Peptides 30 (2009) 839848

Contents lists available at ScienceDirect

Peptides
journal homepage: www.elsevier.com/locate/peptides

Structure and function of a custom anticancer peptide, CB1a

Jiun-Ming Wu b,c,1, Pey-Shynan Jan a,1, Hui-Chen Yu a, Hsu-Yuang Haung a,
Huey-Jen Fang b, Yuan-I Chang b, Jya-Wei Cheng c, Hueih Min Chen a,b,*
a

Nano Biomemdical and MEMS Technology Division, National Nano Device Laboratories, Hsinchu, Taiwan
Agricultural Biotechnology Research Center, Academia Sinica, Taipei, Taiwan
c
Institute of Biotechnology and Department of Life Sciences, National Tsing Hua University, Hsinchu, Taiwan
b

A R T I C L E I N F O

A B S T R A C T

Article history:
Received 10 November 2008
Received in revised form 2 February 2009
Accepted 3 February 2009
Available online 13 February 2009

Several natural antimicrobial peptides including cecropins, magainins and melittins have been found to
kill cancer cells. However, their efcacy may not be adequate for their development as anticancer agents.
In this study, we used a natural antimicrobial peptide, cecropin B (CB), as a template to generate a novel
anticancer peptide. Cecropin B is an amphipathic and polycationic peptide derived from the hemolymph
of Hyalophora cecropia with well-known antimicrobial and cytolytic properties. The signature pattern of
cecropins is W-x-(0,2)-[KDN]-x-{L}-K-[KRE]-[LI]-E-[RKN] (PROSITE: PS00268), and this signature
sequence is located at N-terminus of CB. CB1a was constructed by repeating the N-terminal ten amino
acids of CB three times and including a hinge near C-terminus. The circular dichroism spectra showed
that CB1a is unstructured in aqueous solution, but adopt a helical conformation in membrane-like
environment. The solution structure of CB1a in a polar solvent was also studied by NMR. CB1a formed a
helixhingehelix in 20% HFIP solution, and it was found the bent angle between two helical segments
was induced ranging from 608 to 1108. A heparin-binding motif is located in the central part of helix 1.
Isothermal titration calorimetry reveals the association constant of CB1a bound to low molecular weight
heparin is 1.66  105 M1 at physiological ionic strength at 25 8C. Binding of CB1a to heparin produces a
large conformational change toward a more structural state. CB1a demonstrated promising activity
against several cancer cells but low toxicity against non-cancer cells. The IC50 of CB1a on leukemia and
stomach carcinoma cells were in the range of 28-fold lower than those of CB. Besides, CB1a exhibited
low hemolytic activity against human red blood cells. Due to these properties, CB1a has the potential to
become a promising anticancer agent.
2009 Elsevier Inc. All rights reserved.

Keywords:
Cecropin B
Antimicrobial peptide
Anticancer activity
Heparin-binding motif
CD
NMR solution structure
ITC

1. Introduction
Hundreds of natural peptides have been found to show
antimicrobial properties that can kill a wide spectrum of Grampositive and Gram-negative bacteria, protozoa and fungi and can
potentially act as antibiotics [20,21,23]. Most of these antimicrobial peptides (AMPs) have cationic and amphipathic properties

* Corresponding author at: Nano Biomemdical and MEMS Technology Division,

National Nano Device Laboratories, Hsinchu, Taiwan. Tel.: +886 3 572 6100x7712;
fax: +886 3 572 6109.
E-mail address: hmchen@ndl.org.tw (H.M. Chen).
1
These authors contributed equally.
Abbreviations: AMP, antimicrobial peptide; ATCC, American Type Culture Collection; CB, cecropin B; CLSM, confocal laser scanning microscopy FACSuorescenceactivated cell sorter; FITC, uorescein isothiocyanate; HFIP, 1,1,1,3,3,3-hexauoroisopropanol; HPLC, high pressure liquid chromatography; HSPG, heparan sulfate
proteoglycans; ITC, isothermal titration calorimetry; MTT, 3-(4,5-dimethylthiazol2-yl)-2,5-diphenyl tetrazolium bromide; NMR, nuclear magnetic resonance; RBC,
red blood cell; HFIP, 1,1,1,3,3,3-hexauoro-2-propanol.
0196-9781/$ see front matter 2009 Elsevier Inc. All rights reserved.
doi:10.1016/j.peptides.2009.02.004

[22,32] that allow interaction with bacterial cytoplasmic membrane. Some of AMPs which have also been found to have the
ability to kill cancer cells, include cecropins [13,51], magainins
[3,14,35,59], melittins [43,44], and human LL-37 [16,29].
Cecropins are a family of antimicrobial peptides which are
widely found in the immune hemolymph of Hyalophora cecropia
[6,27,28]. They were constructed/synthesized of 3439 amino
acids and have high sequence homology [51,54]. The sequences of
natural cecropins have basic residues in their N-terminal segments
and hydrophobic residues in their C-terminal segments. Among
the cecropin family (cecropins A, B, D, E and F), cecropin B (CB)
possesses the strongest antimicrobial activity [27]. Derivatives of
CB, cecropin B1 (CB1) and cecropin B3 (CB3), were created to
investigate the effects on cells and synthetic liposomes [11,55].
CB1 was constructed by replacing the C-terminal segment with Nterminal sequence of CB, and CB3 was constructed by replacing the
N-terminal segment with C-terminal sequence of CB. Previous
studies showed that CB and its analogues can disrupt membranes
[11,12,55,56], and some of them have the ability to kill cancer cells

840

J.-M. Wu et al. / Peptides 30 (2009) 839848

[13,49,50]. CB1 retained the membrane lytic activity of its natural

origin [11] and had better anticancer properties against several
cancer cell lines than that of CB or CB3 [13,50]. This is an indication
that the N-terminal sequence of CB was more important than the
C-terminal sequence from a functional point of view. Furthermore,
the signature pattern of cecropins is W-x-(0,2)-[KDN]-x-{L}-K[KRE]-[LI]-E-[RKN] (PROSITE: PS00268) and it is located near the
N-terminus of CB. The N-terminal ten amino acids, KWKVFKKIEK,
is the ngerprint region of cecropin B.
An articially designed peptide CB1a having 33 amino acids with
a net +12 charge was developed. It was constructed by repeating the
original N-terminal ten amino acids of CB, KWKVFKKIEK, three times
and retaining a conserved hinge sequence (Ala-Gly-Pro) of cecropins
between the second and the third repeat. The proline residue used
here in kink section might be responsible for ion channel gating and
oligomerization [10,52] of peptides. The solution structures of the
natural cecropin A [26], cecropin P1 [46] and our previously
developed custom peptides CB1 [49] and CB3 [50] had been
determined by nuclear magnetic resonance (NMR) spectroscopy.
With the exception of cecropin P1 (from pig intestine), which has a
linear helical structure, all other peptides mentioned above were
found to have a common helixhingehelix motif. Although the
mechanism by which the cationic antimicrobial peptides cause cell
death are not yet fully understood, many reports have shown that
peptides may interact with lipid bilayers of cell membrane and
consequently lead to cell death by membranous pore perforation or
by a carpet-like manner [36,60].
CB and CB1 are highly active against cancer cells [13,49], in
contrast to CB3 that does not possess signicant anticancer activity
[50]. In this study, we focused on the structural properties and the
anticancer activities of CB1a, which was found to exhibit good
anticancer activity against several cancer cells but little cytotoxic
effects against non-cancer cells when compared with CB. In
addition, the hemolytic effect of CB1a on red blood cells is weaker
than that of CB and CB1. These properties might contribute to CB1a
becoming a promising drug of the future.
2. Materials and methods
2.1. Materials
Cell culture media and fetal bovine serum (FBS) were purchased
from Gibco (Rockville, MD). The RPMI-1640 medium, sodium
bicarbonate, hydrochloride acid and sodium hydroxide were
products of Sigma. Penicillin and streptomycin were purchased
from Gibco (Rockville, MD). Both 1,1,1,3,3,3-hexauoroisopropanol (HFIP) and heparins (3 kDa) were purchased from Sigma (St.
Louis, MO). Sodium 3-(trimethyl silyl)[2,2,3,3-2H] propionate
(99.5% purity) and deuterated water (D2O) (99.9% purity) were
purchased from Cambridge Isotope Laboratory (Andover, MA,
USA). Deuterated hexauoroisopropanol (d2-HFIP; 98% purity) was
purchased from Aldrich while polycarbonate lters with 0.4 mm
pores were products of Avanti Polar Lipids, Inc. (Alabaster, AL). 3(4,5-dimethylthiazol-2-yl)-2,5-diphenyl tetrazolium bromide
(MTT) was obtained from Sigma (St. Louis, MO). The cancer cells
CCRF-CEM, HL-60, AGS, NCI-H520, NCI-H661, broblast cells 3T3
and RPMI-7666 were purchased from the American Type Culture
Collection (ATCC; USA). Water used in these experiments was
deionized and distilled.
2.2. Preparation of peptides
Preparations of natural peptides, cecropin B (CB), melittin and
magainin II and customized peptides, CB3, CB1 and CB1a used
throughout this study were synthesized [13,56] by an Applied
Biosystems (ABI) peptide synthesizer. Fmoc chemistry was applied

with HBTU/HOBT coupling. The nal products were de-protected

and cleaved from the resin using a solution containing triuoroacetic
acid, water, phenol, thioanisole and ethanedithiol. Resin was
removed by ltration and the peptides were precipitated with
diethyl ether after the organic solvents were evaporated. The crude
peptides were desalted on Sephadex G-10 (20% acetic acid) and
puried using reverse phase high pressure liquid chromatography
(HPLC) (Vydac C-18 column, 0.1% TFA in H2Oacetonitrile). The
purity of the peptides was determined by mass spectrometry. A
small organic molecule, uorescein isothiocyanate (FITC), was
labeled by the conjugation of peptides via primary amines (i.e.,
lysines). Usually, between 3 and 6 FITC molecules were conjugated
to each peptide. The resulting reagents were compared for brightness in order to choose the optimal conjugation ratio. Fluorescein is
typically excited by 488 nm argon laser and the emission is collected
at 530 nm. Sequences of these synthetic peptides are shown in
Table 1. CB1a (net charge: +12) was designed by replacing the
segments of MGRNIRNGIVK and AIAVLGEAKAL of CB with
KWKVFKKIEK. The purity of peptides was determined by RP-HPLC;
all peptides were found to be >95% pure. These peptides were
lyophilized and stored at 20 8C. Prior to experiments, the peptides
were dissolved in PBS buffer at pH 7.4. The molecular weight of
peptides was investigated by mass spectrometry and both
theoretical mass and empirical mass were found to be nearly
identical. The concentration of peptide stock solutions was
determined from the net weight of peptides and their molecular
masses (the weight of the associated counter ions was not taken into
consideration). The concentration of peptides measured by the
method above was conrmed with a bicinchoninic acid assay (Micro
BCA protein assay, Pierce Chemical Co., Rockford, IL). A negligible
deviation between these two methods was observed.
2.3. Circular dichroism (CD) measurements
CD spectra of peptide solutions including HFIP or heparin in
5 mM phosphate buffer at pH 5.5 were obtained by using a Jasco J715 spectropolarimeter (Jasco, Easton, MD). Each spectrum (190
260 nm) consisted of an average of 4 scans using a water-jacketed
quartz cell of 1 mm path length at room temperature. Concentration of peptides dissolved in different solutions including HFIP (0
40%, v/v) and heparin was maintained at 20 mM. The scanning rate
of CD was set at 20 nm/min and the experimental conditions were:
step size 0.1 nm, 2 s response time and 1.0 nm bandwidth. The
content of secondary structure in the peptides was calculated by
CDPro software package [47,48]. Heparin titration CD experiment
was also recorded on Jasco J-715 spectropolarimeter. Besides, the
full length CD spectra were used to monitor the heparin-induced
conformational change at 215 nm (25 8C).
2.4. Nuclear magnetic resonance spectroscopy
Nuclear Overhauser effect spectroscopy (NOESY) and total
correlation spectroscopy (TOCSY) NMR experiments were perTable 1
Amino acid sequences of peptides shown in this study.
Peptide
a

CB
CB1aa,b
CB1
CB3
Melittin
Magainin II

Sequence

Net charge

KWKVFKKIEKMGRNIRNGIVKAGPAIAVLGEAKAL-NH2
KWKVFKKIEK KWKVFKKIEKAGPKWKVFKKIEK-NH2
KWKVFKKIEKMGRNIRNGIVKAGPKWKVFKKIEK-NH2
AIAVLGEAKALMGRNIRNGIVKAGPAIAVLGEAKAL-NH2
GIGAVLKVLTTGLPALISWIKRKRQQ-NH2
GIGKFLHSAKKFGKAFVGEIMNS-NH2

+7
+12
+11
+3
+5
+3

a
The underlined residues show the repeating sequence of the N-terminal
cecropin ngerprint segment of CB.
b
The sequence of heparin-binding motif in CB1a is in bold italics.

J.-M. Wu et al. / Peptides 30 (2009) 839848

formed on a Bruker Avance 600-MHz spectrometer at 303K. The

phase-sensitive double-quantum ltered correlation spectroscopy
(DQF-COSY) NMR was used in this study to elucidate the structure
of the compounds and the experiment was performed on a Bruker
Avance 500-MHz spectrometer at 303K. Samples were dissolved in
500 ml of 20% d-HFIP with H2O/D2O (95:5) at pH 4.9 to a
concentration of 3 mM. NMR spectra were collected in the phasesensitive mode with time-proportional phase incrementation
(TPPI) [34]. NOESY spectra were acquired at three different mixing
times of 100, 150, and 200 ms. Solvent suppression of peptide was
conducted by pre-saturation of the solution in the relaxation and
mixing periods. The NOESY spectra were collected with 2048
points in the t2 dimension and 512 points in the t1 dimension
respectively. Each t1 increment was segmented with 16 scans with
2.0 s relaxation interval. DQF-COSY spectra [39] were collected
with 2048 points in t2 and 512 points in t1 which contained 16
scans in it. TOCSY spectra [4] were recorded using the MLEV17
pulse sequence with mixing times of 60 and 90 ms at 2048 points
in t2 and 512 points in t1. Distance restraints were acquired from
NOEs, which are assigned by a 2D NOESY spectrum with a 150-ms
mixing time. The NOE cross-peak intensities were calibrated using
aromatic cross peaks from Trp-2 and classied as strong, medium
or weak corresponding to the distance limits of 1.82.8, 1.83.6
and 1.85.0 A, respectively. Backbone dihedral angle restraints
were derived from 3JHNHa coupling constants measured from line
shape analysis of anti-phase cross-peak splitting in the DQF-COSY
spectrum. Angles were restrained to 1208  40 for 3JHNHa > 9 Hz
and to 608  30 for 3JHNHa < 5 Hz. H/D exchange NMR experiments
were carried out in 20% d-HFIP in D2O at pH 5.5 sodium phosphate
buffer. The highly protected proton was taken into account after 12 h
exchange.
2.5. Structure calculation
Structure calculations were carried out by using the programs
XPLOR-NIH 2.1.09 [42] on a Linux platform. Peptide template
structures were generated with extended forms by using the
program Insight II 950 and then calculated and rened with the
simulated annealing (SA) protocol. All force constants and
molecular parameters were set to their default values (due to
the original sa.inp and rene.inp protocols) except for the time
step, which was decreased to 0.001 ps throughout the calculations.
Simulated annealing was performed using 10,000 steps at 1500 K
and 5000 steps by gradually cooling to 100 K. The structures
generated by simulated annealing procedures were rened using
SA renement. The temperature in rene.inp was set to 1000 K and
gradually cooled to 100 K in 20,000 steps. Finally, these structures
were energy minimized using 500 steps of Powell energy
minimization. The nal 20 energy lowest structures contained
no distance constraint violations greater than 0.2 A and no dihedral
angle constraint violations greater than 58. Structural visualization
and analysis were performed using MOLMOL [30] and PROCHECKNMR [31].
2.6. Isothermal titration calorimetry (ITC)
ITC experiments were performed by using a VP-ITC isothermal
titration calorimeter (MicroCal Inc., Northhampton, MA) [57]. All
experiments were conducted at 25 8C in sodium phosphate buffer
with 150 mM NaCl at pH 7.4. The solutions were degassed under
vacuum for 10 min prior to experiment. The peptide solution was
placed in the 1.43 ml reaction cell and titrated with the heparin
delivered from a 250 ml motor-driven syringe. The cell contents
were stirred continuously at 300 rpm during the experiment. The
ITC measurements were carried out by injections of heparin
suspension into the sample cell containing the peptides (35.8

841

71.9 mM). In control experiments, the corresponding heparin

solution was injected into buffer without peptide. The heats of
dilution of heparin were subtracted from the heats determined in
the corresponding peptide-heparin-binding experiments. Data
were analyzed with the Origin version 7.0 software (MicroCal).
2.7. Cell culture
Anchorage-independent non-cancer cells such as lymphoblast
RPMI-7666 cells, leukemia cell lines (HL-60 and CCRF-CEM) and
lung cancer cell lines (NCI-H520 and NCI-H661) were maintained
in RPMI-1640 culture media containing 10% FBS and 1% penicillin
streptomycinneomycin (PSN) antibiotic mixtures (Gibco BRL,
Rockville, MD). Tissue cell line (3T3) and stomach carcinoma cell
line (AGS) were cultured in DMEM medium. These Cell lines were
seeded in 25 cm2 ask (Falcon, BD Biosciences) and cultured at
37 8C in the presence of 5% CO2 for two days. Afterward, the cells
were centrifuged at 1000 rpm for 10 min. After washing with cell
medium, the resultant pellets were resuspended in the medium
and cultured in 75 cm2 ask at 37 8C overnight. The cells were
grown to about 80% full of the ask and were used for later
experiments.
2.8. Cytotoxicity assays
Each of the prepared cell suspensions above was adjusted to
contain 1  105 cells/ml before they were transferred to a 96-well
plate (90 ml/well). The suspension in each well was mixed with
10 ml of a culture medium containing peptides freshly prepared
from 500 mM stock solutions to nal concentrations varying
between 0, 1, 5, 10, 15, 20, 30, 50, 100 and 200 mM of peptides. Each
sample was collected in triplicate. After incubating for 48 h, a MTTbased colorimetric assay was conducted (MTT can be metabolized
by mitochondrial dehydrogenases in metabolically active cells to
form a formazan salt with strong absorption at 570 nm). The assay
was performed by adding 10 ml, 5 mg/ml MTT into each well
containing sample cells. In order to solubilize formazan crystals
formed after 4 h incubation, a solution of 100 ml containing 10%
sodium dodecyl sulfate (SDS, a detergent) was added to each well
and re-incubated the mixture at 37 8C overnight. The absorbance at
570 nm (A570nm) of the mixture in each well was measured by
using a Bio-Rad model 450 microtiter plate reader. An average cell
survival rate was determined based on DA570nm ([sample with
peptide]A570nm  [control]A570nm). The survival rates obtained
were plotted against peptide concentrations and the survival
curve was used to calculate the IC50 value (the concentration of
peptide of which the cell survival is 50%).
2.9. Confocal laser scanning microscopy (CLSM)
Adherent lung cancer cells NCI-H520 and NCI-H661 were
maintained in RPMI-1640 supplemented with 10% (v/v) FBS and 1%
penicillinstreptomycinneomycin antibiotic mixtures (Gibco
BRL, Rockville, MD). Cells (8  105 cells/ml NCI-H520 or 2  105
cells/ml NCI-H661) were cultured on glass cover-slips overnight
under 5% CO2 at 37 8C. The uoresced peptides were then added to
the cell cultures and incubated an additional hour. The concentrations of the N-terminal FITC-conjugated CB1a in the culture
medium was 30 mM. Following incubation, the medium was
removed. The cells were subsequently washed three times with
PBS without any xation procedure [40]. The stained cells were
then mounted with 80% glycerol. Fluorescence images were
captured with a Zeiss LSM5 PASCAL confocal laser scanning
microscope (Carl Zeiss, Oberkochen, Germany) with an Axioskop 2
uorescence microscope (Carl Zeiss, Oberkochen, Germany).
Optical excitation was carried out with 488 argon laser beam

842

J.-M. Wu et al. / Peptides 30 (2009) 839848

and 543 HeNe laser beam for FITC-labeled peptide and iso-electric
point (PI), respectively. Images were processed with Adobe
Photoshop 6.0.
2.10. Fluorescence-activated cell sorter (FACS) analysis
NCI-H520 and NCI-H661 cells grown on a 60-mm dish were
incubated with 10 mM FITC-CB1a at 4 8C or at 37 8C for 30 min.
After the incubation, cells were washed with PBS solution and were
treated with trypsin for 10 min [40]. The cells were centrifuged at
1000 rpm for 10 min and then washed with PBS solution. After
washing, the cells were resuspended in 500 ml PBS, ltered
through a 30 mm pore size lter, and analyzed by using
FACSCalibur (Becton Dickinson, Franklin Lakes, NJ). FITC uorescence was detected in FL1 channel (ex 488 nm/em 530 nm).
2.11. Hemolysis assay
Fresh venous blood was stored in an EDTA bottle. The plasma was
removed by centrifugation at 15,000 g for 5 min. The pellet of red
blood cells (RBCs) was washed with sterile Dulbeccos phosphate
buffer saline (D-PBS) four times. The stock solutions of peptides were
diluted to 1, 10, 25, 50, 150 and 200 mM with D-PBS buffer in an
Eppendorf tube (200 ml each containing 5 ml RBC solution). The
solutions were then centrifuged at 3000 g for 5 min. The supernatant
was collected. The amount of hemoglobin released from the dead
RBCs killed by peptides or anticancer agents in the solutions was
determined by measuring their absorption at A414nm. Zero hemolysis
(blank) and 100% hemolysis were determined in D-PBS buffer and 1%
Triton-X 100, respectively. Based on these two reference points, the
percentages of hemolysis of human RBCs at various concentrations
of peptides were obtained.
3. Results
3.1. Circular dichroism measurements
CD spectra were used to study the content of peptide secondary
structures. The conformational behavior of CB1a and other CB
analogues in aqueous buffer and HFIP/water solution was
investigated by using a CD spectrometer. Fig. 1A presents the
Far-UV CD spectra of CB1a at room temperature in HFIP/water
mixtures at various ratios (v/v, from 0% to 40%). The results showed
that CB1a had the highest helical content in 40% HFIP/H2O solution.
At between 15% and 30% of HFIP/H2O, the CD spectra of CB1a retain
a very similar shape to that of CB1a in 40% of HFIP/H2O. In aqueous
solution, the CD spectrum of CB1a showed a negative band around
200 nm, which is typical for the back bond of a peptide in a random
coil conformation. These observations indicated that signicant
negative ellipticities of CB1a at 208 nm and 222 nm in the presence
of lipid-mimetic solvent HFIP and that the peptide had folded into
an a-helical conformation. This conformational change from a
random-coil in aqueous buffer to an a-helix structure in
membrane-mimetic environments is common for many membrane-binding peptides [5,33]. In contrast, in solution containing
less than 15% of HFIP/H2O, the shapes of CD spectra were
dramatically similar to that of CB1a in aqueous solution (0% of
HFIP/H2O). Fig. 1B presents the CD spectra of CB analogues in 40%
HFIP. The results indicated that CB3 has the highest helical
propensity. In the same condition, comparing to CB, CB1a and CB1,
CB3 has higher helical content.
3.2. NMR analysis
Two-dimensional 1H NMR spectroscopy was used to study
structure of CB1a in polar solvent d-HFIP. Both TOCSY and NOESY

Fig. 1. (A) CD spectra of CB1a in hexauoroisopropanol (HFIP). CD spectra of CB1a

were obtained at various ratios of HFIP/water (v/v) in solution (0%, 5%, 10%, 15%,
20%, 30% and 40%). (B) CD spectra of cecropin B and its analogues in 40% HFIP. The
concentration of the peptides used was maintained at 20 mM.

spectra were collected for CB1a at 303 K, pH 4.9. Spin systems for
each amino acid were identied based on a TOCSY spectrum with a
mixing time of 90 ms. Sequential assignment of the residues was
achieved using the NOESY spectrum obtained with a mixing time
of 150 ms. The amide resonance of the N-terminal lysine was not
observed. Fig. 2A represents the NOSEY spectrum of CB1a in 20% dHFIP.
a
a
b
The riched H (i)-HN(i + 3) and H (i)-H (i + 3) NOE connectivities denote the a-helical structures. The available daN (i,i + 4) NOE
connectivities are shown for residues 421 of CB1a. These NOE
connectivities indicate the presence of a regular a-helical
conformation. Summaries of the sequential and medium-range
NOEs observed for CB1a in 20% d-HFIP are depicted in Fig. 2B.
Distance constraints were derived by the NOEs and were used to
carry out the structure determination. Deviations from the
reference random coil values for the a-protons chemical shifts
also provided information on the conformational status of a
peptide or protein. The a-proton chemical shift index (CSI) [58]
result strongly supports the helical conformation of the peptide.
There are two helical segments spanning residues 421 and 2532.
When H/D exchange NMR experiments were performed, some
slow exchange amide protons were observed in the TOCSY spectra.
The high protected protons can be obtained after 12 h H/D
exchange. These amide protons form hydrogen bonds with the
backbone carbonyl groups, which were also used as constraints in
the structure calculations.

J.-M. Wu et al. / Peptides 30 (2009) 839848

843

Fig. 3. Calculated structures of CB1a. 20 superimposed structures of CB1a are

aligned by two sections which are helix 1 from residues 421, and helix 2 from
residue 2532. The diagrams were generated in MOLMOL [30]. (A) and (C) are
aligned by helix 1, (B) and (D) are aligned by helix 2.

Fig. 2. NOESY spectrum and NOE connectivities of CB1a. (A) NOESY spectrum of HNa
H connectivities. Spectrum was recorded on 600 MHz NMR at 150-ms mixing time
of CB1a (3 mM) in 20% d-HFIP at 30 8C. (B) Schematic representations of short-range
and medium-range interactions of NOEs observed for CB1a in 20% d-HFIP. NOE
intensities are indicated by the thickness of the bars. Slow exchange amides are
represented as lled circles. The highly protected proton (slow exchange) was taken
into account after 12 h exchange. The small (J < 6.0 Hz) coupling constants are
represented as lled circles.

3.3. Structure calculation

The solution structure of CB1a was calculated using 366 total
NOE constraints derived from the information in the NOESY
spectra of the peptide obtained at various mixing times (100, 150
and 200 ms). 54 dihedral angle constraints were obtained from
DQF-COSY NMR spectrum. Structure calculations were carried out
by Xplor-NIH 2.1.09 [42] using simulated annealing procedure
based on NOE distance and dihedral angle constraints. The 20
lowest energy conformers were selected to undergo renement.
The nal 20 structures contained no distance constraint violations
greater than 0.2 A and no dihedral angle constraint violations
greater than 58. The 20 overlapping structures of CB1a show that
the helical conformation is discontinuous. Fig. 3 shows that the 20
superimposed structures are aligned by two sections which are
denoted as helix 1 and helix 2. It is not possible to t helix1 and
helix2 at the same time due to the high exibility in the hinge
region (positions 2224).

The 20 nal structures resulted in a RMSD about the mean

coordinate positions for helix 1 (residues 421) of 0.384 A for the
backbone atoms and 1.513 A for the heavy atoms. The 20
structures aligned over helix 2 (residues 2532) with a RMSD of
0.346 A for the backbone and 1.248 A for the heavy atoms.
Structural statistics and the root mean square deviations for the 20
lowest energy structures of CB1a are shown/given in Table 2. The
20 overlapping structures of CB1a depicted that the helical
conformation is discontinuous. Gly-22, Pro-23 and Lys-24 are in
the hinge region, and the helixhingehelix structure of CB1a is
very similar to the structure of cecropin A [26]. Based on the NOE
connectivities of CB1a, Lys-24 is in the edge of the C-terminal
helical section; however, Lys-24 is only partially incorporated in
hinge region.
Unlike the structures of CB1 and CB3, the bent angle of CB1a
between two helical segments is not converged due to no longrange NOEs (ji  jj > 5) observed between helix 1 (residues 421)
and helix 2 (residues 2532). According to the determined
structures, CB1a was found to have a bent angle from 608 to
1108 between two helical segments (helix angles were calculated
by MOLMOL). The bent angle of CB1a is wider than that of CB1 [49]
and CB3 [50]. The Ramachandran plot computed by PROCHECKNMR shows that most of the residues fall in the allowed
conformational regions. Residues that fall in the disallowed
regions of the plot correspond mostly to the disordered regions
of the structure of CB1a. The coordinates of the CB1a had been
deposited at the Protein Data Bank (PDB ID code: 2IGR).
3.4. Heparin-induced secondary structure change of CB1a
Conformational change in heparin solution was monitored by
CD. Upon the addition of heparin (20 mM), the CD spectra of CB1a

J.-M. Wu et al. / Peptides 30 (2009) 839848

844
Table 2
Summary of structural constraints and structure statistics.
NOE constraints
Intraresidues (ji  jj = 0)
Sequential (ji  jj = 1)
Medium range (2 2 ji  jj24)
Long range (ji  jj > 4)
Hydrogen bond constraints
Dihedral angles
Total constraint numbers

79
145
126
0
16
54
420

Structural Statistics (20 Structures)

NOE violations, number > 0.2 A
Dihedral angel violations, number > 58

0
0

RMSD for geometrical analysis

Bond lengths (A)
Bond angles (degree)
Impropers (degree)

0.00235  0.000042
0.412  0.003
0.270  0.006

RMSD from experimental constraints

Distance (A)
Dihedral angles (deg)
Mean total energy (kcal mol1)

0.0347  0.0005
0.398  0.051
54.340  0.806

Atomic RMSD for structured region

Helix 1 (421)
Backbone
Heavy atoms

0.384
1.306

Helix 2 (2532)
Backbone
Heavy atoms

0.346
1.513

Ramachandran statistics
Most favoured region (%)
Additionally allowed (%)
Generously allowed (%)
Disallowed (%)

87.4
11.4
0.7
0.5

changed from a strong negative band near 198 nm (Fig. 4A, solid
line) to a negative band centered at 215 nm (Fig. 4A, dash line). This
is an indication that the CD spectrum of CB1a shifted from a
random coil (in aqueous solution) to an a/b structure (in heparin
solution). Further experiments were conducted to study the
conformational changes exhibited by CB1a in the presence of
heparin by systematically increasing the content of heparin in
solution. A three-phase shift of ellipticity at 215 nm was observed
when heparin was mixed with CB1a at different heparin
concentrations. The results indicated that the secondary structure
of the peptide changed between the two stages.

cell and titrated with heparin solution containing 150 mM NaCl

using a motor-driven syringe. The binding constants determined
for CB1a and CB bound to heparin were 1.66  105 M1 and
7.53  103 M1, respectively. The binding strength between CB1a
and heparin is over 20-fold bigger than that of CB.

3.5. Isothermal titration calorimetry analysis

3.6. Anticancer activity assays

To quantify the heparin-binding afnity, the isothermal

titration calorimetry experiments were carried out. The data on
heat enthalpy versus molar ratio (heparin/CB1a) of which curve
was tted with a theoretical prediction implied a single set model
of identical heparin-binding sites on the peptide. Fig. 5 displays the
ITC data determined when CB1a or CB was placed in the reaction

Cytotoxicity assays were conducted on cancer cells as well as

non-cancer cells and the IC50 of CB1a, CB, melittin, and magainin II
and the cancer chemotherapeutic agent doxorubicin (DOX) are
summarized in Table 3. As shown in Table 3, CB1a displayed higher
cytotoxic activity (i.e., a lower IC50 value) than CB and magainin II
against leukemia cells and stomach carcinoma. The cytotoxic

Fig. 4. Heparin-induced conformational change of CB1a. (A) CD spectra of 20 mM

CB1a with (dash line) and without (solid line) 20 mM heparin. (B) CD experiment of
CB1a titrating with heparin was monitored at 215 nm by CD.

Table 3
IC50a values of various peptides and doxorubicin on different cell lines.
Peptide designation

CB1a
CB
Melittin
Magainin II
DOX
a

Cancer cells

Non-cancer cells

Stomach carcinoma

Leukemia cells

Lymphoblast

Tissue cell

Lung cell

AGS

HL-60

CCRF-CEM

NCI-H520

Lung cancer cells

NCI-H661

RPMI-7666

3T3

WI-38

5.6  0.5
>50
0.6  0.1
28.7  5.0
4.1  0.4

6.7  1.1
18.5  1.7
0.8  0.1
38.3  3.2
0.2  0.0

4.4  0.3
11.7  1.3
0.5  0.1
32.9  0.9
0.4  0.1

22.7  0.2
58.8  2.0
14.1  0.7
46.9  2.7
<10

35.9  0.2
>100
2.6  0.2
>100
<10

>100
>200
4.5  0.2
> 50
1.9  0.1

>50
>50
6.0  0.1
>50
2.0  0.7

>100
>100
5.8  0.5
>100
<10

The results are the average of at least three experiments and mean deviations are shown. The unit used for the IC50 value is mM.

J.-M. Wu et al. / Peptides 30 (2009) 839848

845

Fig. 5. ITC titration of CB1a and CB with heparin. Titration calorimetry of (A) CB1a and (B) CB with heparin solution in phosphate buffer containing 150 mM NaCl at 25 8C. Top
panel shows the experimental trace of heat energy absorbed per injection. Bottom panel indicates the integrated heat energy for each injection. The heat energy of dilution
was subtracted from the control for calculation of the binding isotherm.

activities of CB1a against AGS and leukemia cell lines are about 8fold and 23 fold stronger than that of CB, respectively. Melittin
showed cytotoxic effects against cancer cells as well as non-cancer
cells. CB1a and CB are highly potent against cancer cells but not
against normal mammalian cells. DOX showed good activity
against both cancer cells and normal mammalian cells.

efciency at 4 8C or at 37 8C suggested that an endosomal pathway

was involved in the cellular uptake [24]. The kind of endocytosis
was also an evidence of heparin sulfate proteoglycans (HSPGs)mediating cell-penetrating process.

3.7. Confocal images and ow cytometry analysis of peptide uptake

Effects of CB, CB1, CB1a and melittin on human RBCs were also
examined (Fig. 8). It was found that CB1a exhibited less hemolytic
activity than melittin, CB, CB1 and doxorubicin. As shown in Fig. 8,
only about 18% of RBCs were lysed even after being treated with
CB1a at a concentration between 100 mM and 200 mM. While,
about 100%, 57%, 55%, and 18% of RBCs were lysed after being
treated by melittin, CB, CB1 and CB1a, respectively, at concentration of 200 mM.

Fig. 6 represents the confocal image results of cancer cells

treated with FITC-CB1a peptide. NCI-H661 (panels A and B) and NCIH520 (panels C and D) cells were treated with 30 mM FITC-CB1a and
observed without xation. Cells were incubated with FITC-CB1a for
5 min (panels A and C) and for 1hr (panels B and D). The images
indicated that CB1a stayed lodged onto the cell membrane in the
beginning. After 1 h, the peptide will translocate into the cancer
cells. Confocal images revealed that FITC-conjugated CB1a nally
present in both the cytoplasm and nucleus of cells.
The results of FACS analysis on trypsin-treated cells are present
in Fig. 7. NCI-H661 and NCI-H520 cells were incubated with FITCCB1a (10 mM) at 4 8C or at 37 8C for 30 min and treated with
trypsin before FACS analysis [40]. The difference in translocation

3.8. Hemolytic activity assays

4. Discussion
4.1. Structural characterization of CB1a
Using the PROSITE data base for sequence analysis [7,15,45],
where the signature pattern of cecropins is W-x(0,2)-[KDN]-x-{L}-

Fig. 6. Confocal images of cancer cells treated by CB1a. NCI-H661 (panels A and B) and NCI-H520 (panels C and D) cells were treated with 30 mM FITC-CB1a and observed
without xation by confocal laser scanning microscopy (scale = 20 mm). Cells were incubated with FITC-CB1a for 5 min (panels A and C) and for 1 hr (panels B and D). The
images indicated that CB1a stayed attached to the cell membrane in the beginning. After treated the cancer cells with FITC-CB1a for 1 hr, the peptide penetrated into the cell.
FITC-conjugated CB1a nally present in both the cytoplasm and nucleus of cells (panels B and D). The results show that CB1a has the ability to translocate into the cancer cells.

846

J.-M. Wu et al. / Peptides 30 (2009) 839848

Fig. 7. Cellular uptake analyzed by FACS. NCI-H661 (A) and NCI-H520 (B) cells were incubated with FITC-CB1a (10 mM) at 4 8C (blue curves) or at 37 8C (red curves) for 30 min,
and treated with trypsin before FACS analysis [40]. The black curves correspond to cells incubated in the absence of the peptide. (For interpretation of the references to color in
this gure legend, the reader is referred to the web version of the article.)

K-[KRE]-[LI]-E-[RKN], we found that the signature of cecropins is

located in the N-terminal segment of CB as shown in Table 1. Our
custom designed peptide CB1a was therefore obtained by
repeating CBs N-terminal of ten residues (Am = KWKVFKKIEK)
three times. Between the second and the third repeats, a kink
segment Ala-Gly-Pro remained (e.g., CB1a = Am1-Am2-(Ala-GlyPro)-Am3, where Am1 = Am2 = Am3). The solution NMR structure
of CB1a in membrane-mimetic solvent conrmed that it contains
two a-helices with a kink between them. As shown in Fig. 3, a
discrete structure (Ala-Gly-Pro-Lys) exists in CB1a. This discrete
fragment may provide CB1a with the high exibility. The bent
angle between two helical segments is 601108, and it is more
exible than that of CB1 (1001108) [49] and CA (701008) [26].
The CD results showed that CB1a is unstructured in bulk
solution, but adopts a helical conformation in a polar solvent. This
kind of conformational transition was also observed for other CB
analogues. The secondary structure propensity in 40% HFIP
solution followed the order CB3 > CB > CB1aCB1. Furthermore,
heparin can induce the formation of a secondary structure in CB1a.

The secondary structure content of the peptide has been analyzed

by using the CDPro program package and the results showed that
the helical content of CB1a was 10.87% and 25.22% in bulk solution
and 1:1 molar ratio of heparin, respectively. However, CB did not
show such signicant secondary structure change.
4.2. Heparin-binding property of CB1a
The ITC results (Fig. 5) showed that the heparin-binding afnity
of CB1a is about 20-fold greater than that of CB at physiological
ionic strength. The binding constant of CB1a-heparin is
1.66  105 M1, and the afnity is almost similar when compared
with heparan sulfate binding to melittins [19]. Consensus of
heparin-binding sequences XBBXBX and XBBBXXBX were previously described [8,25], where B represents a basic amino acid and
X indicates any other amino acid. In this study, we intentionally
designed a pattern EKKWKV for CB1a (see Table 1) expectedly
tting with the pattern XBBXBX above. The motif is located in the
N-terminal helix of CB1a. Previous studies reported that the
XBBXBX motif was in cooperation with a b-strand conformation in
heparin binding [9]. Our CD results (Fig. 4) showed that the b-sheet
content of CB1a in heparin is 13.64% while the helical propensity of
CB1a in heparin is 22.73%. This kind of conformational change was
also observed for the melittin analogue [19]. A previous study has
shown that cell-penetrating peptides (CPPs), like (Arg)9 and Tat
[17,53], requires cell surface heparan sulfate proteoglycans (HSPG)
for cellular internalization. The internalization of melittin is likely
to involve HSPG on the cell surface. Interestingly, the confocal
image indicated that CB1a do possess the ability to translocate
across cancer cell membrane (see Figs. 6 and 7). The endocytic
uptake process may relate to the heparin-binding property of CB1a.
4.3. Anticancer and hemolytic activity

Fig. 8. Hemolytic measurements. Hemolysis (%) of red blood cells versus the
hemolytic agents concentrations (mM) are shown for various lytic peptides such as
CB (open square), CB1 (open circle), CB1a(up-open triangle), Mel (melittin; doneopen triangle), and chemotherapeutic agent, Dox (open diamond).

Ghiselli et al. previously conducted animal experiments with CB

in the blood and reported that CB could enhance beta-lactam activity
and cure Gram-negative septic shock in rats [18]. This result implies
that antimicrobial peptides can function in vivo (i.e., in the blood
system). Compared with CB, CB1a was observed to reduce hemolytic
activity by about 18% at 200 mM compared to the 57% activity of CB
(see Fig. 8). This indicated that CB1a is less hemolytic than CB in vivo.
It also suggested that CB1a could be used intravenously in the future.
Compared with the other anticancer peptides or organic compound
agents studied in this work, CB1a exhibited as a competitive
candidate with clinical-used anticancer drug, doxorubicin, due to its
lower hemolytic effect. CB1a exhibited relatively weak activity

J.-M. Wu et al. / Peptides 30 (2009) 839848

847

Fig. 9. Helical wheel diagrams of N-terminal helices of CB1a and CB. The hydrophobic residues of CB1a (A) and CB (B) are shown in black box. CB has a more complete
amphipathic segment than CB1a does.

against non-cancer cells (Table 3). This high selectivity suggests that
CB1a has the potential to become a promising anticancer agent.
Generally, the anticancer activity of CB analogues against leukemia
cell lines followed the order CB1a > CB1 > CB > CB3. According to
the CD result (Fig. 1B), the helical propensity of CB1a and CB1 is less
than that of CB3 and CB indicating that the secondary structure
content is not important for anticancer activity. The positive charge
of the peptide may play a crucial role in the anticancer effect. Since
the net charge of CB1a, CB1 and CB are +12, +11 and +7, respectively.
The positive charge of anticancer peptides may allow stronger
binding to the negative charged membrane of cancer cells (e.g.,
phosphatidylserine) [37,38]. This is the reason why CB1a is more
effective against cancer cells than non-cancer cells. However, the
mechanism of action of cationic peptide is still not clear.
Previous studies indicated that heparin-binding peptides show
antimicrobial activity [1]. However, the current custom peptide,
CB1a, showed no signicant bacteriocidal effect with the comparison to CB (MIC of CB1a is about 128 mg/ml on E. coli ATCC 25922
compared with MIC of CB is 1 mg/ml). The amphiphatic distribution
may play a key role. The diagrams of helical wheels [41] of CB1as and
CBs N-terminal helices show that the amphiphatic character and the
hydrophilic face of CB1a are discrete (Fig. 9). The reduction in
amphipathicity and discrete hydrophilic face may partially explain
why CB1a has poor antimicrobial activity. Moreover, CB1a lacks the
amino acid sequence of the second section (MGRNIRNGIVK) of Nterminal helix in CB that might be the main enhancer of
antimicrobial activity. The complete helix 1 of CB seems to be
required for breaking the bacterial membrane. Previous studies also
report similar observations for cecropin A [2,26].
To summarize, we have constructed a novel anticancer peptide
CB1a modied from natural antimicrobial peptide CB and
determined its anticancer activity and structural characters. The
polycationic property of CB1a demonstrated promising activity
against several cancer sell lines with low toxicity to the non-cancer
cells. In addition, the hemolytic side effect had been reduced. Due
to these properties, CB1a has the potential to become a promising
anticancer agent.
References
[1] Andersson E, Rydengard V, Sonesson A, Morgelin M, Bjorck L, Schmidtchen A.
Antimicrobial activities of heparin-binding peptides. Eur J Biochem 2004;271:
121926.
[2] Andreu D, Merrield RB, Steiner H, Boman HG. N-terminal analogues of
cecropin A: synthesis, antibacterial activity, and conformational properties.
Biochemistry 1985;24:16838.
[3] Baker MA, Maloy WL, Zasloff M, Jacob LS. Anticancer efcacy of Magainin2 and
analogue peptides. Cancer Res 1993;53:30527.

[4] Bax A, Davis DG. MLEV-17 based two-dimensional homonuclear magnetization transfer spectroscopy. J Magn Reson 1985;65:355.
[5] Bechinger B. Structure and functions of channel-forming peptides: magainins,
cecropins, melittin and alamethicin. J Membr Biol 1997;156:197211.
[6] Boman HG, Wade D, Boman IA, Wahlin B, Merrield RB. Antibacterial and
antimalarial properties of peptides that are cecropin-melittin hybrids. FEBS
Lett 1989;259:1036.
[7] Bucher P, Bairoch A. A generalized prole syntax for biomolecular sequence
motifs and its function in automatic sequence interpretation. Proc Int Conf
Intell Syst Mol Biol 1994;2:5361.
[8] Capila I, Linhardt RJ. Heparin-protein interactions. Angew Chem Int Ed Engl
2002;41:391412.
[9] Cardin AD, Weintraub HJ. Molecular modeling of protein-glycosaminoglycan
interactions. Arteriosclerosis 1989;9:2132.
[10] Chang DK, Cheng SF, Trivedi VD, Lin KL. Proline affects oligomerization
of a coiled coil by inducing a kink in a long helix. J Struct Biol 1999;128:
2709.
[11] Chen HM, Clayton AH, Wang W, Sawyer WH. Kinetics of membrane lysis by
custom lytic peptides and peptide orientations in membrane. Eur J Biochem
2001;268:165969.
[12] Chen HM, Leung KW, Thakur NN, Tan A, Jack RW. Distinguishing between
different pathways of bilayer disruption by the related antimicrobial peptides
cecropin B, B1 and B3. Eur J Biochem 2003;270:91120.
[13] Chen HM, Wang W, Smith D, Chan SC. Effects of the anti-bacterial peptide
cecropin B and its analogs, cecropins B-1 and B-2, on liposomes, bacteria, and
cancer cells. Biochim Biophys Acta 1997;1336:1719.
[14] Cruciani RA, Barker JL, Zasloff M, Chen HC, Colamonici O. Antibiotic magainins
exert cytolytic activity against transformed cell lines through channel formation. Proc Natl Acad Sci USA 1991;88:37926.
[15] Falquet L, Pagni M, Bucher P, Hulo N, Sigrist CJ, Hofmann K, et al. The PROSITE
database, its status in 2002. Nucleic Acids Res 2002;30:2358.
[16] Frohm M, Agerberth B, Ahangari G, Stahle-Backdahl M, Liden S, Wigzell H, et al.
The expression of the gene coding for the antibacterial peptide LL-37 is
induced in human keratinocytes during inammatory disorders. J Biol Chem
1997;272:1525863.
[17] Fuchs SM, Raines RT. Pathway for polyarginine entry into mammalian cells.
Biochemistry 2004;43:243844.
[18] Ghiselli R, Giacometti A, Cirioni O, Mocchegiani F, Orlando F, DAmato G, et al.
Cecropin B enhances betalactams activities in experimental rat models of
gram-negative septic shock. Ann Surg 2004;239:2516.
[19] Goncalves E, Kitas E, Seelig J. Structural and thermodynamic aspects of the
interaction between heparan sulfate and analogues of melittin. Biochemistry
2006;45:308694.
[20] Hancock RE. Cationic antimicrobial peptides: towards clinical applications.
Expert Opin Investig Drugs 2000;9:17239.
[21] Hancock RE. Cationic peptides: effectors in innate immunity and novel antimicrobials. Lancet Infect Dis 2001;1:15664.
[22] Hancock RE, Chapple DS. Peptide antibiotics. Antimicrob Agents Chemother
1999;43:131723.
[23] Hancock RE, Patrzykat A. Clinical development of cationic antimicrobial peptides: from natural to novel antibiotics. Curr Drug Targets Infect Disord
2002;2:7983.
[24] Henriques ST, Melo MN, Castanho MA. Cell-penetrating peptides and antimicrobial peptides: how different are they? Biochem J 2006;399:17.
[25] Hileman RE, Fromm JR, Weiler JM, Linhardt RJ. Glycosaminoglycan-protein
interactions: denition of consensus sites in glycosaminoglycan binding
proteins. Bioessays 1998;20:15667.
[26] Holak TA, Engstrom A, Kraulis PJ, Lindeberg G, Bennich H, Jones TA, et al. The
solution conformation of the antibacterial peptide cecropin A: a nuclear
magnetic resonance and dynamical simulated annealing study. Biochemistry
1988;27:76209.

848

J.-M. Wu et al. / Peptides 30 (2009) 839848

[27] Hultmark D, Engstrom A, Bennich H, Kapur R, Boman HG. Insect immunity:

isolation and structure of cecropin D and four minor antibacterial components
from Cecropia pupae. Eur J Biochem 1982;127:20717.
[28] Hultmark D, Steiner H, Rasmuson T, Boman HG. Insect immunity. Purication
and properties of three inducible bactericidal proteins from hemolymph of
immunized pupae of Hyalophora cecropia. Eur J Biochem 1980;106:716.
[29] Johansson J, Gudmundsson GH, Rottenberg ME, Berndt KD, Agerberth B.
Conformation-dependent antibacterial activity of the naturally occurring
human peptide LL-37. J Biol Chem 1998;273:371824.
[30] Koradi R, Billeter M, Wuthrich K. MOLMOL: a program for display and analysis
of macromolecular structures. J Mol Graph 1996;14(515):2932.
[31] Laskowski RA, Rullmannn JA, MacArthur MW, Kaptein R, Thornton JM. AQUA
and PROCHECK-NMR: programs for checking the quality of protein structures
solved by NMR. J Biomol NMR 1996;8:47786.
[32] Lehrer RI, Ganz T. Antimicrobial peptides in mammalian and insect host
defence. Curr Opin Immunol 1999;11:237.
[33] Maloy WL, Kari UP. Structure-activity studies on magainins and other host
defense peptides. Biopolymers 1995;37:10522.
[34] Marion D, Wuthrich K. Application of phase sensitive two-dimensional correlated spectroscopy (COSY) for measurements of 1H-1H spin-spin coupling
constants in proteins. Biochem Biophys Res Commun 1983;113:96774.
[35] Ohsaki Y, Gazdar AF, Chen HC, Johnson BE. Antitumor activity of magainin
analogues against human lung cancer cell lines. Cancer Res 1992;52:35348.
[36] Oren Z, Shai Y. Mode of action of linear amphipathic alpha-helical antimicrobial peptides. Biopolymers 1998;47:45163.
[37] Papo N, Shai Y. New lytic peptides based on the D,L-amphipathic helix motif
preferentially kill tumor cells compared to normal cells. Biochemistry
2003;42:934654.
[38] Papo N, Shai Y. Host defense peptides as new weapons in cancer treatment.
Cell Mol Life Sci 2005;62:78490.
[39] Rance M, Sorensen OW, Bodenhausen G, Wagner G, Ernst RR, Wuthrich K.
Improved spectral resolution in cosy 1H NMR spectra of proteins via double
quantum ltering. Biochem Biophys Res Commun 1983;117:47985.
[40] Richard JP, Melikov K, Vives E, Ramos C, Verbeure B, Gait MJ, et al. Cellpenetrating peptides. A reevaluation of the mechanism of cellular uptake. J
Biol Chem 2003;278:58590.
[41] Schiffer M, Edmundson AB. Use of helical wheels to represent the structures of
proteins and to identify segments with helical potential. Biophys J 1967;7:
12135.
[42] Schwieters CD, Kuszewski JJ, Tjandra N, Clore GM. The Xplor-NIH NMR
molecular structure determination package. J Magn Reson 2003;160:6573.
[43] Sharma SV. Melittin resistance: a counterselection for ras transformation.
Oncogene 1992;7:193201.
[44] Sharma SV. Melittin-induced hyperactivation of phospholipase A2 activity and
calcium inux in ras-transformed cells. Oncogene 1993;8:93947.

[45] Sigrist CJ, Cerutti L, Hulo N, Gattiker A, Falquet L, Pagni M, et al. PROSITE: a
documented database using patterns and proles as motif descriptors. Brief
Bioinform 2002;3:26574.
[46] Sipos D, Andersson M, Ehrenberg A. The structure of the mammalian antibacterial peptide cecropin P1 in solution, determined by proton-NMR. Eur J
Biochem 1992;209:1639.
[47] Sreerama N, Woody RW. Estimation of protein secondary structure from
circular dichroism spectra: comparison of CONTIN, SELCON, and CDSSTR
methods with an expanded reference set. Anal Biochem 2000;287:25260.
[48] Sreerama N, Woody RW. On the analysis of membrane protein circular
dichroism spectra. Protein Sci 2004;13:10012.
[49] Srisailam S, Arunkumar AI, Wang W, Yu C, Chen HM. Conformational study of a
custom antibacterial peptide cecropin B1: implications of the lytic activity.
Biochim Biophys Acta 2000;1479:27585.
[50] Srisailam S, Kumar TK, Arunkumar AI, Leung KW, Yu C, Chen HM. Crumpled
structure of the custom hydrophobic lytic peptide cecropin B3. Eur J Biochem
2001;268:427884.
[51] Steiner H, Hultmark D, Engstrom A, Bennich H, Boman HG. Sequence and
specicity of two antibacterial proteins involved in insect immunity. Nature
1981;292:2468.
[52] Tieleman DP, Shrivastava IH, Ulmschneider MR, Sansom MS. Proline-induced
hinges in transmembrane helices: possible roles in ion channel gating. Proteins 2001;44:6372.
[53] Tyagi M, Rusnati M, Presta M, Giacca M. Internalization of HIV-1 tat requires
cell surface heparan sulfate proteoglycans. J Biol Chem 2001;276:325461.
[54] van Hofsten P, Faye I, Kockum K, Lee JY, Xanthopoulos KG, Boman IA, et al.
Molecular cloning, cDNA sequencing, and chemical synthesis of cecropin B
from Hyalophora cecropia. Proc Natl Acad Sci U S A 1985;82:22403.
[55] Wang W, Smith DK, Chen HM. The effect of pH on the structure, binding and
model membrane lysis by cecropin B and analogs. Biochim Biophys Acta
1999;1473:41830.
[56] Wang W, Smith DK, Moulding K, Chen HM. The dependence of membrane
permeability by the antibacterial peptide cecropin B and its analogs, CB-1 and
CB-3, on liposomes of different composition. J Biol Chem 1998;273:274384.
[57] Wiseman T, Williston S, Brandts JF, Lin LN. Rapid measurement of binding
constants and heats of binding using a new titration calorimeter. Anal Biochem 1989;179:1317.
[58] Wishart DS, Sykes BD. The 13C chemical-shift index: a simple method for the
identication of protein secondary structure using 13C chemical-shift data. J
Biomol NMR 1994;4:17180.
[59] Zasloff M. Magainins, a class of antimicrobial peptides from Xenopus skin:
isolation, characterization of two active forms, and partial cDNA sequence of a
precursor. Proc Natl Acad Sci U S A 1987;84:544953.
[60] Zasloff M. Antimicrobial peptides of multicellular organisms. Nature
2002;415:38995.