Rational Design and Synthesis of Modified Teixobactin Analogues: In Vitro Antibacterial Activity against Staphylococcus aureus, Propionibacterium acnes and Pseudomonas aeruginosa

Article abstract of DOI:10.1002/chem.201801423

Teixobactin, a recently discovered depsipeptide that binds to bacterial lipid II and lipid III, provides a promising molecular scaffold for the design of new antimicrobials. Herein, we describe the synthesis and antimicrobial evaluation of systematically modified teixobactin analogues. The replacement of the Ile¹¹ residue with aliphatic isosteres, the modification of the guanidino group at residue 10 and the introduction of a rigidifying residue, that is, dehydroamino acid, into the macrocyclic ring generated useful structure–activity information. Extensive antimicrobial susceptibility assessment against a panel of clinically relevant Staphylococcus aureus and Propionibacterium acnes strains led to the identification of the new lead compound, [Arg(Me)¹⁰,Nle¹¹]teixobactin, with an excellent bactericidal activity (minimum inhibitory concentration (MIC)=2–4 μg mL^?1). Significantly, the antimicrobial activity of several of the teixobactin analogues against the pathogenic Gram-negative Pseudomonas aeruginosa was “restored” when combined with the sub-MIC concentration of the outer membrane-disruptive antibiotic colistin. The antimicrobial effectiveness of a [Tfn¹⁰,Nle¹¹]teixobactin (32 μg mL^?1)–colistin (2 μg mL^?1; 0.5×MIC) combination against P. aeruginosa PAO1 reveals, for the first time, an alternative therapeutic option in the treatment of Gram-negative infections.

Full text of DOI:10.1002/chem.201801423

A Journal of
Accepted Article
Title: Rational design and synthesis of modified teixobactin analogues:
in vitro antibacterial activity against Staphylococcus aureus,
Propionibacterium acnes and Pseudomonas aeruginosa
Authors: Vivian Ng, Sarah A Kuehne, and Weng Chan
This manuscript has been accepted after peer review and appears as an
Accepted Article online prior to editing, proofing, and formal publication
of the final Version of Record (VoR). This work is currently citable by
using the Digital Object Identifier (DOI) given below. The VoR will be
published online in Early View as soon as possible and may be different
to this Accepted Article as a result of editing. Readers should obtain
the VoR from the journal website shown below when it is published
to ensure accuracy of information. The authors are responsible for the
content of this Accepted Article.
To be cited as: Chem. Eur. J. 10.1002/chem.201801423
Link to VoR: http://dx.doi.org/10.1002/chem.201801423
Supported by

10.1002/chem.201801423
Chemistry - A European Journal
FULL PAPER
Rational design and synthesis of modified teixobactin analogues:
in vitro antibacterial activity against Staphylococcus aureus,
Propionibacterium acnes and Pseudomonas aeruginosa
*
[a]
Vivian Ng,[a] Sarah A. Kuehne[b] and Weng C. Chan
Abstract: Teixobactin, a recently discovered depsipeptide that binds
to bacterial lipid II and lipid III, provides a promising molecular
scaffold for the design of new antimicrobials. Herein, we describe the
synthesis and antimicrobial evaluation of systematically modified
teixobactin analogues. The replacement of Ile11 residue with aliphatic
isosteres, the modification of the guanidino group at residue 10 and
the introduction of a rigidifying residue, dehydroamino acid into the
macrocyclic ring generated useful structure-activity information.
Extensive antimicrobial susceptibility assessment against a panel of
clinically relevant Staphylococcus aureus and Propionibacterium
demonstrated excellent bactericidal activity against methicillin-
resistant Staphylococcus aureus (MRSA) which is associated
with a wide range of infections in both the community (e.g.
cellulitis, abscesses) and the hospital settings (e.g. bacteraemia,
pneumonia).[4–6] Vancomycin is currently the last line of defence
against MRSA infections but strains with reduced susceptibility
to this antibiotic have surfaced.[7–10] Teixobactin offers a potential
solution to this predicament since it remains effective against
MRSA, as well as vancomycin-intermediate S. aureus (VISA)
due to its unique mode of action.[4] It has been shown to
synergistically block the biosyntheses of peptidoglycan and
teichoic acid, thereby resulting in a weakened cell wall and
autolysin-mediated cell lysis.[4,11]
Teixobactin 1 is also a potent antimicrobial against another
human skin commensal, Propionibacterium acnes.[4] This Gram-
positive anaerobe is commonly associated with acnes
vulgaris.[12] In recent years, however, it is increasingly
recognised as an opportunistic pathogen that can cause
invasive infections, especially those associated with medical
implants.[13,14] There have been several reports on the isolation
of P. acnes from prosthetic joints, cardiovascular devices and
ophthalmic implants.[14–18] To aggravate matters, the widespread
use of antibiotics to treat acne vulgaris has led to the emergence
of P. acnes strains that are resistant to numerous antibiotics,
including the macrolides, tetracycline and metronidazole.[13,19–22]
The need for novel antimicrobials is therefore more pressing
than ever. It is hoped that teixobactin and its analogues may
serve as a timely solution to this clinically important pathogen.
acnes led to the identification of
a new lead compound,
[Arg(Me)10,Nle11]teixobactin 63, with excellent bactericidal activity
(MIC 2–4 μg/mL). Significantly, the antimicrobial activity of several of
the teixobactin analogues against the pathogenic Gram-negative
Pseudomonas aeruginosa was ‘restored’ when combined with sub-
MIC concentration of the outer membrane-disruptive antibiotic,
colistin. The antimicrobial effectiveness of [Tfn10,Nle11]teixobactin 66
(32 μg/mL)-colistin (2 μg/mL; 0.5x MIC) combination against P.
aeruginosa PAO1 reveals, for the first time, an alternative
therapeutic option in the treatment of Gram-negative infections.
Introduction
Life-saving antibiotics are rapidly losing the race against the
development of bacterial resistance to most, if not all, antibiotics.
The resultant health and financial implications have spurred the
deployment of antimicrobial stewardship programmes across the
globe to ensure evidence-based prescribing of antibiotics that
are still effective.[1–3] Meanwhile, scientists are working hand-in-
hand to tackle the resistance crisis through drug discovery and
development initiatives.
Natural antimicrobial peptides serve as invaluable
molecular scaffolds for the development of the next generation
of antimicrobial therapeutics. The recently discovered
depsipeptide, teixobactin 1 (Figure 1), has great potential as a
lead compound due to its favourable potency against many
Gram-positive pathogens.[4] Among them, teixobactin has
Figure 1. Structure of teixobactin 1 and the four sites (blue) of modification
presented in this work.
Teixobactin 1, comprised of a 13-membered depsipeptide
core and a tethered linear heptapeptide, offers multiple sites for
synthetic modifications to improve its potency and efficacy. In
less than three years since its discovery, more than a hundred
analogues have been synthesized by various research groups in
the hope of elucidating its structure-activity relationships
(SARs).[23–37] The biological activities of these analogues and the
[a]
[b]
V. Ng, Dr W.C. Chan
School of Pharmacy, Centre for Biomolecular Sciences,
University of Nottingham, University Park,
Nottingham, U.K., NG7 2RD.
E-mail: weng.chan@nottingham.ac.uk
Dr. S.A. Kuehne
School of Dentistry, and Institute for Microbiology and Infection,
University of Birmingham,
Birmingham, U.K., B5 7EG.
different
synthetic strategies
reviewed.[38,39]
reported
X-ray
have
been
comprehensively
crystallographic,
molecular dynamic and NMR structural studies have also been
conducted to construct possible binding models of the native
Supporting information for this article is given via a link at the end of
the document.
This article is protected by copyright. All rights reserved.

10.1002/chem.201801423
Chemistry - A European Journal
FULL PAPER
peptide and its analogues.[26,27,40] Additionally, in a recent mini-
review, we provided an insight into the structural similarities of
teixobactin with other lipid II inhibitors.[41] Together, these
resources provide tremendous information that could aid the
design of optimised analogues.
Early synthetic endeavours focused primarily on the
exocyclic tail and the backbone stereochemistry of the native
peptide. The replacement of any D-amino acid residues with its
L-counterparts abolished activity, suggesting
a significant
contribution of these residues for the optimal conformation of
teixobactin.[25,26,30] Yang et al. further demonstrated the
importance of the N-terminal tail as the removal of the first five
residues detrimentally affected antimicrobial potency.[24]
Teixobactin appears to bind to the pyrophosphate and N-
acetylmuramic acid amino sugar of lipid II. As such, its cyclic
ring is believed to act as the main site of recognition.[4] With
these considerations in mind, we have developed a series of
analogues with modifications mainly on the macrocyclic core to
examine the significance of hydrophobicity at position 11, the
cationic feature of the guanidino group at position 10, and the
effect of introducing conformational rigidity at position 9. The N-
Me-D-Phe1 was also replaced with D-Trp in an attempt to
investigate the contribution of the phenyl group.
Apart from replacing the Ile residue at position 11 with
readily available aliphatic isosteres, we sought to investigate the
effect of introducing fluorine atoms and unsaturated side-chain
at this position. Thus, Fmoc-(S)-6,6,6-trifluoronorleucine-OH
(Fmoc-Tfn-OH) and Fmoc-(S)-homoallylglycine-OH (Fmoc-Hag-
OH) were synthesized and their preparation will be discussed
prior to the synthesis of the teixobactin analogues. All analogues
were extensively evaluated for their antimicrobial activity against
both S. aureus and P. acnes. Thus far, most biological
assessments of reported teixobactin analogues are focused on
S. aureus and only the activity of the native peptide is known
against P. acnes. Herein, we report detailed antimicrobial activity
of teixobactin analogues against several P. acnes strains.
Although teixobactin and analogues thereof are considered
inactive against Gram-negative bacteria (MIC >256 μg/mL), the
effect of using teixobactin analogues in combination with colistin
was also investigated against the Gram-negative pathogen
Pseudomonas aeruginosa.
Scheme
1.
An
optimized
synthesis
of
Ni(II)-Gly-(S)-2-[N-(N-
benzylprolyl)amino]-benzophenone (BPB).
Thus, using the protocol developed by Belokon et al., the
first step in the synthesis progressed smoothly to give N-
benzylated L-proline (S)-(3) in high yield.[43] Although the
condensation between (S)-(3) and 2-aminobenzophenone did
not proceed to completion, a reasonable yield of 45–60 % was
obtained. To our dismay, the use of KOH in the final step, i.e.
the transformation of (S)-4 to (S)-5, gave a disappointing 50 %
recovery of (S)-5 after three recrystallizations. A review of the
literature indicated that K2CO3 was previously employed by
Soloshonok and co-workers to prepare a closely related Schiff
base,[44] thereby suggesting that this alternative base could be
more effective for synthesizing (S)-5. Gratifyingly, K2CO3 (20
equiv.) drove the final reaction step to completion within an hour
and (S)-5 was recrystallized from MeOH/H2O in >85 % yield.
Results and Discussion
Synthesis of Fmoc-(S)-6,6,6-trifluoronorleucine-OH and
Fmoc-(S)-homoallylglycine-OH
Scheme 2. An optimised synthesis of Fmoc-6,6,6-trifluoronorleucine-OH.
An operationally simple and cost-effective approach for the
asymmetric synthesis of Fmoc-Tfn-OH (S)-7 and Fmoc-Hag-OH
(S)-9 is by alkylation of an achiral auxiliary reagent Ni(II)-glycine
Schiff base (S)-5.[42,43] The Ni(II)-complex (S)-5 was synthesized
in large scale in three straightforward steps (Scheme 1). The
coordination of Ni(II) ion to the glycine greatly increased the
acidity of the α-proton, enabling subsequent rapid alkylation with
an alkyl halide.[42]
Having successfully prepared the Ni(II)-Schiff base (S)-5,
we then sought to optimise the alkylation of the complex with
1,1,1-trifluoro-4-iodobutane (Scheme 2). Wang et al. have
previously reported
a high diastereoselectivity (97%) was
achieved with only 1.1 equiv. of NaOH.[45] We have, however,
obtained contradictory results. Although the rate of reaction
increased with an increased amount of NaOH, the
This article is protected by copyright. All rights reserved.

10.1002/chem.201801423
Chemistry - A European Journal
FULL PAPER
diastereoselectivity between (S,S)-6 and (S,R)-6 was
disappointingly low (Table 1). It has been shown in other studies
These Fmoc-protected amino acids were used in the synthesis
of our teixobactin analogues.
that when
a
monoalkylated complex is subjected to
epimerisation under basic condition, thermodynamic control
would dominate and eventually drives the equilibrium towards
the favoured (S,S)-epimer.[46–48] Thus, a mixture of the (S,S)-
and (S,R)-alkylated complex 6 was partially purified in a reaction
work-up, and subsequently treated with K2CO3 in MeOH at
60 °C (Table 1, entry 3). A satisfactory diastereomeric ratio of
97:3 was obtained. In a series of pilot experiments, the addition
of K2CO3 directly into the DMF reaction mixture at 60 °C did not
afford the desirable level of diastereoselectivity whereas
increasing the amount of NaOH led to the formation of other by-
products. As such, the rate of epimerization and the shift of
equilibrium towards the (S,S)-epimer seemed to be highly
dependent on the use of a polar protic solvent.
Synthesis of teixobactin analogues
The substitution of the L-allo-enduracididine10 in teixobactin 1
with arginine has been shown to retain appreciable activity.[23–25]
Hence, (Arg10)teixobactin 55 was initially prepared (Scheme 4)
to serve as the positive control in our microbiological evaluation.
The robust synthetic method reported by Yang et al. was
adapted with several adjustments.[24] This synthetic strategy
enabled the use of the resin-anchored linear decapeptide 16 as
a common intermediate for subsequent on-resin esterification
with different protected amino acid building blocks, thus
providing an expedient access to teixobactin analogues 57–62.
The 2-chlorotrityl chloride resin was used as the polymer
support as it allowed the liberation of the branched peptide by
mild acidolysis while retaining the side-chain protecting groups.
In the synthesis of (Arg10)teixobactin 55, the resin was first
loaded with Fmoc-Arg(Pbf)-OH and the peptide chain was
elongated stepwise by acylation with Fmoc-amino acids that
were pre-activated by HATU (1-[bis(dimethylamino)methylene]-
1H-1,2,3-triazolo[4,5-b]pyridinium 3-oxide hexafluorophosphate)
in the presence of iPr2EtN. In order to minimise steric hindrance
effects, Fmoc-D-Thr-OH and Fmoc-D-Gln-OH were used without
side-chain protecting groups and no significant problems were
observed during their couplings.
Table 1. Conditions investigated for the alkylation of (S)-5 with 1,1,1-trifluoro-4-
iodobutane.[a]
Entry
Condition
Total
(S,S):(S,R)[b] Yield (%)
reaction time
(h)
1.5
0.5
2.5
1
2
3
NaOH (2 equiv.), DMF, r.t.
NaOH (5 equiv.), DMF, r.t.
NaOH (5 equiv.), DMF, r.t.
followed by K2CO3 (5 equiv.),
MeOH, 60 °C
77/23
77/23
97/3
75
61
73
The esterification of the D-Thr residue in intermediate 16
was then carried out with a pre-formed symmetrical anhydride of
Fmoc-Ile-OH. It was found that the esterification step required
up to four repeated couplings, each requiring overnight (15–18
[a] Reactions were performed using 0.40 mmol (S)-5 and 0.44 mmol alkyl halide.
[b] Diastereomeric ratio determined using analytical RP-HPLC.
h) exposure, in order to achieve approximately 50
%
transformation as monitored by RP-HPLC. The long reaction
time could be attributed to the steric bulk of the Ile side-chain
and the entrenched hydroxyl group of the D-Thr residue. In
contrast, greater than 70 % O-acylation of the D-Thr residue was
achieved with Nle, Nva, Abu and Ala as the acylating amino
acids (analogues 57–60), further corroborating the significance
of steric effects. Following Fmoc-deprotection, the branched
peptide 25 was cleaved from the resin in preparation for
solution-phase intramolecular cyclisation between the Arg10 and
Ile11 residues.
The use of (1-cyano-2-ethoxy-2-oxoethylidenamino-
oxy)dimethylamino-morpholino-carbenium hexafluorophosphate
(COMU) as the carboxy-activating reagent enabled the
cyclisation to be visually monitored through colour changes.[49,50]
Upon addition of base, the solution changed from yellowish-
orange to colourless in under an hour and the reaction was
Scheme 3. The synthetic route to Fmoc-homoallylglycine-OH.
completed within
2 h. Global side-chain deprotection by
Finally, the de-assembly of (S,S)-6 to yield the desired amino
acid was achieved by microwave-assisted acid-mediated
hydrolysis. Without further purification, the released amino acid
was N-protected with a Fmoc group. Following a simple work-up,
trituration with hexane afforded sufficiently pure Fmoc-Tfn-OH
(S)-7 for incorporation into the macrocycle of teixobactin. Fmoc-
Hag-OH (S)-9 was prepared under similar conditions by
alkylating Schiff base (S)-5 with 4-bromo-but-1-ene (Scheme 3).
acidolysis of the macrocyclic intermediate, followed by RP-HPLC
purification and lyophilisation afforded 55 as a white solid.
All the other analogues were similarly prepared by
replacing Ile11 with various aliphatic residues, including the use
of the synthesized Fmoc-Tfn-OH and Fmoc-Hag-OH (57–62),
Arg10 with several arginine derivatives and Tfn (63–66), Ala9 with
Abu and Z-dehydrobutyrine (Dhb) (67–68), and finally, N-Me-D-
Phe1 with D-Trp (69) (Figure 2).
This article is protected by copyright. All rights reserved.

10.1002/chem.201801423
Chemistry - A European Journal
FULL PAPER
Scheme 4. Synthetic route for the preparation of teixobactin analogues. The synthesis of branched peptides 25–39 were performed solely on a polymer support
(2-chlorotrityl chloride resin) using a 9-fluorenylmethyloxycarbonyl (Fmoc/tBu) strategy. The cleaved peptides 40–54 were then subjected to solution-phase
macrolactamisation.
This article is protected by copyright. All rights reserved.

10.1002/chem.201801423
Chemistry - A European Journal
FULL PAPER
Figure 2. The chemical structures of teixobactin analogues synthesized for antimicrobial evaluation.
This article is protected by copyright. All rights reserved.

10.1002/chem.201801423
Chemistry - A European Journal
FULL PAPER
Broth microdilution and growth inhibition assays
Table 2. Bacterial strains used for antimicrobial susceptibility testing of
teixobactin analogues.
Minimal inhibitory concentration (MIC) is the most commonly
used parameter to define in vitro antimicrobial susceptibility. It is
defined as the lowest concentration of a compound that inhibits
visible growth. Using the broth microdilution method outlined in
the Clinical and Laboratory Standards Institute (CLSI)
guidelines[51], the MIC values of our teixobactin analogues
against five prominent isolates of S. aureus from different clinical
settings and with varying antibiotic sensitivity were obtained.
Among them, S. aureus USA300 JE2 is particularly virulent due
to the expression of Panton-Valentine Leukocidin (PVL). In fact,
the incidence of skin and soft tissue infections caused by S.
aureus has increased substantially since the late 1990s, which is
largely driven by the emergence of the USA300 clone.[52,53] The
MIC was also determined for three P. acnes strains and the
Gram-negative Pseudomonas aeruginosa PAO1. A summary
description of the bacteria strains used in this study is provided
in Table 2.
Since MIC is determined at a fixed incubation time-point
(typically 16 h or 24 h), detailed information on how bacterial
growth rate is affected by antimicrobials at different
concentrations is unavailable.[65] Hence, a sensitive assay that
involved the generation of growth curves was concurrently used
to evaluate the synthesized teixobactin analogues.
S. aureus SH1000, a fully sequenced strain that is
representative of the species, was used as the test
microorganism in our growth inhibition assay. The starting
bacterial concentration was adjusted to give approximately 106
colony forming units/mL (CFU/mL) in each well of a 96-well
microtitre plate. Following treatment with different concentrations
of the compounds, bacterial growth was monitored for 20 h by
measuring the optical density at 600 nm. The normalized
percentage growth at 13 h was then used to construct a dose-
response curve in order to determine the concentration at which
50 % of the bacterial growth was inhibited (IC50). As illustration,
Organism
Description
Reference
S. aureus
SH1000
Newman
S. aureus 8325-4 strain with repaired rsbU gene
[54]
Clinical MSSA isolate with fully sequenced genome [55,56]
and lacks in antibiotic resistant genes
PM64
Healthcare-associated MRSA, isolate of the
[57]
epidemic MRSA type 16 clonal group (EMRSA-16)
USA300 JE2 Community-acquired MRSA USA300 LAC cured of
three plasmids
[58]
Mu50
Vancomycin-intermediate resistant strain (VISA)
with thickened cell walls
[8,9,59]
P. acnes
ATCC 11828 Representative strain of the type II phylotype
[60,61]
[62]
ATCC 6919
Asn12
Representative strain of the type IA1 phylotype
Representative strain of type III phylotype isolated
from intervertebral disc material
[63]
P. aeruginosa
PAO1
Gram-negative, wild-type laboratory strain,
Nottingham sub-line
[64]
(Arg10,Nle11)teixobactin 57
100
Replicate 1
Replicate 2
Replicate 3
Vancomycin
50
the
dose-response
curves
of
two
analogues,
(Arg10,Nle11)teixobactin 57 (IC50
=
7.38 ±
0.09 μM) and
(Arg10,Ala11)teixobactin 60 (IC50 >100 μM), are shown in Figure
3; detailed growth curves and their corresponding dose-
response curves are provided in the Supporting Information (p.
S37–S40).
-1
0
1
2
3
log [compound, µM]
(Arg10,Ala11)teixobactin 60
The sigmoidal dose-response curves of all active
analogues showed a sharp drop in percentage growth as the
concentration of the test compounds were increased. This was
also observed with vancomycin, a well-established lipid II
inhibitor which was used as the positive control. This
phenomenon where a small change in concentration causes
substantial growth inhibition appeared to be a common attribute
of lipid II and/or lipid III binders. In addition to a measurement of
potency, the IC50 values (Table 3) provided useful and invaluable
SAR information on the effect of subtle changes in the chemical
composition of our teixobactin analogues.
Replicate 1
Replicate 2
Replicate 3
100
50
Vancomycin
-1
0
1
2
3
log [compound, µM]
Figure 3. Dose-response curves of analogues 57 and 60 against S. aureus
SH1000. Three independent experiments were conducted, with vancomycin
(IC50 = 0.45 ± 0.13 μM) as the positive control.
This article is protected by copyright. All rights reserved.

10.1002/chem.201801423
Chemistry - A European Journal
FULL PAPER
The (Arg10,Nle11)teixobactin (57) was also tested using the
same methodology against P. acnes ATCC 11828 (Figure 4).
Unlike S. aureus, the anaerobe P. acnes showed a more
gradual decrease in growth with increasing concentrations of the
test compound. This distinct dose-response pattern may be due
to the slow-growing nature of the bacteria and/or the difference
in its cell wall composition compared to other Gram-positives.[13]
In future, further work will be performed to establish the lipid
contents of the cell wall of P. acnes in order to shed light on the
binding efficacy of teixobactin analogues.
with Nva and Abu (compounds 58 and 59) followed by a
complete loss of activity when Ala was installed (compound 60).
This indicates that a minimum of four carbons are necessary for
optimal hydrophobic interactions. The importance of a non-polar
group at position 11 was also evident in previous studies which
showed that the substitution of Ile with a polar residue, such as
Lys, abolished antibacterial activity.[27,31]
Having identified Nle as a beneficial substitute of Ile, we
next investigated the effects of introducing fluorine atoms and an
unsaturated functional group in the hydrocarbon side-chain. It is
hypothesised that the replacement of the terminal methyl in Nle
with
a
trifluoromethyl moiety could provide the extra
hydrophobicity needed for effective binding. However, the
antimicrobial potency against S. aureus SH1000 dropped by
almost 4.0-fold when Nle was replaced with Tfn in compound 61.
The MIC values of (Arg10,Tfn11)teixobactin 61 against the other S.
aureus strains were also observed to be 2- to 4-fold higher
compared to (Arg10,Nle11)teixobactin 57. The electronegativity of
the fluorine atoms appeared to detrimentally affect the
interaction of the (Arg10,Tfn11)-analogue 61 with its molecular
targets. Alternatively, the steric effect from the bulkier CF3
moiety might have caused the decrease in activity. Further work
could be carried out by synthesizing a shorter trifluoromethylene
analogue since the -CF3 is considered an isosteric replacement
of an ethyl moiety.[66–68] A similar increase in both IC50 and MIC
was observed when a terminal alkenic bond was introduced
(compound 62). Nle was therefore chosen as the optimal
residue at this position for the subsequent six analogues 63–68,
in which the effect of replacing the guanidine moiety in Arg10 was
investigated. The less sterically demanding Nle additionally
Figure 4. Dose-response curves of analogue 57 against P. acnes ATCC
11828 (IC50 = 3.56 ± 0.23 μM). Three independent trials were conducted with
vancomycin (IC50 = 0.29 ± 0.03 μM) as the positive control.
Both arginine and lysine have been used as a substitute
for L-allo-enduracididine at position 10. Yang et al. reported that
(Lys10)teixobactin showed a 2- to 4-fold lower MIC values than
(Arg10)teixobactin against specific strains of Staphylococcus
epidermidis, Streptococcus salivarus, Enterococcus durans and
Bacillus subtilis.[24] A similar result against MRSA ATCC 33591
has recently been observed by Singh and co-workers.[37]
However, based on the MIC values, we observed that the
(Lys10)teixobactin was in fact consistently 2-fold less potent
compared to (Arg10)teixobactin across the S. aureus strains
tested; the IC50 data similarly reported a 2.2-fold loss of activity.
Since (Arg10)teixobactin showed a better antibacterial profile
here, Arg was maintained at this position in analogues 57 to 62
in our initial SAR study.
The Ile11 was initially replaced with its aliphatic isostere,
Nle. The resultant (Arg10,Nle11)teixobactin 57 (IC50 = 7.38 ± 0.09
μM) was found to have comparable potency, if not slightly better
than (Arg10)teixobactin 55 (IC50 = 7.96 ± 0.36 μM) in inhibiting the
growth of S. aureus SH1000. Additionally, both analogues
showed the same MIC of 8 μg/mL against three different strains
of S. aureus. It was surprising to observe that
(Arg10,Nle11)teixobactin was 2-fold more potent against MRSA
PM64 but 2-fold less active against VISA Mu50. Nevertheless,
the overall results suggest that the flexible linear aliphatic chain
of Nle seemed to confer similar hydrophobic interactions as the
branched chain of Ile. Decreasing the alkyl chain length,
however, has a negative impact. There was an increasing
reduction in antimicrobial potency with the replacement of Nle
provided
compared to Ile.
The cationic nature of L-allo-enduracididine10 in the native
a synthetic advantage in the esterification step
peptide is thought to be crucial in the electrostatic interaction
with the negatively-charged pyrophosphate moiety in lipid II.[4] In
this unusual amino acid, the terminal NH group of its guanidine
moiety is connected to the γ-carbon to afford a heterocyclic
structure. As such, it was envisaged that additional methyl
group(s) on the guanidine in the side-chain would closely mimic
the natural interaction of teixobactin with lipid II. Gratifyingly,
compared
to
(Arg10,Nle11)-analogue
57,
the
[Arg(Me)10,Nle11]teixobactin 63 displayed a 2-fold increase in
antimicrobial potency against several S. aureus strains (4-fold in
S. aureus Mu50) when evaluated using the IC50 and MIC values.
Evich et al. has shown that the pKa of the guanidino group of Arg
is not significantly altered by methylation.[69] Hence, the increase
in hydrophobicity rather than basicity resulting from the NG-
methylation has contributed to a greater binding affinity. On the
other hand, the asymmetric and symmetric NG-dimethylated
arginine analogues, 64 and 65, respectively showed 3.4- and
2.4-fold reduction in potency (IC50) compared to the
corresponding Arg analogue 57. The introduction of the second
methyl group appeared to be detrimental, possibly due to
disruption of potential hydrogen bond(s) or significant steric
hindrance.[69] These results are consistent with other studies that
showed a lack of activity when the guanidino group was tetra-
alkylated.[35]
This article is protected by copyright. All rights reserved.

10.1002/chem.201801423
Chemistry - A European Journal
FULL PAPER
Table 3. The antibacterial activity of teixobactin analogues and several antibiotics.
Minimum inhibitory concentration (μg/mL)[b]
IC50 (μM)
P.
S. aureus
P. acnes
against
S. aureus
SH1000[a]
aeruginosa
Compound
USA300
JE2
ATCC
11828
ATCC
6919
SH1000 Newman
PM64
Mu50
Asn12
PAO1
(Arg10)teixobactin 55
7.96 ± 0.36
17.43 ± 2.31
7.38 ± 0.09
8.53 ± 0.37
14.74 ± 0.25
> 100
8
16
8
8
16
8
8
16
8
8
16
4
8
16
16
16
32
>32
32
32
4
4
8
4
8
4
8
>256
>256
>256
>256
>256
>256
>256
>256
>256
>256
>256
>256
>256
>256
>256
(Lys10)teixobactin 56
(Arg10,Nle11)teixobactin 57
(Arg10,Nva11)teixobactin 58
(Arg10,Abu11)teixobactin 59
(Arg10,Ala11)teixobactin 60
(Arg10,Tfn11)teixobactin 61
(Arg10,Hag11)teixobactin 62
(Arg(Me)10,Nle11)teixobactin 63
(ADMA10,Nle11)teixobactin 64
(SDMA10,Nle11)teixobactin 65
(Tfn10,Nle11)teixobactin 66
(Abu9,Arg10,Nle11)teixobactin 67
(Dhb9,Arg10,Nle11)teixobactin 68
(D-Trp1,Arg10)teixobactin 69
4
4
4
8
8
16
16
>32
32
32
4
8
8
8
4
16
>32
32
32
4
16
>32
32
32
4
16
>32
16
32
2
8
8
8
32
16
16
2
>32
32
32
4
32
32
32
2
29.16 ± 1.51
28.55 ± 0.24
3.84 ± 0.26
24.83 ± 4.47
17.82 ± 3.42
8.02 ± 0.26
7.52 ± 0.24
25.69 ± 4.57
> 100
16
16
8
32
16
8
32
16
8
16
16
8
32
32
8
8
8
8
8
8
8
8
4
4
8
8
8
8
8
4
4
4
32
32
32
16
>32
32
>32
16
>32
16
>32
16
>32
>32
-
>32
-
>32
Teixobactin[c]
Meropenem
Vancomycin
Colistin
-
n.d.
0.5[11]
-
-
-
0.08[4]
n.d.
2
-
>100[4]
8
<0.0625 <0.0625 <0.0625
16
n.d.
4
n.d.
1
n.d.
1
0.45 ± 0.13
n.d.
1
0.5
n.d.
n.d.
0.5
n.d.
8
0.25
n.d.
n.d.
>256
4
n.d.
n.d.
n.d.
n.d.
n.d.
n.d.
n.d.
n.d.
n.d.
n.d.
Tunicamycin
n.d.
n.d.
[a]
[b]
IC50 values were expressed as the mean ± SD of three independent growth inhibition assay; n.d. = not determined. MIC determined by broth microdilution
[c]
performed in triplicates according to CLSI guidelines.
MIC values of teixobactin were obtained from literature; Nle = norleucine; Nva = norvaline; Abu =
aminobutyric acid; Tfn = trifluoronorleucine; Hag = homoallylglycine; Arg(Me) = monomethylarginine; ADMA = asymmetric dimethylarginine; SDMA = symmetric
dimethylarginine; Dhb = dehydrobutyrine.
This article is protected by copyright. All rights reserved.

10.1002/chem.201801423
Chemistry - A European Journal
FULL PAPER
Given the likely favourable electronic and hydrophobic
properties of our synthesized trifluoromethyl-containing amino
acid, the (Tfn10,Nle11)teixobactin 66 was synthesized and found
to be only marginally less active than (Arg10,Nle11)teixobactin 57
(IC50 8.02 ± 0.26 cf 7.38 ± 0.09 μM). This result reveals that
position 10 is a good location for the introduction of other
fluorinated hydrocarbon amino acids. In fact, Chen et al.
peptide compounds to the plastic wall of the 96-well
plate.[4,11,73,74] However, no difference in MIC was observed for
[Arg(Me)10,Nle11]teixobactin 63 in the presence of 0.002 % or
0.1 % Tween 80.
We next evaluated the antimicrobial utility of our
teixobactin analogues against the opportunistic pathogen P.
acnes. The majority of the analogues showed a 2-fold higher
potency against the three P. acnes strains tested compared to S.
aureus (Table 3). In the broth microdilution assay, the bacteria
inoculum for P. acnes was approximately 100-fold higher than S.
aureus. As such, the MIC values of the teixobactin analogues
are likely to be considerably lower if a similar inoculum size was
used. Overall, the results are in agreement with teixobactin
which showed a 3-fold higher potency against P. acnes
compared to S. aureus.[4] It is worth highlighting that the MIC of
our most potent analogue, [Arg(Me)10,Nle11]teixobactin (63), is
comparable to that of vancomycin. These results demonstrate
the promising application of our teixobactin analogue against the
emerging pathogenic bacterium P. acnes.
reported
that
(Ala10)teixobactin
unexpectedly
retained
considerable activity against S. aureus; it was proposed that a
positively-charged residue is not essential at position 10.[34]
The role of the Ala9 residue is not fully known except that it
could be replaced with polar amino acids, such as Lys or
Orn.[27,29,31,35] We observed that replacement of the methyl group
with an ethyl moiety did not significantly change the antimicrobial
activity
since
(Abu9,Arg10,Nle11)teixobactin
67
and
(Arg10,Nle11)teixobactin 57 are essentially equipotent against the
tested S. aureus strains. Subsequently, the possibility of
improving the conformational characteristics of the peptide
through the installation of an α,β-dehydroamino acid was
investigated. The alkenic amino acid residue is known to rigidify
the backbone of macrocyclic rings and could lead to a desirable
peptide conformation that enhances binding to its target(s).[70] Z-
Dhb was chosen as the test residue due to its natural
abundance in antimicrobial peptides, including the lipid II-binder
nisin.[71,72] However, compared to the corresponding Abu-
Time-kill assay to establish bactericidal activity against S.
aureus and P. acnes
10
(a)
9
DMSO control
8
containing
analogue
67,
the
potency
of
Tunicamycin control
7
6
5
4
3
2
1
0
(Dhb9,Arg10,Nle11)teixobactin 68 was decreased by 3.4-fold;
using the MIC values, a reduction of 2- to 4-fold in activity was
observed against all the tested S. aureus strains. A similar
result was found by Albericio and co-workers when 4-
aminoproline was introduced at position 10 to constrict the
macrocyclic ring.[35] Collectively, these results showed that
conformational restriction has an unfavourable effect on target
binding.
Vancomycin 4x MIC
[Arg(Me)10,Nle11]teixobactin 4x MIC
Vancomycin 4x MIC + tunicamycin
Bactericidal level
[Arg(Me)10,Nle11]teixobactin 4x MIC
+ tunicamycin
0
4
8
12
16
20
24
Time (h)
10
9
8
7
6
5
4
3
2
1
0
(b)
Meanwhile, hydrophobicity was previously found to play a
significant role at position 1. Although the removal of the N-
methyl group from N-Me-D-Phe1 has a minimal effect on potency,
activity was lost when Wu et al. replaced D-Phe with D-Tyr.[28]
The same observation was obtained with analogue 69, in which
the phenyl was substituted with an indole moiety; our (D-
Trp1,Arg10)-analogue 69 is essentially inactive.
DMSO control
Vancomycin 10x MIC
[Arg(Me)10,Nle11]teixobactin 10x MIC
Bactericidal level
0
4
8
12
16
20
24
Time (h)
The treatment of MRSA infection is increasingly difficult
due to the co-emergence of VISA strains. It is therefore not
surprising that the MIC of vancomycin is increased by 4- to 16-
fold when tested against VISA Mu50 (Table 3). This is attributed
to thicker peptidoglycan layers and a lower degree of cross-
linking that exposes more of the D-alanyl-D-alanine binding site
Figure 5. Time-kill curves of S. aureus USA300 JE2 treated with (a) 4x MIC
[Arg(Me)10,Nle11]teixobactin 63 or 4x MIC vancomycin in the presence and
absence of tunicamycin (0.4 μg/mL) and (b) 10x MIC 63 or 10x MIC
vancomycin. Data are presented as the mean log10 viable counts (CFU/mL) ±
SD from two independent experiments.
of vancomycin.[8,9]
A large proportion of the vancomycin
molecules are consequently trapped or titrated out by the mature
peptidoglycan, preventing them from reaching their vital targets
near the cytoplasmic membrane.[9] In contrast, most of the
teixobactin analogues tested showed the same MIC across all
the S. aureus strains, including Mu50. Hence, these teixobactin
analogues, similar to the parent teixobactin, are unaffected by
the thicker peptidoglycan layers in the VISA strains.[11] On a side
note, studies have suggested the addition of a surfactant in the
broth microdilution assay to prevent possible adsorption of the
Teixobactin has been shown to possess excellent bactericidal
activity against S. aureus.[4,11] Although both IC50 and MIC
values are good indicator of potency, it remains unknown if our
teixobactin analogues are bacteriostatic or bactericidal.[65] Hence,
the killing kinetics of our most potent analogue,
[Arg(Me)10,Nle11]teixobactin 63, were assessed against the
highly virulent pathogen S. aureus USA300 JE2 using a
validated time-kill method. Thus, S. aureus USA300 JE2 grown
to mid-exponential phase was challenged with suprainhibitory
This article is protected by copyright. All rights reserved.

10.1002/chem.201801423
Chemistry - A European Journal
FULL PAPER
concentrations (4x or 10x MIC) of 63. Vancomycin, an
established antibiotic for treating challenging S. aureus
infections, was used as a comparator. Viable counts (CFU/mL)
of the bacteria were then determined at specific times over 24 h.
A compound that reduces bacterial viability by ≥ 3 log10 CFU/mL
is defined as bactericidal.[75]
At 4x MIC, the [Arg(Me)10,Nle11]teixobactin 63 showed a
greater killing rate than vancomycin in the first six hours (Figure
5 (a)). There was, however, a regrowth of bacteria from 8 h
onwards for compound 63. To our delight, a decrease of >3 log10
compound 63 and tunicamycin are likely to perturb different
stages of the WTA biosynthesis. The rapid killing activity of 63
could therefore be due to a more lethal target within the WTA
biosynthetic pathway and further work is required to establish its
target. Despite being viable in vitro, bacterial cells without WTA
tend to show an altered morphology and defects in cellular
division.[83,86] This may explain why the killing effects of both
vancomycin and 63 were noticeably potentiated when used in
combination with tunicamycin.
The time-kill assay was similarly conducted on P. acnes
ATCC 11828. As expected, [Arg(Me)10,Nle11]teixobactin 63 was
bactericidal and exhibited a faster killing rate than vancomycin
(Figure 6). Unlike in S. aureus, however, no regrowth of bacteria
was observed up to 48 h with 4x MIC of compound 63. Once
again, this confirms our analogue 63 holds great promise for the
treatment of P. acnes infection.
CFU/mL
was rapidly
obtained with 10x
MIC of
[Arg(Me)10,Nle11]teixobactin 63, within an hour (Figure 5(b)). In
contrast, vancomycin required up to 4 h to achieve bactericidal
activity at 10x MIC. Thus, our lead compound 63 at 10x MIC
showed highly efficient and sustained bactericidal activity
against the pathogenic S. aureus USA300 JE2.
Vancomycin acts by binding solely to lipid II, an important
intermediate for the synthesis of bacterial peptidoglycan.[10,76] On
the other hand, teixobactin is reported to bind to both lipid II and
lipid III, which may account for its remarkable bactericidal activity
over vancomycin.[4,11] Lipid III is the precursor for the synthesis
of wall teichoic acid (WTA) which is covalently attached to
peptidoglycan.[77,78] It is formed by the TarO-mediated
attachment
pyrophosphate.
of
N-acetylglucosamine
Subsequently, N-acetylmannosamine
to
undecaprenyl
is
transferred to lipid III by TarA. The resultant lipid-linked
disaccharide intermediate is then modified by a series of other
Tar enzymes before being transported through the cell
membrane to be linked to the peptidoglycan network.[77,79,80]
Paradoxically, the deletion of genes encoding TarO and/or TarA
has no effects on the in vitro viability of bacteria. These early-
stage enzymes have been deemed non-essential and are useful
for studying potential WTA inhibitors.[80–83]
Under normal circumstances, WTA plays a pivotal role in
anchoring the major autolysin Atl to prevent unregulated self-
digestion. The absence of WTA results in the delocalisation of
this enzyme throughout the cell surface that eventually leads to
bacterial lysis.[84,85] It is anticipated that the rapid bactericidal
activity of [Arg(Me)10,Nle11]teixobactin 63 might also be due to
the synergistic inhibition of both peptidoglycan and WTA
synthesis.
Figure 6. Time-kill curves of P. acnes ATCC 11828 treated with 4x MIC
[Arg(Me)10,Nle11]teixobactin 63 or 4x MIC vancomycin. Data are presented as
the mean log10 viable count (CFU/mL) ± SD of two independent experiments.
Checkerboard assay to determine antimicrobial effect of
teixobactin analogues-colistin combination against P.
aeruginosa
Given the size (M.W. 1200–1300) and hydrophobic nature of the
teixobactin analogues reported herein, it is not surprising that
these compounds failed to display effective antimicrobial activity
against the Gram-negative pathogen, Pseudomonas aeruginosa
(MIC >256 μg/mL, Table 3). The bacterial outer membrane
presents a significant permeability barrier[87] and hence, the
teixobactin analogues are unable to reach their molecular target,
i.e. lipid II, that is located in the periplasm; lipid III and teichoic
acid are not found in Gram-negatives. The Gram-negative outer
membrane is an asymmetric bilayer comprised of
To demonstrate the significance of WTA for bacterial
survival, we determined the killing kinetics of both
[Arg(Me)10,Nle11]teixobactin and vancomycin (4x MIC) in the
presence of sub-lethal tunicamycin (0.05x MIC),
a highly
selective inhibitor of TarO[86]. At the sub-MIC concentration,
tunicamycin showed no effects on cell viability and growth of S.
aureus USA300 JE2 (Figure 5(a)), though it is predicted that
WTA production is substantially suppressed. In the initial eight
hours, the bactericidal profile was very similar for vancomycin
regardless of the presence of tunicamycin. However, the viable
count was reduced to a greater extent (<0.1 log10 CFU/mL) at 24
h when vancomycin was used with tunicamycin compared to
vancomycin alone (ca. 1 log10 CFU/mL remaining). Tunicamycin
lipopolysaccharides (lipid
A and O-antigen moieties) and
glycerophospholipids. Colistin, a cationic polymyxin antibiotic, is
known to disrupt Gram-negative outer membrane via its
interactions
with
anionic
lipopolysaccharides
and
phospholipids.[88] Thus, we hypothesized that the membrane
disruptive capability of colistin would enable permeation of
teixobactin analogues through the pseudomonal outer
membrane and hence ‘restore’ antimicrobial susceptibility.
Indeed, growth inhibition was observed when several of
the analogues were tested in combination with 0.5x MIC colistin
against P. aeruginosa PAO1 (Figure 7). In fact, the MIC values
of (Arg10,Nle11)teixobactin 57 and (Tfn10,Nle11)teixobactin 66
also
enhanced
the
killing
activity
of
4x
MIC
[Arg(Me)10,Nle11]teixobactin 63 since there was no regrowth of
bacteria throughout the 24 h (Figure 5(a)). This suggests that
This article is protected by copyright. All rights reserved.

10.1002/chem.201801423
Chemistry - A European Journal
FULL PAPER
were markedly reduced from >256 μg/mL to 64 μg/mL and 32
µg/mL, respectively.
susceptibility of (Tfn10,Nle11)teixobactin 66 by sub-MIC level of
colistin offers, for the first time, an unique therapeutic option for
the treatment of infections caused by MDR P. aeruginosa and
possibly other Gram-negative pathogens.
(a)
(b)
Figure 7. Antimicrobial susceptibility testing of analogues (Arg10)teixobactin 55,
(Lys10)teixobactin 56, (Arg10,Nle11)teixobactin 57, (Arg10,Nva11)teixobactin 58,
(ADMA10,Nle11)teixobactin 64, (SDMA10,Nle11)teixobactin 65 and (Tfn10,Nle11)-
teixobactin 66 in combination with colistin (2 μg/mL, 0.5x MIC) against P.
aeruginosa PAO1. Wells A2–A8 and wells B2–B8 were set up as control wells
containing colistin at 4 μg/mL and 2 μg/mL respectively. +ve = media and
bacteria only, -ve = media only.
Figure 8. Checkerboard assay of (a) (Tfn10,Nle11)teixobactin 66 and (b)
[Arg(Me)10,Nle11]teixobactin 63 with varying concentration of colistin. +ve =
media and bacteria only, -ve = media only.
In order to comprehensively establish a synergistic or
additive effect of the test compounds-colistin combination, an
antimicrobial checkerboard assay against P. aeruginosa PAO1
was performed on analogue 66 and our lead compound
[Arg(Me)10,Nle11]teixobactin 63 (Figure 8); the results for
analogue 57 and the comparator antibiotic vancomycin can be
found in Supporting Information (Figure S1, p. S36). Surprisingly,
despite its high potency against the tested Gram-positive
pathogens, compound 63 remained inactive at 256 μg/mL even
when combined with colistin (Figure 8(b)). The vancomycin (256
µg/mL)-colistin (0.125–0.5x MIC) combinations were similarly
found to be ineffective. In sharp contrast, when
(Tfn10,Nle11)teixobactin 66 was used in combination with colistin,
Conclusions
In summary, a series of 14 unique teixobactin analogues have
been designed and synthesized to critically examine the roles of
the residues at positions 9, 10 and 11 within the macrocyclic
structure. The antimicrobial activity of the analogues was
determined against a panel of clinically important S. aureus
isolates, including the highly virulent USA300 JE2 and Mu50 (a
vancomycin intermediate-resistant S. aureus, VISA) strains. The
teixobactin analogues were also tested against a panel of P.
acnes strains and they were found to be more potent than
against S. aureus. The in vitro antimicrobial effect of these
analogues provided valuable SARs information and
[Arg(Me)10,Nle11]teixobactin 63 (IC50 = 3.84 ± 0.26 μM against S.
aureus SH1000; MIC 2–4 μg/mL against 5 different S. aureus
strains and 3 different P. acnes strains) has been identified as a
an additive effect (FICI
= 0.63) was observed in the
checkerboard assay against P. aeruginosa PAO1 (Figure 8(a)).
Multidrug resistant (MDR) P. aeruginosa is increasingly
responsible for infection in the critically ill patients. In many of
these cases, colistin is used as the last resort treatment option.
However, the use of high doses of colistin is limited by its
nephrotoxicity.[88] The pronounced restoration of antimicrobial
lead
candidate.
The
impressive
bactericidal
activity
(characterised by the rate of kill) of compound 63 at 10x MIC,
exceeding that of vancomycin, further reinforces its therapeutic
potential. Importantly, when used in combination with an outer
This article is protected by copyright. All rights reserved.

10.1002/chem.201801423
Chemistry - A European Journal
FULL PAPER
[16] V. Zeller, A. Ghorbani, C. Strady, P. Leonard, P. Mamoudy, N.
Desplaces, J. Infect. 2007, 55, 119–124.
membrane-disruptive antibiotic, colistin at 0.5x MIC, the
antimicrobial activity of a subset of teixobactin analogues
against P. aeruginosa was ‘restored’, thereby providing a
potential option for the treatment of infections caused by MDR P.
aeruginosa and possibly other Gram-negative pathogens. In
light of the pronounced antimicrobial effectiveness of the
[Tfn10,Nle11]teixobactin 66 (32 μg/mL)-colistin (2 μg/mL; 0.5x
MIC) combination, its potency should be determined against a
wider spectrum of clinical isolates of not only P. aeruginosa but
also MDR Acinetobacter baumannii and Klebsiella pneumoniae.
Furthermore, the antibiotic combination should be evaluated
against colistin-resistant Gram-negatives. These future studies
will be reported in due course.
[17] M. Rohacek, M. Weisser, R. Kobza, A. W. Schoenenberger, G. E.
Pfyffer, R. Frei, P. Erne, A. Trampuz, Circulation 2010, 121, 1691–1697.
[18] Y. Achermann, F. Sahin, H. K. Schwyzer, C. Kolling, J. Wüst, M. Vogt,
Infection 2013, 41, 613–620.
[19] M. Patel, W. P. Bowe, C. Heughebaert, A. R. Shalita, J Drugs Dermatol.
2010, 9, 655–664.
[20] K. Nakase, H. Nakaminami, N. Noguchi, S. Nishijima, M. Sasatsu, J.
Dermatol. 2012, 39, 794–796.
[21] G. A. Denys, R. C. Jerris, J. M. Swenson, C. Thornsberry, Antimicrob.
Agents Chemother. 1983, 23, 335–337.
[22] S. H. Moon, H. S. Roh, Y. H. Kim, J. E. Kim, J. Y. Ko, Y. S. Ro, J.
Dermatol. 2012, 39, 833–837.
[23] Y. E. Jad, G. A. Acosta, T. Naicker, M. Ramtahal, A. El-Faham, T.
Govender, H. G. Kruger, B. G. de la Torre, F. Albericio, Org. Lett. 2015,
17, 6182–6185.
Experimental section
[24] H. Yang, K. H. Chen, J. S. Nowick, ACS Chem. Biol. 2016, 11, 1823–
1826.
For full experimental procedures, spectroscopic and analytical
data for all new compounds, including copies of NMR spectra,
see the Supporting Information.
[25] A. Parmar, A. Iyer, C. S. Vincent, D. Van Lysebetten, S. H. Prior, A.
Madder, E. J. Taylor, I. Singh, Chem. Commun. 2016, 52, 6060–6063.
[26] A. Parmar, S. H. Prior, A. Iyer, C. S. Vincent, D. Van Lysebetten, E.
Breukink, A. Madder, E. J. Taylor, I. Singh, Chem. Commun. 2017, 53,
2016–2019.
Acknowledgements
[27] H. Yang, D. R. Du Bois, J. W. Ziller, J. S. Nowick, Chem. Commun.
2017, 53, 2772–2775.
We would like to acknowledge financial support from Vice-
Chancellor’s Scholarship for Research Excellence, University of
Nottingham (to VN). We thank Dr Andrew McDowell, University
of Ulster for the kind gift of the Propionibacterium acnes strains
and for expert advice.
[28] C. Wu, Z. Pan, G. Yao, W. Wang, L. Fang, W. Su, RSC Adv. 2017, 7,
1923–1926.
[29] (a) K. Jin, I. H. Sam, K. H. L. Po, D. Lin, E. H. G. Zadeh, S. Chen, Y.
Yuan, X. Li, Nature Commun. 2016, 12394; (b) K. Jin, K. H. L. Po, S.
Wang, J. A. Reuven, C. N. Wai, H. T. Lau, T. H. Chan, S. Chen, X. Li,
Bioorg. Med. Chem. 2017, 25, 4990–4995; (c) K. Jin, K. H. L. Po, W. Y.
Kong, C. H. Lo, C. W. Lo, H. Y. Lam, A. Sirinimal, J. A. Reuven, S.
Chen, X. Li, Bioorg. Med. Chem. 2018, 26, 1062–1068.
[30] S. A. H. Abdel Monaim, Y. E. Jad, G. A. Acosta, T. Naicker, E. J.
Ramchuran, A. El-Faham, T. Govender, H. G. Kruger, B. G. de la Torre,
F. Albericio, RSC Adv. 2016, 6, 73827–73829.
Keywords: macrocyclic peptides • lipid II inhibitors • teixobactin
• antimicrobial • colistin
[1]
A. P. Johnson, D. Ashiru-Oredope, E. Beech, Antibiotics 2015, 4, 467–
479.
[31] S. A. H. Abdel Monaim, Y. E. Jad, E. J. Ramchuran, A. El-Faham, T.
Govender, H. G. Kruger, B. G. de la Torre, F. Albericio, ACS Omega
2016, 1, 1262–1265.
[2]
[3]
[4]
S. Doron, L. E. Davidson, Mayo Clin. Proc. 2011, 86, 1113–1123.
D. A. Goff, M. J. Rybak, Infect. Dis. Ther. 2015, 4, 1–3.
L. L. Ling, T. Schneider, A. J. Peoples, A. L. Spoering, I. Engels, B. P.
Conlon, A. Mueller, T. F. Schäberle, D. E. Hughes, S. Epstein, et al.,
Nature 2015, 517, 455–459.
[32] C. E. Schumacher, P. W. R. Harris, X.-B. Ding, B. Krause, T. H. Wright,
G. M. Cook, D. P. Furkert, M. A. Brimble, Org. Biomol. Chem. 2017, 15,
8755–8760.
[33] S. A. H. Abdel Monaim, S. Noki, E. J. Ramchuran, A. El-Faham, F.
Albericio, B. G. de la Torre, Molecules 2017, 22, 1632.
[34] K. H. Chen, S. P. Le, X. Han, J. M. Frias, J. S. Nowick, Chem. Commun.
2017, 53, 11357–11359.
[5]
[6]
[7]
F. R. DeLeo, M. Otto, B. N. Kreiswirth, H. F. Chambers, Lancet 2010,
375, 1557–1568.
K. Alrabiah, S. Al Alola, E. Al Banyan, M. Al Shaalan, S. Al Johani, Int.
J. Pediatr. Adolesc. Med. 2016, 3, 71–77.
[35] S. A. H. Abdel Monaim, E. J. Ramchuran, A. El-Faham, F. Albericio, B.
G. de la Torre, J. Med. Chem. 2017, 60, 7476–7482.
B. P. Howden, J. K. Davies, P. D. R. Johnson, T. P. Stinear, M. L.
Grayson, Clin. Microbiol. Rev. 2010, 23, 99–139.
[36] A. Parmar, A. Iyer, S. H. Prior, D. G. Lloyd, E. T. L. Goh, C. S. Vincent,
T. Palmai-Pallag, C. Z. Bachrati, E. Breukink, A. Madder, et al., Chem.
Sci. 2017, 8, 8183–8192.
[8]
[9]
H. Hanaki, K. Hiramatsu, Jpn J Antibiot. 1997, 50, 794–798.
H. Hanaki, H. Labischinski, Y. Inaba, N. Kondo, H. Murakami, K.
Hiramatsu, J. Antimicrob. Chemother. 1998, 42, 315–320.
[37] A. Parmar, A. Iyer, D. G. Lloyd, C. S. Vincent, S. H. Prior, A. Madder, E.
J. Taylor, I. Singh, Chem. Commun. 2017, 53, 7788–7791.
[38] S. A. H. Abdel Monaim, Y. E. Jad, A. El-Faham, B. G. de la Torre, F.
[10] S. Fletcher, W. Yu, J. Huang, S. M. Kwasny, J. Chauhan, T. J.
Opperman, A. D. MacKerell, E. P. H. de Leeuw, Drug Des. Devel. Ther.
2015, 9, 2383–2394.
Albericio,
Bioorganic
Med.
Chem.
2017,
DOI
[11] T. Homma, A. Nuxoll, A. B. Gandt, P. Ebner, I. Engels, T. Schneider, F.
Götz, K. Lewis, B. P. Conlon, Antimicrob. Agents Chemother. 2016, 60,
6510–6517.
10.1016/j.bmc.2017.09.040.
[39] C. Guo, D. Mandalapu, X. Ji, J. Gao, Q. Zhang, Chem. Eur. J. 2018, 24,
5406–5422.
[12] H. C. Williams, R. P. Dellavalle, S. Garner, Lancet 2012, 379, 361–372.
[13] Y. Achermann, E. J. C. Goldstein, T. Coenye, M. E. Shirtliff, Clin.
Microbiol. Rev. 2014, 27, 419–440.
[40] Y. Liu, Y. Liu, M. B. Chan-Park, Y. Mu, Sci. Rep. 2017, 7, 17197.
[41] V. Ng, W. C. Chan, Chem. Eur. J. 2016, 22, 12606–12616.
[42] A. E. Sorochinsky, J. L. Aceña, H. Moriwaki, T. Sato, V. A. Soloshonok,
Amino Acids 2013, 45, 691–718.
[14] E. Jakab, R. Zbinden, J. Gubler, C. Ruef, A. von Graevenitz, M. Krause,
1996, 69, 477–482.
[43] Y. N. Belokon, V. I. Tararov, V. I. Maleev, T. F. Savel’eva, M. G.
Ryzhov, Tetrahedron: Asymmetry 1998, 9, 4249–4252.
[15] V. A. Deramo, T. D. Ting, Curr. Opin. Ophthalmol. 2001, 12, 225–229.
This article is protected by copyright. All rights reserved.

10.1002/chem.201801423
Chemistry - A European Journal
FULL PAPER
[44] M. Jörres, X. Chen, J. L. Aceña, C. Merkens, C. Bolm, H. Liu, V. A.
Soloshonok, Adv. Synth. Catal. 2014, 356, 2203–2208.
[45] J. Wang, D. Lin, S. Zhou, X. Ding, V. A. Soloshonok, H. Liu, J. Org.
Chem. 2011, 76, 684–687.
[66] M. Jagodzinska, F. Huguenot, G. Candiani, M. Zanda, ChemMedChem
2009, 4, 49–51.
[67] K. Müller, C. Faeh, F. Diederich, Science. 2007, 317, 1881–1886.
[68] F. Leroux, ChemBioChem 2004, 5, 644–649.
[46] Y. Nian, J. Wang, S. Zhou, W. Dai, S. Wang, H. Moriwaki, A.
Kawashima, V. A. Soloshonok, H. Liu, J. Org. Chem. 2016, 81, 3501–
3508.
[69] M. Evich, E. Stroeva, Y. G. Zheng, M. W. Germann, Protein Sci. 2016,
25, 479–486.
[70] J. M. Humphrey, A. R. Chamberlin, Chem. Rev. 1997, 97, 2243–2266.
[71] G. Jung, Angew. Chem. Int. Ed. Engl. 1991, 30, 1051–1068.
[72] D. Siodłak, Amino Acids 2015, 47, 1–17.
[47] R. Takeda, A. Kawamura, A. Kawashima, T. Sato, H. Moriwaki, K.
Izawa, K. Akaji, S. Wang, H. Liu, J. L. Aceña, et al., Angew. Chem. Int.
Ed. 2014, 53, 12214–12217.
[73] F. F. Arhin, I. Sarmiento, A. Belley, G. A. McKay, D. C. Draghi, P.
Grover, D. F. Sahm, T. R. Parr, G. Moeck, Antimicrob. Agents
Chemother. 2008, 52, 1597–1603.
[48] V. A. Soloshonok, T. K. Ellis, H. Ueki, T. Ono, J. Am. Chem. Soc. 2009,
131, 7208–7209.
[49] A. El-Faham, R. Subirós Funosas, R. Prohens, F. Albericio, Chem. Eur.
J. 2009, 15, 9404–9416.
[74] R. P. Rennie, L. Koeth, R. N. Jones, T. R. Fritsche, C. C. Knapp, S. B.
Killian, B. P. Goldstein, J. Clin. Microbiol. 2007, 45, 3151–3154.
[75] NCCLS, Methods for Determining Bactericidal Activity of Antimicrobial
Agents; Approved Guideline, NCCLS Document M26-A, National
Committee For Clinical Laboratory Standards, Wayne, PA, 1999.
[76] J. C. Barna, D. H. Williams, Annu. Rev. Microbiol. 1984, 38, 339–57.
[77] S. Brown, Y.-H. Zhang, S. Walker, Chem. Biol. 2008, 15, 12–21.
[78] F. C. Neuhaus, J. Baddiley, Microbiol. Mol. Biol. Rev. 2003, 67, 686–
723.
[50] A. El-Faham, F. Albericio, J. Pept. Sci. 2010, 16, 6–9.
[51] CLSI, Methods for Dilution Antimicrobial Susceptibility Tests for
Bacteria That Grow Aerobically; Approved Standard — Ninth Edition,
CLSI Document M07-A9, Clinical And Laboratory Standards Institute,
Wayne, PA, 2012.
[52] M. D. King, B. J. Humphrey, Y. F. Wang, E. V. Kourbatova, S. M. Ray,
H. M. Blumberg, Ann. Intern. Med. 2006, 144, 309–317.
[53] F. C. Tenover, R. V. Goering, J. Antimicrob. Chemother. 2009, 64,
441–446.
[79] A. Müller, H. Ulm, K. Reder-Christ, H.-G. Sahl, T. Schneider, Microb.
Drug Resist. 2012, 18, 261–270.
[54] M. J. Horsburgh, J. L. Aish, I. J. White, L. Shaw, J. K. Lithgow, S. J.
Foster, J. Bacteriol. 2002, 184, 5457–5467.
[80] J. G. Swoboda, T. C. Meredith, J. Campbell, S. Brown, T. Suzuki, T.
Bollenbach, A. J. Malhowski, R. Kishony, M. S. Gilmore, S. Walker,
ACS Chem. Biol. 2009, 4, 875–883.
[55] T. Baba, T. Bae, O. Schneewind, F. Takeuchi, K. Hiramatsu, J.
Bacteriol. 2008, 190, 300–310.
[81] L. W. Pasquina, J. P. Santa Maria, S. Walker, Curr. Opin. Microbiol.
2013, 16, 531–537.
[56] E. S. Duthie, L. L. Lorenz, J. Gen. Microbiol 1952, 6, 95–107.
[57] P. C. L. Moore, J. A. Lindsay, J. Med. Microbiol. 2002, 51, 516–521.
[58] P. D. Fey, J. L. Endres, V. K. Yajjala, T. J. Widhelm, R. J. Boissy, J. L.
Bose, W. Bayles, MBio 2013, 4, e00537–12.
[82] M. A. D’Elia, J. A. Henderson, T. J. Beveridge, D. E. Heinrichs, E. D.
Brown, J. Bacteriol. 2009, 191, 4030–4034.
[83] J. Campbell, A. K. Singh, J. G. Swoboda, M. S. Gilmore, B. J. Wilkinson,
S. Walker, Antimicrob. Agents Chemother. 2012, 56, 1810–1820.
[84] M. Schlag, R. Biswas, B. Krismer, T. Kohler, S. Zoll, W. Yu, H. Schwarz,
A. Peschel, F. Götz, Mol. Microbiol. 2010, 75, 864–873.
[59] M. Kuroda, T. Ohta, I. Uchiyama, T. Baba, H. Yuzawa, I. Kobayashi, L.
Cui, A. Oguchi, K. Aoki, Y. Nagai, et al., Lancet 2001, 357, 1225–1240.
[60] B. Horváth, J. Hunyadkürti, A. Vörös, C. Fekete, E. Urbán, L. Kemény, I.
Nagy, J. Bacteriol. 2012, 194, 202–203.
[85] S. H. Lee, H. Wang, M. Labroli, S. Koseoglu, P. Zuck, T. Mayhood, C.
Gill, P. Mann, X. Sher, S. Ha, et al., Sci. Transl. Med. 2016, 8, 329ra32.
[86] J. Campbell, A. K. Singh, J. P. Santa Maria, Y. Kim, S. Brown, J. G.
Swoboda, E. Mylonakis, B. J. Wilkinson, S. Walker, ACS Chem. Biol.
2011, 6, 106–116.
[61] I. Nagy, A. Pivarcsi, K. Kis, A. Koreck, L. Bodai, A. McDowell, H.
Seltmann, S. Patrick, C. C. Zouboulis, L. Kemény, Microbes Infect.
2006, 8, 2195–2205.
[62] E. Barnard, J. Liu, E. Yankova, S. M. Cavalcanti, M. Magalhães, H. Li,
S. Patrick, A. McDowell, Sci. Rep. 2016, 6, 31968.
[87] G. Krishnamoorthy, I. V Leus, J. W. Weeks, D. Wolloscheck, V. V.
Rybenkov, H. I. Zgurskaya, MBio 2017, 8, e01172–17.
[63] A. McDowell, A. L. Perry, P. A. Lambert, S. Patrick, J. Med. Microbiol.
2008, 57, 218–224.
[88] F. Rabanal, Y. Cajal, Nat. Prod. Rep. 2017, 34, 886–908.
[64] B. W. Holloway, J. Gen. Microbiol. 1955, 13, 572–581.
[65] M. Mueller, A. de la Peña, H. Derendorf, Antimicrob. Agents Chemother.
2004, 48, 369–377.
This article is protected by copyright. All rights reserved.

10.1002/chem.201801423
Chemistry - A European Journal
FULL PAPER
Entry for the Table of Contents (Please choose one layout)
Layout 1:
FULL PAPER
Text for Table of Contents
Author(s), Corresponding Author(s)*
Page No. – Page No.
Title
((Insert TOC Graphic here: max.
width: 5.5 cm; max. height: 5.0 cm))
Layout 2:
FULL PAPER
Author(s), Corresponding Author(s)*
Page No. – Page No.
Title
Breaking through the superbugs! Teixobactin analogues: the potential solution to
both Gram-positive and Gram-negative infections.
This article is protected by copyright. All rights reserved.