Matrix metalloproteases (MMPs) have been found to be highly expressed in a variety of malignant tumor tissues. Noninvasive visualization of MMP activity may play an important role in the diagnosis of MMP associated diseases. Here we report the design and synthesis of a set of fluorine-19 dendron-based magnetic resonance imaging (MRI) probes for real-time imaging of MMP-2 activity. The probes have the following features: (a) symmetrical fluorine atoms; (b) the number of fluorine atoms can be increased through facile chemical modification; (c) readily accessible peptide sequence as the MMP-2 substrate; (d) activatable (19)F signal (off/on mode) via paramagnetic metal ion incorporation. Following optimization for water solubility, one of the probes was selected to evaluate MMP-2 activity by (19)F magnetic resonance spectroscopy (MRS). Our results showed that the fluorine signal increased by 8.5-fold in the presence of MMP-2. The specific cleavage site was verified by mass spectrometry. The selected probe was further applied to detect secreted MMP-2 activity of living SCC7 squamous cell carcinoma cells. The fluorine signal was increased by 4.8-fold by MRS analysis after 24 h incubation with SCC7 cells. This type of fluorine probe can be applied to evaluate other enzyme activities by simply tuning the substrate structures. This symmetrical fluorine dendron-based probe design extends the scope of the existing (19)F MRI agents and provides a simple but robust method for real-time (19)F MRI application.
Matrix metalloproteases (MMPs) have been found to be highly expressed in a variety of malignant tumor tissues. Noninvasive visualization of MMP activity may play an important role in the diagnosis of MMP associated diseases. Here we report the design and synthesis of a set of fluorine-19 dendron-based magnetic resonance imaging (MRI) probes for real-time imaging of MMP-2 activity. The probes have the following features: (a) symmetrical fluorine atoms; (b) the number of fluorine atoms can be increased through facile chemical modification; (c) readily accessible peptide sequence as the MMP-2 substrate; (d) activatable (19)F signal (off/on mode) via paramagnetic metal ion incorporation. Following optimization for water solubility, one of the probes was selected to evaluate MMP-2 activity by (19)F magnetic resonance spectroscopy (MRS). Our results showed that the fluorine signal increased by 8.5-fold in the presence of MMP-2. The specific cleavage site was verified by mass spectrometry. The selected probe was further applied to detect secreted MMP-2 activity of living SCC7squamous cell carcinoma cells. The fluorine signal was increased by 4.8-fold by MRS analysis after 24 h incubation with SCC7 cells. This type of fluorine probe can be applied to evaluate other enzyme activities by simply tuning the substrate structures. This symmetrical fluorine dendron-based probe design extends the scope of the existing (19)F MRI agents and provides a simple but robust method for real-time (19)F MRI application.
Entities:
Keywords:
19F magnetic resonance imaging; 19F magnetic resonance spectroscopy; activatable probe; matrix metalloproteases
Matrix metalloproteases (MMPs) are a family
of endopeptidases composed
of over 25 enzymes that require zinc or calcium to express their catalytic
activities.[1,2] MMPs take part in a set of biological functions
such as extracellular matrix degradation and remodeling, and bioactive
molecules processing and release.[3] In normal
tissues, the expression of active MMPs is low due to strict control
mechanisms, whereas MMPs are overexpressed in numerous pathological
diseases, such as inflammation, and tumor metastasis.[4]Among the MMPs, the gelatinase subfamily including
MMP-2 and MMP-9
have received great attention in the development of anticancer drugs.[5,6] Furthermore, augmented expression and activity of MMP-2 has been
found in a variety of malignant tumor tissues of various organs.[7] Studies have shown that many types of diseases,
such as cancer,[8] diabetes,[9,10] and hypertension,[11,12] are accompanied by elevated blood
level of MMP-2.Noninvasive visualization of activated MMPs
has been a hot pursuit
in recent years. Different imaging modalities, including optical imaging,[13−17] iron oxide-based magnetic resonance imaging (MRI),[18−22] and positron emission tomography,[23,24] were developed
in recent years for the detection of protease activity with promising
results. While different imaging strategies are complementary, optical
imaging is the most widely studied approach to detect protease activity in vivo. Protease activatable near-infrared fluorescent
probes that are specifically cleaved by MMPs to produce fragments
with altered fluorescence have been reported both in chemical conjugates
and in nanoparticle platforms.[13−17] However, optical imaging has limited translational potential into
clinical practice due to the shallow penetration of the excitation/emission
light, and the scattering of light in living subjects.[25]19F magnetic resonance imaging
(19F MRI)
is a promising quantitative imaging technique. 19Fnuclide
has 100% natural isotopic abundance with similar sensitivity to 1H. Notably, negligible endogenous 19F background
in biological systems makes 19F MRI an ideal modality to
monitor protease activity,[26−29] track immune cells,[30] and
quantitatively evaluate neovascular expansion using a clinical scanner.[31] Most current 19F imaging agents are
based on perfluorocarbon emulsions. Perfluoro-15-crown-5-ether (PFCE)
is the most commonly used fluorous agent in 19F MRI,[32] but PFCE is insoluble in most solvents and is
not amenable to chemical modifications. Because of the potential importance
of 19F MRI for disease diagnosis, new 19F probe
design strategies are needed for real-time imaging of biological events
to overcome the shortcomings of perfluorocarbon formulation. To this
end, we designed and synthesized a set of probes with symmetrical
fluorine moiety to detect MMP-2 activity. Our probe features tunable
fluorine content while retaining its symmetrical structure and a single
fluorine resonance. Furthermore, our activatable probe was able to
detect MMP-2 activity through observation of increased 19F signal intensity by magnetic resonance spectroscopy (MRS) and magnetic
resonance imaging (MRI) upon exposure to the enzyme.
Experimental
Section
Reagents and Instrumentation
Unless otherwise stated,
all chemicals were used as received, and the reactions were monitored
by analytical thin layer chromatography (TLC), on Merck precoated
silica gel 60 F254 plates with visualization by ultraviolet irradiation
at λ = 254 nm or staining with KMnO4. Purifications
were performed by silica gel chromatography. Intermediate peptide 10 CGRVGLPG-DOTA was synthesized on an automatic peptide synthesizer
(CS Bio Co., Menlo Park, CA. DOTA is 1,4,7,10-tetraazacyclododecane-1,4,7,10-tetraacetic
acid) by fluorenylmethoxycarbonyl (Fmoc) chemistry at a 1 mmol scale
using a Gly-2-ClTrt resin. 1H, 19F, and 13C NMR spectra were carried out on a Bruker 300 MHz NMR spectrometer,
equipped with a 1H/19F/13C 5 mm broad
band probe. The 1H, 19F, and 13C
NMR spectra were recorded at 300, 282, and 75.5 MHz, respectively.
Mass analysis was conducted on a Waters LC–MS system (Waters,
Milford, MA) that included an Acquity UPLC unit coupled to the Waters
Q-Tof Premier high-resolution mass spectrometer.
p-Toluenesulfonyl chloride (TsCl)
(11.4 g, 60 mmol) was added to a mixture of tetraethylene glycol (21
mL, 23.4 g, 120.6 mmol), Et3N (12.6 mL, 90 mmol), and DMAP
(0.72 g, 6 mmol) in anhydrous CH2Cl2 (300 mL)
at 0 °C in one portion. The reaction mixture was then stirred
at room temperature (rt) overnight. H2O was added to quench
the reaction, and the mixture was then extracted with CH2Cl2. Removing the combined solvent under vacuum followed
by silica gel flash chromatography using CH2Cl2/MeOH (20/1) as the eluent afforded compound 2 (13.1
g, 63% yield) as a yellow oil. 1H NMR (300 MHz, CDCl3) δ 7.70–7.67 (m, 2H), 7.28–7.24 (m, 2H),
4.07–4.04 (m, 2H), 3.60–3.54 (m, 6H), 3.53–3.51
(m, 2H), 3.50–3.46 (m, 6H), 3.03 (br, 1H), 2.34 (s, 3H).
2-(2-(2-(2-Azidoethoxy)ethoxy)ethoxy)ethanol, 3
To a solution of tetraethylene glycol monotosylate 2 (13 g, 37.4 mmol) in DMF (200 mL) was added sodium azide
(3.6 g, 56 mmol) carefully. The reaction mixture was stirred at 60
°C overnight. After the solution was cooled to rt, water was
added slowly to dissolve minor remaining solid. The mixture was then
extracted with ethyl acetate. The combined organic mixture was washed
successively with water and brine and concentrated by rotary evaporation.
The residue was purified by silica gel column chromatography using
CH2Cl2/MeOH (20/1) as the eluent to afford compound 3 (6.9 g, 84% yield) as a light yellow oil. 1H
NMR (300 MHz, CDCl3) δ 3.76–3.59 (m, 12H),
3.57–3.54 (m, 2H), 3.36–3.33 (m, 2H), 2.91 (br, 1H).
To the azide compound 3 (9.32 g, 42.6
mmol) in CH2Cl2 (200 mL) was added Et3N (8.4 mL, 60.0 mmol), p-toluenesulfonyl chloride
(11.4 g, 59.9 mmol) successively 0 °C, the mixture was stirred
overnight at rt. The mixture was poured into water, and the organic
layer was separated and evaporated through rotary evaporation. The
residue was purified by silica gel column chromatography using hexane/EtOAc
(1/1) as the eluent to afford compound 4 (7.6 g, 93%
yield) as a light yellow liquid. 1H NMR (300 MHz, CDCl3) δ 7.82–7.78 (m, 2H), 7.36–7.33 (m, 2H),
4.17–4.14 (m, 2H), 3.67–3.58 (m, 8H), 3.59–3.58
(m, 4H), 3.40–3.37 (m, 2H), 2.45 (s, 3H).
To a suspension of 60% sodium hydride (1.6 g, 40
mmol)
in 90 mL of anhydrous THF at 0 °C was added tetraethylene glycol 1 (6.98 g, 3.6 mmol) in 20 mL of anhydrous THF dropwise under
nitrogen. The reaction was stirred at 0 °C for 1 h and continued
at rt for another 2 h. Tosylate 4 (6.7 g, 1.8 mmol) in
THF (20 mL) was added dropwise to the refluxing solution of sodium
alcoholate; then the mixture was refluxed for another 10 h. After
cooling to rt, the mixture was quenched with H2O carefully.
Upon solvent evaporation, the residue was extracted with EtOAc, washed
with H2O, and saturated sodium chloride solution successively;
the combined organic phase was concentrated through rotary evaporation
and subjected to silica gel chromatography using CH2Cl2/MeOH as the eluent to afford compound 5 (5.2
g, 73% yield) as a clear oil. 1H NMR (300 MHz, CDCl3) δ 3.69–3.66 (m, 4H), 3.65–3.61 (m, 20H),
3.58–3.56 (m, 4H), 3.35 (t, J = 4.2 Hz, 4H),
2.62 (br, 1H).
To the azide compound 5 (3 g, 7.6
mmol) in CH2Cl2 (35 mL) was added Et3N (1.5 mL, 11 mmol), then p-toluenesulfonyl chloride
(2.1 g, 11 mmol) was added in one portion at 0 °C; the reaction
was stirred overnight at rt. The mixture was poured into water, and
the organic layer was separated and evaporated through rotary evaporation.
The residue was purified by silica gel column chromatography using
hexane/EtOAc (1/1) as the eluent to afford compound 6 (3.2 g, 76% yield) as a light yellow liquid. 1H NMR (300
MHz, CDCl3) δ 7.62 (d, J = 8.1 Hz,
2H), 7.19 (d, J = 8.1 Hz, 2H), 4.00–3.97 (m,
2H), 3.52–3.48 (m, 24H), 3.47–3.40 (m, 4H), 3.21 (t, J = 5.1 Hz, 2H), 2.28 (s, 3H). 13C NMR (75.5
MHz, CDCl3) δ 144.5, 132.6, 129.5, 127.6, 70.3, 70.24,
70,22, 70.17, 70.10, 69.6, 69.1, 68.2, 50.3, 21.2.
To the azide compound 6 (1.0 g, 1.8
mmol) in dry DMF (9 mL) was added sodium perfluoro-tert-butoxide (0.94 g, 3.6 mmol), which was prepared according to the
literature.[33] The reaction mixture was
stirred at 65 °C overnight, and H2O was added to the
cool down mixture. The mixture was extracted with EtOAc, the pooled
organic extracts were washed with water and brine, the combined organic
phase was concentrated under reduced pressure through rotary evaporation,
and the residue was purified by silica gel flash chromatography using
hexane/EtOAc (1/1) as the eluent to afford compound 7 (0.78 g, 70% yield) as a clear liquid. 1H NMR (300 MHz,
CDCl3) δ 4.11 (t, J = 4.8 Hz, 2H),
3.71–3.64 (m, 2H), 3.63–3.59 (m, 26H), 3.34 (t, J = 5.1 Hz, 2H). 19F NMR (282 MHz, CDCl3) δ 70.52. 13C NMR (75.5 MHz, CDCl3)
δ 120.4 (q, J = 292.9 Hz), 80.8–79.3
(m), 71.1, 70.75, 70.73, 70.69, 70.64, 70.08, 69.44, 69.38, 69.36,
69.34, 69.32, 50.7. Mass (ESI) m/z 614.2 [M + H]+.
To the azide compound 7 (0.78 g, 1.3
mmol) in dry THF (10 mL) was added Ph3P (0.6 g, 2.3 mmol).
Upon the completion of the reaction as confirmed by TLC, water (0.23
mL) was added and the reaction continued overnight at rt. After removal
of the solvent in vacuum, the residue was purified by silica gel column
chromatography first using CH2Cl2/MeOH (16/1)
then MeOH as the eluent to give compound 8 (0.7 g, 94%
yield). 1H NMR (300 MHz, CDCl3) δ 4.00
(t, J = 4.5 Hz, 2H), 3.58 (t, J =
4.8 Hz, 2H), 3.50 (s, 24H), 3.39 (t, J = 5.4 Hz,
2H), 2.73 (br, 2H), 2.37 (br, 2H). 19F NMR (282 MHz, CDCl3) δ 70.68. 13C NMR (75.5 MHz, CDCl3) δ 120.1 (q, J = 295.2 Hz), 80.2–78.6
(m), 72.5, 70.9, 70.42, 70.36, 70.34, 70.31, 70.1, 69.2, 41.4. Mass
(ESI) m/z 588.7 [M + H]+.
To the amino compound 8 (235 mg, 0.4 mmol) in CH2Cl2 (4 mL) was added N,N-diisopropylethylamine (0.11 mL, 0.6
mmol); then 3-(maleimido)propionic acid N-hydroxysuccinimide
ester (80 mg, 0.36 mmol) was added at 0 °C, and the mixture was
stirred at rt for 4 h. The mixture was concentrated under vacuum,
and the residue was purified by silica gel column chromatography using
CH2Cl2/MeOH (16/1) as the eluent to afford compound 9 (196 mg, 73% yield). 1H NMR (300 MHz, CDCl3) δ 6.70 (s, 2H), 6.58 (br, 1H), 4.14 (t, J = 4.2 Hz, 2H), 3.84–3.79 (m, 2H), 3.74–3.70 (m, 2H),
3.64–3.59 (m, 24H), 3.53 (t, J = 5.1 Hz, 2H),
3.42–3.37 (m, 2H), 2.50 (t, J = 6.9 Hz, 2H). 19F NMR (282 MHz, CDCl3) δ 70.95. 13C NMR (75.5 MHz, CDCl3) δ 170.7, 170.2, 134.4, 120.4
(q, J = 297.5 Hz), 80.3–79.1 (m), 70.8, 70.7,
70.4, 69.83, 68.76, 68.4, 66.0, 65.6, 46.4, 39.6, 34.7, 34.6, 29.9.
Mass (ESI) m/z 739.2 [M + H]+.
Conjugation of DOTA Conjugated Peptide PEP-DOTA 10 with F9-PEG-Mal 9
To the peptide 10 (253 mg, 0.22 mmol) in degassed PBS (150 mL) was added
a solution
of compound 9 (196 mg, 0.27 mmol) in degassed EtOH (30
mL). The mixture was stirred at rt under argon and monitored by analytical
reversed-phase high performance liquid chromatography (RP-HPLC). The
mixture was quenched by 0.1% aqueous TFA and concentrated through
rotary evaporation. The residue was purified by preparative HPLC.
The proper fraction was collected and lyophilized to afford fluorine-containing
peptide as a white solid (316 mg, 76% yield). Mass (ESI) m/z 942.6 [M + 2H]2+. 19F NMR
(282 MHz, D2O) δ 70.50. For semipreparative HPLC,
a Beckman Ultrasphere C18 column (10 × 250 mm) and
a gradient elution profile were used with 0.5% phosphoric acid in
water (solvent A) and 0.5% phosphoric acid in CH3CN (solvent
B). The elution profile was isocratic at 5% solvent B for 5 min, then
a gradient to 80% solvent B over 45 min. The flow rate was 4 mL/min.
The major peak at about 27.0 min was collected. The purity of the
resulting compound was conducted by analytical HPLC.
Synthesis
of Probe F9-PEG-Mal-PEP-DOTA-Gd, 11
A DOTA-containing
peptide (75 mg) was dissolved in PBS, GdCl3·6H2O (5 equiv) was added, and the pH of the
solution was adjusted to 4–5. The mixture was heated at 80
°C, and the reaction was monitored by HPLC; typically the reaction
was completed in 4 h. The mixture was centrifuged and subject to semipreparative
HPLC. A Beckman Ultrasphere C18 column (10 × 250 mm)
and a gradient elution profile were used with 0.5% phosphoric acid
in water (solvent A) and 0.5% phosphoric acid in CH3CN
(solvent B). The gradient elution profile was from 5% solvent B to
80% solvent B in 50 min, then to 100% solvent over the next 5 min.
The flow rate was 4 mL/min. The major peak at 34.4 min was collected
and lyophilized. Analytical HPLC was used to confirm the purity (Condition:
4.6 × 150 mm Phenomenex C18 column, 1 mL/min, detection
wavelength at 214 nm, elution profile, a gradient from 5% solvent
B to 60% solvent B in 15 min, and to 100% solvent B over next 5 min).
The retention time of the fluorinated peptide was 13.7 min. Mass (ESI) m/z 1019.6 [M + 2H]2+.
To a suspension of 60% sodium hydride (0.16 g,
4.0 mmol) in 10 mL of THF at 0 °C was added alcohol 12 (2.04 g, 2.4 mmol), which was prepared according to the reported
procedure,[33] in 6 mL of anhydrous THF dropwise.
The reaction was stirred at 0 °C for 1 h and then at rt for another
2 h. Tosylate 6 (1.1 g, 2.0 mmol) in THF (6 mL) was added
dropwise to the refluxing solution of sodium alcoholate. The mixture
continued to reflux at 70 °C for overnight. Upon being cooled
to rt, the mixture was quenched with H2O. After solvent
evaporation, the residue was extracted with EtOAc, washed with H2O, concentrated through rotary evaporation, and subjected
to silica gel chromatography using hexane/EtOAc as eluent to afford
compound 13 (1.05 g, 43% yield) as a clear oil. 1H NMR (300 MHz, CDCl3) δ 3.91 (s, 6H), 3.56–3.48
(m, 28H), 3.43–3.40 (m, 2H), 3.37–3.25 (m, 4H), 3.24–3.22
(m, 4H), 1.73–1.65 (m, 2H). 19F NMR (282 MHz, CDCl3) δ 71.00. 13C NMR (75.5 MHz, CDCl3) δ 120.1 (q, J = 292.9 Hz), 80.4–78.5
(m), 70.64, 70.61, 70.55, 70.45, 70.3, 70.2, 70.0, 68.5, 68.0, 65.7,
65.4, 50.6, 46.1, 29.6.
To a solution of compound 14 (150
mg, 0.11 mmol) in anhydrous DMF (1 mL) at 0 °C was added 3-maleimidopropionic
acid (29 mg, 0.17 mmol), 1-hydroxybenzotriazole (HOBt) (23 mg, 0.17
mmol), 1-(3-(dimethylamino)propyl)-3-ethylcarbodiimide hydrochloride
(EDC·HCl) (33 mg, 0.17 mmol), 4 Å molecular sieves (100
mg), and DIPEA (30 μL, 0.17 mmol) successively. The mixture
was stirred at 0 °C for 0.5 h and then at rt overnight. The mixture
was quenched with ice water and extracted with CH2Cl2. The organic extracts were washed with water and brine successively.
After drying the organic layer and evaporation of the solvent under
vacuum, the residue was purified by silica gel column chromatography
using CH2Cl2/MeOH to afford the product at 100
mg; 30 mg of starting material 14 was recovered (81%
yield after recovery). 1H NMR (300 MHz, CDCl3) δ 6.70 (s, 2H), 6.42 (br, 1H), 4.03 (s, 6H), 3.86–3.71
(m, 2H), 3.65–3.62 (m, 26H), 3.57–3.52 (m, 4H), 3.50–3.42
(m, 6H), 3.36 (s, 2H), 2.52 (t, J = 7.2 Hz, 2H),
1.87–1.78 (m, 2H). 19F NMR (282 MHz, CDCl3) δ 70.52. 13C NMR (75.5 MHz, CDCl3)
δ 170.6, 170.2, 134.4, 120.3 (q, J = 293.7
Hz), 80.2–79.0 (m), 70.70, 70.66, 70.4, 69.8, 68.8, 68.3, 65.9,
65.6, 46.3, 39.6, 34.6, 34.5, 29.8. Mass (ESI) m/z 1350.4 [M + H]+.
Cell Culture Experiment
Squamous cell carcinoma cells
(SCC7) were cultured in RPMI-1640 medium supplemented with 10% fetal
bovine serum (FBS) (Hyclone) 100 μg/mL penicillin, and 100 U/mL
streptomycin at 37 °C with 5% CO2.
MMP-2 Concentration
Determination
We collected cell
culture medium supernatant to evaluate the capacity of the developed
probe to detect the secreted MMP-2 by SCC7cancer cells. Briefly,
the monolayer of cultured cells were washed two times with ice cold
PBS, followed by adding fresh complete cell culture medium 24 h prior
to supernatant collection. The next day, the supernatant was harvested
and centrifuged at 3000 rpm for 3 min to get rid of cell debris. The
concentration of MMP-2 in the supernatant was quantified by MMP-2
ELISA assay (R&D) following the manufacturer’s protocol.
19F NMR and Imaging
The probe was added
to MMP-2 buffer (26.9 nM) or SCC7 cell medium (1% 2,2,2-trifluoroethanol
with 5% D2O); the final concentration of the probe is 0.25
mM. The mixture was put into an NMR tube and incubated at 37 °C. 19F NMR was acquired at different time points using the same
parameters. A broad spectrum pan-MMP inhibitor (100 μM) was
added to the solution for control experiments. For 19F
NMR imaging, the experiments were performed with a home-built surface
coil tuned to 281.65 MHz, which is the 19F resonance frequency
at 7 T. A 2D FLASH image was acquired on the fluorine-containing sample,
with the following acquisition parameters: TE = 3 ms, TR = 100 ms,
FOV = 4 × 4 cm2, matrix size = 128 × 128, slice
thickness = 2 mm, and FA = 30 deg.
Results
Design of Probes
We synthesized a probe containing
four specific segments. The first is the perfluorinated signal emitting
portion A. This portion is somewhat tunable in that we
are able to chemically synthesize a F9 or F27 moiety while maintaining
the symmetry of 19F signal. The unit B is
a polyethylene glycol chain that enhances water solubility of the
final construction. The unit C is a peptide substrate
of MMP-2. The unit D is a chelated Gd3+ ion
that serves as a signal modulator of the 19F resonance
due to its strong paramagnetic effect and shortening of transverse
relaxation time (T2). As constructed,
the proximity of the Gd3+ to the 19F resulted
in significant attenuation of the 19F signal. Upon MMP-2
incubation, the MMP substrate is site-specifically cleaved and the
paramagnetic effect of Gd3+ toward 19F is canceled,
which induces the extension of T2 and
recovery of 19F signal. The recovery rate of 19F signal is proportional to the MMP-2 activity (Figure 1).
Figure 1
Schematic of 19F-based MMP activatable probe. Fn represents
fluorous unit with different numbers of symmetric fluorinated branches
for signal emitter; PEG is poly(ethylene glycol).
Schematic of 19F-based MMP activatable probe. Fn represents
fluorous unit with different numbers of symmetric fluorinated branches
for signal emitter; PEG is poly(ethylene glycol).
Probe Synthesis and Characterization
The PEG linker
was synthesized starting from commercially available tetraethylene
glycol 1. Selective protection of one of the hydroxyl
groups with tosyl chloride afforded compound 2; reaction
of compound 2 with sodium azide gave compound 3. A second tetraethylene glycol unit was linked to tosylate 4 to produce azide 5 with eight ethylene glycol
units. Symmetrical F9 unit was introduced to the glycol linker by
the SN2 displacement of tosylate 6 with sodium
per-fluoro-tert-butoxide. Reduction of azide 6 by the classical Staudinger reaction afforded free amine 8. A maleimido group, for subsequent thiol conjugation, was
introduced by the condensation of compound 8 with commercially
available 3-(maleimido)propionic acid N-hydroxysuccinimide
ester in high yield to give compound 9. Compounds 7–9 exhibited single, sharp 19F resonances with no loss of symmetry in the fluorinated unit.Next we introduced the MMP-2 substrate into the PEGylated fluorinated
unit. The MMP-2 cleavable peptide substrate 10, Cys-Gly-Arg-Val-Gly-Leu-Pro-Gly-DOTA (abbreviated as PEP-DOTA), was
synthesized by a standard solid-phase Fmoc peptide chemistry. The
reaction between the free thiol group in peptide 10 and
maleimido linker 9 proceeded smoothly with high yield.
The 19F signal modulator Gd3+ was introduced
to the 1,4,7,10-tetraazacyclododecane-1,4,7,10-tetraacetic acid (DOTA)
chelator in high efficiency (Scheme 1). The
purity and identity of 11 were confirmed by analytical
HPLC and LC–MS.
Scheme 1
Synthesis of Activatable Probe F9-PEG-Mal-PEP-DOTA-Gd
Containing
9 Symmetric Fluorine Atoms
Red color in compound 10 represents the specific cleavage site. Reagents and conditions:
(a) TsCl, Et3N, DCM, rt; (b) NaN3, DMF, 65 °C;
(c) TsCl, Et3N, DCM; (d) tetraethylene glycol, NaH, THF,
rt to reflux; (e) TsCl, Et3N, THF; (f) (CF3)3CONa, DMF; (g) Ph3P, THF, then H2O,
rt; (h) 3-(maleimido)propionic acid N-hydroxysuccinimide
ester, DIPEA, DMF; (i) PBS/EtOH (v/v, 4/1), pH 7.4, rt; (j) GdCl3·6H2O, PBS, pH 4–5, 80 °C.
Synthesis of Activatable Probe F9-PEG-Mal-PEP-DOTA-Gd
Containing
9 Symmetric Fluorine Atoms
Red color in compound 10 represents the specific cleavage site. Reagents and conditions:
(a) TsCl, Et3N, DCM, rt; (b) NaN3, DMF, 65 °C;
(c) TsCl, Et3N, DCM; (d) tetraethylene glycol, NaH, THF,
rt to reflux; (e) TsCl, Et3N, THF; (f) (CF3)3CONa, DMF; (g) Ph3P, THF, then H2O,
rt; (h) 3-(maleimido)propionic acid N-hydroxysuccinimide
ester, DIPEA, DMF; (i) PBS/EtOH (v/v, 4/1), pH 7.4, rt; (j) GdCl3·6H2O, PBS, pH 4–5, 80 °C.
19F-NMR Properties
Results
by 19F spectroscopy showed about 11 times difference in 19F
intensity before and after Gd3+ incorporation, indicating
the strong paramagnetic effect of Gd3+ toward 19F nuclei and the potential of this probe to detect enzyme activity.Next we synthesized more highly branched probes with higher fluorine
content. According to a reported strategy,[33] we synthesized alcohol 12 with 27 symmetrical fluorine
atoms. Reaction of 12 with PEGlyated linker 6 afforded azide compound 13. The azide was subjected
to reduction to give free amine 14, then condensation
of amine 14 with 3-maleimidopropionic acid produced 15 (Scheme 2). All the synthesized
compounds 12–15 showed single, sharp 19F resonances. Next we attempted to attach F27 linker 15 to MMP-2 substrate 10 via the sulfide–maleimide
reaction. This reaction failed under the various conditions attempted.
This difficulty could come from the strong hydrophobicity of compound 15, which did not allow its dispersion/dissolvation into the
aqueous media required for this coupling reaction.
Scheme 2
Synthetic Scheme
Towards Activatable Probe Containing Symmetric 27
Fluorine Atoms
First we used the probe to detect activity of
purified MMP-2. Probe 11 (0.25 mM) was incubated with
MMP-2, and 19F
NMR was used to monitor the 19F signal recovery at 37 °C
over time. Time-dependent 19F NMR spectra, including intensity
and signal/noise ratio were recorded (Figure 2). Before adding MMP-2, 19F signal exhibited a broad peak
with full width at half the maximum (fwhm) of around 12.7 Hz due to
the strong paramagnetic effect of Gd3+ toward 19F. Upon MMP-2 incubation, 19F signal started to increase
at 15 min and an apparent sharp/signal appeared at 60 min, which further
increased over time, until reaching a plateau at 17 h. The fwhm dropped
to 2.1 Hz, and the 19F intensity increased by 8.5-fold
(Figure 3). Time-dependent signal/noise ratio
of 19F NMR showed the similar trend with 19F
intensity change. MMP-2 specificity of the probe was confirmed by
coincubation with a broad spectrum pan-MMP inhibitor.[34,35] There was almost no change of 19F intensity when MMP
inhibitor (MMPI) was added to the system, indicating the substrate
specificity. The fluorine signal and signal/noise ratio remained unchanged
in blank buffer, suggesting that the fluorine signal recovery was
induced by MMP specific cleavage. LC–MS further confirmed the
site-specific cleavage of the peptide in the presence of MMP-2.
Figure 2
Time-dependent 19F NMR spectral changes of F9-PEG-Mal-PEP-DOTA-Gd
(0.25 mM) after adding MMP-2 (26.9 nM) at 37 °C. Reaction buffer:
50 mM Tris, 10 mM CaCl2, 150 mM NaCl, 0.05% Brij35, pH
7.8, 5% D2O.
Figure 3
Recovery of 19F signal in the presence of pure MMP,
pure MMP with MMP inhibitor, and probe without MMP treatment (control
in blank buffer). Probe concentration is 0.25 mM.
Time-dependent 19F NMR spectral changes of F9-PEG-Mal-PEP-DOTA-Gd
(0.25 mM) after adding MMP-2 (26.9 nM) at 37 °C. Reaction buffer:
50 mM Tris, 10 mM CaCl2, 150 mM NaCl, 0.05% Brij35, pH
7.8, 5% D2O.Recovery of 19F signal in the presence of pure MMP,
pure MMP with MMP inhibitor, and probe without MMP treatment (control
in blank buffer). Probe concentration is 0.25 mM.
Secretive MMP-2 Activity Detection of Living Cells
With
the purified MMP-2 activity evaluated, we further applied our
probe for real-time monitoring of secreted MMP-2 activity in living
cells. SCC7 cell line was chosen because these cells secrete a high
level of MMP-2. Before the detection of MMP activity, an ELISA MMP
assay kit was used to estimate the MMP concentration, and the level
was found to be 8.5 ng/mL in cell culture medium. The newly synthesized 19F probe was then incubated with the cell medium. The 19F signal recovery was apparent at 60 min, and the signal
reached a plateau at 36 h. Compared with pure MMP, the signal increased
about 4.8-fold (Figure 4 and 5). In contrast, the 19F signal only increased 1.9-fold
in the presence of MMP-2 inhibitor, and no significant 19F signal change was found when the probe was incubated with either
blank buffer or pure water. Enzymes other than MMP-2 in the cell medium
may also lead to nonspecific cleavage of the probe, which explains
why MMP-2 inhibitor was unable to completely suppress 19F signal.
Figure 4
Time-dependent 19F NMR spectral changes of F9-PEG-Mal-PEP-DOTA-Gd
(0.25 mM) after the addition of SCC7 cell medium at 37 °C; MMP-2
concentration was 8.5 ng/mL determined by MMP assay kit.
Figure 5
Recovery of 19F signal in the presence of SCC7
cell
medium, cell medium with MMP inhibitor treatment, and probe incubation
in the presence of blank buffer or pure water. Probe concentration
is 0.25 mM.
Time-dependent 19F NMR spectral changes of F9-PEG-Mal-PEP-DOTA-Gd
(0.25 mM) after the addition of SCC7 cell medium at 37 °C; MMP-2
concentration was 8.5 ng/mL determined by MMP assay kit.Recovery of 19F signal in the presence of SCC7
cell
medium, cell medium with MMP inhibitor treatment, and probe incubation
in the presence of blank buffer or pure water. Probe concentration
is 0.25 mM.
Relaxation Time Determination
To further evaluate the
MMP activity secreted by SCC7cancer cells, we measured the longitudinal
relaxation time T1 and and transverse
relaxation time T2 of the 19F signal under different conditions. The results showed that F9-PEG-Mal-PEP-DOTA
substrate without Gd chelation exhibited long T1 and T2 with strong 19F signal. In contrast, both T1 and T2 were significantly reduced after Gd3+ incorporation, indicating the strong paramagnetic effect of Gd3+ toward 19F, which leads to over 20 times reduction
of T2 and broadening of 19F
signal. However, this effect was dissipated during the probe incubation
in SCC7 conditioned medium, in which the T1 and T2 increased rapidly with significant 19F signal recovery. This signal recovery was correlated with
MMP-2 activity in the conditioned medium. Compared to Gd3+ free conditions, the T1 and T2 values of Gd3+ complex after incubation
with cell medium did not fully recover. This might be due to the fact
that Gd3+ is still in the solution, but that interaction
is reduced due to an increased average distance between Gd3+ ions and 19F atoms[28] (Table 1).
Table 1
T1 and T2 Change of Synthesized
MMP Substrates before
and after SCC7 Cell Medium Incubation
T1 (ms)
T2 (ms)
F9-PEG-Mal-PEP-DOTA
1760
518
F9-PEG-Mal-PEP-DOTA-Gd
33.2
17.3
F9-PEG-Mal-PEP-DOTA-Gd + SCC7 cell medium
245.7
57.7
19F MRI Imaging of MMP-2 Activity
19F MRI imaging was performed to show the 19F signal intensity
change upon treatment with SCC7 cell medium (Figure 6). Before SCC7 conditioned cell medium incubation, there was
no 19F imaging signal except for background noise. After
incubation with conditioned medium overnight, the 19F MRI
was clearly visualized under the same fluorine concentration. The
promising results further showed that our probe was able to visualize
MMP activity.
Figure 6
19F imaging of a solution (2.5 mM) of probe 11 before (left) and after (middle) SCC7 cell medium incubation.
As
a control, a solution (right) of F9-PEG-Mal-PEP-DOTA without Gd3+ incorporation was also imaged. The experiments were performed
with a homemade surface coil tuned to 281.65 MHz, which is the 19F resonance frequency at 7 T. A 2D FLASH image was acquired
on the probe, with the following acquisition parameters: TE = 3 ms,
TR = 100 ms, FOV = 4 × 4 cm2, matrix size = 128 ×
128, slice thickness = 2 mm, and FA = 30 deg.
19F imaging of a solution (2.5 mM) of probe 11 before (left) and after (middle) SCC7 cell medium incubation.
As
a control, a solution (right) of F9-PEG-Mal-PEP-DOTA without Gd3+ incorporation was also imaged. The experiments were performed
with a homemade surface coil tuned to 281.65 MHz, which is the 19F resonance frequency at 7 T. A 2D FLASH image was acquired
on the probe, with the following acquisition parameters: TE = 3 ms,
TR = 100 ms, FOV = 4 × 4 cm2, matrix size = 128 ×
128, slice thickness = 2 mm, and FA = 30 deg.
Discussion
MMPs play an important role in a series
of normal physiological
processes with low expression level. However, MMPs are overexpressed
in numerous pathological diseases, such as inflammation, and tumor
metastasis. Noninvasive imaging of activated MMPs and other proteases
has been extensively studied in recent years for tumor diagnosis.
While optical imaging is the most used strategy for visualization
of MMPs, its combination with iron oxide-based nanoformulation for
multimodality imaging has received great attention.[19,20] Chen et al. designed an Au–Fe3O4 composite
activatable nanoparticle by integrating the excellent quenching property
of gold nanoparticle toward fluorescent dyes and easily tunable surface
chemistry of iron oxide nanoparticle for imaging of MMPs with great
success.[19] Kim et al. reported a MRI/NIRF-based
dual modal imaging probe composed of silica-coated iron oxide nanoparticles
and Cy5.5-MMP unit that can be visualized in vivo by both MRI and optical imaging.[20]1H MRI is extensively used clinically for diagnosis
due to the high concentration of water in the human body (55 M).[36] Traditional anatomical 1H MRI relies
on signal from protons in water of tissues; contrast agents such as
positive Gd-based[37,38] or negative iron oxide-based
contrast agents[39] are used to enhance contrast.
However, for metal-based contrast agents, image artifacts are sometimes
produced due to the difficulty in discrimination between the targeted
tissues and susceptibility associated with surrounding environment. 19F MRI detects 19F nuclei directly with negligible
background and can be used for quantitative purpose in living subjects.
The advancement of 19F MRI has lagged behind that of 1H MRI although 19F was investigated early in the
history of MRI.[40,41] Unlike the protons from water
molecules, most fluorine compounds have nonequivalent fluorine atoms
that spread the resonance signals through a relatively wide chemical
shift range.A series of 19F-based activatable probes
responsive
to pH,[42,43] temperature,[44] enzymatic activity,[27,28] and redox potential[45] have been reported. Most of the developed probes
contain a single trifluoromethyl group; thus, sensitivity is an issue
in these model systems. Hence, increasing the number of fluorine atoms
in a single molecule is one strategy to enhance the sensitivity of
the 19F probes. Hamachi et al. reported several 19F-based self-assembling nanoprobes for protein detection. Results
showed that the sensitivity of the designed probes is linear to the
number of the incorporated fluorine atoms.[29] Our recently developed fluorous dendron-cyanine dye-conjugated nanoprobe
successfully achieved optical/MR bimodal imaging in vivo.[46] Here we described a set of MMP-responsive
probes with highly symmetrical fluorinated branches, which exhibited
a single/sharp resonance with negligible chemical shift change during
incubation with enzyme. Perfluorocarbon-based imaging agents, including
emulsions and polymers, usually exhibit split 19F resonance,
and may cause imaging artifacts.[32] PFCE
with highly symmetric fluorine atoms and considerable fluorine content
makes it the most commonly used fluorous agent in 19F MRI,
but PFCE is insoluble in almost all solvents and is not amenable to
chemical modifications. Complex formulation procedures are required
for imaging purposes. Compared to PFCE and other perfluorocarbon-based
imaging agents, our probe exhibited highly symmetric signal emitter,
and the fluorine content was easily tunable. In addition, water solubility
was modulatable by switching the length of PEGylated chain or the
number of symmetric fluorine branches.Extensive studies of 19F MRI in recent years have pointed
out the comparatively low sensitivity observed when compared to 1H MRI.[36,41] Our activatable probe encountered
similar issue since millimolar concentration was required to provide
enough 19F signal to visualize MMP activity. Further advances
in both instrumentation and chemistry, such as increasing the number
of equivalent fluorine in the probe design for higher sensitivity,[26] are required to improve the detection limit.
Compared to iron oxide nanoparticles, 19F MRI is able to
quantitatively evaluate MMP level with low background noise, but the
sensitivity of 19F MRI agent is lower than iron oxide nanoparticles.
Especially for clinical translation, the sensitivity of 19F MRI remains a challenge because local 19F concentration
at the target site is far below the 1H level in the human
body. The use of higher dose of 19F agents for clinical
application may provide better sensitivity, assuming target uptake
correlates with dose, but introduces the potential trade-off between
sensitivity and toxicity for regulatory approval. However, PFC based
imaging agents have been approved for a clinical trial in the US.[30] Another challenge would be the selectivity and
specificity of the probe for protease detection due to considerable
concentration of MMP-2/MMP-9 in atherosclerotic vessels, major arteries,
and microvasculature. The developed probe needs to be sensitive enough
to discriminate the target pathological site. By appropriate installation
of symmetrical fluorine branches and tuning the substrate to different
targets, we may use this strategy to image other enzymatic pathology in vivo.
Conclusions
In conclusion, a fluorous
probe was developed to detect MMP activity
of living cells. The probe features a highly symmetrical fluorine
signal emitting unit, with a remarkable time-dependent fluorine signal
recovery upon incubation with MMP-2 enzyme, indicating that our probe
is capable of specifically monitoring MMP-2 activity in real-time.
Our reported 19F-based probe can be widely used to detect
enzyme activities by properly changing the substrate structures. This
symmetrical fluorine dendron-based probe design extends the scope
of existing 19F MRI agents and provides another simple
but robust method for real-time 19F MRI application.
Authors: S Zucker; M Hymowitz; C Conner; H M Zarrabi; A N Hurewitz; L Matrisian; D Boyd; G Nicolson; S Montana Journal: Ann N Y Acad Sci Date: 1999-06-30 Impact factor: 5.691
Authors: Francois Fay; Line Hansen; Stefanie J C G Hectors; Brenda L Sanchez-Gaytan; Yiming Zhao; Jun Tang; Jazz Munitz; Amr Alaarg; Mounia S Braza; Anita Gianella; Stuart A Aaronson; Thomas Reiner; Jørgen Kjems; Robert Langer; Freek J M Hoeben; Henk M Janssen; Claudia Calcagno; Gustav J Strijkers; Zahi A Fayad; Carlos Pérez-Medina; Willem J M Mulder Journal: Bioconjug Chem Date: 2017-05-05 Impact factor: 4.774
Authors: Lina A Basal; Matthew D Bailey; Jonathan Romero; Meser M Ali; Lyazat Kurenbekova; Jason Yustein; Robia G Pautler; Matthew J Allen Journal: Chem Sci Date: 2017-10-26 Impact factor: 9.825
Authors: In Ok Ko; Ki-Hye Jung; Mi Hyun Kim; Kyeung Jun Kang; Kyo Chul Lee; Kyeong Min Kim; Insup Noh; Yong Jin Lee; Sang Moo Lim; Jung Young Kim; Ji-Ae Park Journal: Contrast Media Mol Imaging Date: 2017-09-26 Impact factor: 3.161
Authors: Henryk M Faas; James L Krupa; Alexander J Taylor; Francesco Zamberlan; Christopher J Philp; Huw E L Williams; Simon R Johnson; Galina E Pavlovskaya; Neil R Thomas; Thomas Meersmann Journal: Contrast Media Mol Imaging Date: 2019-02-28 Impact factor: 3.161
Authors: Beatriz Brito; Thomas W Price; Juan Gallo; Manuel Bañobre-López; Graeme J Stasiuk Journal: Theranostics Date: 2021-08-07 Impact factor: 11.556