Fluorinated O-phenylserine residues enhance the broad-spectrum antimicrobial activity of ultrashort cationic lipopeptides

Article abstract of DOI:10.1016/j.jfluchem.2020.109685

As a crucial structural element in many antimicrobial lipopeptides, phenylalanine poses a potential, but sensitive, point of modification in the pursuit of enhancing the antimicrobial efficacy of peptide antibiotics. We report the synthesis and applicatio

Full text of DOI:10.1016/j.jfluchem.2020.109685

Journal of Fluorine Chemistry 241 (2021) 109685
Contents lists available at ScienceDirect
Journal of Fluorine Chemistry
journal homepage: www.elsevier.com/locate/fluor
Fluorinated O-phenylserine residues enhance the broad-spectrum
antimicrobial activity of ultrashort cationic lipopeptides
a
a
a
a
Hugh D. Glossop , Gayan Heruka De Zoysa , Lisa I. Pilkington , David Barker ,
a,b,
Vijayalekshmi Sarojini
*
a
School of Chemical Sciences, The University of Auckland, Private Bag 92019, Auckland, 1142, New Zealand
The MacDiarmid Institute for Advanced Materials and Nanotechnology, Wellington, 6140, New Zealand
b
A R T I C L E I N F O
A B S T R A C T
Keywords:
As a crucial structural element in many antimicrobial lipopeptides, phenylalanine poses a potential, but sensitive,
point of modification in the pursuit of enhancing the antimicrobial efficacy of peptide antibiotics. We report the
synthesis and application of a library of fluorinated O-phenyl serine analogues, followed by biological testing on
relevant bacterial and fungal pathogens and the comparison of these analogues to commercial fluorinated
phenylalanine amino acids. The synthetic aromatic residues were investigated for their enhanced potency over
the original Phe when incorporated into a model ultrashort lipopeptide sequence derived from the octapeptin
battacin and were found to enhance broad-spectrum antimicrobial activity across the panel of microbial species
Antimicrobial peptides
Cationic peptides
Fluorinated amino acids
Peptide synthesis
Lipopeptide antibiotics
(
while retaining non-hemolytic activity). While the major drawback of antimicrobial peptides is the cost of
synthesis, they are some of the last antimicrobials to be majorly affected by antimicrobial resistance, making
them a promising class of antibiotic for studying. Antimicrobial activity of the most potent fluorinated peptide
was around 16-fold greater than the unfluorinated Phe-peptide against the fire blight pathogen, with milder
increases in potency toward other pathogens such as S. aureus and E. coli. This indicates a potential future di-
rection for these peptides to be developed as crop treatments to protect rosaceae from fire blight.
1
. Introduction
Current antibiotics are in danger of losing effectiveness due to the
self-association of charged peptides is implicated in generating a
threshold cationic charge density at the membrane prior to pore for-
mation [6]. Additionally, the putative contributing mechanism of action
of cationic helical peptides via “lipid clustering” has been recently
described by Epand et al., whereby a sufficient overall cationic charge is
implicated in the sequestering of anionic lipids from zwitterionic lipids
and is considered to contribute to CAMP specificity toward
Gram-negative pathogens such as E. coli [7,8].
progressive development of antimicrobial resistance by pathogenic
bacteria and fungi [1]. This rise in resistance threatens to complicate
medical treatments of infections should no new drugs be discovered. The
agricultural industries may also be threatened by antimicrobial resis-
tance in livestock and plant pathogens [2,3]. To date, several classes of
antibiotics have been all but retired due to overwhelming antimicrobial
resistance and as such the search for new classes of antimicrobials is an
ongoing task. A current promising class of antimicrobials, cationic
antimicrobial peptides (CAMPs) have the unique property of
non-specific targeting selective to bacterial membranes owing to their
electrostatic and amphipathic related mode of action towards negatively
charged components of bacterial membranes [4]. CAMPs cause bacterial
membrane disruption, typically due to their amphipathic properties
constituted by hydrophobic and charged residues, resulting in pore
formation, as has been well studied for the magainins [5]. The
With typically no specific targets such as enzymes or receptors like
many conventional antibiotics, CAMPs are less likely to draw resistance
from single microbial mutations, making peptides a promising future
alternative to small-molecule antibiotics [9]. Many CAMPs are naturally
occurring such as in the innate immune systems of mammals and in-
vertebrates alike [10]. However, more and more semi-synthetic and
fully synthetic antimicrobial sequences have been designed over recent
years. In particular, the structure-activity relationship studies of poly-
myxin B are exemplary of the efforts in the field of medicinal chemistry
to understand and enhance cationic peptide antibiotics [11,12]. These
*
Corresponding author at: School of Chemical Sciences, The University of Auckland, Private Bag 92019, Auckland, 1142, New Zealand.
E-mail address: v.sarojini@auckland.ac.nz (V. Sarojini).
https://doi.org/10.1016/j.jfluchem.2020.109685
Received 24 June 2020; Received in revised form 7 October 2020; Accepted 5 November 2020
Available online 11 November 2020
0
022-1139/© 2020 Elsevier B.V. All rights reserved.

H.D. Glossop et al.
Journal of Fluorine Chemistry 241 (2021) 109685
studies and many others confirm that net cationic charge and amphi-
pathicity are crucial factors in the activity of this class of antibiotic. In
addition, the presence of unnatural or D-amino acids further enhanced
efficacy by improving the stability of peptide therapeutics as these un-
natural residues have less affinity as substrates for endogenous proteases
demonstrated via Mitsunobu reactions of phenols with the serine side
chain alcohol previously [23,29]. This allowed for us to selectively
append phenolic fragments to serine via the ether linkage without the
necessitation of harsher conditions or overly dangerous reagents.
As fluorinated amino acids have previously been shown to enhance
antimicrobial activity in longer peptides, usually by contributing to
secondary structure formation and enhancing protease stability of the
peptide [17,30], we selected a modest number of fluorinated phenols to
append to serine (Fig. 1) including 4-fluorophenol (1.1), 3,5-difluoro-
phenol (1.2), 3,4,5-trifluorophenol (1.3), 2,3,5,6-tetrafluorophenol
(1.4), and pentafluorophenol (1.5). In addition to this, peptides con-
taining 4-fluorophenylalanine and pentafluorophenylalanine were also
included in the peptide library for comparison as these amino acids are
both commercially available and are fluorinated phenylalanine de-
rivatives commonly used in peptide synthesis [15,31]. Our intention was
to synthesise fluorinated O-phenyl Fmoc-Serꢀ OH derivatives (Fig. 1)
starting from N-tritylated serine methyl ester (Scheme 2), to be used in
conventional Fmoc solid-phase peptide synthesis of a library of peptides
bearing these unnatural fluorinated residues (Table 1) to ascertain the
effect, if any, of an increasing number of fluorines on this unnatural
residue in the model peptide.
[
13].
Fluorination is a common modification in medicinal chemistry, and
has found use in the field of AMPs via the incorporation of fluorinated
amino acids into peptide sequences such as the 5,5,5-trifluoroleucine
melittin analogue reported by Niemz and Tirrell [14]. Magainin ana-
logues have been well studied as scaffolds for incorporating fluorinated
residues [15]. The highly fluorinated magainin analogue,
fluorogainin-1, exhibited enhanced proteolytic stability up to 10 h
compared to its unfluorinated variant, which degraded within 30 min
[
16]. Other prominent fluorinated peptide studies include that of the
combined library of fluorinated buforin and magainin analogues which
had moderate proteolytic resistance and improved antimicrobial activ-
ity, but with higher hemolytic activity observed for the magainin library
[
17]. Fluorinated residues have found plentiful use in helical peptide
design to stabilise the secondary structure [18,19] as well as in β-hairpin
sequences such as protegrin-1 analogues [20].
In this scope, we have designed a library of ultrashort CAMPs based
on the battacin lipopeptide pentapeptide scaffold with the express pur-
pose of substituting the key D-phenylalanine residue with fluorinated
aromatic L-serine derivatives. To date, O-phenyl serine derivatives
While all phenols were successfully conjugated to serine, and further
converted to Fmoc protected residues (Supplementary Data), the
2,3,5,6-tetrafluorophenyl and 2,3,4,5,6-pentafluorophenyl derivates
were unfortunately too unstable for use in typical Fmoc solid phase
peptide synthesis. Attempted syntheses of pentapeptides incorporating
these Fmoc-protected amino acids resulted in major peptide products
lacking fluorinated or aromatic functionality (aside from TFA counter-
ions) post cleavage. This is likely due to beta-elimination during solid-
phase synthesis because pentafluorophenyl esters and ethers are
known to be good leaving groups, and the tetrafluorophenyl moiety
likely exhibits a similar property [32,33].
(
including halogenated variants) have been reported in several synthe-
ses including that of dopamine β-hydroxylase inhibitors [21], biphenyls
22], and substituted pyrrole-2-ones [23]. However, to the best of our
[
knowledge, the use of these serine derivatives themselves as residues in
synthetic antimicrobial peptides has yet to be explored. For this reason,
we see the potential of these amino acids as bioisosteric substitutes for
aromatic residues such as phenylalanine with potential for use in other
aspects of medicinal chemistry in structural modifications of peptide
therapeutics. Additionally, as unnatural halogenated residues, these
amino acids may infer stability to the peptide as well as enhancing
physical properties such as hydrophobicity and halogen-related in-
teractions in aromatic systems [24,25]. Fluorinated amino acids are of
great interest to medicinal chemists as building blocks or starting ma-
terials in syntheses due to their typically enhanced bioactivities
compared to non-fluorinated counterparts, which we wish to explore in
the field of antimicrobial peptide research [26].
NMR characterisation was in support of this hypothesis, revealing
that the major peptide product yielded by the syntheses involving the
pentafluoro and tetrafluorophenoxy residues in fact contained dehy-
droalanine (Supplementary information), a product of beta-elimination
(Scheme 1), also seen for phosphoserine [34,35]. This
dehydroalanine-containing peptide did not have antimicrobial activity
(not shown). Incorporation of these residues into the peptide sequence
may be achieved through use of a different N-terminal protecting group
of the tetra or pentafluorophenoxy residue and a fragment-based
approach to coupling halves of the peptide to avoid the need for basic
treatment to remove Fmoc. However, this falls outside of the intentions
of this study to produce conventionally protected amino acids for Fmoc
solid phase peptide synthesis. We have included NMR spectra of the
tetra and pentafluorophenyl residues (Supplementary information) to
assist any future endeavours in synthesising and applying these amino
acids.
The linear battacin pentapeptide previously identified by our group
[
27] was used as a model peptide sequence for biological testing. Ana-
logues were generated by substituting the original D-Phe residue with
the fluorinated L-Phe and modified L-phenylserine residues. L-amino
acids were chosen due to their ubiquity in nature and an unfluorinated
L-Phe containing peptide was included as a chirality control. Synthesis of
these serine derivatives was effected via Mitsunobu displacement, a
highly versatile reaction that allows for modular syntheses wherein
commercially available fluorinated phenol fragments may be appended
to a seryl alcohol early on in the synthesis, rather than having manually
fluorinate a late-stage product. Both the modified synthesis and bio-
logical testing outcomes are discussed in full in later sections.
2
. Results and discussion
2
.1. Amino acid and peptide design rationale
The previous structure-activity relationship studies that we had un-
dertaken on linear battacin lipopeptides revealed that the hydrophobic
residues in the middle of the sequence, Leu-D-Phe, were crucial for
antimicrobial efficacy, much like the hydrophobic dipeptide motif
similarly found in other lipopeptide families such as the polymyxins and
octapeptins [11,27,28]. Endeavouring to improve antimicrobial potency
via a single amino acid modification, we selected the D-Phe residue as a
point of modification because synthesis of aromatic mimics had been
Fig. 1. Chemical structures of (1) the five fluorinated phenols originally
selected for attachment to serine via Mitsunobu reaction in Scheme 2 and (2)
the generic structure of the intended final Fmoc-serine derivatives.
2

H.D. Glossop et al.
Journal of Fluorine Chemistry 241 (2021) 109685
Table 1
MIC values (
μ
g/mL) of the peptide library against the panel of pathogens.
Minimal inhibitory concentration ( g/mL)
μ
Name
S. aureus
P. aeruginosa
E. coli
C. albicans
C. jejuni
B. cereus
E. amylovora (Ea1501)
E. amylovora (Str4Ea)
Psa16027
1
2
3
4
5
6
7
8
3.1
6.3
1.6
1.6
3.1
6.3
6.3
6.3
0.8
–
50
25
100
100
50
6.3
6.3
3.1
3.1
3.1
25
100
100
50
12.5
25
6.3
6.3
6.3
6.3
3.1
6.3
12.5
6.3
6.3
1.6
1.6
–
50
50
25ꢀ 50
6.3
25ꢀ 50
25
12.5
12.5
12.5
50
12.5
12.5
12.5
100
100
50
50
25
6.3
25
100
>100
>100
>100
–
25
12.5
100
100
50
100
100
100
–
<0.2
–
100
100
100
3.1
–
50
12.5
12.5
–
1.6
–
50
Streptomycin
Polymyxin B
Amphotericin B
–
0.8
>100
<0.2
–
<0.2
–
–
1.6
<0.2
–
–
–
Scheme 1. Proposed mechanism of piperidine-mediated beta elimination of the O-pentafluorophenyl serine residue to form dehydroalanine.
Scheme 2. General synthetic scheme for the preparation of O-phenyl serine derivatives. A) Mitsunobu displacement (DIAD, triphenylphosphine, phenol derivative),
3 2
column chromatography (toluene); B) 5-6 N HCl (aq), reflux, aqueous workup; C) Fmoc-OSu, NaHCO , H O/dioxane, acidification, liquid extraction, recrystalli-
zation (DCM/toluene).
2
.2. O-Phenylserine-containing peptides exhibit similar circular dichroism
2.3. Fluorine modification retains or enhances antimicrobial activity of
the peptide library depending on the microbial species
patterns
When analysed by circular dichroism (CD) spectroscopy in pure
Minimal inhibitory concentration (MIC) screening of the pentapep-
tide (Fig. 2) library against a variety of pathogens (Table 1) reveals that
all substitutions of residue 4 retained or enhanced antimicrobial activity
compared to the original pentapeptide (D-Phe), but to varying magni-
tudes for each species. Peptide 7 (L-Phe) and 8 (O-phenyl-L-serine) had
virtually the same activity as the original D-Phe peptide 6, eliminating
both chirality and the O-phenyl ether of residue 4 as a major antimi-
crobial factor in this library. Overall, decreases in the MICs for fluori-
nated derivatives varied widely in the range of around 2–4 times
(compared to the D-Phe pentapeptide) against P. aeruginosa, E. coli,
B. cereus, and Psa. P. aeruginosa displayed higher susceptibility towards
water, peptides 2, 5, 6 and 7 exhibited the expected spectra for unor-
dered structure with a primary minimum at around 197 nm. However,
the O-phenylserine containing peptides 1, 3, 4, and 8 exhibited a new
minimum around 200ꢀ 202 nm as well as a new peak at around 190 nm
(
Fig. 3).
This difference in the spectra is most likely due to the difference
between the aromatic residue structures rather than any particular
secondary structure formation, given that linear peptides of this class
present unstructured CD spectra, even in high concentrations of tri-
fluoroethanol that would usually induce secondary structures [27].
These deviations in CD spectra are then possibly due to the different
properties of the O-phenylserine residues compared to the phenylala-
nine derivatives. However, there is a dearth of data to support why the
phenyl ether groups of 1, 3, 4, and 8 present the additional CD maxima
and minima observed in these peptides and this phenomenon must be
pursued further in future work.
the pentafluoro-L-Phe substitution (peptide 5, MIC: 25
compared to the monofluorinated variant (peptide 2, MIC: 50
μ
g/mL)
g/mL)
μ
and the unfluorinated pentapeptide (peptide 6, MIC: 100
μg/mL), which
was a pattern observed for E. coli too to a lesser degree when compared
to the rest of the pathogens (Table 1). This suggests that the fluorine
content on the aromatic residue itself is a major driving point behind
3

H.D. Glossop et al.
Journal of Fluorine Chemistry 241 (2021) 109685
the cationic peptides described, which likely act in a similar mechanism
as CAMPs against other Gram-negative pathogens. All fluorinated pep-
tides in the library were more potent than the non-fluorinated 6 by
around 4–8 times, and all exhibited a similar range of MICs, indicating
that C. jejuni is not necessarily sensitive to any of the number of fluorines
of the residues, but rather the presence of an unnatural structure bearing
any fluorine in the first place.
The synthetic modifications of the library had the greatest effect on
inhibition of E. amylovora out of all the pathogens in the panel with
analogues exhibiting up to 16-times more potency than the non-
fluorinated control peptides 6 and 7 (MIC: 100 g/mL). Intriguingly,
μ
the streptomycin-resistant E. amylovora strain showed sensitivity toward
peptides containing fluorinated O-phenyl serine residues (1, 3, and 4)
over the 4-fluoro-L-Phe and pentafluoro-L-Phe residues (2 and 5). An
example of this is between peptide 1, featuring O-(4-fluorophenyl)-L-
serine and 2, instead containing 4-fluoro-L-phenylalanine. Against the
streptomycin resistant E. amylovora strain Str4Ea, 1 (MIC: 6.3
μ
g/mL) is
a more potent inhibitor compared to 2 (25ꢀ 50 g/mL) despite the only
μ
structural difference between the two being the ether linkage between
the 4-fluorophenyl moiety and the peptide backbone. While the overall
MIC data suggests that the phenyl ether linkage of the O-phenylserine
derivatives is not a major factor in antimicrobial activity, it may still
contribute to a lesser degree as evidenced by this sensitivity in
E. amylovora between Phe and O-phenyl groups.
Fig. 2. Chemical structure of the model pentapeptide scaffold. R denotes po-
While no prior work could be found relating the differences between
the Phe side chain and the phenoxyserine moiety towards antimicrobial
efficacy, we posit that the ether linkage may add a further degree of un-
naturality to the residue (in addition to fluorine present), which may in
turn enhance proteolytic stability of the peptides against bacterial en-
zymes, given how unnatural and D-amino acids are considered to impart
this property too [41]. The effect of the phenyl ether linkage on the
amphipathic profile of the peptide is not to be overlooked, either from
the slight polarity of the ether linkage or its potential to form hydrogen
bonding with other moieties. However, this is not certain and will need
to be further investigated in future studies to elucidate the mechanism of
how this relatively minor modification confers such a high degree of
antimicrobial activity against E. amylovora. Overall, the library of fluo-
rinated peptides exhibited broad-spectrum activity across strains of
Gram-positive, Gram-negative, and a fungal species. This is especially
promising for the clinical setting where acquired infections may be
caused by more than one pathogen.
sition of the synthetic residue that differs between analogues.
enhancing inhibition against these two species compared to the rest of
the microbial panel, as discussed later. Inhibition of S. aureus was
enhanced by a similar magnitude by the synthetic modifications of 3 and
4
(MIC: 1.6
.3 g/mL). Psa inhibition was also moderately affected by the synthetic
g/mL)
g/mL), and overall being close to the activity of
the unmodified sequence 6 (MIC: 12.5 g/mL). It appears that sub-
μ
g/mL both) compared to that of the unfluorinated 6 (MIC:
6
μ
modifications, with all analogues presenting the same MIC (6.3
except for 4 (MIC: 3.1
μ
μ
μ
stitutions of D-Phe with putatively similar residues has less influence on
these species.
Against C. albicans, the only fungal pathogen in the panel, all pep-
tides in the library exhibited inhibitory activity between 50ꢀ 100
mL, whereas the original pentapeptide 6 containing D-Phe was inactive
>100 g/mL) in our tested concentration range both in this study and
in our prior investigations on antifungal lipopeptides in this class (re-
ported MIC: >100 M) [36]. In our previous work on linear battacin
μg/
(
μ
μ
2
.4. The fluorinated peptides are non-hemolytic at biologically relevant
lipopeptides it was found that an octapeptide structure (expanded
pentapeptide 6 sequence) featuring a hydrophobic N-terminal group
such as myristoyl or Fmoc, or dimer/trimer octapeptide repeat was
required to potentiate C. albicans inhibition for this class of lipopeptides
concentrations
Hemolysis is a common concern regarding AMPs [42] and must be
assessed early prior to any in vivo testing. The hemolytic activity of the
most active fluorinated peptides 3, 4, and 5 was investigated with mouse
red blood cells (Fig. 4), with 6 serving as the non-fluorinated control as
the original sequence. By comparing hemolytic activity of the peptides
against that of Triton X-100 (100 % hemolysis control), it was revealed
that 3, 4, 5, and 6 were virtually non-hemolytic at all concentrations
(
lowest MIC: 6.25
μ
M) [36]. For this reason, the observed activity
(
MIC ≤ 100 g/mL) of the synthetically modified pentapeptide ana-
μ
logues in this library against C. albicans arising from singular amino acid
substitution of D-Phe with the fluorinated residues is unexpected. This
clearly shows that using fluorinated amino acids in peptides can be a
beneficial strategy to induce antifungal activity in short peptides, which
is encouraging for the clinical setting, especially for nosocomial fungal
infections.
tested (highest concentration: 250
hemolysis at 250 g/mL, whereas 4 exhibited 6% hemolysis at the same
maximum concentration.
μg/mL). 3, 5, and 6 induced up to 1%
μ
The largest shift in antimicrobial potency between the non-
fluorinated and fluorinated analogues was observed against the gastro-
intestinal pathogen C. jejuni and, to an even greater extent, the plant
pathogen E. amylovora Ea1501 and its streptomycin-resistant strain
Str4Ea, which are causal agents for fire blight disease in rosaceae plants.
Current AMPs studied for anti-campylobacter properties include tride-
captin [37], bacteriocin [38], and polymyxin B [39]. Typically,
designing peptide antibiotics to treat intestinal pathogens is challenging
because of proteolytic degradation in the intestine [40] and, as such,
fluid replacement therapy remains the mainstay treatment for campy-
lobacteriosis. As a Gram-negative bacterium, C. jejuni is susceptible to
2
.5. The fluorinated O-phenyl serine residue retains or has higher stability
to phenylalanine in short peptides
It was hypothesised that incorporation of fluorinated phenoxyserine
residues would impart enhanced proteolytic stability to these peptides
compared to the original sequence containing D-Phe. To test this, an MIC
assay was conducted in the presence of 10 % porcine serum to ascertain
the effect of in situ mammalian proteases on the peptide library’s
inhibitory activity of C. jejuni. Overall activity from the experiment in
triplicate revealed no change in MIC against C. jejuni, indicating that any
4

H.D. Glossop et al.
Journal of Fluorine Chemistry 241 (2021) 109685
Fig. 3. Circular dichroism spectra of the peptide library in water at a concentration of 150 g/mL.
μ
potential degradation caused by 10 % porcine serum was not detri-
mental for antimicrobial activity.
pentapeptide 7 (62 % degradation) over 6 h. Peptide 4 (3,4,5-tri-
fluorophenoxyserine analogue) on the other hand exhibited an increase
in proteolytic resistance (34 % degradation) over 6 h, suggesting that
further fluorination of the ring improves degradation resistance. Both
peptides 7 and 8 were found to rapidly degrade up to 90 % by 12 h. Due
to the large time intervals (6 h) employed, this investigation served to
only give a broad understanding of the relative stabilities of these pep-
tides in porcine serum. Octapeptides in this class of linear battacin
peptides were previously reported by us to resist proteolysis by rat
To further investigate proteolytic stability, HPLC methods were
employed. Peptides 1, 2, 4, 7, and 8 were selected as representative
peptides for comparison against control L-Phe pentapeptide 7 and the
unfluorinated O-phenylserine pentapeptide 8. Analytical HPLC moni-
toring of peptides incubated in 22.5 % porcine serum revealed that the
4
-fluorinated phenoxyserine-containing analogue 1 (71 % degradation)
in fact degraded to a similar degree (10 % difference) to L-Phe containing
5

H.D. Glossop et al.
Journal of Fluorine Chemistry 241 (2021) 109685
Fig. 4. The % hemolysis response of mouse red blood cells (as a function of
hemoglobin release) in the presence of peptides 3, 4, 5, and 6 across serial
dilutions from 250 to 0.49 g/mL.
μ
serum proteases up to 6 h [43]. However, potential differences between
rat and porcine serum proteolytic activity should be taken into consid-
eration when comparing these observations. Representative HPLC traces
(
Fig. S1) and a table of stability data (Table S1) are provided in the
supporting information.
Stability monitoring of peptide 1 was also briefly investigated by
recording 19F NMR spectra under proton decoupling, which is a
convenient technique for monitoring changes in fluorinated peptides in
complex biological mixtures by monitoring the clear 19F signal [44–46].
This method of studying peptide stability in porcine serum at 310 K
allowed for in situ monitoring of the same sample at discrete time points,
however, was not viable for the non-fluorinated pentapeptide, for
obvious reasons.
19F NMR monitoring of pentapeptide 1 (containing O-(4-fluo-
rophenyl) serine) revealed the evolution of a second fluorine signal over
the course of 12 h (Fig. 5) However, it is worth noting that there would
be a delay in observing degradation due to the time needed for the
degradation product to accumulate enough to be detected by NMR.
1
9
Fig. 5. Stacked F 1D NMR spectra of peptide 1 (100 g/mL) in the presence
μ
of 22.5 % porcine serum at 310 K. The main signal is centered at -123.70 ppm
and correlates to that expected for the single aromatic fluorine present. The
signal developed after 12 h is centered at -123.94.
Nevertheless,
a clear time-dependent decrease in peak intensity
antibiotics. While current research on the interactions of short antimi-
crobial peptides with E. amylovora and C. jejuni is limited, this work
reported herein may nonetheless contribute to the continued efforts
towards peptide design and eradication of pathogens, particularly those
threatening agricultural crops.
(
-123.70 ppm) becomes noticeable from at least the 6 h time-point
which generally agrees with the trend observed in stability data ac-
quired via HPLC monitoring which suggests that peptide 1 majorly de-
grades (~70 %) by 6 h. This demonstrates the promise of this technique
for monitoring fluorinated peptide serum stability in vitro. However,
quantitative monitoring of time-dependent degradation products by
fluorine NMR requires a higher starting concentration of the peptide.
While the representative fluorine modified peptides are not fully
resistant to porcine serum proteases, this may be of little consequence in
their potential application as treatment against plant pathogens such as
kiwifruit canker (Psa) or fire blight (E. amylovora), both of which are
highly sensitive to the fluorine modifications in the current library.
4
. Experimental
4
.1. General considerations
Diisopropylazodicarboxylate, L-serine methyl ester hydrochloride,
triphenylphosphine, 4-fluorophenol, 2,3-difluorophenol, 3,4,5-trifluor-
ophenol, 2,3,5,6-tetrafluorophenol and pentafluorophenol were pur-
chased from AK Scientific (California, USA). Fmoc-Dab(Boc)ꢀ OH and
Fmoc-D-Dab(Boc)ꢀ OH were purchased from ChemPep (Florida, USA).
Fmoc-rink amide linker was purchased from AK Scientific (California,
USA). HATU was purchased from Gyros Protein Technologies. N,N-
Diisopropylethylamine, piperidine, and trifluoroacetic acid were pur-
3
. Conclusions
The simple modification described on residue 4 of the battacin
pentapeptide confers enhanced broad-spectrum inhibitory activity
against a wide array of pathogenic species from clinically relevant
bacteria such as P. aeruginosa and C. jejuni to plant-specific pathogens
such as streptomycin resistant E. amylovora, responsible for fire blight in
rosaceae; and Psa, which is the cause of kiwifruit vine canker. Sub-
stitutions with fluorinated phenylalanine and fluorinated O-phenyl
serine residues considerably enhanced this antimicrobial potency, even
as L-amino acids replacing D-Phe on the original sequence. This implies
that the benefits of these structural modifications give credence to the
use of fluorinated Phe and mimics thereof in the design of lipopeptide
chased from Sigma-Aldrich. TentaGel S NH
Peptide International (Louisville, USA). BD Difco Mueller Hinton Broth
MHB) was purchased from Fort Richard Laboratories (Auckland, NZ).
2
resin was purchased from
(
Fmoc-4-fluoro-L-phenylalanine and Fmoc-pentafluoro-L-phenylalanine
were purchased from Chem-Impex International, Inc. (Illinois, USA).
Bacterial strains Pseudomonas aeruginosa ATCC27853, Escherichia coli
DH5
α
, Staphylococcus aureus, Bacillus cereus, and Candida albicans
SC5314 were obtained from the University of Auckland School of
6

H.D. Glossop et al.
Journal of Fluorine Chemistry 241 (2021) 109685
Biological Sciences microbial culture collection. Both Erwinia amylovora
strains Ea1501 and Str4Ea were obtained from the International
collection of Microorganisms from Plants (Landcare Research, New
Zealand). P. syringae pv. actinidiae Psa16027 was gifted by Dr Vanneste,
New Zealand Institute for Plant and Food Research. Campylobacter jejuni
Dioxane was removed under vacuum to yield a viscous slurry. Water
(60 mL) was then added, and the mixture acidified to pH 3 by dropwise
addition of aqueous 1 M KHSO
dichloromethane (4 × 30 mL) and the pooled organic extracts were
dried over Na SO , filtered, and then concentrated to yield a colourless
4
. The aqueous phase was extracted with
2
4
3
785 was obtained from the Institute of Environmental Science and
to pale yellow residue which was recrystallised from minimal hot
toluene and dichloromethane (~1:1) under gentle nitrogen flow to yield
the recrystallized title product, which was filtered (washing with mini-
mal ice-cold toluene) to yield Fmoc-(O-3,5-difluorophenyl)-Serꢀ OH
Research, New Zealand.
4
.2. Serine phenoxy derivative synthesis
(
0.122 g, 0.28 mmol, 35 % yield) as an off-white solid which was used
The following is a representative description of the synthesis for the
,5-difluoro variant, which also applies when using the other fluorinated
directly without further purification for solid-phase peptide synthesis.
3
phenol derivatives. The synthesis of the fluorinated O-phenyl-L-serine
residues was undertaken following modified methods from patent
US6358978B1 (2002) [47].
4.2.5. Fmoc-(O-4-fluorophenyl)-L-Ser-OH
◦
1
Pale gray powder; m.p. 146.7–150.3 C; H NMR (400 MHz, DMSO-
) δ 12.99 (s, 1H, COOH), 7.89 (d, J = 7.6, 2H, ArH), 7.74 (d, J = 7.5,
d
6
2
H, ArH), 7.41 (t, J = 7.5, 2H, ArH), 7.31 (td, J = 7.4, 3.4, 2H, ArH),
4
.2.1. N-Trityl-L-serine methyl ester
7.12 (t, J = 8.8, 2H, ArH), 6.96 (dd, J = 9.3, 4.4, 2H, ArH) 4.44ꢀ 4.39
13
Adapted from patent WO2016008946A1 (2016) [48]. To a suspen-
(m, 1H, CH), 4.31 (d, J = 7.5, 2H, CH
NMR (126 MHz, CDCl ) δ 174.2, 158.7, 156.8, 155.1, 154.1, 143.5,
141.3, 127.8, 127.1, 125.0, 120.0, 116.1, 115.9, 115.9, 115.8, 68.5,
2
), 4.25ꢀ 4.20 (m, 3H, H
α
, H
β
);
C
sion of L-serine methyl ester hydrochloride (7.5 g, 48.4 mmol) in
dichloromethane (100 mL), cooled on a water-ice bath, was added
triethylamine (13.8 mL, 96.0 mmol) dropwise followed by triphe-
nylmethyl chloride (13.44 g, 48.3 mmol) in dichloromethane (30 mL).
The mixture was stirred at ice bath temperature for 4 h, then allowed to
warm to ambient temperature overnight. The white precipitate that
formed was filtered off and the filtrate was evaporated to yield a pale
yellow solid, which was dissolved in ethyl acetate and extracted sub-
3
1
9
67.4, 53.8, 47.1; F NMR (376 MHz, CDCl
3
) δ -122.39 (s, 1 F); HRMS
+
calculated for [M + Na]+
+
ESI
C
24
H
19
F
2
NO
5
Na = 444.1123,
found = 444.1220.
4.2.6. Fmoc-(O-3,5-difluorophenyl)-L-Ser-OH
◦
1
Off-white powder; m.p. 110.0–112.4 C; H NMR (400 MHz, DMSO-
) δ 13.03 (s, 1H, COOH), 7.89 (d, J = 7.2, 2H, ArH), 7.73 (d, J = 7.6,
sequently with saturated NaHCO
3
, 10 % citric acid, water, and brine.
SO , filtered and concentrated in
d
6
The organic layer was dried over Na
2
4
2H, ArH), 7.41 (t, J = 7.4, 2H, ArH), 7.31 (td, J = 7.2, 3.0, 2H, ArH),
13
vacuo to yield an pale yellow solid that was triturated in hexanes to give
the title compound as a colourless solid (9.1 g, 52 %), or fractionally
crystallized from the slow evaporation of DCM/hexane to give the title
4.44ꢀ 4.40 (m, 1H, CH), 4.32–4.21 (m, 5H, H
α β
, H ); C NMR (126 MHz,
CDCl ) δ 173.5, 164.5, 162.7, 159.8, 156.1, 143.5, 141.3, 127.8, 127.1,
3
1
9
125.0, 120.1, 98.7, 98.5, 97.4, 68.3, 67.5, 53.6, 47.1; F NMR
+
compound as pale yellow crystals. TLC R
ethyl acetate 2:1.
f
: 0.4ꢀ 0.6; petroleum ether:
(376 MHz, CDCl
3
) δ -108.61 (s, 2 F); HRMS ESI calculated for
[M + Na]+
F
Na = 462.1129, found = 462.1117.
+
24
C H
19
2
NO
5
4
.2.2. N-Trityl-O-(3,5-difluorophenyl)-L-serine methyl ester
Representative procedure for the synthesis of a fluorinated O-phenyl
serine residue. A mixture of N-trityl-L-serine methyl ester (5.33 g,
4.76 mmol), 3,5-difluorophenol (2.11 g, 16.23 mmol), and triphenyl-
4.2.7. Fmoc-(O-3,4,5-trifluorophenyl)-L-Ser-OH
◦
1
Off-white powder; m.p. 114.1–119.5 C; H NMR (400 MHz, CDCl
3
)
δ 7.57 (dd, J = 7.6, 3.7, 2H, ArH), 7.59ꢀ 7.57 (m br, 2H, ArH), 7.39 (d,
J = 6.8, 2H, ArH), 7.29 (t, J = 7.6, 2H, ArH), 6.53 (dd, J = 5.9, 3.0, 2H,
ArH), 4.77 (d br, J = 7.9, 1 H), 4.46 (t br, J = 8.2, 2 H), 4.38 (d br,
1
phosphine (4.26 g, 16.25 mmol) in dry toluene (55 mL) was stirred
under nitrogen for 20 min. The solution was cooled on a water-ice bath
and diisopropylazodicarboxylate (3.35 mL, 16.22 mmol, 95 %) was
added dropwise to the reaction mixture over 20 min. The cooling bath
1
3
J = 7.6, 1 H), 4.23 (t, J = 6.7, 1 H), 4.19 (s br, 1 H); C NMR (126 MHz,
CDCl
3
) δ 172.3, 155.9, 153.2, 152.5, 150.5, 143.6, 143.5, 141.3, 127.8,
19
127.1, 124.9, 120.1, 99.6, 99.5, 68.6, 67.4, 53.4, 47.1; F NMR
◦
expired over 60 min and the mixture was then stirred at 80 C for 48 h
(471 MHz, CDCl
3
) δ -132.76 (d, J = 21.1, 2 F), -169.39 (t, J = 21.1, 1 F);
+
+
+
(
Note: for the more highly fluorinated phenols the reaction proceeded
rapidly at room temperature). The crude was purified by silica gel flash
column chromatography (toluene; R
HRMS ESI calculated for [M + Na]
found = 480.1026.
24
C H
18
F
3
NO
5
Na = 480.1035,
f
0.6ꢀ 0.7, as a dark yellow spot
visualised by Hanessian’s stain) and was recrystallized from equal parts
dichloromethane and n-hexanes to yield a colourless solid (3.77 g,
4.2.8. Fmoc-(O-2,3,5,6-tetrafluorophenyl)-L-Ser-OH
◦
1
Colorless powder; m.p. 163.2–165.8 C; H NMR (400 MHz, DMSO-
) δ 13.11 (s, 1H, COOH), 7.88 (d, J = 7.3, 3H, ArH), 7.70 (d, J = 7.4,
7
.97 mmol 53 % yield).
d
6
2
H, ArH), 7.64ꢀ 7.54 (m, 1H, ArH), 7.40 (t, J = 7.4, 2H, ArH), 7.30, (t,
4
.2.3. O-3,5-Difluorophenyl-L-serine
J = 7.4, 2H, ArH), 4.56 (d, J = 6.3, 1H, H
4.29 (d, J = 1.7, 1H, H ), 4.27 (s, 1H, H ), 4.21 (d, J = 6.9, 1H, H
NMR (126 MHz, DMSO-d ) δ 170.5, 156.0, 146.7, 146.6, 144.8, 144.6,
143.7, 141.5, 141.3, 140.7, 139.5, 139.4, 137.0, 127.6, 127.0, 125.2,
α
), 4.43 (d, J = 6.6, 2H, H
β
),
1
3
The isolated N-trityl-O-(3,5-difluorophenyl)-serine methyl ester was
β
β
α
);
C
suspended in aqueous 5ꢀ 6 N HCl and the suspension was refluxed for
h with rapid mixing. The resultant bright yellow solution was cooled
to room temperature and extracted with diethyl ether (50 mL) and DCM
2 × 50 mL). The aqueous phase was evaporated, re-dissolved in MeCN:
O (1:4) and lyophilised to yield the fully deprotected amino acid
6
3
1
9
125.2, 120.0, 100.7, 100.5, 100.3, 73.4, 65.9, 54.2, 46.5; F NMR
(376 MHz, DMSO-d
(
6
) δ -141.43 (dd, J = 14.5, 9.7, 2 F), -157.65 (dd,
+
+
H
2
J = 13.6, 9.1, 2 F); HRMS ESI
calculated for [M + Na]
+
hydrochloride salt (1.72 g, 6.79 mmol, 85 % yield) as an off-white
powder which was used without further purification.
C
24
H
17
F
4
NO
5
Na = 498.0941, found = 498.0935.
4
.2.9. Fmoc-(O-2,3,4,5,6-pentafluorophenyl)-L-Ser-OH
◦
1
4
.2.4. Fmoc-(O-3,5-difluorophenyl)-L-Ser-OH
Colorless powder; m.p. 152.5–153.4 C; H NMR (500 MHz, MeOD)
The amino acid hydrochloride salt (0.201 g, 0.79 mmol) was
δ 7.79 (d, J = 7.5, 2H, ArH), 7.68 (t, J = 6.9, 2H, ArH), 7.39 (td, J = 7.4,
partially dissolved in dioxane (16 mL) and water (6 mL). NaHCO
0.503 g, 6 mmol) was added and the mixture was cooled on a water-ice
3
3.2, 2H, ArH), 7.31 (t, J = 7.3, 2H, ArH), 4.62–4.50 (m, 3 H), 4.37 (d,
1
3
(
J = 7.1, 2 H), 4.25 (t, J = 7.1, 1 H); C NMR (101 MHz, MeOD) δ 179.9,
bath. Fmoc-OSu (0.407 g, 1.08 mmol) was added and the mixture was
stirred at water-ice bath temperature for 30 min, then at RT overnight.
177.6, 165.5, 153.2, 151.2, 150.2, 137.1, 136.5, 135.7, 129.6, 83.3,
75.4, 63.7, 61.9, 56.0; 19F NMR (471 MHz, CDCl
) δ -156.43 (d,
3
7

H.D. Glossop et al.
Journal of Fluorine Chemistry 241 (2021) 109685
J = 19.6, 2 F), -161.7 (t, J = 21.8, 1 F), -162.49 (t br, J = 39.8, 2 F);
blood was centrifuged to remove the buffy coat. The erythrocytes were
then washed thrice in Tris buffer (10 mM Tris, 150 mM NaCl, pH 7.4)
and then diluted to a 2% v/v solution in Tris buffer. Peptide stock so-
+
+
+
HRMS ESI calculated for [M + Na]
C
24
H
16
F
5
NO
5
Na = 516.0846
Found = 516.0848
lutions (500
μ
g/mL) in Tris were serially diluted across 96-well plates to
4
.2.10. Fmoc-(O-phenyl)-L-Ser-OH
a final volume of 100
μ
L, to which the resuspended blood (100 L) was
μ
◦
1
Colorless powder; m.p. 145.4–148.6 C; H NMR (500 MHz, CDCl
.75 (dd, J = 7.5, 4.1, 2H, ArH), 7.59 (t br, J = 11.9, 2H, ArH), 7.39 (dt,
3
) δ
then added in each well. Buffer alone and 0.1 % v/v Triton X-100 were
used as negative and positive hemolytic controls, respectively. Samples
7
◦
J = 7.4, 4.6, 2H, ArH), 7.32ꢀ 7.27 (m, 4H, ArH), 5.76 (d, J = 8.5, 1 H),
and blood were incubated together at 37 C for 1 h and then the plates
1
3
4
.79 (d br, J = 8.5, 1 H), 4.49ꢀ 4.39 (m, 3 H), 4.28ꢀ 4.23 (m, 2 H);
C
were centrifuged. Absorbance at 540 nm was measured for all super-
natants and the buffer absorbance subtracted. Hemolysis was then
determined as a percentage of the positive hemolytic absorbance (Triton
X-100).
NMR (126 MHz, CDCl ) δ 173.2, 157.9, 156.1, 143.7, 143.6, 141.3,
3
1
29.6, 127.8, 127.1, 125.1, 121.8, 120.0, 114.6, 67.7, 67.4, 53.7, 47.1;
HRMS ESI calculated for [M + Na]+
+
+
24 5
C H21NO Na = 426.1317,
found = 426.1309.
%
hem olysis = [(ASupernatant – ABuffer)/(A100 % - ABuffer)]*100
4
.3. Peptide synthesis
Peptides were synthesised via typical Fmoc/tBu solid-phase peptide
synthesis methods as reported by us previously [27]. Briefly, peptides
were assembled stepwise from C to N-terminus on Tentagel S NH resin
via a rink amide linker. Subsequent Fmoc deprotection was effected with
0 % piperidine, and incoming amino acid activation was effected by
DIPEA and HATU. Final cleavage and deprotection of peptide func-
tionality was performed with a cocktail of TFA:TIS:H O (95:2.5:2.5) and
4
.5.3. Serum stability MIC assay
In vitro serum peptide stability was investigated via MIC assays
2
conducted on the peptide library in the presence of 10 % porcine serum.
C. jejuni was selected for these assays due to the peptide library’s potent
inhibition of the bacterial species and its clinical relevance. Peptide and
bacteria in MHB medium and in the presence of 10 % porcine serum
were incubated overnight (16 h) as per MIC assay conditions described
and assessed identically to typical MIC assay protocol.
2
2
¨
the subsequent crude was purified by HPLC with an AKTA purifier UPC
0 system and a Phenomenex Luna 5
m C18 100 Å 250 mm × 50 mm
column. Purified peptides were characterized by NMR (400 MHz and
00 MHz Bruker NMR spectrometers) and electrospray ionization–mass
1
μ
4
.5.4. Serum stability HPLC assay [43,52]
5
Whole porcine serum (Life Technologies, New Zealand) was mixed
spectrometry (ESI–MS) on a Bruker micrOTOF-QII high-resolution mass
spectrometer as further detailed by us previously [49]. Mass spectra data
and HPLC chromatograms are available in the supplementary
information.
with PBS (10 mM phosphate, 150 mM NaCl) to a final volume of 900
and 25 % v/v serum concentration. To this was then added peptide stock
solution (100 L, 1 mg/mL) to achieve a final concentration of 100 g/
mL peptide and 22.5 % serum. The mixture was then incubated at 37 C,
with agitation, and aliquots of 100 L were collected at set time points. A
μ
L
μ
μ
◦
μ
4
.4. Circular dichroism spectroscopy
t = 0 aliquot was taken immediately after the peptide was mixed with
◦
serum prior to incubation. Aliquots were stored at -80 C post collection.
Circular dichroism (CD) measurements were carried out on 150 g/
μ
Prior to HPLC analysis, 15 % trichloroacetic acid solution (40
μ
L) was
mL peptide solution in pure water in a 2 mm quartz cuvette (Hellma,
Asia Pacific) with a Chirascan V100 CD spectrometer (Applied Photo-
◦
added to each aliquot, followed by at least 10 min incubation at 4 C to
precipitate proteins. Samples were then centrifuged (12,000 RPM,
◦
physics, United Kingdom). Experiments were carried out at 22 C using
5
min) and the supernatant analysed by HPLC at a gradient of 10–65 %
2
O/acetonitrile over 25 min, at a flow rate of 1 mL/min with a Phe-
0
.25 s per data point and a step size and bandwidth of 1 nm. Spectra
H
were recorded as the average of three scans and background and blank
subtracted using Chirascan Q100 software.
nomenex Luna 5
μm C18 100 Å 250 mm × 4.6 mm analytical column.
Unicorn 5.0 software (GE Healthcare) was used to integrate the 214 nm
chromatograms to compare relative peptide peak areas to that of the
serum peak at 11ꢀ 12 min.
4
4
.5. Biological assays
4
.5.5. Serum stability 19F NMR assay
The sample for 19F NMR serum stability assays was prepared iden-
.5.1. Minimal inhibitory concentration (MIC) Assay
Unless otherwise stated, all minimal inhibitory concentration (MIC)
assays were performed via standard broth microdilution procedure [50].
tically to that described for the HPLC stability assay but instead to a total
volume of 600 L, which was then conditioned into an NMR tube
immediately after addition of peptide to serum) and incubated within a
6
Briefly, overnight microbial growth culture was adjusted to 10 CFU/mL
μ
in Mueller Hinton broth according to MacFarlane standard. The adjusted
microbial suspensions were mixed in 96-well plate wells containing the
(
1
9
4
00 MHz NMR spectrometer (Bruker, Germany) at 310 K. 1D F ex-
serially diluted peptide solution (50
μ
L) to reach a final volume of
L per well and concentrations of peptides ranging from 100 to
g/mL, including inhibition control (streptomycin for S. aureus,
periments were carried out at 30 min intervals, each totalling 256 scans,
recording between -70 to -130 ppm over the course of 12 h. Spectra
were processed with TopSpin software (Bruker, Germany).
1
0
00
μ
.19
μ
B. cereus, and E. amylovora; polymyxin B for P. aeruginosa, E. coli,
C. jejuni, and Psa; amphotericin for C. albicans). Each experiment was
repeated in triplicate. Inhibition was assessed via ocular density at
Declaration of Competing Interest
The authors report no declarations of interest.
Acknowledgement
6
25 nm (OD625). MIC assays on C. albicans were performed following
NCCL M27-A2 methodology, with minor modifications [36,51]. Incu-
bation of C. jejuni was performed in an anaerobic environment of 85 %
2 2 2
N , 10 % CO , and 5% O .
HDG is grateful for the support received through the University of
Auckland doctoral scholarship.
4
.5.2. Hemolytic assay
Hemolysis was assessed by a spectrophotometric method comparing
the release of hemoglobin from erythrocytes in the presence of peptides
to that of a known hemolytic substance, in this case the surfactant Triton
X-100, via a previously described protocol [27]. Briefly, fresh mouse
8

H.D. Glossop et al.
Journal of Fluorine Chemistry 241 (2021) 109685
Appendix A. Supplementary data
structure activity relationship studies of battacin peptides, J. Med. Chem. 58
(
2015) 625–639.
[
[
28] H. Tsubery, I. Ofek, S. Cohen, M. Eisenstein, M. Fridkin, Modulation of the
hydrophobic domain of polymyxin B nonapeptide: effect on outer-membrane
permeabilization and lipopolysaccharide neutralization, Mol. Pharmacol. 62
Supplementary material related to this article can be found, in the
online version, at doi:https://doi.org/10.1016/j.jfluchem.2020.10
(
2002) 1036–1042.
29] R. Castelli, M. Tognolini, F. Vacondio, M. Incerti, D. Pala, D. Callegari, et al., Delta
5)-cholenoyl-amino acids as selective and orally available antagonists of the Eph-
9
685.
(
References
ephrin system, Eur. J. Med. Chem. 103 (2015) 312–324.
[
[
30] B.C. Buer, E.N. Marsh, Fluorine: a new element in protein design, Protein Sci. 21
(
2012) 453–462.
[
[
[
[
[
[
[
1] C.L. Ventola, The antibiotic resistance crisis: part 1: causes and threats, Pharm.
Ther. 40 (2015) 277–283.
31] M.D. Díaz, M. Palomino-Sch a¨ tzlein, F. Corzana, C. Andreu, R.J. Carbajo, M. del
Olmo, et al., Antimicrobial peptides and their superior fluorinated analogues:
structure–activity relationships as revealed by NMR spectroscopy and MD
calculations, ChemBioChem 11 (2010) 2424–2432.
2] G.W. Sundin, N. Wang, Antibiotic resistance in plant-pathogenic bacteria, Annu.
Rev. Phytopathol. 56 (2018) 161–180.
3] S. Thanner, D. Drissner, F. Walsh, Antimicrobial resistance in agriculture, mBio 7
[
[
[
32] V.V. Aksenov, V.M. Vlasov, G.G. Yakobson, Interaction of pentafluoropyridine with
(
2016) e02227–15.
4
-nitrophenol and pentafluorophenol in the presence of potassium fluoride and 18-
4] K. Matsuzaki, Control of cell selectivity of antimicrobial peptides, Biochim.
Biophys. Acta 1788 (2009) 1687–1692.
crown-6-ether, J. Fluor. Chem. 20 (1982) 439–458.
33] M.R. Lockett, M.F. Phillips, J.L. Jarecki, D. Peelen, L.M. Smith, A tetrafluorophenyl
activated ester self-assembled monolayer for the immobilization of amine-modified
oligonucleotides, Langmuir 24 (2008) 69–75.
5] K. Matsuzaki, Magainins as paradigm for the mode of action of pore forming
polypeptides, Biochim. Biophys. Acta 1376 (1998) 391–400.
6] H.W. Huang, Free energies of molecular bound states in lipid bilayers: lethal
concentrations of antimicrobial peptides, Biophys. J. 96 (2009) 3263–3272.
7] R.F. Epand, W.L. Maloy, A. Ramamoorthy, R.F.M. Epand, Probing the "charge
cluster mechanism" in amphipathic helical cationic antimicrobial peptides,
Biochemistry 49 (2010) 4076–4084.
34] J.M. Lacombe, F. Andriamanampisoa, A.A. Pavia, Solid-phase synthesis of peptides
containing phosphoserine using phosphate tert.-butyl protecting group, Int. J. Pept.
Protein Res. 36 (1990) 275–280.
[
[
35] Y. Yang, β-Elimination side reactions, Chapter 2, in: Y. Yang (Ed.), Side Reactions
in Peptide Synthesis, Academic Press, Oxford, 2016, pp. 33–42.
[
8] R.F. Epand, L. Maloy, A. Ramamoorthy, R.F.M. Epand, Amphipathic helical
cationic antimicrobial peptides promote rapid formation of crystalline states in the
presence of phosphatidylglycerol: lipid clustering in anionic membranes, Biophys.
J. 98 (2010) 2564–2573.
36] G.H. De Zoysa, H.D. Glossop, V. Sarojini, Unexplored antifungal activity of linear
battacin lipopeptides against planktonic and mature biofilms of C. albicans, Eur. J.
Med. Chem. 146 (2018) 344–353.
[
[
[
37] C.T. Lohans, M.J. van Belkum, J. Li, J.C. Vederas, Characterization of bacterial
antimicrobial peptides active against Campylobacter jejuni, Can. J. Chem. 93
[
9] R.E. Hancock, R. Lehrer, Cationic peptides: a new source of antibiotics, Trends
Biotechnol. 16 (1998) 82–88.
(
2014) 381–388.
[
[
[
[
[
[
[
10] P. Bulet, R. Stocklin, L. Menin, Anti-microbial peptides: from invertebrates to
vertebrates, Immunol. Rev. 198 (2004) 169–184.
38] K.V. Hoang, N.J. Stern, A.M. Saxton, F. Xu, X. Zeng, J. Lin, Prevalence,
development, and molecular mechanisms of bacteriocin resistance in
Campylobacter, Appl. Environ. Microbiol. 77 (2011) 2309–2316.
11] T. Velkov, P.E. Thompson, R.L. Nation, J. Li, Structure—activity relationships of
polymyxin antibiotics, J. Med. Chem. 53 (2010) 1898–1916.
39] T. Keo, J. Collins, P. Kunwar, M.J. Blaser, N.M. Iovine, Campylobacter capsule and
lipooligosaccharide confer resistance to serum and cationic antimicrobials,
Virulence 2 (2011) 30–40.
12] H. Tsubery, I. Ofek, S. Cohen, M. Fridkin, Structure activity relationship study of
polymyxin B nonapeptide, Adv. Exp. Med. Biol. 479 (2000) 219–222.
13] Z. Feng, B. Xu, Inspiration from the mirror: D-amino acid containing peptides in
biomedical approaches, Biomol. Concepts 7 (2016) 179–187.
[
[
40] A. Sharma, D. Singla, M. Rashid, G.P. Raghava, Designing of peptides with desired
half-life in intestine-like environment, BMC Bioinform. 15 (2014) 282.
41] K. Hamamoto, Y. Kida, Y. Zhang, T. Shimizu, K. Kuwano, Antimicrobial activity
and stability to proteolysis of small linear cationic peptides with D-amino acid
substitutions, Microbiol. Immunol. 46 (2002) 741–749.
14] A. Niemz, D.A. Tirrell, Self-association and membrane-binding behavior of
melittins containing trifluoroleucine, J. Am. Chem. Soc. 123 (2001) 7407–7413.
15] E.N.G. Marsh, B.C. Buer, A. Ramamoorthy, Fluorine—a new element in the design
of membrane-active peptides, Mol. Biosyst. 5 (2009) 1143–1147.
[
[
[
42] A. Gautam, K. Chaudhary, S. Singh, A. Joshi, P. Anand, A. Tuknait, et al.,
Hemolytik: a database of experimentally determined hemolytic and non-hemolytic
peptides, Nucleic Acids Res. 42 (2013) D444–D449.
16] L.M. Gottler, H.-Y. Lee, C.E. Shelburne, A. Ramamoorthy, E.N.G. Marsh, Using
fluorous amino acids to modulate the biological activity of an antimicrobial
peptide, ChemBioChem 9 (2008) 370–373.
43] H.D. Glossop, E. Pearl, G.H. De Zoysa, V. Sarojini, Linear analogues of the
lipopeptide battacin with potent in vitro activity against S. aureus, Adv. Protein
Chem. Struct. Biol. 112 (2018) 385–394.
[
[
17] H. Meng, K. Kumar, Antimicrobial activity and protease stability of peptides
containing fluorinated amino acids, J. Am. Chem. Soc. 129 (2007) 15615–15622.
18] H.P. Chiu, Y. Suzuki, D. Gullickson, R. Ahmad, B. Kokona, R. Fairman, et al., Helix
propensity of highly fluorinated amino acids, J. Am. Chem. Soc. 128 (2006)
44] M.V. Christensen, K.T. Kongstad, T.E. Sondergaard, D. Staerk, H.M. Nielsen,
H. Franzyk, et al., (19)F-substituted amino acids as an alternative to fluorophore
labels: monitoring of degradation and cellular uptake of analogues of penetratin by
1
5556–15557.
[
[
19] M. Salwiczek, E.K. Nyakatura, U.I.M. Gerling, S. Ye, B. Koksch, Fluorinated amino
acids: compatibility with native protein structures and effects on protein–protein
interactions, Chem. Soc. Rev. 41 (2012) 2135–2171.
(
19)F NMR, J. Biomol. NMR 73 (2019) 167–182.
[
[
45] E.N.G. Marsh, Y. Suzuki, Using 19F NMR to probe biological interactions of
proteins and peptides, ACS Chem. Biol. 9 (2014) 1242–1250.
20] L.M. Gottler, R. de la Salud Bea, C.E. Shelburne, A. Ramamoorthy, E.N.G. Marsh,
Using fluorous amino acids to probe the effects of changing hydrophobicity on the
physical and biological properties of the β-Hairpin antimicrobial peptide protegrin-
46] D. Maisch, P. Wadhwani, S. Afonin, C. B o¨ ttcher, B. Koksch, A.S. Ulrich, Chemical
labeling strategy with (R)- and (S)-trifluoromethylalanine for solid state 19F NMR
analysis of peptaibols in membranes, J. Am. Chem. Soc. 131 (2009) 15596–15597.
47] O. Ritzeler, H.U. Stilz, B. Neises, J.W.J. Bock, A. Walser, Ga Flynn, et al.
Substituted benzimidazoles, US6358978 (B1). 1999.
1
, Biochemistry 47 (2008) 9243–9250.
[
[
[
[
[
[
21] A. Beliaev, J. Wahnon, D. Russo, Process research for multikilogram production of
etamicastat: a novel dopamine β-hydroxylase inhibitor, Org. Process Res. Dev. 16
48] K. Mills, P. Kelly, J.-G. Tiraby, T. Lioux, E. Perouzel Novel compounds,
WO2016008946A1. 2014.
(
2012) 704–709.
22] K. Urbahns, M. H ¨a rter, A. Vaupel, M. Albers, D. Schmidt, U. Brüggemeier, et al.,
Biphenyls as potent vitronectin receptor antagonists. Part 2: biphenylalanine ureas,
Bioorg. Med. Chem. Lett. 13 (2003) 1071–1074.
49] H.D. Glossop, G.H. De Zoysa, Y. Hemar, P. Cardoso, K. Wang, J. Lu, et al., Battacin-
inspired ultrashort peptides: nanostructure analysis and antimicrobial activity,
Biomacromolecules 20 (2019) 2515–2529.
ˇ
´
ˇ
´
´
23] O. Krenk, J. Kratochvíl, M. Spulak, V. Buchta, J. Kunes, L. Novakova, et al.,
Methodology for synthesis of enantiopure 3,5-disubstituted pyrrol-2-ones, Eur. J.
Org. Chem. 2015 (2015) 5414–5423.
[
[
50] I. Wiegand, K. Hilpert, R.E. Hancock, Agar and broth dilution methods to
determine the minimal inhibitory concentration (MIC) of antimicrobial substances,
Nat. Protoc. 3 (2008) 163–175.
[
[
24] J.C. Biffinger, H.W. Kim, S.G. DiMagno, The polar hydrophobicity of fluorinated
compounds, ChemBioChem 5 (2004) 622–627.
51] E. Chryssanthou, M. Cuenca-Estrella, Comparison of the antifungal susceptibility
testing subcommittee of the European committee on antibiotic susceptibility
testing proposed standard and the E-test with the NCCLS broth microdilution
method for voriconazole and caspofungin susceptibility testing of yeast species,
J. Clin. Microbiol. 40 (2002) 3841–3844.
25] R. Wilcken, M.O. Zimmermann, A. Lange, A.C. Joerger, F.M. Boeckler, Principles
and applications of halogen bonding in medicinal chemistry and chemical biology,
J. Med. Chem. 56 (2013) 1363–1388.
[
[
26] X.-L. Qiu, F.-L. Qing, Recent advances in the synthesis of fluorinated amino acids,
Eur. J. Org. Chem. 2011 (2011) 3261–3278.
[
52] L.T. Nguyen, J.K. Chau, N.A. Perry, L. de Boer, S.A. Zaat, H.J. Vogel, Serum
stabilities of short tryptophan- and arginine-rich antimicrobial peptide analogs,
PLoS One 5 (2010).
27] G.H. De Zoysa, A.J. Cameron, V.V. Hegde, S. Raghothama, V. Sarojini,
Antimicrobial peptides with potential for biofilm eradication: synthesis and
9