Photochemistry and Photobiology, 2008, 84 767
700 (4.34), 744 (sh) nm; 1H NMR (500 MHz, CDCl3) d )0.24 (br s,
2H, NH), 1.82 (d, 12H, CHMe2), 2.01 (tt, 4H, SCH2CH2CH2), 2.20 (tt,
4H, SCH2CH2CH2), 2.21 (tt, 4H, SCH2CH2CH2), 2.58 (t, 4H,
SCH2CH2CH2), 2.67 (t, 4H, SCH2CH2CH2), 2.67 (t, 4H,
SCH2CH2CH2), 3.85 (s, 4H, OCH3), 3.96 (s, 8H, OCH3), 4.05
(t, 4H, SCH2CH2CH2), 4.12 (t, 4H, SCH2CH2CH2), 4.21 (t, 4H,
SCH2CH2CH2), 5.29 (hp, 2H, CHMe2), 7.67 (s, 2H, ArH); 13C NMR
(125 MHz, CDCl3) 25.7, 25.8, 25.9, 33.0, 33.2, 33.3, 34.6, 34.7, 34.8,
66.08, 66.16, 66.18, 72.7, 119.6, 126.9, 138.2, 139.1, 142.3, 151.0, 152.0,
154.2, 155.0, 173.2, 173.3; ESI-MS (m ⁄ z) calcd for C56H73N8O14S6
[M + H]+ 1273.35, found 1273.
Binding constants. For each porphyrazine derivative, a titration
method (25) was used to determine the binding constant, Kb. To a
cuvette containing 2.7 mL water and 2 lL dye stock (1 mM),
gradually increasing quantities of lipid were added. The mixture
was shaken, after which a fluorescence spectrum was taken. It was
important to make sure that binding had reached equilibrium before
taking the reading, as described in the previous paragraph.
Consequently, after each addition of lipid, the solution was placed
on a shaker for the time taken to reach equilibrium. This varied
from one derivative to another. At high concentrations of lipid,
after shaking for a few minutes, the solution was left to stand for a
few minutes to allow air bubbles to disappear. Fluorescence
excitation and emission wavelengths varied from sample to sample.
For A2B2 and A3B compounds, excitation invariably centered
around 660 nm, and the emission peak, depending on the sensitizer
and its solvent, ranged from 720 to 820 nm. An integral of the
Gaussian emission curve obtained was found to provide the most
accurate representation of the fluorescence intensity. This area was
then plotted on a graph against the lipid concentration. A nonlinear
line was fitted as detailed in Eq. (1).
2,3,7,8,12,13,17,18-Octakis(octyl)-21H,23H-porphyrazine
The porphyrazine was prepared from MNT{S-n-C8H17
(B52).
under the
}
2
usual Linstead, magnesium-ion template (22). The product was
carefully purified by chromatography on silica gel (5% MeOH in
CH2Cl2 eluent): UV–Vis (CH2Cl2) kmax (log ꢀ) 356 (4.66), 502 (2.23),
636 (3.05), 710 (4.08) nm; 1H NMR (500 MHz, CDCl3) d 0.08 (br s,
2H, NH), 0.88 (t, 24H, Me), 1.28–1.38 (m, 80H, CH2), 1.67 (tt, 16H,
SCH2CH2-), 2.68 (t, 16H, SCH2CH2–); 13C NMR (125 MHz, CDCl3)
14.3, 22.9, 28.8, 29.2, 29.4, 29.5, 30.0, 32.1, 32.4, 39.4; ESI-MS (m ⁄ z)
calcd. for C80H139N8S8 [M + H]+ 1467.88, found 1467 .
Finit þ FcompKb½Lꢀ
Stock solutions at 1 mM dye were made up in DMF and kept at
)20ꢂC. Rose Bengal (RB) and hematoporphyrin (HP) were obtained
from Sigma (St. Louis, MO), and TPP from Porphyrin Products
(Logan, UT). Stock solutions in DMF of the three reference
sensitizers, RB, HP and TPP were made up at 2, 4 and 1 mM
F ¼
;
ð1Þ
1 þ Kb½Lꢀ
where the three values Finit, F and Fcomp are the fluorescence intensity
of the dye without lipid, with lipid at concentration L, and that which
would be obtained asymptotically at complete binding, respectively. In
order to obtain Kb for each compound, F was plotted against [L] and
fitted to Eq. (1) by a nonlinear regression routine (Origin, OriginLab
Corporation, Northampton, MA), which calculates Kb and Fcomp. Kb
concentrations, respectively. DMA, used as
a target for singlet
oxygen, was obtained from Sigma. A 2 mM stock solution was
prepared in DMF and kept in the freezer ()20ꢂC), away from the
light.
Solvents. Methanol and DMF were obtained from Frutarom Ltd.
(Haifa, Israel). Diethyl ether (>99.8%) was obtained from Fluka
Chemie (Buchs, Switzerland). n-Octanol was obtained from Merck
(Darmstadt, Germany).
is given in units of (mg lipid mL)1 )1
) .
Singlet oxygen quantum yield. Samples were prepared containing
2 lM porphyrazine dye and 2 lM DMA in 3 mL methanol or benzene
in a 1 cm cuvette. When liposomes were used, the dye and DMA were
added to 3 mL of liposomes (0.2 mg lipid mL)1
) and binding
Lipids. L-a-Phosphatidylcholine (L-a-lecithin) type XIII-E, from
frozen egg yolk, was obtained from Sigma as 100 mg, pre-dissolved in
ethanol at 100 mg mL)1. According to Sigma, the proportion of fatty
acids in solution is as follows: 33% C16:0 (palmitic), 13% C18:0
(stearic), 31% C18:1(oleic) and 15% C18:2 (linoleic) (other fatty acids
being minor contributors), which would give an average weighted
molecular weight of approximately 768.
Preparation of liposomes. To produce a suspension of liposomes, at
a lipid concentration of 5 mg mL)1, 200 lL of phosphatidylcholine
were dispensed into a 15 mL scintillation vial and dried in a stream of
nitrogen. To aid in the drying process, about 0.5 mL of diethyl ether
was added. After 30 min in dry nitrogen, a film of fatty acids coated
the inside of the vial. 5 mL of double-distilled water was added, the
vial was vortexed for 60 s, and placed in a probe sonicator (Soniprep
150, Sanyo ⁄ MSE, Crawley, UK). Sonication was performed at an
amplitude of 10 lm for 10 min in an ice-cold water bath, after which a
clear solution was obtained. Titanium particles from the probe were
removed by centrifugation.
Spectroscopic measurements. Absorption spectra were measured on
a Perkin-Elmer Lambda-9 UV–Vis-near-IR spectrophotometer (Nor-
walk, CT) controlled by a PC or on a Agilent 8453 UV–Visible
Spectrophotometer (Agilent Technologies, Foster City, CA). The
fluorescence intensity, excitation and emission corrected spectra,
anisotropy and time-drive were all measured on a Perkin-Elmer digital
fluorimeter, model LS-50 B, equipped with polarizers in the excitation
and emission beams, and controlled by a PC. Deconvolution of the
absorption spectra was carried out with the aid of the program Peakfit
(Seasolve Software, Inc., Framingham, MA). 1H and 13C NMR
spectra were obtained using a Varian Inova 500 NMR spectrometer
(Varian, Inc., Palo Alto, CA). Electron spray ionization mass spectra
(ESI-MS) were acquired on a LCQ Advantage mass spectrometer
(Thermo Scientific, Waltham, MA).
Liposome binding experiments—Binding kinetics. To make sure the
binding constant was obtained at equilibrium conditions, a kinetic
experiment was carried out at a high concentration of liposomes
(167 lg lipid mL)1). 2 lL of the dye’s stock solution (1 mM) was
added to the liposome solution, and an increase in fluorescence was
observed while stirring (or periodic shaking). No further significant
increase in fluorescence observed gave the minimal time to reach
equilibrium conditions.
conditions were met. A laser was chosen such that its radiation
wavelength was in an absorption band of the sensitizer. Photosensi-
tization was effected for the three reference dyes at 514.5 nm, which
came from an Ar+ laser (Spectra-Physics, Mountain View, CA, model
Beamlok-2060). For most porphyrazines, 675 nm (Sanyo diode laser)
was used for photosensitization. To determine the power of the laser, a
power meter (Ophir Nova, Jerusalem, Israel) was used. A time-drive
experiment was performed while irradiating the sample, monitoring
the decreasing fluorescence intensity of DMA, because of its photo-
oxidation by singlet oxygen. RB, HP and TPP were used as standards
for the solvents methanol, liposomes and benzene, respectively, and
given standard values of FD = 0.80
0.16 (26) for RB,
FD = 0.77
0.13 (27) for HP and FD = 0.66 0.13 (28) for TPP,
against which all measurements of FD for porphyrazines were scaled,
keeping the same standard in each solvent.
All samples were air-saturated and kept in dim light to prevent
inadvertent singlet oxygen production. Samples were stirred with a
magnet during the time-drive to ensure proper homogeneous illumi-
nation throughout the cuvette. The time-drive data were transferred to
the program Origin for graphic and curve-fitting analyses. The rate of
photon absorption by the sensitizer, kpho, is given by Eq. (2) (29):
ꢀ
ꢁ
0:98ꢁPꢁ 1 ꢂ 10ðꢂabsꢁLÞ
kpho
¼
;
ð2Þ
EꢁV
where P is the power of the laser in mW, abs is the optical density of
the sample at the irradiated wavelength, L is the path length traversed
by the laser beam through the sample, E is the Einstein units of light
energy per second per watt of light at the irradiating wavelength and
V is the volume of the sample in the cuvette in mL. The factor 0.98
corrects for the light reflected at the air ⁄ liquid interface.
The disappearance of DMA’s fluorescence follows first-order decay
kinetics according to Eq. (3):
DMAꢁt
DMAflu ¼ Aꢁeꢂk
;
ð3Þ
where kDMA is the rate constant for the decrease of DMA fluorescence,
DMAflu and t represents time in seconds. For each sensitizer, the
quantum yield is proportional to the value of (kDMA ⁄ kpho). If FD,std of