Neomycin-phenolic conjugates: Polycationic amphiphiles with broad-spectrum antibacterial activity, low hemolytic activity and weak serum protein binding

Article abstract of DOI:10.1016/j.bmcl.2012.01.025

Here we present a proof-of-concept study, combining two known antimicrobial agents into a hybrid structure in order to develop an emergent cationic detergent-like interaction with the bacterial membrane. Six amphiphilic conjugates were prepared by copper

Full text of DOI:10.1016/j.bmcl.2012.01.025

Bioorganic & Medicinal Chemistry Letters 22 (2012) 1499–1503
Contents lists available at SciVerse ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Neomycin–phenolic conjugates: Polycationic amphiphiles
with broad-spectrum antibacterial activity, low hemolytic
activity and weak serum protein binding
Brandon Findlay a, George G. Zhanel b, Frank Schweizer a,b,
⇑
a Department of Chemistry, University of Manitoba, Winnipeg, MB, Canada R3T 2N2
b Department of Medical Microbiology, University of Manitoba, Health Sciences Centre MS673-820 Sherbrook St., Winnipeg, MB, Canada R3A 1R9
a r t i c l e i n f o
a b s t r a c t
Article history:
Here we present a proof-of-concept study, combining two known antimicrobial agents into a hybrid
structure in order to develop an emergent cationic detergent-like interaction with the bacterial mem-
brane. Six amphiphilic conjugates were prepared by copper (I)-catalyzed 1,3-dipolar cycloaddition
between a neomycin B-derived azide and three alkyne-modified phenolic disinfectants. Three conjugates
displayed good activity against a variety of clinically relevant Gram positive and Gram negative bacteria,
including MRSA, without the high level of hemolysis or strong binding to serum proteins commonly
observed with other cationic antimicrobial peptides and detergents.
Received 5 December 2011
Revised 4 January 2012
Accepted 9 January 2012
Available online 14 January 2012
Keywords:
Aminoglycosides
Quaternary ammonium chlorides
Cationic amphiphiles
Huisgen condensation
Phenolic disinfectants
Ó 2012 Elsevier Ltd. All rights reserved.
Over the years, many first-line antibiotics have been relegated
to the back benches as an increased presence of drug resistant bac-
teria rendered them ineffective. At the same time, a reduced focus
on antibiotic research at the major pharmaceutical companies has
drastically reduced the rate of drug discovery, leaving us more in
need of new antimicrobial agents and scaffolds than ever.1,2 The
need for new scaffolds is especially great, as widespread resistance
to most antibiotics appears shortly after the drugs are introduced
into clinical use, with a gap of roughly twenty years for penicillins
and less than nine years for fluoroquinolones (five years for cipro-
floxacin).3,4 This resistance may affect many drugs with similar
scaffolds, limiting the effectiveness of new drugs before they even
enter the clinic. Resistance seems to arise from small numbers of
bacteria already present in the population at large. Attempts to
determine the age of common resistance mechanisms has found
that, like the secondary metabolites many current antibiotics are
based upon, the genes which code for antibiotic resistance are
ancient.5 The potential for widespread drug resistance is thus
latent in every bacterial population, and limiting its emergence
will require improved education in the use of antibiotics and the
creation of antibiotic classes that have been designed with antimi-
crobial resistance in mind.
inhibiting their function. Bacterial resistance arises from mecha-
nisms which disrupt the drug–target complex, either through
modification of the binding site (via DNA mutation or chemical
alteration), the drug (via acetylases, phosphates and others) or by
simply preventing the drug from entering the cell and encounter-
ing its target (efflux pumps).5,6 Drug efflux is of particular concern,
Most commercial antibiotics are molecular inhibitors, binding
to enzymes, cellular receptors or nucleic acids within the cell and
Figure 1. Benzethonium chloride, 1, served as the initial template for the hybrids
presented in this Letter. Neomycin B, 2, was used for its cationic charges, RNA-
binding properties and self-promoted uptake, whereas the phenolics chloroxylenol,
3, triclosan, 4, and clofoctol, 5, are expected to induce hydrophobic membrane
interactions in the hybrids.
⇑
Corresponding author.
E-mail address: schweize@ms.umanitoba.ca (F. Schweizer).
0960-894X/$ - see front matter Ó 2012 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2012.01.025

1500
B. Findlay et al. / Bioorg. Med. Chem. Lett. 22 (2012) 1499–1503
surfactants to act as antimicrobial agents.8–16 Found throughout
nature and long used as antimicrobial detergents, cationic amphi-
philes do not act on any single target within the cell, instead they
disrupt DNA replication, protein synthesis and bacterial membrane
integrity.12 Widely varying in structure and size, all CAs are based
on two common features: a hydrophobic face that interacts with
the lipid bilayer and a polar, cationic face that is drawn via electro-
static interactions to anionic moieties such as some lipid head
groups and nucleic acids. Little in vitro resistance to these amphi-
philes has been observed, due to their multiple modes of action
and ability to form pores in the bacterial membrane, but their clin-
ical use has been severely limited due to issues with selectivity,
protease susceptibility and toxicity.17 The therapeutic ratio of
these amphiphiles has been improved, reducing their ability to lyse
red blood cells and increasing selectivity towards Gram positive
and Gram negative bacteria.18 Nevertheless, the presence of non-
specific binding to human serum proteins remains a major limita-
tion of these agents, resulting in loss of antibacterial activity in
vivo.19 Starting from the structure of an amphiphilic disinfectant,
the quaternary ammonium compound benzethonium chloride, 1 (
Fig. 1), we devised alternatives to the classical CAs, in order to cre-
ate agents with similar characteristics but devoid of their limita-
tions.20 This work has cumulated in neomycin–phenolic
conjugates presented here (Fig. 2).
Figure 2. Structures of the neomycin–phenol conjugates produced in this study.
Conjugates 6 and 7 are based upon chloroxylenol, conjugates 8 and 9 on triclosan
and 10 and 11 on clofoctol. All six conjugates use neomycin B to provide the
polycationic charge.
Our conjugates use two known antimicrobial agents to create
the cationic and hydrophobic faces required for interaction with
bacterial membranes. The hybrid molecules are intended to display
three distinct modes of action; one from each participating agent
and one from an emergent CA-like behavior from the superstruc-
ture itself. This triple mode-of-action should lead to broad spec-
trum activity and resilience against bacterial resistance, as
resistance to one agent will not alter susceptibility to the other half
of the conjugate or to the CA-like mode of action. In optimal cases
this will allow the conjugates to retain activity against even MDR
bacteria. This strategy has been attempted previously, but has been
hampered by the large size of the conjugates, which reduces diffu-
sion across the cellular membrane.21,22 To maintain permeability
we chose phenolic disinfectants as the hydrophobic segment, as
phenolics are known to derive at least part of their activity from
moving small cations across the bacterial membrane, which
requires rapid diffusion in and out of the cell.20 When combined
with the polycationic aminoglycoside neomycin, which is known
Scheme 1. Synthesis of the alkyne linkers. Conditions: (a) TsCl, KOH, Et2O (41–
64%). (b) 3–5, K2CO3, DMF (51–96%).
as the poor selectivity of efflux pumps can easily lead to broad anti-
biotic resistance, with a single pump effective against whole clas-
ses of antibiotics.7 Bacteria which endogenously express a large
number of drug efflux pumps, such as Pseudomonas aeruginosa,
are able to withstand most common antimicrobial agents and are
often the cause of multi-drug resistant (MDR) infections.
As a means of circumventing these resistance mechanisms,
work in our lab and others has explored the potential for (poly)cat-
ionic amphiphiles (CAs) such as antimicrobial peptides, lipids and
Scheme 2. Synthesis of conjugates 6–11. Conditions: (a) Boc2O (10 equiv), TEA/MeOH/H2O (61%). (b) TIPS-Cl (31 equiv), pyridine (41%). (c) NaN3 (10 equiv), DMF/H2O (94%).
(d) 16a–c, 17a–c (1.2 equiv), CuI (0.2 equiv), DIPEA (3 equiv), ACN (50–88%). (e) TFA/H2O (80–91%).

B. Findlay et al. / Bioorg. Med. Chem. Lett. 22 (2012) 1499–1503
1501
Table 1
Antimicrobial activitya and hemolysis of the conjugates and drug standards
Organism
Compound
1
2r
3
4
5s
6
7
8
9
10
11
S. aureusb
MRSAc
2
2
2
8
4
4
2
32
32
32
64
64
32
32
32
77.3
1
32
64
32
16
0.5
60.25
1
60.25
8
—
—
—
—
—
—
—
—
—
—
—
—
—
—
—
16
256
8
8
64
128
64
16
16
128
8
8
128
16
64
32
16
128
512
256
>512
512
4
32
128
16
16
64
16
256
64
4
8
4
2
16
8
64
16
16
64
128
64
>512
128
4
8
8
4
1
8
8
64
16
64
64
128
64
512
64
64
27.5
16
16
8
8
64
8
64
64
64
128
128
64
512
256
32
22.5
256
0.5
60.25
16
4
32
4
1
8
512
256
>512
64
0.25
0.69
MSSEd
MRSEe
E. faecalisf
E. faeciumg
S. pneumoniaeh
E .colii
256
256
128
256
256
256
512
256
128
128
256
—
16
128
60.25
1
1
>512
64
512
8
1
E. colij
16
64
E. colik
128
512
256
>512
>512
8
128
512
256
>512
>512
32
P. aeruginosal
P. aeruginosam
S. maltophilian
A. baumanniio
K. pneumoniaep
Hemolysisq
—
0.75
0.75
0.99
1.62
a
b
c
MIC90, reported in
ATCC 29213.
ATCC 33592.
81388 CANWARD 2008.
CAN-ICU 61589.
ATCC 29212.
ATCC 27270.
ATCC 49619.
ATCC 25922.
lg/mL.
d
e
f
g
h
i
j
CAN-ICU 61714.
CAN-ICU 63074.
ATCC 27853.
k
l
m
n
o
p
q
r
CAN-ICU 62308.
CAN-ICU 62584.
CAN-ICU 63169.
ATCC 13883.
Percent hemolysis at 100
Neomycin trisulfate hydrate.
lg/mL of compound.
s
Compound 5 was poorly soluble in water and its activity could not be accurately assessed.
to self-promote its own uptake into bacterial cells by disrupting
polysaccharide–cation interactions,23 we expected our conjugates
to display good diffusion kinetics, despite their larger size and high
number of hydrogen bond donors and acceptors. Moreover, the
RNA-binding motif of neomycin may induce intracellular modes
of antibacterial action in the conjugates.24
The single primary hydroxyl group of neomycin was selected as
the point of attachment as previous studies have shown that mod-
ifications at this position allow the aminoglycoside to retain anti-
bacterial activity and RNA-binding.14,31
Briefly, neomycin sulfate was treated with di-tert-butyl-dicar-
bonate in a mixture of methanol, water and triethylamine to pro-
tect the amino groups. Flash chromatography of crude 18 was
then used to separate Boc protected neomycin B from neomycin
C, and the primary hydroxyl moiety was activated using a large ex-
cess of triisopropylsulfonyl chloride (TIPS-Cl) to afford sulfonate
ester 19. Addition of sodium azide in a mixture of DMF and water
at 70 °C cleanly produced the required azide 20. The two halves of
the conjugate were then linked through copper (I) catalyzed
1,3-dipolar cycloaddition. Deblocking with trifluoroacetic acid
gave the neomycin–phenol conjugates 6–11.
Antibacterial activity was assessed using macrobroth dilution
assays according to standard CLSI methodology.32 Compound
activity was determined against a combination of reference strains
of Gram-positive and Gram-negative bacteria and clinically rele-
vant pathogens from the national surveillance CAN-ICU and CAN-
WARD studies.32,33 The inclusion of clinically relevant bacteria is
especially important in light of the rapid increase in antimicrobial
resistance with varied resistance mechanisms. Bacteria from cur-
rent hospital environments are far more likely to be resistant to
a variety of antibiotics and disinfectants with different chemical
structures and mechanism(s) of action, and testing with only labo-
ratory strains can produce misleadingly effective antimicrobial
activities. Full results are summarized in Tables 1 and 2.
Work with other CAs has found that altering the size and shape
of the hydrophobic domain can greatly influence antimicrobial
activity,12,25 and so we linked three distinct phenolic disinfectants,
3–5, to neomycin B to create the hybrid compounds 6–11 (Fig. 2).
While triclosan, 4, has been found to inhibit fatty acid synthesis by
blocking the key enzyme FabI,26 the targets of chloroxylenol, 3, and
clofoctol, 5, are unknown. It appears that as a class the phenols, like
CAs, have a number of cellular interactions but derive much of
their activity from interactions with the bacterial membrane.20
Easily ionized, the phenols seem to ferry small cations across the
bacterial membrane, dispersing the membrane polarization. While
we expect much of this activity to be inhibited by the ether linkage
used to connect these phenols to neomycin B, an analysis of the
structure of 1 suggests that interactions with the cellular mem-
brane will be maintained, allowing the hybrids to pass through
the membrane and maintain a high intracellular concentration.
To produce the conjugates the two antimicrobials were linked
via a copper (I)-catalyzed 1,3-dipolar cycloaddition reaction be-
tween the phenol-modified alkynes (Scheme 1),27,28 and the neo-
mycin-based azide 20 (Scheme 2). The length of the phenolic
alkyne linker was varied, and attached to the phenols by displace-
ment of the corresponding alkyne sulfonate esters.29 Azide-
functionalized neomycin B was prepared as a Boc-protected
derivative using previously established methodology.10,14,30
The most active hybrids were compounds 9 and 10, one of
which had triclosan as the phenolic with the longer of our two

1502
B. Findlay et al. / Bioorg. Med. Chem. Lett. 22 (2012) 1499–1503
Table 2
Antimicrobial activitya of conjugates and drug standards in the presence of bovine serum albumin (BSA)
Organism
Compound
1
2q
3
4
5r
6
7
8
9
10
11
S. aureusb
MRSAc
16
32
128
128
32
0.5
128
0.5
60.25
16
4
8
0.25
1
512
512
512
512
0.5
1
1
—
—
—
—
—
—
—
—
—
—
—
—
—
—
—
32
512
8
8
128
64
128
32
16
128
512
256
>512
>512
4
32
256
8
8
128
32
256
32
16
128
>512
256
>512
512
4
64
256
16
16
128
64
512
64
64
256
512
256
>512
>512
32
16
32
4
8
16
2
4
16
8
128
64
128
256
512
128
>512
256
64
16
32
4
4
64
16
64
64
128
256
512
256
>512
>512
32
MSSEd
MRSEe
1
2
E. faecalisf
E. faeciumg
S. pneumoniaeh
E. colii
>512
>512
>512
>512
>512
>512
>512
>512
>512
>512
>512
512
512
256
16
16
64
>512
>512
>512
128
32
32
32
128
16
8
128
512
256
>512
256
2
32
16
256
256
256
512
512
256
256
256
E. colij
E. colik
16
P. aeruginosal
P. aeruginosam
S. maltophilian
A. baumanniio
K. pneumoniaep
256
256
>512
32
60.25
a
b
c
MIC90, reported in
ATCC 29213.
ATCC 33592.
81388 CANWARD 2008.
CAN-ICU 61589.
ATCC 29212.
ATCC 27270.
ATCC 49619.
ATCC 25922.
lg/mL.
d
e
f
g
h
i
j
CAN-ICU 61714.
CAN-ICU 63074.
ATCC 27853.
k
l
m
n
o
p
q
r
CAN-ICU 62308.
CAN-ICU 62584.
CAN-ICU 63169.
ATCC 13883.
Neomycin trisulfate hydrate.
Compound 5 was poorly soluble in water and its activity could not be accurately assessed.
linkers, and the other which had clofoctol and a short linker. Both
molecules displayed similar or improved activity against the Gram
positive bacteria in our study, with an improved activity of 64 lg/
against P. aeruginosa, we observed little difference in activity. Our
cationic amphiphile standard, 1, in contrast had an average eight-
fold reduction in efficacy, with the median Gram positive MIC ris-
mL observed against the normally highly neomycin B resistant P.
aeruginosa strain CAN-ICU 62308. Unlike many other CAs, com-
pound 9 was not appreciably hemolytic at near MIC concentra-
tions. The optimal spacing between cationic and hydrophobic
domains appears to differ between hydrophobic domains and must
be determined on a case-by-case basis.
ing from 2 to 16
from 32 to 256
l
l
g/mL and the median Gram negative MIC moving
g/mL. This reduction was expected, given the
importance of hydrophobicity on nonspecific membrane interac-
tions. As they are not greatly inhibited by BSA, we infer that much
of the antimicrobial effect in the conjugates is from the action of
either neomycin B or the various phenolics. Interestingly, the
shorter neomycin–triclosan conjugate 9 retains potent activity in
the presence of BSA against two Escherichia coli strains (MIC 6 16)
while benzethonium chloride is only weakly active (MIC = 256) un-
der these conditions.
Further characterization of the conjugates’ mode of action can
be determined by examining the activities of triclosan. Triclosan
is known to possess several modes of action, with much of its
activity stemming from inhibition of FabI, a key component of fatty
acid synthesis.26 Crystal structures of triclosan bound to FabI sug-
gest that hydrogen bonding between the phenol and enzyme is
important to enzyme binding.34 While our unaltered triclosan
Inspecting the antimicrobial results as a whole revealed a num-
ber of trends. The chloroxylenol conjugates 6 and 7 were broadly
ineffective, displaying reduced activities compared to conventional
neomycin sulfate. The triclosan conjugate with a short linker, 8,
was similarly inactive, though the longer triclosan conjugate 9
was more active than neomycin sulfate against MRSA and two P.
aeruginosa strains. Both conjugates of clofoctol were active, but
both displayed increased hemolytic activity, suggesting that the
hydrophobic tail of clofoctol mediates non-specific interactions,
similar to our standard CA, 1. The intermediate hemolytic activity
of 10 and 11 is likely due to the influence of the cationic aminogly-
coside face, which may preferentially target the hybrids to nega-
tively charged bacterial membrane lipids, reducing interactions
with zwitterionic eukaryotic cells. In general, it appears that the
conjugates are less active against neomycin susceptible bacteria
such as MRSE and Klebsiella pneumoniae, but more active than
neomycin against drug resistant strains such as MRSA and P.
aeruginosa.
standard had an MIC against E. coli ATCC 25922 of 60.25 lg/mL,
both triclosan conjugates 8 and 9 are less active, suggesting that
using the phenol of triclosan to form an ether linkage has removed
some of the site specific antimicrobial activity. When this informa-
tion is combined with the BSA results, it seems likely that for con-
jugates 8 and 9 improvements in the antibacterial activity over
neomycin B must therefore relate to either increased binding to
cellular targets, a new resilience to enzymatic inactivation, or an
increased concentration of the conjugate in the cell.
To test the influence of non-specific interactions on conjugate
activity antimicrobial testing was repeated in the presence of 4%
bovine serum albumin (BSA). Many cationic amphiphiles show
greatly reduced activity in the presence of BSA due to protein bind-
ing,19 but with the exception of the results for compounds 9–11
Of course, each of these options may come into play. Com-
pounds 9–11 remain relatively active against MRSA, while our ami-
noglycoside control, neomycin sulfate has its activity reduced over

B. Findlay et al. / Bioorg. Med. Chem. Lett. 22 (2012) 1499–1503
1503
250-fold from Staphylococcus aureus to MRSA. The bacterial strain’s
resistance is likely due to the presence of neomycin-modifying en-
zymes that modify the drug, blocking effective binding to RNA or
decreasing the drug’s intracellular concentration.5 The presence
of a large hydrophobic moiety in compounds 9–11 likely prevents
successful binding to the inactivating enzymes, leading to
sustained activity. When we examine the conjugates created with
the smaller chloroxylenol moiety we see that they show
the expected decrease in activity against MRSA, suggesting the sin-
gle aromatic ring lacks the bulk required to prevent enzyme
interactions.
The increased activity of conjugates 9–11 against the two
strains of P. aeruginosa is somewhat more difficult to explain. All
three compounds were roughly fourfold more active than neomy-
cin sulfate, but while P. aeruginosa does possess inactivating en-
zymes, most of its drug resistance stems from the expression of
efflux pumps.5,7 A triclosan-specific interaction is unlikely, due to
the presence of a non-susceptible analogue of FabI, FabV,26 and
the inferences made by examining the activity of conjugates 8
and 9 against E. coli (vide supra). One possible explanation is that
attaching the hydrophobic phenolics to neomycin has increased
the drugs diffusion into the cell, partially overcoming the effect
of the efflux pumps. This hypothesis fits well with the reduced effi-
cacy of conjugates 9–11 against P. aeruginosa in the presence of
BSA, as we would expect the hydrophobic protein to reduce the
concentration of free conjugate outside of the cell, slowing diffu-
sion and aiding efflux. The effect of efflux pumps and permeability
could be further characterized using strains of P. aeruginosa with
reduced efflux, but the relatively low activity of the conjugates
may complicate matters.
Natural Sciences and Engineering Research Council of Canada
(NSERC) and the Canadian Foundation for Innovation (CFI).
Supplementary data
Supplementary data (experimental details) associated with this
article can be found, in the online version, at doi:10.1016/
j.bmcl.2012.01.025.
References and notes
1. Arias, C. A.; Murray, B. E. N. Eng. J. Med. 2009, 360, 439.
2. Payne, D. J.; Gwynn, M. N.; Holmes, D. J.; Pompliano, D. L. Nat. Rev. Drug Disc.
2007, 6, 29.
3. Vatopoulos, A. C.; Kalapothaki, V.; Legakis, N. J. Emerg. Infect. Dis. 1999, 5, 471.
4. Lowy, F. D. J. Clin. Invest. 2003, 111, 1265.
5. Wright, G. D. Nat. Rev. Microbiol. 2007, 5, 175.
6. Mulvey, M. R.; Simor, A. E. CMAJ 2009, 180, 408.
7. Li, X. Z.; Nikaido, H. Drugs 2004, 64, 159.
8. Bera, S.; Zhanel, G. G.; Schweizer, F. J. Antimicrob. Chemother. 2010.
9. Bera, S.; Zhanel, G. G.; Schweizer, F. J. Med. Chem. 2010.
10. Bera, S.; Zhanel, G. G.; Schweizer, F. Bioorg. Med. Chem. Lett. 2010, 20, 3031.
11. Bera, S.; Zhanel, G. G.; Schweizer, F. J. Med. Chem. 2008, 51, 6160.
12. Findlay, B.; Zhanel, G. G.; Schweizer, F. Antimicrob. Agents Chemother. 2010, 54,
4049.
13. Hanessian, S.; Pachamuthu, K.; Szychowski, J.; Giguere, A.; Swayze, E. E.;
Migawa, M. T.; Francois, B.; Kondo, J.; Westhof, E. Bioorg. Med. Chem. Lett. 2010,
20, 7097.
14. Zhang, J.; Chiang, F. I.; Wu, L.; Czyryca, P. G.; Li, D.; Chang, C. W. J. Med. Chem.
2008, 51, 7563.
15. Ouberai, M.; El Garch, F.; Bussiere, A.; Riou, M.; Alsteens, D.; Lins, L.; Baussanne,
I.; Dufrene, Y. F.; Brasseur, R.; Decout, J. L.; Mingeot-Leclercq, M. P. Biochim.
Biophys. Acta 2011, 1808, 1716.
16. Baussanne, I.; Bussiere, A.; Halder, S.; Ganem-Elbaz, C.; Ouberai, M.; Riou, M.;
Paris, J. M.; Ennifar, E.; Mingeot-Leclercq, M. P.; Decout, J. L. J. Med. Chem. 2010,
53, 119.
17. Marr, A. K.; Gooderham, W. J.; Hancock, R. E. Curr. Opin. Pharmacol. 2006, 6, 468.
18. Rotem, S.; Mor, A. Biochim. Biophys. Acta 2008, 1788, 1582.
19. Svenson, J.; Brandsdal, B. O.; Stensen, W.; Svendsen, J. S. J. Med. Chem. 2007, 50,
3334.
20. McDonnell, G.; Russell, A. D. Clin. Microbiol. Rev. 1999, 12, 147.
21. Pokrovskaya, V.; Belakhov, V.; Hainrichson, M.; Yaron, S.; Baasov, T. J. Med.
Chem. 2009, 52, 2243.
22. Waring, M. J. Bioorg. Med. Chem. Lett. 2009, 19, 2844.
23. Hancock, R. E.; Bellido, F. J. Antimicrob. Chemother. 1992, 29, 235.
24. Schroeder, R.; Waldsich, C.; Wank, H. EMBO J. 2000, 19, 1.
25. Chen, Y.; Guarnieri, M. T.; Vasil, A. I.; Vasil, M. L.; Mant, C. T.; Hodges, R. S.
Antimicrob. Agents Chemother. 2007, 51, 1398.
26. Zhu, L.; Lin, J.; Ma, J.; Cronan, J. E.; Wang, H. Antimicrob. Agents Chemother.
2010, 54, 689.
27. Meldal, M.; Tornoe, C. W. Chem. Rev. 2008, 108, 2952.
28. Kolb, H. C.; Finn, M. G.; Sharpless, K. B. Angew. Chem., Int. Ed. 2004, 2001, 40.
29. Dickschat, J. S.; Reichenbach, H.; Wagner-Döbler, I.; Schulz, S. Eur. J. Org. Chem.
2005, 2005, 4141.
In conclusion, we have produced six novel aminoglycoside–
phenolic conjugates, in order to test the viability of combining
known hydrophobic drugs and aminoglycosides to create com-
pounds with an emergent activity similar to that of the cationic
antibacterial peptides and cationic detergents. In general the con-
jugates displayed improved activity against neomycin sulfate resis-
tant bacteria and slightly reduced activity against neomycin
susceptible strains. For several conjugates activity against MRSA
was found comparable to that against S. aureus, while activity
against P. aeruginosa was moderately improved. Unlike previous
work with analogues of CAs like cationic detergents and amphi-
philic aminoglycosides,8,11,12 our most active compounds were
not appreciably hemolytic, and activity was retained in the pres-
ence of BSA. Work is currently in progress to optimize the antimi-
crobial activity of non-hemolytic neomycin phenol conjugates and
to explore the likelihood of resistance development.
30. Boer, J.; Blount, K. F.; Luedtke, N. W.; Elson-Schwab, L.; Tor, Y. Angew. Chem., Int.
Ed. 2005, 44, 927.
31. Hanessian, S.; Masse, R.; Capmeau, M. L. J. Antibiot. (Tokyo) 1977, 30, 893.
32. Zhanel, G. G.; DeCorby, M.; Laing, N.; Weshnoweski, B.; Vashisht, R.; Tailor, F.;
Nichol, K. A.; Wierzbowski, A.; Baudry, P. J.; Karlowsky, J. A.; Lagace-Wiens, P.;
Walkty, A.; McCracken, M.; Mulvey, M. R.; Johnson, J.; Hoban, D. J.Canadian
Antimicrobial Resistance Alliance (CARA) Antimicrob. Agents Chemother. 2008,
52, 1430.
33. Zhanel, G. G.; Adam, H. J.; Low, D. E.; Blondeau, J.; Decorby, M.; Karlowsky, J. A.;
Weshnoweski, B.; Vashisht, R.; Wierzbowski, A.; Hoban, D. J.Canadian
Antimicrobial Resistance Alliance (CARA) Microbiol. Infect. Dis. 2011, 69, 291.
34. Stewart, M. J.; Parikh, S.; Xiao, G.; Tonge, P. J.; Kisker, C. J. Mol. Biol. 1999, 290,
859.
Acknowledgments
We are indebted to Dr. E. Lattová for her assistance with high
resolution mass spectrometry, to N. Liang for aid in hemolytic
and microbial testing and to L.K. Freeman for helpful discussions.
This research was supported by the Canadian Institutes of Health
Research (CIHR), Manitoba Health Research Council (MHRC),