Novel cyclopeptides for the design of MMP directed delivery devices: A novel smart delivery paradigm

Article abstract of DOI:10.1007/s11095-010-0164-0

Purpose: Matrix metalloproteinases (MMP) are a family of proteolytic enzymes, the expression of which in a key step of tumor progression has been better defined recently. The studies highlighted the ongoing need for very specific inhibitors, substrates or release devices designed to be selective for one or at least very few MMPs. Methods: This report deals with the design, synthesis and in vitro evaluation of linear and especially novel cyclic peptidic moieties, embodying MMP cleavable sequences designed to answer these questions. FRET (fluorescence resonance energy transfer) labelling via chromophore-modified amino-acids was used to give access to enzyme kinetics. Results: Evaluation of these peptides showed that cyclisation gives rise to high specificity for certain MMP, suggesting that this approach could provide very specific MMP substrate. Moreover, cyclic structures present a very good plasma stability. Conclusions: These original derivatives could allow the design of MMP-controlled delivery devices, the specificity of which will be retained in complex biological media and in vivo.

Full text of DOI:10.1007/s11095-010-0164-0

Pharm Res (2010) 27:1713–1721
DOI 10.1007/s11095-010-0164-0
RESEARCH PAPER
Novel Cyclopeptides for the Design of MMP Directed Delivery
Devices: A Novel Smart Delivery Paradigm
El-Farouck Moustoifa & Mohamed-Anis Alouini & Arnaud Salaün & Thomas Berthelot & Aghleb Bartegi & Sandra Albenque-Rubio &
Gérard Déléris
Received: 23 February 2010 /Accepted: 19 April 2010 /Published online: 8 May 2010
#
Springer Science+Business Media, LLC 2010
Conclusions These original derivatives could allow the design
of MMP-controlled delivery devices, the specificity of which will
be retained in complex biological media and in vivo.
ABSTRACT
Purpose Matrix metalloproteinases (MMP) are a family of
proteolytic enzymes, the expression of which in a key step of
tumor progression has been better defined recently. The
studies highlighted the ongoing need for very specific inhibitors,
substrates or release devices designed to be selective for one
or at least very few MMPs.
.
.
.
KEY WORDS cancer cyclic peptide delivery device
.
fluorescent substrate MMP
Methods This report deals with the design, synthesis and in
vitro evaluation of linear and especially novel cyclic peptidic
moieties, embodying MMP cleavable sequences designed to
answer these questions. FRET (fluorescence resonance energy
transfer) labelling via chromophore-modified amino-acids was
used to give access to enzyme kinetics.
Results Evaluation of these peptides showed that cyclisation
gives rise to high specificity for certain MMP, suggesting that
this approach could provide very specific MMP substrate.
Moreover, cyclic structures present a very good plasma
stability.
ABBREVIATIONS
ACN
APMA
DAC
DCM
DIEA
DMF
acetonitrile
4-aminophenylmercuric acetate
7-diethylamino coumarin-3-carboxylic acid
methylene chloride
N,N′-diisopropyldiethylamine
N,N-dimethylformamide
DMSO
EDC
dimethylsulfoxide
1-ethyl-3-(3-dimethylaminopropyl) carbodiimide
hydrochloride
ECM
Et2O
Fmoc
FRET
HBTU
extracellular matrix
diethyl ether
(9H-fluoren-9-ylmethoxycarbonyl)
fluorescence resonance energy transfer
N-[1H-benzotriazol-1-yl)dimethylamino)methylene]-
N-methylmethanaminium hexafluorophosphate
N-oxide
:
:
:
:
E.-F. Moustoifa M.-A. Alouini A. Salaün S. Albenque-Rubio
G. Déléris (*)
Université Victor Segalen Bordeaux 2
CNRS UMR 5084 Bio-Organic Chemistry Group
146 rue Léo Saignat
F-33076 Bordeaux Cedex, France
e-mail: gerard.deleris@u-bordeaux2.fr
HOBt
HPLC
MALDI-
TOF
N-hydroxybenzotriazole
high performance liquid chromatography
matrix-assisted laser desorption/ionization—time
of flight
:
M.-A. Alouini T. Berthelot
MC
7-methoxy coumarin-3-carboxylic acid
Matrix metalloprotease
MMP inhibitor
CEA, IRAMIS, LSI Irradiated Polymers Group
UMR 7642 CEA/CNRS/Ecole Polytechnique
F-91128 Palaiseau Cedex, France
MMP
MMPI
MS
mass spectrometry
:
M.-A. Alouini A. Bartegi
NMM
PyBOP
tBu
N-methylmorpholine
benzotriazol-1-yl-oxy-tris-pyrrolidinophosphonium
t-butyl
Unité de Recherche 02/UR/09-01 Biochimie des Protéines
et Interactions Moléculaires
Avenue Taher Hadda, BP 74, Monastir 5000, Tunisie

1714
Moustoifa et al.
TFA
THF
TIS
TRIS
Trt
Trifluoroacetic acid
tetrahydrofuran
triisopropylsilane
tris(hydroxymethyl)aminomethane
trityl
solid phase peptide synthesis
the development of specific and selective substrate and/or
inhibitors for one or a group of MMPs still remains of high
interest for tumor targeting via mediation by MMP.
MMP activities have been previously addressed with
synthetic fluorogenic derivatives both as biochemical and
pharmacological tools, as well as for the design of targeted
delivery systems. They are most often FRET-based
substrates which were designed from natural collagen-
identified sequences cleaved by MMPs while assuming
additive thermodynamic contributions of different amino
acids. Experimental conclusions highlight that this rationale
is not fully convenient. To overcome this drawback,
selectivity was addressed by G.B. Fields and coworkers
with the synthesis of homotrimeric, fluorogenic triple-
helical peptide (THP) models which include MMP cleavage
sites (19).
Moreover, several delivery devices have also been
designed, using polymeric vehicles (20,21), hydrogel matrix
(22) or liposomal devices functionalized with peptidic
moieties embodying MMP cleavage sequences. These
previous studies confirmed that MMP-based release devices
offer opportunities even if they present poor specificity
(23,24). Moreover, as soon as peptidic structures are
involved, requiring collagen-mimicking helicoidal entities,
large numbers of amino-acids are included. Specific
delivery at tumor areas upon MMP activation obviously
remains a crucial challenge for the design of targeted
anticancer therapies or imaging. It requires that reasonable
size molecules, designed as MMP-cleavable devices, present
a high level of selectivity for a very small number of selected
MMP as well as high plasma stability.
SPPS
INTRODUCTION
Matrix Metalloproteinases (MMPs) are a family of Zinc
metalloenzymes involved in the catabolism of extracellular
matrix (ECM). The tissue remodelling occurs in normal
physiological processes such as development and morpho-
genesis but also during the course of diseases such as
arthritis, cardiovascular diseases and cancer. MMPs display
a large set of ECM and non-ECM substrates. Such
enzymes have long been associated with many types and
stages of cancer and were thought to be essential but
limited to basement-membrane penetration during tumor
progression and metastasis (1–3). The relationship between
specific MMP overexpression and tumor progression has
been very recently highlighted (4,5). It has been shown that
dramatic tumor escaping from antiangiogenic chemothera-
pies is correlated with MMP-2 and MMP-9 over-expression
concomitant with MMP-1 downregulation. MMP monitor-
ing could thus be envisaged as providing tools for diagnosis
or prognosis. MMP expression profile can be considered as
a tumor fingerprint. Matrix metalloproteinase are present
globally in very low concentration, but concentrated on the
surface of cells in tumor areas in high concentrated and
activated form (6). For that reason, research for MMP
inhibitors was very intensive about a decade ago. Several
MMP-inhibitor (MMPI) drugs advanced up to phase III
clinical trials in patients with advanced cancer, but the trials
failed mainly because of muscle and joint pains which
prevented reaching their end points of increased survival
(7–10).
One of the most frequently used approaches to improve
peptide stability in biological medium is the cyclization of
linear peptides, which prevents unspecific degradation by
proteases. But direct cyclization of linear fluorogenic
FRET-based peptides is not an acceptable approach, as
the hydrolysis of sequence does not result in significant
change of donor/acceptor proximity and thus does not
allow enzymatic cleavage to be measured.
These results highlight the fact that previous MMPI
development programs were initiated without adequate
target validation and without identifying in vivo MMP
substrates or their physiological roles. As previously pointed
out, MMPs promote tumor progression not only through
single ECM degradation, but also through a number of
signaling functions (11–14). They are submitted to highly
specific over-expression in dramatic phenomena linked to
invasive phenotypic changes following anti-angiogenic
therapies (4). The complexity of MMP over-expression
and activities highlights their important role in metastatic
diseases, especially in highly aggressive late-stage tumors
with poor clinical outcome (15–18) and in dramatic
situations when chemotherapy escaping occurs. Therefore,
For these reasons, we herein propose a novel approach
based on the design of cyclic peptides which incorporate
two MMP substrate sequences. In order to test them, each
fluorescent amino acid of a FRET probes pair was
introduced between the substrate sequences. When hydro-
lyzed, the cyclic peptide releases fluorescent probes, and
change in FRET signal provides access to cleavage kinetics.
In order to validate the proof of concept for this approach,
linear and cyclic MMP substrate peptides were synthesized
and evaluated both in vitro and in plasmatic medium. We
focused our attention on MMP-1 for its implication in
metastatic process (15,25,26) and MMP-2 and MMP-9 for
their over-expression in numerous steps of tumor progres-
sion (27).

Novel Cyclopeptides as MMP-Directed Delivery Devices
1715
MATERIALS AND METHODS
General Methods
Peptides were cleaved from the resin by treatment with
TFA/H2O/TIS (95:2.5:2.5) mixture for 1 h 30 at room
temperature. The resin was removed by filtration and washed
with TFA (10 mL). The filtrate was evaporated under reduced
pressure. The product was precipitated with cold diethyl ether.
The crude peptides were collected by filtration and purified by
preparative HPLC (C18) to yield S1 (55 mg) MALDI-TOF
m/e (M+H+) 1484.64 (theoretical:1483.73 Da), S2 (96 mg)
MALDI-TOF m/e (M+H+) 1541.56 (theoretical:1540.69 Da)
and S9 (102 mg) MALDI-TOF m/e (M+H+) 1411.65
(theoretical:1410.68 Da).
All standard chemicals and solvents were of analytical grade
and purchased from Sigma Aldrich. 2-Chloro chlorotrityl resin
and all Fmoc-amino acids were purchased from Novabiochem.
Absorption spectra were recorded on a U-2010 Hitachi
spectrophotometer (Hitachi instruments, USA). Enzymatic
assays were recorded on a Luminescence Spectrometer
model LS55 with a 4-position automatic cell changer
including water thermostatting and strirring for each
sample position. Solid Phase Peptide Synthesis was carried
out on an Applied Biosystems 433A automated peptide
synthesizer. Reverse-phase HPLC was performed on a
Hitachi LaChrom Elite equipped with an Organizer, Diode
Array detector L-2450, Autosampler L-2200, pump L-2130
with a SATISFACTION RP18AB 5 µm 250×4.6 mm C18
column for analytical session. Preparative HPLC was
performed on a Shimadzu instrument equipped with a
SCL-10 AVP system controller, LC8A HPLC pumps and
SPD-10 AVP UV-vis detector probing at 214 nm on a
SATISFACTION RP18AB 5 µm 250×20 mm C18
column (C.L.I Cluzeau) for preparation. The following
solvent systems were used for the elution in a linear
gradient mode at a flow rate of 1 or 15 ml/min (for
analytical and preparative HPLC respectively): (A) 0.1%
aqueous trifluoroacetic acid (TFA) and (B) 0.1% TFA in
70% aqueous acetonitrile (ACN). MALDI-TOF mass
spectrometry was performed on a BRUKER BIFLEX III
mass spectrometer (CESAMO, Bordeaux, France)
equipped with a nitrogen laser (337 nm, 3 ns pulse width).
Synthesis of Sc1, Sc2 and Sc9 by on Resin Cyclisation
Sc1, Sc2 and Sc9 were synthesized on a 0.2 mmol scale
using Fmoc strategy. Fmoc-Glu-OAllyl (818 mg, 2 mmol)
was dissolved in dry DCM (10 mL). After complete
dissolution, DIEA (367 µL, 2.2 mmol) was added. After
10 min, 2-chloro chlorotrityl resin (1 g, 1.3 mmol/g) was
added. The resulting suspension was stirred at room
temperature for 5 h. The resin was filtered off and
successively washed with DCM (3×10 mL), DCM/metha-
nol/DIEA (17/2/1) (4×10 mL), DMF (2×10 mL) and
DCM (3×10 mL). Resin was dried under high vacuum
over KOH. 0.23 mmol/g substitution level was estimated
by UV determination of the concentration of liberated
dibenzofulvene after Fmoc group cleavage with piperidine.
The Fmoc-Glu-OAllyl-trityl resin was packed in the
reaction column of the automatic peptide synthesizer
(437 mg, 0.1 mmol). Nα-Fmoc amino acids (Gly, Leu, Pro,
Gln(Trt), Ala, Val, Lys(DAC), Lys(MC)) were used in a
four-fold excess using HBTU in the presence of HOBt and
DIEA. The synthesis was performed without capping,
giving N-protected peptide-bound resin. This peptidyl resin
was dried at 40°C under high vacuum for 4 h and then
flushed with a stream of argon. Pd(PPh3)4 (693 mg,
0.6 mmol) was dissolved by bubbling a stream of Ar in a
mixture of chloroform/acetic acid/N-methylmorpholine
(8 mL, 37:2:1). Then, this mixture was transferred into
reaction column containing the resin and was occasionally
stirred with gentle agitation for 2 h at room temperature.
Resin was filtered off and washed consecutively with 0.5%
DIEA in DMF (3×10 mL) and sodium diethyldithiocarba-
mate (0.5% w/w) in DMF (3×10 mL) to remove the
catalyst. Resin was successively washed with DMF (2×
10 mL) and DCM (3×10 mL). After removal of Fmoc
protecting group, resin was washed with 1 M HOBt in
DMF. Peptidyl resin was mixed with a solution of PyBOP
(260 mg, 0.5 mmol), HOBt (69 mg, 0.5 mmol) and DIEA
(180 µL, 1 mmol) in N-methylpyrrolidone (10 mL). Mixture
was swelled at room temperature for 48 h. Peptidyl resin
was washed with DMF (2×10 mL) and DCM (3×10 mL).
Peptides were cleaved from the resin by treatment with
Fluorescent Probes
Fluorescent amino acids, Nε-(7-methoxycoumarin-3-car-
boxyl)-L-Fmoc lysine (i.e. Fmoc-Lys(MC)-OH) and Nε-(7-
diethylamninocoumarin-3-carboxyl)-L-Fmoc lysine (i.e.
Fmoc- Lys(DAC)-OH) were synthesized as previously
described by Berthelot et al. (28).
Peptide Synthesis (Table I)
Synthesis of S1, S2 and S9
S1, S2 and S9 were synthesized using Fmoc strategy. A
preloaded resin Fmoc-Ala-Wang Resin (0.72 mmol/g) was
used for the synthesis (0,25 mmol; 350 mg). Nα-Fmoc
amino acids (Gly, Leu, Pro, Gln(Trt), His(Trt), Tyr(tBu),
Ala, Val, Lys(DAC), Lys(MC)) were used in a four-fold
excess using HBTU in the presence of HOBt and DIEA.
The synthesis was performed without capping, giving N-
deprotected peptide-bound resin.

1716
Moustoifa et al.
Table I Structure of Linear (S1, S2, S9) and Cyclic (Sc1, Sc2, Sc9) Peptides
Peptide
Structure
S1
H2N-Lys(DAC).Gly-Pro-Gln-Gly-Leu-Leu-Gly-Ala-Lys(MC)-Ala-COOH
H2N-Lys(DAC)-Pro-Pro-Gly-Ala-Tyr-His-Gly-Ala-Lys(MC)-Ala-COOH
H2N-Lys(DAC)-Gly-Pro-Gly-Gly-Val-Val-Gly-Pro-Lys(MC)-Ala-COOH
S2
S9
Sc1
Sc2
Sc9
Cyclo(Gly-Pro-Gln-Gly-Leu-Leu-Gly-Ala-Lys (DAC)-Gly-Pro-Gln-Gly-Leu-Leu-Gly-Ala-Lys(MC)-Glu)
Cyclo(Pro-Pro-Gly-Ala-Tyr-His-Gly-Ala-Lys(DAC)-Pro-Pro-Gly-Ala-Tyr-His-Gly-Ala-Lys(MC)-Glu)
Cyclo(Gly-Pro-Gly-Gly-Val-Val-Gly-Pro-Lys( DAC)-Gly-Pro-Gly-Gly-Val-Val-Gly-Pro-Lys(MC)-Glu)
TFA/H2O/TIS (95:2.5:2.5) mixture for 1 h 30 at room
temperature. Crude peptide was collected by filtration and
purified by HPLC (C18). Preparative HPLC yielded Sc1
(6 mg) MALDI-TOF m/e (M+Na+) 2240.13 (theoretical:
2217.11 Da) and Sc9 (12 mg) MALDI-TOF m/e (M+H+)
2072.67 (theoretical: 2071.005 Da) .
In order to perform cyclization, crude peptide (0.25 mmol)
was mixed with 500 mL DCM, EDC (1-ethyl-3-(3-dimethy-
laminopropyl) carbodiimide hydrochloride) (230 mg,
1.2 mmol) and HOBt (183.6 mg, 1.2 mmol) for 48 h. The
final estimated concentration of peptide was close to 0.5 mM.
After evaporation under reduced pressure to around 10 mL,
20 mL were added to wash the organic solution. After
extraction and total evaporation of DCM, the sample was
triturated and filtered with cold diethyl ether to reduce the
presence of HOBt. Peptides were fully deprotected by
treatment with TFA/H2O/TIS (95:2.5:2.5) mixture for 1 h
at room temperature. TFA was evaporated under reduced
pressure, and the product was precipitated with cold diethyl
ether. Crude peptide was collected by filtration and purified
by preparative HPLC (C18) to yield Sc1 (18 mg) MALDI-
TOF m/e (M+Na+) 2240.13 (theoretical: 2217.11 Da), Sc2
(135 mg) MALDI-TOF m/e (M+H+) 2332.00 (theoretical:
2331.038 Da) and Sc9 (61 mg) MALDI-TOF m/e (M+H+)
2072.67 (theoretical: 2071.005 Da).
Synthesis of Sc1, Sc2 and Sc9 by Cyclisation in Solution
Sc1 , Sc2 and Sc9 were synthesized on a 0.25 mmol scale
using Fmoc strategy. Fmoc-Glu(OtBu)-OH (850 mg,
2 mmol) was dissolved in dry DCM (10 mL). After complete
dissolution, DIEA (367 µL, 2.2 mmol) was added. After
10 min, 2-chloro chlorotrityl resin (1 g, 1.3 mmol/g) was
added. The resulting suspension was stirred at room
temperature for 5 h. The resin was filtered off and
successively washed with DCM (3×10 mL), DCM/metha-
nol/DIEA (17/2/1) (4×10 mL), DMF (2×10 mL) and
DCM (3×10 mL). Resin was then dried under high
vacuum over KOH. 0.76 mmol/g substitution level for
this preloaded resin was estimated by UV determination
of the concentration of liberated dibenzofulvene, after the
cleavage of the Fmoc group with piperidine. Fmoc-Glu
(OtBu)-trityl resin (329 mg) was packed into the reaction
column of the automatic peptide synthesizer. Nα-Fmoc
amino acids (Gly, Leu, Pro, Gln(Trt), His(Trt), Tyr(tBu),
Ala, Val, Lys(DAC), Lys(MC)) were used in a four-fold
excess using HBTU in the presence of HOBt and DIEA.
Synthesis was performed without capping, giving N-depro-
tected peptide-bound resin. Repetitive treatment of pep-
tidyl resin with a 0.6% TFA solution in dichloromethane
(16×10 mL) was carried out in a sealable sintered glass
funnel (1 min per treatment). Resin was filtered into a flask
containing 10% of pyridine in DCM (2 mL of pyridine in
20 mL of DCM). After evaporation under reduced
pressure, 40 mL of water were added and the mixture
cooled with ice to aid precipitation of the product. In the
case of Sc9, the crude peptide was washed with 30 mL
water without adding organic solvent. Resulting product
was dried overnight.
Fluorescence Spectra
All spectra were carried out in a total 2 mL volume. Slit
width was 3.4 for emission and 12 for the excitation.
Spectra were recorded in 0.1 M TRIS, 0.1 M NaCl,
10 mM CaCl2, 0.05% of Igepal(CA-630), pH=7.4. Peptide
concentrations were calculated using the molar extinction
coefficient of MC and DAC in peptide: λ360nm (MC)=
18,500 M−1cm−1 and λ436nm (DAC)=26,300 M−1cm−1 for
linear peptide. This coefficient is different in cyclic peptide,
λ360nm (MC)=10,714 M−1cm−1 and λ436nm (DAC)=
19,291 M−1cm−1
.
Enzyme Assays
MMP assays were performed in the same buffer described
below. 20 µL aliquot of 220 nM commercial MMP-1,
MMP-2 and MMP-9 as zymogen form were activated with
10 µM solution of 4-aminophenylmercuric acetate (APMA)
(2 µL, Buffer TRIS 0,1 M pH=9) at 37°C for 3 h. Stock

Novel Cyclopeptides as MMP-Directed Delivery Devices
1717
peptide solutions were prepared in DMSO (3 mg peptide in
200 µL pure DMSO). For each measurement, peptide
concentration was calculated using the molar extinction
coefficient previously described. All assays were carried out
in a total 2 mL volume with 2 nM activated MMP. After
addition of enzyme to initiate reaction, the initial rate of
substrate hydrolysis was determined by monitoring the
increase in 401 nm fluorescence emission using a 350 nm
excitation wavelength. Initial velocities were calculated in
order to fit a Lineweaver-Burk plot for determination of
kinetics parameter Kcat and Km.
to monitor MMP activity (29). For that purpose, our
substrates were designed from described sequences arising
from Nagase et al. (30), which were described as MMP
natural cleavage sites (Table II).
Peptide Synthesis
Fluorogenic amino acids were introduced as any other
amino acid in the sequence, by solid phase synthesis, in
order to provide the corresponding FRET peptide (Fig. 1).
Linear fluorogenic substrates S1, S2 and S9 were obtained
in satisfactory yields (Table III).
Plasma Assays
As tridimensional structural requirements are impor-
tant for proteolytic activity and selectivity of MMPs, the
design of conformationnally constrained derivatives was
an obvious target in our strategy. We herein report the
first study involving conformationally restrained peptides
after cyclization.
Cyclic moieties were first synthesized through a
“head-to-tail” on resin cyclization. This synthetic ap-
proach is based upon the anchoring of the peptide to the
resin through a side chain. Glutamic acid was used,
protected with allyl group as an orthogonal α-carboxyl
protection (34–36). Solid-phase synthesis of this cyclo-
peptide was performed on 2-chloro chlorotrityl resin.
Loading of the resin was achieved to yield a low loaded
resin (0.23 mmol/g) in order to improve intramolecular
cyclization. Unfortunately, only cyclopeptides Sc1, Sc9
were obtained in poor yields (Table IV) after a tricky
HPLC purification step.
Because of the poor yields obtained from on-resin
cyclisations, we prepared cyclic peptide through in-
solution cyclisations. Fmoc-Glu(OtBu)-OH was loaded on
a 2-chloro chlorotrityl resin. After automatic solid phase
peptide synthesis, the fully protected peptide was removed
from resin with diluted TFA (1%) on DCM. Cyclisation
was then performed at 0.5 mM of peptide in presence of
EDC and HOBt during 48 h in DCM (Fig. 2). No
oligomeric fraction was observed in these conditions, and
cyclic peptides were obtained with better yield after classical
purification with water and acetonitrile/water (70/30)
mixtures in presence of 0.1% TFA (Table IV). Even if
peptide cyclization appeared to be more efficient in
solution, cyclization step appeared here again to be strongly
dependent on peptide sequence.
Blood was collected from voluntary donor at “Etablisse-
ment Français du Sang” (Bordeaux Site). Donors were
tested and determined negative for HIV-1 and HIV-2,
Hepatitis B and C HTLV-I/II viruses, and syphilis.
Freshly collected plasma (immediately after control)
containing 1.8 mg/ml EDTA 2 K (ethylene diamine
tetraacetic acid dipotassium salt) were centrifuged 15 min
at 1300 x g. Plasma was then isolated. For every
experiment, plasma tubes coming from three donors
were pooled. All assays were carried out in a total 2 mL
volume with 4 µl of crude plasma and 4 µM substrate
added. The increase in fluorescence emission was
monitored during 1 h at 401 nm using a 350 nm
excitation wavelength. For the corresponding assays,
activated MMP (1, 2 or 9) was subsequently added to
the solution, and the fluorescence emission was moni-
tored as described above again for 1 h.
RESULTS AND DISCUSSION
Novel FRET peptidic substrates were obtained by incor-
poration of fluorescent amino acids, Nε-(7-methoxycou-
marin-3-carboxyl)-L-Fmoc lysine (Fmoc-Lys(MC)-OH) as
the donor and Nε-(7-diethylamninocoumarin-3-carboxyl)-L-
Fmoc lysine (Fmoc- Lys(DAC)-OH) as acceptor (28). These
probes appear to be attractive due to their extended
spectral range, high emission quantum yields, photostabil-
ity, good solubility in the safest solvents and easy incorpo-
ration in a peptide sequence through solid phase peptide
synthesis (SPPS) (29). These moieties previously allowed us
Table II Natural MMP Cleavage
Sequence
Sites
Natural substrate
MMP
Ref
Gly-Pro-Gln-Gly775∼Leu776-Leu-Gly-Ala
Pro-Pro-Gly-Ala62∼Ty r 63-His-Gly-Ala
Gly-Pro-Gly-Gly439∼Val440-Val-Gly-Pro
α2 chain—Human collagen I
Human galectin-3
MMP-1,−8
MMP-2
(31)
(32)
(33)
α1 chain - Human collagen XI
MMP-9

1718
Moustoifa et al.
Fig. 1 Synthesis of linear MMP-1,
MMP-2 and MMP-9 substrates.
Fmoc-Ala-Wang preloaded resin
was used for synthesis.
Enzymatic Assays
to S2, S9 did not undergo significant cleavage when treated
with MMP-1, while there is no significant difference
Linear peptide embodying Lys (DAC) and Lys (MC) at N
and C terminus revealed to be slightly hydrophobic. A stock
solution of peptide in DMSO had yet to be achieved. Final
concentration of DMSO (0.04–0.01% v/v) in the assay
solution should not interfere with enzyme performance
(37).
Peptides were treated with 2 nM of each activated
enzyme (MMP-1, MMP-2 and MMP-9). Initial rates of
hydrolysis were determined at various peptide concentra-
tions. Data were used to construct Lineweaver-Burk plots in
order to calculate kcat and Km (Fig. 3).
The second-order rate constant, kcat/Km, was deter-
mined for each enzyme and each substrate (Table V).
Suprisingly, S1 from collagenase 1 natural substrate
sequence exhibited a high affinity for gelatinases MMP2
and MMP-9. kcat/Km values were, respectively, 5.0*106
M−1.s−1 and 3.4*106M−1.s−1 for MMP-2 and MMP-9.
This represents a 20-fold increased ratio for MMP-2 and
between MMP-2 and MMP-9 (respectively, 6.2*104M−1
.
s−1 and 3.8*104M−1.s−1). Thus, S2 and S9 exhibited a very
good selectivity for gelatinase family (MMP-2 and MMP-9),
and S2 offers a good specificity for MMP-2. The use of
coumarin-modified amino acids seems to play a key role in
the affinity of these peptides to the enzyme catalytic site.
When focusing on cyclic peptides, we could observe that
Sc1 cleavage properties versus MMP-1 were roughly
unchanged compared with S1 (1.7*105M−1.s−1). However,
interestingly, MMP-2 and MMP-9 activities on this deriv-
ative were notably decreased with respect to kcat/Km
6.1*105M−1.s−1 and 6.9*105M−1.s−1
.
While almost no selectivity could be observed between
the three MMP towards Sc9, cyclic peptide Sc2 was only
recognized by MMP-9.
Indeed, no significant cleavage was observed either with
MMP-1 or MMP-2, while kcat/Km ( 9.3*105M−1.s−1) is
only two times less than the value of linear form versus this
enzyme.
14-fold for MMP-9 with regard to MMP-1 (2.5*105M−1
.
s−1). S2 did not undergo significant cleavage when treated
with MMP-1. As expected, S2 is three times more
recognized by MMP-2 with respective values 6.1*105M−1
.
Table IV Cyclic Substrates Yields from On-Resin (A) and In-Solution (B)
Cyclisations
s−1 and 1.9*105M−1.s−1 for MMP-2 and MMP-9. Similar
Compound Sequence
Yield %
A
B
Table III Sequences of Novel Substrates
Sc1
Sc2
Sc9
Cyclo[GPQGLLGAK(DAC)GPQGLLGAK
(MC)E]
Cyclo[PPGAYHGAK(DAC)PPGAYHGAK
(MC)E]
Cyclo[GPGGVVGPK(DAC)GPGGVVGPK
(MC)E]
2.6
–
3.5
23
Compounds
Sequences
Yields
S1
S2
S9
H2N-K(DAC)GPQG∼LLGAK(MC)A-COOH
H2N-K(DAC)PPGA∼YHGAK(MC)A-COOH
H2N-K(DAC)GPGG∼VVGPK(MC)A-COOH
27%
46%
42%
5.8 11

Novel Cyclopeptides as MMP-Directed Delivery Devices
1719
Fig. 2 Cyclopeptides synthesis
via in solution cyclization. 2-
chloro chlorotrityl resin was used
for synthesis.
These results highlight that reasonably sized novel
modified structures can reach a very high level of selectivity
towards one specific MMP.
Plasma Stability
Table V Kinetic Parameters of Enzyme (N.S.C: No Significant Cleavage)
Substrate
kcat (s−1
)
Km (µM)
kcat/Km (M−1.s−1
)
The use of our substrate in a drug delivery system implies a
relatively good stability in biological sample in order to
avoid non-specific release.
MMP-1
S1
1.57
6.3
2.5*105
N.S.C
S2
N.S.C
N.S.C
1.09
N.S.C
N.S.C
6.3
S9
N.S.C
1.7*105
Sc1
Sc2
Sc9
MMP-2
S1
N.S.C
0.28
N.S.C
5.4
N.S.C
5.2*104
15.8
2.84
0.29
2.19
N.S.C
0.18
3.1
5,0*106
6.1*105
6.2*104
6.5*105
N.S.C
S2
4.6
S9
4.7
Sc1
Sc2
Sc9
MMP-9
S1
3.39
N.S.C
3.1
5.6*104
17.6
0.96
0.43
5.11
0.95
0.33
5.1
3.4*106
1.9*105
3.8*104
6.9*105
9.3*104
4.9*104
S2
4.9
S9
11.3
7.4
Sc1
Sc2
Sc9
10.2
6.8
Fig. 3 Lineweaver-Burk analysis of S1 hydrolysis by activated MMP-1
(2 nM). Fluorometric assay monitoring at 37°C, and substrate concentra-
tion range 1-4 µM.

1720
Moustoifa et al.
Fig. 4 Substrates digestion in plasma. Peptides were treated for 1 h with
4 µl crude plasma.
Fig. 6 Stability of S2 and Sc2 in plasma and subsequent treatment with
MMP-1, MMP-2 or MMP-9. After 1 h plasma digestion, activated MMP-1,
MMP-2 or MMP-9 was added to Sc2 assay.
In order to evaluate the stability, substrates were treated
with plasma for 1 h (Fig. 4).
When activated MMP-1 or MMP-2 were added to the
mixture, peptide cleavage remained very limited. On the
contrary, hydrolysis of this specific substrate increased
substantially upon addition of activated MMP-9.
While linear compounds underwent rapid cleavage from
aspecific plasma proteases, cleavage kinetics for Sc1 and
Sc9, respectively, decreased by a 10 and 13 factor. Sc2 did
not significantly cleave during the time of assays.
Addition of activated MMP-9 (1nM) into the previous
assays led to a dramatic increase in cyclic peptide hydrolysis
(Fig. 5). Results correlate with the ones obtained with pure
enzymes, i.e Sc1 is more sensitive than Sc2 or Sc9 to the
presence of activated MMP-9.
CONCLUSION
Proposed design of original chemical moieties highlights a
novel paradigm for specific delivery at tumor areas.
Original reasonably sized cyclo-peptidic derivatives proved
to be specifically cleaved by only one or at least selectively
by one MMP. It has been proved that selective MMP over-
expression is correlated with crucial stages or tumor
progression. Therefore, these derivatives could behave as
smart delivery devices at the zone of MMP expression
under MMP activity itself, due to their plasma stability.
Our results show that other proteolytic enzymes could be
targeted after optimization.
Results presented in Fig. 6 clearly demonstrated that,
while linear S2 is immediately proteolyzed within plasma,
MMP addition highlights a very specific cleavage of Sc2 by
only one MMP, MMP-9 here.
Design of delivery devices will involve either covalent
binding through glutamic acid side chain or via supramo-
lecular design. Indeed, we have data (not published)
proving that coumarine aromatic part may play an essential
role in device stabilization through host guest interactions.
ACKNOWLEDGMENTS
Authors wish to warmly thank Dr. Pierre Voisin and Dr.
Philippe Mellet for helpful discussions, Collectivité Dépar-
tementale de Mayotte, Conseil Régional d’Aquitaine and
Ligue Contre le Cancer for financial support.
Fig. 5 Addition of 1nM activated MMP-9 to assay described Fig. 5. After
1 h digestion with plasma, activated MMP-9 was added to Sc1 and Sc9
plasmatic assay.

Novel Cyclopeptides as MMP-Directed Delivery Devices
1721
20. Vartak DG, Gemeinhart RA. Matrix metalloproteases: underu-
tilized targets for drug delivery. J Drug Target. 2007;15:1–20.
21. Chau Y, Tan FE, Langer R. Synthesis and characterization of
dextran-peptide-methotrexate conjugates for tumor targeting via
mediation by matrix metalloproteinase II and matrix metal-
loproteinase IX. Bioconjug Chem. 2004;15:931–41.
22. Tauro JR, Lee BS, Lateef SS, Gemeinhart RA. Matrix
metalloprotease selective peptide substrates cleavage within hydrogel
matrices for cancer chemotherapy activation. Peptides. 2008;29:1965–
73.
REFERENCES
1. Liotta LA. Metastatic potential correlates with enzymatic
degradation of basement membrane collagen. Nature.
1980;284:67–8.
2. Stetler-Stevenson WG. Progelatinase A activation during tumor
cell invasion. Invasion Metastasis. 1994;14:259–68.
3. Nagaseand H, Woessner JF. Matrix metalloproteinases. J Biol
Chem. 1999;274:21491–4.
4. Lucio-Eterovic AK, Piao Y, de Groot JF. Mediators of glioblas-
toma resistance and invasion during antivascular endothelial
growth factor therapy. Clin Cancer Res. 2009;15:4589–99.
5. Roy R, Louis G, Loughlin KR, Wiederschain D, Kilroy SM,
Lamb CC et al. Tumor-specific urinary matrix metalloprotei-
nase fingerprinting: identification of high molecular weight
urinary matrix metalloproteinase species. Clin Cancer Res.
2008;14:6610–7.
6. Brooks PC, Stromblad S, Sanders LC, von Schalscha TL, Aimes
RT, Stetler-Stevenson WG et al. Localization of matrix metal-
loproteinase MMP-2 to the surface of invasive cells by interaction
with integrin alpha v beta 3. Cell. 1996;85:683–93.
23. Tauro JR, Gemeinhart RA. Matrix metalloprotease triggered
delivery of cancer chemotherapeutics from hydrogel matrices.
Bioconjug Chem. 2005;16:1133–9.
24. Tauro JR, Gemeinhart RA. Extracellular protease activation of
chemotherapeutics from hydrogel matrices: a new paradigm for
local chemotherapy. Mol Pharm. 2005;2:435–8.
25. Murray GI, Duncan ME, O'Neil P, Melvin WT, Fothergill JE.
Matrix metalloproteinase-1 is associated with poor prognosis in
colorectal cancer. Nature Med. 1996;2:461–2.
26. Overall CM, Kleifeld O. Validating matrix metalloproteinases as
drug targets and anti-targets for cancer therapy. Nat Rev Cancer.
2006;6:227–39.
7. Zucker S, Cao J, Chen WT. Critical appraisal of the use of matrix
metalloproteinase inhibitors in cancer treatment. Oncogene.
2000;19:6642–50.
8. Coussens LM, Fingleton B, Matrisian LM. Matrix metalloprotei-
nase inhibitors and cancer: trials and tribulations. Science (New
York, NY). 2002;295:2387–92.
27. Folgueras AR, Pendas AM, Sanchez LM, Lopez-Otin C. Matrix
metalloproteinases in cancer: from new functions to improved
inhibition strategies. Int J Dev Biol. 2004;48:411–24.
28. Berthelot T, Lain G, Latxague L, Deleris G. Synthesis of novel
fluorogenic L-Fmoc lysine derivatives as potential tools for
imaging cells. J Fluoresc. 2004;14:671–5.
9. Overall CM. Molecular determinants of metalloproteinase sub-
strate specificity: matrix metalloproteinase substrate binding
domains, modules, and exosites. Mol Biotechnol. 2002;22:51–86.
10. Fingleton B. Matrix metalloproteinase inhibitors for cancer
therapy: the current situation and future prospects. Expert Opin
Ther Targets. 2003;7:385–97.
11. McCawley LJ, Matrisian LM. Matrix metalloproteinases: they’re not
just for matrix anymore! Curr Opin Cell Biol. 2001;13:534–40.
12. Aoki T, Sumii T, Mori T, Wang X, Lo EH. Blood-brain barrier
disruption and matrix metalloproteinase-9 expression during
reperfusion injury: mechanical versus embolic focal ischemia in
spontaneously hypertensive rats. Stroke. 2002;33:2711–7.
13. Overall CM. Dilating the degradome: matrix metalloproteinase-2
cuts to the heart of the matter. Biochem J. 2004;383:E5–7.
14. Parks WC, Wilson CL, Lopez-Boado YS. Matrix metalloprotei-
nases as modulators of inflammation and innate immunity. Nature
Rev Immunol. 2004;4:617–29.
29. Berthelot T, Talbot JC, Lain G, Deleris G, Latxague L.
Synthesis of N epsilon-(7-diethylaminocoumarin-3-carboxyl)-
and N epsilon-(7-methoxycoumarin-3-carboxyl)-L-Fmoc lysine as
tools for protease cleavage detection by fluorescence. J Pept Sci.
2005;11:153–60.
30. Nagase H, Gregg BF. Human matrix metalloproteinase specificity
studies using collagen sequence-based synthetic peptides. Pept Sci.
1996;40:399–416.
31. Fields GB. A model for interstitial collagen catabolism by
mammalian collagenases. J Theor Biol. 1991;153:585–602.
32. Ochieng J, Fridman R, Nangia-Makker P, Kleiner DE, Liotta LA,
Stetler-Stevenson WG et al. Galectin-3 Is a Novel Substrate for
Human Matrix Metalloproteinases-2 and-9. Biochemistry.
2002;33:14109–14.
33. Niyibizi C, Chan R, Wu JJ, Eyre D. A 92 kDa gelatinase (MMP-
9) cleavage site in native type V collagen. Biochem Biophys Res
Commun. 1994;202:328–33.
15. Minn AJ. Genes that mediate breast cancer metastasis to lung.
Nature. 2005;436:518–24.
16. Minn AJ, Kang Y, Serganova I, Gupta GP, Giri DD, Doubrovin
M et al. Distinct organ-specific metastatic potential of individual
breast cancer cells and primary tumors. J Clin Invest.
2005;115:44–55.
17. Radisky DC, Levy DD, Littlepage LE, Liu H, Nelson CM, Fata
JE et al. Rac1b and reactive oxygen species mediate MMP-3-
induced EMT and genomic instability. Nature. 2005;436:123–7.
18. Weigelt B, Peterse JL, van 't Veer LJ. Breast cancer metastasis:
markers and models. Nat Rev Cancer. 2005;5:591–602.
19. Lauer-Fields JL, Broder T, Sritharan T, Chung L, Nagase H,
Fields GB. Kinetic analysis of matrix metalloproteinase activity
using fluorogenic triple-helical substrates. Biochemistry.
2001;40:5795–803.
34. Flouzat C, Marguerite F, Croizet F, Percebois M, Monteil A,
Combourieu M. Solid-phase synthesis of “head-to-side chain”
cyclic tripeptides using allyl deprotection. Tetrahedron Lett.
1997;38:1191–4.
35. Planas M, Bardají E, Barany G. Synthesis of cyclic peptide
hybrids with amino acid and nucleobase side-chains. Tetrahedron
Lett. 2000;41:4097–100.
36. Napolitano A, Bruno I, Rovero P, Lucas R, Peris MP, Gomez-
Paloma L et al. Synthesis, structural aspects and bioactivity of the
marine cyclopeptide hymenamide C. Tetrahedron. 2001;57:6249–
55.
37. Gogly B, Groult N, Hornebeck W, Godeau G, Pellat B. Collagen
Zymography as a Sensitive and Specific Technique for the
Determination of Subpicogram Levels of Interstitial Collagenase.
Anal Biochem. 1998;255:211–6.