Synthesis of anticancer heptapeptides containing a unique lipophilic β2,2-amino acid building block

Article abstract of DOI:10.1002/psc.1434

We report a series of synthetic anticancer heptapeptides (H-KKWβ^2,2WKK-NH₂) containing eight different central lipophilic β^2,2-amino acid building blocks, which have demonstrated high efficiency when used as scaffolds in small cationic antimicrobial peptides and peptidomimetics. The most potent peptides in the present study had IC₅₀ values of 9-23μm against human Burkitt's lymphoma and murine B-cell lymphoma and were all nonhaemolytic (EC₅₀>200μm). The most promising peptide 10e also demonstrated low toxicity against human embryonic lung fibroblast cells and peripheral blood mononuclear cells and exceptional proteolytic stability.

Full text of DOI:10.1002/psc.1434

Research Article
Received: 22 August 2011
Revised: 21 October 2011
Accepted: 31 October 2011
Published online in Wiley Online Library: 16 January 2012
(wileyonlinelibrary.com) DOI 10.1002/psc.1434
Synthesis of anticancer heptapeptides
containing a unique lipophilic b2,2-
amino acid building block
Veronika Tørfoss,a Dominik Ausbacher,a Cristiane de A. Cavalcanti-Jacobsen,a
Terkel Hansen,a Bjørn-Olav Brandsdal,b Martina Havelkovaa and
Morten B. Strøma*
We report a series of synthetic anticancer heptapeptides (H-KKWb2,2WKK-NH2) containing eight different central lipophilic
b
2,2-amino acid building blocks, which have demonstrated high efficiency when used as scaffolds in small cationic antimicrobial
peptides and peptidomimetics. The most potent peptides in the present study had IC50 values of 9–23 mM against human Burkitt’s
lymphoma and murine B-cell lymphoma and were all nonhaemolytic (EC50 > 200 mM). The most promising peptide 10e also
demonstrated low toxicity against human embryonic lung fibroblast cells and peripheral blood mononuclear cells and exceptional
proteolytic stability. Copyright © 2012 European Peptide Society and John Wiley & Sons, Ltd.
Supporting information may be found in the online version of this article.
Keywords: peptide synthesis; anticancer peptides; b2,2-amino acids; heptapeptides
intracellular target proteins, CAPs may have a lower probability of
inducing resistance compared with traditional chemotherapeutics.
Introduction
Alkylating agents, antimetabolites, and plant alkaloids are tradition-
ally used as anticancer chemotherapeutics but are often associated
with numerous side effects and low patient compliance [1]. Several
targeted therapies such as monoclonal antibodies and small mole-
cule inhibitors have also been introduced, which are less toxic and
better tolerated than traditional anticancer chemotherapy [2]. A key
concern in anticancer chemotherapy is, however, the often rapid
development of resistance, involving expression of efflux pumps or
mutations in intracellular or extracellular target proteins [3,4]. Clearly,
there is a strong need for continued development of anticancer
drugs with new modes of action, low toxicity, and low probability
of inducing resistance.
Cationic antimicrobial peptides (CAPs) is a promising class of anti-
cancer agents having a unique mode of action by targeting and
disrupting the integrity of the cancer cell membrane or the mito-
chondrial membrane, the latter demonstrated by swelling of the
mitochondria and induction of apoptosis [5–7]. Anticancer CAPs have
also been shown to trigger innate immunity, kill resistant cancer cells,
and enhance the cytotoxicity of traditional chemotherapeutics in
multidrug-resistant tumours [8,9].
Although the literature on anticancer CAPs of less than nine resi-
dues is somewhat limited, an inspiring aspect when designing small
anticancer CAPs is that the antimicrobial core sequence of bovine
lactoferricin (LfcinB) (H-RRWQWR-NH2) displays anticancer activity
when delivered intracellularly to T-leukaemia cells [13]. Furthermore,
the peptide contains a lipophilic core sequence with cationic ends,
and despite its small size, it adopts a highly defined amphipathic
structure upon interaction with negatively charged micelles [14].
We have recently investigated the antimicrobial activity of small
CAP-based b-peptidomimetics consisting of a newly developed
series of achiral, lipophilic b2,2-amino acids, which were potent
against both Gram-positive and Gram-negative bacteria and excep-
tionally potent against MRSA [15,16]. Noteworthy, b2,2-amino acids
have been shown to induce turns when incorporated into longer
*
Correspondence to: Morten B. Strøm. Department of Pharmacy, Faculty of Health
Sciences, University of Tromsø, NO-9037 Tromsø, Norway. E-mail: morten.strom
@uit.no
a Department of Pharmacy, Faculty of Health Sciences, University of Tromsø,
NO-9037 Tromsø, Norway
The preference of CAPs for cancer cells compared with normal
cells has been attributed to an electrostatic attraction between the
cationic residues of the peptides and the negatively charged outer
membrane constituents of cancer cells [6]. The difference in suscep-
tibility between normal and cancerous cells is explained by a 3–9%
excess of the acidic phospholipid phosphatidylserine, elevated levels
of O-glycosylated mucins, and increased sialylation of glycoproteins
and glycolipids on cancer cells [10,11]. Additionally, increased fluidity
of the cell membrane and increased surface area owing to more
microvill on cancer cells is believed to affect the action of CAPs
[12]. Thus, by evading the use of specific extracellular and
b Department of Chemistry, Faculty of Science and Technology, NO-9037 Tromsø,
Norway
Abbreviations used: Ramos, human Burkitt’s lymphoma; A20, murine B-cell
lymphoma; MRC-5, human embryonic lung fibroblasts; CAPs, cationic antimicrobial
peptides; RBC, red blood cells; PBMCs, human peripheral blood mononuclear
cells; MTT, 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide; MRSA,
methicillin-resistant Staphylococcus aureus; ATCC, American Type Culture Collection;
HBTU, O-benzotriazole-N,N,N′,N′-tetramethyl-uronium-hexafluoro-phosphate; HOBt,
N-hydroxybenzotriazole; TFFH, fluoro-N,N,N′,N′-tetramethylformamidinium hexa-
fluorophosphate; ITC, isothermal titration calorimetry; PI propidium iodide.
J. Pept. Sci. 2012; 18: 170–176
Copyright © 2012 European Peptide Society and John Wiley & Sons, Ltd.

ANTICANCER HEPTAPEPTIDES
peptides, and by including bulky lipophilic substituents, the turn
inducing effect may be enhanced [17,18]. To design an even more
distinctive amphipathic motif around the b2,2-amino acids, they were
flanked by tryptophans to increase the bulkiness and then by lysines
as cationic residues, resulting in a palindromic model heptapeptide
H-KKWb2,2WKK-NH2. Lysine was preferred as cationic residue
because Arg in CAPs has been reported to interact more strongly
with zwitterionic phospholipids in healthy cells and thereby promote
cell toxicity [19].
In the present study, we have synthesized and studied the effects
of incorporating various lipophilic b2,2-amino acid residues into the
palindromic model heptapeptide to develop small anticancer CAPs
(Figure 1). The anticancer potency of the peptides was investigated
against human Burkitt’s lymphoma (Ramos) and murine B-cell lym-
phoma (A20), and toxicity was examined against human red blood
cells (RBC), healthy human embryonic lung fibroblasts (MRC-5), and
healthy human peripheral blood mononuclear cells (PBMCs). High
proteolytic stability was furthermore demonstrated by degradation
studies using trypsin and a-chymotrypsin.
LC quadropole MS equipped with an electrospray ion source (Micro-
mass, Manchester, UK) and analyzed using the MASSLYNX v3.4 soft-
ware. All samples were analyzed in the positive ion mode.
1H-NMR and 13C-NMR experiments were recorded on a Varian 400
NMR spectrometer or a Varian 600 NMR spectrometer with CDCl3
or CD3OD as solvents and analyzed using VNMR software (Varian Inc.,
Palo Alto, CA, USA). Chemical shifts are expressed in parts per million
(ppm, d) and referred to the solvent signal [20]. The preparative
HPLC system consisted of a Waters 600 E System Controller, a Waters
In-Line Degasser, a Waters 717 auto sampler, and a Waters 2487
Dual l Absorbance Detector controlled by Empower Pro software
(Waters Corporation). A SunFire Prep OBD 5-mm C18 RP column
(19 Â 250 mm) (Waters Corporation) was used. The absorbance
detector was operated at 214 nm. The mobile phase was different
combinations of 5% acetonitrile in Milli-Q water (A) and 95% acetoni-
trile in Milli-Q water (B), both containing 0.1% TFA, and the flow rate
was 15 ml/min. The analytical HPLC system consisted of a Waters
2695 Separations Module and a Waters 996 Photodiode Array
(PDA) Detector controlled by Empower Pro software. A SunFire
5mm C18 RP column (4.6 Â 250 mm) was used for purity analysis. A
YMC-Pack Pro 5 mm C18 RP column (4.6 Â 250 mm) was used for re-
tention time analysis. The compounds were analyzed at wavelengths
214 and 254 nm with the PDA detector spanning from wavelengths
210 to 310 nm The mobile phase was different combinations of 5%
acetonitrile in Milli-Q water (A) and 95% acetonitrile in Milli-Q water
(B), both containing 0.1% TFA, and the flow rate was 1 ml/min.
Materials and Methods
General
All chemicals were purchased from Sigma-Aldrich and used without
further purification. HRMS spectra were recorded on a Waters
Micromass LCT Premier time-of-flight mass spectrometer equipped
with an electrospray ion source and analyzed using MASSLYNX v4.1
software (Waters Corporation, Milford, MA, USA). The samples were
introduced to the mass spectrometer using a Waters 2795 analytical
HPLC with an XTerra MS 3.5-mm C18 RP column (2.1 Â 50 mm)
(Waters Corporation). The mobile phase consisted of different combi-
nations of Milli-Q water (Millipore Corporation, Billerica, MA, USA) and
acetonitrile, both containing 0.1% formic acid, and a gradient
running for 5 min was used. Leu-enkephalin was infused through
the reference probe and used as lock mass for internal calibrations
throughout the data acquisitions. All data were acquired in the
positive ion mode. MS-spectra were recorded on a Waters Quattro
Synthesis
For synthesis, yields, and characterization data of 1a, 1c-1h, 3a, 3c,
3d, 3g, 4a, 4c, 4d, and 4g, please see Hansen et al. [15]. Synthesis,
yields, and characterization data of all other compounds are given
in the Supporting Information. All peptides tested had a purity of
>95% as determined by analytical RP-HPLC.
Maintenance of cell lines
Murine B-cell lymphoma, MRC-5, and Ramos cell lines were kindly
provided by Prof. ꢀystein Rekdal (Department of Medical Biology,
NH3
H3N
O
O
NH2
H3N
HN
O
HN
O
NH
NH
O
HN
O
O
NH3
H3N
NH
HN
HN
R
R
R =
10a-h
F
CF3
a
b
c
d
e
f
g
h
Figure 1. General structure of the synthesized heptapeptides 10a-10h (H-KKWb2,2WKK-NH2).
J. Pept. Sci. 2012; 18: 170–176 Copyright © 2012 European Peptide Society and John Wiley & Sons, Ltd. wileyonlinelibrary.com/journal/jpepsci

TꢀRFOSS ET AL.
University of Tromsø). A20, a B-cell lymphoma of BALB/c mice origin
(TIB-208, ATCC, Manassas, VA, USA), and the Ramos cell line, a human
Burkitt’s lymphoma B lymphocyte, were maintained in RPMI-1640
medium containing 10% heat-inactivated foetal bovine serum
(FBS). The human embryonic fibroblast cell line MRC-5 (CCL-171,
ATCC) was maintained in Minimal Essential Medium (Sigma-Aldrich,
St. Louis, MO, USA) supplemented with 10% FBS. Both culture media
were without antibiotics. The cell cultures were maintained in a
humidified atmosphere of 5% CO2 at 37 ꢀC, and all cell lines were
regularly tested for viruses (RapidMAPTM-27, Taconic, Europe) and
mycoplasma (MycoAlert Assay Control Set, Cat # NLT07-518, Lonza
Rockland Inc., Rockland, ME, USA). The adherent cell line MRC-5
was detached from the solid support using trypsin and suspended
in RPMI-1640 medium with 10% FBS and 1% L-glutamine prior to use.
available test tube containing EDTA (BD vacutainer, 7.2 mg K2 EDTA,
BD, Heidelberg, Germany) and into a 10 ml reaction vial containing
40 ml of a heparin solution (1000 U/ml). After 30 min, the haematocrit
(Hct) of the EDTA-treated blood was determined (Sysmex K-1000,
Sysmex Corporation, Kobe, Japan), and according to the measured
value, the necessary amounts of blood and PBS for a suspension with
10% Hct were calculated. The heparinized blood sample was centri-
fuged for 10 min at 437 g (1500 r.p.m)., and the supernatant was
carefully removed and replaced with prewarmed (37 ꢀC) PBS. The
procedure was repeated three times, and the sample was finally
diluted with PBS to 10% Hct. The peptides were dissolved in PBS
and were pipetted in reaction vials in end concentrations from
1mg/ml up to 1000 mg/ml. A positive control with an end concentra-
tion of 0.1% Triton X-100 and a negative control containing pure PBS
buffer were included. The erythrocytes (1% v/v) were added, and the
vials were incubated under agitation at 37 ꢀC for 1h. The samples
were centrifuged for 5 min at 3112 g (4000 r.p.m)., and 100 ml of each
reaction vial was transferred to a 96-well plate, and intensely
coloured samples were diluted with PBS. After measuring the absor-
bance at 405 nm with a microplate reader (VersaMaxTM, Molecular
Devices, Sunnyvale, CA, USA), the percentage of haemolysis was
calculated as the ratio of the absorbance in the peptide-treated
and surfactant-treated samples. The absorbance of the negative
control was used for baseline corrections.
PBMCs separation
Buffy coat samples were obtained from the University Hospital of
North Norway Blood Bank. PBMCs were isolated by centrifugation in
Lymphoprep (1.077 g/ml; Nycomed Pharma, Oslo, Norway). Briefly,
buffy coat was diluted 1 : 4 in PBSA (0.2% bovine serum albumin in
PBS) without calcium or magnesium. Afterwards, 35 ml of the diluted
buffy coat was layered over 15 ml of Lymphoprep, followed by a
15-min centrifugation at 800 g at room temperature. The interface
was collected and mixed gently with 40 ml of PBSA. The sample was
centrifuged two times at 400 g, for 6 min, at room temperature. After
each centrifugation, the supernatant was eliminated and another
40 ml of PBSA was added. The pellet, containing the PBMCs, was
resuspended in RPMI 1640 culture medium (Sigma-Aldrich, St. Louis,
MO, USA) and immediately used on MTT assay.
Proteolytic Stability Assays
The proteolytic stability of 10e against trypsin and a-chymotrypsin
was determined in a solution of 100 mg/ml of 10e, 2mg/ml of trypsin
or a-chymotrypsin, 100 mM tris-HCl and 10 mM CaCl2. The enzyme
digestions were incubated at 37 ꢀC. Samples of 15 ml were collected
at different time intervals, and 100 ml 10% (v/v) formic acid was
added to stop the enzyme activity. A negative control without
enzyme was incubated to determine whether the degradation
was due to the enzyme or other factors. As positive control for
a-chymotrypsin, a solution of succinyl-Ala-Ala-Pro-Phe-p-nitroaniline
(0.62 mg/ml) was added to a final concentration of 0.48 mM instead
of the peptide solution. Similarly, 0.5-mM Na-Benzoyl-L-arginine ethyl
ester hydrochloride (Sigma-Aldrich, St. Louis, MO, USA) to a final
concentration of 11.8 mg/ml was used as positive control for trypsin.
Propranolol hydrochloride was used as internal standard and added
to the 200 mM tris-HCl buffer to a final concentration of 57.8 mM. In the
digestions containing succinyl-Ala-Ala-Pro-Phe-p-nitroaniline, the
internal standard was 5-nitroimidazole at a final concentration of
40 mg/ml in the buffer solution. All samples were run on an analytical
HPLC-PDA (Waters 2695 Separations Module and a Waters 996 PDA
Detector controlled by Empower Pro software with an YMC-Pack
Pro 5 mm C18 RP column (4.6 Â 250 mm)) immediately after prepara-
tion for quantitative analyses of remaining 10e. The gradient for
separation started with an isocratic elution with 80% A and 20% B
for 3 min, then a linear gradient to 40% A and 60% B over the next
17 min. Solvent A consisted of 95% water, 5% acetonitrile, and 0.1%
TFA, whereas solvent B consisted of 5% water, 95% acetonitrile,
and 0.1% TFA. Detection of 10e, Na-Benzoyl-L-arginine ethyl ester
hydrochloride and propranolol was performed at 260 nm, and detec-
tion of succinyl-Ala-Ala-Pro-Phe-p-nitroaniline and 5-nitroimidazole
was performed at 350 nm. The relative ratio of the area of 10e over
the area of propranolol was determined, and the degree of degrada-
tion of 10e was calculated using the ratio at t= 0 as reference. To
determine the degradation product of trypsin, we purified the
enzyme digestion on a preparative HPLC-UV and analyzed the
collected fractions using MS.
MTT Assay
All cells, including the suspension cells, were counted and centri-
fuged, and the resulting pellets were suspended in serum-free
RPMI-1640 medium [21]. The cells were seeded in 96-well plates,
100 ml per well of cell suspension. A20 cancer cells were seeded at
a concentration of 6 Â 105 cells/ml, MRC-5 cells at 1 Â 105 cells/ml,
Ramos cancer cells at 4 Â 105 cells/ml, and PBMCs at 2 Â 106 cells/
ml. Ramos, A20, and PBMCs were stimulated immediately, whereas
the MRC-5 cells were incubated for 24h and washed twice with
serum-free RPMI-1640 medium before stimulation.
To determine the cytotoxic effect of the peptide, we used the
colorimetric MTT viability assay, and 100 ml of different concentra-
tions of peptides were added to all the cell lines. Cells added with
100 ml serum-free medium were used as negative controls, and cells
added with 100 ml 1% Triton X-100 in serum-free medium were used
as positive controls. After 4 h of incubation, 10 ml MTT solution (5mg
MTT per mL PBS) was added to the MRC-5 cells, whereas 20 ml was
added to the Ramos, A20, and PBMCs, followed by another 2 h of in-
cubation. A volume of 70 ml was removed from the MRC-5 cells, and
130 ml was removed from Ramos, A20, and PBMCs, before 100 ml of
isopropanol with 0.04 N HCl was added, and the plates were shaken
for 1 h at room temperature. The absorbance at 590 nm was mea-
sured on a microtitre plate reader. Cell survival was determined from
the ΔA590 nm relative to the negative control, and the 50% inhibitory
concentrations (IC50) were determined from the dose–response
curves. The procedures per sample were carried out in triplicates.
RBC Toxicity Assay
The RBC toxicity testing was conducted as previously described
[15,22,23]. In brief, 8 ml of blood was collected from adult male
donors, and the blood was divided equally into a commercially
wileyonlinelibrary.com/journal/jpepsci Copyright © 2012 European Peptide Society and John Wiley & Sons, Ltd. J. Pept. Sci. 2012; 18: 170–176

ANTICANCER HEPTAPEPTIDES
Although different b2,2-amino acids were prepared, method B gave,
in general, higher yields, involved fewer steps, and was quicker and
easier to perform than method A. The final peptides 10a-10h
(Figure 1) were synthesized by Fmoc SPPS using HBTU/HOBt or
TFFH as activating agents. (See Supporting Information for experi-
mental details and comparison of yields.)
Results and Discussion
Synthesis
Synthesis of the Fmoc-b2,2-amino acids 9a-9h was performed by two
different methods, A and B, as shown in Scheme 1. Both methods
started with dialkylation of methyl cyanoacetate and reduction of
the nitrile group by catalytic hydrogenation to achieve the unpro-
tected b2,2-amino acid methyl esters 2a-2h [24]. Method A was based
on our previous synthesis of antimicrobial b-peptidomimetics in
which the b2,2-amino acid methyl esters 2 were further Boc-
protected before the methyl esters were hydrolyzed [15]. The Boc
protections were removed, and the resulting b2,2-amino acids (5)
were precipitated as tosylate salts (6) before the amino groups were
protected with Fmoc groups (9) [25].
Anticancer Activity
The peptides were screened for anticancer activity against Ramos
and A20 cells. As a measurement of toxicity, the peptides were tested
for haemolytic activity against human RBC, and for cytotoxicity,
against normal MRC-5 and human PBMCs.
Although only the side chains of the central b2,2-amino acid
residue were altered in the eight peptides prepared, the results
nevertheless revealed a substantial variation in anticancer potency
and toxicity. We also found that the anticancer potency of the
peptides was comparable with potencies reported for longer
anticancer CAPs, thereby supporting the hypothesis on the effi-
ciency of including a central lipophilic core sequence with cationic
N-terminal and C-terminal ends in our model heptapeptides [9,26].
When investigating the individual peptides, 10h was the overall
most cytotoxic peptide prepared, followed by 10g and 10f (Table 1).
Peptide 10h was the only peptide containing a b2,2-amino acid
residue with two purely aliphatic lipophilic side chains (Figure 1),
and its high cytotoxicity against all cell types may be a result of a
deeper penetration into the phospholipid acyl chain region of both
normal and cancer cell membranes.
A more efficient method was developed during the project,
method B, which involved precipitation of the b2,2-amino acid
methyl esters 2 as tosylate salts (7) before hydrolysis of the methyl
esters and subsequent Fmoc protection of the amino groups.
O
N
O
a
O
N
1a-h
O
Crude yield: 42->99%
R
R
b
O
O
H2N
2a-h
With respect to the remaining peptides containing b2,2-amino
acids with aromatic side chains, we observed that the trends in
anticancer potency followed the same trends as observed for our
small antimicrobial b-peptidomimetics with similar b2,2-amino acid
derivatives in the order from high to low potency: 10g, 10f, 10e,
10d, and 10c, and with peptides 10b and 10a showing poor or no
anticancer potency (Table 1) [15,16]. Importantly, peptides 10f and
10g contained the most bulky b2,2-amino acid derivatives, which
might explain their high anticancer potency when presuming a
membrane-disrupting mode of action as reported for larger CAPs [6].
Although only being the fourth most potent peptide, 10e was
perhaps the most promising peptide prepared when considering
overall anticancer potency and toxicity, that is, nontoxic against
RBC and PBMCs and low toxicity against MRC-5 cells (Table 1).
Peptide 10e contained a b2,2-amino acid derivative with two
aromatic p-trifluoromethyl benzylic side chains, which also has
proven highly efficient when incorporated into short antimicrobial
peptidomimetics [15,16]. Noteworthy, incorporation of aromatic or
benzylic fluorine substituents, as found in 10e, is widely used during
drug development to improve the pharmacokinetic properties of
drug candidates [27]. The effects on anticancer potency and toxicity
of the trifluoromethyl substituents of the b2,2-amino acid side chains
of 10e are not fully understood. However, the trifluoromethyl groups
may alter the electronic distribution of the aromatic side chains and
contribute to increased lipophilicity, which together can affect the
peptides’ interaction with various cell membrane constituents,
improve membrane permeability, and thereby permit interaction
with intracellular targets as reported for LfcinB [7,28]. Importantly,
we observed an increased influx of the noncell membrane perme-
able dye PI in Ramos cells treated for 4 h with 10e at its IC50
concentration, indicating loss of membrane integrity (see Supporting
Information for details). However, more extensive studies are needed
to elucidate the detailed mechanism of action.
R
R
Method A
Method B
c
O
h
Boc
Boc
3a,c,d,g
Yield: 55-79%
O
N
H
R
R
R
R
R
O
7b,e,f,h
Yield: 8-48%
d
Ts- +H3N
O
O
O- Li+
R
R
4a,c,d,g
Yield: 56-76%
N
H
R
e
i
O
CF3COO- +H3N
OH
5a,c,d,g
O
R
H2N
O- Li+
8b,e,f,h
f
O
R
R
Ts- +H3N
OH
6a,c,d,g
R
g
j
O
9a-h
Fmoc
Yield: 18-100%
Total yield 1-9: 4-30%
OH
N
H
R
R
Scheme 1. Synthesis of the Fmoc-protected b2,2-amino acids 9a-9h.
Method A gave total yields of 4–9%, and method B gave total yields of
10–30% for synthesis of 9a-9h. R groups are shown in Figure 1 and in
the Supporting Information. Isolated compounds are italicized. a: NaOMe,
R-Br (performed twice). b: Ra/Ni, H2(g), HOAc. c: TEA, Boc2O. d: LiOH, H2O/
dioxane. e: TFA:TIS:H2O (95:2.5:2.5). f: TsOH. g: TEA, Fmoc-OSu. h: TsOH. i:
LiOH, H2O/dioxane. j: Fmoc-OSu.
J. Pept. Sci. 2012; 18: 170–176 Copyright © 2012 European Peptide Society and John Wiley & Sons, Ltd. wileyonlinelibrary.com/journal/jpepsci

TꢀRFOSS ET AL.
Table 1. Anticancer potency of peptides 10a-10h against human Ramos and murine A20 cancer cells (IC50) and toxicity against normal human RBC
(EC50), MRC-5 cells (IC50), and PBMCs (EC50
)
Anticancer potency
Toxicity
Selectivity indexi
MRC-5/Ramos
Rtj
Entry
Ramosa,d
A20a,e
RBCc,f
MRC-5b,g
PBMCsa,h
RBC/Ramos
PBMCs/Ramos
(min)
10a
10b
10c
10d
10e
10f
—
—
—
—
—
n.d.
n.d.
—
—
—
—
—
—
3.24
3.13
4.69
4.99
5.95
4.82
9.44
9.29
—
113 Æ 7
24 Æ 2
24 Æ 1
22 Æ 2
14 Æ 2
10 Æ 1
9 Æ 1
—
44 Æ 3
33 Æ 1
23 Æ 3
22 Æ 5
16 Æ 1
9 Æ 2
—
115 Æ 9
124 Æ18
140 Æ 9
58 Æ 6
29 Æ 5
27 Æ 6
—
>13
>17
>23
19
2.6
3.8
6.1
2.6
1.8
3.0
>2.6
0.3
—
10 Æ 3
—
—
>4.7
1.4
425
232
214
31 Æ14
7 Æ 2
12 Æ 3
10g
10h
15
0.4
24
1.3
All values are in micromolar. Highest concentrations tested were the following: a200 mg/ml, b500 mg/ml, and c1000 mg/ml. All peptides were isolated as
their pentatrifluoroacetate salts, and the molar concentrations were calculated as such. dHuman Burkitt’s lymphoma B-lymphocyte eB-cell lymphoma of
BALB/c mice origin (ATCC TIB-208). fHuman red blood cells. gNormal human embryonic lung fibroblast cells (ATCC CCL-171). hHealthy human
peripheral blood mononuclear cells. iThe selectivity index was calculated as the IC50 values against PBMCs, MRC-5 cells, or RBC divided by the IC50
values against Ramos cancer cells. jRetention time on an analytical RP-HPLC C18 column. The symbol ‘—’denotes no detectable activity (IC50 or EC50)
within the concentration range tested, and ‘n.d.’ denotes value not determined.
Peptides 10c and 10d displayed similar potencies against the
A20 cancer cell line as 10e but were a little less potent against
Ramos cancer cells. However, the differences in structure of the
b2,2-amino acids resulted in a somewhat higher potency of 10d
against Ramos cancer cells compared with 10c. As described previ-
ously, 10a was inactive against all cell lines tested, whereas 10b
showed detectable activity against the A20 cancer cell line and
thereby revealed a small effect of the fluorine substituents in 10b.
Proteolytic Stability
A major obstacle when designing peptide-based drugs is often
their intrinsically low in vivo stability. Unmodified a-peptides are
in general unable to survive systemic circulation for more than a
few minutes because of extensive proteolytic degradation [30].
On the other hand, studies involving pure b-peptides show that
these peptides are highly resistant to proteolytic degradation by a
wide range of proteases [31].
By incorporating a disubstituted b2,2-amino acid into an otherwise
pure a-peptide, an extra methylene group was added to the current
peptide backbone, which in combination with two bulky residues
was suspected to improve proteolytic stability compared with pure
a-peptides. To investigate how the introduction of the b2,2-amino
acids affected proteolytic stability, we examined our most promising
peptide 10e by ITC and degradation by trypsin and a-chymotrypsin.
Trypsin cleaves C-terminal to cationic amino acids such as Lys and
Arg, whereas a-chymotrypsin cleaves C-terminal to large lipophilic
amino acids such as Trp, Phe, and Tyr. Thus, several of the amide
bonds in the present peptides were theoretically susceptible to
hydrolysis by these proteases.
The results from the stability experiments on 10e were rather
striking. No binding interactions at all were detected in the ITC
studies (see Supporting Information for details). However, the proteo-
lytic stability assay showed that 10e was in fact slowly degraded by
trypsin, with a half-life of about 22 h and with approximately 13%
of the peptide remaining after 4 days (Figure 2). Studies by prepara-
tive RP-HPLC-UV and MS revealed that only the N-terminal Lys-Lys
bond in H-KKWb2,2WKK-NH2 was susceptible to hydrolysis by trypsin,
resulting in the two fragments H-K-OH and H-KWb2,2WKK-NH2. No
other amide bonds C-terminal to the remaining Lys residues were
hydrolyzed.
Toxicity
Haemolytic activity is a widespread method for briefly assessing the
toxicity of CAPs and is often used when designing peptide-based
drugs. In general, a peptide is considered to be nonhaemolytic if
it displays less than 10% lysis of RBC at therapeutic concentrations
[29]. In the present study, only peptides 10f, 10g, and 10h
displayed measurable haemolytic activity but with EC50 values
above 200 mM and well above their IC50 values against the Ramos
and A20 cancer cell lines (Table 1). Noteworthy, even the most
haemolytic peptide 10h still displayed more than 27-fold
preference for both cancer cell lines compared with human RBC.
Because only three of the peptides displayed measurable EC50
values against human RBC, the data set was fairly insufficient to
rank all eight peptides with respect to a selectivity index (Table 1).
Given that RBCs are fragile, nondividing cells, a measurement of
toxicity with respect to metabolic active cells was therefore
obtained by measuring the toxicity against normal human MRC-5
cells and PBMCs.
The results showed that the anticancer potency against the
human Ramos cancer cell line was approximately twofold to more
than sixfold higher than the toxicity against normal MRC-5 cells
depending on the peptide in question and with 10c, 10d, and 10e
displaying the lowest measured toxicity against MRC-5 cells (Table 1).
Similar results were also obtained with respect to the murine A20
cancer cell line (calculations not shown). Furthermore, 10e was non-
toxic against PBMCs resulting in the highest selectivity index of all
peptides tested, whereas peptides 10d, 10f, 10g, and 10h with
similar or higher anticancer potencies were rather cytotoxic against
the PBMCs. Although 10c also was nontoxic against PBMC, peptide
10e was nevertheless considered the most promising peptide for
further studies on the basis of its higher anticancer potency.
Peptide 10e was seemingly completely stable against proteolytic
degradation by a-chymotrypsin, and after 4 days, 92% of the peptide
was still intact. The results thereby demonstrated how a single
b2,2-amino acid could protect at least the two adjacent residues on
each side from being hydrolyzed. Thus, lipophilic b2,2-amino acids
may be valuable building blocks also in other classes of peptides
by inducing unique conformations, altered lipophilicity, and biolog-
ical profiles, as well as protecting scissile peptide bonds against
proteolytic degradation.
wileyonlinelibrary.com/journal/jpepsci Copyright © 2012 European Peptide Society and John Wiley & Sons, Ltd. J. Pept. Sci. 2012; 18: 170–176

ANTICANCER HEPTAPEPTIDES
peptide. Elucidating the details of these interactions are challenging,
but high-resolution NMR experiments are in progress to reveal possi-
ble active conformations of the peptides upon interaction with
model micelles.
Conclusions
The study demonstrates how a recently developed series of lipophilic
b2,2-amino acids can be used to design small anticancer peptides.
The most promising peptide was 10e, which showed profound
potency against Ramos and A20 cancer cell lines, low toxicity against
human RBC, MRC-5, and PBMCs, and exceptional proteolytic stability
against trypsin and a-chymotrypsin. Peptide 10e may therefore form
the basis for future anticancer drug candidates that can either be
used individually or in combination with existing anticancer
chemotherapeutics. Because of the relative ease of synthesis of the
Fmoc-protected b2,2-amino acids 9a-9h, these lipophilic building
blocks may also find applications in other biological active peptides
involving optimization of pharmacokinetic properties and formation
of extraordinary secondary conformations.
Figure 2. Proteolytic stability of peptide 10e. The peptide was digested
with trypsin (crosses) or a-chymotrypsin (diamonds). A peptide sample
without enzyme was used as negative control (triangles). Percentage of
the remaining peptide 10e was measured using analytical RP-HPLC-UV.
Structural Considerations
When considering the primary structure of the peptides, the central
b2,2-amino acid was flanked by two tryptophan residues and thereby
formed a lipophilic core sequence with cationic Lys residues at the
N-terminal and C-terminal ends, similar to the core structure of
LfcinB. Linear a-peptides are in general unstructured in solution,
but b-peptides consisting of only 3–4 residues have been shown to
form highly stable secondary conformations even in aqueous
environment [17,32]. Because b2,2-amino acids have been shown to
be turn-inducing residues [17,18], the present peptides were
designed to be able to adopt a wedge-shaped amphipathic second-
ary conformation upon interaction with cancer cell membranes.
However, the most obvious parameters that varied between the
peptides were lipophilicity and the structures of the side chains
within the central b2,2-amino acid building blocks. As a measurement
of lipophilicity of the peptides, we compared their retention time on
an analytical C18-RP-HPLC column eluted with a linear gradient of
acetonitrile in water (Table 1) in which a correlation between overall
lipophilicity and anticancer potency as well as toxicity was revealed
(Supporting Information). However, the correlation was not
completely linear because 10e, being the third overall most lipophilic
peptide, showed lower toxicity against MRC-5 cells and PBMCs than
expected from its retention time. Also, 10f was more potent than 10c
and 10d, although they had similar retention times. Thus, these
results emphasized how retention time alone was unable to explain
all the observed effects of the peptides. Clearly, size and bulkiness
had a vital effect on anticancer potency.
There was also another important aspect of the present peptides
related to the close proximity of the aromatic side chains of the
b2,2-amino acids, especially in peptides 10a-10c and 10e-10g. The
ortho-protons in these side chains are likely to cause steric repulsions,
which may affect both the bulkiness of the b2,2-amino acids and
the overall peptide conformations. This may further reinforce the
turn-inducing potential of the b2,2-amino acids through a nearly
perpendicular orientation of these side chains. In addition to peptide
10e, an especially intriguing structure to solve is that of 10f, which
contained a b2,2-amino acid with two 2-naphthyl methylene side
chains and, as described previously, was more potent than presumed
from its retention time. A quite complex p–p stacking pattern involv-
ing the tryptophan residues and the 2-naphthyl methylene side
chains, in combination with side-chain repulsive forces within the
b2,2-amino acid building block could be imagined for this specific
Acknowledgements
The project was supported by the KOSK II grant 185141/V30 from
the Norwegian Research Council (NFR). We also thank Dr Espen
Hansen (Marbio, University of Tromsø) for providing help in
acquiring HRMS spectra, Prof. ꢀystein Rekdal (Department of
Medical Biology, University of Tromsø) for supplying Ramos, A20,
and MRC-5 cells, and the University Hospital of North Norway Blood
Bank for supplying the buffy coat.
References
1 Smith L. L., Brown K., Carthew P., Lim C.-K., Martin E. A., Styles J., White
I. N. H. Chemoprevention of breast cancer by tamoxifen: risks and
opportunities. Crit. Rev. Toxicol. 2000; 30: 571–594. DOI: 10.1080/
10408440008951120
2 Gerber D. E. Targeted therapies: a new generation of cancer treatments.
Am. Fam. Physician 2008; 77: 311–319.
3 Pérez-Tomás R. Multidrug resistance: retrospect and prospects in anti-
cancer drug treatment. Curr. Med. Chem. 2006; 13: 1859–1876.
4 Ellis L. M., Hicklin D. J. Resistance to targeted therapies: refining
anticancer therapy in the era of molecular oncology. Clin. Cancer
Res. 2009; 15: 7471–7478.
5 Mader J. S., Hoskin D. W. Cationic antimicrobial peptides as novel cy-
totoxic agents for cancer treatment. Expert Opin. Investig. Drugs 2006;
15: 933–946.
6 Hoskin D. W., Ramamoorthy, A. Studies on anticancer activities of an-
timicrobial peptides. BBA - Biomembranes 2008; 1778: 357–375.
7 Eliassen L. T., Berge G., Leknessund A., Wikman M., Lindin I., Løkke C.,
Ponthan F., Johnsen J. I., Sveinbjørnsson B., Kogner P., Flægstad T.,
Rekdal ꢀ. The antimicrobial peptide, lactoferricin B, is cytotoxic to
neuroblastoma cells in vitro and inhibits xenograft growth in vivo.
Int. J. Cancer 2006; 119: 493–500.
8 Kim S., Kim S. S., Bang Y. J., Kim S. J., Lee B. J. In vitro activities of native and
designed peptide antibiotics against drug sensitive and resistant tumor
cell lines. Peptides 2003; 24: 945–953.
9 Berge G., Eliassen L., Camilio K., Bartnes K., Sveinbjørnsson B., Rekdal ꢀ.
Therapeutic vaccination against a murine lymphoma by intratumoral
injection of a cationic anticancer peptide. Cancer Immunol. Immun. 2010;
59: 1285–1294. DOI: 10.1007/s00262-010-0857-6
10 Papo N., Shai Y. Host defense peptides as new weapons in cancer
treatment. Cell. Mol. Life Sci. 2005; 62: 784–790.
11 Fuster M. M., Esko J. D. The sweet and sour of cancer: glycans as novel
therapeutic targets. Nat. Rev. Cancer 2005; 5: 526–542.
12 Leuschner C., Hansel W. Membrane disrupting lytic peptides for cancer
treatments. Curr. Pharm. Design 2004; 10: 2299–2310.
J. Pept. Sci. 2012; 18: 170–176 Copyright © 2012 European Peptide Society and John Wiley & Sons, Ltd. wileyonlinelibrary.com/journal/jpepsci

TꢀRFOSS ET AL.
13 Richardson A., de Antueno R., Duncan R., Hoskin D. W. Intracellular de-
livery of bovine lactoferricin’s antimicrobial core (RRWQWR) kills T-leu-
kemia cells. Biochem. Bioph. Res. Co. 2009; 388: 736–741.
14 Schibli D. J., Hwang P. M., Vogel H. J. The structure of the antimicrobial
active center of lactoferricin B bound to sodium dodecyl sulfate micelles.
FEBS Lett. 1999; 446: 213–217.
15 Hansen T., Alst T., Havelkova M., Strøm M. B. Antimicrobial activity of small
b-Peptidomimetics based on the pharmacophore model of short cationic
antimicrobial peptides. J. Med. Chem. 2010; 53: 595–606. DOI:10.1021/
jm901052r
16 Hansen T., Ausbacher D., Flaten G. E., Havelkova M., Strøm M. B. Synthesis
of cationic antimicrobial b2,2-amino acid derivatives with potential for oral
administration. J. Med. Chem. 2011; 54: 858–868. 10.1021/jm101327d
17 Seebach D., Abele S., Sifferlen T., Hänggi M., Gruner S., Seiler P. Preparation
and structure of b-peptides consisting of geminally disubstituted b2,2- and
b3,3-amino acids: A turn motif for b-peptides. Helv. Chim. Acta 1998; 81:
2218–2243.
18 Mollica A., Paglialunga Paradisi M., Torino D., Spisani S. Lucente G. Hybrid
a/b-peptides: For-Met-Leu-Phe-OMe analogues containing geminally
disubstituted b2,2- and b3,3-amino acids at the central position. Amino
Acids 2006; 30: 453–459.
19 Yang S.-T., Shin S. Y., Lee C. W., Kim Y.-C., Hahm K.-S., Kim J. I. Selective
cytotoxicity following Arg-to-Lys substitution in tritrpticin adopting a
unique amphipathic turn structure. FEBS Lett. 2003; 540: 229–233.
DOI: 10.1016/s0014-5793(03)00266-7
20 Gottlieb H. E., Kotlyar V., Nudelman A. NMR chemical shifts of common lab-
oratory solvents as trace impurities. J. Org. Chem. 1997; 62: 7512–7515.
21 Mosmann T. Rapid colorimetric assay for cellular growth and survival:
application to proliferation and cytotoxicity assays. J. Immunol. Methods
1983; 65: 55–63.
23 Svenson J., Karstad R., Flaten G. E., Brandsdal B.-O., Brandl M. Svendsen
J. S. Altered activity and physicochemical properties of short cationic
antimicrobial peptides by incorporation of arginine analogues. Mol.
Pharmaceutics 2009; 6: 996–1005. DOI: 10.1021/mp900057k
24 Cronin J. R., Yuen G. U., Pizzarello S. Gas chromatographic-mass spec-
tral analysis of the five-carbon b-, g- and d-amino alkanoic acids. Anal.
Biochem. 1982; 124: 139–149.
25 Bodanszky M., Bodanszky A. The Practice of Peptide Synthesis. 2nd ed,
Berlin Heidelberg: Springer-Verlag, 1994.
26 Iwasaki T., Ishibashi J., Tanaka H., Sato M., Asaoka A., Taylor D.
Yamakawa M. Selective cancer cell cytotoxicity of enantiomeric
9-mer peptides derived from beetle defensins depends on negatively
charged phosphatidylserine on the cell surface. Peptides 2009; 30:
660–668.
27 Hagmann W. K. The many roles for fluorine in medicinal chemistry. J.
Med. Chem. 2008; 51: 4359–4369. DOI: 10.1021/jm800219f
28 Giménez D., Andreu C., Olmo M. d., Varea T., Diaz D., Asensio G. The
introduction of fluorine atoms or trifluoromethyl groups in short cat-
ionic peptides enhances their antimicrobial activity. Bioorgan. Med.
Chem. 2006; 14: 6971–6978.
29 Amin K., Dannenfelser R. M. In vitro hemolysis: guidance for the pharma-
ceutical scientist. J. Pharm. Sci. 2006; 95: 1173–1176. 10.1002/jps.20627
30 Adessi C., Soto C. Converting a peptide into a drug: strategies to im-
prove stability and bioavailability. Curr. Med. Chem. 2002; 9: 963–978.
31 Frackenpohl J., Arvidsson P. I., Schreiber J. V., Seebach D. The outstand-
ing biological stability of b- and g-peptides toward proteolytic enzymes:
an in vitro investigation with fifteen peptidases. Chembiochem 2001; 2:
445–455.
32 Arvidsson P. I., Ryder N. S., Weiss H. M., Hook D. F., Escalante J., Seebach D.
Exploring the atibacterial and hemolytic activity of shorter- and longer-
chain b-, a,b-, and g-peptides, and of b-peptides from b2-3-aza- and
b3-2-methylidene-amino acids bearing proteinogenic side chains – A
survey. Chem. Biodiversity 2005; 2: 401–420.
22 Eliassen L. T., Berge G., Svinbjørnsson B., Svendsen J. S., Vorland L. H.,
Rekdal ꢀ. Evidence for a direct antitumor mechanism of action of bo-
vine lactoferricin. Anticancer Res 2002; 22: 2703–2710.
wileyonlinelibrary.com/journal/jpepsci Copyright © 2012 European Peptide Society and John Wiley & Sons, Ltd. J. Pept. Sci. 2012; 18: 170–176