Lysine N ε-trimethylation, a tool for improving the selectivity of antimicrobial peptides

Article abstract of DOI:10.1021/jm100261r

The effects of lysine N^?-trimethylation at selected positions of the antimicrobial cecropin A?melittin hybrid peptide KWKLFKKIGAVLKVL-amide have been studied. All five monotrimethylated, four bis-trimethylated plus the per-trimethylated analogues have been synthesized and tested for antimicrobial activity on Leishmania parasites and on Gram-positive and -negative bacteria, as well as for hemolysis of sheep erythrocytes as a measure of cytotoxicity. The impact of trimethylation on the solution conformation of selected analogues has been evaluated by NMR, which indicates a slight decrease in the ?-helical content of the modified peptides, particularly in the N-terminal region. Trimethylation also enhances the proteolytic stability of mono- and bis-trimethylated analogues by 2?3-fold. Although it tends to lower antimicrobial activity in absolute terms, trimethylation causes an even higher decrease in hemolytic activity and therefore results in improved selectivity for several analogues. The monotrimethylated analogue at position 6 shows the overall best selectivity against both the Leishmania donovani protozoan and Acinetobacter baumannii, a Gram-negative bacterium of increasing clinical concern.

Full text of DOI:10.1021/jm100261r

J. Med. Chem. 2010, 53, 5587–5596 5587
DOI: 10.1021/jm100261r
Lysine Nε-Trimethylation, a Tool for Improving the Selectivity of Antimicrobial Peptides
†
Marıa Fernandez-Reyes, Dolores Dıaz, Beatriz G. de la Torre, Ania Cabrales-Rico, Mariona Valles-Miret,
†
‡
‡,§
‡,
ꢀ
ꢁ
Jesus Jimenez-Barbero, David Andreu,*,‡ and Luis Rivas*,†
†
ꢀ
ꢀ
†Centro de Investigaciones Bioloꢀgicas CSIC, Ramiro de Maeztu 9, 28040 Madrid, Spain, and ‡Departament de Cieꢁncies Experimentals i de la
§
Salut, Universitat Pompeu Fabra, Dr. Aiguader 88, 08003 Barcelona, Spain. Present address: Centro de Ingenierıa Genetica y Biotecnologıa,
La Habana, Cuba. Present address: School of Chemistry, University of Edinburgh, Edinburgh EH9 3JJ, U.K.
´
ꢀ
´
Received March 1, 2010
The effects of lysine Nε-trimethylation at selected positions of the antimicrobial cecropin A-melittin
hybrid peptide KWKLFKKIGAVLKVL-amide have been studied. All five monotrimethylated, four
bis-trimethylated plus the per-trimethylated analogues have been synthesized and tested for antimicro-
bial activity on Leishmania parasites and on Gram-positive and -negative bacteria, as well as for
hemolysis of sheep erythrocytes as a measure of cytotoxicity. The impact of trimethylation on the
solution conformation of selected analogues has been evaluated by NMR, which indicates a slight
decrease in the R-helical content of the modified peptides, particularly in the N-terminal region.
Trimethylation also enhances the proteolytic stability of mono- and bis-trimethylated analogues by
2-3-fold. Although it tends to lower antimicrobial activity in absolute terms, trimethylation causes an
even higher decrease in hemolytic activity and therefore results in improved selectivity for several
analogues. The monotrimethylated analogue at position 6 shows the overall best selectivity against
both the Leishmania donovani protozoan and Acinetobacter baumannii, a Gram-negative bacterium of
increasing clinical concern.
Introduction
insertion into the pathogen membrane, leading to its disrup-
tion or to AMP translocation and eventual damage to intra-
cellular structures.4 In either case, development of resistance is
less likely than for other antimicrobials, involving as it would
major changes in phospholipid composition and architecture,
in turn requiring substantial overhauling of membrane-based
enzymatic and transport systems.5
While AMPs have raised expectations as alternatives to
conventional antibiotics in the face of a worldwide increase in
pathogen multiresistance6 and of a relatively exhausted pipe-
line of antimicrobial leads, they still have to gain a foothold in
therapeutics, largely due to limitations intrinsic to their peptide
nature such as short in vivo half-lives, or side-effects unrelated
to antimicrobial activity,7 all of them requiring careful opti-
mization before clinical assays can be implemented.1,8
Residue-specific modification of bioactive peptides is one of
the most effective and broadly used strategies for peptide lead
optimization.9 For a significant number of small and even
medium-sized bioactive peptides, including quite a few AMPs,
both the size and the scarcity of complex posttranslational
modifications (i.e., disulfide bridges) underpin the synthesis-
based strategies and have thus resulted in a wealth of analogues
incorporating modifications such as non-natural residues, site-
specific epimerization, acylation, glycosylation, or mobility res-
triction (e.g., by internal disulfide, helix stapling, etc.) aimed at
preserving privileged conformational features10-12
The ε-amino group of Lys has been frequently used for
conjugation to other biomolecules (fatty acids, nucleic acids)
or drugs, or for cross-linking to other proteins or peptides.
It has also recently received attention for its in vivo role in a
vast variety of modifications, e.g., methylation, acetylation,
ubiquitination, sumoylation, or citrullination (reviewed in ref 13),
Eukaryotic antimicrobial peptides (AMPsa), either present
orinducedina host after initialcontact with a pathogen,1 pose
a first chemical barrier against infection.2 Pathogen recogni-
tion by AMPs involves molecular traits common to a wide
variety of microorganisms. For instance, anionic phospho-
lipids exposed at the outer leaflet of the plasma membrane
underscore AMP specificity for prokaryotes and lower eu-
karyotes, because in higher eukaryotes, these phospholipids
are confined to the cytoplasmic face of the membrane.3 The
strong cationic nature of most AMPs, plus their tendency to
adopt amphipathic conformations, favors accumulation and
*To whom correspondence should be addressed. For D.A.: phone,
þ34-933160868; fax, þ34-933160901; e-mail, david.andreu@upf.edu.
For L.R.: phone, þ34-918373112; fax, þ34-915360432; e-mail, luis.rivas@
cib.csic.es.
aAbbreviations: AMP, antimicrobial peptide; ATCC, American Type
Culture Collection; Boc, tert-butoxycarbonyl; CA, cecropin A; CA-M,
cecropin A-melittin hybrid peptide; CECT, Spanish Type Culture Collec-
tion; COSY, correlated spectroscopy; DIEA, N,N-diisopropylethyla-
mine; DMF, N,N-dimethylformamide; Fmoc, 9-fluorenylmethyloxy-
carbonyl; HBTU, 2-(1H-benzotriazole-1-yl)-1,1,3,3-tetramethyluronium
hexafluorophosphate; HC50, peptide concentration causing 50% hemo-
globin release; HOBt, N-hydroxybenzotriazole; HPLC, highperformance
liquid chromatography; IC50, peptide concentration causing 50% inhibi-
tion of MTT reduction at 4 h incubation; LC-MS, liquid chromatography
coupled to mass spectrometry; LC50, peptide concentration causing 50%
inhibition of Leishmania proliferation; M, melittin; MALDI-TOF, ma-
trix-assisted laser desorption ionization, time-of-flight analysis; MIC50
,
peptide concentration causing 50% inhibition of bacterial growth; MS,
mass spectrometry; MTT, 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetra-
zolium bromide; NOESY, nuclear Overhauser enhancement spectrosco-
py; rmsd, root-mean-square deviation; SI, selectivityindex (=HC50/LC50
or HC50/MIC50); TFA, trifluoroacetic acid; TOCSY, total correlated
spectroscopy; TSP, 2,2,3,3-tetradeutero-3-trimethylsilylpropionic acid.
r
2010 American Chemical Society
Published on Web 07/09/2010
pubs.acs.org/jmc

5588 Journal of Medicinal Chemistry, 2010, Vol. 53, No. 15
Fernaꢀndez-Reyes et al.
Table 1. Structures and Chemical Characterization of CA(1-7)M(2-9) and Trimethyl-Lys Analogues
mass
peptide
sequence/replacement
sequencea
tR (min)b
expected
foundc
1
CA(1-7)M(2-9)
K1 (Me3)
K3 (Me3)
KWKLFKKIGAVLKVL
KWKLFKKIGAVLKVL
KWKLFKKIGAVLKVL
KWKLFKKIGAVLKVL
KWKLFKKIGAVLKVL
KWKLFKKIGAVLKVL
KWKLFKKIGAVLKVL
KWKLFKKIGAVLKVL
KWKLFKKIGAVLKVL
KWKLFKKIGAVLKVL
KWKLFKKIGAVLKVL
9.53
9.05
9.11
9.10
9.24
9.22
9.12
9.03
9.14
9.19
9.01
1769.18
1812.24
1812.24
1812.24
1812.24
1812.24
1855.29
1855.29
1855.29
1855.29
1984.45
1770
1813
1813
1813
1813
1813
1856
1856
1856
1856
1985
2
3
4
K6 (Me3)
K7 (Me3)
K13 (Me3)
5
6
7
K
K
1,3 (Me3)2
3,6 (Me3)2
8
9
K6,7 (Me3)2
K1,13 (Me3)2
10
11
K
1,3,6,7,13 (Me3)5
a All peptides in C-terminal carboxamide form. Lysine (K) residues in bold are trimethylated. b Linear 5-95% B into A gradient over 20 min.
A = 0.1% formic acid in water; B = 0.08% formic acid in acetonitrile. For all peptides, HPLC purity was >95%. c Determined by LC-MS.
Table 2. Leishmanicidal and Hemolytic Activities of CA(1-7)M(2-9) and Its Trimethylated Derivatives
L. donovani promastigotes
L. pifanoi amastigotes
erythrocytes
peptide
sequence/replacement
IC50 (μM)a
LC50 (μM)a
SIb
IC50 (μM)a
LC50 (μM)a
SIb
HC50 (μM)a
1
CA(1-7)M(2-9)
K1 (Me3)
K3 (Me3)
1.8 ((0.0)
3.9 ((0.3)
3.3 ((0.0)
3.7 ((0.0)
3.7 ((0.2)
3.5 ((0.1)
12.8 ((0.4)
5.1 ((0.2)
6.1 ((0.1)
8.8 ((0.3)
>50
1.5 ((0.1)
2.8 ((0.0)
3.1 ((0.6)
1.7 ((0.0)
2.3 ((0.0)
2.5 ((0.0)
15.5 ((0.6)
6.7 ((0.2)
5.3 ((0.6)
7.6 ((0.1)
>50
26.8 (1.0)
27.6 (1.0)
27.6 (1.0)
92.7 (3.4)
71.3 (2.6)
63.0 (2.3)
6.7 (0.2)
17.4 (0.6)
25.3 (0.9)
10.8 (0.4)
nd
4.3 ((0.2)
6.0 ((0.8)
9.4 ((0.8)
13.4 ((0.1)
11.4 ((0.8)
10.1 ((0.6)
16.2 ((0.1)
7.4 ((0.3)
9.2 ((2.0)
17 0.0((3.0)
>50
4.5 ((0.2)
6.4 ((1.2)
8.2 ((1.0)
16.3 ((1.30)
12.1 ((1.2)
11.6 ((0.5)
24.8 ((1.2)
16.5 ((1.6)
27.0 ((1.2)
13.5 ((1.4)
>50
8.9 (1)
40.2 ((2.2)
77.2 ((11.1)
85.6 ((4.6)
157.6 ((7.3)
164.0 ((7.8)
157.4 ((1.0)
105.0 ((3.5)
117.3 ((13.9)
134.0 ((8.8)
82.6 ((12.4)
>200
2
12.0 (1.3)
10.4 (1.2)
9.6 (1.1)
13.5 (1.5)
13.5 (1.5)
4.2 (0.5)
7.1 (0.8)
4.9 (0.5)
6.1 (0.7)
nd
3
4
K6 (Me3)
K7 (Me3)
5
6
K
13 (Me3)
K1,3 (Me3)2
3,6 (Me3)2
7
8
K
9
K6,7 (Me3)2
K1,13 (Me3)2
K1,3,6,7,13 (Me3)5
10
11
a In parentheses, standard deviation. b Selectivity index (SI) as defined in text; in parentheses, improvement in SI relative to parent peptide 1.
of the histone proteins that regulate the transcriptional
status of chromatin.14-17 This interest in Lys side chain
modification, particularly methylation, by the epigenetics
community contrasts with the relative lack of references to
Lys methylation in the AMP field, excepting (1) histone-
related sequences with reported antimicrobial activities,18-24
(2) a methylated analogue of the decapeptide KSL used in
chewing gum,25 (3) lactoferricin,26 or (4) melittin, which en-
tirely loses its hemolytic activity upon trimethylation of all
its Lys residues.27
In this work, we have further investigated the effect of Lys tri-
methylation on antimicrobial activity and hemolysis (as a mea-
sure of cytotoxicity) of AMPs. To this end we have chosen a well-
tested, simple AMP platform, namely the cecropin A-melittin
(CA-M) hybrid CA(1-7)M(2-9), KWKLFKKIGAVLKVL-
amide (reviewed in refs 28,29). In this peptide,30,31 the N-termi-
nal (first seven residues) cationic region of nonhemolytic cecro-
pin A is fused to the more hydrophobic, also nonhemolytic
N-terminus of melittin, the resulting hybrid structure showing an
improvement in both antimicrobial and cytotoxic activities
relative to the parent structures. CA(1-7)M(2-9) and some
congeners have shown activity on a variety of microorganisms,
including bacteria,32-34 fungi,35,36 or protozoa,31,37-39 as well as
in animal33,40,41 and plant disease models.36,42
CA-M hybrids as potential antimicrobial candidates against
multiresistant pathogens.
Results
Trimethylated Derivatives of CA(1-7)M(2-9). Structures
of the parent peptide (1), all five singly trimethylated (2-6),
plus four doubly trimethylated (7-10) and the fully trimethy-
lated (11) analogues are shown in Table 1. All peptides were
made by Fmoc solid phase synthesis methods,43 using com-
mercially available Fmoc-Lys(Me3)-OH at the designated
positions. After full deprotection and cleavage from the resin,
peptides were purified to near-homogeneity (>95%) by pre-
parative RP-HPLC and characterized by electrospray MS,
as they showed poor desorption by MALDI-TOF. In selec-
ted cases (see below), additional characterization by NMR
was done.
Activity against Leishmania and Erythrocytes. Leishmania
was chosen as the model target microorganism for measuring
antimicrobial activity due to its lack of tightly packed external
barriers such as the outer membrane or peptidoglycan layer of
Gram-negative and Gram-positive bacteria, respectively. For
measuring hemolysis, sheep erythrocytes were selected. Activities
of peptides 1-11 on the two forms (promastigote, amastigote) of
the parasite and on erythrocytes are shown in Table 2.
Predictable effects of Lys trimethylation include (1) retain-
ing the same global charge, (2) increased bulkiness and
hydrophobicity of the side chains, and (3) ε-amino groups
(quaternary) with no hydrogen bonding ability. Our explora-
tion of these aspects on the CA(1-7)M(2-9) model under-
pins Lys trimethylation as a useful tool for modulating
antimicrobial activity and half-life, and restates the case for
A general trend observed was the gradual decrease in both
antimicrobial and cytotoxic activity as the number of Lys-
(Me3) residues in the sequence increased. The nontrimethy-
lated lead 1, CA(1-7)M(2-9), was most active, while full
trimethylation (analogue 11) caused complete inhibition of
all activities within the range of concentrations tested. With-
in this general trend, subtle variations could be observed.

Article
Journal of Medicinal Chemistry, 2010, Vol. 53, No. 15 5589
Figure 1. Morphological damage and distribution of trimethylated Lys peptides on Leishmania. Upper panels: Electron microscopy of L.
donovani promastigotes ((A) control parasites; (B) treated with 4 μM CA(1-7)M(2-9) (1); (C) treated with 6 μM K7(Me3) analogue (5)
showing depletion of intracitoplasmic electron-dense material (bar = 1 μm). Lower panels: Cytofluorometric assay of the incorporation of
Sytox Green under the same conditions assayed for electron microscopy. Percentage of cells incorporating the vital dye (peaks on the right) is
indicated inside each panel.
For instance, for monosubstituted analogues tested on
Leishmania promastigotes, antiparasitic activity was rela-
tively independent of the location of the Lys(Me3) resi-
due, whereas in the disubstituted series, analogue 7, with
Lys(Me3) replacements at the positions closest to the
N-terminus, underwent the most significant activity loss.
In amastigotes, although a similar trend was observed, dif-
ferences between analogues were less clear. In the hemolysis
assays, the drop in activity was particularly acute for three
monosubstituted analogues (4-6) yet not so pronounced for
the four doubly trimethylated peptides (7-10) tested.
Contrasting the cytotoxic and antiparasitic activities of
each peptide by means of the selectivity index, (SI, defined as
HC50/LC50, Table 2), it is clear that, for some analogues and
particularly against the promastigote forms of Leishmania,
the Lys(Me3) replacement entails a substantial improvement
in SI. Thus, for analogues 4-6, 2-3-fold SI enhancements
were found relative to parent peptide 1, with antimicrobial
activities still well into the low-micromolar range. For
amastigotes, two of the above peptides (5 and 6) also
displayed enhanced SIs, but to a lower extent. On the other
hand, all of the four double Lys(Me3) replacements investi-
gated (analogues 7-10) resulted in lower SIs.
Electron Microscopy. Electron microscopy provides useful
insights into the morphological effects of peptides on mi-
crobes, and those may in turn provide useful clues on the
lethal mechanism. The damage inflicted by the peptides to
Leishmania promastigotes is illustrated in Figure 1. To avoid
any bias due to the differential killing ability of the peptides,
they were assayed at equipotent concentrations, i.e., equiva-
lent to those inducing a 70% inhibition of MTT reduction. In
all cases, and regardless of the peptide assayed, extensive
permeabilization, with translucent cytoplasm and extensive
damage to the plasma membrane, was observed. This pattern
was followed by most, though not all, parasites in the same
field, with some displaying an apparently undamaged plas-
ma membrane and no evidence of leakage of cytoplasmic
material.
Investigation of Lethal Mechanisms. In view of the electron
microscopy data, and in accordance with previous studies on
the mechanism of action of other CA-M peptides,39,44 we
explored the membrane permeation activity of Lys(Me3)-
substituted peptides on Leishmania promastigotes by two
complementary methods: (1) enhanced fluorescence of
the vital dye Sytox Green (MW = 600 Da), which binds
exclusively the nucleic acids of permeabilized cells, and (2)
membrane depolarization. As seen in Figure 2, all active
peptides were able to induce the entrance of the dye into
parasites in a dose-dependent manner, as illustrated for
peptides 1, 5, and 10 on both promastigotes and amastigotes.
For all of them, a rapid increase in Sytox Green fluorescence
followed peptide addition, reaching a maximal value in less
than 10 min.
An additional conclusion to be drawn from Table 2 is that
the IC50/LC50 ratio for most peptides on promastigotes is
close to 1, which means that the peptide effect after the first 4
h incubation is basically irreversible. In contrast, for three of
the doubly substituted analogues tested (7-9), the short-
term inhibition (IC50) of the peptides on amastigotes is more
pronounced than the inhibition of proliferation (LC50),
suggesting that the amastigote has a substantial ability to
recover from the initial blow.
Plasma membrane depolarization experiments furnished
comparable results. An increase in fluorescence, irreversible

5590 Journal of Medicinal Chemistry, 2010, Vol. 53, No. 15
Fernaꢀndez-Reyes et al.
Figure 2. Permeation of L. donovani promastigote plasma membrane by peptides CA(1-7)M(2-9) (1), K7(Me3) (5), and K1,13(Me3)2 (10).
Upper panels: kinetics of Sytox Green entrance after peptide addition (t = 0), measured by the increase in fluorescence after binding of the dye
to intracellular nucleic acids (λexc = 504 nm; λems = 524 nm). Full permeation (100%) is achieved after addition of 0.1% Triton X100 (indicated
by arrow). Peptide concentrations (μM): 0.5 (b), 1.0 (O); 1.5 (1); 2.0 (4); 4.0 (9); 6.0 (0). Lower panels: Promastigote plasma membrane
depolarization measured by the increase in fluorescence of bisoxonol (λexc = 540 nm; λems = 580 nm). Full depolarization was defined as that
achieved by addition of 3 μM CA(1-8)M(1-18) (see ref 37).
Figure 3. Real-time depletion of intracellular ATP levels upon peptide addition in living L. donovani 3-Luc promastigotes. See ref 39 for
details. Peptide nomenclature and concentration as in Figure 2. Peptide added at t = 0 (luminescence = 100%). Luminiscence was corrected
relative to control parasites.
over the entire observation period, was found (Figure 2,
lower panels). Depolarization has the advantage over vital
dye uptake experiments of its dynamic character, i.e.,
changes in bisoxonol fluorescence reflect dye insertion into
the parasite plasma membrane, driven by the collapse of
membrane potential, and hence are potentially reversible if
the parasite recovers from its membrane lesions. In contrast,
Sytox Green uptake relies on the high-affinity interaction of
the dye with intracellular nucleic acids, a practically irrever-
sible process impervious to any membrane-repairing effort
after peptide insertion. In this particular case, however, both
approaches yielded comparable results, indicating that the
peptides caused irreversible damage to the membrane.
To account for the electron microscopy observation of
apparently unharmed organisms at peptide concentrations
that diminish parasite proliferation, the entrance of Sytox
Green in individual cells was also assessed by cytofluorome-
try (Figure 1, lower panels). Two distinct cell populations
were readily identified, a minor one, essentially unlabeled,
and a major one (>70%), incorporating the dye, a situation
suggesting an all-or-none effect of the peptides.
To illustrate the rapid physiological impairment of the
parasite caused by the peptides, variations in the ATP pool
were monitored. To this end, 3-Luc strain promastigotes
expressing a cytoplasmic form of luciferase39 were used in
conjunction with caged-free permeable DMNPE-luciferin, a
setting with ATP as the limiting factor for luminescence. As
shown in Figure 3, all three analogues tested (1, 5, 10) caused
a fast, dose-dependent drop in luminescence upon addition
to the parasite suspension, confirming fast bioenergetic
collapse upon peptide addition.
Bactericidal Activity of Trimethylated Analogues. Given
the success of Lys trimethylation in improving the SI of
CA(1-7)M(2-9) on Leishmania, it seemed appropriate to
test also the activity of Lys(Me3)-substituted peptides on
bacteria. To this end, a representative Gram positive
(Staphylococcus aureus) and an opportunistic Gram negative
(Acinetobacter baumannii) were chosen as targets. Results are

Article
Journal of Medicinal Chemistry, 2010, Vol. 53, No. 15 5591
Table 3. Bactericidal and Hemolytic Activities of CA(1-7)M(2-9) and Its Trimethylated Derivatives
S. aureus
A. baumannii
MIC50 (μM)a
erythrocytes
peptide
sequence/replacement
MIC50 (μM)a
SIb
SIb
HC50 (μM)a
1
CA(1-7)M(2-9)
K1 (Me3)
K3 (Me3)
2.9 ((0.1)
3.4 ((0.1)
2.4 ((0.1)
4.6 ((0.2)
7.2 ((0.6)
4.4 ((0.2)
8.0 ((0.0)
6.2 ((0.5)
10.7((2.9)
4.5 ((0.3)
>30
13.8 (1.0)
22.7 (1.6)
35.7 (2.6)
34.3 (2.5)
22.8 (1.6)
35.8 (2.6)
13.1 (0.9)
18.9 (1.4)
12.5 (0.9)
18.4 (1.3)
nd
1.1 ((0.3)
1.3 ((0.1)
1.7 ((0.3)
1.4 ((0.1)
1.9 ((0.4)
2.1 ((0.1)
3.5 ((0.5)
2.0 ((0.1)
2.6 ((0.1)
1.8 ((0.5)
>30
36.5 (1.0)
59.4 (1.7)
50.3 (1.4)
112.2 (3.0)
86.3 (2.3)
74.9 (2.0)
30.0 (0.8)
58.6 (1.6)
51.5 (1.4)
45.8 (1.25)
nd
40.2 ((2.2)
77.2 ((11.1)
85.6 ((4.6)
157.6 ((7.3)
164.0 ((7.8)
157.4 ((1.0)
105.0 ((3.5)
117.3 ((13.9)
134.0 ((8.8)
82.6 ((12.4)
>200
2
3
4
K6 (Me3)
K7 (Me3)
K13 (Me3)
5
6
7
K
1,3 (Me3)2
3,6 (Me3)2
8
K
9
K6,7 (Me3)2
K1,13 (Me3)2
10
11
K1,3,6,7,13 (Me3)5
a In parentheses, standard deviation. b Selectivity index (SI) as defined in text; in parentheses, improvement in SI relative to parent peptide 1.
shown in Table 3. For S. aureus, the activity tended to drop
slightly as the level of Lys(Me3) substitution increased, similar
to Leishmania, although in this case the trend was less uniform
(compare, e.g., disubstituted analogue 10 with monosubsti-
tuted 4 or 6). For A. baumannii, a comparable behavior was
seen, although the range of variation was less pronounced than
for the Gram-negative organism. As found in Leishmania,
when antibacterial and hemolytic activities were contrasted,
2-3-fold enhancements in SI were found for some analogues,
particularly 4 (against either bacteria, also for Leishmania
promastigotes), as well as 3 and 6 (for S. aureus).
Proteolytic Stability of Trimethylated Analogues. As tri-
methylation protects the Lys residue from proteolysis by
trypsin-like enzymes, it was expected that the Lys(Me3)-
substituted analogues would be somewhat more resistant to
proteolysis than parent peptide CA(1-7)M(2-9). The time-
course of trypsin digests of mono-, disubstituted, and fully
substituted analogues of 1 was monitored by LC-MS. As
expected, the pertrimethylated analogue 11 was quite imper-
vious to proteolysis, with no appreciable decay throughout
the full period of incubation (60 min). Conversely, the parent
peptide 1 was 64% degraded after 1 min digestion, with the
Lys6-Lys7 peptide bond shown as the most vulnerable (De la
Torre, manuscript in preparation). Among the monotri-
methylated analogues, 4 was particularly resistant (29%
degraded at 1 min, 2.2-fold higher proteolytic stability than 1),
arguably because the highly labile Lys6-Lys7 bond was pro-
tected by trimethylation at Lys6. Other monotrimethylated
analogues (e.g., 5, 6) were about 50% degraded after 1 min.
On the other hand, analogues 8, 9, and 10, representative
of bis-trimethylation, were degraded 20%, 25%, and 19%,
respectively, after 1 min. At the 5 min time point, all peptides
(except 11) were largely or totally degraded.
stable conformation. In the presence of 30% (v/v) TFE, however,
both chemical shifts and NOE connectivities clearly revealed
that stable secondary structures had been induced. First,
upfield shifts in the R-protons, characteristic of R-helical
structure, were observed for all three peptides. Comparison
with the random coil values in the chemical shift index
(CSI)47 (Figure SF1, Supporting Information) showed for
most residues deviations strongly indicative of predomi-
nantly helical conformation, with the CA less structured
than the M segment. The pattern of chemical shift deviations
suggested similar conformations for all peptides. In Figure S2
(Supporting Information), a comparison between the RCH
chemical shifts of analogues 1, 5, and 11 and equivalent resi-
dues in the reported NMR data for CA48 and M49 suggests
relatively less stable helix patterns for all three hybrids, either
trimethylated (5, 11) or not (1).
Analysis of NOE cross-peak patterns revealed similarities
among 1 and its two analogues, such as almost continuous
stretches of dNN(i, iþ1) and dRHN(i, iþ3) connectivities. A series
of strong, helix-specific dRHN(i,iþ4)-type NOE crosspeaks
were also detected for all three peptides. CYANA structure
calculations50 using interproton distance restraints deter-
mined from the NOE intensities, plus nonambiguous cou-
pling constants (3JNHR) as angular restraints, allowed
proposal of 3D structures for all three peptides. In each case,
the 20 best low-energy conformers, all with residues located
only in the “most favored” and “allowed” regions of the
Ramachandran plot, were chosen for analysis, validated by
PROCHECK,51 and are shown as superimposed represen-
tative structures in Figure 4. Inspection of these and previous
data suggests that trimethylation induces minimal, nondras-
tic modifications in secondary structure. Though subtle,
these changes may nonetheless suffice to modulate peptide-
membrane interaction by increasing flexibility, particularly
in the CA region.
NMR Structural Analysis of Representative Analogues.
Further insights into the influence of Lys trimetylation on
the structure of CA-M peptides were obtained by NMR
analysis of mono- and fully trimethylated analogues 5 and 11,
respectively, in comparison with lead peptide 1. Resonance
assignments were derived from the COSY/TOCSY and NOESY
spectra according to the sequential assignment methodology45
and using the CARA software46 (see Tables ST1-ST3, Sup-
porting Information).
Discussion
The biological relevance of Lys methylation has received
increasing attention since the discovery of its role as epigenetic
modulator of the transcriptional state of chromatin, as well as
regulator of the activity of transcriptional factors.52-54 In
these instances, Lys methylation is subtly and tightly regu-
lated, i.e., the biological response appears to rely on the
creation of a new specific recognition motif, or on the
competition with other Lys modifications,55 but not strictly
on changes in the physicochemical parameters of the Lys side
chain. When Lys(Me3) is incorporated onto AMPs, selectivity
of action depends on the initial recognition between a cationic
For all three peptides, 1H NMR spectra in water showed
small chemical shift dispersion in the amide region, indicat-
ing a predominance of flexible, nonordered structures in fast
exchange on the chemical shift NMR time scale. Besides, the
two-dimensional NOESY spectra did not contain medium-
or long-range NOEs characteristic of any defined and

5592 Journal of Medicinal Chemistry, 2010, Vol. 53, No. 15
Fernaꢀndez-Reyes et al.
This is not unusual for AMPs of this type, as shown also in
Leishmania for magainin analogues,63 or in artificial mem-
branes for CA-M hybrids65,66 and their parent structures
melittin (M)60,67 and cecropin A (CA).68
Having thus established membrane permeation as the lethal
mechanism, it is interesting to discuss whether the process is
significantly or not affected by Lys trimethylation. Our
experimental data show that it lowers the biological activity
against all targets, either protozoan or bacterial or eukaryotic.
Activity diminishes upon increasing trimethylation and is
totally lost in the pertrimethylated analogue 11, a situa-
tion with parallels in melittin27 and lactoferricin.26 However,
this decreased antimicrobial activity is offset by an even
greater loss in activity against eukaryotic cells (measured as
hemolysis), which thus results in improved selectivity, parti-
cularly for singly trimethylated analogues (Tables 2 and 3).
For the two test bacteria (Table 3), as well as for protozoan
Leishmania promastigotes (though not for amastigotes,
Table 2), trimethylation at the more internal Lys residues
(e.g., analogues 4 and 5) is more effective than at the outer ones.
A straightforward interpretation of this general decay of
membrane activity with trimethylation is risky given the
structural complexity of external barriers such as the pepti-
doglycan of Gram-positive and the outer membrane of Gram-
negative bacteria, both hampering peptide access to the
primary plasma membrane target. Despite these caveats,
several factors, not mutually exclusive, can still be considered
to be at play: (1) Similar to other AMPs with cationic side
chains, the present analogues can use their Nε-trimethyl Lys
groups to establish electrostatic interactions with anionic
phospholipids, common to all the different types of cells tested.
(2) Lysine “snorkeling”,69 a term describing the orientation
of Lys ε-amino groups to achieve favorable energetics in a
membrane interphase, is also possible for Nε-trimethyl groups.
In contrast, (3) hydrogen bonds between Lys ε-amino groups
and phospholipid phosphate groups will not exist in the
trimethylated derivatives. The relative weight of each of these
factors requires further investigations that are outside the
scope of the present work.
Another structural property that may be affected by tri-
methylation is hydrophobicity. In the present case, however,
the hypothesis is not quite borne out by experimental data. As
seen in Table 1, all analogues, including pertrimethylated 11,
elute slightly faster in RP-HPLC than lead peptide 1.
Although the differences in retention times are modest, they
may somehow be related with the general trend of reduced
hemolytic potency observed. The fact that erythrocytes are
zwitterionic in their outward interface suggests that electro-
static interactions may play a lesser role than hydrophobicity
in AMP insertion. If so, the lower hydrophobicity of ana-
logues 2-11 could translate in poorer insertion (vs parent
peptide 1) and thus account for the differences in SI.
Finally, another area where Lys trimethylation has a pre-
sumable impact is peptide conformation. Adoption of an
R-helix upon contact with membranes or membrane-like
systems is regarded as a key step in peptide-membrane
interaction, in contrast with the often unstructured conforma-
tion of peptides in water.30,70,71 At moderate peptide-lipid
ratios, lead peptide 1 localizes parallel to the bilayer at the
membrane-water interface, with Lys residues exposed and
interacting with phospholipid head groups and hydrophobic
residues buried into the hydrophobic matrix.66,70,71 Replacing
Lys by considerably bulkier Lys(Me3) side chains is bound to
have conformational consequences, at least in the vicinity of
Figure 4. Superimposition of the 20 lowest energy conformers of
peptides CA(1-7)M(2-9) (1), K7(Me3) (5) and K1,3,6,7,13(Me3)5 (11).
Structures are superimposed over the backbone N, CR, and C0 atoms
of residues 1 through 15.
peptide and the anionic external leaflet of the pathogen
membrane, a scenario with much looser specificity require-
ments.
Prior to the crucial peptide-membrane interaction, the
amount of available peptide is controlled by the pathogen’s
own proteolytic enzymes and/or by peptidases present in the
biological fluids that substantially impair the efficacy of the
peptide.56-58 In line with observations on other peptides,25 in
the present case the introduction of one or two Lys(Me3)
residues resulted in 2-3-fold increases in proteolytic resistance,
particularly for replacements at Lys6, the most protease-sensi-
tive position in lead peptide 1. While this enhanced protease
stability cannot be unequivocally used to decode the anti-
microbial activity data (Tables 2 and 3), it contributes to the
overall performance of the analogues and may be of further
relevance in vivo, where the peptide is in prolonged contact
with potent degradation systems such as, e.g., leishmaniolysin,
a metalloprotease involved in Leishmania resistance against
AMPs,59 or other proteases from biological fluids.
Assuming protease resistance to be a coadjuvant though
not a decisive factor for the increased SIs of most trimethy-
lated analogues, a more appropriate explanation should look
into how the structural changes brought about by trimethyla-
tion influence interaction of AMPs with their final target.
For CA-M peptides, ample evidence points to permeation
of the plasma membrane as the key step in the lethal acti-
vity.37,44,60,61 Here, using Leishmania as a model microorgan-
ism, membrane permeation has been conclusively linked to
peptide killing activity on several counts: (1) trimethylated
analogues permeabilize the parasite plasma membrane as
assessed by the fast Sytox Green uptake after peptide addi-
tion, (2) permeation is also consistent with the fast and severe
depletion of intracellular ATP observed upon peptide addi-
tion and inducing a lethal bioenergetic collapse, (3) early
damage (i.e., at 4 h) to the parasite is in most cases irreversible,
as shown by the IC50/LC50 ratios close to 1 for most analo-
gues, ruling out a time-dependent mechanism, (4) electron
microscopy, particularly with promastigotes, showed plasma
membrane permeation and loss of cytoplasmic material after
4 h incubation with peptide, a typical scenario of membrane-
active peptides.37,39,62-64 Membrane permeation appears in
this case to take place in an all-or-none fashion, rather than by
steadily increasing damage to the entire parasite population.

Article
Journal of Medicinal Chemistry, 2010, Vol. 53, No. 15 5593
the replaced residue. While NMR results for analogues 5
and 11 indicate a predominantly helical geometry, they also
suggest slightly different levels of flexibility with respect to 1.
This subtle helix-destabilizing action of the Lys(Me3) residue
affects mostly the CA moiety of the peptide and can be
plausibly correlated with the observed biological activities
by assuming that a slightly more flexible N-terminal (CA)
segment is more detrimental for the hemolytic (i.e., increasing
HC50) than for the microbicidal activityof the singly trimethy-
lated analogue 5. According to this view, hemolysis and
antimicrobial activity would be related to the structural/
dynamic features of the CA and M regions, respectively.
The hypothesis would be further supported by the fact that
destabilization of both regions, as in analogue 11, impairs the
interaction with both eukaryotic and bacterial membranes
and renders the peptide totally ineffective. In summary, NMR
resultssuggestthatLys trimethylation entails a subtle, gradual
(with the number of Lys(Me3) residues) loss in the helical
content and rigidity of the peptide, which may explain, within
the current consensus theory of AMP-membrane interac-
tion, the slight decrease in potency of monotrimethylated
analogues 2-6, the more substantial reduction observed in
the four doubly trimethylated analogues 7-10 and, especially,
the total loss of activityin the pertrimethylated analogue 11. A
related example of helix distortion bringing about a decrease
in hemolytic activity and thus an improved SI is provided
by the diasteromeric analogues of R-helical AMPs described
by Shai.72 From a more practical point of view, one can also
conclude that AMP Nε-trimethylation offers the possibility of
improved activity on bacteria of serious clinical concern such
as S. aureus and A. baumannii.73
After chain assembly, full deprotection and cleavage was carried
out with TFA-water-triisopropylsilane (95:2.5:2.5 v/v, 90
min, rt). Peptides were isolated by precipitation with cold diethyl
ether and separated by centrifugation, dissolved in 0.1 M acetic
acid and lyophilized. LC-MS was performed in a LC-MS
2010EV instrument (Shimadzu) fitted with a Jupiter C18 col-
umn (4.6 mm ꢀ 250 mm, 5 μm; Phenomenex) eluted with a
5-95% linear gradient of B into A (A = 0.1% formic acid in
water; B = 0.08% formic acid in acetonitrile) over 20 min (see
Table 1) at a flow rate of 1 mL/min, with UV detection at 220 nm
Preparative HPLC runs were performed on a Luna C18 column
(21.2 mm ꢀ 250 mm, 10 μm; Phenomenex), using linear gradi-
ents of solvent B (0.1% in ACN) into A (0.1% TFA in water), as
required, with a flow rate of 25 mL/min. Fractions of high
(>95%) HPLC homogeneity and with the expected mass were
combined, lyophilized, and used in subsequent experiments.
For proteolytic stability determination, peptides (1 mg/mL in
50 mM NH4HCO3) were incubated with trypsin (Promega) at
37 °C in a 1:100 enzyme-peptide ratio. Incubation was termi-
nated by acetic acid addition. The remaining amount of origi-
nal peptide was determined by HPLC peak integration (elution
conditions as above).
Peptide-Membrane Interaction Studies. Unless otherwise
stated, reagents were purchased from Sigma (Madrid, Spain)
and Merck (Darmstadt, Germany). Bisoxonol [bis-(1,3-diethyl-
thiobarbituric) trimethine oxonol], Sytox Green, and DMNPE-
luciferin [1-(4,5-dimethoxy-2-nitrophenyl)-D-luciferin ethyl ester]
were from Invitrogen (Barcelona, Spain). Fluorescence and
luminescence measurements were recorded in a Polarstar Galaxy
microplate reader (BMG Labotechnologies, Germany), fit-
ted with the corresponding optical setting, unless otherwise
stated.
Determination of Antimicrobial Activities. Staphylococcus
aureus strain CECT 240 (kindly provided by the Microbiology
Department, University of Granada, Spain) was used to test the
bactericidal effect on Gram-positive bacteria. The activity on
Gram-negatives was assayed on the opportunistic bacteria
Acinetobacter baumannii (ATCC 19606). Bacteria were main-
tained as frozen stocks in order to minimize variations during
continuous liquid growth, thawed before the experiment, and
grown in Mueller-Hinton II broth (MHB), cation-adjusted
(Becton-Dickinson, Cockeysville, MD) at 37 °C. Bactericidal
activities were assayed in solution, as described.74 Briefly,
bacteria were collected in exponential growth phase, washed
twice with phosphate buffered saline, and resuspended at 1 ꢀ
106 CFU/mL in fresh MHB. Aliquots of this suspension were
transferred into 96-microwell plates (50 μL/well), and the same
volume of a peptide solution at 2 ꢀ final concentration in MHB
was added to each well. Bacteria were allowed to proliferate
overnight at 37 °C, and growth was assessed by turbidimetry at
600 nm in a model 680 microplate reader (Bio-Rad Labora-
tories, Hercules. CA). Measurements were made by triplicate,
and experiments were repeated at least twice.
To assess leishmanicidal activity, protocols previously devel-
oped in our laboratory were used.39,44 Briefly, the L. donovani
promastigote strain MHOM/SD/00/1S-2D, or the L. pifanoi
axenic amastigote strain (MHOM/VE/60/Ltrod) were grown in
their respective growth media at 26 or 32 °C, respectively.
Parasites were harvested at the late exponential growth phase,
washed twice in Hank’s (136 mM NaCl, 4.2 mM Na2HPO4,
4.4 mM KH2PO4, 5.4 mM KCl, 4.1 mM NaHCO3, pH 7.2),
supplemented with 20 mM D-glucose (Hank’s þ Glc), resus-
pended in the same medium (2 ꢀ 107 parasites/mL), aliquoted
into a 96-microwell plate (120 μL/well final volume), and then
treated with peptide. Standard incubation conditions were 4 h at
26 or 32 °C for promastigotes and amastigotes, respectively.
After peptide incubation, a 20 μL aliquot was transferred
from each well into its respective complete medium and incu-
bated for 72 h at 26 °C or 96 h at 32 °C to allow proliferation
of the surviving promastigotes or amastigotes, respectively.
Conclusions
This work serves as a proof-of-principle of the versatility
and potentiality of Lys trimethylation as a strategy for fine-
tuning AMP biological activity. Trimethylation can hopefully
expand the existing armamentarium of subtle modifications
aimed at improving the SI of these peptides by either boosting
their antimicrobial activity or, as in the present case, dimin-
ishing their toxicity. Furthermore, judicious trimethylation of
Lys ε-amino groups decreases susceptibility to proteases in
biological fluids and/or target pathogens, which results in
improved half-lives and thus helps to overcome one of the
major hurdles for AMP druggability.
Experimental Section
Peptide Synthesis and Characterization. Fmoc-protected ami-
no acids were obtained from Senn Chemicals (Dielsdorf, Switzer-
land). Fmoc-Lys(Me3)-OH was from Iris Biotech (Marktredwitz,
Germany). Fmoc-Rink-amide (MBHA) resin was from Novabio-
€
chem (Laufelfingen, Switzerland), and 2-(1H-benzotriazol-1-yl)-
1,1,3,3-tetramethyluronium hexafluorophosphate (HBTU) and
N-hydroxybenzotriazole (HOBt) from Matrix Innovation (Mon-
treal, Canada). HPLC-grade acetonitrile (ACN), and peptide
synthesis-grade N,N-dimethylformamide (DMF), N,N-diisopro-
pylethylamine (DIEA), and trifluoroacetic acid (TFA) were from
SDS (Peypin, France).
Solid phase peptide synthesis was done by Fmoc-based
chemistry on Fmoc-Rink-amide (MBHA) resin (0.1 mmol) in
a model 433 synthesizer (Applied Biosystems, Foster City, CA)
running FastMoc protocols. Couplings used 8-fold molar excess
each of Fmoc-amino acid, HBTU and HOBt, and 16-fold molar
excess of DIEA. Side chains of Lys and Trp residues were
protected with the Boc group. Lys(Me3) was incorporated at
the specified positions using the same synthetic file than for Lys.

5594 Journal of Medicinal Chemistry, 2010, Vol. 53, No. 15
Fernaꢀndez-Reyes et al.
Growth was assessed by incubation of parasites with 0.5 mg/mL
of MTT for 2 h at their respective temperature, and final solubiliza-
tion of the resulting formazan with 5% SDS, which was quanti-
tated spectrophotometrically in a 680 Bio-Rad microplate ELISA
reader equipped with a 600 nm filter. For short-time effects,
immediately after the 4 h incubation with peptide, a 90 μL aliquot
of the parasite suspension was transferred into a 96-microwell
plate, an equal volume of MTT (1 mg/mL) added, and incubation
continued for an additional 2 h followed by quantitation as above.
All experiments were performed by triplicate and repeated at least
twice. To illustrate short- and long-term leishmanicidal activity,
the IC50 and LC50 parameters were used (see Abbreviations).
Electron Microscopy. L. donovani promastigotes were incu-
bated with peptides under standard conditions, except for a scale
up of the incubation volume to 800 μL. Parasites were collected
by centrifugation, washed twice with 1 mL of PBS, fixed in 5%
(w/v) glutaraldehyde in the same medium, and included with
2.5% (w/v) OsO4 for 1 h. Next, the cells were gradually dehy-
drated in ethanol [30, 50, 70, 90, and 100% (v/v); 30 min each],
included with propylene oxide (1 h), embedded in Epon 812 resin,
and observed in a Jeol-1230 electron microscope.63
NMR Spectroscopy. Samples for NMR experiments were
dissolved in PBS buffer (pH 6.3) or in TFE-containing aqueous
solution (30% v/v trifluoroethanol-d3, Cambridge Isotope
Laboratories) to make a final peptide concentration of ca.
3 mM. Spectra were recorded at 298K on a Bruker DRX-500
spectrometer (1H frequency of 500 MHz) using a 5 mm triple-
resonance Z gradient probe and processed using XWIN-NMR
software (Bruker). The transmitter frequency was set on the
HDO/H2O signal, and the TSP resonance was used as chemical
shift reference (δ TSP = 0 ppm). One-dimensional (1D) and
two-dimensional (2D) spectra were acquired by standard pulse
sequences using WATERGATE-based solvent suppression se-
quences. Total correlation spectroscopy (TOCSY), and nuclear
Overhauser effect spectroscopy (NOESY), were all collected in
the phase-sensitive mode using 512 time-proportional phase
increments in t1 for each experiment. The free induction decay
in t2 consisted of 2K complex data points over a spectral width
of 6009.615 Hz. Usually 4096 ꢀ 1024 data points were collected
for each block, 96 transients were collected for two-dimensional
experiments. The spectral width was 10 ppm, and the relaxation
delays were set to 1.5 and 2 s in the TOCSY and NOESY
experiments, respectively. TOCSY spectra were recorded using
the MLEV-17 pulse sequence with mixing times (spin-lock) of
65-80 ms. NOESY experiments were acquired with mixing time
of 150-200 ms to avoid any possible contribution from spin
diffusion. The NH-CRH scalar couplings were obtained from
high-resolution 1H monodimensional spectra. The experimental
data were acquired and processed using the TopSpin program
on a PC station. The data matrices were multiplied by a qsine
function in both dimensions and then zero-filled to 1024 data
points in F1 prior to Fourier transformation.
Real Time Monitoring of in Vivo Changes in Intracellular Levels
of ATP in L. donovani Promastigotes. A previously described
protocol63 was followed. Briefly, L. donovani promastigotes ofthe
3Luc strain, expressing a cytoplasmic form of luciferase, were
resuspended at 2 ꢀ 107 promastigotes/mL in Hanks þ Glc and
incubated at 25 °C with the membrane-permeant luciferase
substrate DMNEP-luciferin at 25 μM. When luminescence
reached a plateau, peptide was added (t = 0) and the ensuing
decayinluminescence was monitored ina BMGPolarstar Galaxy
€
microwell reader (Offenburg, Germany) equipped with lumines-
cence setting. Luminescence data were referred to the percentage
of luminescence of the corresponding untreated promastigotes.
In vitro inhibition of Photinus pyralis luciferase (EC 1.13.12.7) by
the peptides was tested with purified enzyme (Roche Applied
Sciences, Madrid, Spain) as described.39
Peak lists for the NOESY spectra recorded with a 0.15-0.2 s
mixing time were generated by interactive peak picking using the
CARA software.46 NOESY cross-peak volumes were deter-
mined by the automated peak integration routine implemented
in CARA. Three-dimensional structures were determined by the
standard protocol of the CYANA program (version 2.1),50
using seven cycles of combined automated NOESY assignment
and structure calculations followed by a final structure calcula-
tion. Since the Lys(Me3) residue is not included in the standard
CYANA libraries, it was built using the MOLMOL76 program,
also used to visualize the three-dimensional structures. For each
CYANA cycle, 1000 randomized conformers and the standard
simulated annealing schedule were used. The 20 conformers
with the lowest final score were retained for analysis and passed
on to the next cycle. Weak restraints on ψ/ψ torsion-angle pairs
and on side-chain torsion angles between tetrahedral carbon
atoms were applied temporarily during the high-temperature
and cooling phases of the simulated annealing schedule in order
to favor the permitted regions of the Ramachandran plot and
staggered rotamer positions, respectively. The list of upper-
distance bonds for the final structural calculation consists of
unambiguously assigned upper-distance bonds and does not
require the possible swapping of diastereotopic pairs. Next, the
20 conformers with the lowest final CYANA target function
values were subjected to restrained energy-minimization in a
water shell using the AMBER 8.0 program.77 The protein was
immersed in a shell of water molecules created using the TIP3P
model, with a thickness of 10 A (1 A = 0.1 nm). The number of
additional water molecules oscillated between 2081 and 2991
over the 20 conformers of the three peptides. The restrained
energy minimization was performed in three stages. In the first
stage, only the water molecules were optimized. Subsequently,
the peptide alone was relaxed, maintaining the water molecules
fixed, and, finally, the whole system was minimized. In the last
stage, a maximum of 1500 steps of restrained energy minimiza-
tion and a combination of the steepest descent and conjugate
gradient algorithms were applied using, in addition to the force
field of Cornell et al.,78 a parabolic or linear penalty function
for the NOE upper distance bonds and torsion-angle restraints.
Hemolysis Assay. The release of hemoglobin from sheep
erythrocytes was measured as described.75 Washed erythrocytes
were resuspended in Hank’s medium at 2 ꢀ 107 cells/mL. Then
100 μL aliquots of this suspension were incubated in 1.5 mL
Eppendorf tubes with the peptides (4 h, 37 °C). After centrifuga-
tion (14000 rpm, 5 min, 4 °C), 80 μL aliquots of the supernatant
were transferred into a 96-well culture microplate and hemoglo-
bin was measured at 550 nm in a Bio-Rad 680 microplate reader.
Full lysis was taken as that caused by incubation with 0.1%
Triton X-100 (TX-100). HC50, defined as the peptide concentra-
tion causing 50% hemoglobin release relative to that obtained
with TX-100, was calculated using SigmaPlot v. 11.0. The
selectivity index (SI) was defined as the ratio between HC50 and
LC50 (for Leishmania) or betweenHC50 and MIC50 (for bacteria).
Plasma Membrane Permeation. The loss of Leishmania plas-
ma membrane potential was monitored by the increase in
fluorescence of the potential-sensitive, anionic dye bisoxonol,
upon its insertion into the depolarized parasite membrane. The
above standard assay conditions were maintained except for
inclusion of 0.2 μM bisoxonol in the incubation medium.
Maximal depolarization was defined as that achieved by peptide
CA(1-8)M(1-18) at 3.0 μM.39
˚
˚
In complementary experiments, the production of larger
lesions in the parasite plasma membrane by the peptides was
assessed by access of the membrane-impermeable vital dye
Sytox Green to the cytoplasm. The above standard assay
conditions were maintained except for inclusion of 1 μM final
concentration of the probe in the incubation medium. Fluores-
cence changes were monitored at λex = 485 nm and λem
=
520 nm. Maximal permeation was defined as that achieved by
0.1% Triton X-100. Sytox Green incorporation into individual
parasite cells was assessed by cytofluorometry under the same
conditions used above for bulk dye entrance, using a Coulter XL
EPICS cytofluorometer.

Article
Journal of Medicinal Chemistry, 2010, Vol. 53, No. 15 5595
The force constants were chosen such that restraint violations of
˚
peptides by hydrocarbon staples. J. Am. Chem. Soc. 2009, 131,
4622–4627.
(13) Smith, B. C.; Denu, J. M. Chemical mechanisms of histone lysine
and arginine modifications. Biochim. Biophys. Acta 2009, 1789,
45–57.
(14) Cedar, H.; Bergman, Y. Linking DNA methylation and histone
modification: patterns and paradigms. Nat. Rev. Genet. 2009, 10,
295–304.
(15) Fingerman, I. M.; Du, H. N.; Briggs, S. D. Controlling his-
tone methylation via trans-histone pathways. Epigenetics 2008, 3,
237–242.
(16) Vaissiere, T.; Sawan, C.; Herceg, Z. Epigenetic interplay between
histone modifications and DNA methylation in gene silencing.
Mutat. Res. 2008, 659, 40–48.
(17) Vaillant, I.; Paszkowski, J. Role of histone and DNA methylation
in gene regulation. Curr. Opin. Plant Biol. 2007, 10, 528–533.
(18) Cho, J. H.; Sung, B. H.; Kim, S. C. Buforins: Histone H2A-derived
antimicrobial peptides from toad stomach. Biochim. Biophys. Acta
2009, 1788, 1564–1569.
(19) Koo, Y. S.; Kim, J. M.; Park, I. Y.; Yu, B. J.; Jang, S. A.; Kim,
K. S.; Park, C. B.; Cho, J. H.; Kim, S. C. Structure-activity
relations of parasin I, a histone H2A-derived antimicrobial pep-
tide. Peptides 2008, 29, 1102–1108.
(20) Birkemo, G. A.; Luders, T.; Andersen, O.; Nes, I. F.; Nissen-
Meyer, J. Hipposin, a histone-derived antimicrobial peptide in
Atlantic halibut (Hippoglossus hippoglossus L.). Biochim. Biophys.
Acta 2003, 1646, 207–215.
0.1 A and torsion violations of 1° contributed by 0.84 kJ/mol
(0.2 kcal/mol) to the potential energy. The resulting 20 energy-
minimized conformers represent the solution structure of CA-M
and derivatives.
Root-mean-square deviation (rmsd) values were calculated
using CYANA for superpositions of the backbone N, CR, and
C0 atoms; the heavy atoms over the whole protein, or over its
structured regions. To obtain the rmsd of a structure repre-
sented by a bundle of conformers, all conformers were super-
imposed upon the first one and the average of the rmsd values
between the individual conformers and their average coordi-
nates was calculated. Conformational energies were calculated
using AMBER 8.0 and applying the force field of Cornell et al.78
The statistics regarding the quality and precision of the 20
energy minimized conformers that represent the solution structure
of CA-M and derivatives are summarized in Supporting Informa-
tion Table S4. The quality of the structures was also reflected by
the presence of 59.4-87.1% of the (ψ/ψ) backbone torsion-angle
pairs in the most favored regions, 12.9-49.6% of which were
found within the regions additionally permitted by the Rama-
chandran plot, according to PROCHECK51 conventions.
(21) Parseghian, M. H.; Luhrs, K. A. Beyond the walls of the nucleus:
the role of histones in cellular signaling and innate immunity.
Biochem. Cell Biol. 2006, 84, 589–604.
(22) Kawasaki, H.; Iwamuro, S. Potential roles of histones in host
defense as antimicrobial agents. Infect. Dis. Drug Targets 2008, 8,
195–205.
(23) Wartha, F.; Beiter, K.; Normark, S.; Henriques-Normark, B.
Neutrophil extracellular traps: casting the NET over pathogenesis.
Curr. Opin. Microbiol. 2007, 10, 52–56.
(24) Fernandes, J. M.; Saint, N.; Kemp, G. D.; Smith, V. J. Oncorhyn-
cin III: a potent antimicrobial peptide derived from the non-histone
chromosomal protein H6 of rainbow trout, Oncorhynchus mykiss.
Biochem J. 2003, 373, 621–628.
(25) Na, D. H.; Faraj, J.; Capan, Y.; Leung, K. P.; DeLuca, P. P.
Stability of antimicrobial decapeptide (KSL) and its analogues for
delivery in the oral cavity. Pharm. Res. 2007, 24, 1544–1550.
(26) Haug, B. E.; Strom, M. B.; Svendsen, J. S. The medicinal chemistry
of short lactoferricin-based antibacterial peptides. Curr. Med.
Chem. 2007, 14, 1–18.
Acknowledgment. This work was supported by the
European Union (HEALTH-2007-223414, Leishdrug, to L.R.
and D.A.), the Spanish Ministry of Science and Innovation
(PET2006-0139 to D.A. and L.R., CTQ2009-0856 to J.J.B.,
BIO2005-07592-CO2-02 and BIO200804487-CO3-02 to D.A.),
Fondo de Investigaciones Sanitarias (PI061125, PS09-01928,
and RD 06/0021/0006 to L.R., PI040885 to D.A.), and by the
regional governments of Madrid (S-BIO-0260/2006 to L.R.,
BIPPED to J.J.B.) and Catalonia (SGR2005-00494). D.D. isa
ꢀ
recipient of a Ramon y Cajal contract from the Spanish Ministry
of Science and Innovation.
Supporting Information Available: NMR data and details on
structure calculations. This material is available free of charge
via the Internet at http://pubs.acs.org.
(27) Ramalingam, K.; Bello, J. Effect of permethylation on the haemo-
lytic activity of melittin. Biochem. J. 1992, 284 (Pt 3), 663–665.
(28) Merrifield, R. B.; Merrifield, E. L.; Juvvadi, P.; Andreu, D.;
Boman, H. G. Design and synthesis of antimicrobial peptides.
Ciba Found. Symp. 1994, 186, 5-20; discussion 20-6.
(29) Rivas, L.; Andreu, D. Cecropin-melittin hybrid peptides as ver-
satile templates in the development of membrane active antibiotics
agents. In Pore-Forming Peptides and Protein Toxins; Menestrina,
G.; Dalla Serra, M., Eds.; Harwood Academic Publishers: Reading,
Berkshire, UK, 2003; pp 215-219.
(30) Andreu, D.; Ubach, J.; Boman, A.; Wahlin, B.; Wade, D.;
Merrifield, R. B.; Boman, H. G. Shortened cecropin A-melittin
hybrids. Significant size reduction retains potent antibiotic activity.
FEBS Lett. 1992, 296, 190–194.
(31) Boman, H. G.; Wade, D.; Boman, I. A.; Wahlin, B.; Merrifield,
R. B. Antibacterial and antimalarial properties of peptides that are
cecropin-melittin hybrids. FEBS Lett. 1989, 259, 103–106.
(32) Scott, M. G.; Gold, M. R.; Hancock, R. E. Interaction of cationic
peptides with lipoteichoic acid and gram-positive bacteria. Infect.
Immun. 1999, 67, 6445–6453.
(33) Friedrich, C.; Scott, M. G.; Karunaratne, N.; Yan, H.; Hancock,
R. E. Salt-resistant alpha-helical cationic antimicrobial peptides.
Antimicrob. Agents Chemother. 1999, 43, 1542–1548.
(34) Piers, K. L.; Brown, M. H.; Hancock, R. E. Improvement of outer
membrane-permeabilizing and lipopolysaccharide-binding activ-
ities of an antimicrobial cationic peptide by C-terminal modifica-
tion. Antimicrob. Agents Chemother. 1994, 38, 2311–2316.
(35) Cavallarin, L.; Andreu, D.; San Segundo, B. Cecropin A-derived
peptides are potent inhibitors of fungal plant pathogens. Mol.
Plant-Microbe Interact. 1998, 11, 218–227.
(36) Yevtushenko, D. P.; Romero, R.; Forward, B. S.; Hancock, R. E.;
Kay, W. W.; Misra, S. Pathogen-induced expression of a cecropin
A-melittin antimicrobial peptide gene confers antifungal resistance
in transgenic tobacco. J. Exp. Bot. 2005, 56, 1685–1695.
(37) Diaz-Achirica, P.; Ubach, J.; Guinea, A.; Andreu, D.; Rivas, L.
The plasma membrane of Leishmania donovani promastigotes is the
References
(1) Jenssen, H.; Hamill, P.; Hancock, R. E. W. Peptide antimicrobial
agents. Clin. Microbiol. Rev. 2006, 19, 491–511.
(2) Lievin-Le Moal, V.; Servin, A. L. The front line of enteric host
defense against unwelcome intrusion of harmful microorganisms:
mucins, antimicrobial peptides, and microbiota. Clin. Microbiol.
Rev. 2006, 19, 315–337.
(3) Matsuzaki, K. Control of cell selectivity of antimicrobial peptides.
Biochim. Biophys. Acta 2009, 1788, 1687–1692.
(4) Melo, M. N.; Ferre, R.; Castanho, M. A. Antimicrobial peptides:
linking partition, activity and high membrane-bound concentra-
tions. Nat. Rev. Microbiol. 2009, 7, 245–250.
(5) Nizet, V. Antimicrobial peptide resistance mechanisms of human
bacterial pathogens. Curr. Issues Mol. Biol. 2006, 8, 11–26.
(6) Pachon, J.; Vila, J. Treatment of multiresistant Acinetobacter
baumannii infections. Curr. Opin. Invest. Drugs 2009, 10, 150–156.
(7) Lai, Y.; Gallo, R. L. AMPed up immunity: how antimicrobial
peptides have multiple roles in immune defense. Trends Immunol.
2009, 30, 131–141.
(8) Mookherjee, N.; Rehaume, L. M.; Hancock, R. E. Cathelicidins
and functional analogues as antisepsis molecules. Expert Opin.
Ther. Targets 2007, 11, 993–1004.
(9) Zelezetsky, I.; Tossi, A. Alpha-helical antimicrobial peptides;
using a sequence template to guide structure-activity relationship
studies. Biochim. Biophys. Acta 2006, 1758, 1436–1449.
(10) Giuliani, A.; Pirri, G.; Bozzi, A.; Di Giulio, A.; Aschi, M.; Rinaldi,
A. C. Antimicrobial peptides: natural templates for synthetic
membrane-active compounds. Cell. Mol. Life Sci. 2008, 65,
2450–2460.
(11) Sitaram, N. Antimicrobial peptides with unusual amino acid
compositions and unusual structures. Curr. Med. Chem. 2006, 13,
679–696.
(12) Kutchukian, P. S.; Yang, J. S.; Verdine, G. L.; Shakhnovich, E. I.
All-atom model for stabilization of alpha-helical structure in

5596 Journal of Medicinal Chemistry, 2010, Vol. 53, No. 15
Fernaꢀndez-Reyes et al.
main target for CA(1-8)M(1-18), a synthetic cecropin A-melittin
hybrid peptide. Biochem. J. 1998, 330 (Pt 1), 453–460.
(38) Haines, L. R.; Hancock, R. E.; Pearson, T. W. Cationic antimi-
crobial peptide killing of African trypanosomes and Sodalis glossi-
nidius, a bacterial symbiont of the insect vector of sleeping sickness.
Vector Borne Zoonotic Dis. 2003, 3, 175–186.
(39) Luque-Ortega, J. R.; Saugar, J. M.; Chiva, C.; Andreu, D.; Rivas,
L. Identification of new leishmanicidal peptide lead structures by
automated real-time monitoring of changes in intracellular ATP.
Biochem. J. 2003, 375, 221–230.
aureus-derived proteinases. Antimicrob. Agents Chemother. 2004, 48,
4673–4679.
(59) Kulkarni, M. M.; McMaster, W. R.; Kamysz, E.; Kamysz, W.;
Engman, D. M.; McGwire, B. S. The major surface-metallo-
protease of the parasitic protozoan, Leishmania, protects against
antimicrobial peptide-induced apoptotic killing. Mol. Microbiol.
2006, 62, 1484–1497.
(60) Ladokhin, A. S.; White, S. H. Folding of amphipathic alpha-helices
on membranes: energetics of helix formation by melittin. J. Mol.
Biol. 1999, 285, 1363–1369.
(40) Alberola, J.; Rodriguez, A.; Francino, O.; Roura, X.; Rivas, L.;
Andreu, D. Safety and efficacy of antimicrobial peptides against
naturally acquired leishmaniasis. Antimicrob. Agents Chemother.
2004, 48, 641–643.
(41) Nos-Barbera, S.; Portoles, M.; Morilla, A.; Ubach, J.; Andreu, D.;
Paterson, C. A. Effect of hybrid peptides of cecropin A and melittin
in an experimental model of bacterial keratitis. Cornea 1997, 16,
101–106.
(42) Osusky, M.; Zhou, G.; Osuska, L.; Hancock, R. E.; Kay, W. W.;
Misra, S. Transgenic plants expressing cationic peptide chimeras
exhibit broad-spectrum resistance to phytopathogens. Nature Bio-
technol. 2000, 18, 1162–1166.
(43) Fields, G. B.; Noble, R. L. Solid phase peptide synthesis utilizing
9-fluorenylmethoxycarbonyl amino acids. Int. J. Pept. Protein Res.
1990, 35, 161–214.
(61) Wieprecht, T.; Beyermann, M.; Seelig, J. Binding of antibacterial
magainin peptides to electrically neutral membranes: thermody-
namics and structure. Biochemistry 1999, 38, 10377–10387.
(62) Hernandez, C.; Mor, A.; Dagger, F.; Nicolas, P.; Hernandez, A.;
Benedetti, E. L.; Dunia, I. Functional and structural damage in
Leishmania mexicana exposed to the cationic peptide dermaseptin.
Eur. J. Cell Biol. 1992, 59, 414–424.
(63) Guerrero, E.; Saugar, J. M.; Matsuzaki, K.; Rivas, L. Role
of positional hydrophobicity in the leishmanicidal activity of
magainin 2. Antimicrob. Agents Chemother. 2004, 48, 2980–2986.
(64) Granados, E. N.; Bello, J. Interactions of poly (N epsilon, N
epsilon, N epsilon, -trimethyllysine) and poly(lysine) with poly-
nucleotides: circular dichroism and A-T sequence selectivity.
Biochemistry 1981, 20, 4761–4765.
(65) Mancheno, J. M.; Onaderra, M.; Martinez del Pozo, A.;
Diaz-Achirica, P.; Andreu, D.; Rivas, L.; Gavilanes, J. G. Release
of lipid vesicle contents by an antibacterial cecropin A-melittin
hybrid peptide. Biochemistry 1996, 35, 9892–9899.
(66) Bastos, M.; Bai, G.; Gomes, P.; Andreu, D.; Goormaghtigh, E.;
Prieto, M. Energetics and partition of two cecropin-melittin
hybrid peptides to model membranes of different composition.
Biophys. J. 2008, 94, 2128–2141.
(44) Chicharro, C.; Granata, C.; Lozano, R.; Andreu, D.; Rivas, L.
N-Terminal fatty acid substitution increases the leishmanicidal
activity of CA(1-7)M(2-9), a cecropin-melittin hybrid peptide.
Antimicrob. Agents Chemother. 2001, 45, 2441–2449.
€
(45) Wuthrich, K. NMR of Proteins and Nucleic Acids; John Wiley &
Sons: New York, 1986.
(46) Keller, R. L. J. The Computer Aided Resonance Assignment Tutorial;
Cantina Verlag: Goldau, Switzerland, 2004.
(67) Benachir, T.; Lafleur, M. Study of vesicle leakage induced by
(47) Wishart, D. S.; Sykes, B. D.; Richards, F. M. The chemical shift
index: a fast and simple method for the assignment of protein
secondary structure through NMR spectroscopy. Biochemistry
1992, 31, 1647–1651.
(48) Holak, T. A.; Engstrom, A.; Kraulis, P. J.; Lindeberg, G.; Bennich,
H.; Jones, T. A.; Gronenborn, A. M.; Clore, G. M. The solution
conformation of the antibacterial peptide cecropin A: a nuclear
magnetic resonance and dynamical simulated annealing study.
Biochemistry 1988, 27, 7620–7629.
melittin. Biochim. Biophys. Acta 1995, 1235, 452–460.
(68) Gregory, S. M.; Cavenaugh, A.; Journigan, V.; Pokorny, A.;
Almeida, P. F. A quantitative model for the all-or-none permea-
bilization of phospholipid vesicles by the antimicrobial peptide
cecropin A. Biophys. J. 2008, 94, 1667–1680.
(69) Killian, J. A. Synthetic peptides as models for intrinsic membrane
proteins. FEBS Lett. 2003, 555, 134–138.
(70) Abrunhosa, F.; Faria, S.; Gomes, P.; Tomaz, I.; Pessoa, J. C.;
Andreu, D.; Bastos, M. Interaction and lipid-induced conforma-
tion of two cecropin-melittin hybrid peptides depend on peptide
and membrane composition. J. Phys. Chem. B 2005, 109, 17311–
17319.
(49) Bazzo, R.; Tappin, M. J.; Pastore, A.; Harvey, T. S.; Carver, J. A.;
Campbell, I. D. The structure of melittin. A 1H-NMR study in
methanol. Eur. J. Biochem. 1988, 173, 139–146.
(50) Guntert, P. Automated NMR structure calculation with CYANA.
(71) Bhargava, K.; Feix, J. B. Membrane binding, structure, and
localization of cecropin-mellitin hybrid peptides: a site-directed
spin-labeling study. Biophys. J. 2004, 86, 329–336.
(72) Oren, Z.; Hong, J.; Shai, Y. A repertoire of novel antibacterial
diastereomeric peptides with selective cytolytic activity. J. Biol.
Chem. 1997, 272, 14643–14649.
Methods Mol. Biol. 2004, 278, 353–378.
(51) Laskowski, R. A.; MacArthur, M. W.; Moss, D. S.; Thornton,
J. M. Procheck: a program to check the stereochemical quality of
protein structures. J. Appl. Crystallogr. 1993, 26, 283–291.
(52) Subramanian, K.; Jia, D.; Kapoor-Vazirani, P.; Powell, D. R.;
Collins, R. E.; Sharma, D.; Peng, J.; Cheng, X.; Vertino, P. M.
Regulation of estrogen receptor alpha by the SET7 lysine methyl-
transferase. Mol. Cell 2008, 30, 336–347.
(53) Scoumanne, A.; Chen, X. Protein methylation: a new mechanism
of p53 tumor suppressor regulation. Histol. Histopathol. 2008, 23,
1143–1149.
(54) Yang, X. D.; Huang, B.; Li, M.; Lamb, A.; Kelleher, N. L.; Chen,
L. F. Negative regulation of NF-kappaB action by Set9-mediated
lysine methylation of the RelA subunit. EMBO J. 2009, 28, 1055–
1066.
(55) Kurash, J. K.; Lei, H.; Shen, Q.; Marston, W. L.; Granda, B. W.;
Fan, H.; Wall, D.; Li, E.; Gaudet, F. Methylation of p53 by Set7/9
mediates p53 acetylation and activity in vivo. Mol. Cell 2008, 29,
392–400.
(56) Haiko, J.; Suomalainen, M.; Ojala, T.; Lahteenmaki, K.; Korho-
nen, T. K. Invited review: Breaking barriers;attack on innate
immune defences by omptin surface proteases of enterobacterial
pathogens. Innate Immun. 2009, 15, 67–80.
(57) Bachrach, G.; Altman, H.; Kolenbrander, P. E.; Chalmers, N. I.;
Gabai-Gutner, M.; Mor, A.; Friedman, M.; Steinberg, D. Resis-
tance of Porphyromonas gingivalis ATCC 33277 to direct killing by
antimicrobial peptides is protease independent. Antimicrob. Agents
Chemother. 2008, 52, 638–642.
(58) Sieprawska-Lupa, M.; Mydel, P.; Krawczyk, K.; Wojcik,
K.; Puklo, M.; Lupa, B.; Suder, P.; Silberring, J.; Reed, M.;
Pohl, J.; Shafer, W.; McAleese, F.; Foster, T.; Travis, J.; Potempa, J.
Degradation of human antimicrobial peptide LL-37 by Staphylococcus
(73) Gootz, T. D. The global problem of antibiotic resistance. Crit. Rev.
Immunol. 2010, 30, 79–93.
(74) Saugar, J. M.; Rodriguez-Hernandez, M. J.; de la Torre, B. G.;
Pachon-Ibanez, M. E.; Fernandez-Reyes, M.; Andreu, D.; Pachon,
J.; Rivas, L. Activity of cecropin A-melittin hybrid peptides
against colistin-resistant clinical strains of Acinetobacter bauman-
nii: molecular basis for the differential mechanisms of action.
Antimicrob. Agents Chemother. 2006, 50, 1251–1256.
(75) Solanas, C.; de la Torre, B. G.; Fernandez-Reyes, M.; Santiveri,
C. M.; Jimenez, M. A.; Rivas, L.; Jimenez, A. I.; Andreu, D.;
Cativiela, C. Therapeutic index of gramicidin S is strongly modu-
lated by D-phenylalanine analogues at the beta-turn. J. Med. Chem.
2009, 52, 664–674.
(76) Koradi, R.; Billeter, M.; Wuthrich, K. MOLMOL: a program
for display and analysis of macromolecular structures. J. Mol.
Graphics 1996, 14 (51-5), 29–32.
(77) Case, D. A.; Darden, T. E.; Cheatham III, T. E.; Simmerling, C. L.;
Wang, J.; Duke, R. E.; Luo, R.; Merz, K. M.; Wang, B.; Pearlman,
D. A.; Crowley, M.; Brozell, S.; Tsui, V.; Gohlke, H.; Mongan, J.;
Hornak, V.; Cui, G.; Beroza, P.; Schafmeister, C.; Caldwell, J. W.;
Ross, W. S.; Kollman, P. A. AMBER 8; University of California: San
Francisco, 2004.
(78) Cornell, W. D.; Cieplak, P.; Bayly, C. I.; Gould, I. R.; Merz, K. M.;
Ferguson, D. M.; Spellmeyer, D. C.; Fox, T.; Caldwell, J. W.;
Kollman, P. A. A Second Generation Force Field for the Simula-
tion of Proteins, Nucleic Acids, and Organic Molecules. J. Am.
Chem. Soc. 1995, 117, 5179–5197.