Anti-angiogenesis activities of novel peptide complexes: Mitochondria-disruptive 9MER peptides conjugated with the integrin alpha v beta 3-homing cyclic RGD

Article abstract of DOI:10.1271/bbb.120397

RGD peptides are popular drug delivery tools in treating integrin αVβ3-expressing malignant tumors and tumor vasculature cells. We investigated the specific delivery and pharmacological potential of enantiomeric mitochondria-disruptive peptides (RLYLRIGRR

Full text of DOI:10.1271/bbb.120397

Biosci. Biotechnol. Biochem., 76 (11), 2044–2048, 2012
Anti-Angiogenesis Activities of Novel Peptide Complexes:
Mitochondria-Disruptive 9mer Peptides Conjugated
with the Integrin alpha V beta 3-Homing Cyclic RGD Motif
y
2;
Takashi IWASAKI,1 Minoru YAMAKAWA,2 Ai ASAOKA,2 Tsuyoshi KAWANO,1 and Jun ISHIBASHI
1Faculty of Agriculture, Tottori University, 4-101 Koyama-Minami, Tottori 680-8553, Japan
2Insect Mimetics Research Unit, National Institute of Agrobiological Sciences, Tsukuba, Ibaraki 305-8634, Japan
Received May 17, 2012; Accepted August 7, 2012; Online Publication, November 7, 2012
[doi:10.1271/bbb.120397]
RGD peptides are popular drug delivery tools in
treating integrin ꢀVꢁ3-expressing malignant tumors
and tumor vasculature cells. We investigated the specific
delivery and pharmacological potential of enantiomeric
mitochondria-disruptive peptides (RLYLRIGRR-NH2,
RLRLRIGRR-NH2, ALYLAIRRR-NH2, and RLLLRIGRR-
NH2) after conjugation with an integrin ꢀVꢁ3-homing
peptide, cyclic pentameric RGD peptide. The cyclic
RGD-conjugated mitochondria-disruptive peptides ex-
hibited specific internalization, apoptosis induction, and
cytotoxicity against integrin ꢀVꢁ3-high-expressing
human umbilical vein endothelial cells. Our findings
indicate that these novel peptide complexes might prove
good anti-angiogenesis reagents.
DA (RLYLRIGRR-NH2), DB (RLRLRIGRR-NH2), DC
(ALYLAIRRR-NH2), and DD (RLLLRIGRR-NH2).8–11)
Their sequences were synthesized using D-amino acids
and were amidated in the C-terminus. All these D9mer
peptides were designed and synthesized on the basis of
insect AMPs, defensins, using D-amino acids and the C-
terminal amidated structure.12,13) They showed cationic
and amphipathic structures14) and have been reported to
have multiple functions, including antimicrobial, anti-
protozoan, and antitumor activities, on the basis of
disruption of negatively charged target membranes.8–10)
Furthermore, our latest study revealed that they can
disrupt the mitochondrial membrane by the mechanism
described above, i.e., they disrupt negatively charged
mitochondrial membranes by charge interactions.11)
These observations suggest that D9mer peptides are
good candidates for mitochondria-targeting strategies in
combination with a DDS. Hence a new peptide named
DC2 (ALYLAIRRRRRRRR-NH2) was designed and syn-
thesized by connecting DC to D-octa-arginine
(RRRRRRRR-NH2) via sharing of arginine residues in
the C-terminus to target mitochondria. Octa-arginine is a
major cell-penetrating peptide (CPP), and was reported
to show effective uptake into cells via interactions
between the cationic octa-arginine and anionic proteo-
glycans on the membrane surface.15) D-Octa-arginine
was also found to have effective membrane permeability
similar to L-octa-arginine,16) because arginine residues
take on the same net charges regardless of the presence
of the L- or D-amino acid. In fact, DC2 penetrated the
cell membrane, caused mitochondrial disruption, and
showed higher cytotoxicity against various cell lines
than DC,14) but DC2 showed no cytoselectivity, and
killed not only tumor cell lines but also a normal
fibroblastic cell line (MRC-5). These findings indicate
that octa-arginine is not a suitable DDS tool for D9mer
peptides due to its non-specific cell-penetrating proper-
ties, although mitochondria targeting is an effective
strategy for D9mer peptides.
Key words: antimicrobial peptide; cyclic RGD motif;
mitochondria; integrin ꢀVꢁ3; drug delivery
system
Mitochondria are organelles that play important roles
in, for example, energy metabolism, reactive oxygen
species production, intracellular calcium control, and
apoptosis regulation1) in all eukaryote cells. Hence,
mitochondria have drawn attention as a promising target
for drug discovery in combination with a drug delivery
system (DDS).2) In particular, antimicrobial peptides
(AMPs) are regarded as suitable materials for mitochon-
dria-targeting strategies. AMPs are short cationic pep-
tides that disrupt negatively charged bacterial mem-
branes by charge interactions.3,4) In a recent study, an
AMP-like D-peptide (KLAKLAKKLAKLAK-OH) was
found to disrupt mitochondria and to exhibit significant
cytotoxicity when it was delivered artificially into the
cytoplasm.5) These effects are considered to be attrib-
utable to negative charges in the mitochondrial outer
membrane. Mitochondria are posited to be derived from
aerobic bacteria in the endosymbiotic theory,6) and have
a negatively charged outer membrane similar to the
bacterial membrane.7) Therefore, cationic AMP-like
peptides are thought to interact with and to disrupt the
negatively charged mitochondrial membrane by charge
interactions.
In this study, we tried to control the cytoselectivity of
mitochondria-disruptive D9mer peptides using a target
cell-specific DDS. We focused on a DDS peptide, cyclic
Arg-Gly-Asp (cRGD) peptide, as a target cell-specific
transporter for D9mer peptides. The cRGD peptide is
Previously, we reported the synthesis and functions of
four AMP-like short peptides termed D9mer peptides,
y
To whom correspondence should be addressed. Fax: +81-29-838-6028; E-mail: bishiba@nias.affrc.go.jp
Abbreviations: AMP, antimicrobial peptide; CPP, cell-penetrating peptide; DDS, drug delivery system; FBS, fetal bovine serum; FITC, fluorescein
isothiocyanate; HUVEC, human umbilical vein endothelial cell; LDH, lactate dehydrogenase; TAMRA, tetramethyl rhodamine

Selective Disruption of Mitochondria in a Target Cell
2045
Integrin αVβ3 homing
Mitochondria-disruptive
D9mer peptides
known to translocate into cells expressing an extrac-
ellular target molecule, integrin ꢀVꢁ3, by binding to it
and inducing endocytosis in it.17) Integrin ꢀVꢁ3 is an
extracellular cell-adhesion protein complex that is
significantly overexpressed on vascular cells during
tumor angiogenesis, but is absent from the quiescent
endothelium and normal tissues.18) This indicates that it
can be a specific target for antitumor angiogenesis drugs.
Here we report the synthesis, cytoselectivity, and
cytotoxicity of integrin ꢀVꢁ3-targeting mitochondria-
disruptive peptides, cRGD-cys-D9mer peptides, against
various cell lines.
cRGD peptide
CRLYLRIGRR-NH2
CRLRLRIGRR-NH2
CALYLAIRRR-NH2
(cRGD-cys-DA)
F
(cRGD-cys-DB)
(cRGD-cys-DC)
(cRGD-cys-DD)
D
C
disulfide
G
R
bond CRLLLRIGRR-NH2
Fig. 1. Structures of cRGD-cys-D9mer Peptides.
cRGD-cys-D9mer peptides consist of two functional peptides: an
integrin ꢀVꢁ3-homing cRGD motif and mitochondria-disruptive
D9mer peptides combined with a disulfide bond. Italicized bold
characters indicate D-amino acids.
kidney cancer cells, RERF-LC-AI human lung cancer cells, HepG2
human liver cancer cells, and NIH-3T3 mouse fibroblasts) were
evaluated by flow cytometric analysis. Integrin ꢀVꢁ3 expressed on the
membrane surface was stained with an anti-integrin ꢀVꢁ3 antibody
conjugated with fluorescein isothiocyanate (FITC) (Bio Vision,
Mountain View, CA) and detected by flow cytometry (Epics XL-
MCL Flow Cytometry System; Beckman Coulter, Fullerton, CA). The
cells were collected and washed twice with FACS medium (1 mM
CaCl2, 0.5 mM MgCl2, 5% FBS, 0.5% bovine serum albumin, and
0.1% NaN3 in phosphate-buffered saline) by centrifugation at 300 ꢀ g
for 5 min. Next, 100 mL of antibody solution (25 mg/mL) was added to
the cells (2:0 ꢀ 105). After 30 min of incubation on ice, the cells were
washed twice with FACS medium, suspended in 1 mL of FACS
medium, and analyzed by flow cytometry within 2 h. The mean
fluorescence intensity (MFI) of more than 1 ꢀ 104 cells was deter-
mined, and the integrin ꢀVꢁ3 expression level was quantified for each
cell line.
Materials and Methods
Cell lines. Cos-1 African green monkey kidney cancer (Riken
Cell Bank, RCB0143), HepG2 human liver cancer (RCB1886), and
NIH-3T3 mouse fibroblastic (RCB2767) cell lines were cultured
in DMEM (Invitrogen, Carlsbad, CA). The RERF-LC-AI human
carcinoma (RCB0444) cell line was cultured in MEM (Invitrogen).
Human umbilical vein endothelial cells (HUVECs) (Lonza, Basal,
Switzerland) were cultured in EBM-2 (Lonza). All media contained
10% (v/v) fetal bovine serum (FBS).
Peptide synthesis, cyclization, and purification. D9mer peptides DA
(RLYLRIGRR-NH2), DB (RLRLRIGRR-NH2), DC (ALYLAIRRR-NH2),
and DD (RLLLRIGRR-NH2) and cys-D9mer peptides cys-DA, cys-DB,
cys-DC, and cys-DD were synthesized by a 9-fluorenylmethoxycar-
bonyl (Fmoc) solid-phase peptide synthetic method on Rink-amide-
MBHA resin (Merck Chemicals, Darmstadt, Germany) using an APEX
396 automatic peptide synthesizer (Advanced ChemTech, Louisville,
KT). Subsequently, the peptides were cleaved from the resin, and the
side chains were deprotected using a mixture of trifluoroacetic acid/
m-cresol/thioanisole/ethane-1,2-dithiol (35:2:2:1, v/v) for 4 h at room
temperature.
Cytotoxicity assay. The various cell lines were seeded onto 96-well
plates at a density of 8:0 ꢀ 103 cells/well in 100 mL of medium
containing various concentrations of peptides (12.5, 25, 50, 75, and
100 mM). The cells were incubated under 5% CO2 for 24 h at 37 ꢁC, and
then 10 mL of Cell Counting Kit-8 Reaction Solution (Dojindo,
Kumamoto, Japan) was added to each well. After 4 h of incubation, the
absorbance was measured at 450 nm using an automated plate reader
(Appliskan; Thermo Fisher Scientific, Waltham, MA). Cell viability
relative to the untreated cells was determined, and the cytotoxicities of
peptides were evaluated.
A side chain-protected linear RGD peptide, H-Asp(OtBu)-D-Phe-
Cys(Trt)-Arg(Pbf)-Gly-OH, was synthesized on HMB resin (Merck)
by the method described above and cleaved from the dried resin using
a mixture of dichloromethane/methanol/acetic acid (8:1:1, v/v) for 2 h
at room temperature. The mixture was evaporated, and the lyophilized
linear RGD peptide was dissolved in N,N-dimethylformamide (1 mg/
mL). Next, six equivalents of 2-(7-aza-1H-benzotriazole-1-yl)-1,1,3,3-
tetramethyluronium hexafluorophosphate (HATU) and six equivalents
N,N-diisopropylethylamine (DIPEA) were added to the peptide solu-
tion. After 24 h, cyclization was complete. The solvent was evaporated,
and the crude products were precipitated with deionized water. The
cyclized RGD peptide was deprotected under the conditions described
above.
Observation of internalization. The internalization of the cRGD-
cys-D9mer peptides into integrin ꢀVꢁ3-high-expressing and ꢀVꢁ3-
low-expressing cell lines was observed. Tetramethyl rhodamine
(TAMRA) (Invitrogen) was introduced to the amino group of cRGD-
cys-DC in 0.1 M sodium tetraborate (pH 8.5) over 6 h at room
temperature. After incubation, cRGD-cys-DC-TAMRA was purified
by the HP1100 HPLC system and reverse-phase PEGASIL ODS
column. Cos-1 cells and HUVECs were used as integrin ꢀVꢁ3-high-
expressing and ꢀVꢁ3-low-expressing cell lines respectively. Both
types of cultured cells were seeded onto 96-well plates at
8:0 ꢀ 103 cells/well at a final volume of 100 mL, and incubated with
25 mM cRGD-cys-DC-TAMRA under 5% CO2 for 24 h at 37 ꢁC.
Internalization of cRGD-cys-DC-TAMRA into the cytoplasm of Cos-1
cells and HUVECs was observed under a fluorescence microscope
(IX71; Olympus, Tokyo) by detection of rhodamine (excitation/
emission = 540/570 nm).
The crude products were precipitated with cold diethyl ether and
purified using an HP1100 HPLC system (Hitachi, Tokyo) with a
reverse-phase PEGASIL ODS column (Senshu-Pak, Tokyo). The
purified D9mer peptides, cys-D9mer peptides, and cRGD peptide,
cyclo(-Asp-D-Phe-Cys-Arg-Gly-), were characterized by matrix-assist-
ed laser desorption/ionization time-of-flight mass spectrometry (Auto-
flex III; Bruker Daltonics, Billerica, MA).
The lyophilized cRGD peptide was dissolved in 2-propanol/acetic
acid (1:1, v/v) and added to 3 mM 2,20-dithiodipyridine in the same
solution to introduce an S-pyridyl (S-pyr) group to the thiol group of
the cysteine residue in the cRGD peptide. The resulting cRGD(S-pyr)
peptide, cyclo[-Asp-D-Phe-Cys(S-pyr)-Arg-Gly-], was purified using
AKTA Explore (GE Healthcare, Little Chalfont, UK) and a Superdex
Peptide column (GE Healthcare), lyophilized, and resolved in 0.5 M
Tris(hydroxymethyl)aminomethane-HCl (pH 8.5). Two equivalents of
cys-D9mer peptides in 5 mM HCl were added dropwise to this solution
with gentle stirring. The combined peptides, cRGD-cys-D9mer
peptides (cRGD-cys-DA, DB, DC, and DD), were obtained after 24 h
at room temperature (Fig. 1). The final compounds were purified using
AKTA Explore and a Superdex Peptide column, and high-purity
peptides (theoretical purity, >95%) were prepared.
Mitochondrial membrane potential assay. The effects of cRGD-cys-
DC on the cytoplasmic mitochondria in Cos-1 cells and HUVECs were
determined by measuring changes in the mitochondrial membrane
potential. Cos-1 cells and HUVECs were seeded in 96-well plates at
5:0 ꢀ 104 cells/well at a final volume of 100 mL, and were incubated
with and without 50 mM cRGD-cys-DC under 5% CO2 for 24 h at 37 ꢁC.
As an indicator of mitochondrial membrane potential, 10 mL of JC-1
staining solution (Cayman Chemical, Ann Arbor, MI) was added to
each well. JC-1 spontaneously forms complexes (JC-1 J-aggregates)
and shows red fluorescence on healthy mitochondria with high
mitochondrial membrane potential, but it remains in monomeric form
(as JC-1 monomers) and shows green fluorescence on damaged
mitochondria with loss of mitochondrial membrane potential. After
Measurements of integrin ꢀVꢁ3 expression. The expression levels
of integrin ꢀVꢁ3 in the various cell types (HUVECs, Cos-1 monkey

2046
T. IWASAKI et al.
incubation under 5% CO2 for 30 min at 37 ꢁC, JC-1 signal colors of red
and green fluorescence were observed under the IX71 fluorescence
microscope by detection of red fluorescence (excitation/emission =
540/590 nm) and green fluorescence (excitation/emission = 485/
535 nm) respectively.
HUVEC
MFI
=18.1
Cos-1
100
50
100
50
a
b
MFI
=1.7
Cell membrane leakage assay and apoptosis assay. Cos-1 cells and
HUVECs were seeded onto 96-well plates at 8:0 ꢀ 103 cells/well in a
final volume of 100 mL and incubated with various concentrations of
cRGD-cys-DC (12.5, 25, and 50 mM) under 5% CO2 for 24 h at 37 ꢁC.
Subsequent assays were performed after peptide treatment.
0
0
100 101 102 103 104
100 101 102 103 104
RERF-LC-AI
HepG2
100
100
c
d
MFI
=1.3
MFI
=2.0
In the cell membrane leakage assay, cell membrane damage was
quantified by measuring the amounts of lactic dehydrogenase (LDH)
released from the cytoplasm using a Cytotoxicity Detection Kit
(Roche, Basel, Switzerland) following the manufacturer’s protocol.
Basal LDH release was determined by examination of untreated cells
and maximum LDH release was determined from cells treated with 2%
Triton X-100 (Sigma-Aldrich, St. Louis, MO) in the medium over
30 min. After treatment of the cells with peptides or Triton X-100, a
tetrazolium salt reaction mixture (100 mL/well) was added to each
supernatant and this was incubated for 30 min at room temperature.
The spectrophotometric absorbance of the colored formazan product,
which indicates the amount of LDH released from the cytoplasm, was
determined by automated plate reader at a wavelength of 490 nm.
In the apoptosis assay, apoptosis induction was determined by
measuring apoptosis indicator caspase 3/7 activity by Caspase-Glo 3/7
Assay (Promega, Madison, WI) according to the manufacturer’s
protocol. A mixture (100 mL/well) containing a luminogenic caspase
3/7 substrate with a caspase cleavage DEVD sequence was added to
each well of cells treated with various concentrations of the peptide.
After incubation for 30 min at room temperature, caspase 3/7 activity
was determined by measuring the luminescence signals produced from
the cleaved caspase 3/7 substrate.
50
50
0
0
100 101 102 103 104
100 101 102 103 104
NIH-3T3
100
e
MFI
=3.3
50
0
100 101 102 103 104
Log FITC fluorescence intensity
Fig. 2. Integrin ꢀVꢁ3-Dependent Internalization of cRGD-cys-
D9mer Peptides.
The expression levels of integrin ꢀVꢁ3 on the membrane surface
of five cell types (a, HUVECs; b, Cos-1 monkey kidney cancer cells;
c, RERF-LC-AI human lung cancer cells; d, HepG2 human liver
cancer cells; e, NIH-3T3 mouse fibroblasts) were determined by
FITC-labeled antibody staining and flow cytometric analysis. Blank
peaks indicate unstained cells and gray filled peaks indicate
antibody-stained cells. The expression levels of integrin ꢀVꢁ3 were
evaluated by the mean fluorescence intensity (MFI) ratio (stained
cells/unstained cells).
Results
Expression of integrin ꢀVꢁ3
Table 1. Cytotoxicities of D9mer, cys-D9mer, and cRGD-cys-D9mer
Peptides
To evaluate the extracellular expression levels of
integrin ꢀVꢁ3, a target molecule for the cRGD motif, we
stained various cell lines with a FITC-labeled anti-
integrin ꢀVꢁ3 antibody and quantified the expression of
it by flow cytometric analysis. The mean fluorescent
intensity (MFI) of FITC in stained and unstained cells
was determined, and the relative integrin ꢀVꢁ3 expres-
sion level was evaluated by calculating the MFI ratio
(stained/unstained cells) (Fig. 2). HUVECs showed high
expression of integrin ꢀVꢁ3 (MFI ¼ 18:1), while the
four cell lines Cos-1, RERF-LC-AI, HepG2, and NIH-
3T3 showed lower expression levels (MFI ¼ 1:3{3:3).
HUVEC Cos-1 RERF HepG2 NIH-3T3
Integrin ꢀVꢁ3 levelꢂ
18.1
1.7
2.0
1.3
3.3
ꢂꢂ
Cytotoxicity IC50 (mM)
DA
DB
DC
DD
>100 >100 >100 >100
>100 >100 >100 >100
>100 >100 >100 >100
>100 >100 >100 >100
>100
>100
>100
>100
cys-DA
cys-DB
cys-DC
cys-DD
>100 >100 >100 >100
>100 >100 >100 >100
>100 >100 >100 >100
>100 >100 >100 >100
>100
>100
>100
>100
Cytotoxicities of cRGD-cys-D9mer peptides
cRGD-cys-DA
cRGD-cys-DB
cRGD-cys-DC
cRGD-cys-DD
45.6
40.6
45.8
24.2
>100 >100 >100
>100 >100 >100
>100 >100 >100
>100 >100 >100
>100
>100
>100
>100
The cytotoxic activities of three kinds of peptides,
D9mer, cys-D9mer, and cRGD-cys-D9mer peptides, were
determined against integrin ꢀVꢁ3-high and low-ex-
pressing cells. The D9mer and cys-D9mer peptides
showed no cytotoxicity, with 50% inhibitory concen-
tration (IC50) values above 100 mM, toward all cell lines
with and without integrin ꢀVꢁ3 expression. In contrast,
all the cRGD-cys-D9mer peptides exhibited integrin
ꢀVꢁ3 expression-dependent cytotoxicity against HU-
VECs (IC50 ¼ 24:2{45:8 mM) (Table 1). The four
cRGD-containing peptides showed effective inhibitory
activities against the proliferation of HUVECs in the
following order: DD > DB > DA ; DC.
ꢂIntegrin ꢀVꢁ3 levels were calculated by the mean fluorescence intensity
ratio (anti-integrin ꢀVꢁ3 antibody-FITC stained/non-stained cells) in flow
cytometric analyses.
ꢂꢂIC50 means the concentration of peptide that is required for 50% inhibition
of cell viability relative to untreated cells.
the integrin ꢀVꢁ3-specific internalization of a D9mer
peptide combined with the cRGD motif, we observed
the localization of TAMRA-labeled cRGD-cys-DC in
integrin ꢀVꢁ3-high-expressing HUVECs (MFI ¼ 18:1)
and ꢀVꢁ3-low-expressing Cos-1 cells (MFI ¼ 1:7).
After 24 h of incubation with 25 mM cRGD-cys-DC-
TAMRA, an accumulation of red fluorescence of
TAMRA was observed in the cytoplasm of the HUVECs
Internalization of cRGD-cys-DC
The cRGD motif has been reported to act as an
integrin ꢀVꢁ3-specific delivery sequence. To confirm

Selective Disruption of Mitochondria in a Target Cell
2047
Cos-1
HUVEC
(Integrin αVβ3
high-expressing cells)
A
B
(Integrin αVβ3
Cos-1
5
120
100
80
60
40
20
0
low-expressing cells)
Cos-1
a
4
3
2
1
0
a
b
20µm
20µm
HUVEC
5
4
3
2
1
0
120
100
80
60
40
20
0
HUVEC
b
c
d
0
12.5 25
Conc. of cRGD-cys-DC (µM)
50
Fig. 4. Integrin ꢀVꢁ3-Dependent Apoptosis Induction by cRGD-cys-
DC.
Fig. 3. Integrin ꢀVꢁ3-Dependent Internalization of cRGD-cys-DC-
TAMRA.
A, The effects of cRGD-cys-DC on cytoplasmic mitochondria
were investigated by JC-1 staining. Cos-1 cells and HUVECs were
incubated with 50 mM cRGD-cys-DC under 5% CO2 at 37 ꢁC for
24 h. Red fluorescence indicates healthy mitochondria and green
fluorescence indicates damaged mitochondria with loss of mito-
chondrial membrane potential. Scale bars indicate 10 mm. B, Cell
viability (circles), LDH release (squares), and caspase 3/7 activity
(diamonds) were measured to examine how cRGD-cys-DC kills
target cells. The cell viability and caspase 3/7 activity are shown as
values relative to control cells (untreated cells), and LDH release is
shown as values relative to maximum LDH release from the cells
treated with 2% Triton X-100. Data shown are means ꢃ SD (n ¼ 3).
Cos-1 cells (integrin ꢀVꢁ3-low-expressing cell line) (a, c) and
HUVECs (integrin ꢀVꢁ3-high-expressing cells) (b, d) were incu-
bated with 25 mM cRGD-cys-DC-TAMRA at 37 ꢁC for 24 h. Red
fluorescence accumulation indicates internalization of cRGD-cys-
DC-TAMRA into the cytoplasm (c, d). Scale bars indicate 20 mm.
(Fig. 3d), whereas the Cos-1 cells showed much less red
fluorescence (Fig. 3c) under fluorescence microscopy.
These observations suggest this D9mer peptide com-
bined with the cRGD peptide translocated into the
cytoplasm that in an integrin ꢀVꢁ3-dependent manner.
rin ꢀVꢁ3-high-expressing HUVECs, whereas the D9mer
and cys-D9mer peptides showed no cytotoxicity toward
any of the cell lines examined.
Effects of cRGD-cys-DC on mitochondrial potential
and apoptosis induction
Furthermore, as for the D9mer peptide sequences
conjugated with the cRGD peptide, the four peptides
showed effective inhibitory activities against the pro-
liferation of HUVECs with the following tendency:
DD > DB > DA ; DC. In particular, cRGD-cys-DD
showed the lowest IC50 value (highest cytotoxicity)
against HUVECs. The DD sequence (RLLLRIGRR-NH2)
contains the largest degree of hydrophobicity and a
middle degree of positive charges among the D9mer
peptides. On the other hand, DB contains the largest
positive charges and a middle degree of hydrophobicity.
This suggests that the degrees of hydrophobicity and
positive charges in the D9mer sequences are associated
with the mitochondria disruptive activities.
We found previously that synthesized peptides DC2
(ALYLAIRRRRRRRR-NH2) and DC (ALYLAIRRR-NH2)
conjugated with CPP (octa-arginine: RRRRRRRR-NH2)
exhibited clear cytotoxic effects by internalization into
the cytoplasm and disruption of the mitochondria.14)
Hence in the present study we used cRGD-cys-DC as a
typical peptide complex from among the four cRGD-
cys-D9mer peptides for comparison with DC2, although
cRGD-cys-DC showed the lowest cytotoxicity against
HUVECs. Similarly to the case for DC2-TAMRA,
cRGD-cys-DC-TAMRA also penetrated the cell mem-
brane and became localized to the cytoplasm of the
HUVECs. These findings suggest that cRGD-cys-DC can
show effects similar to DC2, viz., mitochondria disrup-
tion, in HUVECs. To confirm this, we observed the
mitochondrial membrane potential and carried out three
assays, for cell viability, cell membrane leakage, and
apoptosis induction, in cells treated with cRGD-cys-DC.
The cRGD-cys-DC induced, a loss of mitochondrial
membrane potential (a change in the JC-1 fluorescence
To examine the intracellular effects of cRGD-cys-
D9mer peptides, mitochondrial membrane status was
observed by JC-1 changes from J-aggregates (red
fluorescence) to JC-1 monomers (green fluorescence)
in Cos-1 cells and HUVECs treated with 50 mM cRGD-
cys-DC. The integrin ꢀVꢁ3-low-expressing Cos-1 cells
showed low green fluorescence and high accumulation
of red fluorescence, indicating healthy mitochondria
with high mitochondrial membrane potential (Fig. 4Aa).
On the other hand, a large increase in green fluorescence
and a lack of red fluorescence, which indicates damaged
mitochondria with loss of mitochondrial membrane
potential, were observed in the HUVECs treated with
cRGD-cys-DC (Fig. 4Ab). This indicates, that cRGD-
cys-DC damaged the mitochondria and decreased the
mitochondrial membrane potential in the HUVECs.
Cell membrane leakage and apoptosis assays were
also performed in conjunction with a cell viability assay
to elucidate the cytotoxic mechanism of cRGD-cys-DC.
The cRGD-cys-DC inhibited the growth of HUVECs at
lower concentrations than Cos-1 cells with an increase
in apoptosis indicator caspase 3/7 activity and without
cell membrane damage, as determined by LDH release.
This indicates that cRGD-cys-DC killed integrin ꢀVꢁ3-
high-expressing HUVECs by inducing apoptosis, rather
than by disrupting the cell membrane.
Discussion
In this study, we identified selective mitochondrial
disruption and cytotoxicity using mitochondria-disrup-
tive D9mer peptides and a tumor-homing cRGD peptide.
The cRGD-cys-D9mer peptides effectively killed integ-

2048
T. IWASAKI et al.
color from red to green), and decreased cell viability
with an increase in caspase 3/7 activity in an integrin
ꢀVꢁ3 expression-specific manner. In contrast, the
amount of LDH released from the cytoplasm was not
affected. Our previous study revealed that cationic
D9mer peptides, including DC, disrupt the tumor cell
membrane and increase LDH derived from tumor cells,
because large amounts of negatively charged phospha-
tidylserine are expressed on the target tumor membrane
surface.12) However, in the present study, cRGD-cys-DC
killed HUVECs with an increase in caspase 3/7 activity
without any increase in LDH release. This indicates that
cRGD-cys-DC exerts its cytotoxicity by inducing apop-
tosis without damaging the outer cell membrane, and
that uptake of the peptide into HUVECs via the cRGD
motif takes place in preference to outer-cell membrane
disruption by the DC sequence.
In this study, we succeeded in controlling the
cytotoxicity of mitochondria-disruptive peptides by
conjugating them with the cRGD motif. Our results
suggest that cRGD-cys-D9mer peptides are promising as
anti-angiogenesis drugs against integrin ꢀVꢁ3-high-
expressing tumor vasculature cells. In addition, this
study illustrates that D9mer peptides have broad utility
for mitochondria-targeting strategies focusing on other
cell types using other DDSs. It might be worthwhile to
test novel combinations of D9mer peptides and other
DDSs in the future.
References
1) Desagher S and Martinou JC, Trends Cell Biol., 10, 369–377
(2000).
2) Gogvadze V, Orrenius S, and Zhivotovsky B, Semin. Cancer
Biol., 19, 57–66 (2009).
3) Hancock RE and Chapple DS, Antimicrob. Agents Chemother.,
43, 1317–1323 (1999).
4) Zasloff M, Nature, 415, 389–395 (2002).
5) Ellerby HM, Arap W, Ellerby LM, Kain R, Andrusiak R, Rio
GD, Krajewski S, Lombardo CR, Rao R, Ruoslahti E, Bredesen
DE, and Pasqualini R, Nat. Med., 5, 1032–1038 (1999).
6) Margulis L, Symp. Soc. Exp. Biol., 29, 21–38 (1975).
7) Trapp S and Horobin RW, Eur. Biophys. J., 34, 959–966 (2005).
8) Iwasaki T, Saido-Sakanaka H, Asaoka A, Taylor D, Ishibashi J,
and Yamakawa M, J. Insect Biotechnol. Sericology, 76, 25–29
(2007).
9) Iwasaki T, Ishibashi J, Tanaka H, Sato M, Asaoka A, Taylor D,
and Yamakawa M, Peptides, 30, 660–668 (2009).
10) Yamage M, Yoshiyama M, Grab DJ, Kubo M, Iwasaki T, Kitani
H, Ishibashi J, and Yamakawa M, Biosci. Biotechnol. Biochem.,
73, 1520–1526 (2009).
11) Iwasaki T, Ishibashi J, Kubo M, Taylor D, and Yamakawa M,
Biosci. Biotechnol. Biochem., 73, 683–687 (2009).
12) Miyanoshita A, Hara S, Sugiyama M, Asaoka A, Taniai K,
Yukuhiro F, and Yamakawa M, Biochem. Biophys. Res.
Commun., 220, 526–531 (1996).
13) Ishibashi J, Saido-Sakanaka H, Yang J, Sagisaka A, and
Yamakawa M, Eur. J. Biochem., 266, 616–623 (1999).
14) Saido-Sakanaka H, Ishibashi J, Sagisaka A, Momotani E, and
Yamakawa M, Biochem. J., 338, 29–33 (1999).
15) Futaki S, Biopolymers, 84, 241–249 (2006).
16) Rothbard JB, Kreider E, VanDeusen CL, Wright L, Wylie BL,
and Wender PA, J. Med. Chem., 45, 3612–3618 (2002).
17) Temming K, Schiffelers RM, Molema G, and Kok RJ, Drug
Resist. Updat., 8, 381–402 (2005).
Acknowledgments
This work was supported in part by a Grant-in-Aid
for Scientific Research (B) (20380039, to M.Y.) and
by a Research Fellowship from the Japan Society for
the Promotion of Science for Young Scientists (19-1733,
to T.I.).
18) Robinson SD and Hodivala-Dilke KM, Curr. Opin. Cell Biol.,
23, 630–637 (2011).