Synthetic antimicrobial peptidomimetics with therapeutic potential

Article abstract of DOI:10.1021/jm701600a

A series of synthetic antimicrobial peptidomimetics (SAMPs) have been prepared and found to be highly active against several Gram-negative and Gram-positive bacterial strains. These derivatives comprise the minimal structural requirements for cationic antimicrobial peptides and showed high selectivity for Gram-negative and/or Gram-positive bacteria compared to human red blood cells. We have found that SAMPs share many of the attractive properties of cationic antimicrobial peptides inasmuch that a representative SAMP was found to insert into the bilayers of large unilamellar vesicles, permeabilized both the outer and cytoplasmic membrane of Escherichia coli ML-35p, and displayed an extremely rapid bacterial killing for Staphylococcus aureus. However, while antimicrobial peptides are prone to proteolytic degradation, high in vitro stability in human blood plasma was shown for SAMPs. A combination of high antibacterial activity against methicillin-resistant staphylococci and low toxicity against human erythrocytes makes these molecules promising candidates for novel antibacterial therapeutics.

Full text of DOI:10.1021/jm701600a

4306
J. Med. Chem. 2008, 51, 4306–4314
Synthetic Antimicrobial Peptidomimetics with Therapeutic Potential
Bengt Erik Haug,*,†,‡ Wenche Stensen,†,§ Manar Kalaaji,| Øystein Rekdal,§,| and John S. Svendsen†,§
Department of Chemistry and Department of Biochemistry, UniVersity of Tromsø, N-9037 Tromsø, Norway, Department of Chemistry,
UniVersity of Bergen, Alle`gaten 41, N-5007 Bergen, Norway, and Lytix Biopharma AS, P.O. Box 6447,
Tromsø Science Park, N-9294 Tromsø, Norway
ReceiVed December 20, 2007
A series of synthetic antimicrobial peptidomimetics (SAMPs) have been prepared and found to be highly
active against several Gram-negative and Gram-positive bacterial strains. These derivatives comprise the
minimal structural requirements for cationic antimicrobial peptides and showed high selectivity for Gram-
negative and/or Gram-positive bacteria compared to human red blood cells. We have found that SAMPs
share many of the attractive properties of cationic antimicrobial peptides inasmuch that a representative
SAMP was found to insert into the bilayers of large unilamellar vesicles, permeabilized both the outer and
cytoplasmic membrane of Escherichia coli ML-35p, and displayed an extremely rapid bacterial killing for
Staphylococcus aureus. However, while antimicrobial peptides are prone to proteolytic degradation, high in
vitro stability in human blood plasma was shown for SAMPs. A combination of high antibacterial activity
against methicillin-resistant staphylococci and low toxicity against human erythrocytes makes these molecules
promising candidates for novel antibacterial therapeutics.
Introduction
(for reviews, see refs 4–6). At present, more than 500 riboso-
mally synthesized antimicrobial peptides (sometimes referred
to as RAMPs) are listed in the antimicrobial peptide database
(APD),7 and although they are structurally very different, some
common features can be identified. In general, CAPs are fairly
large molecules, ranging from 12 to 50 residues5 (average length
in APD is 28 residues),7 and they carry a net positive charge
(average charge in APD is +4.5)7 and contain about 50%
hydrophobic residues. CAPs are proposed to act by associating
with the negatively charged components of the cytoplasmic
membrane of bacteria, thereby increasing the permeability to
ions and solutes, events that may lead to cell death. Several
mechanisms have been proposed to describe the peptide-lipid
interactions,8–12 but it is still a matter of controversy whether
the final killing event of CAPs is due to a damaged cytoplasmic
membrane or a combination of membrane permeabilization and
peptide interaction with intracellular targets.13,14 In addition, a
number of antimicrobial peptides have also been found to induce
an immune response in vivo.15 Seemingly, CAPs utilize an array
of means to kill bacteria, and in addition, CAPs show an
unusually broad antimicrobial activity spectrum (for a recent
review, see ref 16). Thus, CAPs constitute an attractive class
of compounds for further development of novel therapeutics
against infections caused by bacterial strains resistant to the
antibiotics of today.4 In spite of all their attractive features, few
(if any) CAPs have made it through clinical trials. This may at
least partly be connected to the poor drugability of large
peptides.
The number of infections caused by bacterial strains resistant
to conventional antibiotics has been rising steadily, and infec-
tions caused by methicillin resistant Staphylococcus aureus
(MRSA,a also denoted multidrug-resistant S. aureus) previously
only associated with hospitalized patients have started to spread
into community-acquired infections.1 A recent report on the in
vivo evolution of multidrug resistance in S. aureus revealed the
passage of a mere 2.5 months from the initial treatment of an
MRSA infection with vancomycin to the identification of isolates
with a considerably decreased susceptibility.2 The resistance
resulted from 18 sequential point mutations, also leading to the
development of resistance toward daptomycin, which had never
been administered to the patient.2
As a consequence of resistance development, the search for
novel antibiotics has attracted an enormous effort in the
pharmaceutical industry and by research groups in academia.3
In recent years, several cationic antimicrobial peptides (CAPs)
have been investigated clinically as potential future antibiotics.
Antimicrobial peptides (AMPs) are widespread in nature and
have been isolated from essentially all species investigated,
ranging from insects to humans. It has been well established
that these compounds play an important role in innate immunity
* To whom correspondence should be addressed at University of Bergen.
Phone:
+ 47 55 58 34 68. Fax + 47 55 58 94 90. E-mail:
bengt-erik.haug@farm.uib.no.
†
Department of Chemistry, University of Tromsø.
Department of Chemistry, University of Bergen.
Lytix Biopharma AS.
‡
§
|
Department of Biochemistry, University of Tromsø.
For a number of years our research group has focused on the
development of extremely short CAPs (for a recent review, see
ref 17). In our endeavors we have elucidated a pharmacophore
for short cationic antimicrobial peptides that is surprisingly
small.18 For antistaphylococcal activity the presence of two
cationic units and two lipohilic and bulky units approximately
the size of a phenyl group is required. We have previously
shown that replacing tryptophan residues with large noncoded
aromatic amino acids can result in a dramatic increase in
antibacterial activity,19–23 and in this respect ꢀ-(2,5,7-tri-tert-
a Abbreviations: ATCC, American Type Culture Collection; CAP,
cationic antimicrobial peptide; CCUG, culture collection of the University
of Gothenburg; CENTA, (6R,7R)-3-[[(3-carboxy-4-nitrophenyl)thio]methyl]-
8-oxo-7-[(2-thienyl-acetyl)amino]-5-thia-1-azabicyclo[4.2.0]oct-2-ene-2-car-
boxylic acid; GISA, glycopeptide intermediate resistant S. aureus; LUV;
large unilamellar vesicles; MIC, minimal inhibitory concentration; MLV,
multilamellar vesicles; MRSA, methicillin resistant S. aureus; MRSE,
methicillin resistant S. epidermidis; ONPG, 2-nitrophenyl-ꢀ-D-galactopy-
ranoside; POPC, 1-palmitoyl-2-oleoyl-sn-glycero-3-phosphocolin; POPG,
1-palmitoyl-2-oleoyl-sn-glycero-3-phosphoglycerol; RP-HPLC, reversed
phase high performance liquid chromatography; SAMP, synthetic antimi-
crobial peptidomimetic; Tbt, ꢀ-(2,5,7-tri-tert-butylindol-3-yl)alanine.
10.1021/jm701600a CCC: $40.75
2008 American Chemical Society
Published on Web 06/21/2008

Peptidomimetics with Therapeutic Potential
Journal of Medicinal Chemistry, 2008, Vol. 51, No. 14 4307
Scheme 1. Alkylation of Trp To Produce Tbt25
butylindol-3-yl)alanine24,25 (Tbt; see Scheme 1) has proven to
be especially useful.
In a drug development perspective, it is vital that most of
the attractive properties of CAPs are still contained in our series
of extremely short synthetic CAPs. Even though these molecules
contain amide bonds, they can hardly be regarded as antimi-
crobial peptides in the classical sense. In consequence of this
we propose to use the term synthetic antimicrobial peptidomi-
metics (SAMPs) to describe this class of compounds.
Thus, the focus of the present work has been to establish
whether SAMPs share the same properties as CAPs. The
interaction with large unilamellar vesicles (LUVs) of different
lipid composition has been used as a model system for an initial
investigation of the bactericidal mechanism. The time-course
of bacterial killing and the effects of a representative SAMP
on the integrity of the outer and cytoplasmic membranes of E.
coli ML-35p have also been investigated. To further assess their
potential as novel antibiotics, in vitro toxicity has been assessed
by measuring the lytic activity against human red blood cells,
antibacterial activity against a range of Gram-positive and Gram-
negative bacteria has been measured, and proteolytic stability
in human blood plasma, which is an inherent problem for CAPs
and naturally occurring peptides, has been investigated for
selected SAMPs.
Figure 2. Fluorescence spectra of RTbtR-OMe (A) and RWR-OMe
(B) recorded in buffer (b), POPC (2), POPC/POPG (3:1) ([), and
POPG (9).
Results
Fluoresence Spectroscopy. In order to shed some light on
the bactericidal mechanism of SAMPs, fluorescence emission
spectroscopy was used to study the interactions of RTbtR-OMe
7 (Figure 1) with large unilamellar vesicles of different lipid
composition (Figure 2). The liposomes were prepared using only
negatively charged phospholipids with a phospoglycerol head-
group (POPG), only zwitterionic phosphlipids with a phospho-
colin headgroup (POPC), or a mixture of the two phospholipids.
An inactive tripeptide, RWR-OMe was also included in order
to investigate whether active and inactive tripeptides show
different peptide-lipid interactions.
Figure 3. Leakage of calcein from POPG vesicles induced by RTbtR-
OMe ([) and RWR-OMe (2). Peptide concentration is in µg/mL.
Concentration of liposomes is 0.3 mg/mL.
The excitation maxima for RWR-OMe (inactive peptide) and
RTbtR-OMe 7 were found to be 291 and 282 nm, respectively,
and both compounds showed fluorescence emission in buffer
with maxima at 355 and 353 nm, respectively. The fluorescence
emission maximum of RTbtR-OMe 7 was blue-shifted in the
presence of POPC, POPG, and POPC/POPG (3:1) vesicles
(Figure 2A), commonly associated with the embedding of
tryptophan residues into the hydrophobic environment of the
phospholipid bilayers.26 RWR-OMe demonstrated a 14 nm blue
shift in fluorescence emission when the LUVs contained only
negatively charged phospholipid head-groups (Figure 2B),
whereas a 28 nm blue shift was observed for RTbtR-OMe 7 in
POPG relative to buffer. RWR-OMe displayed a 7 nm blue shift
in POPC/POPG (3:1), but no change in fluorescence emission
maximum was detected in the presence of liposomes composed
solely of POPC. RTbtR-OMe 7 displayed 26 and 12 nm blue
shifts in the fluorescence emission maximum in these liposome
systems, respectively.
Calcein Release. In order to measure membrane permeabi-
lization caused by the peptides, we examined the induced
leakage of calcein by RTbtR-OMe 7 from vesicles of various
lipid compositions (Figure 3). A moderate, yet concentration
dependent leakage of calcein from POPG vesicles induced by
RTbtR-OMe 7 was observed, whereas RWR-OMe did not
display any significant leakage (Figure 3). No leakage of calcein
Figure 1. Structure of tripeptide methyl ester RTbtR-OMe (7).

4308 Journal of Medicinal Chemistry, 2008, Vol. 51, No. 14
Haug et al.
in Table 1. The antibacterial activity of compounds 1, 2, and 4
against E. coli and S. aureus has been reported previously,22
and even though they were found to be overall less active than
the tripeptides, the most active dipeptides showed MIC values
as low as 2.5-5 µg/mL against the various staphylococci strains.
The dipeptides showed higher antibacterial activity against the
Gram-positive strains than against the Gram-negative bacteria.
Moreover, the dipeptide benzylamide derivative TbtR-NHBn 4
was found to be the overall most active dipeptide, and notably
this dipeptide derivative displayed MIC values of 10 and 7.5
µg/mL against E. coli and P. aeruginosa, respectively. The
tripeptides were also found to be somewhat more active against
the methicillin resistant strains than against the normal S. aureus.
Furthermore, the MRSE strain was found to be the most
susceptible bacterium to all peptides, of which both SAMPs
RTbtR-OMe 7 and RTbtR-NHBn 9 displayed MIC values of
1.0 µg/mL. The tripeptide derivatives were also active against
E. coli and P. aeruginosa, although the MIC values were
considerably higher against these Gram-negative bacteria than
those observed against the Gram-positive strains. As was found
for the dipeptides, the benzylamide derivative had the highest
antibacterial activity of the tripeptides, and RTbtR-NHBn 9
showed MIC values between 7.5 and 1.0 µg/mL against the
different bacterial strains.
Hemolytic Activity. The concentrations of the tested di- and
tripeptides that induced 50% hemolysis of human erythrocytes
(EC50) were found to range between 60 to above 1000 µg/mL
and are compiled in Table 1. The dipeptides displayed the lowest
EC50 values, and among the tripeptide derivatives, SAMP 9 was
the most hemolytic.
Plasma Stability. Incubation of SAMP 7 with human blood
plasma resulted in a degradation of the peptide to produce one
main metabolite. After treatment of RTbtR-OMe 7 with an
excess of LiOH in water, the obtained tripeptide carboxylic acid
was found to coelute with the metabolite. The half-life (t1/2) of
RTbtR-OMe 7 was found to be between 2 and 4 h. In contrast,
RTbtR-NH2 5 and RTbtR-NHBn 9did not show any degradation,
even after incubation for up to 4 days. No attempts at assessing if
any enzymes were still active after 4 days were performed.
Figure 4. Bacterial kill-kinetics for 7 against S. aureus.
from POPC vesicles was observed upon exposure to either
RWR-OMe or RTbtR-OMe 7.
Bacterial Kill-Kinetics. Exposure of S. aureus cultures to
SAMP 7 at concentrations corresponding to 1× and 2× the
measured MIC values gave a slow onset of antibacterial activity
and the minimum cfu/mL was read after 300 min (Figure 4).
At a peptide concentration corresponding to 8× the MIC value
(Figure 4), a rapid onset of bacterial killing was observed and
after 30 min the number of bacteria had dropped from 106 to
104 cfu/mL.
E. coli Membrane Destabilization. In order to explore
further the ability of SAMP 7 to cause membrane destabilization,
its action on E. coli ML-35p bacteria was investigated. For the
ML-35p strain of E. coli the integrity of the outer and the
cytoplasmic membrane could be assessed directly by measuring
the amount of substrate hydrolysis for the periplasmic enzyme
ꢀ-lactamase and the cytoplasmic enzyme ꢀ-galactosidase, re-
spectively.27 It was found that outer membrane and inner
membrane integrity was lost at a SAMP 7 concentrations of
128 and 256 µg/mL, respectively, whereas the MIC of SAMP
7 against E. coli ML-35p was found to be 32 µg/mL.
Peptide Hydrophobicity. The retention times obtained from
RP-HPLC analyses of the peptides are compiled in Table 1.
All peptides were analyzed in triplicate, and the deviation in
measured retention time was less than 0.15 min between the
parallels. The dipeptides were found to have a higher effective
hydrophobicity, i.e., displayed longer retention times than the
corresponding tripeptide derivatives, and di- and tripeptides
containing an unsubstituted C-terminal amide were found to be
less hydrophobic than the corresponding peptide methyl esters.
Furthermore, the peptide derivatives with a C-terminal benzy-
lamide were found to be the most hydrophobic.
To further investigate the configurational dependence of the
measured hydrophobicity of diastereoisomeric dipeptides 2 and
3 and diastereoisomeric tripeptides 7 and 8, the hydrophobicity
parameter log kw was determined for these four peptides.28 From
the retention times obtained from analyses of peptides at various
isocratic gradients a series of log k values were calculated for
each peptide in the different mobile phase compositions. An
extrapolation to 100% water gave the log kw values (data not
shown). Also here both di- and tripeptides containing the
L-enantiomer of Tbt were found to have higher log kw values
than their diastereoisomers, i.e., displayed higher effective
hydrophobicity.
Discussion
In a previous study we have shown that tetrapeptides
composed of tryptophan and arginine are efficient antibacteri-
als18 and that these peptides constitute the lower limit as to
how short antimicrobial peptides composed of only naturally
occurring amino acids can be made. In the present study we
report the high antibacterial activity of a series of synthetic
antimicrobial peptidomimetics (SAMPs), which have been
prepared by further elaboration of our previously reported
pharmacophore18 for antimicrobial peptides. In earlier com-
munications we have identified a supertryptophan residue,
ꢀ-(2,5,7-tri-tert-butylindol-3-yl)alanine24,25 (Tbt; see Scheme
1) that when combined with arginine derivatives gave rise
to highly active antimicrobial dipeptide derivatives.22 At-
tachment of a second cationic amino acid to produce
derivatives containing three cationic charges and a superbulky
moiety led not only to a further increase in antibiotic potency
but also to a decrease in hemolytic activity (see Table 1).
The overall cationic charge of CAPs is a requisite for their
efficient electrostatic interaction with the negatively charged
bacterial surface and thereby also their antimicrobial activity.
Thus, to eliminate the counterproductive negative charge17 on
the C-terminal and to ensure a high net positive charge of the
SAMPs of this study, we decided to mask the carboxylic acid
Antibacterial Activity. The antibacterial activities of the
peptides against the Gram-negative E. coli and P. aeruginosa
and the Gram-positive S. aureus, the two methicillin resistant
staphylococci strains (MRSA and MRSE), and the S. aureus
strain intermediate resistant to vancomycin (GISA) are shown

Peptidomimetics with Therapeutic Potential
Journal of Medicinal Chemistry, 2008, Vol. 51, No. 14 4309
Table 1. Antibacterial and Hemolytic Activity of SAMPs
minimal inhibitory concentration (MIC)b
SAMPa
E. coli
P. aeruginosa
S. aureus
MRSA
MRSE
GISA
hemolysisc (EC50)
tR (min)d
TbtR-NH2
1
2
3
4
5
6
7
8
9
Nae
30 (57)
30 (55)
50 (92)
7.5 (12)
30 (44)
50 (76)
15 (21)
40 (57)
5 (6.5)
15 (28)
5 (9.2)
15 (28)
5 (9.2)
7.5 (14)
2.5 (4.6)
7.5 (14)
2.5 (4.0)
2.5 (3.7)
2.5 (3.8)
1.0 (1.4)
2.5 (3.6)
1.0 (1.3)
30 (57)
7.5 (14)
30 (55)
5 (8.1)
15 (22)
30 (46)
7.5 (11)
15 (21)
2.5 (3.2)
305 (578)
245 (451)
273 (503)
60 (97)
Nhf
13.85
19.15
13.45
23.88
11.05
10.11
15.69
9.97
TbtR-OMe
TbtR-OMe
TbtR-NHBn
RTbtR-NH2
KTbtR-NH2
RTbtR-OMe
RTbtR-OMe
RTbtR-NHBn
30 (55)
Nae
15 (28)
2.5 (4.0)
7.5 (11)
15 (23)
5 (7.2)
15 (28)
2.5 (4.0)
7.5 (11)
15 (23)
2.5 (3.6)
5 (7.2)
10 (16)
60 (88)
70 (107)
20 (29)
50 (72)
7.5 (9.7)
Nhf
705 (1008)
5 (7.2)
2.5 (3.2)
Nhf
2.5 (3.2)
90 (116)
19.52
a Underlined residues are of D-configuration. b Minimal inhibitory concentration of peptides in µg/mL; values in parentheses are in µM. c Peptide concentration
that induces 50% hemolysis of human erythrocytes, in µg/mL; values in parentheses are in µM. d Hydrophobicity measured as retention time (tR) on a linear
1% acetonitrile/min gradient. e Na: no antibacterial activity within concentration range, i.e. up to 150 µg/mL. f Nh: no hemolytic activity within concentration
range, i.e. up to 1000 µg/mL.
as a primary amide or as a methyl ester. We have previously
found that the C-terminal offers an excellent attachment point
for modulation of hydrophobicity, and with this in mind a
benzylamide derivative was also included.18
short peptides containing tryptophan did not display any
antibacterial activity (vide infra).
Liposomes can also be used as a model system to study the
potential permeability the exposure to CAPs can induce in a
phospholipid bilayer. The encapsulation of fluorescent dyes into
liposomes and the subsequent measurement of their release as
a function of CAP concentration may be used for this purpose.
In our case, we found that RWR-OMe was not able to induce
leakage of the fluorescent dye calcein from POPG vesicles,
which indicated that an association of the peptide to the
membrane is not enough to affect membrane permeabilization.
On the other hand, SAMP 7 was found to induce a concentration
dependent, albeit relatively low, release of calcein from the
negatively charged liposomes. Seemingly, the presence of only
one tryptophan residue in a peptide is not sufficient to induce
membrane destabilization, whereas the presence of a Tbt residue
fulfills the structural requirements (i.e., provides hydrophobic
bulk) previously identified as a pharmacophore for antibacterial
activity and renders the SAMPs sufficiently hydrophobic to
induce destabilization of the vesicle bilayer. It should be noted
that RTbtR-OMe 7 and calcein are of comparable size, and this
may diminish the ability of the SAMP to render the liposomes
permeable to the fluorescence probe. As could be expected from
the low ability to interact with POPC liposomes, no leakage of
calcein from the zwitterionic POPC vesicles was observed by
treatment with RTbtR-OMe 7, indicating a selective action on
negatively charged bacteria-like membranes compared to zwit-
terionic mammalian-like membranes.
The mode of bacterial killing of CAPs is generally believed
to involve the bacterial membranes, and in a recent review the
mode of membrane permeabilization was collectively denoted
the Shai-Matsuzaki-Huang mechanism.4 However, the effects
imposed on various membrane systems by CAPs are vastly
different, and intracellular targets have been established as
possible targets for antibiotic action (for a recent review, see
ref 29). There has yet to be established a consensus over the
mechanism of action, and most likely this varies among different
peptides, some acting solely by membrane disruption and others
by several mechanisms also involving immuno modulator
properties.15,30
To determine if peptide-lipid interactions also are crucial
for the antibiotic activity of SAMPs, the interactions of a
representative SAMP (RTbtR-OMe 7) with large unilamellar
vesicles, a liposome system that can be made to mimic the
properties of various phospholipid bilayers, were investigated.
For comparison an inactive tripeptide was also included in these
studies. Both compounds demonstrated a blue-shifted fluores-
cence maximum in the presence of the negatively charged POPG
vesicles compared to buffer alone, an effect that is commonly
associated with the embedding of the tryptophan residues into
the bilayers.26 The SAMP had a higher propensity to partition
into phospholipid bilayers than the inactive peptide, and the
inactive peptide did not associate with the zwitterionic POPC
membranes, whereas RTbtR-OMe 7 did. The association to
POPC is presumably a hydrophobic effect, as it has been shown
that association to zwitterionic membranes compared to nega-
tively charged membranes is more governed by the lipophilicity
of peptides than their charge.31 The higher blue shift of the Tbt-
containing SAMP compared with the inactive tripeptide contain-
ing tryptophan in all liposome systems indicated an overall
higher ability of the SAMP to associate to phospholipid
membranes, a feature that is essential for membrane active
peptide. However, for the POPC model system, a higher fraction
of the SAMP apparently remains unbound in the buffer solution.
Furthermore, the higher blue shift of Tbt in POPG and POPC/
POPG (3:1) compared to POPC suggests a deeper burial of Tbt
than tryptophan into membranes containing negatively charged
phospholipids. A deeper location of the Tbt residue would lead
to a more effective disruption of the phospholipid packing, and
furthermore, since the side chain of Tbt is 2.5 times larger than
indole itself,20 this may be an explanation of why derivatives
containing Tbt were antibacterial whereas the corresponding
One of the attractive features of CAPs as a novel class of
antibiotics is their fast bacterial killing kinetics, which in most
cases is superior to commercially available antibiotics. To
establish whether this ability also is inherent to SAMPs, the
time-course of bacterial killing was studied by exposing S.
aureus to various concentrations of derivative 7 (Figure 4).
At 8 MIC, which is a therapeutically relevant concentration,
most of the bacteria were killed within 30 min, an extremely
rapid bactericidal effect often also seen for CAPs.32,33 The fast
bacterial killing suggests that at this concentration the antibacte-
rial effect is mediated through a significant permeabilization or
lysis of the bacterial membranes.33 Killing of the bacteria
through the interaction of SAMPs with intracellular targets
would lead to a much slower onset, and consequently intracel-
lular targets cannot be reached before the bacteria have
disintegrated. It should be noted that some extent of membrane
permeabilization is required for the compounds to reach
additional intracellular targets, and for CAPs the kinetics of
bacterial killing and the rate of membrane permeabilization are
not necessarily well correlated.32 At 2 MIC and 1 MIC the

4310 Journal of Medicinal Chemistry, 2008, Vol. 51, No. 14
Haug et al.
maximum bacterial killing is not achieved before 300 min have
passed. Clearly, at high concentrations, SAMPs share the
exceptionally fast bacterial killing kinetics of CAPs, but whether
SAMPs also can benefit from the multitarget mechanism of
action of many CAPs remains to be elucidated.
negative bacteria, presumably because of the additional barrier
afforded by the outer membrane of E. coli and P. aeruginosa.
In general, the tripeptidomimetics showed higher antibiotic
potency relative to the corresponding dipeptidomimetics, which
is most likely attributable to their higher net positive charge.
Higher net charge is expected to result in a stronger initial
interaction of the SAMPs with bacterial surfaces, which is a
crucial step in the bacterial killing mechanism. We have chosen
to use arginine as the cationic residue for construction of highly
active SAMPs because this amino acid is preferred over lysine,36
possibly because of the more dispersed positive charge of the
guanidine compared to the amine and the possibility of multiple
hydrogen bond formation of the former.26 As expected, SAMP
6 containing a N-terminal lysine residue showed lower anti-
bacterial activity than SAMP 5 containing a N-terminal arginine.
Interestingly, the lysine containing SAMP is also less hydro-
phobic (see Table 1).
The mechanism of action for CAPs in many cases involves
a direct destabilizing effect on the bacterial membranes,34 and
the ability of SAMPs to interact with phospholipid membranes
was demonstrated in the liposome studies (vide supra). In order
to gain some insight into the interaction with bacterial mem-
branes, the membrane integrity of a strain of E. coli when
exposed to SAMP 7 was investigated. With an intact outer
membrane, bacteria of the E. coli ML-35p strain are only able
to cause a minimal ꢀ-lactamase induced hydrolysis of the
chromogenic cephalosporin CENTA and are also unable to
hydrolyze 2-nitrophenyl ꢀ-D-galactopyranoside (ONPG), which
is a substrate for the cytoplasmic ꢀ-galactosidase enzyme.27 We
found that outer and inner membrane integrity was lost at a
SAMP 7 concentration of 128 and 256 µg/mL, respectively,
indicating that SAMPs may have a direct bacterial killing effect
through membrane lysis at high concentrations. The relatively
large difference in SAMP concentration required for bacterial
killing and for loss of membrane integrity may indicate that at
concentrations closer to the MIC value mechanisms other than
membrane permeabilization contribute to the antimicrobial
activity. On the contrary, the bacterial membranes may well be
permeable enough to cause a lethal effect at concentrations lower
than those where efficient hydrolysis of the chromogenic
substrates was observed.
To evaluate the in vitro toxicity of the SAMPs, their hemolytic
activity against human red blood cells was measured (Table
1). In general, tripeptidomimetic derivatives were found to be
overall less hemolytic than the dipeptidomimetic derivatives.
It thus seems that a higher net positive charge disfavors
interaction with the zwitterionic membranes of erythrocytes. The
ability of CAPs to discriminate between bacterial surfaces and
the surfaces of our own cells is determined by the overall
hydrophobicity of the peptides, and highly hydrophobic CAPs
are often found to be toxic also to human cells. This was also
found to be the case for the SAMPs presented in this study and
was in accordance with Kondejewski et al.,37 who also reported
a correlation between hydrophobicity and hemolytic activity.
Clearly the super-tryptophan residue Tbt confers high anti-
bacterial activity for both the di- and tripeptidomimetics (Table
1) and fulfills the lipophilic bulk requirements of the pharma-
cophore by itself. By comparison, the tryptophan analogues of
peptides 2 and 7 (see Figure 1), WR-OMe and RWR-OMe, both
showed MIC values above 500 µg/mL against E. coli and S.
aureus (data not shown).
The two sets of diastereomeric compounds, SAMPs 2 and 3
and SAMPs 7 and 8, showed different biological activity (Table
1), and changing the chirality of the Tbt residue dramatically
affected the RP-HPLC retention times.38 Also for longer CAPs
it has been shown that introduction of D-amino acids can alter
the hydrophobicity and biological activity as a result of the
structural changes induced.31 We have also recently reported a
good correlation between the antibacterial activity and RP-HPLC
retention times for tripeptide derivatives containing novel biaryl
amino acids.23 A similar hydrophobicity-antibacterial activity
relationship was found for all bacterial strains. Furthermore, the
least hydrophobic diastereoisomers of TbtR-OMe 2 and TbtR-
OMe 3 and of RTbtR-OMe 7 and RTbtR-OMe 8 were also
found be less hemolytic. Apparently, also for SAMPs the
different configuration of the diastereoisomers influences both
their interactions with the C18-matrix of the RP-HPLC column
and their interactions with bacteria and human erythrocytes.
Moreover, it is noteworthy that the MIC and EC50 values are
differently affected by configurational changes in the SAMPs,
an observation that opens up the possibility of fine-tuning the
biological activity by manipulating the configuration.
All the SAMPs prepared in this study were found to be
equally active against the MRSA strain and the normal S.
aureus. The MRSE strain proved to be the most susceptible
bacterium, and the GISA strain was somewhat less sensitive
the SAMPs. Normally bacterial resistance against vancomycin
is caused by a change in the peptide side chain of the
peptidoglycan, which in resistant strains ends in D-Ala-D-Lac
rather than the D-Ala-D-Ala, an alteration that makes the binding
affinity of vancomycin a 1000-fold lower (for a review, see ref
35). This reprogramming of the vancomycin target is not
expected to influence the antibacterial activity of the SAMPs
against this bacterium. Apparently, the difference in the lowest
obtained MIC value against each of the four staphylococci
strains stems from susceptibility variations between the bacteria
rather than being connected to antibiotic resistance.
Most of the antibacterial active SAMPs were found to be
bactericidal, as the minimal bactericidal concentrations (MBC)
generally were found to be identical to or one titer-step higher
than the MIC values obtained (data not shown). However, in
some cases and mainly for derivatives displaying relatively high
MIC values, the MBC values against the Gram-negative bacteria
E. coli and P. aeruginosa were several titer-steps higher than
the MIC values. Whether the additional barrier afforded by the
outer membrane of Gram-negative bacteria causes this difference
or if this variance is inherent in the method for determination
of antibacterial activity remains to be elucidated. In addition
both di- and tripeptides generally showed higher antibacterial
activity against the Gram-positive strains than against the Gram-
Most naturally occurring antimicrobial peptides are rather
large molecules, and therefore, many do not have the physio-
chemical properties that are desirable for new therapeutics.
Peptides are also prone to enzymatic degradation in vivo and
thus may require encapsulation for systemic use; alternatively,
use of antimicrobial peptides may be limited to topical indica-
tions. One class of antimicrobial peptides that shows promising
proteolytic stability are the cyclic D,L-R-peptides, which are
proposed to assemble into tubular structures upon interaction
with bacterial cell membranes.39 Presumably, this unique
architecture and the alternating D- and L-amino acids provide
the stability of these peptides in vivo. In addition, aggregation
of peptide monomers is also reported to increase the stability

Peptidomimetics with Therapeutic Potential
Journal of Medicinal Chemistry, 2008, Vol. 51, No. 14 4311
ꢀ-(2,5,7-tri-tert-Butylindol-3-yl)alanine (Tbt).25 A mixture of
H-Trp-OH (19.14 g, 94 mmol) and t-BuOH (74.18 g, 1.0 mol) in
trifluoroacetic acid (290 mL) was stirred at room temperature for
69 h (Scheme 1). The resulting dark solution was evaporated to a
black oil under reduced pressure, and water (100 mL) was added.
To the resulting suspension was added potassium hydrogen carbon-
ate until the pH was neutral. Constant trituration of the oily material
using a spatula during the addition of base resulted in the formation
a pinkish gumlike material. The aqueous layer was decanted off,
and the gumlike material was crystallized from 50% ethanol in
water. The first crop of crystalline material (21.48 g) was recrystal-
lized from 50% ethanol in water to give the title compound as a
colorless crystalline solid (11.74 g, 34%). Analytical data were in
accordance with literature data.25
Boc-Tbt-OH. H-Tbt-OH was Boc protected following a standard
procedure.42 To a solution of H-Tbt-OH (3.41 g, 9.15 mmol) in
dioxane/water, 1:1 (80 mL), was added sodium hydrogen carbonate
(2.34 g, 27.85 mmol), and the resulting mixture was stirred at room
temperature overnight. The reaction mixture was evaporated to a
solid residue under reduced pressure and separated between ethyl
acetate and a 5% aqueous solution of potassium hydrogen carbonate.
The aqueous layer was extracted twice more with fresh ethyl acetate,
and the combined organic phases were washed with a 5% aqueous
solution of potassium hydrogen carbonate and a saturated aqueous
solution of sodium chloride before it was dried over anhydrous
magnesium sulfate. After removal of the drying agent by filtration
and evaporation under reduced pressure the title compound was
isolated as a light-yellowish foam (4.14 g, 95% crude yield). The
crude material was used without any further purification. Analytical
data were in accordance with literature data.43
against proteases of other antimicrobial peptides.31 In any case,
the in vivo biological activity is often affected by proteolytic
hydrolysis, and this is an issue that needs to be resolved before
these compounds can enter the clinic.30 In this respect, a stability
study of SAMPs 5, 7, and 9 in human blood plasma was
undertaken in order to assess their potential as novel drug
candidates. Incubation of RTbtR-OMe 7 with human blood
plasma resulted in degradation producing one main metabolite,
RTbtR-OH; thus, it would appear that the SAMP was subject
to cleavage by esterases in blood plasma. The antibacterial
activity of the main metabolite was found to be >100 and 75
µg/mL against E. coli and S. aureus, respectively (data not
shown); i.e., metabolism of RTbtR-OMe 7 renders it consider-
ably less active. Even though the carboxylic acid derivative has
the same net positive charge as the dipeptidomimetics, the
antibacterial activity was considerably lower. The negative
charge of the carboxylate seemingly disfavors interactions with
bacteria, an observation that underscores the importance of a
high net positive charge as a structural element in highly active
SAMPs, and clearly demonstrates the importance of avoiding
zwitterionic structures when designing novel SAMPs. The amide
RTbtR-NH2 5 and the benzylamide derivative RTbtR-NHBn 9
did not show any enzymatic degradation in human blood plasma.
These results clearly demonstrate that it is possible to design
SAMPs that may be pharmacologically useful. The Tbt residue
used in the present study has previously been found to induce
proteolytic protection to peptides in which it has been incor-
porated.40
Binding to albumin, as has recently been reported for a series
of related compounds,41 will influence systemic antibacterial
activity but will likely also contribute to a reduced clearance.
Further studies into the pharmacokinetics of this novel class of
compounds will establish their potential for clinical use;
however, the ability to modulate their stability in blood plasma
gives further promise to the progression of this class of
compounds toward new antibacterial drugs.
General Procedure for Peptide Couplings. Boc-Tbt-OH (0.474
g, 1.0 mmol), H-Arg-OMe dihydrochloride (0.276 g, 1.06 mmol)
and 1-HOBt (0.165 g, 1.22 mmol) were dissolved in DMF (4 mL)
in a vial. DIPEA (0.755 mL, 4.41 mmol) was added and the solution
cooled in ice/water. HBTU (0.458 g, 1.21 mmol) was added, and
the solution was cooled in ice/water for another 10 min before the
vial was agitated on an orbital shaker for 1.5-2 h at room
temperature. The reaction mixture was diluted with ethyl acetate
(21 mL) and extracted with 2 × 8 mL of citric acid in aqueous
sodium chloride, 2 × 8 mL of sodium bicarbonate in aqueous
sodium chloride, and 20 mL of saturated aqueous sodium chloride.
All extractions were performed using an automated Myriad ALLEX
(Mettler-Toledo, U.K.) liquid-liquid extractor. The ethyl acetate
phase was dried over MgSO4, filtered, and evaporated to yield
0.63 g of a colorless solid. The Boc protecting group was removed
by treatment with 10 mL of TFA-H2O (95:5) by stirring in the
dark for 2.5 h. The oil obtained after evaporation was treated with
a solution of p-toluenesulfonic acid (2.5 equiv) in diethyl ether,
and the resulting colorless solid material was triturated again before
it was dried in vacuo.
Conclusions
In conclusion, we have prepared novel antibiotic compounds,
SAMPs, that display high specificity toward bacteria relative
to eukaryotic red blood cells and show antibacterial activity in
the low micromolar range against several clinically relevant
pathogens. As for cationic antimicrobial peptides, from which
SAMPs have been derived through systematic investigations
of structure-activity relationships, both antibacterial activity
and toxicity toward human cells are modulated by hydrophobic-
ity. Our preliminary investigations into the antibiotic mechanism
suggest that peptidomimetic-lipid interactions, mediated through
the presence of a supertryptophan residue, play an important
role in the mechanism of action of the SAMPs reported. Our
results show that the attractive features of CAPs can be
condensed to produce SAMPs, compounds that can give the
additional proteolytic stability that is often desired for clinical
use and thus provide superior drugability compared to antimi-
crobial peptides. Further evaluation of these compounds in
relevant biological systems will be reported in due course.
Purification. Peptide crude products were either taken directly
to the next coupling step or purified by reversed-phase high
performance liquid chromatography (RP-HPLC) using a C18-column
(Delta-PakTM C18, 100 Å, 15 µm, 25 mm × 100 mm, Waters Corp.,
Milford, MA) with a mixture of water and acetonitrile (both
containing 0.1% TFA) as mobile phase and UV detection at 254
nm. The homogeneity of the purified peptides was analyzed by RP-
HPLC on an analytical C18-column (Delta-Pak C18, 100Å, 5 µm,
3.9 mm × 150 mm, Waters Corp., Milford, MA), and all peptides
were found to be >96.5% pure. Correct peptide masses (Table 1)
were confirmed by positive ion electrospray ionization mass
spectrometry on a VG Quattro quadrupole mass spectrometer (VG
Instruments Inc., Altringham, U.K.).
Experimental Section
H-Arg-Tbt-Arg-NH2 (5). 1H NMR (400 MHz, methanol-d3) δ:
7.84 (1H, d, J ) 8.2 Hz), 7.44 (1H, d, J ) 1.5 Hz), 7.17 (1H, d,
J ) 1.5 Hz), 4.60 (1H, dd, J ) 10.6, 5.1 Hz), 4.26-4.19 (1H, m),
4.01 (1H, app t, J ) 6.1 Hz), 3.49 (1H, dd, J ) 14.2, 10.6 Hz),
3.29 (3H, t, J ) 7.1 Hz), 3.18-3.09 (2H, m), 2.06-1.95 (2H, m),
1.89-1.65 (3H, m), 1.56 (9H, s), 1.52 (9H, s), 1.40 (9H, s),
Amino acids were purchased from Bachem (Bubendorf, Swit-
zerland), and coupling reagents, DMF, DIPEA, and TFA were
purchased from Fluka (Buchs, Switzerland). The ꢀ-lactamase
substrate CENTA was purchased from Calbiochem (Merck KGaA,
Darmstadt, Germany), and the ꢀ-galactosidase substrate ONPG was
obtained from Sigma-Aldrich (Schnelldorf, Germany).

4312 Journal of Medicinal Chemistry, 2008, Vol. 51, No. 14
Haug et al.
1.56-1.40 (3H, m). ESMS Calcd for C35H61N11O3 (average isotope
composition) 683.93 [M]+. Found 684.7. HPLC: tR ) 14.51 min,
100%.
a scan speed of 60 nm/s. The concentrations of the peptides and
lipids were 20 µg/mL and 0.3 mg/mL, respectively. In all cases
the fluorescence spectra were corrected by subtracting the residual
fluorescence of the medium.
H-Lys-Tbt-Arg-OMe (6). 1H NMR (400 MHz, methanol-d3)
δ: 7.81 (1H, d, J ) 8.4 Hz), 7.44 (1H, d, J ) 1.5 Hz), 7.17 (1H,
d, J ) 1.5 Hz), 4.59 (1H, dd, J ) 10.8, 5.0 Hz), 4.26-4.18 (1H,
m), 3.97 (1H, t, J ) 6.48 Hz), 3.49 (1H, dd, J ) 14.2, 10.8 Hz),
3.34-3.30 (1H, m), 3.18-3.10 (2H, m), 3.06-3.00 (2H, m),
2.03-1.89 (2H, m), 1.89-1.70 (3H, m), 1.63-1.53 (5H, m), 1.56
(9H, s), 1.52 (9H, s), 1.41 (9H, s). ESMS Calcd for C35H61N9O3
(average isotope composition) 655.92 [M]+. Found 656.5. HPLC:
tR ) 14.26 min, 99.1%.
Calcein Release. Liposomes (see above) were separated from
unentrapped calcein by gel filtration on a Sephadex G 50 (20 cm
× 1.5 cm) column (Pharmacia Biotech AB, Sweden) eluting with
calcein free buffer (see above). The lipid concentration was
determined by quantitative phosphorus analysis.45 Aliquots of
liposome suspension were then diluted in calcein-free buffer to a
final concentration of 50 µM lipid and incubated for 5 min with
different concentrations of different peptide solutions. Calcein
release from LUVs was assessed fluorometrically by measuring the
increase in fluorescence intensity resulting from the decrease in
the level of self-quenching of calcein (excitation at 489 nm and
emission at 520 nm). The fluorescence intensity corresponding to
100% calcein release was determined by addition of a 10% solution
(w/v) of Triton X-100. Relative leakage was calculated using the
formula ((Fx - F0) × 100)/(Ft - F0) where Fx is the intensity
measured at a given concentration of peptide, F0 is the intensity of
the liposomes (background), and Ft is the intensity after lysis by
Triton X-100.
1
H-Arg-Tbt-Arg-OMe (7). H NMR (400 MHz, methanol-d3)
δ: 7.60-7.55 (1H, d, J ) 8.0 Hz), 7.32 (1H, d, J ) 1.2 Hz), 7.12
(1H, d, J ) 1.2 Hz), 4.59 (1H, dd, J ) 10.5, 5.3 Hz), 4.32-4.24
(1H, m), 4.01 (1H, app t, J ) 6.1 Hz), 3.44-3.36 (1H, m), 3.40
(3H, s), 3.32-3.26 (3H, m), 3.14-3.08 (2H, m), 2.06-1.94 (2H,
m), 1.88-1.65 (3H, m), 1.62-1.53 (3H, m), 1.56 (9H, s), 1.53
(9H, s), 1.40 (9H, s). ESMS Calcd for C36H62N10O4 (average isotope
composition) 698.94 [M]+. Found 700.0. HPLC: tR ) 15.55 min,
96.9%.
1
H-Arg-D-Tbt-Arg-OMe (8). H NMR (600 MHz, H2O/D2O,
9:1) δ: 8.77 (1H, d, J ) 7.2 Hz), 8.36 (1H, s), 7.66 (1H, d, J ) 6.0
Hz), 7.14 (2H, bs), 7.08 (1H, s), 6.82-6.79 (1H, m), 6.65-6.34
(5H, bs), 4.48-4.42 (1H, m), 3.95-3.93 (1H, m), 3.70-3.67 (1H,
m), 3.54 (3H, s), 3.25-3.17 (1H, m), 3.16 (1H, dd, J ) 14.7, 5.1
Hz), 3.09-3.00 (2H, m), 2.72-2.66 (2H, m), 1.80-1.71 (2H, m),
1.50-1.45 (2H, m), 1.34 (9H, s), 1.31 (9H, s), 1.25-1.18 (1H,
m), 1.22 (9H, s), 0.92-0.85 (1H, m), 0.73-0.65 (2H, m). ESMS
Calcd for C36H62N10O4 (average isotope composition) 698.94 [M]+.
Found 700.0. HPLC: tR ) 14.02 min, 97.1%.
Time-Kill Studies. S. aureus ATCC 25923 was grown in 2%
Bacto Peptone water (Difco 1807-17-4) until exponential growth
and aliquots of bacterial culture (inoculum of 106 cfu/mL) were
exposed to peptide 7 at concentrations corresponding to 1×, 2×,
and 8× the measured MIC value. The cfu/mL was monitored, and
readings were taken at 0, 10, 30, 60, 120, and 300 min. Untreated
bacterial culture was used as control.
E. coli ML-35p Membrane Integrity. The outer and cytoplas-
mic membrane permeabilization assays on E. coli ML-35p were
performed following a literature procedure with minor modifica-
tions.27 Briefly, the assay mixtures contained 50 µL of 10 mM
NaPBS buffer (pH 7.4, with and without 100 mM NaCl), 4 × 106
cfu/mL of E. coli ML-35p in 50µL in 10 mM NaPBA buffer (pH
7.4), either 50 µL of the ꢀ-lactamase substrate (80 µM in NaBPS
buffer) or the ꢀ-galactosidase substrate (10 mM in NaPBS buffer)
and 50 µL of SAMP 7 in various concentrations. After incubation
at 37 °C with shaking every 10 s for 1 h, the absorbance at 420 nm
was read. In the present study (6R,7R)-3-[[(3-carboxy-4-nitrophe-
nyl)thio]methyl]-8-oxo-7-[(2-thienylacetyl)amino]-5-thia-1-
azabicyclo[4.2.0]oct-2-ene-2-carboxylic acid (CENTA) and 2-ni-
trophenyl-ꢀ-D-galactopyranoside (ONPG) were used as substrates
for the ꢀ-lactamase and ꢀ-galactosidase, respectively. As control
experiments, incubation without bacteria, without SAMP 7, and
with mellitin instead of SAMP 7 was performed.
H-Arg-Tbt-Arg-NHBn (9). 1H NMR (400 MHz, methanol-d3)
δ: 7.48 (1H, d, J ) 1.4 Hz), 7.33-7.21 (3H, m), 7.19 (1H, d, J )
1.4 Hz), 7.08 (2H, d, J ) 7.4 Hz), 4.70 (1H, dd, J ) 10.1, 5.3 Hz),
4.38-4.30 (1H, m), 4.17-4.08 (1H, m), 3.99 (1H, app t, J ) 5.8
Hz), 3.94-3.87 (1H, m), 3.49 (1H, dd, J ) 14.4, 10.1 Hz),
3.33-3.29 (1H, m), 3.29-3.21 (2H, m), 3.18-3.05 (2H, m),
2.05-1.91 (2H, m), 1.89-1.65 (3H, m), 1.64-1.51 (3H, m), 1.58
(9H, s), 1.53 (9H, s), 1.41 (9H, s). ESMS Calcd for C42H67N11O3
(average isotope composition) 774.05 [M]+. Found 774.0. HPLC:
tR ) 19.52 min, 98.4%.
For analytical data on all compounds see Supporting Information.
Preparation of Liposomes. Large unilamellar vesicles (LUVs)
of 1-palmitoyl-2-oleoyl-sn-glycero-3-phosphoglycerol (POPG) am-
monium salt, 1-palmitoyl-2-oleoyl-sn-glycero-3-phosphocholine
(POPC) (Genzyme Pharmaceuticals, Switzerland), and a mixture
of POPC/POPG (3:1) were prepared as previously described.44
Briefly, a defined amount of lipid was dissolved in a chloroform/
methanol solvent mixture (2:1 v/v). The solvents were evaporated
in vacuo at 40 °C, forming a thin lipid film that was hydrated with
a buffer comprising 10 mM Hepes, 150 mM NaCl, and 1 mM
EDTA at pH 7.4. For dye release experiments, the film was hydrated
with calcein buffer (25 mM calcein, 10 mM Hepes, 150 mM NaCl,
1 mM EDTA and adjusted to pH 7.4). After evaporation and
rehydration, the produced liposome dispersion was frozen and
thawed three times. To make LUVs by extrusion, the multilamellar
vesicles (MLVs) or frozen and thawed preparations were passed
10 times through Isopore membrane filters with decreasing pore
size of 0.8, 0.4, 0.2, 0.1 µm, respectively, using a miniextruder
(Avestin, Ottawa, Canada). Sizes of liposomes were determined
by light scattering using a Nicomp submicrometer particle sizer
(Nicomp Particle Sizing System, Santa Barbara, CA). Monomodal
size distribution were fitted in all cases with a diameter between
and 62 and 152 nm for empty liposomes and between 150 and 166
nm for liposomes with entrapped calcein.
Hydrophobicity. All peptides were analyzed by RP-HPLC (see
above) using a linear gradient ranging from 30% to 60% acetonitrile
in water (both containing 0.1% TFA) in 30 min on an analytical
C18-column (see above). The column temperature was set to 30 °C
for all experiments and the flow rate set to 1 mL/min.
Antibacterial Activity. The bacterial strains Pseudomonas
aeruginosa ATCC 27853, Escherichia coli ATCC 25922, Staphy-
lococcus aureus ATCC 25923, methicillin resistant Staphylococcus
aureus (MRSA) ATCC 33591, methicillin resistant Staphylococcus
epidermidis (MRSE) ATCC 27626, and glycopeptide (vancomycin)
intermediate-resistant Staphylococcus aureus (GISA) CCUG 43316
or 43315 (peptide 4 and 9) were grown in 2% Bacto Peptone water
(Difco 1807-17-4) until growth was exponential. A standard
microdilution technique with an inoculum of 2 × 106 cfu/mL was
used. The minimal inhibitory concentration (MIC) of the peptides
was determined in 1% Bacto Peptone water after incubation
overnight at 37 °C. The concentration range used for the peptides
against the Gram-positive strains was 50, 30, 20, 15, 10, 7.5, 5.0,
2.5, 1.0, and 0.5 µg/mL. In addition the concentration series 150,
100, 90, 80, 70, 60, 50, 40, 30, and 20 µg/mL was used against the
Gram-negative strains. All peptides were tested at least twice in
parallel. Gentamicin was used as a positive control in the antibacte-
rial assay and MIC values of 0.1-1 and <0.05 µg/mL were found
for E. coli and S. aureus, respectively.
Fluorescence Spectroscopy. Intrinsic fluorescence of peptides
in buffer solution (10 mM Hepes/150 mM NaCl, 1 mM EDTA at
pH 7.4) was measured using a Perkin-Elmer luminescence fluo-
rimeter LS-50 B. Individual emission spectra of the peptides with
and without lipid vesicles were taken between 310 and 450 nm
(both excitation and emission bandwidths were set to 10 nm) with

Peptidomimetics with Therapeutic Potential
Journal of Medicinal Chemistry, 2008, Vol. 51, No. 14 4313
(13) Hancock, R. E. W.; Chapple, D. S. Peptide antibiotics. Antimicrob.
Agents Chemother. 1999, 43, 1317–1323.
(14) Wu, M. H.; Maier, E.; Benz, R.; Hancock, R. E. W. Mechanism of
interaction of different classes of cationic antimicrobial peptides with
planar bilayers and with the cytoplasmic membrane of Escherichia
coli. Biochemistry 1999, 38, 7235–7242.
(15) McPhee, J. B.; Hancock, R. E. W. Function and therapeutic potential
of host defence peptides. J. Pept. Sci. 2005, 11, 677–687.
(16) Jenssen, H.; Hamill, P.; Hancock, R. E. W. Peptide Antimicrobial
Agents. Clin. Microbiol. ReV. 2006, 19, 491–511.
(17) Haug, B. E.; Strøm, M. B.; Svendsen, J. S. The medicinal chemistry
of short lactoferricin-based antibacterial peptides. Curr. Med. Chem.
2007, 14, 1–18.
(18) Strøm, M. B.; Haug, B. E.; Skar, M. L.; Stensen, W.; Stiberg, T.;
Svendsen, J. S. The pharmacophore of short cationic antibacterial
peptides. J. Med. Chem. 2003, 46, 1567–1570.
(19) Haug, B. E.; Svendsen, J. S. The role of tryptophan in the antibacterial
activity of a 15-residue bovine lactoferricin peptide. J. Pept. Sci. 2001,
7, 190–196.
(20) Haug, B. E.; Skar, M. L.; Svendsen, J. S. Bulky aromatic amino acids
increase the antibacterial activity of 15-residue bovine lactoferricin
derivatives. J. Pept. Sci. 2001, 7, 425–432.
(21) Haug, B. E.; Andersen, J.; Rekdal, Ø.; Svendsen, J. S. Synthesis of a
2-arylsulfonylated tryptophan: the antibacterial activity of bovine
lactoferricin peptides containing Trp(2-Pmc). J. Pept. Sci. 2002, 8,
307–313.
Hemolytic Activity. Freshly isolated human red blood cells
(RBC), isolated as described by Dathe and co-workers,46 were
incubated for 1 h at 37 °C with 1, 50, 100, 500, and 1000 µg/mL
solutions of peptides dissolved in PBS. The samples were centri-
fuged at 4000 rpm for 5 min before the absorbance of the
supernatant was measured at 540 nm by a microtiter plate reader
(Thermomax Molecular Devices, NJ). The 0% hemolysis and 100%
hemolysis were determined in PBS and 1% Triton X-100, respec-
tively. Peptide concentrations corresponding to 50% hemolysis
(EC50) were determined from the dose-response curves.
Plasma Stability Assay. Heparinized blood plasma (0.5 mL)
and peptide (20 µL of 1.09-1.25 mg/mL stock solutions of peptide
in water) were mixed and incubated at 37 °C. After different time
intervals the blood samples were applied onto an Oasis extraction
cartridge (Waters Corp., Milford, MA) using 1 mL of water
containing 0.1% TFA. After sample application the cartridge was
washed with 1 mL of water containing 0.1% TFA, 1 mL of 5%
methanol in water containing 0.1% TFA and eluted with 1 mL of
40% acetonitrile in water (both containing 0.1% TFA). External
standard was added directly to the eluted solutions. Samples were
analyzed by RP-HPLC with UV detection at 280 nm using the same
analytical column and data system as above.
(22) Haug, B. E.; Stensen, W.; Stiberg, T.; Svendsen, J. S. Bulky
nonproteinogenic amino acids permit the design of very small and
effective cationic antibacterial peptides. J. Med. Chem. 2004, 47, 4159–
4162.
(23) Haug, B. E.; Stensen, W.; Svendsen, J. S. Application of the Suzuki-
Miyaura cross-coupling to increase antimicrobial potency generates
promising novel antibacterials. Bioorg. Med. Chem. Lett. 2007, 17,
2361–2364.
(24) Wu¨nsch, E.; Jaeger, E.; Kisfaludy, L.; Lo¨w, M. Side reactions in
peptide synthesis: tert-butylation of tryptophan. Angew. Chem., Int.
Ed. Engl. 1977, 16, 317–318.
(25) Lo¨w, M.; Kisfaludy, L.; Jaeger, E.; Thamm, P.; Knof, S.; Wu¨nsch, E.
Direct tert-butylation of tryptophan: preparation of 2,5,7-tri-tert-
butyltryptophan. Hoppe-Seyler’s Z. Physiol. Chem. 1978, 359, 1637–
1642.
(26) Vogel, H. J.; Schibli, D. J.; Jing, W.; Lohmeier-Vogel, E. M.; Epand,
R. F.; Epand, R. M. Towards a structure-function analysis of bovine
lactoferricin and related tryptophan- and arginine-containing peptides.
Biochem. Cell Biol. 2002, 80, 49–63.
(27) Lehrer, R. I.; Barton, A.; Ganz, T. Concurrent assessment of inner
and outer membrane permeabilization and baceriolysis in E. coli by
multiple-wavelength spectrophotometry. J. Immunol. Methods 1988,
108, 153–158.
(28) Braumann, T. Determination of hydrophobic parameters by reversed-
phase liquid chromatography: theory, experimental techniques, and
application in studies on quantitative structure-activity relationships.
J. Chromatogr. 1986, 373, 191–225.
(29) Hancock, R. E. W.; Rozek, A. Role of membranes in the activities of
antimicrobial cationic peptides. FEMS Microbiol. Lett. 2002, 206, 143–
149.
Acknowledgment. This work was funded by the Research
Council of Norway (grant to B.E.H.). The authors thank Lars
Vorland and Ørjan Samuelsen at the University Hospital of
Tromsø, Trine Stiberg at the Department of Chemistry, Uni-
versity of Tromsø, Petra Dalheim and Mari Wikman at the
Department of Biochemistry, University of Tromsø for providing
us with the MIC and hemolysis data, Martin Brandl at the
Institute for Pharmacy for assistance with liposome studies, Fred
Leeson for running NMR experiments and for linguistic
assistance, and Hilde Ulvatne Marthinsen at Toslab AS for
performing the time-kill studies and the E. coli ML-35p
bacterial assays.
Supporting Information Available: HPLC traces, 1H NMR
spectra, analytical data, and structures for all compounds. This
material is available free of charge via the Internet at http://
pubs.acs.org.
References
(1) Hancock, R. E. W. The end of an era. Nat. ReV. Drug DiscoVery 2007,
6, 28.
(2) Mwangi, M. M.; Wu, S. W.; Zhou, Y.; Sieradzki, K.; Lencastre, H.
d.; Richardson, P.; Bruce, D.; Rubin, E.; Myers, E.; Siggia, E. D.;
Tomasz, A. Tracking the in vivo evolution of multidrug resistance in
Staphylococcus aureus by whole-genome sequencing. Proc. Natl.
Acad. Sci. U.S.A. 2007, 104, 9451–9456.
(3) A joint focus on the need for new antibacterials was recently
featured by the journals Nature Biotechnology and Nature ReViews
Drug DiscoVery. For Web site, see http://www.nature.com/focus/
antibacterials/index.html.
(4) Zasloff, M. Antimicrobial peptides of multicellular organisms. Nature
2002, 415, 389–395.
(30) Hancock, R. E. W.; Sahl, H.-G. Antimicrobial and host-defence
peptides as new anti-infective therapeutic strategies. Nat. Biotechnol.
2006, 24, 1551–1557.
(31) Shai, Y.; Oren, Z. From “carpet” mechanism to de-novo designed
diastereomeric cell-selective antimicrobial peptides. Peptides 2001,
22, 1629–1641.
(32) Friecrich, C. L.; Moyles, D.; Beveridge, T. J.; Hancock, R. E. W.
Antibacterial action of structurally diverse cationic peptides on Gram-
positive bacteria. Antimicrob. Agents Chemother. 2000, 44, 2086–
2092.
(5) Hancock, R. E. W.; Scott, M. G. The role of antimicrobial peptides
in animal defenses. Proc. Natl. Acad. Sci. U.S.A. 2000, 97, 8856–
8861.
(33) Sader, H. S.; Fedler, K. A.; Rennie, R. P.; Stevens, S.; Jones, R. N.
Omiganan pentahydrochloride (MBI 226), a topical 12-amino-acid
cationic peptide: spectrum of antimicrobial activity and measurements
of bactericidal activity. Antimicrob. Agents Chemother. 2004, 48,
3112–3118.
(34) Sitaram, N.; Nagaraj, R. Interaction of antimicrobial peptides with
biological and model membranes: structural and charge requirements
for activity. Biochim. Biophys. Acta 1999, 1462, 29–54.
(35) Walsh, C. Molecular mechanisms that confer antibacterial drug
resistance. Nature 2000, 406, 775–781.
(36) Kang, J. H.; Lee, M. K.; Kim, K. L.; Hahm, K.-S. Structure-biological
activity relationships of 11-residue highly basic peptide segment of
bovine lactoferrin. Int. J. Pept. Protein Res. 1996, 48, 357–363.
(37) Kondejewski, L. H.; Lee, D. L.; Jelokhani-Niaraki, M.; Farmer, S. W.;
Hancock, R. E. W.; Hodges, R. S. Optimization of microbial specificity
in cyclic peptides by modulation of hydrophobicity within a defined
structural framework. J. Biol. Chem. 2002, 277, 67–74.
(6) Raj, P. A.; Dentino, A. R. Current status of defensins and their role
in innate and adaptive immunity. FEMS Microbiol. Lett. 2002, 206,
9–18.
(7) For a complete and updated list, see http://aps.unmc.edu/AP/main.php.
(8) Ludtke, S.; He, K.; Huang, H. Membrane thinning caused by magainin
2. Biochemistry 1995, 34, 16764–16769.
(9) Hancock, R. E. W.; Bell, A. Antibiotic uptake into Gram-negative
bacteria. Eur. J. Clin. Microbiol. Infect. Dis. 1988, 7, 713–720.
(10) Shai, Y. Mechanism of the binding, insertion and destabilization of
phospholipid bilayer membranes by R-helical antimicrobial and cell
non-selective membrane-lytic peptides. Biochim. Biophys. Acta 1999,
1462, 55–70.
(11) Matsuzaki, K. Why and how are peptide-lipid interactions utilized
for self-defense? Magainins and tachyplesins as archetypes. Biochim.
Biophys. Acta 1999, 1462, 1–10.
(12) Huang, H. W. Action of antimicrobial peptides: two-state model.
Biochemistry 2000, 39, 8347–8352.

4314 Journal of Medicinal Chemistry, 2008, Vol. 51, No. 14
Haug et al.
(38) The D-Tbt was prepared by alkylation of the unnatural enantiomer of
tryptophan.
(43) Lo¨w, M.; Kisfaludy, L.; Soha´r, P. tert-Butylation of the tryptophan
indole ring during the removal of the tert-butyloxycarbonyl group.
Hoppe-Seyler’s Z. Physiol. Chem. 1978, 359, 1643–1651.
(39) Fernandez-Lopez, S.; Kim, H.-S.; Choi, E. C.; Delgado, M.; Granja,
J. R.; Khasanov, A.; Kraehenbuehl, K.; Long, G.; Weinberger, D. A.;
Wilcoxen, K. M.; Ghadiri, M. R. Antibacterial agents based on the
cyclic D,L-R-peptide architecture. Nature 2001, 412, 452–455.
(40) Sajgo, M.; Lo¨w, M. Enzymatic digestibility of peptides containing
tert-butyltryptophan residues. Hoppe-Seyler’s Z. Physiol. Chem. 1979,
360, 9–12.
(44) Mayer, L. D.; Hope, M. J.; Cullis, P. R. Vesicles of variable size
produced by a rapid extrusion procedure. Biochim. Biophys. Acta 1986,
858, 161–168.
(45) Eibl, H.; Lands, W. E. A new, sensitive determination of phosphate.
Anal. Biochem. 1969, 30, 51–57.
(41) Svenson, J.; Brandsdal, B. O.; Stensen, W.; Svendsen, J. S. Albumin
binding of short cationic antimicrobial micropeptides and its influence
on the in vitro bactericidal effect. J. Med. Chem. 2007, 50, 3334–
3339.
(42) Keller, O.; Keller, W. E.; van Look, G.; Wersin, G. tert-Butoxycar-
bonylation of amino acids and their derivatives: N-tert-butoxycarbonyl-
L-phenylalanine. Org. Synth. 1985, 63, 160–170.
(46) Dathe, M.; Schu¨mann, M.; Wieprecht, T.; Winkler, A.; Beyermann,
M.; Krause, E.; Matsuzaki, K.; Murase, O.; Bienert, M. Peptide helicity
and membrane surface charge modulate the balance of electrostatic
and hydrophobic interactions with lipid bilayers and biological
membranes. Biochemistry 1996, 35, 12612–12622.
JM701600A