Photosensitized Oxidation in Vesicles
J. Am. Chem. Soc., Vol. 122, No. 11, 2000 2447
estimated that the deactivation of 1O2 generated by the
membrane-bound sensitizer occurred efficiently (90%) in the
lipid layer of the lecithin liposomes. Hoebeke et al.5 also
concluded that with a sensitizer bound to dimyristoyl phos-
formation by transmission electron microscopy with negative
technique (stained with uranyl acetate). The unilamellar layer
of the vesicle was clearly shown in the electron micrograph,
and its thickness was measured to be ∼4-5 nm. The vesicles
were polydisperse with radii ranging from ∼80 to 150 nm, and
the average vesicle radius was ∼100 nm. Vesicles formed in
this way were stable, and the solution was optically clear.10a
The substrates DMOS, TS, DPB and TMP can all be easily
incorporated into the bilayer membranes of the vesicles by
sonication, since they are hydrophobic. Generally, the concen-
tration of the olefins was ∼2.0 × 10-3 M, and that of TMP
was ∼2.0 × 10-2 M, corresponding to thousands of substrate
molecules in each vesicle (see below). The sensitizer TPP can
be solubilized in the bilayer membranes, while MB can be
encapsulated in the aqueous inner compartments of the vesicles.
The concentration of the sensitizer was generally ∼1.0 × 10-4
M.
In the case of TPP as the sensitizer, the vesicles only
incorporating the substrate and those only solubilizing the
sensitizer were prepared separately as described above. Equal
volumes of the two samples were then mixed. Although
sonication was carried out during preparation of the component
solutions, the final mixture was not sonicated. In this way inter-
mixing of solubilizates was prevented. In the case of MB as
the sensitizer, we encapsulated this water-soluble dye in the inner
compartments of one set of vesicles and solubilized the
substrates in the bilayers of another set of vesicles. The MB-
containing vesicles were prepared as described above with the
substitution of saturated MB solution for water. Vesicles formed
in the presence of MB were equal in size to those formed in
pure water. These MB-containing vesicles were chromato-
graphed through a Sephadex G-25 column to remove the MB
from the exterior of the vesicles,11 then the eluting vesicle
dispersions were diluted to 1.0 × 10-4 M. Mixing of the
substrate- and the MB-containing vesicles gave the irradiation
samples. Irradiation of the samples saturated with oxygen by
bubbling the gas was carried out by using a 450-W Hanovia
Hg lamp as the light source, and a glass filter was used to cut
off the light with a wavelength below 400 nm, ensuring the
absence of direct excitation of the alkene and the amine
substrates. For the alkene samples, after irradiation the products
were extracted with ether and analyzed by GC. Generally,
material balance was greater than 95%. For TMP samples, the
nitroxide radical (a product of reaction of 1O2 with TMP)12 was
analyzed by ESR spectroscopy. The yield of this radical was
measured on the basis of the intensity of the ESR signal by
comparison with that of a calibrated concentration of the radical
in solution.
1
phatidylcholine liposomes, O2 spent more than 87% of its
lifetime in the liposome environment. On the other hand,
Reddi,3g Kanofsky,6 Ehrenberg,2c Grossweiner,3f,7 Rodgers,8
Nonell,9 and their co-workers concluded that a significant
fraction of the 1O2 generated by liposome-bound sensitizer
escapes from the lipid bilayer to the water phase and there exists
fast exchange of 1O2 between vesicles and aqueous solution and
even among vesicles.
In the present paper, we report the results of a study of
photosensitized oxidations of three olefins and a sterically
hindered amine in mixed surfactant vesicles. The olefins we
studied were trans-1,2-dimethoxystilbene (DMOS), trans-stil-
bene (TS), and trans,trans-1,4-diphenyl-1,3-butadiene (DPB),
and the amine was 2,2,6,6-tetramethyl piperidine (TMP). The
photosensitizer we used was either a hydrophobic dye, tetra-
phenylporphyrin (TPP), or a cationic dye, methylene blue (MB).
We incorporated the photosensitizers in the bilayer membranes
or the aqueous inner compartments of one set of vesicles and
solubilized the substrate molecules in the bilayer regions of
another set of vesicles. These two sets of vesicle dispersions
were then mixed to give the irradiation samples. By isolating
the products in photooxidation of the olefins, and by detecting
the electron spin resonance (ESR) spectrum of the stable
nitroxide radical in the reaction of 1O2 and the amine, we
demonstrated that singlet oxygen produced in the bilayer region
or in the inner water pool of a vesicle was capable of diffusing
out of the vesicle and reacting with substrates which were
located in other vesicles. Furthermore, the olefin oxidation
products were quite different from those observed with the same
substrates in homogeneous solutions.
Results and Discussion
General. The vesicles selected for investigation were prepared
by sonicating the equimolar mixture of a cationic surfactant
(octyltrimethylammonium bromide, 8.2 × 10-2 M) and an
anionic surfactant (sodium laurate, 8.2 × 10-2 M) in buffered
solution (pH ) 9.2) for 30 min at 50 °C.10 Formation of vesicles
from the mixture of these cationic and anionic surfactants arises
from the strong electrostatic interaction between the oppositely
charged headgroups of the components. As a result, the mean
effective headgroup area decreases considerably, while the mean
hydrophobic volume of the tails remains the same. Thus, this
dynamic ion pairing yields a pseudo-double-tailed zwitterionic
surfactant, which is known to have the preferred geometry of a
vesicle-forming surfactant.10a We demonstrated the vesicle
The above photosensitized oxidations for the samples in
which water was replaced by D2O as dispersion medium were
also performed. The products for all the substrates were identical
to those in the vesicles in water dispersions. However, the
efficiencies for product formation were significantly enhanced.
Reaction of Singlet Oxygen Generated in the Bilayer
Regions of One Set of Vesicles with Olefins Located in
Another Set of Vesicles. As mentioned above, in the case of
TPP as the sensitizer, the vesicles only incorporating the
substrate and those only solubilizing the sensitizer were prepared
separately and then mixed to give the irradiation samples.
(5) Hoebeke, M.; Piette, J.; Vorst, A. J. Photochem. Photobiol., B 1991,
9, 281.
(6) (a) Baker, A.; Kanofsky, J. R. Arch Biochem, Biophys. 1991, 286,
70. (b) Baker, A.; Kanofsky, J. R. Photochem. Photobiol. 1993, 57, 720.
(c) Fu, Y.; Kanofsky, J. R. Photochem. Photobiol. 1995, 62, 692.
(7) (a) Blum, A.; Grossweiner, L. I. Photochem. Photobiol. 1985, 41,
27. (b) Goyal, G. C.; Blum, A.; Grossweiner, L. I. Cancer Res. 1983, 43,
5826.
(8) Rodgers, M. A. J.; Bates, A. L. Photochem. Photobiol. 1982, 35,
473.
(9) Nonell, S.; Braslavsky, S. E.; Schaffner, K. Photochem. Photobiol.
1990, 51, 551.
(10) (a) Kaler, E. W.; Murthy, A. K.; Rodriguez, B. E.; Zasadzinski, J.
A. Science 1989, 245, 1371. (b) Oberdisse, J. Langmuir 1996, 12, 1212.
(c) Hoffman, H.; Thunig, C.; Schmiedel, P.; Munkert, U. Langmuir 1994,
10, 3972. (d) Talhout, R.; Engberts, J. B. F. N. Langmuir 1997, 13, 5001.
(e) Duque, D.; Tarazona, P.; Chacon, E. Langmuir 1998, 14, 6827. (f)
So¨derman, O.; Herrington, K. L.; Kaler, E. W.; Miller D. D. Langmuir
1997, 13, 5531.
(11) Lymar, S. V.; Parmon, V. N.; Zamaraev, K. I. Photoinduced Electron
Transfer Across Membranes. Top. Curr. Chem. 1991, 159, 6.
(12) (a) Lion, Y.; Delmelle, M.; Vorst, A. Nature 1976, 263, 442. (b)
Moan, J.; Wold, E. Nature 1979, 279, 450. (c) Zang, L. Y.; Frederik, J. G.
M. K.; Bibhu, R. M.; Hara, P. M. Biochem. Mol. Biol. Int. 1995, 37(2),
283.