Antimicrobial peptides from plants: Stabilization of the γ core of a tomato defensin by intramolecular disulfide bond

Article abstract of DOI:10.1002/psc.2479

Cysteine-containing antimicrobial peptides of diverse phylogeny share a common structural signature, the γ core, characterized by a strong polarization of charges in two antiparallel β sheets. In this work, we analyzed peptides derived from the tomato defensin SolyC07g007760 corresponding to the protein γ core and demonstrated that cyclization of the peptides, which results in segregation of positive charges to the turn region, produces peptides very active against Gram negative bacteria, such as Salmonella enterica and Helicobacter pylori. Interestingly, these peptides show very low hemolytic activity and thus represent a scaffold for the design of new antimicrobial peptides.

Full text of DOI:10.1002/psc.2479

Special Issue Article
Received: 30 October 2012
Revised: 6 December 2012
Accepted: 6 December 2012
Published online in Wiley Online Library: 19 February 2013
(wileyonlinelibrary.com) DOI 10.1002/psc.2479
Antimicrobial peptides from plants:
stabilization of the g core of a tomato
defensin by intramolecular disulfide bond{
C. Avitabile,a R. Capparelli,b M. M. Rigano,b A. Fulgione,b A. Barone,b
C. Pedonea and A. Romanellia*
Cysteine-containing antimicrobial peptides of diverse phylogeny share a common structural signature, the g core, characterized
by a strong polarization of charges in two antiparallel b sheets. In this work, we analyzed peptides derived from the tomato
defensin SolyC07g007760 corresponding to the protein g core and demonstrated that cyclization of the peptides, which results
in segregation of positive charges to the turn region, produces peptides very active against Gram negative bacteria, such as
Salmonella enterica and Helicobacter pylori. Interestingly, these peptides show very low hemolytic activity and thus represent a
scaffold for the design of new antimicrobial peptides. Copyright © 2013 European Peptide Society and John Wiley & Sons, Ltd.
Keywords: tomato defensin; cyclization; antimicrobial
bacteria, killing only the minority of the probiotic tested. In
addition, SolyC downregulated the proinflammatory cytokines
Introduction
A tremendous effort has been recently devoted to the discovery
of biologically active peptides from natural sources [1–3]. Plants
produce a wide variety of antimicrobial peptides and proteins
showing activity against either plant or human pathogens [4].
Many of these, such as thionins and knottins, are rich in cysteines
and are stabilized by a number of disulfide bonds [5,6]. Interest-
ingly, some of the structural features displayed by peptides pro-
duced in plants are well conserved also within proteins produced
by evolutionarily diverse organisms, and in fact, the existence of
multidimensional signatures for antimicrobial peptides has been
hypothesized [7–9]. The g core motif has recently been indicated as
a three dimensional signature for antimicrobial peptides or proteins
and, generally, as a feature shown by membrane-active proteins. In
particular, the g core motif, characterized by a conserved CXG
sequence and two antiparallel b sheets, is common to host defence
polypeptides showing multiple disulfide bonds such as toxins,
kinocidins, and thionins [10,11]. This structural motif can be
generated by different amino acid sequences and can exist in the
dextromeric or levomeric form, depending on whether it is derived
by the amino acid forward or reverse sequence (Figure 1). In all
cases, g core motifs show polarization of positive charge and
segregation of hydrophobic amino acids. There are some cases in
which the only g core is sufficient for the peptide antimicrobial
activity, as observed for protegrins, θ defensin RTD-1, and other
cases in which the g core represents the scaffold around which
helices and sheets packs, as for kinocidins [12,13].
to an extent comparable with the known anti-inflammatory drug
acetyl salicylic acid and did not cause red blood cell hemolysis
and was not toxic toward eukaryotic cells.
SolyC is a peptide composed by 17 amino acids and contains
three cysteines in positions 6, 13, and 15. The results described
so far were obtained employing the peptide in its linear form.
The presence of three cysteines that might be subjected to
spontaneous oxidation and the observation that in typical g cores
those cysteines are involved in disulfide bonds urged us to
investigate how formation of disulfide bonds in a controlled
fashion could affect the secondary structure and the antimicrobial
activity of the peptide. Two peptides in which Cys13 or Cys15 was
mutated into a serine were obtained in the linear and oxidized
form. Disulfide bonds were formed between C6 and C13 (SolyC1)
or C6 and C15 (SolyC2); in both cases, one cysteine was missing.
Truncated peptides, starting from the conserved GXC motif, were
also obtained in the linear and oxidized form. The secondary
structure of all peptides was analyzed by circular dichroism (CD);
the antimicrobial activity of the peptides was tested against Gram
positive and Gram negative bacteria. All the oxidized peptides
*
Correspondence to: Alessandra Romanelli, University of Naples ‘Federico II’,
School of Biotechnological Sciences, Department of Biological Sciences, Via
Mezzocannone 16, 80134, Naples, Italy. E-mail: alessandra.romanelli@unina.it
In a recent paper, we identified, after a bioinformatics analysis
of the tomato genome, the tomato defensin SolyC07g007760
(iTAG v.2.3) and demonstrated that a synthetic peptide (SolyC)
corresponding to its g core exerts antimicrobial activity against
both Gram positive and Gram negative bacteria, being particu-
larly active against different strains of the Gram negative bacte-
rium Helicobacter pylori [14,15]. Interestingly, the peptide SolyC
was able to discriminate between pathogens and probiotic
{
Special issue devoted to contributions presented at the 13th Naples Workshop on
Bioactive Peptides, June 7–10, 2012, Naples.
a University of Naples ‘Federico II’, School of Biotechnological Sciences, Depart-
ment of Biological Sciences, Via Mezzocannone 16, 80134, Naples Italy
b University of Naples ‘Federico II’, School of Biotechnological Sciences, Depart-
ment of Soil, Plant, Environmental and Animal Production Sciences, Via Univer-
sità 100, 80055, Portici Italy
J. Pept. Sci. 2013; 19: 240–245
Copyright © 2013 European Peptide Society and John Wiley & Sons, Ltd.

ANTIMICROBIAL PEPTIDES FROM PLANTS
SolyC (Da): calculated 1994.30; [M + 2H]2+: 998.16; [M + 3H]3+
:
665.77; found: [M + 2H]2+: 997.55; [M + 3H]3+: 665.64; retention
time: 5.40 min.
SolyC-t (Da): calculated 1703.01; [M + 2H]2+: 853.50; [M + 3H]3+
568.67; found: [M + 1H]1+: 1702.58; [M + 2H]2+: 852.29; [M + 3H]3+
568.52; retention time: 12.0 min.
:
:
SolyC1 (Da): calculated 1978.26; [M + 2H]2+: 990.13; [M + 3H]3+
:
660.42; found: [M + 2H]2+: 990,13; [M + 3H]3+: 660.41; retention
time: 5.12 min.
SolyC1-t (Da): calculated 1686.95; [M + 2H]2+: 844.47; [M + 3H]3+
:
Figure 1. Schematic representation of the gamma core isoforms.
563.31; found: [M + 2H]2+: 844.40; [M + 3H]3+: 563.21; retention
time: 5.12 min.
are highly active against Gram negative bacteria; the linear
peptide SolyC2 shows antimicrobial activity against Gram
negative bacteria comparable with that of the reference peptide
SolyC, whereas linear SolyC1 exhibits an antimicrobial activity
against Gram negative bacteria at higher concentrations. The
activity of modified peptides against Gram positive bacteria was
found in all cases lower as compared with that of the reference
peptide. These results suggest that the interaction of the peptides
with the membranes of Gram negative bacteria is stronger when
the positive charges are exposed, and this likely occurs when
the peptides are in the oxidized form, in which the antiparallel
b sheets are constrained by the disulfide bond.
SolyC2 (Da): calculated 1978.26; [M + 2H]2+: 990.13; [M + 3H]3+
:
660.42; found: [M + 2H]2+: 989.90; [M + 3H]3+: 660.21; retention
time: 5.33 min.
3+
SolyC2-t (Da): calculated 1686.95; [M+ 2H]2+: 844.47; [M + 3H]
:
563.31; found: [M+ 2H]2+: 844.03; [M+ 3H]3+: 563.23; retention
time: 4.93 min.
Oxidation Reaction
Disulfide-bridged analogs were obtained from the purified linear
peptides after an oxidation reaction. Linear peptides were
dissolved in a mixture of H2O/acetic acid (95/5, v/v) at a final con-
centration of 2.5 mM (9 mL total volume); pH was adjusted to 6
with ammonium hydrogen carbonate and then DMSO (1 mL)
was added. The reaction was complete after 24 h. Analysis of
the crudes was performed by LC-MS using a gradient of acetoni-
trile (0.1% TFA) in water (0.1% TFA) from 5% to 70% in 15 min.
Purification was performed by semipreparative RP-HPLC using a gra-
dient of acetonitrile (0.1% TFA) in water (0.1% TFA) from 5% to 70%
in 15 min. Peptides were characterized by ES mass spectrometry.
SolyC1-ox (Da): calculated 1976.26; [M + 2H]2+: 989.13; [M + 3H]3+:
659.75; found: [M + 2H]2+: 988.70; [M + 3H]3+: 659.60; retention time:
5.35 min.
Materials and Methods
Reagents
The amino acids used for the peptide synthesis, Fmoc-Arg(Pbf)-OH,
Fmoc-Asn(Trt)-OH, Fmoc-Cys(Trt)-OH, Fmoc-Gly-OH, Fmoc-Lys
(Boc)-OH, Fmoc-Phe-OH, Fmoc-Ser(OtBu)-OH, Fmoc-Thr(tBu)-OH,
the Rink amide MBHA resin and the activators N-hydroxybenzotria-
zole (HOBT), and O-benzotriazole-N,N,N0,N0-tetramethyl-uronium-
hexafluoro-phosphate (HBTU), were purchased from Novabiochem
(Gibbstown, NJ, USA). Acetonitrile was from Reidel-deHaën (Seelze,
Germany) and dry N,N-dimethylformamide (DMF) from LabScan
(Dublin, Ireland). All other reagents were from Sigma Aldrich (Milan,
Italy). LC-MS analyses were performed on a LC-MS Thermo Finnigan
with an electrospray source (MSQ) on a Phenomenex Jupiter 4m
Proteo (50 ꢀ 4.6mm) column for all peptides except for SolyC-t that
was analyzed on a Jupiter 4m Proteo (150 ꢀ 4.6 mm) column.
Purification was carried out on an Onyx monolithic semi-prep C18
(100 ꢀ 10 mm) column.
SolyC1-t-ox (Da): calculated 1684.94; [M + 2H]2+
: 843.47;
[M + 3H]3+: 562.64; found: [M + 2H]2+: 843.23; [M + 3H]3+: 562.53;
retention time: 4.96 min.
Solyc2-ox (Da): calculated 1976.26; [M + 2H]2+: 989.13; [M + 3H]3+:
659.75; found: [M + 2H]2+: 988.60; [M + 3H]3+: 659.60; retention time:
5.10 min.
SolyC2-t-ox (Da): calculated 1684.94; [M + 2H]2+: 843.47; [M+ 3H]3+
:
562.64; found: [M+ 2H]2+: 843.26; [M+ 3H]3+: 562.50; retention
time: 5.15 min.
Synthesis
Bacteria
Peptides were synthesized on solid phase by Fmoc chemistry on
the MBHA (0.54 mmol/g) resin by consecutive deprotection,
coupling, and capping cycles [16]. Deprotection: 30% piperidine
in DMF, 5 min (2ꢀ). Coupling: 2.5 equivalents of amino acid + 2.49
equivalent of HOBT/HBTU (0.45 M in DMF/DMSO 75/25 v/v) + 3.5
equivalents NMM, 40 min. Capping: acetic anhydride/DIPEA/DMF
15/15/70 v/v/v, 5 min.
Peptides were cleaved off from the resin and deprotected by
treatment of the resin with a solution of TFA/TIS/EDT/H2O 94/1/
2.5/2.5 v/v/v/v, 3 h, rt. TFA was concentrated, and peptides were pre-
cipitated in cold ethylic ether. Analysis of the crudes was performed
by LC-MS using a gradient of acetonitrile (0.1% TFA) in water
(0.1% TFA) from 5% to 70% in 15min. Purification was performed
by semipreparative RP-HPLC using a gradient of acetonitrile (0.1%
TFA) in water (0.1% TFA) from 5% to 70% in 15 min.
Bacterial isolates were from patients hospitalized at the Medical School
of the University of Naples ‘Federico II’ and at the ‘Villa Betania’ hospital
(Naples, Italy). Species identification was carried out by PCR [17–21]. Bac-
teria were grown at 37 ꢁC in Tryptic Soy Broth (TSB) or Luria Broth (LB)
medium, harvested in the exponential phase (OD 600 nm 0.6–0.8),
centrifuged (10 min at 8 ꢀ 103 g), and resuspended in Muller Hinton
broth at the concentration of ~105 CFU/ml. The resistance/susceptibility
of the different strains used in this study to conventional antibiotics was
also determined using the disk diffusion method on Mueller–Hinton
agar according to the NCCLS guidelines 2002 (data not shown).
Antibacterial and Hemolytic Activity
Bacteria were distributed in triplicate into plates (60 ml/well),
mixed with dilutions of the peptides (5–100 mg/ml; 40 ml/well),
J. Pept. Sci. 2013; 19: 240–245 Copyright © 2013 European Peptide Society and John Wiley & Sons, Ltd. wileyonlinelibrary.com/journal/jpepsci

AVITABILE ET AL.
and incubated at 37 ꢁC for 20 h. The minimal concentration of
peptides causing 100% growth inhibition (MIC100) was
determined by measuring the absorbance at 600 nm (Biorad
microplate reader model 680, CA) [22]. The test was performed
in triplicate. The hemolytic activity of the peptides was tested
using mouse red blood cells. The hemolytic activity was
measured according to the formula OD peptide ꢂ OD negative
control/OD positive control ꢂ OD negative control ꢀ 100, where the
negative control (0% hemolysis) was represented by erythrocytes sus-
pended in saline and the positive control (100% hemolysis) was repre-
sented by the erythrocytes lysed with 1% triton X100 [22].
The LC50 value relative to the SolyC was calculated as
described [23].
there is a preference for the formation of one of the two disul-
fides and also whether there is a difference in the activity of
the peptides with cycles of different size (Table 1). Because in g
cores cysteines contained in sequences CXC are rarely connected
by disulfide bonds [26], we explored only two of the possible
combinations of disulfides, namely C6–C13 (SolyC1) and C6–
C15 (SolyC2) and mutated the cysteine not involved in the
disulfide into a serine. To investigate the effect of reducing the
hydrophobicity of the peptide, we also obtained truncated pep-
tides, lacking three amino acids at the N-terminus. In the trun-
cated peptides, the residues that characterize the g core motif
GXC were conserved (Table 1). HPLC retention times for
truncated peptides are similar or lower as compared with those
of the full length peptides, as expected comparing the mean
hydrophobicity calculated for each peptide (Table 1) [27].
All peptides were obtained in the linear and oxidized form.
Oxidation was carried out incubating linear peptides with DMSO
(10%) at pH 6.0. LC-MS analysis of the peptides, showing for the
majority of the peptides a decrease of the retention time and a
reduction of 2 units in the mass, confirmed that the reaction
occurred (Figure 2).
Circular dichroism
Circular dichroism spectra were recorded at 25 ꢁC using a
1 cm quartz cell with the Jasco-810 spectropolarimeter using a
260–200 nm measurement range, 100 nm/min scanning speed,
1 nm bandwidth, 4 s response time, 0.5 nm data pitch.
Peptides concentration for CD measurement was 25 mM. CD
spectra were registered in 10 mM sodium phosphate buffer, pH
7 and in 10 mM sodium phosphate, 20 mM SDS buffer, pH 7.
Results
Peptide Design and Synthesis
The SolyC peptide (Table 1) contains the g core sequence of the
tomato defensin Solyc07g007760 and shows strong similarity in
the primary sequence with the levomeric form of porcine prote-
grin PG-1. In PG-1, four cysteines form two disulfide bonds, one
involving C6 and C15 and the other C8 and C13, which force
the peptide to fold into two antiparallel beta sheets; the oxida-
tion of cysteines induces the formation of a turn characterized
by a strong concentration of positive charges [24,25].
SolyC contains only three cysteines; therefore, to induce the
peptide to assume a PG-1-like structure, only one disulfide bond
can be formed. We designed two peptides with a different disul-
fide pattern, with the aim to explore the effect of cyclization on
the antimicrobial activity of the peptide, to understand whether
Table 1. Peptide sequences, names, reference number, and mean
hydrophobicity
Mean
hydrophobicity
Peptide sequence
Name
SolyC
Number
FSGGNCRGFRRRCFCTK
GNCRGFRRRCFCTK
FSGGNCRGFRRRCFSTK
GNCRGFRRRCFSTK
1
2
3
4
5
6
ꢂ0.71
ꢂ0.98
ꢂ0.91
ꢂ1.22
ꢂ0.91
ꢂ1.22
SolyC-t
SolyC1
SolyC1-t
SolyC2
SolyC2-t
FSGGNCRGFRRRSFCTK
GNCRGFRRRSFCTK
SolyC1-ox
SolyC1-t-ox
SolyC2-ox
SolyC2-t-ox
7
8
ꢂ0.91
ꢂ1.22
ꢂ0.91
ꢂ1.22
9
10
Figure 2. Comparison of the LC profiles for SolyC2 in the linear (A) and
oxidized form (B). Zoom between 4 and 6 min of the LC run.
wileyonlinelibrary.com/journal/jpepsci Copyright © 2013 European Peptide Society and John Wiley & Sons, Ltd. J. Pept. Sci. 2013; 19: 240–245

ANTIMICROBIAL PEPTIDES FROM PLANTS
Figure 3. CD spectra of SolyC2 (blue) and SolyC2-ox (red) (50 mM)
recorded in 10 mM phosphate buffer pH 7 (continuous line) and in
10 mM phosphate buffer, 30 mM SDS pH 7 (dotted lines).
Secondary Structure Studies
Linear and oxidized peptides were analyzed by CD at a 50 mM
concentration in phosphate buffer, pH7. Peptides are mostly in
an unordered form; spectra for the linear peptides, in fact, show
one minimum around 204 nm and a very shallow minimum
around 230 nm; upon oxidation, the minimum shifts toward
207 nm. Spectra recorded in the presence of SDS 30 mM are in
most cases very similar to those recorded in buffer but more
intense (Figure 3).
Antimicrobial Activity Test
The antimicrobial activity of the peptides was tested against the
Gram positive bacteria Staphylococcus aureus, Staphylococcus
epidermidis, Listeria monocytogenes and against the Gram nega-
tive bacteria H. pylori and Salmonella enterica (Table 2). For the
standard peptide SolyC, truncation results in loss of activity
against the Gram positive Staphilococci, whereas no difference
was found when the peptides SolyC and SolyC-t were tested
against Gram negative bacteria. When tested against Gram posi-
tive bacteria, the analogs containing two cysteines (SolyC1 and
SolyC2) show a reduced antimicrobial ability as compared with
the reference peptide SolyC; MIC values are in all cases higher
than that of SolyC, and there is no large difference between the
activity of linear and oxidized peptides. Peptides are in general
more active against Gram negative bacteria; linear peptides show
lower MIC in the full length version as compared with the
truncated form. Noteworthy, SolyC2, the peptide having the di-
sulfide between C6 and C15, shows activity identical to that of
the reference peptide. Oxidation results in an increase of the
Figure 4. Hemolytic activity of the peptides measured at different
concentrations.
antimicrobial activity of the modified peptides against Gram
negative bacteria, either in the extended or in the truncated
form: all oxidized peptides exhibit activity identical or compara-
ble with that of the reference peptide.
The hemolytic activity of the peptides was tested for all
peptides at several concentrations (Figure 4). Up to 90 mg/ml, all
peptides have hemolytic activity below 30%. At low
Table 2. Antimicrobial activity of the peptides against Gram positive and Gram negative bacteria
Strains
MIC100 (mg/ml)
Peptide number
1
2
3
4
5
6
7
8
9
10
Staphylococcus aureus
Staphylococcus epidermidis
Listeria monocytogenes
Salmonella serovar paratyphi B
Helicobacter pylori
40
40
80
15
15
50
80
80
15
15
50
100
80
50
100
80
80
100
80
80
100
80
80
100
80
50
100
80
100
100
80
80
100
80
40
80
15
40
15
15
15
20
15
15
15
20
15
15
20
20
J. Pept. Sci. 2013; 19: 240–245 Copyright © 2013 European Peptide Society and John Wiley & Sons, Ltd. wileyonlinelibrary.com/journal/jpepsci

AVITABILE ET AL.
Table 3. LC50 for the SolyC and SolyC analogs
Peptide number
1
2
3
4
5
6
7
8
9
10
LC50 (mg/ml)
275.2
266.7
212.3
228.5
226.5
213.0
229.6
231.0
197.4
235.1
concentrations, all long linear peptides and the oxidized form of
SolyC1 show hemolytic activity comparable with that of trun-
cated peptides. Interestingly, SolyC2 has in the long oxidized
form (SolyC2-ox) very low hemolytic activity, comparable with
that of the standard peptide.
LC50 measured for all peptides was found lower for the peptides
containing only two cysteines as compared with SolyC (Table 3).
The lowest LC50 value was observed for SolyC2 in the oxidized form
(SolyC2-ox).
Figure 5. Schematic representation of the peptide SolyC2-ox; in red is
the positively charged amino acids in the turn, in yellow is the cysteines
involved in the disulfide bond, and in green is the cysteine residue
mutated into serine.
Discussion
Peptides containing a g core motif and, in particular, analogs of
protegrins have widely been studied as they are capable to exert
antimicrobial activity even at high salt concentrations, unlike
other antimicrobial peptides that form sheets such as b defensins
[28,29]. Interestingly, the PG-1 analog IB-367 has been tested
in phase II clinical trials for the treatment of oral mucositis, and
PG-1-based peptidomimetics were found to be active at nano-
molar concentrations against Pseudomonas aeruginosa [30,31]. It
has been reported that the stabilization of the hairpin is
necessary for the activity: PG-1 analogs lacking cysteines, in
which the hairpin is stabilized by a tryptophan zipper motif,
possess antimicrobial activity comparable with PG-1 [32]. In PG-1,
the number of disulfide bonds and their position affect the
activity; analogs of PG-1 with a single disulfide bond possess
antimicrobial activity comparable with PG-1 in the linear form.
The g core of the tomato defensin family, to which SolyC pep-
tides belong, has a strong similarity in the amino acid sequence
with the g core of other plant defensins as the Medicago sativa
defensin 4 (MtDef4), and also with the levomeric form of PG-1
[9,33]. The three-dimensional structures of plant defensins, which
are composed of about 50–60 amino acids, are very complex, and
cysteines belonging to the g core form disulfide bonds with
cysteines located outside of the g core [34–36]. In self-consistent
g cores as PG-1 and RTD-1, instead, the beta-hairpin structure is
stabilized by two disulfide bonds between cysteines inside the
g core. SolyC is a fragment of a tomato defensin and, as most
of the plant defensins, has a very conserved GXCRG motif within
the g core with three cysteines. It is reasonable to think that SolyC
has a structural organization similar to that of plant defensin in
which cysteines in the g core are not connected between them.
As we have demonstrated in a previous paper that the linear
fragment SolyC possess high antimicrobial activity, we now
explored the possibility of stabilizing the peptide structure
introducing in a controlled fashion one disulfide bond and
evaluated the effects of cyclization on the peptide antimicrobial
activity and secondary structure. Furthermore, similar studies
were carried out on truncated peptides.
buffer containing SDS: both linear and oxidized peptides seem to
be in an unordered form (Figure 3). All the analogs are more ac-
tive against Gram negative than against Gram positive bacteria;
these results are in line to what has been reported for PG-1 and
PG-1 analogs with a single disulfide bond [28]. Interestingly, cyclic
peptides are generally more active than linear ones against Gram
negative bacteria. Cyclization induces a strong polarization of the
peptides, confining the positive charges in the turn and the hydro-
phobic residues on the ‘tails’ and reasonably allows the peptide to
interact with the bacterial LPS, neutralizing and permeating
it
(Figure 5). Recent studies aimed to characterize the interaction of
rationally designed peptides with the LPS demonstrate that amphi-
patic peptides are active antimicrobials as they assume
the structure of a b boomerang upon interaction with LPS, exert-
ing their activity by disaggregating it [37]. These peptides are
mimics of the outer membrane, as they possess a polar head
and hydrophobic tails and easily intercalate into LPS. The SolyC
cyclic analogs are already in the active conformation with the
segregation of charges, and therefore, it is likely that they exert
their activity as the b boomerang peptides do.
SolyC2 in the full length and truncated version shows antimi-
crobial activity identical to that of the reference peptide SolyC.
These result may be interpreted hypothesizing that, in the
experimental conditions employed for the activity test, SolyC
spontaneously oxidizes, and the formation of the disulfide bond
between Cys in position i and i + 9 is favored. As in SolyC2, only
the disulfide bridge between Cys i and i + 9 can be formed, it is
reasonable to think that the linear peptide spontaneously
converts into the oxidized form that corresponds also to the
active form. This hypothesis is consistent with the observation
that full length SolyC2 in the linear and oxidized form has the
same antimicrobial activity as the reference SolyC. Interestingly,
in SolyC2, the cysteines involved in the disulfide bond are located
in positions identical to external cysteines of PG-1, reproducing
the cyclization pattern of the highly active ‘bullet’ peptide pro-
posed by Harwig [28].
The observation that there is no difference in the activity
against Gram negative bacteria of cyclic SolyC1 and SolyC2
derivatives, which differ in the number of amino acids included
in the cycle, suggests that the charge segregation and not the
dimension of the cycle is crucial for activity.
In SolyC, cysteines are located in position i, i + 7, and i + 9; pep-
tides with disulfide bonds between Cys i – i + 7 (SolyC1) and
i ꢂ i + 9 (SolyC2) were obtained. The secondary structure of all
peptides was analyzed by CD in phosphate buffer and in phosphate
wileyonlinelibrary.com/journal/jpepsci Copyright © 2013 European Peptide Society and John Wiley & Sons, Ltd. J. Pept. Sci. 2013; 19: 240–245

ANTIMICROBIAL PEPTIDES FROM PLANTS
genes, exfoliative toxin genes, and toxic shock syndrome toxin 1
gene. J. Clin. Microbiol. 1998; 36: 2548–2553.
All peptides at concentrations up to threefold that of MIC show
very low hemolytic activity, indicating that these peptides are selec-
tive for bacterial outer membrane. The lowest percentage of hemo-
lytic activity was found for the oxidized SolyC2. LC50 calculated for
all peptides on murine red blood cells is lower for SolyC2, confirming
the selectivity for negatively charged membrane (as the bacterial LPS)
versus zwitterionic membranes (as the erythrocytes).
In conclusion, we have demonstrated that isolated g cores of
plant defensins possess strong antimicrobial activity against Gram
negative bacteria; formation of an intramolecular disulfide bond
stabilizes the peptides in the ‘active’ conformation. We found that
the peptide SolyC2, which reproduces the disulfide bond pattern
observed in PG-1, shows a high selectivity towards bacterial outer
membrane and possess very low hemolytic activity at concentra-
tions that are threefold the MIC. These results encourage future
studies on the antimicrobial activity of isolated g core peptides
from plants due to their strength and specificity of action.
18 Hong Y, Liu T, Lee MD, Hofacre CL, Maier M, White DG, Ayers S,
Wang L, Berghaus R, Maurer JJ. Rapid screening of Salmonella
enterica serovars Enteritidis, Hadar, Heidelberg and Typhimurium us-
ing a serologically-correlative allelotyping PCR targeting the O and H
antigen alleles. BMC Microbiol. 2008; 8: 178.
19 Tomasini ML, Zanussi S, Sozzi M, Tedeschi R, Basaglia G, De Paoli P.
Heterogeneity of cag genotypes in Helicobacter pylori isolates from
human biopsy specimens. J. Clin. Microbiol. 2003; 41: 976–980.
20 Bubert A, Hein I, Rauch M, Lehner A, Yoon B, Goebel W, Wagner M.
Detection and differentiation of Listeria spp. by a single reaction
based on multiplex PCR. Appl. Environ. Microbiol. 1999; 65: 4688–4692.
21 Martineau F, Picard FJ, Roy PH, Ouellette M, Bergeron MG. Species-
specific and ubiquitous DNA-based assays for rapid identification of
Staphylococcus epidermidis. J. Clin. Microbiol. 1996; 34: 2888–2893.
22 Capparelli R, Romanelli A, Iannaccone M, Nocerino N, Ripa R, Pensato S,
Pedone C, Iannelli D. Synergistic antibacterial and anti-inflammatory
activity of temporin A and modified temporin B in vivo. PLoS One
2009; 4: e7191.
23 Vinhal Costa Orsine J, Vinhal da Costa R, Carvalho da Silva R, de Fatima
Menezes Almeida Santos M, Carvalho Garbi Novaes MR. The acute
cytotoxicity and lethal concentration (LC50) of Agaricus sylvaticus
through hemolytic activity on human erythrocyte. Intl. J. Nutr.
Metabol. 2012; 4: 19–23.
24 Fahrner RL, Dieckmann T, Harwig SS, Lehrer RI, Eisenberg D, Feigon J.
Solution structure of protegrin-1, a broad-spectrum antimicrobial
peptide from porcine leukocytes. Chem. Biol. 1996; 3: 543–550.
25 Mani R, Cady SD, Tang M, Waring AJ, Lehrer RI, Hong M. Membrane-
dependent oligomeric structure and pore formation of a beta-hairpin
antimicrobial peptide in lipid bilayers from solid-state NMR. Proc. Natl.
Acad. Sci. U.S.A. 2006; 103: 16242–16247.
References
1 Ganz T. Defensins: antimicrobial peptides of innate immunity. Nat.
Rev. Immunol. 2003; 3: 710–720.
2 Zanetti M, Gennaro R, Romeo D. Cathelicidins: a novel protein family
with a common proregion and a variable C-terminal antimicrobial
domain. FEBS Lett. 1995; 374: 1–5.
3 Schittek B, Hipfel R, Sauer B, Bauer J, Kalbacher H, Stevanovic S, Schirle
M, Schroeder K, Blin N, Meier F, Rassner G, Garbe C. Dermcidin: a novel
human antibiotic peptide secreted by sweat glands. Nat. Immunol.
2001; 2: 1133–1137.
26 Yount NY, Yeaman MR. Emerging themes and therapeutic prospects
for anti-infective peptides. Annu. Rev. Pharmacol. Toxicol. 2012; 52:
337–360.
4 Castro MS, Fontes W. Plant defense and antimicrobial peptides.
Protein Pept. Lett. 2005; 12: 13–18.
5 Stec B. Plant thionins--the structural perspective. Cell. Mol. Life Sci.
2006; 63: 1370–1385.
27 Kyte J, Doolittle RF. A simple method for displaying the hydropathic
character of a protein. J. Mol. Biol. 1982; 157: 105–132.
6 Chiche L, Heitz A, Gelly JC, Gracy J, Chau PT, Ha PT, Hernandez JF,
Le-Nguyen D. Squash inhibitors: from structural motifs to macrocyclic
knottins. Curr. Protein Pept. Sci. 2004; 5: 341–349.
7 Yeaman MR, Yount NY. Unifying themes in host defence effector
polypeptides. Nat. Rev. Microbiol. 2007; 5: 727–740.
8 Yount NY, Andres MT, Fierro JF, Yeaman MR. The gamma-core
motif correlates with antimicrobial activity in cysteine-containing
kaliocin-1 originating from transferrins. Biochim. Biophys. Acta 2007;
1768: 2862–2872.
9 Yount NY, Yeaman MR. Multidimensional signatures in antimicrobial
peptides. Proc. Natl. Acad. Sci. U.S.A. 2004; 101: 7363–7368.
10 Yeaman MR, Yount NY, Waring AJ, Gank KD, Kupferwasser D, Wiese R,
Bayer AS, Welch WH. Modular determinants of antimicrobial activity
in platelet factor-4 family kinocidins. Biochim. Biophys. Acta 2007;
1768: 609–619.
11 Pelegrini PB, Franco OL. Plant gamma-thionins: novel insights on the
mechanism of action of a multi-functional class of defense proteins.
Int. J. Biochem. Cell Biol. 2005; 37: 2239–2253.
12 Steinberg DA, Hurst MA, Fujii CA, Kung AH, Ho JF, Cheng FC, Loury DJ,
Fiddes JC. Protegrin-1: a broad-spectrum, rapidly microbicidal peptide
with in vivo activity. Antimicrob. Agents Chemother. 1997; 41: 1738–1742.
13 Tran D, Tran PA, Tang YQ, Yuan J, Cole T, Selsted ME. Homodimeric
theta-defensins from rhesus macaque leukocytes: isolation, synthesis,
antimicrobial activities, and bacterial binding properties of the cyclic
peptides. J. Biol. Chem. 2002; 277: 3079–3084.
28 Harwig SS, Waring A, Yang HJ, Cho Y, Tan L, Lehrer RI. Intramolecular
disulfide bonds enhance the antimicrobial and lytic activities of
protegrins at physiological sodium chloride concentrations. Eur. J.
Biochem. 1996; 240: 352–357.
29 Goldman MJ, Anderson GM, Stolzenberg ED, Kari UP, Zasloff M,
Wilson JM. Human beta-defensin-1 is a salt-sensitive antibiotic in lung
that is inactivated in cystic fibrosis. Cell 1997; 88: 553–560.
30 Mosca DA, Hurst MA, So W, Viajar BS, Fujii CA, Falla TJ. IB-367, a
protegrin peptide with in vitro and in vivo activities against the micro-
flora associated with oral mucositis. Antimicrob. Agents Chemother.
2000; 44: 1803–1808.
31 Srinivas N, Jetter P, Ueberbacher BJ, Werneburg M, Zerbe K,
Steinmann J, Van der Meijden B, Bernardini F, Lederer A, Dias RL,
Misson PE, Henze H, Zumbrunn J, Gombert FO, Obrecht D, Hunziker
P, Schauer S, Ziegler U, Kach A, Eberl L, Riedel K, DeMarco SJ, Robinson
JA. Peptidomimetic antibiotics target outer-membrane biogenesis in
Pseudomonas aeruginosa. Science 2010; 327: 1010–1013.
32 Lai JR, Huck BR, Weisblum B, Gellman SH. Design of non-cysteine-
containing antimicrobial beta-hairpins: structure-activity relationship
studies with linear protegrin-1 analogues. Biochemistry 2002; 41:
12835–12842.
33 Sagaram US, Pandurangi R, Kaur J, Smith TJ, Shah DM. Structure-
activity determinants in antifungal plant defensins MsDef1 and
MtDef4 with different modes of action against Fusarium graminearum.
PLoS One 2011; 6: e18550.
14 The tomato genome sequence provides insights into fleshy fruit
evolution. Nature 2012; 485: 635–641.
15 Rigano MM, Romanelli A, Fulgione A, Nocerino N, D’Agostino N, Avitabile
C, Frusciante L, Barone A, Capuano F, Capparelli R. A novel synthetic
peptide from a tomato defensin exhibits antibacterial activities against
Helicobacter pylori. J. Pept. Sci. 2012; 12: 755–762.
34 Janssen BJ, Schirra HJ, Lay FT, Anderson MA, Craik DJ. Structure of
Petunia hybrida defensin 1, a novel plant defensin with five disulfide
bonds. Biochemistry 2003; 42: 8214–8222.
35 Fant F, Vranken W, Broekaert W, Borremans F. Determination of the
three-dimensional solution structure of Raphanus sativus antifungal
protein 1 by 1H NMR. J. Mol. Biol. 1998; 279: 257–270.
16 Romanelli A, Moggio L, Montella RC, Campiglia P, Iannaccone M,
Capuano F, Pedone C, Capparelli R. Peptides from Royal Jelly: studies
on the antimicrobial activity of jelleins, jelleins analogs and synergy
with temporins. J. Pept. Sci. 2011; 17: 348–352.
17 Becker K, Roth R, Peters G. Rapid and specific detection of toxigenic
Staphylococcus aureus: use of two multiplex PCR enzyme immunoas-
says for amplification and hybridization of staphylococcal enterotoxin
36 Fant F, Vranken WF, Borremans FA. The three-dimensional solution
structure of Aesculus hippocastanum antimicrobial protein 1 determined
by 1H nuclear magnetic resonance. Proteins 1999; 37: 388–403.
37 Bhunia A, Mohanram H, Domadia PN, Torres J, Bhattacharjya S.
Designed beta-boomerang antiendotoxic and antimicrobial peptides:
structures and activities in lipopolysaccharide. J. Biol. Chem. 2009;
284: 21991–22004.
J. Pept. Sci. 2013; 19: 240–245 Copyright © 2013 European Peptide Society and John Wiley & Sons, Ltd. wileyonlinelibrary.com/journal/jpepsci