Photodynamic Molecular Beacons triggered by MMP-2 and MMP-9: Influence of the distance between photosensitizer and quencher onto photophysical properties and enzymatic activation

Article abstract of DOI:10.2174/092986712803833128

Angiogenesis is a key step in the tumoral progression process. It is characterized by an over-expression of a number of matrix metalloproteinases (MMP). Among these MMPs, gelatinases (MMP-2 and MMP-9) are known to play a critical role in tumor angiogenesis and the growth of many cancers. Photodynamic Molecular Beacons (PMB) can be designed for cancer treatment by associating a chlorin-like photosensitizer and a black hole quencher linked by a gelatinase substrate peptide with the aim of silencing photosensitizer toxicity in non-targeted cells and restore its toxicity only in surrounding gelatinases. This article provides a report on the synthesis and photophysical and biochemical studies of new families of PMB, using tetraphenylchlorin and a black hole quencher as a donor-acceptor pair, and MMP specific sequence (H-Gly-Pro-Leu-Gly-Ile-Ala-Gly-Gln-Lys-OH or H-Pro-Leu-Gly-Leu-OH) to keep them in close proximity. Different spacers were used to evaluate the influence of the distance between the photosensitizer and the quencher on the photophysical properties and enzymatic activation of the PMB. Time-resolved quenching experiments were performed and FRET energy transfer could be observed. Photosensitizers' triplet state band in transient absorption disappears in PMB. However, even if both MMP-2 and MMP-9 were found to efficiently cleave the peptide alone, no cleavage was observed for all PMB. Further studies would be required to assess the ability of the PMB constructs to retain the sensitivity of the peptide linker to be cleaved by matrix metalloproteinases.

Full text of DOI:10.2174/092986712803833128

Send Orders of Reprints at reprints@benthamscience.org
5580
Current Medicinal Chemistry, 2012, 19, 5580-5594
Photodynamic Molecular Beacons Triggered by MMP-2 and MMP-9: Influence of
the Distance Between Photosensitizer and Quencher onto Photophysical Properties
and Enzymatic Activation
M. Verhille1,2,3,4, H. Benachour5,6, A. Ibrahim7, M. Achard3,4, P. Arnoux1,2, M. Barberi-Heyob5,6,8,9
J.-C. André1,2, X. Allonas7, F. Baros1,2, R. Vanderesse3,4,8 and C. Frochot1,5,8
,
1Université de Lorraine, Laboratoire Réactions et Génie des Procédés (LRGP), UPR 3349, Nancy, F-54000, France; 2CNRS, LRGP,
3
UPR 3349, Nancy, F-54000, France; Université de Lorraine, Laboratoire de Chimie Physique Macromoléculaire (LCPM), UMR
4
5
CNRS 7568, Nancy, F-54000, France; CNRS, LCPM, UMR CNRS 7568, Nancy, F-54000, France; Université de Lorraine, Centre
de Recherche en Automatique de Nancy (CRAN), UMR 7039, Campus Science, BP 70239, Vandœuvre-lès-Nancy, F-54500, France;
6CNRS, CRAN, UMR CNRS 7039, Centre Alexis Vautrin, Vandœuvre-lès-Nancy, F-54500, France; 7Université de Haute Alsace, Dé-
8
partement de Photochimie Générale, UMR CNRS 7525, Mulhouse, F-68100, France; GDR CNRS 3049 "Médicaments Photoacti-
vables-Photochimiothérapie (PHOTOMED)", 9 Centre Alexis Vautrin, CRLCC, avenue de Bourgogne, Vandœuvre-lès-Nancy, F-
54511, France
Abstract: Angiogenesis is a key step in the tumoral progression process. It is characterized by an over-expression of a number of matrix
metalloproteinases (MMP). Among these MMPs, gelatinases (MMP-2 and MMP-9) are known to play a critical role in tumor angiogene-
sis and the growth of many cancers. Photodynamic Molecular Beacons (PMB) can be designed for cancer treatment by associating a
chlorin-like photosensitizer and a black hole quencher linked by a gelatinase substrate peptide with the aim of silencing photosensitizer
toxicity in non-targeted cells and restore its toxicity only in surrounding gelatinases. This article provides a report on the synthesis and
photophysical and biochemical studies of new families of PMB, using tetraphenylchlorin and a black hole quencher as a donor-acceptor
pair, and MMP specific sequence (H-Gly-Pro-Leu-Gly-Ile-Ala-Gly-Gln-Lys-OH or H-Pro-Leu-Gly-Leu-OH) to keep them in close prox-
imity. Different spacers were used to evaluate the influence of the distance between the photosensitizer and the quencher on the photo-
physical properties and enzymatic activation of the PMB. Time-resolved quenching experiments were performed and FRET energy trans-
fer could be observed. Photosensitizers’ triplet state band in transient absorption disappears in PMB. However, even if both MMP-2 and
MMP-9 were found to efficiently cleave the peptide alone, no cleavage was observed for all PMB. Further studies would be required to
assess the ability of the PMB constructs to retain the sensitivity of the peptide linker to be cleaved by matrix metalloproteinases.
Keywords: Energy transfer, molecular beacon, matrix metalloproteinases activities, photodynamic therapy, photosensitizer, targeting,
quenching.
INTRODUCTION
supply the tumor with life-sustaining oxygen and nutriments [12],
vascular targeted photodynamic therapy (VTP) can lead to the death
of vascularized tumors [13-20]. Regulation of ROS production is
thus the key step in PDT selectivity and recent works have tended
to increase targeted versus non-targeted cells damage [21].
Matrix metalloproteinases (MMP) are known to be over-
expressed in many different kinds of cancer because of their major
implication in neo-vascularization of tumors [11, 22, 23]. Among
MMPs, the MMP-2 and MMP-9 gelatinases are highly expressed
for their role in extracellular matrix degradation [24]. Gelatinases
are able to recognize and cleave specific peptidic sequences and
have already been used for prodrugs activation [7, 25, 26].
Photodynamic Therapy (PDT) is an emerging non-invasive
technique for the treatment of a variety of cancer tumors, involving
the light activation of a photoactivatable molecule called a photo-
sensitizer in the presence of oxygen [1, 2]. After light irradiation of
the photosensitizer, the energy from the excited photosensitizer can
be transferred to surrounding molecules, allowing the generation of
reactive oxygen species (ROS) including singlet oxygen (1O2),
which is believed to be the major cytotoxic agent [3, 4]. Thus, de-
livering the photosensitizer into the cancer-associated cells means
the tumor can be specifically destroyed.
Increase in protease expression and activity has been reported
in many human cancers [5-7]. Tumor-associated proteases are
known to function at multiple steps of tumor progression, affecting
tumor establishment, growth, neovascularization, intravasation,
extravasation, and metastasis. Moreover, proteolytic degradation of
the extracellular matrix is crucial for cancer development, invasion
and metastasis, all of them associated with increased activities of
several different protease families [5, 8]. Reports have shown that
several proteases (e.g. cathepsins and MMP-2 and -9) are up-
regulated in many cancer types [9-11] and potentially, these tumor-
associated proteases could act as activators of the protease-
mediated PDT agents.
In our concept study phase, we designed a Photodynamic Mo-
lecular Beacon (PMB), made to silence the photoactive properties
of photosensitizers until the tissue target. If no damage occurs be-
fore reaching its target, selective delivery is increased and non-
triggered cells killing are limited. Molecular beacons are target-
activatable probes based on energy transfer allowing fluorescence
1
emission and O2 production in response to a specific cancer target,
such as tumor-associated endopeptidases. The photosensitizer is
associated with a quencher leading to an energy transfer and thus
1O2 quenching and kept in close proximity by using a peptidic link.
The peptide, specifically cleaved by a certain type of protease, re-
leases the two moieties which enables photoactivity to be restored.
The development of “on/off” switchable probes for cancer imaging
using fluorescence resonance energy transfer (FRET) activatable
fluorescent pairs has been well described in literature on the subject
[27], but surprisingly only few molecular beacons for treatment
purposes have been designed. Zheng et al. reported on the first
family of PMB using Pyropheophorbide a and carotenoid or Black
Hole Quencher 3 (BHQ3) as the donor/acceptor pair, targeting dif-
Certain strategies have emerged aimed at enhancing PDT selec-
tivity and efficiency and have mainly been based on active or pas-
sive cancer cells addressing. Among the existing strategies, target-
ing and destroying the neo-vasculature of tumors by PDT is thought
to be promising. Indeed, by the destruction of the neo-vessels that
*Address correspondence to this author at the Laboratoire Réactions et Génie des
Procédés, 1 rue Grandville BP 20451, F-54001, Nancy, France; Tel: +33 3 83 17 51 15;
Fax: +33 3 37 81 20; E-mail: celine.frochot@ensic.inpl-nancy.fr
1875-533X/12 $58.00+.00
© 2012 Bentham Science Publishers

Photodynamic Molecular Beacons Triggered by MMP-2 and MMP-9
Current Medicinal Chemistry, 2012 Vol. 19, No. 32
5581
ferent substrates [28-34]. Only one type of photosensitizer (pyro-
pheophorbide) was used for elaborating PMB for photodynamic
therapy applications with two types of quencher (carotenoid and
BHQ-3). Zheng’s team used different types of peptides
(GDEVDGSGK, GPLGLARK, TSGPNQEQK) or phospholipid
[35] for the linker. In this latest study, the specificity of different
phospholipases was modulated by the insertion of spacers (C6, C12)
between Pyropheophorbide and the lipid backbone. The study
showed that the C12-spacered PyroC12-PtdEtn-BHQ demonstrated a
remarkable selectivity for phospholipase A2 (PLA2) and the best
relative PLA2/PLC sensitivity, significantly outperforming previ-
ously known probes. To our knowledge this is the only paper de-
scribing the use of different spacers in self-quenching constructs
and no studies have been made concerning the photophysical prop-
erties.
Chemistry
Synthesis of Photosensitizers
Synthesis of 5-(4-carboxyphenyl)-10,15,20-triphenyl porphy-
rin (1) and tetraphenylporphyrin (2): 4-carboxybenzaldehyde (238
mg; 2.5 mmol) and freshly distillated pyrrole (0.69 mL; 10 mmol)
and benzaldehyde (0.78 mL; 7.5 mmol) were dissolved in 700 mL
dichloromethane and stirred for 15 minutes in a round bottom flask
isolated from light and under inert atmosphere. Trifluoroborate
etherate (BF3.OEt2, 185 ꢀL) was then added and the solution was
stirred for another 2 hours. Then p-chloranil (1.845 g; 7.5 mmol)
was added and the solution stirred for 2 hours at dichloromethane
reflux. The solution was filtered off on silica and the crude mixture
was purified by two column chromatographies (dichloro-
methane/ethanol [97:3] and dichloromethane/acetone [95:5]) to
give first tetraphenylporphyrin 2 (320 mg; 0.52 mmol; 20%) as a
purple solid (Rf = 0.9 (dichloromethane/ethanol [97:3])) and then 5-
(4-carboxyphenyl)-10,15,20-triphenyl porphyrin 1 (115 mg; 0.175
mmol; 7%) as a purple solid (Rf = 0.4 (dichloromethane/ethanol
Zheng and al. synthesized [29] MMP-7 triggered PMB with
pyro as the PS, black hole quencher as a dual fluorecence and O2
1
quencher and a short peptide GPLGLARK as the MMP-7 cleavage
linker. Indeed, MMP-7 belongs to matrilysins family and has been a
particularly important target because of its epithalial origin and its
high expression in pancreatic, colon, breast and nonsmall-cell lung
cancer. We chose MMP-2 and MMP-9 since in cancer research,
much interest has been devoted recently to a gelatinase subgroup of
MMPs that include MMP-2 (gelatinase A, type IV collagenase) and
MMP-9 (gelatinase B, type IV collagenase). This is partly because
of their ability to degrade type IV collagen and because of their
assumed important role in cellular invasion. In numerous studies,
expression of MMP-2 has been shown to be related to invasive
phenotype and metastatic potential of tumor cells [36, 37]. Moreo-
ver, based on impressive preclinical studies which showed an im-
portant role for MMPs in cancer, the pharmaceutical industry suc-
cessfully developed several synthetic MMP inhibitors that could be
readily administered orally to patients with cancer (lung, pancreas,
prostate, stomach, breast) [38, 39]. Even if the role of MMPs in
early stage cancer versus initiation of metastasis and growth at sec-
ondary sites remains to be more clearly defined, MMP-2 and -9
capacity to degrade the extracellular matrix, resulting in, e.g. migra-
tion of endothelial cells, is known to play a role in angiogenesis,
tumor growth and metastasis.
1
[97:3]). (1): H NMR (300 MHz, CDCl3): ꢀ(ppm) -2.61 (s, 2 H);
7.72 (s, 9 H); 8.21 (m, 6 H); 8.37 (d, 4 H, J = 8.2 Hz); 8.80 (m, 8
H). MS (ESI) m/z calculated 658.7, found 659.6 [M+H]+. (2): H
1
NMR (300 MHz, CDCl3): ꢀ (ppm) -2.79 (s, 2 H); 7.75 (d, 4 H);
7.89 (m, 8 H); 8.28 (m, 8 H); 8.84 (m, 8 H). MS (ESI) m/z calcu-
lated 614.7, found 615.8 [M+H]+.
Synthesis of 5-(4-carboxyphenyl)-10,15,20-triphenyl-21,23-H
chlorin (TPC, 3): 1 (200 mg; 0.30 mmol), K2CO3 (1.26 g; 9 mmol)
and p-toluenesulfonylhydrazide (1.70 g; 9 mmol) were dissolved in
freshly distillated pyridine (20 mL) and stirred at pyridine reflux
(110°C) in a round bottom flask isolated from light and under inert
atmosphere. New starting materials were added every hour until the
analysis of UV-visible spectra revealed the complete formation of
bacteriochlorin (appearance of an absorption peak at 730 nm).
Then, 50 mL ethyl acetate and 50 mL water were added and the
mixture was stirred for one hour at 80°C. After cooling at room
temperature, the two layers were separated, the organic layer was
washed with HCl (0.1 M) until the aqueous layer turned acidic and
then washed with 2 x 25 mL water. The organic layer was dried
over MgSO4, filtered and concentrated. The red solid was dissolved
in 30 mL dry toluene, and stirred in the dark in an inert atmosphere;
o-chloranil was slowly added and UV-visible spectra were recorded
every 15 minutes to monitor chlorin formation. Then 30 mL water
was added, the layers were separated and the organic layer was
washed with 3 x 25 mL NaOH (0.1 M), 3 x 25 mL Na2CO3 (0.1 M)
and 3 x 25 mL water, then filtered and concentrated. The crude
product was purified by HPLC (preparative reverse phase HPLC
Varian ProStar PrepStar 218, preparative column C18 (5 ꢁm, 250 x
21.2 mm, Varian Pursuit). Detection device: fluorescence detector
ProStar 363 (ꢁexc = 415 nm, ꢁem = 650 nm) and diode bar ProStar
335 using a acetonitrile/water [75:25] to acetonitrile 100% gradient
in 15 minutes, followed by 15 minutes of isocratic acetonitrile. Rt =
10 minutes. Yield in 3 (79 mg, 0.12 mmol, 40%). 1HNMR (300
MHz, CDCl3): ꢀ(ppm) -1.56 (s, 1 H); -1.46 (s, 1 H); 4.10 (s, 4 H);
7.62 (s, 9 H); 8.02 (s, 6 H); 8.34 (m, 4 H); 8.84 (m, 6 H). MS (ESI)
m/z calculated 660.7, found 661.7 [M+H]+.
This study provides a report on the synthesis, photophysical
properties and biological evaluation of a new family of PMB, trig-
gering neo-vessels proteases. Our constructs aimed at using MMP-2
and MMP-9 peptidase activity to release the photosensitizer from
its quencher using the specific peptide sequence H-Gly-Pro-Leu-
Gly-Ile-Ala-Gly-Gln-Lys-OH [40-44]. The photophysical proper-
ties of the photosensitizer / quencher pair (i.e. tetraphenylchlorin
and a Black Hole Quencher: BBQ650) were investigated using
different lengths of peptides and spacers to determine the type and
efficiency of the energy transfer which occurred.
MATERIALS AND METHODS
Chemicals
Peptide synthesis chemicals were purchased from Merck
Novabiochem. Other chemicals were purchased from Sigma-
Aldrich, NovaBioChem, Accros, Merck, Invitrogen. BBQ650-NHS
ester was purchased from Berry & Associates (Dexter, USA).
HPLC solvent was purchased from Sigma-Aldrich and spectro-
scopic solvent from Carlo Erba. MMP-2 pro-enzyme and the active
MMP-9 were purchased from Calbiochem® (Darmstadt, Germany).
The Mca-Pro-Leu-Gly-Leu-Dpa-Ala-Arg-NH2 [(7-Methoxy couma
rin-4-yl)acetyl-PLGL-N-3-(2,4-Dinitrophenyl)-L-2,3-diaminopro
pionyl-AR-NH2] fluorogenic peptide substrate was purchased from
R&D Systems (Abingdon, United Kingdom).
Synthesis of 5-(4-carboxyphenyl succinimide ester)-10,15,20-
triphenyl-21,23-H chlorin (4): 3 (70 mg; 0.11 mmol), dicyclo-
hexylcarboxydiimide (24.7 mg; 0.12 mmol) and N-
hydroxysuccinimidine (13.8 mg; 0.12 mmol) were dissolved in 10
mL dichloromethane and stirred at room temperature for 4 hours in
a round bottom flask isolated from light and under inert atmos-
phere. The crude mixture was concentrated and purified by column
chromatography (dichloromethane/ethanol [97:3]) to give a red
1
solid (Rf = 0.9; 65 mg; 0.086 mmol; 78%). H NMR (300 MHz,
CDCl3): ꢀ(ppm) -1.56 (s, 1 H); -1.46 (s, 1 H); 2.93 (s, 4 H); 4.10 (s,
4 H); 7.63 (s, 9 H); 8.01 (s, 6 H); 8.34 (m, 4 H); 8.42 (m, 6 H). MS
(ESI) m/z calculated 757.8, found 758.9 [M+H]+.

5582 Current Medicinal Chemistry, 2012 Vol. 19, No. 32
Verhille et al.
Synthesis of 5-(4-aminophenyl)-10,15,20-triphenyl porphyrin
(5): 2 (200 mg; 0.30 mmol) was dissolved in 10 mL TFA in a round
bottom flask isolated from light and under inert atmosphere. NaNO2
(44 mg; 0.6 mmol) was added carefully, the mixture was allowed to
stir for 3 minutes and then poured into 200 mL of cold water. The
stage, light exposure was minimized by sealing the reaction vessel
in aluminium foil to limit the occurrence of unwanted side reac-
tions. The SPPS reactor was shaken for 15 hours and the resin was
washed with NMP, DMF and dichloromethane. The red resin was
dried under vacuum and then cleaved using TFA (2 mL) for 2 hours
and 30 minutes. The acidic green resin was filtered, washed with
another 1 mL TFA and N-methylpyrrolidone (NMP) and dichloro-
methane. The filtrate was dried under vacuum and lyophilisated.
The crude product was further purified by preparative HPLC using
acetonitrile/water (0.1% TFA) [75:25] to 100% acetonitrile gradient
in 15 minutes, followed by isocratic acetonitrile for 15 minutes. Rt
= 16 minutes. Pure product was isolated as a red powder (51 mg;
mono-nitrated product was extracted with 6
25 mL DCM, the
solvent was removed and the compound was purified on column
chromatography (dichloromethane/hexanes [80:20]) to give a pur-
ple solid (Rf = 0.61; 103 mg; 0.16 mmol; 62%). 1H NMR (300
MHz, CDCl3): ꢀ(ppm) -2.76 (s, 2 H); 7.79 (s, 9 H); 8.21 (s, 6 H);
8.70 (m, 4 H); 8.86 (m, 6 H). MS (ESI) m/z calculated 659.7, found
660.6 [M+H]+.
1
0.034 mmol; 40%). H NMR (300 MHz, DMSO d6): ꢀ(ppm) -1.58
The product was then dissolved in 10 mL HCl 37% and vigor-
ously stirred. Slowly, SnCl2 (240 mg; 1.2 mmol) was added and the
reaction was stirred for another 90 minutes. The crude mixture was
poured into 100 mL of cold water and Na2CO3 was slowly added
until no more CO2 was produced. Extraction with dichloromethane
until clear aqueous phase was done and solvent removal gave a
crude product, which was allowed to be purified by column chro-
matography (dichloromethane) to give a purple solid (Rf = 0.8; 62
and -1.56 (s, 2 H); 0.77-0.90 (m, 9 H); 1.17-2.27 (m, 32 H); 2.72 (s,
2 H); 3.71 (m, 3 H); 4.07-4.70 (m, 6 H); 4.14 (s, 4 H); 6.71 and 7.28
(s, 2 H); 7.70-8.56 (m, 9 H); 7.73 (s, 9 H); 8.07 (s, 6 H); 8.31 (m, 4
H); 8.58 (m, 6 H). MS (ESI) m/z calculated 1482.7, found 1483.5
[M+H]+.
Synthesis of TPC-DGA-GPLGIAGQK-OH (9a): the peptide H-
GPLGIAGQ(Trt)K(Boc) coupled on the Wang resin was synthe-
sized as described previously on a 0.033 mmol scale. After resin
swelling with dichloromethane and removal of the solvent, a solu-
tion of 7 (21 mg; 0.028 mmol), HATU (35 mg; 0.092 mmol) and
DIEA (0.1 mL) in DMF was added and stirred overnight. The same
procedure as for compound 8a was used. Rt = 6 minutes. Pure
product was isolated as a red powder (28 mg; 0.018 mmol; 65%).
1H NMR (300 MHz, DMSO d6): ꢀ(ppm) -1.58 (s, 2 H); 0.80-0.87
(m, 9 H); 1.22-1.89 (m, 26 H); 2.10 (m, 6 H); 2.72 (s, 2 H); 3.72
(m, 3 H); 3.80-4.37 (m, 6 H); 4.12 (s, 4 H); 4.22 (s, 2 H); 6.76 and
7.2 (s, 2 H); 7.72-8.55 (m, 9 H); 7.76 (s, 9 H); 8.03 (s, 6 H); 8.29
(m, 4 H); 8.56 (m, 6 H). MS (ESI) m/z calculated 1569.8, found
1
mg; 0.1 mmol; 62%). H NMR (300 MHz, CDCl3): ꢀ(ppm) -2.74
(s, 2 H); 4.07 (s, 2 H); 7.69 (s, 9 H); 8.13 (s, 6 H); 8.76 (m, 6 H);
8.86 (d, 4 H, J = 4.8 Hz). MS (ESI) m/z calculated 629.7, found
630.8 [M+H]+.
Synthesis of 5-[[(4-phenylcarbamoyl)methoxy]-acetic acid]-
10,15,20-triphenyl porphyrin (6): Aminoporphyrin 5 (143 mg, 0.23
mmol) was dissolved in 2 mL DMF in a round bottom flask isolated
from light and under inert atmosphere. Diglycolic anhydride (100
mg, 0.86 mmol) was then added and the solution was stirred at
room temperature overnight. The reaction mixture was diluted in 10
mL CHCl3 and hexane was added until precipitation occurred. The
precipitate was filtered off and washed with water to remove resid-
ual anhydride and then dried under vacuum and purified on column
chromatography (dichloromethane/ethanol [97:3]) to give a purple
785.3 [M+2H] 2+
.
Synthesis of TPC-PEG9-GPLGIAGQK-OH (10a): the peptide
H-GPLGIAGQ(Trt)K(Boc) was synthesized as described previ-
ously on a 0.033 mmol scale. Then, after resin swelling with di-
chloromethane and removal of the solvent, a solution of Fmoc-
O2Oc-OH (Fmoc-PEG9-COOH) (26.1 mg; 0.068 mmol), HATU
(27.4 mg; 0.072 mmol) and DIEA (0.1 mL) in DMF was added and
stirred overnight. The resin was washed with NMP, DMF and di-
chloromethane. Fmoc deprotection was undergone using
piperidine/DMF [60:40]. The resin was washed and dried and 7 (28
mg; 0.042 mmol), HATU (27 mg; 0.071 mmol) and DIEA (0.1 mL)
in DMF were added and stirred overnight. The same procedure as
for compound 8a was used. Rt = 4 minutes. Pure product was iso-
1
solid (Rf = 0.8; 102 mg; 0.16 mmol; 70%). H NMR (300 MHz,
CDCl3): ꢀ(ppm) -2.91 (s, 2 H); 4.22 (s, 2 H); 4.34 (s, 2 H); 7.82 (s,
9 H); 8.22 (m, 6 H); 8.70 (m, 4 H); 8.82 (m, 6 H); 11.10 (s, 1H).
MS (ESI) m/z calculated 745.8, found 746.8 [M+H]+.
Synthesis of 5-[[(4-phenylcarbamoyl)methoxy]-acetic acid]-
10,15,20-triphenyl-21,23-H chlorin (TPC-DGA, 7): Same proce-
dure as 3 was followed, using 6 (122 mg; 0.63 mmol), p-
toluenesulfonylhydrazine (950 mg; 5.1 mmol) and K2CO3 (1.5 g;
10.8 mmol). The product was purified by HPLC using metha-
nol/water [75:25] to methanol 100% gradient in 15 min, followed
by 15 minutes of isocratic methanol flow. (Rt = 12 min) giving pure
7 (51 mg, 0.12 mmol, 41%). 1H NMR (300 MHz, CDCl3): ꢀ(ppm) -
1.53 and -1.54 (s, 2 H); 3.87 (s, 4H); 4.14 (s, 2 H); 4.16 (s, 2 H);
7.72 (s, 9 H); 8.12 (m, 6 H); 8.29 (m, 4 H); 8.55 (m, 6 H); 10.21 (s,
1H). MS (ESI) m/z calculated 747.8, found 748.9 [M+H]+.
1
lated as a red powder (14 mg; 0,009 mmol; 26%). H NMR (300
MHz, DMSO d6): ꢀ(ppm) -1.59 (s, 2 H); 0.79-0.93 (m, 9 H); 1.17-
1.97 (m, 26 H); 2.12 (m, 6 H); 2.73 (s, 2 H); 3.05 (d, 8 H); 3.46 (s,
2 H); 3.68 (m, 3 H); 3.73-4.13 (m, 6 H); 4.13 (s, 4 H); 6.76 and 6.79
(s, 2 H); 7.72-8.59 (m, 36 H). MS (ESI) m/z calculated 1627.9,
found 814.7 [M+2H] 2+
.
Conjugate Photosensitizer-Peptide Synthesis
Synthesis of TPC-PLGLK-OH (11a): the peptide Fmoc-
PLGLK(Boc)-OH was synthesized by automatic SPPS using the
same procedure; compound 4 (60 mg; 0,079 mmol) was then added
using DIEA (0.1 mL) in dichloromethane. The SPPS reactor was
shaken overnight and the resin was washed with NMP, DMF and
dichloromethane. The same procedure as for compound 8a was
used. Rt = 12 minutes. Pure product was isolated as a red powder
(16 mg; 0,014 mmol; 42%). 1H NMR (300 MHz, DMSO d6):
ꢀ(ppm) -1.52 and -1.57 (s, 2 H); 0.73-0.87 (m, 12 H); 1.41-2.85 (m,
12 H); 2.30 (m, 2 H); 2.70 (m, 2 H); 2.90 (m, 2 H); 3.61 (m, 2 H);
3.8-4.70 (m, 6 H); 4.14 (s, 4 H); 7.70-8.61 (m, 29 H). MS (ESI) m/z
calculated 1169.4, found 1170.2 [M+H]+.
Synthesis of TPC-GPLGIAGQK-OH (8a): The side chains of
glutamine and lysine were respectively protected by trityl (Trt) and
N-tert-butyloxy carbonyl (Boc) groups. H-GPLGIAGQK- and the
peptide conjugated photosensitizer TPC–GPLGIAGQK- were syn-
thesized on a multichannel peptide synthesizer, according to a clas-
sical Fmoc/tBu solid phase methodology, using Fmoc-Lys(Boc)-
Wang resin. Double coupling was performed using a 3-fold excess
of N-Fmoc-amino acid (except during the first glycine coupling
stage: 2-fold excess and six times coupling) and activation reagents
2-(1H-benzotriazol-1-yl)-1,1,3,3-tetramethyl-uronium hexafluoro-
phosphate (HBTU) (3 eq.), 1-hydroxybenzotriazole (HOBt) (3 eq.)
and N,N-diisopropylethylamine (DIEA) (9 eq.) in dimethylforma-
mide (DMF).
Molecular Beacons Synthesis
Synthesis of compounds 8b, 9b, 10b and 11b: in a round bot-
tom flask isolated from light and under inert atmosphere, the ade-
quate conjugate was dissolved in 2 mL DMF and 3 mL carbonate
On a 0.033 mmol scale, final Fmoc protection was removed and
TPC-NHS (65 mg; 0.086 mmol) was then added with triethylamine
(1 mL) in dichloromethane. During the photosensitizer coupling

Photodynamic Molecular Beacons Triggered by MMP-2 and MMP-9
Current Medicinal Chemistry, 2012 Vol. 19, No. 32
5583
buffer pH = 9.3 and the solution was stirred for 15 minutes. Com-
mercially available BBQ650-NHS ester (25 mg; 0.035 mmol) was
added, completed with another 2 mL DMF and the solution was
stirred from several days at 36°C. Reaction was monitored by ana-
lytical HPLC (conditions as follow). Quenching of the buffer was
carried out using 2 mL HCL (0.5 M) and the solution was concen-
trated. Purification was done by HPLC:
Time-resolved experiments were performed using for excita-
tion: a pulsed laser diode LDH-P-C-405 coupled with a driver PDL
800-D (both PicoQuant GmbH, Berlin, Germany) emitting at 408
nm (FWHM < 70 ps, 1 MHz) and an avalanche photodiode SPCM-
AQR-15 (EG & G, Vaudreuil CANADA) coupled with a 650 nm (
10 nm) interference long-wave pass filter as detection system. The
acquisition system was a PicoHarp 300 module with a 4 channels
router PHR-800 (both PicoQuant GmbH, Berlin, Germany). Vari-
ous fluorescence decays were recorded using the single photon
counting method (4 optical faces quartz cells for pure products in
ethanol and triangular quartz cells for FRET experiments in DMF).
Data was collected up to 1000 counts accumulated in the maximum
channel and analyzed using Time Correlated Single Photon Count-
ing (TCSPC) software Fluofit (PicoQuant GmbH, Berlin, Germany)
based on iterative reconvolution using a Levensberg-Marquandt
algoritm, enabling the obtention of multi-exponential profiles
(mainly one or two exponentials in our cases).
8b: From 8a (50 mg; 0.033 mmol). The same purification pro-
cedure as for compound 8a was used. The pure product was iso-
lated as a green/blue powder (Rt = 16 min; 34 mg; 0.017 mmol;
48%). MS (ESI) m/z calculated 2081.37, found 1041.42 [M+2H]2+.
9b: From 9a (27 mg; 0.016 mmol). The same purification pro-
cedure as for compound 8a was used. The pure product was iso-
lated as a green/blue powder (Rt = 11 min; 17 mg; 0.008 mmol;
51%). MS (ESI) m/z calculated 2168.45, found 1085.37 [M+2H]2+.
10b: From 10a (13 mg; 0.009 mmol). The same purification
procedure as for compound 8a was used. The pure product was
isolated as a green/blue powder (Rt = 8 min; 12.5 mg; 0.006 mmol;
62%). MS (ESI) m/z calculated 2226.53, found 1114.42 [M+2H]2+.
The theoretical calculation of Förster radius can be carried out
using the following equation:
(3)
11b: From 11a (15 mg; 0.014 mmol). The same purification
procedure as for compound 8a was used. The pure product was
isolated as a green/blue powder (Rt = 10 min; 16 mg; 0,009 mmol;
65%). MS (ESI) m/z calculated 1768.06, found 1768.75 [M+H]+
and 885.00 [M+2H]2+.
where R0 is the Förster radius expressed in nm, ꢄ2 is the orientation
factor, taken as 2/3 [47], n is the refraction index of ethanol, taken
as 1.3594. ꢂD is the fluorescence quantum yield of the donor
(TPC). The overlap integral J is described as follows:
(4)
Photophysical Properties
where FD(ꢅ) is the emission spectra of TPC, ꢆA(ꢅ) is the extinction
coefficient of the acceptor (BBQ) in M-1.cm-1 and ꢅ is the wave-
length in nm.
Absorption spectra were recorded on a Perkin-Elmer (Lambda
EZ 210) double beam UV-visible spectrophotometer. Fluorescence
spectra were recorded on a Horiba Jobin Yvon Fluorolog-3 (FL3-
222) spectrofluorimeter equipped with a 450 W Xenon lamp, a
thermostated cell compartment (25°C), a R928 (HAMAMATSU
Japan) UV-visible photomultiplier and a liquid nitrogen-cooled
InGaAs infrared detector (DSS-16A020L Electro-Optical System
Inc, Phoenixville, PA, USA). The excitation spectrometer is a
SPEX double grating monochromator (1200 grating/mm blazed 330
nm). The fluorescence was measured by the UV-Visible detector
through a SPEX double grating monochromator (1200 grating/mm
blazed 500 nm). The singlet oxygen production was measured by
the IR detector through a SPEX double grating monochromator
(600 grating/mm blazed 1 ꢀm). All spectra were recorded using 4
optical faces quartz cells. Fluorescence quantum yields (ꢂf) were
determined using a tetraphenyl porphyrin (TPP) solution in toluene
as a fluorescence standard (ꢂf = 0.11 [45]). Rose Bengal was cho-
The nanosecond transient absorption setup was based on a
pulsed solid state Nd:Yag laser (Surelite II-10, Continuum, Santa
Clara, USA) operating at 10 Hz. After amplification, the output
frequency was doubled and then tripled. This laser can provide
nanosecond pulses at 355 nm with an energy of about 160 mJ
which can be used to pump an optical parametric oscillator (SL
OPO, Continuum, Santa Clara, USA). Nanosecond pulses were
then delivered with a tunable wavelength from 410 to 2200 nm and
energy input reduced to about 6-7 mJ to excite the sample. The
transient absorption analysis system (LKS 60, AppliedPhotophys-
ics, UK) used a 150 W pulsed xenon arc lamp, a Czerny-Turner
monochromator and a 5-stage based photomultiplier (1P28). The
instrument response was about 20 ns using a transient digitizer
(Infiniuum, Agilent). The observation wavelength is indicated in
each case. Experiments were performed in acetonitrile under air, Ar
or O2 bubbling. The absorbance of the dyes was adjusted to values
between 0.3-0.5 in order to prevent self quenching effects.
1
sen as a reference solution thanks to its high O2 production quan-
tum yield in ethanol (ꢂꢃ = 0.68 [46]). The absorbance value at the
excitation wavelength (415 nm) of both the reference and the sam-
ple solutions were set to around 0.2.
The fluorescence and triplet state quenching were calculated us-
ing Stern-Volmer’s equation:
The quantum yield of fluorescence is determined thanks to
equation (1):
(5)
where ꢀ0 and ꢀ stand for the lifetimes of the excited state in the
absence and in the presence of quencher, respectively, K is the
Stern-Volmer quenching constant, kq is the bimolecular quenching
constant and [Q] the quencher concentration. In the case of steady-
state fluorescence experiments, the Stern-Volmer quenching con-
stant can be determined from F0 and F which stand for the fluores-
cence intensities in the absence and in the presence of quencher,
respectively.
(1)
where ꢂf and ꢂfo, If and Ifo, DO and DOo, n and no are the quantum
yields, the intensities, the absorbances and the refraction indices of
the sample and of the reference respectively.
The quantum yield of singlet oxygen production is determined
thanks to equation (2):
Incubation Studies with Recombinant Active MMP-2 / MMP-9
Before enzymatic cleavage assay, the enzyme activity of both
MMP-2 and MMP-9 was confirmed by their ability to cleave the
fluorogenic peptide substrate (7-Methoxycoumarin-4-yl)acetyl-
PLGL-N-3-(2,4-Dinitrophenyl)-L-2,3-diaminopropionyl-AR-NH2
(2)
where ꢂꢃ and ꢂꢃo, I and Io, DO and DOo are the luminescence
quantum yields of singlet oxygen, the luminescence intensities, the
absorbances of the sample and of the reference respectively.

5584 Current Medicinal Chemistry, 2012 Vol. 19, No. 32
Verhille et al.
[Mca-PLGL-Dpa-AR-NH2] (R&D Systems) [39] used as positive
control substrate. The PMB compounds 8b, 9b, 10b, 11b and the
GPLGIAGQK peptide were dissolved in dimethylsulfoxyde and
then diluted in the assay buffer (50 mM Tris, 10 mM CaCl2, 150
mM NaCl, 0.05% Brij 35, pH 7.5). The ability of the H-
GPLGIAGQK-OH peptide linker to be cleaved was analyzed first.
The peptide samples at 10 or 50 ꢀM were incubated without or with
1.5 or 1.2 nM (v:v) of active MMP-2 and MMP-9, respectively, for
1 and 2 hours at 37°C. The incubation products were analyzed by
Liquid Chromatography coupled to High Resolution Mass Spec-
trometry (LC-HRMS) using a Bruker micrOTO-Q system (Bruker
Daltonik GmbH, Bremen, Germany). The same incubation condi-
tions were then applied to compounds 8b, 9b, 10b and 11b. After
each incubation period, the fluorescence of the incubation products
samples was assessed by fluorescence spectroscopy with an excita-
tion wavelength of 415 nm and emission between 450 and 800 nm.
conjugates were then synthesized by SPPS; successful synthesis of
Fmoc-Gly-Pro-Leu-Gly-Ile-Ala-Gly-Gln(Trt)-Lys(Boc)-Wang
Resin (for later compounds 8a, 9a and 10a) and Fmoc-Pro-Leu-
Gly-Leu-Lys(Boc)-Wang Resin (for later compound 11a) were
achieved by classical peptide synthesis on solid phase using Fmoc
strategy and a six-times coupling of first glycine residue.
Compounds 8a and 11a were obtained by coupling NHS-
activated photosensitizer (4) on deprotected N-terminus peptide in
its solid phase, followed by a cleavage from the resin using TFA,
also allowing complete side-chains deprotection Fig. (2).
Compound 9a was obtained by synthesizing first porphyrin 6
using Vicente et al. procedure [52], and then Whitlock reduction to
give 7. Coupling of 7 on peptide was performed with good yields
using the HATU coupling reagent. 10a was obtained after peptide
synthesis, addition during the solid phase of {2-[Fmoc-ethoxy]-
ethoxy}-acetic acid, noted Fmoc-NH-PEG9-COOH (where 9 refers
to the number of atoms of the chain), and then 3, using HATU as
coupling reagent for both.
RESULTS
Synthetic Strategy
For the synthesis of Photodynamic Molecular Beacons Fig. (2),
TPC-peptide conjugates were dissolved in a DMF/carbonate buffer
(pH 9.3) mixture allowing a coupling in a solution of commercially
available BBQ650-NHS ester on peptide lysine ꢀNH2, as described
in literature in the field of study [53]. The compounds were purified
by reverse phase preparative HPLC using both UV-visible and fluo-
rescence detections.
The aim of this study was to synthesize a family of Photody-
namic Molecular Beacons Fig. (1) using MMP-2 and MMP-9 sub-
strate peptide –GPLGIAGQK-, described by Mansour et al. [31],
using different lengths of spacer between the photosensitizer and
the first amino acid. As our “control” molecular beacon, we chose
the fluorogenic peptide substrate containing PLGL sequence, de-
scribed by Knight et al. [26, 48]. Synthesis was basically performed
in three steps as shown in Fig. (2) namely: 1) Photosensitizer syn-
thesis; 2) Coupling to the peptide on solid phase and cleavage; 3)
Coupling to the quencher BBQ NHS-ester in solution.
Photophysical Properties
Photosensitizer Derivates
As shown in Fig. (4), no significant change was observed be-
tween TPC and its derivates: UV-visible spectra remained un-
changed with a maximum corresponding to the Soret band at 415
nm (ꢀ = 85 000 M-1.cm-1) and an intense QI band at 650 nm (ꢀ = 21
000 M-1.cm-1) in ethanol Fig. (4a). Fluorescence emission and sin-
glet oxygen production were similar. For all compounds, fluores-
cence lifetimes were 9.1 ns, and estimated fluorescence and singlet
oxygen quantum yields were 0.3 and 0.7 in EtOH respectively Figs.
(4a, 4b and 4c). Transient absorption spectra were detected at 440 –
450 nm which is typical for the triplet state of such chlorins as
shown in Fig. (4d) [31].
A variety of photosensitizers deriving from tetraphenylchlorin
monocarboxylic acid (TPC) were synthesized by a commonly used
method previously described by our team [49], using successively
Lindsey method [50] to produce porphyrin 1 and Witlock reduction
[51] to obtain the chlorin 3 Fig. (3). First, a complete reduction of
porphyrin and chlorin into bacteriochlorin was carried out, using 30
equivalents of p-toluenesulfonyhydrazide under pyridine reflux.
Bacteriochlorin was then slowly converted into chlorin by carefully
adding an oxidizing agent (o-chloranil). The reaction could be eas-
ily monitored by checking UV-visible spectra after each addition.
Those reactions gave variable yields ranging from 30% to 90%. The
PLGLK
GPLGIAGQK
R1
R1
R2
R2
8a
8b
9a
9b
10a
10b
11a
11b
-H
-H
Fig. (1). Conjugates and molecular beacons synthesized.

Photodynamic Molecular Beacons Triggered by MMP-2 and MMP-9
Current Medicinal Chemistry, 2012 Vol. 19, No. 32
5585
1. Deprotection
2. Elongation
Peptide
Activation
X = -COOH : 3
-NHC(O)CH2OCH2COOH: 7
Peptide
TFA
Peptide
BBQ-NHS
Peptide
Fig. (2). General procedure for PMB synthesis.
(a) BF3.OEt2, DCM, o-chloranil, 4h, (1) 7%, (2) 20%; (b) p-TSH, pyridine, K2CO3, 60%; (c) p-chloranil, AcOEt, 40%; (d) DCC, NHS, DCM, 78%; (e) NaNO2,
TFA, 3 min, 70%; (f) SnCl2, HCl, 1 h, 60%; (g) diglycolic anhydrid, DMF, 15h, 70%.
Fig. (3). Synthesis of photosensitizers.

5586 Current Medicinal Chemistry, 2012 Vol. 19, No. 32
Verhille et al.
A
B
Absorption
Fluorescence
D
C
Fig. (4). Photophysical properties of conjugates. (A) UV-visible absorption (full line) and fluorescence emission spectra (dash line, ꢂ = 415 nm) of 3 (pur-
ple), 8a (dark blue), 9a (red), 10a (white blue) and 11a (green) in ethanol. (B) Fluorescence decay in ethanol, ꢂexc = 408 nm. (C) Singleextc oxygen luminescence
spectra in ethanol, ꢂexc = 415 nm. (D) Transient absorption in ethanol, ꢂexc = 415 nm.
The quenching of the photosensitizers’ triplet state by oxygen
was investigated using time-resolved transient absorption. Fig. (5)
shows the general shape of all photosensitizers’ transient absorption
decays. As expected, triplet state lifetime depends on oxygen con-
centration in the media; the observed triplet lifetimes of TPC and
TPC-peptide were around 55 ns in an oxygen saturated solution
([O2] = 10.1 mmol/L [45]), 240 ns in a solution under atmospheric
pressure ([O2] = 2.1 mmol/L [54]) and 110 ꢀs in an oxygen free
solution. The Stern-Volmer plots Fig. (5c) were linear and calcu-
PMB
The presence of the quencher can be directly detected on UV-
visible spectra by adding BBQ spectrum onto the TPC spectrum.
The light shift of the Soret band was observed leading to a maxi-
mum at 418 nm, as shown in Fig. (7a), representing the general
shape of all PMB. Efficient quenching of fluorescence intensity and
thus quantum yield was observed with values decreasing by 90%
for all PMB (8b, 9b, 10b and 11b) as compared to the correspond-
ing photosensitizer derivates (8a, 9a, 10a and 11a, respectively)
Figs. (7b, 7c, 7d). However, fluorescence decay was found to show
a double decay corresponding to both quenched construct and re-
sidual photosensitizer expression, decay time being respectively of
2 ns and 9.1 ns Fig. (7b).
lated oxygen quenching constants kq O2 were around 1.8
109 M-
1.s-1, which indicates that the quenching observed was due to a bi-
molecular diffusion-controlled process. All values are summarized
in (Table 1).
FRET Experiments
Photosensitizers’ triplet state band in transient absorption (ꢂ =
To determine if FRET could be involved in the energy transfer
process between the photosensitizer and BBQ, the fluorescence
quenching efficiency was investigated using compound 8a and
BBQ in DMF solution. As shown in Fig. (6a), BBQ UV-visible
spectrum shows a good level of spectral overlap with TPC fluores-
cence emission spectrum.
440 – 450 nm) disappears in PMB.
Generation of singlet oxygen dramatically decreased, quantum
yields were observed from ꢁꢀ = 0 to ꢁꢀ = 0.06 (Table 2).
Enzymatic Cleavage
As shown in Figs. (8-9), both MMP-2 and MMP-9 were found
to efficiently cleave the H-GPLGIAGQK-OH peptide. Following
the 2 hour incubation of the peptide at 50 ꢀM with active MMP-2
or MMP-9, a significant reduction of the mass spectrometry peak
intensity of the native peptide at 8.3-8.9 minutes was observed and
a new distinct mass spectrometry peak was detected at 9.2-9.8 min-
utes. The new peak displayed a mass of 343 corresponding to the
cleaved tetrapeptide H-GPLG-OH Figs. (8-9). Since BBQ650 effi-
The spectral overlap J of the pair TPC/BBQ was calculated us-
ing equation (3) and estimated to be: J = 4.9 1015 M-1.cm-1.nm4.
The Förster radius was calculated using equation (2) and we found
R0 = 54 nm.
Experiments in time-resolved fluorescence spectroscopy using
8a and BBQ in solution showed linear Stern-Volmer plots Fig. (6b)
and a quenching constant kq BBQ = 4.3 1010 M-1.s-1, indicating that
the energy transfer is limited by a diffusion process.

Photodynamic Molecular Beacons Triggered by MMP-2 and MMP-9
Current Medicinal Chemistry, 2012 Vol. 19, No. 32
5587
A
B
C
Fig. (5). (A) Time-resolved transient absorption of 8a monitored at 450 nm in ethanol saturated in oxygen (red line) and under atmospheric pressure (blue line),
ꢀexc = 415 nm. (B) Time-resolved transient absorption of 8a monitored at 450 nm in ethanol saturated with argon ), ꢀexc = 415 nm. (C) Stern-Volmer plots of
conjugates triplet state quenched by oxygen in ethanol.
Table 1.
Summary of Photosensitizers’ Conjugates Photophysical Properties in Ethanol. a Oxygen Saturated Solution, [O2] = 10.1 mmol/L. b Solu-
tion Under Atmospheric Pressure, [O2] = 2.1 mmol/L. c Argon Saturated Solution, [O2] = 0 mmol/L.
ciently quenches TPC fluorescence in the intact PMB molecules,
we tested to find out whether active MMP-2 and/or MMP-9 were
able to specifically cleave the peptide linkers in the different PMB
constructs (8b, 9b, 10b and 11b) and to restore the fluorescence.
However, no increase of the fluorescence intensity for 8b, 9b, 10b
and 11b compounds was observed after incubation with MMP-2 or
MM-9 (data not shown), suggesting that any cleavage occurred.
Additional incubation conditions were tested (see supplementary
material) but no restoration of the fluorescence intensity was ob-
served for any construct.
DISCUSSION
The control of singlet oxygen production is a key step for PDT
applications, and protease-mediated activatable probes are of great
interest. Several studies involving photosensitizer aggregation as
quenching process were realized by Weissleder team [55-57] and
Lange team [58-60]. Zheng et al. developed the first photosensi-
tizer-quencher and enzyme-active targeting probes [28-34] where a
photosensitizer was linked to a quencher via a disease targeted
biomolecule. Recently, phospholipases activatable probes compris-

5588 Current Medicinal Chemistry, 2012 Vol. 19, No. 32
Verhille et al.
Figure 6: (A) Spectral overlap of 8a emission (solid line, ꢀexc = 415 nm) and BBQ650 absorption (dash line) in ethanol. (B) Quenching of 8a by BBQ in DMF
solution, ꢀexc = 408 nm, ꢀem = 650 nm, in ethanol.
Fig. (7). (A) UV-visible spectra of TPC (solid line), Black Berry Quencher (small dash line) and PMB (big dashed line) in ethanol. (B) Fluorescence decays of
TPC (above decay) and PMB (beneath decay), ꢀexc = 415 nm, ꢀ = 650 nm, in ethanol. (C) Normalized fluorescence of conjugates and their associated PMB.
In ethanol, ꢀexc = 415 nm,ꢀem = 650 nm, in ethanol. (D) Normaliemzed singlet oxygen luminescence from conjugates and their associated PMB. In ethanol, ꢀexc
415 nm ,ꢀem = 650 nm, in ethanol.
=
ing different spacer lengths between pyropheophorbide-a and
BHQ3 were synthesized and their biological activity reported [35].
malignant gliomas [61], the use of PMB could provide another way
of controlling toxic species generation in order to prevent non tar-
geted cells destruction.
The aim of this study was to synthesize a family of PMB using
peptidase activity of gelatinases MMP-2 and MMP-9, with a sub-
strate peptide H-GPLGIAGQK-OH, first described by Mansour et
al. [40] and to evaluate the influence of the length of the linker on
the photophysical properties and enzymatic cleavage sensitivity.
The peptide H-GPLGIAGQK-OH has been used as gelatinases
substrate for drug release by many studies [41-44]. As a “control”
molecular beacon, we chose H-PLGL-OH peptide sequence, de-
scribed by Knight et al. [48], known to be cleaved by numerous
MMPs and other proteases [26, 43]. Since gelatinases are normally
over-expressed in the neo-vasculature of many cancers such as
An ideal PMB should be composed of a photosensitizer with a
high singlet oxygen production that can be efficiently quenched.
Zheng’s team used pheophorbide-a as a photosensitizer. This is a
chlorophyll analogue with a long wavelength at 667 nm which is
1
also an efficient O2 producer with a quantum yield > 50% [28].
Our team specializes in the synthesis of porphyrin and chlorin and
therefore we decided to use a monocarboxylic tetraphenyl chlorin
[62]. The choice of photosensitizer was made because of its high
singlet oxygen quantum yield value (0.69 in ethanol) and its absorp-
tion in the red part of the spectrum (650 nm) [63]. The quenching of

Photodynamic Molecular Beacons Triggered by MMP-2 and MMP-9
Table 2. Summary of Molecular Beacons’ Photophysical Properties in Ethanol
Current Medicinal Chemistry, 2012 Vol. 19, No. 32
5589
Fig. (8). LC-MS spectra of peptide GPLGIAGQK (50 ꢀM) incubated in the absence (A) or presence (B) of 1.5 nM active MMP-2 for 2 hours at 37°C.
singlet oxygen is of major interest in the construction of such com-
pounds so the Black Berry quencher, previously analyzed by Lovell
et al. [31], was chosen for its good (high) quenching properties. The
synthesis of different PMB comprising several spacers was carried
out along with those made up of a short peptide to illustrate the
relationship between the distance of the acceptor/donor pairs and
quenching efficiencies. Different spacers were tested in order to
change the length between the first amino acid of the peptide and
the photosensitizer, and to establish a structure/enzyme activity
relationship. The spacers were also chosen according to their hy-
drophilic properties.
ized by a good singlet oxygen quantum yield (ꢁꢀ = 0.69) and a
strong absorption in the red part of the spectrum, which is important
from the photodynamic therapy standpoint.
As expected, compounds 10a and 10b with the PEG9 spacer
were found to be more hydrophilic (Rt = 4 and 8 min) than with a
short spacer (9a, 9b; 6 and 11 min) and than compounds with no
spacer (11a, 11b; 12 and 10 minutes. 8a, 8b; 14 and 16 min) (Table
3). Increasingly less accessibility to the ꢂNH2 of lysine was ob-
served proportionally to water solubility. Indeed, compounds 8b
and 11b, were synthesized within 24 hours from 8a and 11a respec-
tively, whereas 9b was synthesized within a week and the reaction
1
PMBs were obtained with poor overall yields due both to por-
phyrin asymmetry and variable yields of chlorin synthesis. The
Lindsey reaction for porphyrins was preferred to other routes be-
cause of its easy one-step procedure although a great deal of purifi-
cation was required. Nevertheless, TPC was found to be character-
time of 10b exceeded a month. H NMR showed a temperature-
dependant massif in labile proton domain (4 – 7 ppm) which may
indicate that water can be caged by the molecule. Special peptide
conformation may be involved in this reactivity issue.

5590 Current Medicinal Chemistry, 2012 Vol. 19, No. 32
Verhille et al.
Fig. (9). LC-MS spectra of peptide GPLGIAGQK (50 ꢀM) incubated in the absence (A) or presence (B) of 1.2 nM active MMP-9 for 2 hours at 37°C.
Table 3.
Retention Times of Compounds 3, 8a, 9a, 10a, 11a, 8b, 9b, 10b, 11b in HPLC Using [75:25] Acetonitrile/water (0.1% TFA) to 100% Ace-
tonitrile Gradient in 15 min, and then Isocratic Acetonitrile for 15 min
From a photophysical point of view, our investigations showed
disturbed by the presence of a peptide moiety since kq O2 constants
were around 2 ꢀ 109 M-1.s-1. Singlet oxygen production was thus
found to be unchanged, ꢁꢂ = 0.7 for all compounds.
that the coupling of peptides to TPC did not modify its photophysi-
1
cal properties, or its fluorescence and O2 quantum yields. As illus-
trated in (Table 1), fluorescence quantum yield and lifetime of
products 3 and 7 were unchanged after functionalization (com-
pounds 8a, 9a, 10a and 11a). If we compare 8a and 11a, no signifi-
cant changes can be observed, meaning that the type of peptide
linked does not influence the photophysical properties of TPC.
Moreover, the same analysis can be made for 8a, 9a and 10a mean-
ing that the size of the spacer does not influence the photophysical
properties of TPC. Similar observations were made concerning the
triplet state of those chlorins. The band at 440 – 450 nm in transient
absorption remained the same and the quenching of this triplet state
by oxygen to produce singlet oxygen was still controlled by a diffu-
sion process, as can be deduced from the kq O2 constants. TPC and
its derivates thus was found to have a high Inter System Crossing
rate and a good singlet oxygen formation efficiency which was not
To determine whether the FRET energy transfer process oc-
curred in our system, quenching experiments were performed in
solution with compound 8a and increasing amounts of BBQ. Al-
though there are few studies describing energy transfer processes
involving black hole quenchers, Lovell et al. investigated the FRET
efficiency between a pyropheophorbide-a (Pyro) and several black
hole quenchers [31]. In this study, the quenchers were attached to
the photosensitizer using a lysine linker. Förster radii, fluorescence
and singlet oxygen quenching rates were calculated, showing that
quenchers absorbing in the red part of the spectrum were the most
efficient. Those Pyro-quenchers, linked together by a simple lysine,
resulted in 95% to 99% of fluorescence quenching and 91% to 99%
of singlet oxygen quenching. Among them, Black Berry Quencher

Photodynamic Molecular Beacons Triggered by MMP-2 and MMP-9
Current Medicinal Chemistry, 2012 Vol. 19, No. 32
5591
650 (BBQ650) showed particularly good properties, quenching
99% of Pyro fluorescence and 97% of Pyro production of singlet
oxygen.
We investigated whether active MMP-2 and/or MMP-9 were
able first to cleave peptide linkers in the different PMB constructs
and secondly, to restore the fluorescence of the conjugated photo-
sensitizer. Although, as expected, both active enzymes efficiently
cleaved the free GPLGIAGQK peptide, no cleavage was measured
for compounds 8b, 9b, 10b and 11b Fig. (10). Different incubation
times and substrate concentrations were tested using MMP-2 and
MMP-9 active enzymes, but unfortunately no restoration of the
fluorescence intensity was observed whatever the incubation condi-
tions. A cleavage test was also performed with compound 8a
(quencher-free construct) to determine whether cleavage was pre-
vented by the photosensitizer molecule or by the quencher com-
pound. No cleavage was evidenced with MMP-2, as showed by
HPLC analysis (data not shown). This lack of recognition could
hypothetically be explained by a constrained peptide conformation
in the PMB constructs or at least a non-natural peptide conforma-
tion due to the presence of TPC since the free H-GPLGIAGQK-OH
peptide is efficiently cleaved by both MMP-2 and MMP-9. Al-
though different lengths of the linker were tested in different con-
structs in the absence or presence of the quencher without any re-
sponse towards MMP-2 and MMP-9, the problem may come from a
steric hindrance being too high because of the TPC molecule itself.
We could also speculate that the peptide substrate in the PMB con-
structs adopts a conformation adapted for interaction only with non-
catalytic domains of the gelatinases, thus impeding their enzymatic
cleavage. Interactions with cell surface integral membrane proteins
(CD44, ꢀVꢁ/ꢀꢁ1/ꢀꢁ2 integrins and Ku protein,…) can occur
through the gelatinases active site but also via their non-catalytic
domain known as exosite or allosteric region, such as hemopexin-
like C-terminal domain [64]. The non-enzymatic, signal transduc-
tion roles of surface-bound gelatinases (e.g. MMP-2 and MMP-9)
and their subsequent effects on cell survival, migration and angio-
genesis were recently reported [64, 65]. Moreover, beside control of
MMPs’ enzymatic activity, recent study described the conjugation
of gelatinase-binding peptides, identified by phage display, to
liposome by covalent peptide-PEG-PE anchor creating an efficient
targeted drug delivery vehicle [66]. Kratz et al. used a maleimide
moiety right after the first glycine to attach a doxorubicin [41] and
Terada et al. functionalized this same glycine with a PEG [42], in
both cases cleavage of –GPLGIAGQ- occurred. It must also be
remembered that enzymatic cleavage activity in the buffer solution
may be significantly different when compared to biologic tissues.
Jang and Choi (2012) recently demonstrated enzymatic activation
of pyropheophorbide-a photosensitizer-conjugated gold nanorod,
linked through a GPLGVRG MMP-2 specific sequence, after incu-
bation with MMP-2-positive HT1080 cells [67]. Targeting MMPs
was also recently used for several targeted drug delivery approaches
[68, 69]. Singh and colleagues reported the synthesis of bovine
serum albumin nanoparticles stabilized with coatings formed using
peptide domains of varying sensitivity to MMP-2. It was found that
cleavage domain type influenced the release of the model drug from
the nanoparticles [69].
On the basis of this study, the photophysical properties of the
tetraphenylchlorin and BBQ650 pair were investigated. BBQ650
was found to have broad absorption spectra with a maximum in
ethanol at 596 nm, which presents an overlap with TPC fluores-
cence emission, as shown in Fig. (6a). Using equations (2) and (3),
the spectral overlap J was calculated to be 4.9 ꢀ 1015 M-1.cm-1.nm4
and the Förster radius R0 = 54 nm. This appears large enough for a
spatial separation that, in our case, was that of a 9 amino acids long
peptide and a 9 atoms long polyethylene glycol moiety. Given that
peptide folding can bring both ends closer, we thus conclude that
PMB constructs can have an efficient energy transfer.
Due to the strong BBQ absorption at 415 nm (wavelength of
excitation), it was not possible to carry out a FRET experiment in
fluorescence intensity as most of the light is absorbed directly by
the high concentration of BBQ. Time-resolved quenching experi-
ments were then performed in the DMF solution and the Stern-
Volmer plots are shown in Fig. (6b). From the value of quenching
constant K and the fluorescence lifetime the bimolecular quenching
constant, kq BBQ = 4.3 ꢀ 1010 M-1.s-1 can be calculated. This is the
value expected for the diffusion-controlled bimolecular rate con-
stant between TPC and BBQ.
As expected, photophysical properties of conjugates 8b, 9b,
10b, 11b showed great differences with their homologues 8a, 9a,
10a, 11a after coupling of BBQ650. Dramatic decreases of fluores-
cence and singlet oxygen quantum yields, as well as fluorescence
lifetimes were observed in accordance with quenching. Quenching
rate values were between 77 and 100%, indicating that a singlet-
singlet energy transfer occurs between chlorin and BBQ650. This
result was confirmed by transient absorption measurements. Indeed,
the triplet state band at 440 – 450 nm previously observed in TPC
disappears in PMB. Values for all PMB are shown in (Table 2). If
we compare the results obtained with 8b (peptide GPLGIAGQK)
and 11b (peptide PLGLK), it seems that the shortest peptide length
induces better fluorescence and 1O2 quenching, respectively 91%
and 100%. Compounds 9b (short spacer) and 10b (long spacer)
present the less 1O2 quenching respectively 77% and 88%. The
introduction of a spacer does not seem to induce a diminution of
1O2 quenching. This might be due to an increase of the flexibility of
the molecules which prevents close proximity of the photosensitizer
and BBQ650. Further modeling will be carried out in order to better
understand this phenomenon.
Fluorescence lifetimes showed complex fluorescence decays. A
first short decay was observed in the range 1 to 2 ns corresponding
to the complex quenched construct (diffusion controlled reaction or
long distance energy transfer, etc.) and a second mono-exponential
decay corresponding to the long-lived fluorescence of the free TPC
alone. “Double” decays were observed for compounds 8b (peptide
GPLGIAGQK, no spacer) and 11b (peptide PLGLK, no spacer),
while compounds 9b (peptide GPLGIAGQK, small spacer) and 10b
(peptide GPLGIAGQK, long spacer) showed very little signal am-
plitude of quenched construct. Since compounds 9b and 10b had
additional spacers compared to compounds 8b and 11b, it can be
hypothesized that the longer the spacer, the less energy transfer that
occurs. This can be observed by both static spectroscopy (fluores-
cence quantum yields are lower for 9b and 10b than for 8b and
11b) and fluorescence decays (more residual TPC expression ob-
served). The presence of a long lifetime may hide the real signal
linked with the quenching process and the signal processing associ-
ated is not able to robustly eliminate this component in order to
calculate the “pure” decay of the fluorophore. Work on this point is
in progress along with a simulation of the reaction kinetics (to be
published).
In Zheng’s work, no spacer has been added between the photo-
sensitizer and the peptide. They could observed an effective 1O2
quenching (94%) as well as a cleavage by MMP-7 that restored the
quenched 1O2. Unfortunately, we could not observe any cleavage of
our peptide by MMP-2 and MMP-9. We used a related quencher
(BBQ650) and the peptides are of similar size (8 or 9 amino acids).
Different explanations could be envisaged:
•
Conformational structure of the two peptides are different and
in our case, prevent the photosensitizer and the quencher to be
in close proximity,
•
Phenyl groups of the chlorin bring a steric hindrance.
Further modelling will be carried out in order to better under-
stand this phenomenon. Other peptides and photosensitizers could
also be used.

5592 Current Medicinal Chemistry, 2012 Vol. 19, No. 32
Verhille et al.
Fig. (10). Summary of the photophysical and biochemical studies.
CONCLUSION
REFERENCES
[1]
Dougherty, T. J.; Dougherty, T. J.; Gomer, C. J.; Jori, G.; Kessel, D.; Korbe-
lik, M.; Moan, J.; Peng, Q. Photodynamic Therapy. JNCI J Natl Cancer Inst
1998, 90, 889–905.
Macdonald, I. J.; Dougherty, T. J. Basic principles of photodynamic therapy.
Journal of Porphyrins and Phthalocyanines 2001, 5, 105–129.
Weishaupt, K. R.; Gomer, C. J.; Dougherty, T. J. Identification of singlet
oxygen as the cytotoxic agent in photoinactivation of a murine tumor. Can-
cer Res. 1976, 36, 2326–2329.
Mitchell, J. B.; McPherson, S.; DeGraff, W.; Gamson, J.; Zabell, A.; Russo,
A. Oxygen dependence of hematoporphyrin derivative-induced photoinacti-
vation of Chinese hamster cells. Cancer Res. 1985, 45, 2008–2011.
Berdowska, I. Cysteine proteases as disease markers. Clin. Chim. Acta 2004,
342, 41–69.
Duffy, M. J.; Duggan, C. The urokinase plasminogen activator system: a rich
source of tumour markers for the individualised management of patients with
cancer. Clin. Biochem. 2004, 37, 541–548.
Overall, C. M.; Kleifeld, O. Validating matrix metalloproteinases as drug
targets and anti-targets for cancer therapy. Nature Reviews Cancer 2006, 6,
227–239.
van Kempen, L. C. L.; de Visser, K. E.; Coussens, L. M. Inflammation,
proteases and cancer. European Journal of Cancer 2006, 42, 728–734.
DeClerck, Y. A.; Mercurio, A. M.; Stack, M. S.; Chapman, H. A.; Zutter, M.
M.; Muschel, R. J.; Raz, A.; Matrisian, L. M.; Sloane, B. F.; Noel, A.; Hen-
drix, M. J.; Coussens, L.; Padarathsingh, M. Proteases, extracellular matrix,
and cancer: a workshop of the path B study section. Am. J. Pathol. 2004,
164, 1131–1139.
New photodynamic molecular beacons were developed to
achieve multi-level control of 1O2 production by targeting over-
expressed matrix metalloproteinases activity in malignant tissues in
order to provide a higher degree of PDT selectivity. These PMBs
were based on a chlorin singlet excited state quenched by a Black
Berry quencher when the two moieties are connected through pep-
tide sequences specific to the target enzymes. Two peptides have
been chosen: GPLGIAGQK, used as gelatinases substrate for drug
release, and PLGLK as control peptide, known to be cleaved by
numerous MMPs and other proteases. Two different lengths were
selected for the spacers. In total, 4 PMBs were synthesized. We
demonstrated that both fluorescence intensity and 1O2 production
were efficiently quenched in the native PMBs. Quenching rate val-
ues were between 77 and 100 % indicating that a singlet-singlet
energy transfer occurs between the chlorin and BBQ650. The intro-
duction of a spacer, whichever length, does not seem to induce a
[2]
[3]
[4]
[5]
[6]
[7]
[8]
[9]
1
diminution of O2 quenching; this maybe due to an increased flexi-
bility of the molecules. To date, this is the first distance-related
photophysical study concerning PMB and more specifically ge-
latinases-targeted probes involving TPC and BBQ650. Whereas
MMP-2 and MMP-9 efficiently cleaved GPLGIAGQK peptide as
revealed by LC-HRMS, none of the elaborated PMBs were cleaved
by these enzymes, which was probably due to a constrained peptide
conformation in the PMB constructs. Further modelling should be
carried out in order to achieve a better understanding of the phe-
nomena. In the near future we plan to elaborate new PMBs with
others photosensitizers and/or peptides linkage to help understand
the structure/function relationship of MMP-2 and MMP-9.
[10]
[11]
[12]
[13]
Koblinski, J. E.; Ahram, M.; Sloane, B. F. Unraveling the role of proteases in
cancer. Clin. Chim. Acta 2000, 291, 113–135.
Lynch, C. C.; Matrisian, L. M. Matrix metalloproteinases in tumor-host cell
communication. Differentiation 2002, 70, 561–573.
Folkman, J. Angiogenesis in cancer, vascular, rheumatoid and other disease.
Nature Medicine 1995, 1, 27–30.
Kurohane, K.; Tominaga, A.; Sato, K.; North, J. R.; Namba, Y.; Oku, N.
Photodynamic therapy targeted to tumor-induced angiogenic vessels. Cancer
Letters 167, 49–56.
[14]
[15]
[16]
[17]
Chen, B.; Pogue, B. W.; Luna, J. M.; Hardman, R. L.; Hoopes, P. J.; Hasan,
T. Tumor vascular permeabilization by vascular-targeting photosensitization:
effects, mechanism, and therapeutic implications. Clin. Cancer Res. 2006,
12, 917–923.
Fingar, V. H.; Taber, S. W.; Haydon, P. S.; Harrison, L. T.; Kempf, S. J.;
Wieman, T. J. Vascular damage after photodynamic therapy of solid tumors:
a view and comparison of effect in pre-clinical and clinical models at the
University of Louisville. In Vivo 2000, 14, 93–100.
Fingar, V. H.; Wieman, T. J.; Wiehle, S. A.; Cerrito, P. B. The role of mi-
crovascular damage in photodynamic therapy: the effect of treatment on ves-
sel constriction, permeability, and leukocyte adhesion. Cancer Res. 1992, 52,
4914–4921.
Frochot, C.; Di Stasio, B.; Vanderesse, R.; Belgy, M.-J.; Dodeller, M.;
Guillemin, F.; Viriot, M.-L.; Barberi-Heyob, M. Interest of RGD-containing
CONFLICT OF INTEREST
The author(s) confirm that this article content has no conflicts
of interest.
ACKNOWLEDGEMENTS
We would like to thank François Dupire and Cédric Paris for
their expert assistance in mass spectroscopy analysis, and Aurélie
Stallivieri for her assistance in molecular beacons synthesis.

Photodynamic Molecular Beacons Triggered by MMP-2 and MMP-9
Current Medicinal Chemistry, 2012 Vol. 19, No. 32
5593
linear or cyclic peptide targeted tetraphenylchlorin as novel photosensitizers
for selective photodynamic activity. Bioorg. Chem. 2007, 35, 205–220.
liposomes for hepatocellular carcinoma-selective targeting. J Control Re-
lease 2006, 111, 333–342.
[18]
Huang, Z.; Chen, Q.; Luck, D.; Beckers, J.; Wilson, B. C.; Trncic, N.; Larue,
S. M.; Blanc, D.; Hetzel, F. W. Studies of a vascular-acting photosensitizer,
Pd-bacteriopheophorbide (Tookad), in normal canine prostate and spontane-
ous canine prostate cancer. Lasers Surg Med 2005, 36, 390–397.
McMahon, K. S.; Wieman, T. J.; Moore, P. H.; Fingar, V. H. Effects of
photodynamic therapy using mono-L-aspartyl chlorin e6 on vessel constric-
tion, vessel leakage, and tumor response. Cancer Res. 1994, 54, 5374–5379.
Wieman, T. J.; Mang, T. S.; Fingar, V. H.; Hill, T. G.; Reed, M. W.; Corey,
T. S.; Nguyen, V. Q.; Render, E. R., Jr Effect of photodynamic therapy on
blood flow in normal and tumor vessels. Surgery 1988, 104, 512–517.
Verhille, M.; Couleaud, P.; Vanderesse, R.; Brault, D.; Barberi-Heyob, M.;
Frochot, C. Modulation of photosensitization processes for an improved tar-
geted photodynamic therapy. Curr. Med. Chem. 2010, 17, 3925–3943.
Verma, R. P.; Hansch, C. Matrix metalloproteinases (MMPs): chemical-
biological functions and (Q)SARs. Bioorg. Med. Chem. 2007, 15, 2223–
2268.
Berthelot, T.; Lasne, M.-C.; Déléris, G. New trends in molecular imaging of
tumor angiogenesis. Anticancer Agents Med Chem 2008, 8, 497–522.
Brown, S.; Bernardo, M. M.; Li, Z.-H.; Kotra, L. P.; Tanaka, Y.; Fridman,
R.; Mobashery, S. Potent and Selective Mechanism-Based Inhibition of Ge-
latinases. J. Am. Chem. Soc. 2000, 122, 6799–6800.
Mook, O. R. F.; Frederiks, W. M.; Van Noorden, C. J. F. The role of ge-
latinases in colorectal cancer progression and metastasis. Biochim. Biophys.
Acta 2004, 1705, 69–89.
Kline, T.; Torgov, M. Y.; Mendelsohn, B. A.; Cerveny, C. G.; Senter, P. D.
Novel antitumor prodrugs designed for activation by matrix metalloprotein-
ases-2 and -9. Mol. Pharm. 2004, 1, 9–22.
Stefflova, K.; Chen, J.; Zheng, G. Using molecular beacons for cancer imag-
ing and treatment. Front. Biosci. 2007, 12, 4709–4721.
Chen, J.; Stefflova, K.; Niedre, M. J.; Wilson, B. C.; Chance, B.; Glickson, J.
D.; Zheng, G. Protease-Triggered Photosensitizing Beacon Based on Singlet
Oxygen Quenching and Activation. J. Am. Chem. Soc. 2004, 126, 11450–
11451.
Zheng, G.; Chen, J.; Stefflova, K.; Jarvi, M.; Li, H.; Wilson, B. C. Photody-
namic molecular beacon as an activatable photosensitizer based on protease-
controlled singlet oxygen quenching and activation. Proc. Natl. Acad. Sci.
U.S.A. 2007, 104, 8989–8994.
Chen, J.; Jarvi, M.; Lo, P.-C.; Stefflova, K.; Wilson, B. C.; Zheng, G. Using
the singlet oxygen scavenging property of carotenoid in photodynamic
molecular beacons to minimize photodamage to non-targeted cells.
Photochem. Photobiol. Sci. 2007, 6, 1311–1317.
Lovell, J. F.; Chen, J.; Jarvi, M. T.; Cao, W.-G.; Allen, A. D.; Liu, Y.; Tid-
well, T. T.; Wilson, B. C.; Zheng, G. FRET Quenching of Photosensitizer
Singlet Oxygen Generation. J. Phys. Chem. B 2009, 113, 3203–3211.
Lo, P.-C.; Chen, J.; Stefflova, K.; Warren, M. S.; Navab, R.; Bandarchi, B.;
Mullins, S.; Tsao, M.; Cheng, J. D.; Zheng, G. Photodynamic molecular bea-
con triggered by fibroblast activation protein on cancer-associated fibroblasts
for diagnosis and treatment of epithelial cancers. J. Med. Chem. 2009, 52,
358–368.
Chen, J.; Liu, T. W. B.; Lo, P.-C.; Wilson, B. C.; Zheng, G. “Zipper” mo-
lecular beacons: a generalized strategy to optimize the performance of acti-
vatable protease probes. Bioconjug. Chem. 2009, 20, 1836–1842.
Lovell, J. F.; Chen, J.; Huynh, E.; Jarvi, M. T.; Wilson, B. C.; Zheng, G.
Facile synthesis of advanced photodynamic molecular beacon architectures.
Bioconjug. Chem. 2010, 21, 1023–1025.
Popov, A. V.; Mawn, T. M.; Kim, S.; Zheng, G.; Delikatny, E. J. Design and
Synthesis of Phospholipase C and A2-Activatable Near-Infrared Fluorescent
Smart Probes. Bioconjugate Chem. 2010, 21, 1724–1727.
Nakajima, M.; Welch, D. R.; Belloni, P. N.; Nicolson, G. L. Degradation of
basement membrane type IV collagen and lung subendothelial matrix by rat
mammary adenocarcinoma cell clones of differing metastatic potentials.
Cancer Res. 1987, 47, 4869–4876.
Sreenath, T.; Matrisian, L. M.; Stetler-Stevenson, W.; Gattoni-Celli, S.;
Pozzatti, R. O. Expression of matrix metalloproteinase genes in transformed
rat cell lines of high and low metastatic potential. Cancer Res. 1992, 52,
4942–4947.
Nikkola, J.; Vihinen, P.; Vuoristo, M.-S.; Kellokumpu-Lehtinen, P.; Kähäri,
V.-M.; Pyrhönen, S. High serum levels of matrix metalloproteinase-9 and
matrix metalloproteinase-1 are associated with rapid progression in patients
with metastatic melanoma. Clin. Cancer Res. 2005, 11, 5158–5166.
Coussens, L. M.; Fingleton, B.; Matrisian, L. M. Matrix metalloproteinase
inhibitors and cancer: trials and tribulations. Science 2002, 295, 2387–2392.
Mansour, A. M.; Drevs, J.; Esser, N.; Hamada, F. M.; Badary, O. A.; Unger,
C.; Fichtner, I.; Kratz, F. A new approach for the treatment of malignant
melanoma: enhanced antitumor efficacy of an albumin-binding doxorubicin
prodrug that is cleaved by matrix metalloproteinase 2. Cancer Res. 2003, 63,
4062–4066.
[43]
[44]
Netzel-Arnett, S.; Sang, Q. X.; Moore, W. G.; Navre, M.; Birkedal-Hansen,
H.; Van Wart, H. E. Comparative sequence specificities of human 72- and
92-kDa gelatinases (type IV collagenases) and PUMP (matrilysin). Biochem-
istry 1993, 32, 6427–6432.
Jang, J. H.; Kim, M. Y.; Lee, J.-W.; Kim, S. C.; Cho, J. H. Enhancement of
the cancer targeting specificity of buforin IIb by fusion with an anionic pep-
tide via a matrix metalloproteinases-cleavable linker. Peptides 2011, 32,
895–899.
Seybold, P. G.; Gouterman, M. Porphyrins: XIII: Fluorescence spectra and
quantum yields. Journal of Molecular Spectroscopy 1969, 31, 1–13.
Redmond, R. W.; Gamlin, J. N. A compilation of singlet oxygen yields from
biologically relevant molecules. Photochem. Photobiol. 1999, 70, 391–475.
Dale, R. E.; Eisinger, J.; Blumberg, W. E. The orientational freedom of
molecular probes. The orientation factor in intramolecular energy transfer.
Biophys. J. 1979, 26, 161–193.
Knight, C. G.; Willenbrock, F.; Murphy, G. A novel coumarin-labelled
peptide for sensitive continuous assays of the matrix metalloproteinases.
FEBS Lett. 1992, 296, 263–266.
Di Stasio, B.; Frochot, C.; Dumas, D.; Even, P.; Zwier, J.; Müller, A.; Dide-
lon, J.; Guillemin, F.; Viriot, M.-L.; Barberi-Heyob, M. The 2-
aminoglucosamide motif improves cellular uptake and photodynamic activity
of tetraphenylporphyrin. Eur J Med Chem 2005, 40, 1111–1122.
Lindsey, J. S.; Hsu, H. C.; Schreiman, I. C. Synthesis of tetraphenylporphy-
rins under very mild conditions. Tetrahedron Letters 1986, 27, 4969–4970.
Whitlock, H. W.; Hanauer, R.; Oester, M. Y.; Bower, B. K. Diimide reduc-
tion of porphyrins. J. Am. Chem. Soc. 1969, 91, 7485–7489.
Sibrian-Vazquez, M.; Jensen, T. J.; Hammer, R. P.; Vicente, M. G. H. Pep-
tide-mediated cell transport of water soluble porphyrin conjugates. J. Med.
Chem. 2006, 49, 1364–1372.
Valanne, A.; Malmi, P.; Appelblom, H.; Niemelä, P.; Soukka, T. A dual-step
fluorescence resonance energy transfer-based quenching assay for screening
of caspase-3 inhibitors. Anal. Biochem. 2008, 375, 71–81.
Murov, S. L.; Hug, G. L.; Carmichael, I. Handbook Of Photochemistry; M.
Dekker, 1993.
Weissleder, R.; Tung, C.-H.; Mahmood, U.; Bogdanov, A. In vivo imaging of
tumors with protease-activated near-infrared fluorescent probes. Nature Bio-
technology 1999, 17, 375–378.
Tung, C. H.; Bredow, S.; Mahmood, U.; Weissleder, R. Preparation of a
cathepsin D sensitive near-infrared fluorescence probe for imaging. Biocon-
jug. Chem. 1999, 10, 892–896.
Tung, C.-H.; Mahmood, U.; Bredow, S.; Weissleder, R. In Vivo Imaging of
Proteolytic Enzyme Activity Using a Novel Molecular Reporter. Cancer Res
2000, 60, 4953–4958.
Campo, M. A.; Gabriel, D.; Kucera, P.; Gurny, R.; Lange, N. Polymeric
photosensitizer prodrugs for photodynamic therapy. Photochem. Photobiol.
2007, 83, 958–965.
Gabriel, D.; Campo, M. A.; Gurny, R.; Lange, N. Tailoring protease-
sensitive photodynamic agents to specific disease-associated enzymes. Bio-
conjug. Chem. 2007, 18, 1070–1077.
Gabriel, D.; Zuluaga, M.-F.; Martinez, M.-N.; Campo, M. A.; Lange, N.
Urokinase-plasminogen-activator sensitive polymeric photosensitizer pro-
drugs: Design, synthesis and in vitro evaluation. J. Drug Delivery Sci. Tech-
nol. 2009, 19, 15–24.
Forsyth, P. A.; Wong, H.; Laing, T. D.; Rewcastle, N. B.; Morris, D. G.;
Muzik, H.; Leco, K. J.; Johnston, R. N.; Brasher, P. M.; Sutherland, G.; Ed-
wards, D. R. Gelatinase-A (MMP-2), gelatinase-B (MMP-9) and membrane
type matrix metalloproteinase-1 (MT1-MMP) are involved in different as-
pects of the pathophysiology of malignant gliomas. Br. J. Cancer 1999, 79,
1828–1835.
Tirand, L.; Frochot, C.; Vanderesse, R.; Thomas, N.; Trinquet, E.; Pinel, S.;
Viriot, M.-L.; Guillemin, F.; Barberi-Heyob, M. A peptide competing with
VEGF165 binding on neuropilin-1 mediates targeting of a chlorin-type pho-
tosensitizer and potentiates its photodynamic activity in human endothelial
cells. J Control Release 2006, 111, 153–164.
[19]
[20]
[21]
[22]
[45]
[46]
[47]
[48]
[49]
[23]
[24]
[25]
[26]
[50]
[51]
[52]
[27]
[28]
[53]
[54]
[55]
[29]
[30]
[56]
[57]
[58]
[59]
[60]
[31]
[32]
[33]
[34]
[35]
[36]
[61]
[62]
[63]
[37]
[38]
Couleaud, P.; Bechet, D.; Vanderesse, R.; Barberi-Heyob, M.; Faure, A.-C.;
Roux, S.; Tillement, O.; Porhel, S.; Guillemin, F.; Frochot, C. Functionalized
silica-based nanoparticles for photodynamic therapy. Nanomedicine (Lond)
2011, 6, 995–1009.
Bauvois, B. New facets of matrix metalloproteinases MMP-2 and MMP-9 as
cell surface transducers: outside-in signaling and relationship to tumor
progression. Biochim. Biophys. Acta 2012, 1825, 29–36.
Redondo-Muñoz, J.; Ugarte-Berzal, E.; Terol, M. J.; Van den Steen, P. E.;
Hernández del Cerro, M.; Roderfeld, M.; Roeb, E.; Opdenakker, G.; García-
Marco, J. A.; García-Pardo, A. Matrix metalloproteinase-9 promotes chronic
lymphocytic leukemia b cell survival through its hemopexin domain. Cancer
Cell 2010, 17, 160–172.
Medina, O. P.; Pillarsetty, N.; Glekas, A.; Punzalan, B.; Longo, V.; Gönen,
M.; Zanzonico, P.; Smith-Jones, P.; Larson, S. M. Optimizing tumor target-
ing of the lipophilic EGFR-binding radiotracer SKI 243 using a liposomal
nanoparticle delivery system. J Control Release 2011, 149, 292–298.
Jang, B.; Choi, Y. Photosensitizer-Conjugated Gold Nanorods for Enzyme-
Activatable Fluorescence Imaging and Photodynamic Therapy. Theranostics
2012, 2, 190–197.
[64]
[65]
[39]
[40]
[66]
[67]
[41]
[42]
Kratz, F.; Drevs, J.; Bing, G.; Stockmar, C.; Scheuermann, K.; Lazar, P.;
Unger, C. Development and in vitro efficacy of novel MMP2 and MMP9
specific doxorubicin albumin conjugates. Bioorg. Med. Chem. Lett. 2001, 11,
2001–2006.
Terada, T.; Iwai, M.; Kawakami, S.; Yamashita, F.; Hashida, M. Novel PEG-
matrix metalloproteinase-2 cleavable peptide-lipid containing galactosylated

5594 Current Medicinal Chemistry, 2012 Vol. 19, No. 32
Verhille et al.
[68]
[69]
Zhang, Y.; Gemeinhart, R. A. Improving matrix metalloproteinase-2 specific
response of a hydrogel system using electrophoresis. Int J Pharm 2012, 429,
31–37.
matically cleavable domains for controlling the release of model protein drug
from “soft” nanoparticles. Acta Biomater 2012, 8, 636–645.
Singh, H. D.; Bushnak, I.; Unsworth, L. D. Engineered peptides with enzy-
Received: June 18, 2012
Revised: September 04, 2012
Accepted: September 10, 2012