New amphiphilic neamine conjugates bearing a metal binding motif active against MDR E.?aerogenes Gram-negative bacteria

Article abstract of DOI:10.1016/j.ejmech.2016.10.054

Structure of bacterial envelope is one of the major factors contributing to Gram negative bacterial resistance. To develop new agents that target the bacterial membranes, we synthesized, by analogy with our previous peptide conjugates, new amphiphilic 3′,4′,6-trinaphthylmethylene neamines functionalized at position 5 through a short spacer by a chelating group, tris(2-pyridylmethyl)amine (TPA) and di-(picolyl)amine (DPA) and tetraazacyclotetradecane (Cyclam). ESI⁺mass spectrometry analyses showed that neither Zn(II)(NeaDPA) nor Cu(II)(NeaCyclam) were stable in the Mueller Hinton (MH) medium used for antibacterial assays. In contrast Zn(NeaTPA) was stable in the MH medium. Interestingly, in MH, the free ligand NeaTPA was found bound to zinc, the zinc salt being the most abundant salt in this medium. Thus, the antibacterial activities of all compounds were evaluated as free ligands against E.?coli strains, wild type AG100 and E.?aerogenes EA289 (a clinical MDR strain that overexpresses AcrAB-TolC efflux pump), as well as AG100A an AcrAB- E.?coli strain and EA298 a TolC- derivative. NeaCyclam and Zn(NeaTPA) were by far the most efficient compounds active against resistant isolate EA289 with MICs in the range 16–4 and 4?μM, respectively, while usual antibiotics such as β-lactams and phenicols were inactive (MICs?>?128) and ciprofloxacin was at 64?μM. Zn(NeaTPA) and NeaCyclam were shown to target and permeabilize the outer membrane of EA289 by promoting the cleavage of nitrocefin by periplasmic β-lactamase. Moreover, all the neamine conjugates were able to block the efflux of 1,2’-dinaphthylamine in EA289, by acting on the efflux transporter located in the inner membrane. These membranotropic properties contribute to explain the activities of these neamine conjugates toward the MDR EA289 strain.

Full text of DOI:10.1016/j.ejmech.2016.10.054

Accepted Manuscript
New amphiphilic neamine conjugates bearing a metal binding motif active against
MDR E. aerogenes Gram-negative bacteria
Anas Allam, Laure Maigre, Rodolphe Alves de Sousa, Estelle Dumont, Julia Vergalli,
Jean-Marie Pagès, Isabelle Artaud
PII:
S0223-5234(16)30916-3
10.1016/j.ejmech.2016.10.054
EJMECH 9019
DOI:
Reference:
To appear in: European Journal of Medicinal Chemistry
Received Date: 16 June 2016
Revised Date: 29 September 2016
Accepted Date: 23 October 2016
Please cite this article as: A. Allam, L. Maigre, R. Alves de Sousa, E. Dumont, J. Vergalli, J.-M. Pagès,
I. Artaud, New amphiphilic neamine conjugates bearing a metal binding motif active against MDR E.
aerogenes Gram-negative bacteria, European Journal of Medicinal Chemistry (2016), doi: 10.1016/
j.ejmech.2016.10.054.
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
New Amphiphilic Neamine Conjugates Bearing a Metal Binding Motif
Active Against MDR E. aerogenes Gram-negative Bacteria
Anas Allam,‡1 Laure Maigre,‡2 Rodolphe Alves de Sousa,1 Estelle Dumont2, Julia Vergalli2,
Jean-Marie Pagès,*2 and Isabelle Artaud.*1
1Laboratoire de Chimie et Biochimie Pharmacologiques et Toxicologiques, UMR8601, CNRS Université
Paris Descartes, PRES Paris Cité, 45 rue des St Pères 75270 Paris Cedex 06 France.
2UMR_MD1 Aix Marseille Univ, IRBA, Transporteurs Membranaires, Chimiorésistance et Drug-
Design, Facultés de Médecine et de Pharmacie, 27 boulevard jean Moulin 13385 Marseille Cedex 05,
France.
Corresponding Authors
*Jean-Marie Pagès: jean-marie.pages@univ-amu.fr Phone: 33-4-91 32 45 87 Fax: 33-4-91 42 46 06.
*Isabelle Artaud : isabelle.artaud@parisdescartes.fr Phone: 33-1-42 86 21 89 Fax: 33-1-42 86 83 87.
Author Contributions
J.M.P. and I.A. designed the research and wrote the manuscript. A.A. and R.A. synthezised and characterized the
neamine conjugates. L.M., E.D. J.V. measured the antibiotic susceptibilities and performed nitrocefin hydrolysis
as well as DNA efflux experiments. All authors have given approval to the final version of the manuscript.
‡These authors contributed equally.

ACCEPTED MANUSCRIPT
Abstract
Structure of bacterial envelope is one of the major factors contributing to Gram negative bacterial resistance. To
develop new agents that target the bacterial membranes, we synthesized, by analogy with our previous peptide
conjugates, new amphiphilic 3’,4’,6-trinaphthylmethylene neamines functionalized at position 5 through a short
spacer by a chelating group, tris(2-pyridylmethyl)amine (TPA) and di-(picolyl)amine (DPA) and
tetraazacyclotetradecane (Cyclam). ESI+ mass spectrometry analyses showed that neither Zn(II)(NeaDPA) nor
Cu(II)(NeaCyclam) were stable in the Mueller Hinton (MH) medium used for antibacterial assays. In contrast
Zn(NeaTPA) was stable in the MH medium. Interestingly, in MH, the free ligand NeaTPA was found bound to
zinc, the zinc salt being the most abundant salt in this medium. Thus, the antibacterial activities of all compounds
were evaluated as free ligands against E. coli strains, wild type AG100 and E. aerogenes EA289 (a clinical MDR
strain that overexpresses AcrAB-TolC efflux pump), as well as AG100A an AcrAB- E. coli strain and EA298 a
TolC- derivative. NeaCyclam and Zn(NeaTPA) were by far the most efficient compounds active against resistant
isolate EA289 with MICs in the range 16-4 and 4 µM, respectively, while usual antibiotics such as β-lactams and
phenicols were inactive (MICs > 128) and ciprofloxacin was at 64 µM. Zn(NeaTPA) and NeaCyclam were
shown to target and to permeabilize the outer membrane of EA289 by promoting the cleavage of nitrocefin by
periplasmic β-lactamase. Moreover, all the neamine conjugates were able to block the efflux of 1,2’-
dinaphthylamine in EA289, by acting on the efflux transporter located in the inner membrane. These
membranotropic properties contribute to explain the activities of these neamine conjugates toward the MDR
EA289 strain.
Keywords: Antibacterial activity, Gram-negative bacteria, Metal-binding groups, Membrane permeability,
Multidrug resistant bacteria, Neamine derivatives.

ACCEPTED MANUSCRIPT
1. Introduction
The widespread dissemination of Gram-negative bacterial pathogens resistant to available antibiotics requires the
development of new classes of antibiotics with new modes of action [1,2]. The particular structure of Gram-
negative cell wall is one of the major factors responsible for this resistance. Modifications of the outer membrane
(OM) permeability reduce the uptake of both hydrophilic and hydrophobic compounds [3], while the efflux
transporters, located in the two bacterial membranes, pump antibacterial agents out of the cell before they can
reach their target [4]. As a result, the membrane barrier appears as a target of interest to circumvent drug
resistance. The first drugs described to specifically target the membrane barrier were the naturally occurring
antimicrobial peptides (AMPs) that are cationic and amphiphilic [5,6]. They interact with the negatively charged
lipopolysaccharide (LPS) of the OM leading to membrane disruption and, eventually, they translocate across the
cytoplasmic (inner) membrane to interact with cytoplasmic targets [7]. Recently it has been shown that grafting
an ATCUN (Amino Terminal Cu and Ni) metal binding site on these AMPs improved their antibacterial
efficiency [8-10]. This increase of activity was attributed to a dual action of these derivatives that, in addition to
their usual mode of action, could promote the formation of radical oxygen species [9] and / or DNA cleavage
[10]. In a related approach, we have developed tetra- and hexa-peptide analogues of AMPs with tris(2-
pyridylmethyl)amine metal binding group that were more efficient than the parent peptides against E. coli and
multidrug resistant (MDR) E. aerogenes bacterial strains [11]. They could also be used under a Co(III) or Zn(II)
metalated form as drug carrier, but metalation was not absolutely required since the free system was readily
linked to a metal cation in the culture medium used for bacterial growth [12]. However the efficiency of these
compounds remained moderate and to improve it, we proposed to replace the moderatly active lipophilic cationic
peptides by the more active lipophilic cationic neamines [13-19]. Indeed, recently, several groups have shown
that converting the free hydroxyl groups of aminoglycosides into lipophilic ethers [13-18], or thioethers [19],
while leaving the amines free, produced a new class of molecules active against a panel of Gram-negative
bacteria. For instance, 3’,4’,6-trinaphthylmethylene neamine was active against susceptible and resistant Gram-
negative bacteria. Décout’s group has shown it was unable to bind to 16S rRNA but able to bind to LPS and to
induce P. aeruginosa membrane depolarization, probably explaining its antibacterial activity [17].
Herein, and related to our previous work [11,12], we synthesized new neamine compounds, with amphiphilic
character, coupled to several chelating groups, such as tris(2-pyridylmethyl)amine (TPA) and di-(picolyl)amine
(DPA) and tetraazacyclotetradecane (Cyclam). Their antibacterial activities were evaluated against Escherichia
coli wild type AG100 and Enterobacter aerogenes EA289 (a clinical MDR strain found in nosocomial infections
that overexpresses AcrAB-TolC efflux pumps), as well as AG100A an AcrAB- E. coli derivative strain and
EA298 a TolC- E. aerogenes derivative. Then, we evaluated their metal binding abilities in the Mueller Hinton
(MH) medium used for bacterial growth and drug susceptibility determination. Then, to determine their
membranotropic activity, the neamine conjugates were evaluated against EA289 for their ability to permeabilize
the OM and to alter the efflux of 1,2’-dinaphthylamine (1,2’-DNA), a substrate of AcrB pump which is located in
the inner membrane.

ACCEPTED MANUSCRIPT
2. Results
2.1 Synthesis of neamine conjugates
3’,4’,6-trinaphthylmethylene neamine was previously prepared by Décout et al.[18]. Its membranotropic
properties [17] make 3’,4’,6-trinaphthyl-methylene neamine a candidate of choice to introduce at position 5 of the
neamine scaffold a chain and further a chelating group. Syntheses of all neamine conjugates are depicted in
Scheme 1. Preparation of 3’,4’,6-trinaphthylmethylene neamine 1 was performed from neamine as previously
described by Jackowski et al.[20]. The less reactive hydroxyl at position 5 of 1 (Scheme 1) was alkylated with
1,6-dibromohexane in DMF using NaH as base. Nucleophilic substitution of the bromide of 2 with sodium azide
led to the azide 3 that was further reduced into the amine 4 with triphenyl phosphine. Compound 4 was the
precursor to introduce the different chelating groups, TPA, DPA and Cyclam. They were grafted through a
glutaryl spacer for Cyclam 5 (under its Boc protected form) and TPA 6, and through a valeric one for DPA 7. To
assess the effect of the chelating group we also prepared the tri-benzylamine (TBA) derivative 8 equivalent to
TPA but without any complexing ability. Syntheses of compounds 6, 7 and 8 were described in our previous work
[12]. Synthesis of the Cyclam derivative 5 was achieved in two steps (Scheme S1) from tri-tert-butyl 1,4,8,11-
tetraazacyclotetradecane-1,4,8-tricarboxylate as previously described by Fabbrizzi et al. [21]. After protecting
three of the amine functions as tert-butyl carbamate, reaction of the last free amine with methyl 5-chloro-5-
oxopentanoate followed by saponification with aqueous LiOH afforded the expected compound 5. Coupling of
the neamine derivative 4 with the acids 5, 6, 7 and 8 was achieved in DMF in the presence of DIEA as base and,
HOAT and HATU as coupling agents. Then deprotection with TFA in CH2Cl2 afforded the various neamine
conjugates bearing a Cyclam (NeaCyclam), a TPA (NeaTPA) and a DPA (NeaDPA) binding motif and the non-
chelating TBA moiety (NeaTBA). Deprotected azide (NeaN3) and deprotected amine (NeaNH2) were also
prepared for control biological assays.
Insert Scheme 1
2.2 Binding of a metal cation in the culture medium
Metallodrugs although promising are always assumed to be toxic and this leads to suspiciousness of
pharmaceutical companies while highly efficient metallodrugs are already applied to therapy or diagnosis of
diseases [22-24]. Using strong chelating metal-binding groups, our neamine conjugates could readily bind a metal
cation in the rich MH medium broth used for bacterial growth in the antibacterial assays. TPA [25] and DPA [25]
have a strong affinity for metal cations but decreasing in the order Cu(II) > Zn(II) > Fe(II). Cyclam with a N-
substituted carboxamide binds efficiently copper [26]. While iron and zinc salts are present in high amounts in the
MH medium, the content in copper salt is low [27,28]. In this regard, the best metal complexes candidates for this
study are [Zn(NeaTPA)]2+, [Zn(NeaDPA)]2+ and [Cu(NeaCyclam]2+ whose structures are shown in Scheme S2.
So, we prepared zinc and copper complexes of our neamine conjugates by mixing the neutral free ligand with
Zn(ClO4)2 in MeOH or Cu(CF3SO3)2 in DMF. The three complexes, as well as the free ligands, were analyzed by

ACCEPTED MANUSCRIPT
ESI+ HRMS and their mass spectra effectively displayed the expected molecular peak. Then, we studied their
stability upon dilution in MH at a concentration close to that used for the evaluation of their antibacterial
activities. Concomitantly, we studied, under the same conditions, the direct binding of zinc or copper to the
neamine conjugates in the MH broth. The mass spectra shown in Figure 1 and in Figures S1 and S2 as well as the
analysis of the data listed in Table S1 revealed that [Zn(NeaDPA)]2+ and [Cu(NeaCyclam)]2+ dissociated upon
incubation in MH. The only stable complex was [Zn(NeaTPA)]2+ and, interestingly, the free NeaTPA ligand
readily binds zinc upon incubation in the MH broth (Figure 1). As a result, we only tested the free ligands,
bearing in mind that the species tested with TPA was Zn(NeaTPA).
Insert Figure 1
2.3 Antibacterial activities of neamine conjugates against E. coli and E. aerogenes strains
Minimal inhibitory concentrations (MICs) of neamine conjugates were determined against isogenic strains: two
Escherichia coli, the wild type AG100 and its derivative AG100A devoid of AcrAB efflux pump, and two
Enterobacter aerogenes, EA289 a clinical MDR strain overexpressing the AcrAB-TolC efflux pumps and its tolC-
derivative EA298 (Table 1). Neamine was not significantly active on strains that expressed efflux pumps with
MIC in the range 32-16 µM and was inactive on strains devoid of efflux pump (MIC > 128 µM). The NeaN3
derivative used as control in our series exhibited activities similar to those of neamine on active efflux strains but
was more efficient on efflux- AG100A and EA298. Among the neamine conjugates, NeaTBA with no metal
binding ability was the less potent derivative. The other neamine conjugates showed similar activities against the
parental strain and its efflux– derivative indicating that there were not well-recognized substrates of AcrAB
pump. Their activity increased in the order: DPA < NH2 < Cyclam
≈ TPA. It is worth noticing that MIC values
obtained toward the resistant clinical strain (EA289) with NeaTPA and NeaCyclam were lower than those
obtained with the usual antibiotics belonging to β-lactams, quinolones, and phenicol families (Table 1). Addition
of sub-inhibitory concentration of polymyxin B nonapeptide (PMBN), a molecule known to permeabilize the
membrane [29], did not significantly improve NeaTBA and NeaDPA activities (1 or 2 dilutions). All these data
support a possible permeabilizing effect of the neamine conjugates and indicate that they are able to bypass the
resistance mechanisms acting in this MDR strain.
Insert Table 1
2.4 Effect of neamine conjugates on the hydrolysis rate of nitrocefin.
The effect of the neamine conjugates on the OM permeability was analyzed using nitrocefin, a well-known
substrate of the β-lactamase periplasmic enzyme [30] that has been previously used to assay the OM permeability
during incubation with various compounds [31,32]. EA289 was incubated with different concentrations of
neamine conjugates and carbonyl cyanide m-chlorophenylhydrazone (CCCP) at 3.3 µM to prevent the efflux of
molecules. After addition of nitrocefin, the OD at 490 nm was recorded for 1 h. An increase in the optical density
(OD) indicated nitrocefin hydrolysis associated with its translocation across the OM. In the absence of neamine

ACCEPTED MANUSCRIPT
conjugates, no significant variation of the absorbance was observed showing that nitrocefin did not pass the OM
under these conditions. In contrast, addition of the neamine conjugates promoted a concentration-dependent
increase of the rate of nitrocefin hydrolysis (Figure 2 and Figure S3). Among all neamine conjugates, NeaCyclam
was the most efficient. While 90 % of nitrocefin was hydrolyzed after 50 min with 32 µM of NeaTPA, the
reaction was complete with only 2 µM of NeaCyclam (Figure 2). Addition of NeaDPA or NeaTBA at 64 µM
(Figure 2) promoted only 60% and 20% of nitrocefin hydrolysis, respectively. In the same assays, NeaNH2
(Figure S3) exhibited an intermediate profile between NeaTPA and NeaCyclam. Under the same conditions, we
also assayed polymyxin B (PMB) (Figure 2), a well-known antibiotic that efficiently permeabilizes the bacterial
OM [33]. A rapid and strong permeabilizing effect was observed with PMB at only 3 µM within 20 min, when
only a weak action was noted with polymyxin B nonapeptide (PMBN) (Figure S3) known to exhibit a lower
membranotropic action [34]. NeaCyclam with an activity profile quite close to that of PMB presented the higher
permeabilizing effect. Finally, the activity range toward the bacterial OM was the following: PMB > NeaCyclam
> NeaNH2 > NeaTPA >> NeaDPA > NeaTBA ≈ PMBN.
Insert Figure 2
2.5 Effect of neamine conjugates on the efflux of 1,2’ DNA.
We investigated the effect of the neamine conjugates on the efflux pump activity using the fluorophore 1,2'
DiNaphtyl-Amine (1,2’-DNA), a substrate of AcrAB-TolC system, which is non-fluorescent in aqueous solution
but strongly fluorescent upon partitioning into the phospholipid bilayer [35]. EA289 overexpressing acrAB efflux
pump, was pre-incubated in the presence of 1,2'-DNA and CCCP (CCCP collapses the proton motive force that
energizes the efflux pump [4,36]). After centrifugation and resuspension in phosphate buffer, the release of 1,2'-
DNA was assayed in the absence or in the presence of different concentrations of neamine conjugates. Since
AcrAB pump, as other RND pumps, utilizes the Proton Motive Force (PMF) as an energy source driving the drug
transport [37], we used both, as controls, phenylalanine arginine ß-naphthylamide (PAßN), an AcrAB inhibitor,
and CCCP, an energy uncoupler [38]. Glucose was added to provide energy to efflux pumps [32] and start the
expel (t = 0, Figure 3): the fluorescence intensity decrease was related to 1,2'-DNA efflux outside the bacterial
membrane. The addition of neamine conjugates NeaTPA, NeaDPA and NeaCyclam was able to totally impair
1,2’-DNA efflux at around 8 and 16 µM (Figure 3) compared to the limited action of NeaTBA at 32 µM (Figure
S4). PAßN was inefficient (Figure S4), but under the same conditions, CCCP blocked the efflux in the same range
of concentrations (Figure 3) as neamine conjugates demonstrating that 1,2’-DNA efflux was dependent on the
energy source. In addition, the respective profiles of the curves obtained with NeaDPA and NeaCyclam
derivatives were only a little bit different from CCCP that supported an effect on the efflux transporter activity
located in the inner membrane.
Insert Figure 3

ACCEPTED MANUSCRIPT
3. Discussion
From the above-mentioned results, two of the neamine conjugates tested for their antibacterial activities, NeaTPA
and NeaCyclam are highly effective against E. coli strain and resistant E. aerogenes strains. NeaDPA is less
efficient than NeaNH2 while NeaTBA remains inactive. The antibacterial activities of the neamine conjugates
follow quite closely their effect on the OM. Hence, they clearly affect the OM membrane stability by altering the
permeability barrier as demonstrated by the increase of nitrocefin hydrolysis following the addition of NeaTPA
or NeaCyclam, in contrast to the absence of any effect of NeaTBA. However, the difference in metal binding
affinity of the neamine conjugates cannot by itself explain the variations of antibacterial activity, since the only
stable complex in the biological conditions is [Zn(NeaHnTPA)](n+2)+ (n being the number of positive charges
arising from the neamine core, 1 ≤ n ≤ 4). In contrast, Cyclam with a nitrogen substituted carboxamide remains
highly efficient while binding relatively poorly Cu(II), since the copper complex is unstable in the MH broth.
Thus, the effect on the OM seems to be modulated by the number of positive charge exhibited by the compounds.
So, we calculated pKa of all the amines in each derivative using Discovery Studio software (Table S2). While
pKa values of the aliphatic amine(s) in the metal binding groups are all around 6.6, those of the neamine core are
clearly modulated by the nature of the nitrogen ligand. At pH 7.0, the tertiary amine of TBA and DPA in NeaTBA
and NeaDPA, respectively, is mainly deprotonated, while the cyclam ring in NeaCyclam is partly protonated and
can have an average charge of +1. Moreover, three of the neamine amines are protonated in NeaCyclam against
two in NeaTPA and only one in NeaDPA and NeaNH2, the others being partly protonated. Interaction with the
OM is also modulated by the amphipatic nature of the neamine conjugates. The naphthyl residues of the neamine
core, and the positive charge associated with the protonation of the NH2 groups, facilitate the interaction with the
lipids and the negatively charged phospholipid headgroups, respectively. The neamine conjugates bring additional
positive charges. In that regard, incorporation of zinc in NeaTPA brings two additional positive charges (global
charge +4), explaining its interaction with the OM as previously reported for other DPA [39,40] or phthalocyanine
[41,42] zinc complexes. NeaCyclam is efficient as free ligand, since it exhibits 4 positive charges due to its
protonation state, whereas NeaDPA brings only one charge. NeaTBA is the less efficient derivative, not only
because of its protonation state, but also because the addition of three benzyl groups drastically impairs the
hydrophobicity / hydrophilicity balance (calculated LogD (LogP at pH 7.0): NeaTBA 7.664, NeaDPA 4.286,
NeaTPA 3.836, NeaNH2 1.941, NeaCyclam -0.657).
All the neamine conjugates have a similar activity on the efflux and this property could be related to their overall
protonation state in the more acidic periplasmic compartment (pH
≈ 6). The reserve of periplasmic protons is
required for the antibiotic-H+ antiport by the AcrB pump [4,43]. Overall protonation of the neamine conjugates
could induce a decrease of free periplasmic protons under the threshold necessary for H+ antiport flux that is
required for an efficient AcrB activity, as previously reported for ethidium bromide extrusion [35].
4. Conclusion
To conclude, in this paper, we reported the synthesis and complete characterization of new amphiphilic neamine
conjugates, the OH at position 5 of which was functionalized by tris(2-pyridylmethyl)amine (TPA) or di-

ACCEPTED MANUSCRIPT
(picolyl)amine (DPA) or tetraazacyclotetradecane (Cyclam) via a short spacer. NeaTPA and NeaCyclam are the
most potent antibacterial agents being active both against E. coli strains and a MDR E. aerogenes strain EA289,
that has lost its susceptibility toward usual antibiotics. Clearly, the neamine core contributes to improve the
activities relative to similar derivatives that we reported with tetra- or hexa-peptide analogues of AMPs [11,12].
In this series, the increase in activity relative to the precursor NeaNH2 is associated with an increase in the
positive charges, due either to a high protonation state of the neamine core in NeaCyclam, or to a direct
metallation of TPA in NeaTPA by Zn(II) cations available in the culture medium. The OM interaction and
membranotropic action, evidenced by the increase of nitrocefin hydrolysis by periplasmic β-lactamase, could
contribute to a significant antibacterial activity on MDR E. aerogenes EA289. Moreover all compounds are able
to target the inner membrane by blocking the efflux of 1,2’-DNA.
5. Experimental
5.1 Chemical syntheses
All solvents and chemicals were purchased from SDS and Aldrich, respectively. DMF (CaH2), MeOH (Mg
turnings and iodine) and CH3CN (CaH2) were dried upon distillation over standard reagents indicated in brackets.
1H NMR spectra were recorded on a Bruker ARX-250 spectrometer or on a Bruker Avance-500 spectrometer
(Bruker, Wissembourg, France) and chemical shifts were reported in ppm downfield from TMS. Electrospray
ionization (ESI) and HRMS mass spectrometry analyses were obtained using a Thermo Finnigan LCD Advantage
spectrometer (Thermo Electron SAS, Courtaboeuf, France). Elemental analyses were carried out by microanalysis
service at Gif-sur-Yvette CNRS.
Neamine was synthesized as previously described [44] with some modifications. Briefly, a concentrated HCl
solution (25 mL, 12.1 N) was added to a MeOH solution (230 mL) of neomycin B (10.0 g, 10.4 mmol). The
reaction was refluxed for 4 h before all reactants were completely consumed. After cooling to r.t., a light-yellow
solution containing a white precipitate was observed. This later was filtered and washed with MeOH to give a
first fraction of neamine. A second fraction was isolated by successive cycles of dissolutions of the concentrated
light-yellow solution in water (3 mL) and precipitations in methanol (50 mL). This allowed obtaining the crude
1
neamine product in 80% yield. H NMR (250 MHz, D2O) δ ppm 5.96 (d, J = 3.9 Hz, 1H), 4.10 – 3.96 (m, 3H),
3.73 (t, J = 8.9 Hz, 1H), 3.67 – 3.57 (m, 2H), 3.57 – 3.44 (m, 3H), 3.44 – 3.25 (m, 2H), 2.54 (dt, J = 12.5 & 4.4
Hz, 1H). 1.95 (q, J = 12.5 Hz, 1H). MS (ESI+, m/z, H2O): 322.9 (100%, [M+H]+).
Compound 1 was prepared as previously described [20,45]. 1H NMR (500 MHz, CDCl3) δ ppm 7.88 – 7.66 (m,
8H), 7.65 – 7.41 (m, 22H), 7.40 – 7.00 (m, 37H), 6.99 – 6.82 (m, 14H), 5.26 (s, 1H), 4.92 (d, J = 11.4 Hz, 1H),
4.88 (d, J = 11.0 Hz, 1H), 4.81 (d, J = 11.1 Hz, 1H), 4.78 (d, J = 11.1 Hz, 1H), 4.56 (d, J = 11.4 Hz, 1H), 4.45 (d,
J = 11.0 Hz, 1H), 4.34 (br, 1H), 4.14 (m, 1H) 4.02 (br, 1H), 3.61(t, J = 9.5 Hz, 2H), 3.14 (t, J = 9.5 Hz, 1H), 3.02
(t, J = 9.5 Hz, 1H), 2.90 (m, 2H), 2.76 (br, 1H), 2.32 (m, 2H), 1.99 – 1.84 (m, 3H), 1.39 – 1.25 (m, 2H), 0.90 (m,
1H). MS (ESI+, m/z, MeOH): 1712.3 (100%, [M+H]+). El. anal. calcd. for C121H106N4O6: C, 84.88; H, 6.24; N,
3.27; found: C, 84.74; H, 6.12; N, 2.88.

ACCEPTED MANUSCRIPT
Compound 2. To a DMF solution (20 mL) of compound 1 (1.95 g, 1.14 mmol) were successively added NaH
(460 mg, 11.5 mmol) and 1,6-dibromohexane (1.86 mL, 11.3 mmol), and the mixture was stirred at 60 °C for 20
h. After evaporation of DMF and dissolution of the residue in CH2Cl2 (200 mL) the solution was washed twice (2
x 150 mL) with saturated aqueous NH4Cl, dried over MgSO4 and evaporated. The residue was purified over basic
1
alumina gel using pentane/CH2Cl2 (70:30, v:v) as eluent to give compound 2 in 66% yield. H NMR (500 MHz,
CDCl3) δ ppm 7.85 – 6.82 (m, 81H), 5.28 (s, 1H), 4.91 – 4.42 (m, 4H), 4.41 – 4.21 (m, 3H), 3.56 (m, 1H), 3.40
(m, 1H), 3.24 (m, 1H), 3.07 (m, 2H), 2.84 – 2.41 (m, 5H), 2.09 – 1.79 (m, 2H), 1.57 – 1.42 (m, 6H), 1.32 – 1.16
(m, 4H), 1.11 – 0.97 (m, 2H), 0.93 – 0.78 (m, 4H). MS (ESI+, m/z, MeOH): 1875.2 (100%, [M+H]+). El. anal.
calcd. for C127H117BrN4O6•1.5H2O: C, 80.19; H, 6.36; N, 2.95; found: C, 80.04; H, 6.50; N, 3.23.
Compound 3. Sodium azide (275 mg, 4.24 mmol) was added to a DMF solution (10 mL) of compound 2 (1.0 g,
0.53 mmol), and the mixture was stirred at r.t. for 20 h. After evaporating DMF, CH2Cl2 (300 mL) was added to
the residue and the solution was washed twice (2 x 200 mL) with saturated aqueous NH4Cl, dried over MgSO4
and evaporated. The residue was purified over basic alumina gel using pentane/CH2Cl2 (70:30, v:v) as eluent to
1
give compound 3 in 81% yield. H NMR (500 MHz, CDCl3)
δ ppm 7.25 – 6.98 (m, 24H), 7.63 – 6.26 (m, 41H),
7.96 – 7.65 (m, 16H), 5.27 (s, 1H), 4.80 – 4.40 (m, 7H), 3.72 (m, 1H), 3.59 (m, 1H), 3.45 – 3.30 (m, 2H), 3.20
(m, 1H), 3.02 (m, 2H), 2.98 – 2.80 (m, 4H), 2.78 – 2.42 (m, 4H), 2.12 (m, 2H), 1.75 – 1.52 (m, 3H), 1.47 – 1.29
(m, 4H), 1.17 – 0.94 (m, 4H). MS (ESI+, m/z, MeOH): 1837.3 (100%, [M+H]+). El. anal. calcd. for
C127H117N7O6•0.6H2O: C, 82.53; H, 6.45; N, 5.31; found: C, 82.91; H, 6.46; N, 4.87.
Compound 4. Triphenyl phosphine (713 mg, 2.72 mmol) was added to a solution of compound 3 (1.0 g, 0.54
mmol) in a mixture of THF/water (95:5 v:v) (30 mL). Then the mixture was refluxed for 5 h. After evaporating
the solvents, the residue was purified over basic alumina gel using CH2Cl2/ MeOH (98:2 v:v) as eluent to afford
compound 4 in 71% yield. 1H NMR (500 MHz, CDCl3)
δ ppm 7.76 – 7.55 (m, 12H), 7.54 – 7.25 (m, 32H), 7.24 –
7.07 (m, 15H), 7.05 – 6.84 (m, 22H), 4.85 – 4.33 (m, 5H), 4.31 – 4.17 (m, 3H), 3.49 (m, 1H), 3.33 (m, 1H), 3.17
(m, 1H), 2.98 (m, 1H), 2.80 – 2.54 (m, 4H), 2.53 – 2.42 (m, 2H), 2.41 – 2.19 (m, 3H), 2.01 – 1.72 (m, 2H), 1.63 –
1.35 (m, 5H), 1.28 – 1.13 (m, 3H), 1.11 – 0.98 (m, 3H), 0.96 – 0.70 (m, 4H) (in CDCl3, 119H attributed including
4 NHTrt and 1 NH2). HRMS (ESI+, m/z, CH2Cl2): calcd for C127H120N5O6 1811.9267 ([M+H]+), found 1811.9282.
El. anal. calcd. for C127H119N5O6•2.6H2O: C, 82.09; H, 6.74; N, 3.77; found: C, 82.43; H, 6.69; N, 3.42.
General procedure for amine deprotection. To a CH2Cl2 solution (5 mL) kept at 0 °C of the amino protected
compound (0.109 mmol) and anisole (250 µL) was added dropwise TFA (5 mL). After stirring the mixture at r.t.
for 2 h, the solvents were evaporated and the product was isolated after precipitation into diethyl ether followed
by centrifugation.
General procedure for coupling. To a DMF solution (2 mL) of compound 4 (0.055 mmol, 100 mg), compound
5, 6, 7 or 8 (0.055 mmol), HOAT (0.0825 mmol, 11 mg) and HATU (0.0825 mmol, 31 mg) was added DIPEA
(0.22 mmol, 38 µL) at 0 °C. After stirring for 3 h at r.t., addition of water to the DMF solution led to precipitation
of the product that was washed with MeOH and used in the deprotection step without further purification.
1
NeaN3. 81% yield (white powder) H NMR (500 MHz, MeOD-d4)
δ ppm 7.88 – 7.77 (m, 9H), 7.76 – 7.69 (m,
3H), 7.52 – 7.39 (m, 9H), 5.52 (d, J = 3.3 Hz, 1H), 5.02 (s, 2H), 4.98 (d, J = 11.8 Hz, 1H), 4.94 (d, J = 11.5 Hz,

ACCEPTED MANUSCRIPT
1H), 4.92 (d, J = 11.8 Hz, 1H), 4.86 (d, J = 11.5 Hz, 1H), 4.24 (ddd, J = 9.0, 7.8 & 3.0 Hz, 1H), 3.95 (d, J = 7.8
Hz, 1H), 3.72 (t, J = 8.9 Hz, 1H), 3.71 (m, 1H), 3.57 (t, J = 8.9 Hz, 1H), 3.54 (t, J = 7.8 Hz, 1H), 3.49 (t, J = 8.9
Hz, 1H), 3.27 (dd, J = 13.3 & 3.0 Hz, 1H), 3.19 (m, 1H), 3.18 (m, 1H), 3.17 (dd, J = 13.3 & 9.0 Hz, 1H), 3.11
(ddd, J = 12.4, 8.9 & 4.3 Hz, 1H), 3.06 (dd, J = 7.8 & 3.3 Hz, 1H), 2.86 (t, J = 6.7 Hz, 2H), 2.23 (dt, 12.4 & 4.3
Hz, 1H), 1.61 (q, 12.4 Hz, 1H), 1.47 – 1.32 (m, 2H), 1.18 (m, 2H), 1.03 (m, 1H), 0.92 – 0.95 (m, 3H) (53H
attributed + 4 NH2 = 61H) HRMS (ESI+, m/z, MeOH): calcd. for C51H62N7O6 [M+H]+ 868.4756, found 868.4752.
El. anal. calcd. for C51H61N7O6•4.45CF3COOH: C, 52.30; H, 4.80; N, 7.13; found: C, 52.76; H, 5.00; N, 6.64.
NeaNH2. 73% yield (White powder) 1H NMR (500 MHz, MeOD-d4)
δ ppm 7.90 – 7.82 (m, 6H), 7.82 – 7.75 (m,
4H), 7.73 (br, 2H), 7.53 – 7.47 (m, 7H), 7.42 (d, J = 8.5 Hz, 1H), 7.37 (d, J = 8.5 Hz, 1H), 5.64 (d, J = 2.1 Hz,
1H), 5.07 (d, J = 11.8 Hz, 1H), 5.02 (d, J = 11.8 Hz, 1H), 4.85 (s, 2H), 4.82 (m, 1H), 4.78 (m, 1H), 4.62 (ddd, J =
10.8, 3.0 & 2.2 Hz, 1H), 4.31 (t, J = 4.0 Hz, 1H), 4.23 (dd, J = 10.2 & 9.2 Hz, 1H), 3.75 (t, J = 9.2 Hz, 1H), 3.73
(dd, J = 4.0 & 2.1 Hz, 1H), 3.71 (m, 2H), 3.70 (m, 1H), 3.67 (dd, J = 14.2 & 10.8 Hz, 1H), 3.60 (t, J = 9.2 Hz,
1H), 3.46 (ddd, J = 12.4, 10.2 & 4.4 Hz, 1H), 3.40 (ddd, J = 12.4, 9.2 & 4.4 Hz, 1H), 3.14 (dd, J = 14.2 & 2.2 Hz,
1H), 2.48 (dt, J = 12.4 & 4.4 Hz, 1H), 2.41 (dd, J = 8.4 & 7.2 Hz, 2H), 2.04 (q, J = 12.4 Hz, 1H), 1.45 (m, 1H),
1.36 (m, 1H), 1.28 – 1.20 (m, 2H), 1.10 – 0.97 (m, 2H), 0.87 – 0.78 (m, 2H) (53H attributed + 5 NH2 = 63H).
HRMS (ESI+, m/z, MeOH): calcd. for C51H65N5O6 [M+ 2H]2+ 421.7462, found 421.7473. El. anal. calcd. for
C51H63N5O6•5.05CF3COOH•0.35H2O: C, 51.53; H, 4.87; N, 4.92; found: C, 51.66; H, 5.00; N, 4.79.
NeaCyclam. 50 % yield (White powder) 1H NMR (500 MHz, MeOD-d4)
δ ppm 7.90 – 7.82 (m, 6H), 7.81 – 7.75
(m, 4H), 7.73 (br, 2H), 7.55 – 7.47 (m, 7H), 7.42 (d, J = 8.6 Hz, 1H), 7.37 (d, J = 8.5 Hz, 1H), 5.64 (d, J = 2.1 Hz,
1H), 5.08 (d, J = 11.7 Hz, 1H), 5.01 (d, J = 11.7 Hz, 1H), 4.85 (m, 1H), 4.77 (m, 1H), 4.82 (s, 2H), 4.62 (d, J =
9.9 Hz, 1H), 4.28 (dd, J = 5.9 & 3.6 Hz, 1H), 4.24 (dd, J = 10.5 & 9.3 Hz, 1H), 3.72 – 3.78 (m, 4 H), 3.68 – 3.70
(m, 2H), 3.58 – 3.60 (m, 3H), 3.43 – 3.46 (m, 3H), 3.40 (m, 1H), 3.30 (m, 2H), 3.20 (m, 2H), 3.14 – 3.17 (m, 4H),
3.11 (m, 1H), 2.88 (m, 2H), 2.84 (m, 2H), 2.78 (m, 2H), 2.50 (m, 1H), 2.34 (m, 2H), 2.15 (t, J = 7.2 Hz, 2H), 2.05
(q, J = 12.4 Hz, 1H), 2.00 (m, 2H), 1.82 – 1.85 (m, 4H), 1.54 – 1.40 (m, 2H), 1.21 – 1.14 (m, 2H), 1.13 – 1.06 (m,
2H), 0.97 (m, 2H) (79H attributed + 4 NH2 + 1 NHCO + 3 NH Cyclam = 91H) . 13C NMR (125.7 MHz, MeOD-
d4)
δ ppm 22.63, 26.30, 26.86, 27.99, 29.61, 29.63, 30.36, 31.61, 33.85, 39.93, 40.35, 45.86, 46.80, 47.28, 48.64,
50.54, 50.92, 51.41, 73.55, 73.85, 74.66, 74.72, 74.74, 75.55, 80.04, 81.70, 84.83, 93.36, 117.18, 119.51, 126-130
(C not listed), 134.63, 134.80, 134.84, 134.89, 134.94, 163.20, 163.48. HRMS (ESI+, m/z, MeOH) calcd. for
C66H93N9O8 [M+2H]2+ 569.8568, found 569.8562. El. anal. calcd. for C66H91N9O8•7.2CF3COOH•2.5H2O: C,
48.18; H, 5.19; N, 6.29; found: C, 48.25; H, 5.41; N, 6.14.
NeaTPA. 53% yield (white powder) 1H NMR (500 MHz, MeOD-d4): δ ppm 8.61 (d, J = 5.1 Hz, 2H), 7.90 – 7.83
(m, 6H), 7.82 (m, 1H), 7.80 – 7.76 (m, 6H), 7.74 – 7.71 (m, 2H), 7.61 (dd, J = 8.1 & 7.5 Hz, 1H), 7.53 – 7.49 (m,
5H), 7.47 (m, 2H), 7.46 – 7.44 (m, 2H), 7.42 (d, J = 8.5 Hz, 1H), 7.41 (dd, J = 7.8 & 5.1 Hz, 2H), 7.36 (d, J = 8.5
Hz, 1H), 7.09 (d, J = 7.4 Hz, 1H), 5.66 (d, J = 2.2 Hz, 1H), 5.07 (d, J = 11.6 Hz, 1H), 4.99 (d, J = 11.6 Hz, 1H),
4.85 (d, J = 11.5 Hz, 1H), 4.82 (s, 2H), 4.77 (d, J = 11.5 Hz, 1H), 4.64 (ddd, J = 10.4, 3.3 & 2.1 Hz, 1H), 4.31 (m,
5H), 4.24 (dd, J = 10.2 & 9.1 Hz, 1H), 4.11 (s, 2H), 3.78 (t, J = 9.1 Hz, 1H), 3.73 – 3.77 (m, 2H), 3.72 (m, 1H),
3.71 (dd, J = 13.9 & 10.4 Hz, 1H), 3.68 (m, 1H), 3.59 (t, J = 9.1 Hz, 1H), 3.45 (ddd, J = 12.5, 10.2 & 4.2 Hz, 1H),

ACCEPTED MANUSCRIPT
3.41 (ddd, J = 12.5, 9.1 & 4.2 Hz, 1H), 3.13 (dd, J = 13.9 & 2.1 Hz, 1H), 2.94 (m, 1H), 2.83 (m, 1H), 2.50 (dt, J =
12.5 & 4.2 Hz, 1H), 2.41 (m, 2H), 2.19 (t, J = 7.7 Hz, 2H), 2.09 (q, J = 12.5 Hz, 1H), 1.93 (m, 2H), 1.52 – 1.39
(m, 2H), 1.20 (m, 1H), 1.12 – 1.21 (m, 2H), 1.06 (m, 1H), 0.98 (m, 2H) (76H attributed + 4 NH2 + 2 NH = 86H).
13C NMR (125.7 MHz, MeOD-d4)
δ ppm 23.03, 26.88, 27.82, 29.87, 30.29, 31.50, 36.45, 37.03, 39.69, 40.00,
48.34, 48.51, 48.64, 50.48, 50.81, 51.57, 60.50, 60.85, 73.85, 74.84, 74.99, 76.42, 81.78, 85.25, 124.55, 125.53,
126.71, 126.92, 127.33, 127.51, 127.54, 127.61, 127.77, 128.41, 128.6-129.8 (C not listed), 134.71, 134.80,
134.85, 135.43, 135.53, 136.51, 139.6, 140.24, 141.67, 175.22. HRMS (ESI+, m/z, MeOH): calcd. for
C77H88N10O8 [M+2H]2+ 622.3388, found 622.3403. El. anal. calcd. for C74H86N10O8•5.8CF3COOH•2.5H2O: C,
52.73; H, 5.00; N, 7.18; found: C, 52.98; H, 5.03; N, 7.16.
NeaTBA. 74% yield (white powder) 1H NMR (500 MHz, MeOD-d4)
δ ppm 8.58 (d, J = 7.1 Hz, 1H), 7.91 – 7.70
(m, 18H), 7.56 – 7.32 (m, 16H), 5.62 (d, J = 2.0 Hz, 1H), 5.07 (d, J = 11.7 Hz, 1H), 5.01 (d, J = 11.7 Hz, 1H),
4.82 (m, 1H), 4.79 (m, 2H), 4.76 (m, 1H), 4.60 (d, J = 10.4 Hz, 1H), 4.41 (br, 6H), 4.25 (t, J = 3.9 Hz, 1H), 4.21
(dd, J = 10.4 & 9.2 Hz, 1H), 3.79 – 3.72 (m, 2H), 3.68 – 3.71 (m, 2H), 3.65 (m, 1H), 3.64 (dd, J = 13.5 & 10.4
Hz, 1H), 3.58 (t, J = 9.2 Hz, 1H), 3.43 (m, 1H), 3.38 (m, 1H), 3.12 (m, 1H), 2.83 (t, J = 7.2 Hz, 2H), 2.46 (dt, J =
12.5 & 4.2 Hz, 1H), 2.08 (t, J = 7.3 Hz, 2H), 2.04 (q, J = 12.5 Hz, 1H), 1.72 (m, 2H), 1.54 (q, 7.3 Hz, 2H), 1.46
(m, 1H), 1.40 (m, 1H), 1.15 (m, 2H), 1.08 (m, 1H), 1.04 (m, 1H), 0.98 (m, 2H) (79H attributed + 4 NH2 + 2 NH =
89H). 13C NMR (125.7 MHz, MeOD-d4)
δ ppm 23.33, 26.71, 27.78, 29.03, 30.32, 30.86, 31.48, 36.46, 37.16,
40.04, 58.77, 58.86, 73.98, 74.80, 75.10, 76.35, 126-131 (C not listed), 134.64, 134.75, 134.79, 134.80, 134.85,
135.51, 136.61, 140.23, 173.97, 175.3. HRMS (ESI+, m/z, MeOH): calcd. for C77H91N7O8 [M+2H]2+ 620.8459,
found 620.8452. El. anal. calcd. for C77H89N7O8•4.6CF3COOH•2H2O: C, 57.48; H, 5.46; N, 5.44; found: C, 57.30;
H, 5.46; N, 5.56.
NeaDPA. 70% yield (white powder) 1H NMR (500 MHz, MeOD-d4)
δ ppm 8.58 (d, J = 5.0 Hz, 2H), 7.90 – 7.83
(m, 6H), 7.82 (m, 2H), 7.80 – 7.76 (m, 4H), 7.74 – 7.71 (m, 2H), 7.53 – 7.47 (m, 7H), 7.46 – 7.44 (m, 2H), 7.42
(d, J = 8.5 Hz, 1H), 7.38 (dd, J = 7.6 & 5.0 Hz, 2H), 7.36 (d, J = 8.5 Hz, 1H), 5.62 (d, J = 2.1 Hz, 1H), 5.07 (d, J
= 11.7 Hz, 1H), 5.01 (d, J = 11.7 Hz, 1H), 4.82 (m, 1H), 4.79 (m, 2H), 4.75 (m, 1H), 4.59 (m, 1H), 4.41 (s, 4H),
4.25 (t, J = 3.8 Hz, 1H), 4.21 (dd, J = 10.5 & 9.3 Hz, 1H), 3.78 – 3.73 (m, 2H), 3.70 (t, J = 9.2 Hz, 1H), 3.69 (m,
1H), 3.66 (m, 1H), 3.64 (dd, J = 13.9 & 10.5 Hz, 1H), 3.58 (t, J = 9.3 Hz, 1H), 3.42 (m, 1H), 3.38 (m, 1H), 3.12 –
3.09 (m, 2H), 3.13 (ddd, J = 13.9 & 1.9 Hz, 1H), 2.83 (t, J = 7.5 Hz, 2H), 2.46 (dt, J = 12.5 & 4.5 Hz, 1H), 2.08
(t, J = 7.3 Hz, 2H), 2.03 (q, J = 12.5 Hz, 1H), 1.71 (m, 2H), 1.54 (m, 2H), 1.46 (m, 1H), 1.37 (m, 1H), 1.20 – 1.12
(m, 2H), 1.08 – 1.02 (m, 2H), 0.96 – 0.89 (m, 2H) (73H attributed + 4 NH2 + 1 NHCO = 82H). 13C NMR (125.7
MHz, MeOD-d4)
δ ppm 23.95, 25.43, 26.81, 27.84, 27.97, 29.71, 30.39, 30.42, 31.65, 36.19, 39.95, 40.35, 51.49,
51.64, 55.98, 59.04, 73.65, 73.88, 74.67, 74.86, 76.31, 76.39, 81.92, 84.86, 84.91, 93.93, 125.37, 125.45, 126-129
(C not listed), 134.84, 134.88, 135.51, 136.78, 139.40, 150.37, 150.42, 163.12, 163.35, 175.13. HRMS (ESI+,
m/z, MeOH): calcd. for C68H83O8N7 [M+H]+ 1123.6379, found 1123.6370. El. anal. calcd. for
C68H82N8O7•5.2CF3COOH•2.5H2O: C, 53.46; H, 5.28; N, 6.36; found: C, 53.48; H, 5.27; N, 6.36.
5.2 Biological studies

ACCEPTED MANUSCRIPT
Bacterial strains and growth
Two Escherichia coli (E. coli) strains and 2 Enterobacter aerogenes (E. aerogenes) strains were previously
described [43,46]. Briefly, AG100 was an E. coli wild type strain and AG100A its AcrAB- derivative. EA289 was
a Kans derivative of an E. aerogenes multi-drug resistant (MDR) clinical isolate overexpressing AcrAB-tolC
efflux pumps, and EA298 its tolC- derivative. Strains were routinely grown at 37°C on Luria-Bertani agar or in
MH broth, supplemented with kanamycin (50 µg.mL-1) for AG100A and EA298.
Susceptibility determination
The minimal inhibitory concentrations (MICs) were determined by broth dilution method as previously described
[36]. The different molecules were solubilized in DMSO, and under the used conditions, the final percentage of
DMSO (≤ 0.5%) had no effect on the bacterial growth [11,12]. Susceptibilities were determined in 96-wells
microplates with an inoculum of 2x105 cfu in 200 µL of MH broth containing two-fold serial dilutions of each
compounds. MICs were realized in the absence and in the presence of PMBN membrane permeabilizer, used at
51.2 mg/L (1/5 of its direct MIC previously determined). MIC was defined as the lowest concentration of each
compound for which no visible growth was observed after 18 h of incubation at 37°C. MIC values were means of
at least three independent experiments. Results were expressed in µM in order to have a direct comparison of
biological activity.
Nitrocefin assay[31]
Bacteria were grown to the exponential phase (OD at 600 nm = 0.5), centrifuged 20 min at 4000 rpm at 20°C and
washed in PPB buffer (K2HPO44 20 mM ; MgCl21mM; pH 7). The OD of the bacterial culture was then adjusted
to 0.375 and 100 µL was added to 50 µL of two-fold serial dilutions of each molecule in PPB buffer at a final
concentration of 2, 4, 8, 16, 32 and 64 µM. Tests were performed at 37°C in microplates of 96 wells. Carbonyl
cyanide m-chlorophenylhydrazone (CCCP) was added at 3.3 µM to prevent the efflux of molecules. Nitrocefin
(50 µL at 0.2 mg / mL) was added at t = 0 and the OD at 490 nm was monitored for 1 h with the microplates
lector TECAN infinite M200 Pro. Each test was performed in duplicate or triplicate.
Real time efflux assay [35]
After an incubation of 24 h at 37°C, bacteria were centrifuged for 20 min at 4000 rpm and 20°C; and washed in
PPB buffer. The pellet was resuspended in PPB buffer and the OD at 600 nm was adjusted to 0.25. The
fluorophore 1,2'-DiNaphtylAmine (1,2'-DNA) and the dissipater of energy Carbonyl Cyanide 3-
ChloroPhenylhydrazone (CCCP) were added at 32 and 5 µM respectively. After an incubation time of 18 h at
37°C protected from light, the bacterial culture was washed in PPB and 99 µL were added to 1 µL of two-fold
serial dilutions of each molecule at a final concentration of 0.5 to 128 µM. Tests were performed at 37°C in black
flat bottom microplates of 96 half-wells (Greiner Bio-One). The fluorescence intensity (λex 370 nm / λem 420
nm) was measured every 4 s during 6 min with the microplates lector TECAN infinite M200 Pro. Five µL of
glucose at 1 M was added at cycle 20 (after about 100 s). Each test was performed in duplicate or triplicate.
Acknowledgements

ACCEPTED MANUSCRIPT
We thank Assia Hessani for mass spectrometry analyses and Anne-Marie Tran for participation in bacteriology
support.
Funding
The research leading to the results discussed here was conducted as part of the translocation consortium
(www.translocation.eu) and has received support from the Innovative Medicines Initiative joint Undertaking
under Grant Agreement n°115525, resources which are composed of financial contribution from the European
Union’s seventh framework programme (FP/2007-2013) and EFPIA companies in kind contributions.
CNRS, Paris Descartes and Aix-Marseille Universities, IRBA are gratefully acknowledged.
Abbreviations
HOAt, 1-hydroxy-7azabenzotriazole; HATU, 1-[Bis(dimethylamino)methylene]-1H-1,2,3-triazolo[4,5-
b]pyridinium-3-oxid hexafluorophosphate. DIEA diisopropylethylamine. TBAF tetrabutylammonium fluoride
Supplementary data
Synthesis of intermediates – Scheme S1 synthesis of protected Cyclam - ESI+ HRMRS of free ligands and
complexes in MeOH and MH medium: Figure S1 NeaDPA and ZnII(NeaDPA); Figure S2 NeaCyclam and
ZnII(NeaCyclam) – Structure of the complexes Scheme S2 –Table S1 analysis of the mass spectra - Microbiology
: Figure S3 nitrocefin hydrolysis in the presence of NeaTBA, NeaNH2 and PMBN - Figure S4 : time-dependent
1,2’-DNA efflux curves of E. aerogenes EA289 in the presence of NeaTBA and PAβN - Table S2 : calculated
1
pKa and logD of neamine derivatives - H NMR of final free ligands: NeaN3; NeaNH2; NeaTPA; NeaDPA;
NeaCyclam. This material is available free of charge at…...
References
[1]
R.A. Stavenger, M. Winterhalter, Translocation project: how to get good drugs into bad bugs, Sci Transl.
Med. 6 (2014) 228ed7.
[2]
R. Laxminarayan, A. Duse, C. Wattal, A.K.M. Zaidi, H.F.L. Wertheim, N. Sumpradit, E. Vlieghe, G.L.
Hara, I.M. Gould, H. Goossens, C. Greko, A.D. So, M. Bigdeli, G. Tomson, W. Woodhouse, E.
Ombaka, A.Q. Peralta, F.N. Qamar, F. Mir, S. Kariuki, Z.A. Bhutta, A. Coates, R. Bergstrom, G.D.
Wright, E.D. Brown, O. Cars, Antibiotic resistance - the need for global solutions, Lancet Infect. Dis. 13
(2013) 1057–1098.
[3]
[4]
J.-M. Pagès, C.E. James, M. Winterhalter, The porin and the permeating antibiotic: a selective diffusion
barrier in Gram-negative bacteria, Nat. Rev. Microbiol. 6 (2008) 893–903.
H. Nikaido, J.-M. Pagès, Broad-specificity efflux pumps and their role in multidrug resistance of Gram-
negative bacteria, FEMS Microbiol. Rev. 36 (2012) 340–363.
[5]
[6]
M. Zasloff, Antimicrobial peptides of multicellular organisms, Nature 415 (2002) 389–395.
R.E.W. Hancock, H.-G. Sahl, Antimicrobial and host-defense peptides as new anti-infective therapeutic
strategies, Nat. Biotechnol. 24 (2006) 1551–1557.
[7]
[8]
C.D. Fjell, J.A. Hiss, R.E.W. Hancock, G. Schneider, Designing antimicrobial peptides: form follows
function, Nat. Rev. Drug Discov. 11 (2012) 37–51.
M.D. Libardo, J.L. Cervantes, J.C. Salazar, A.M. Angeles-Boza, Improved bioactivity of antimicrobial
peptides by addition of amino-terminal copper and nickel (ATCUN) binding motifs, ChemMedChem. 9
(2014) 1892–1901.
[9]
M.D.J. Libardo, S. Nagella, A. Lugo, S. Pierce, A.M. Angeles-Boza, Copper-binding tripeptide motif

ACCEPTED MANUSCRIPT
increases potency of the antimicrobial peptide anoplin via reactive oxygen species generation, Biochem.
Biophys. Res. Commun. 456 (2015) 446–451.
[10]
[11]
J.C. Joyner, W.F. Hodnick, A.S. Cowan, D. Tamuly, R. Boyd, J.A. Cowan, Antimicrobial
metallopeptides with broad nuclease and ribonuclease activity, Chem. Commun. 49 (2013) 2118–2120.
J.-M. Pagès, S. Kascàkovà, L. Maigre, A. Allam, M. Alimi, J. Chevalier, E. Galardon, M. Réfrégiers, I.
Artaud, New Peptide-based antimicrobials for tackling drug resistance in bacteria: single-cell
fluorescence imaging, ACS Med. Chem. Lett. 4 (2013) 556–559.
[12]
[13]
A. Allam, L. Maigre, M. Alimi, R. Alves de Sousa, A. Hessani, E. Galardon, J.-M. Pagès, I. Artaud,
New peptides with metal binding abilities and their use as drug carriers, Bioconjug. Chem. 25 (2014)
1811–1819.
S. Bera, G.G. Zhanel, F. Schweizer, Antibacterial activities of aminoglycoside antibiotics-derived
cationic amphiphiles. Polyol-modified neomycin B-, kanamycin A-, amikacin-, and neamine-based
amphiphiles with potent broad spectrum antibacterial activity, J. Med. Chem. 53 (2010) 3626–3631.
C.W.T. Chang, J.Y. Takemoto, Antifungal amphiphilic aminoglycosides, MedChemComm. 5 (2014)
1048–1057.
G. Sautrey, L. Zimmermann, M. Deleu, A. Delbar, L. Souza Machado, K. Jeannot, F. Van Bambeke,
J.M. Buyck, J.-L. Décout, M.-P. Mingeot-Leclercq, New amphiphilic neamine derivatives active against
resistant Pseudomonas aeruginosa and their interactions with lipopolysaccharides, Antimicrob. Agents
Chemother. 58 (2014) 4420–4430.
[14]
[15]
[16]
[17]
[18]
[19]
[20]
[21]
L. Zimmermann, A. Bussière, M. Ouberai, I. Baussanne, C. Jolivalt, M.-P. Mingeot-Leclercq, J.-L.
Décout, Tuning the antibacterial activity of amphiphilic neamine derivatives and comparison to
paromamine homologues, J. Med. Chem. 56 (2013) 7691–7705.
M. Ouberai, F. El Garch, A. Bussière, M. Riou, D. Alsteens, L. Lins, I. Baussanne, Y.F. Dufrêne, R.
Brasseur, J.-L. Décout, M.-P. Mingeot-Leclercq, The Pseudomonas aeruginosa membranes: a target for
a new amphiphilic aminoglycoside derivative? Biochim. Biophys. Acta. 1808 (2011) 1716–1727.
I. Baussanne, A. Bussière, S. Halder, C. Ganem-Elbaz, M. Ouberai, M. Riou, J.-M. Paris, E. Ennifar,
M.-P. Mingeot-Leclercq, J.-L. Décout, Synthesis and antimicrobial evaluation of amphiphilic neamine
derivatives, J. Med. Chem. 53 (2009) 119–127.
I.M. Herzog, K.D. Green, Y. Berkov-Zrihen, M. Feldman, R.R. Vidavski, A. Eldar-Boock, R. Satchi-
Fainaro, A. Eldar, S. Garneau-Tsodikova, M. Fridman 6"-Thioether tobramycin analogues: towards
selective targeting of bacterial membranes, Angew. Chem. Int. Ed. Engl. 51 (2012) 5652–5656.
O. Jackowski, A. Bussière, C. Vanhaverbeke, I. Baussanne, E. Peyrin, M.-P. Mingeot-Leclercq, J.-L.
Décout, Major increases of the reactivity and selectivity in aminoglycoside O-alkylation due to the
presence of fluoride ions, Tetrahedron 68 (2012) 737–746.
L. Fabbrizzi, F. Foti, M. Licchelli, P.M. Maccarini, D. Sacchi, M. Zema, Light-emitting molecular
machines: pH-induced intramolecular motions in a fluorescent nickel(II) scorpionate complex, Chem.
Eur. J. 8 (2002) 4965–4972.
[22]
[23]
[24]
[25]
K.D. Mjos, C. Orvig, Metallodrugs in medicinal inorganic chemistry, Chem. Rev. 114 (2014) 4540–
4563.
N.P.E. Barry, P.J. Sadler, Exploration of the medical periodic table: towards new targets, Chem.
Commun. 49 (2013) 5106–5131.
N. Graf, S.J. Lippard, Redox activation of metal-based prodrugs as a strategy for drug delivery, Adv.
Drug Deliv. Rev. 64 (2012) 993–1004.
G. Anderegg, E. Hubmann, N.G. Podder, F. Wenk, Pyridinderivate als komplexbildner. XI. Die
thermodynamik der metallkomplexbildung mit bis-, tris- und tetrakis-[(2‐pyridyl)methyl]‐aminen, Helv.
Chim. Acta. 60 (1977) 123–140.
[26]
[27]
C. Schickaneder, F.W. Heinemann, R. Alsfasser, Copper(II) complexes of the tetraazamacrocyclic
tertiary amide ligand alanyl cyclam, Eur. J. Inorg. Chem. (2006) 2357–2363.
‐
R. Girardello, P.J.M. Bispo, T.M. Yamanaka, A.C. Gales, Cation concentration variability of four
distinct Mueller-Hinton agar brands influences polymyxin B susceptibility results, J. Clin. Microbiol. 50
(2012) 2414–2418.
[28]
[29]
C. Fernández-Mazarrasa, O. Mazarrasa, J. Calvo, A. del Arco, L. Martínez-Martínez, High
concentrations of manganese in Mueller-Hinton agar increase MICs of tigecycline determined by Etest,
J. Clin. Microbiol. 47 (2009) 827–829.
A.Z. Sahalan, R.A. Dixon, Role of the cell envelope in the antibacterial activities of polymyxin B and

ACCEPTED MANUSCRIPT
polymyxin B nonapeptide against Escherichia coli, Int. J. Antimicrob. Agents. 31 (2008) 224–227.
[30]
[31]
[32]
J.-M. Pagès, S. Peslier, T.A. Keating, J.-P. Lavigne, W.W. Nichols, Role of the outer membrane and
porins in susceptibility of
β-lactamase-producing Enterobacteriaceae to ceftazidime-avibactam,
Antimicrob. Agents Chemother. 60 (2015) 1349–1359.
Y. Matsumoto, K. Hayama, S. Sakakihara, K. Nishino, H. Noji, R. Iino, A. Yamaguchi, Evaluation of
multidrug efflux pump inhibitors by a new method using microfluidic channels, PLoS ONE. 6 (2011)
e18547.
A. Lieutaud, E. Guinoiseau, V. Lorenzi, M.C. Giuliani, V. Lome, J.-M. Brunel, A. Luccani, J. Casanova,
J.-M. Pagès, L. Berti, M. Bolla, Inhibitors of antibiotic efflux by AcrAB-TolC in Enterobacter
aerogenes, Anti-Infect. Agents 11 (2013) 168-178.
[33]
[34]
[35]
R. Daugelavicius, E. Bakiene, D.H. Bamford, Stages of polymyxin B interaction with the Escherichia
coli cell envelope, Antimicrob. Agents Chemother. 44 (2000) 2969–2978.
T. Velkov, P.E. Thompson, R.L. Nation, J. Li, Structure-activity relationships of polymyxin antibiotics,
J. Med. Chem. 53 (2010) 1898–1916.
J.A. Bohnert, S. Schuster, M. Szymaniak-Vits, W.V. Kern, Determination of real-time efflux phenotypes
in Escherichia coli AcrB binding pocket phenylalanine mutants using a 1,2'-dinaphthylamine efflux
assay, PLoS ONE. 6 (2011) e21196.
[36]
M. Malléa, J. Chevalier, C. Bornet, A. Eyraud, A. Davin-Regli, C. Bollet, J.-M. Pagès, Porin alteration
and active efflux: two in vivo drug resistance strategies used by Enterobacter aerogenes, Microbiol. 144
(Pt 11) (1998) 3003–3009.
[37]
[38]
[39]
L.J.V. Piddock, Clinically relevant chromosomally encoded multidrug resistance efflux pumps in
bacteria, Clin. Microbiol. Rev. 19 (2006) 382–402.
J.-M. Pagès, M. Masi, J. Barbe, Inhibitors of efflux pumps in Gram-negative bacteria, Trends Mol.
Med. 11 (2005) 382–389.
K.M. DiVittorio, W.M. Leevy, E.J. O'Neil, J.R. Johnson, S. Vakulenko, J.D. Morris, K.D. Rosek, N.
Serazin, S. Hilkert, S. Jurley, M. Marquez, B.D. Smith, Zinc(II) coordination complexes as membrane-
active fluorescent probes and antibiotics, ChemBioChem. 9 (2008) 286–293.
W.M. Leevy, J.R. Johnson, C. Lakshmi, J. Morris, M. Marquez, B.D. Smith, Selective recognition of
bacterial membranes by zinc(II)-coordination complexes, Chem. Commun. (Camb.). (2006) 1595–1597.
M.B. Spesia, D.A. Caminos, P. Pons, E.N. Durantini, Mechanistic insight of the photodynamic
inactivation of Escherichia coli by a tetracationic zinc(II) phthalocyanine derivative, Photodiag.
Photodyn. Ther. 6 (2009) 52–61.
[40]
[41]
[42]
[43]
A. Minnock, D.I. Vernon, J. Schofield, J. Griffiths, J.H. Parish, S.B. Brown, Mechanism of uptake of a
cationic water-soluble pyridinium zinc phthalocyanine across the outer membrane of Escherichia coli,
Antimicrob. Agents Chemother. 44 (2000) 522–527.
M. Viveiros, A. Jesus, M. Brito, C. Leandro, M. Martins, D. Ordway, A.M. Molnar, J. Molnar, L.
Amaral, Inducement and reversal of tetracycline resistance in Escherichia coli K-12 and expression of
proton gradient-dependent multidrug efflux pump genes, Antimicrob. Agents Chemother. 49 (2005)
3578–3582.
[44]
W. K. C. Park, M. Auer, A. H. Jaksche, C.-H. Wong, Rapid combinatorial synthesis of aminoglycoside
antibiotic mimetics:ꢀuse of a polyethylene glycol-linked amine and a neamine-derived aldehyde in
multiple component condensation as a strategy for the discovery of new inhibitors of the HIV RNA Rev
Responsive Element, J. Am. Chem. Soc. 118 (1996) 10150-10155.
[45]
[46]
E. Riguet, J. Désiré, C. Bailly, J.-L. Décout, A route for preparing new neamine derivatives targeting
HIV-1 TAR RNA, Tetrahedron 60 (2004) 8053–8064.
E. Pradel, J.-M. Pagès, The AcrAB-TolC efflux pump contributes to multidrug resistance in the
nosocomial pathogen Enterobacter aerogenes, Antimicrob. Agents Chemother. 46 (2002) 2640–2643.

ACCEPTED MANUSCRIPT
Scheme 1. Synthesis of neamine conjugates with metal binding groups
.
(i) 1,6-dibromohexane, NaH, DMF, 60°C, 20 h (ii) NaN3, DMF, r.t., 20 h (iii) TFA/CH2Cl2 (1:1), 0 °C, r.t., 2 h (iv) PPh3,
THF/water (95:5; v:v), reflux, 5 h. (v) HOAT, HATU, DIEA, DMF, r.t., 3 h.

ACCEPTED MANUSCRIPT
Figure 1. ESI+ Mass spectra of NeaTPA and Zn(NeaTPA) complex in MeOH with or without further
incubation in MH.
(a) HRMS of NeaTPA 50 µM in MeOH. (b) HRMS of NeaTPA 50 µM in MeOH after incubation in MH. (c) HRMS of
Zn(NeaTPA) complex prepared from NeaTPA 50 µM and Zn(ClO4)2 500 µM with 400 µM of DIPEA in MeOH. (d) HRMS
of Zn(NeaTPA) complex prepared in MeOH and incubated at 50 µM in MH.

ACCEPTED MANUSCRIPT
Figure 2: percentage of nitrocefin hydrolysis in E. aerogenes EA289
Percentage of the ODmax expected at 490 nm for complete hydrolysis of 50 µg/mL Nitrocefin for 1h - Curves at
0, 2, 4, 8, 16, 32 µM of neamine conjugates and PMB up to 48 µM.

ACCEPTED MANUSCRIPT
Figure 3: Representative 1,2'-DNA efflux curves of E. aerogenes EA289
Each neamine conjugate, as well as CCCP, was added at a final concentration of 0.5 to 16 or 32 µM.
Tests were performed at 37°C. The fluorescence intensity (λex 370 nm / λem 420 nm) was measured
every 4 s during 5 min.

ACCEPTED MANUSCRIPT
CIP
FLE
0.5
CHL
8
CAZ
0.5
Neamine
NeaN3 NeaNH2
NeaCyclam NeaTPA
NeaTBA
NeaDPA
AG100
0.125
32-16
(8)
32
8
4
4-2
128
16-8
(4)
(32)
(8)
(2)
(4-2)
(64-32)
AG100A 0.03
0.125
>128
16
1
0.25
128
32
8
2
2
128-32
(32)
16-8
(4-2)
(128)
(16)
(4)
(2-1)
(2)
EA289
EA298
64
8
>128
32
>128
>128
32
16
16
16-4
(8)
4
128
16
(16)
(32-8) (32-16)
(8-4)
(128)
(8)
> 128
16
4
4-2
(2)
8-2
(2)
128-64
(64-32)
8
(> 128)
(32)
(8-4)
(8-4)
in parentheses values in the presence of PMBN 51.2 µg mL-1
Table 1. Minimal Inhibitory concentrations in µM of neamine conjugates and tested
antibiotics: CIP, ciprofloxacin; FLE, fleroxacin; CHL, chloramphenicol; CAZ, ceftazidime.

ACCEPTED MANUSCRIPT
Highlights
• Synthesis of new amphiphilic neamine conjugates bearing a metal binding motif.
• Evaluation of their specific metal binding ability in the Mueller Hinton medium.
• Significant antibacterial activity against a multiresistant clinical isolate compared to usual antibiotic
• Determination of their membranotropic action: permeabilization of the outer membrane and action
on the efflux transporter located in the inner membrane.