218 Maria A. F. Faustino et al.
oxygen concentrations in the range 0–1.1 ϫ
10 3 mol dm 3 accord-
Ϫ
Ϫ
(Riedel-de-Haen, Seeize, Germany), nitrobenzene (Merck) and p-
toluenesulfonylhydrazine (Aldrich) were used as supplied. Toluene
was dried over sodium wire. Pyridine was dried (NaOH) and dis-
tilled before use. Silica gel (Merck, 35-70 mesh) and neutral alumina
ing to the method previously described by Grewer et al. (13). The
values of the quenching rate constants kT were estimated from the
Q
slope of the plots of 1/T versus [O ] (see [Eq. 1]).
2
0
(
Merck; Brockmann grade III, i.e. deactivated with 4.8% water) were
used for column chromatography. Preparative thin-layer chromatog-
raphy (TLC) was carried out on 20 20 cm glass plates coated
The quantum yields of the S1
S⌬, the efficiency of
quenching of T , were calculated by measuring ⌽⌬ as a function of
→
T1 intersystem crossing
⌽
T
and
1
⌬g formation accompanying the oxygen
ϫ
1
with Merck 60 silica gel (1 mm thick). Analytical TLC was per-
formed using Merck 60 silica gel (precoated sheets 0.2 mm thick).
Reactions were monitored by TLC and UV–visible spectra. Proton
nuclear magnetic resonance (NMR) spectra were obtained in CDCl3
solutions in a 300 MHz Brucker AMX 300 spectrometer; chemical
O concentration in the range 0–5.3
ϫ
10
Ϫ
3
mol dm
Ϫ3
according to
2
the method described earlier (7).
Liposome preparation. The two porphyrin–porphyrin and the por-
phyrin–chlorin dimers were incorporated into DPPC liposomes using
a sonication procedure described elsewhere (14). The photosensitiz-
er:lipid molar ratio in the liposomes was 1:150 in order to obtain a
predominantly monomeric state of the porphyrin in the phospholipid
shifts are expressed in parts per million relative to tetramethylsilane
(TMS). UV–visible spectra were measured in chloroform solution
in a Uvikon spectrophotometer. Mass spectra were obtained on a
VG AutoSpec Q spectrometer.
DL-␣-Dipalmitoylphosphatidylcholine (DPPC, Sigma Chemical
Co, St. Louis, USA) and sodium dodecyl sulfate (SDS, Merck) were
used as received. All other chemicals and solvents were of analytical
grade.
Ϫ1
bilayer and to inject a photosensitizer dose of 1.0 mg kg body
weight. The dimers concentrations in the liposome suspensions were
calculated by diluting a known volume of each suspension into a
known excess of tetrahydrofuran and measuring the corresponding
5
Ϫ
1
Ϫ1
absorbance at 420 nm for dimer D2 (⑀ ϭ 5.18 x 10 M cm ), 421
5
Ϫ
1
Ϫ1
nm for dimer D3 (⑀ ϭ 7.14
ϫ 10 M cm ) and 418 nm for dimer
The measurements of the photophysical properties of the porphy-
rin–porphyrin dimers D2 and D3 and the porphyrin–chlorin dimer
D4 were performed in dichloromethane (Aldrich, spectroscopic
grade), which was purified by column chromatography with basic
Al O (Woelm Pharma, Bad Honnef, Germany) to remove any slight
5
Ϫ
1
Ϫ
1
D4 (⑀ ϭ 5.6
ϫ 10 M cm ).
Animals and tumors. Female Balb/c mice (20–22 g body weight)
were supplied by Charles River (Como, Italy) and kept in cages with
free access to tap water and standard dietary chow. The MS-2 fi-
2
3
brosarcoma was originally supplied by Istituto Nazionale dei Tu-
contamination by hydrochloric acid. Iodopropane (IP), cresyl violet
and phenalenone were purchased from Aldrich and used as supplied.
Absorption and fluorescence spectra. Absorption spectra were re-
corded with a Hewlett–Packard 8452A diode array spectrophotom-
eter. Corrected fluorescence spectra and fluorescence quantum yields
5
mori, Milan, Italy. For tumor implantation 2
ϫ
10 cells in 0.2 mL
of sterile physiological solution were intramuscularly injected into
the right hind leg of the mouse; the tumor growth took place at a
rather aggressive rate, reaching an external diameter of ca 0.7 cm
on the seventh day. All pharmacokinetic studies were performed
within 7–8 days after tumor implantation when no detectable spon-
taneous tumor necrosis had generally occurred. When necessary, the
mice were anesthetized by an intraperitonial injection of Ketalar
were determined with a computer controlled Perkin–Elmer 650-40
0
fluorescence spectrometer. Fluorescence quantum yields
⌽
(in the
F
absence of oxygen) were measured relatively to cresyl violet in air-
saturated methanol used as the reference; ⌽F (cresyl violet) 0.55
0.02 (6).
Rate constant of fluorescence quenching. The rate constant of
ϭ
Ϫ
1
(
150 mg kg ). Animal care was performed according to the guide-
Ϯ
lines established by the Italian Committee for Experimental Ani-
mals.
S
fluorescence quenching by molecular oxygen (k ) was determined
Q
Pharmacokinetic studies. In a typical experimental procedure, the
tumor-bearing mice were i.v.-injected with porphyrin–porphyrin di-
mers or porphyrin–chlorin dimer in DPPC liposomes at a dose of
by measuring the fluorescence lifetimes in dependence of the [O ]
2
Ϫ
3
Ϫ3
in the range 0–5.3
ϫ 10 mol dm according to (Eq. 1)
0
S
1.0 mg kg
Ϫ
1
b.w. At time intervals between 1 and 168 h after in-
1/
s ϭ 1/ ϩ k [O ]
(1)
s
Q
2
jection the mice (three mice at each time point) were sacrificed by
prolonged exposure to ether vapors; blood, tumor and selected nor-
mal tissues (liver, spleen, skin, brain, lung, kidneys and muscle in
the contralateral leg) were quickly removed and washed with saline
0
where S and
are the fluorescence lifetimes in the presence and
s
absence of oxygen, respectively. Fluorescence lifetimes were deter-
mined with a home built emission spectrometer as described previ-
ously (7). Oxygen concentrations were calculated using Bunsen’s
solubility coefficient ␣ ϭ 0.239 (8).
solution, and frozen (Ϫ20ЊC) until the analysis of the photosensitizer
contents were performed. Typically, about 200 mg of tissue was
homogenized with a Polytron in 3 mL of 2% aqueous SDS. After
incubation for 1 h at room temperature under gentle magnetic stir-
ring, the homogenate was diluted (nine-fold for D2 and D4, and
five-fold for D3) with tetrahydrofuran and centrifuged for 10 min at
The method of heavy atom induced intersystem crossing by IP
developed by Dreeskamp et al. (9), was used for the determination
of the S –T energy gap and thus for the determination of the energy
1
1
of the T state of the dimers. IP concentrations were in the range
1
Ϫ
3
0
.1–0.5 mol dm .
3
000 rpm in order to obtain a transparent supernatant. Serum sam-
Rate constants of singlet oxygen (1⌬g) quenching (k ) and quan-
⌬
Q
ples (0.05 mL) isolated from the blood by centrifugation were di-
luted by addition of 2% SDS (0.7 mL) and tetrahydrofuran (1.5 mL
for D2 and D4, and 3 mL for D3) and centrifuged as specified for
the tissue extracts. The tissue and plasma supernatants were analyzed
spectrophotofluorimetrically with a Perkin–Elmer MPF4 instrument.
The samples were excited in the Soret band and the fluorescence
emission of the dimers was monitored in the 600–750 nm range.
The intensities of maximum emission were measured and the pho-
tosensitizer concentrations were calculated by interpolation with a
calibration plot (5).
Synthesis of D2, D3 and D4. The porphyrin–porphyrin dimers D2,
D3 and porphyrin–chlorin dimer D4 were synthesized by the cou-
pling reaction of the acid chloride of porphyrin (2) with, respective-
ly, the amino-substituted compounds (4), (6) and (8). The amino
derivatives (4) and (6) were obtained by reduction of the correspond-
ing nitroporphyrins (3) and (5) with sodium borohydride. The chlor-
in (8) was prepared by reduction reaction of porphyrin (7) with p-
toluenesulfonylhydrazine.
1
⌬
Q
tum yields of ⌬g formation (⌽⌬). The rate constants k were deter-
mined by measuring the ⌬g lifetime via the phosphorescence emis-
1
sion at 1275 nm as a function of porphyrin concentration [PH]. From
1
⌬
Q
the [PH]-dependent ⌬g lifetimes ⌬(PH), k was calculated using
(Eq. 2)
⌬
1/⌬(PH)
ϭ
1/⌬ ϩ k [PH]
(2)
Q
where ⌬ is the 1⌬g lifetime limited only due to quenching by the
solvent (10). Such measurements were carried out in the [PH] range
Ϫ
5
Ϫ4
Ϫ
3
1
ϫ 10 –1.6 ϫ 10 mol dm . Excitation of the dimers was per-
formed at the long wavelength absorption band of the dimers, i.e.
at 646 nm for D2, at 648 nm for D3 and 651 nm for D4 using laser
pulses (dye laser FL 3002 pumped by excimer laser EMG 200;
Lambda Physik, Goettingen, Germany).
⌽⌬
method described by Grewer et al. (11). Phenalenone in air-saturated
CH Cl (12) was used as the reference (exc 436 nm; A 1.2; ⌽⌬
-values were determined in stationary experiments using the
ϭ
ϭ
2
2
ϭ
0.95
Ϯ
0.05).
Porphyrins (1) and (5), the key compounds in the synthesis of
porphyrins (3) and (6), were obtained from the Rothemund and the
crossed Rothemund reactions using the appropriate benzaldehydes
and pyrrole, at reflux in acetic acid and nitrobenzene. Porphyrin (3)
Oxygen quenching of the T states. The oxygen quenching of the
1
T1 state was studied using laser pulse excitation at 516 nm with the
analysis at 450–470 nm. The lifetime T was determined at different
Faustino, Maria A. F.
