bax, C. Page, The future challenges facing the development of new
antimicrobial drugs, Nat. Rev. Drug Discov. 1 (2002) 895–910. doi:10.1038/nrd940.
J.H. Powers, Antimicrobial drug development – the past, the present, and the future, Clin.
Microbiol. Infect. 10 (2004) 23–31. doi:10.1111/j.1465-0691.2004.1007.x.
D.J. Payne, M.N. Gwynn, D.J. Holmes, D.L. Pompliano, Drugs for bad bugs: confronting
the challenges of antibacterial discovery, Nat. Rev. Drug Discov. 6 (2006) 29.
http://dx.doi.org/10.1038/nrd2201.
[5]
L.L. Silver, Are natural products still the best source for antibacterial discovery? The
bacterial entry factor, Expert Opin. Drug Discov. 3 (2008) 487–500.

ACCEPTED MANUSCRIPT
2
8
doi:10.1517/17460441.3.5.487.
[
6]
7]
G.D. Wright, Antibiotic Adjuvants: Rescuing Antibiotics from Resistance, Trends
Microbiol. 24 (2016) 862–871. doi:10.1016/j.tim.2016.06.009.
[
R. Domalaon, T. Idowu, G.G. Zhanel, F. Schweizer, Antibiotic Hybrids: the Next
Generation of Agents and Adjuvants against Gram-Negative Pathogens? Clin. Microbiol.
Rev. 31 (2018) e00077-17. doi:10.1128/CMR.00077-17.
[
[
[
[
[
8]
B.K. Gorityala, G. Guchhait, D.M. Fernando, S. Deo, S.A. McKenna, G.G. Zhanel, A.
Kumar, F. Schweizer, Adjuvants based on hybrid antibiotics overcome resistance in
Pseudomonas aeruginosa and enhance fluoroquinolone efficacy, Angew. Chem., Int. Ed.
Engl. 55 (2016) 555–559. doi:10.1002/anie.201508330.
9]
B.K. Gorityala, G. Guchhait, S. Goswami, D.M. Fernando, A. Kumar, G.G. Zhanel, F.
Schweizer, Hybrid atibiotic overcomes resistance in P. aeruginosa by enhancing outer
membrane penetration and reducing efflux, J. Med. Chem. 59 (2016) 8441–8455.
doi:10.1021/acs.jmedchem.6b00867.
10] X. Yang, S. Goswami, B.K. Gorityala, R. Domalaon, Y. Lyu, A. Kumar, G.G. Zhanel, F.
Schweizer, A tobramycin vector enhances synergy and efficacy of efflux pump inhibitors
against multidrug-resistant Gram-negative bacteria, J. Med. Chem. 60 (2017) 3913–3932.
doi:10.1021/acs.jmedchem.7b00156.
11] X. Yang, R. Domalaon, Y. Lyu, G. Zhanel, F. Schweizer, Tobramycin-Linked Efflux
Pump Inhibitor Conjugates Synergize Fluoroquinolones, Rifampicin and Fosfomycin
against Multidrug-Resistant Pseudomonas aeruginosa, J. Clin. Med. 7 (2018) 158.
doi:10.3390/jcm7070158.
12] Y. Lyu, X. Yang, S. Goswami, B.K. Gorityala, T. Idowu, R. Domalaon, G.G. Zhanel, A.