Effect of structural modification on photodynamic activity of hypocrellins

Article abstract of DOI:10.1562/0031-8655(2001)074<0143:EOSMOP>2.0.CO;2

Aluminum ion complexed 5,8-di-Br-hypocrellin B is a new water-soluble perylenequinonoid derivative with enhanced absorption over hypocrellin B (HB) in the phototherapeutic window (600-900 nm). Electron paramagnetic resonance and 9,10-diphenyl-anthracene bleaching methods were used to investigate the photosensitizing activity of [Al₂(5,8-di-Br-HB)Cl₄]_n in the presence of oxygen. Singlet oxygen, superoxide anion radical and hydroxyl radical can be generated by [Al₂(5,8-di-Br-HB)Cl₄]_n photosensitization. Singlet oxygen (¹O₂) is formed via energy transfer from triplet-state [Al₂(5,8-diBr-HB)Cl₄]_n to ground-state molecular oxygen. ¹O2 participates in the generation of a portion of superoxide anion radical (0₂^.-). Besides superoxide anion radical (O₂^.-) may originate from the electron transfer between the triplet-state [Al₂(5,8-di-Br-HB)Cl₄]_n and the ground-state molecular oxygen. OH is formed through the Fenton-Haber-Weiss reaction and the decomposition of DMPO-¹O₂ adduct. Compared with HB [Al₂(5,8-di-Br-HB)Cl₄]_n primarily remains and enhances the generation efficiency of superoxide anion radical and hydroxyl radical but that of singlet oxygen decreases.

Full text of DOI:10.1562/0031-8655(2001)074<0143:EOSMOP>2.0.CO;2

Effect of Structural Modification on Photodynamic Activity of Hypocrellins
Author(s): Jianghua Ma, Jingquan Zhao, Lijin Jiang
Source: Photochemistry and Photobiology, 74(2):143-148.
Published By: American Society for Photobiology
DOI: http://dx.doi.org/10.1562/0031-8655(2001)074<0143:EOSMOP>2.0.CO;2
URL: http://www.bioone.org/doi/full/10.1562/0031-8655%282001%29074%3C0143%3AEOSMOP
%3E2.0.CO%3B2
BioOne (www.bioone.org) is a nonprofit, online aggregation of core research in the biological, ecological, and
environmental sciences. BioOne provides a sustainable online platform for over 170 journals and books published
by nonprofit societies, associations, museums, institutions, and presses.
Your use of this PDF, the BioOne Web site, and all posted and associated content indicates your acceptance of
BioOne’s Terms of Use, available at www.bioone.org/page/terms_of_use.
Usage of BioOne content is strictly limited to personal, educational, and non-commercial use. Commercial inquiries
or rights and permissions requests should be directed to the individual publisher as copyright holder.
BioOne sees sustainable scholarly publishing as an inherently collaborative enterprise connecting authors, nonprofit publishers, academic institutions, research
libraries, and research funders in the common goal of maximizing access to critical research.

Photochemistry and Photobiology, 2001, 74(2): 143–148
Symposium-in-Print
Effect of Structural Modification on Photodynamic Activity of
Hypocrellins
Jianghua Ma, Jingquan Zhao* and Lijin Jiang
Center for Molecular Science, Institute of Chemistry, The Chinese Academy of Sciences, Beijing, People’s Republic of
China
Received 5 January 2001; accepted 15 May 2001
ABSTRACT
essential factors; specifically, the photosensitizer is a key
factor for PDT. So far the only photosensitizer approved by
the United States Food and Drug Administration is Photofrin
Aluminum ion complexed 5,8-di-Br-hypocrellin B is a
new water-soluble perylenequinonoid derivative with en-
hanced absorption over hypocrellin B (HB) in the pho-
totherapeutic window (600–900 nm). Electron paramag-
netic resonance and 9,10-diphenyl-anthracene bleaching
methods were used to investigate the photosensitizing ac-
tivity of [Al2(5,8-di-Br-HB)Cl4]n in the presence of oxy-
gen. Singlet oxygen, superoxide anion radical and hy-
droxyl radical can be generated by [Al2(5,8-di-Br-
HB)Cl4]n photosensitization. Singlet oxygen (1O2) is
formed via energy transfer from triplet-state [Al2(5,8-di-
௡
II . However, it possesses several serious disadvantages
(2,3). The limitations have led to the investigations of sec-
ond-generation photosensitizers, in which hypocrellins have
attracted more attention (4).
Hypocrellins, including hypocrellin A (HA) and hypo-
crellin B (HB) (Fig. 1), are new types of photosensitive pig-
ments growing abundantly in China (5,6). They exhibit sev-
eral advantages over the presently used hematoporphyrin de-
rivatives, i.e. availability in pure monomeric form, facility
for site-directed chemical modifications, high quantum
yields of singlet oxygen and fast clearance from normal tis-
sue (4).
1
Br-HB)Cl4]n to ground-state molecular oxygen. O2 par-
ticipates in the generation of a portion of superoxide an-
ion radical (O2 ). Besides superoxide anion radical (O2
•Ϫ
•Ϫ
)
may originate from the electron transfer between the
triplet-state [Al2(5,8-di-Br-HB)Cl4]n and the ground-state
molecular oxygen. OH is formed through the Fenton–
Haber–Weiss reaction and the decomposition of DMPO–
1O2 adduct. Compared with HB [Al2(5,8-di-Br-HB)Cl4]n
primarily remains and enhances the generation efficiency
of superoxide anion radical and hydroxyl radical but that
of singlet oxygen decreases.
However, hypocrellins are lipophilic agents and exhibit
weak absorption in the phototherapeutic window (600–900
nm), which prevents further investigations on their biologi-
cal consequences and limits their PDT application clinically.
In order to improve the red absorption and amphiphilicity of
hypocrellins a variety of hypocrellin derivatives have been
synthesized, including the sulfonated (7,8), sulfurized (9,10),
aminated (11–16), halogenated (17,18) and metal ion–com-
plexed (18–20) compounds. Their relevant photophysical
and photochemical properties have been investigated (7–20).
Among these derivatives, brominated hypocrellins not
only retained similar quantum yields of singlet oxygen (1O2)
to the parent compounds (17), but also showed obviously
•
INTRODUCTION
Photodynamic therapy (PDT)† is a novel therapeutic meth-
odology for tumors (1). Selective photosensitizer and strong
light source that matches with the photosensitizer are two
•
enhanced efficiency of hydroxyl radical ( OH) generation un-
der heterogeneous system (23). It was also found that bro-
minated hypocrellins could photodynamically kill Hela cells
(24) and damage calf thymus DNA (25) more efficiently
than their parent compounds.
*To whom correspondence should be addressed at: Institute of
Photographic Chemistry, The Chinese Academy of Sciences,
Beijing 100101, People’s Republic of China. Fax: 86-10-
64879375; e-mail: zhaojq@ipc.ac.cn
†Abbreviations: DABCO, 1,4-diazabicyclo [2.2.2] octane; DMPO,
5,5-dimethyl-1-pyrroline-N-oxide; DMSO, dimethylsulfoxide;
DPA, 9,10-diphenyl-anthracene; DTPA, diethylene triamine pen-
taacetic acid; EPR, electron paramagnetic resonance; HA, hypo-
crellin A; HB, hypocrellin B; NADH, reduced nicotinamide ad-
enine dinucleotide; OD, optical density; PDT, photodynamic ther-
apy; SOD, superoxide dismutase; TEMP, 2,2,6,6-tetramethyl-4-
piperidone; TEMPO, 2,2,6,6-tetramethyl-4-piperidone-N-oxyl
radical.
Investigations demonstrated that the balance between li-
pophilicity and hydrophilicity was an important factor for
photosensitizing efficiencies and tumor or cellular uptake
(21,22). However, brominated hypocrellins are basically hy-
drophobic compounds that prevent effective drug delivery in
vivo, and their redshifts are still not enough to act as agents
on solid tumors. It was known that the complex of HA with
ϩ
᭧
2001 American Society for Photobiology 0031-8655/01
$5.00
ϩ0.00
Al3 has enhanced red absorption and excellent water solu-
143

144 Jianghua Ma et al.
Figure 1. The chemical structures of hypocrellins.
bility over HA (20). Therefore, in combination with the ad-
ϩ
vantages of brominated and Al3 -complexed hypocrellins we
have first designed and synthesized the complex of 5,8-di-
Br-HB with Al3 in high yield (26). The resulting product
ϩ
Figure 2. Absorption spectra of HB and [Al2(5,8-di-Br-HB)Cl4]n
measured in DMSO solution. HB (3.3
10Ϫ5 mol·LϪ1), [Al2(5,8-di-
[Al2(5,8-di-Br-HB)Cl4]n exhibits excellent water-solubility
and enhanced absorption over HB in the phototherapeutic
window (600–900 nm). The new photosensitizer was char-
ϫ
1
5
Br-HB)Cl4]n (0.16 mg·mLϪ 10Ϫ mol·LϪ1).
; Յ 3.3 ϫ
1
acterized by UV–visible, IR, H-nuclear magnetic resonance
Determination of the quantum yield of 1O2 generated by DPA-
bleaching method. The DPA-bleaching method was used to deter-
mine the quantum yields of 1O2 generation, the details of which were
described in an earlier report (28). The photo-oxidation of DPA sen-
sitized by [Al2(5,8-di-Br-HB)Cl4]n was carried out on a ‘merry-go-
round’ apparatus, using a high-pressure mercury lamp (500 W) with
a 578 nm monochromatic filter as the light source. The reactions
were followed spectrophotometrically by observing the decrease of
374 nm absorption peak for DPA (where the sensitizer used has the
lowest absorption) as a function of irradiation time.
and elemental analysis measurements. 5,8-di-Br-HB forms a
ϩ
2:1 type (metal–ligand ratio) complex with Al3 measured
by molar ratio and continuous variation methods. Based on
the above experimental results we first proposed a polymer-
like structural model for [Al2(5,8-di-Br-HB)Cl4]n (26).
As a potential phototherapeutic agent it is necessary to
investigate the photosensitizing activities of [Al2(5,8-di-Br-
HB)Cl4]n. The photoinduced generation of semiquinone an-
ion radical by [Al2(5,8-di-Br-HB)Cl4]n in the absence of ox-
ygen has been discussed in detail elsewhere (26). Herein, we
will report the photosensitized generation of active oxygen
species in the presence of oxygen. The generation efficien-
cies of active oxygen species by [Al2(5,8-di-Br-HB)Cl4]n
were compared with those of HB and their generation mech-
anism was also discussed.
RESULTS
The new photosensitizer—[Al2(5,8-di-Br-HB)Cl4]n—is easi-
ly soluble in polar organic solutions such as DMSO, DMF,
CH3CN, CH3CH2OH, CH3OH, and even in H2O. The ab-
sorption peak of [Al2(5,8-di-Br-HB)Cl4]n shifted red to 614
nm (Fig. 2). It is obvious that the water solubility and red
absorption of [Al2(5,8-di-Br-HB)Cl4]n are both enhanced
over those of HB. Under physiological pH value the com-
plex did not change its absorption spectrum and there was
no precipitation formation for a long time; therefore, it can
be concluded that the complex possesses high hydrolysis sta-
bility in physiological conditions. In order to survey the ef-
fect of chemical modifications on the photosensitization ac-
tivities of the new photosensitizer, it is necessary to inves-
tigate the efficiency and mechanism of production of the
active species. The type-I mechanism of [Al2(5,8-di-Br-
HB)Cl4]n in the absence of oxygen has been discussed else-
where in detail (26). Herein, we will specially concentrate
on the generation of active oxygen species in the presence
of oxygen by [Al2(5,8-di-Br-HB)Cl4]n photosensitization.
MATERIALS AND METHODS
Chemicals. HB and 5,8-di-Br-HB was prepared as described pre-
viously (27,17). Al3ϩ-complexed 5,8-di-Br-HB was synthesized ac-
cording to our previous report (26). 9,10-Diphenyl-anthracene
(DPA), 5,5-dimethyl-1-pyrroline-N-oxide (DMPO), 2,2,6,6-tetrame-
thyl-4-piperidone (TEMP), diethylene triamine pentaacetic acid
(DTPA) and 2,2,6,6-tetramethyl-4-piperidone-N-oxyl radical (TEM-
PO) were purchased from Aldrich Chemical Company (Milwaukee,
WI). 1,4-Diazabicyclo [2.2.2] octane (DABCO) and dimethylsulf-
oxide (DMSO) were purchased from Merck Chemical Company
(Darmstadt, Germany). Catalase and superoxide dismutase (SOD)
were purchased from Sigma Chemical Co (St. Louis, MO). Deuter-
ated solvents and other agents of analytical grades were purchased
from Beijing Chemical Plant (Beijing, China). Water was freshly
distilled before use. pH values of the irradiated solutions were ad-
justed by using KH2PO4–Na2HPO4 buffers (pH 7.4). The working
stock solutions were prepared immediately before use.
Electron paramagnetic resonance measurements. The electron
paramagnetic resonance (EPR) measurements were performed at
room temperature using a Bruker ESP 300E spectrometer. A 532
nm Quanta-Ray Nd:YAG laser was used as the light source. The
following instrumental settings were used: microwave power, 8.08
mW; modulation amplitude, 1 G; sweep width, 100 G; receiver gain,
Generation of singlet oxygen (1O2) by [Al2(5,8-di-Br-
HB)Cl4]n
It has been previously reported that TEMPO, a nitroxide
radical detectable by EPR, could be generated from reaction
of TEMP with singlet oxygen (29). In addition the DPA-
bleaching method was confirmed to be an efficient measure-
1.25
ϫ
104. Photoinduced EPR spectra were measured from the
L) injected quantitatively into quartz capillaries de-
sample (40
␮
signed specially for EPR analysis.
1
ment for the quantum yield of O2 generation (28).
To make the production efficiencies of all spin adducts compa-
rable, the concentrations of [Al2(5,8-di-Br-HB)Cl4]n and HB were
adjusted to keep the same optical density (OD) at 532 nm. To min-
imize measurement errors the same quartz capillary tube was used
throughout the EPR measurements. Magnetic parameters of the rad-
icals detected were obtained from direct measurements of magnetic
field and microwave frequency.
EPR detection of singlet oxygen. When an air-saturated
Ϫ
1
DMSO solution of [Al2(5,8-di-Br-HB)Cl4]n (0.5 mg·mL ),
Ϫ
1
Ϫ1
SOD (40 ␮g·mL ), HCO2Na (0.1 mol·mL ) and TEMP (20
mM) was irradiated at room temperature an EPR spectrum
of triplet peaks with equal intensity, characteristic of a ni-

Photochemistry and Photobiology, 2001, 74(2) 145
Figure 4. Photosensitized DPA bleaching by measuring the absor-
Figure 3. The EPR signal intensity of TEMPO formed in an air-
bance decrease (⌬A) at 374 nm as a function of irradiation time in
saturated DMSO solution of [Al2(5,8-di-Br-HB)Cl4]n (0.5 mg·mLϪ1),
oxygen-saturated DMSO solution containing [Al2(5,8-di-Br-
HB)Cl4]n (0.08 mg·mLϪ1) and DPA (0.3 mM) (line A). Line B: same
as line A except that [Al2(5,8-di-Br-HB)Cl4]n was replaced by HB
SOD (40 ␮
g·mLϪ1), HCO2Na (0.1 mol·mLϪ1) and TEMP (20 mM)
as a function of illumination time using 532 nm light (line A). Line
B: same as line A except that [Al2(5,8-di-Br-HB)Cl4]n, oxygen or
illumination was omitted. Line C: same as line A except that DAB-
CO (5 mM) was added. Line D: same as line A except that DMSO
was replaced by fully deuterated DMSO. The inset is a typical EPR
spectrum of TEMPO formed from [Al2(5,8-di-Br-HB)Cl4]n under the
same conditions as line A.
(39
␮M). Line C: same as line A except that [Al2(5,8-di-Br-
HB)Cl4]n, oxygen or light was omitted. Line D: same as line A
except that DABCO (2 mM) was added.
•Ϫ
Generation of superoxide anion radical (O2
)
•Ϫ
It is known that O2 is relatively stable in aprotic solvents,
ϩ
where no free hydrogen ions (H ) exist for the dismutation
•Ϫ
of O2 . Thus the experiments were performed in DMSO.
troxide radical, was observed (Fig. 3, inset). The hyperfine
When the air-saturated DMSO solution containing [Al2(5,8-
splitting constant and g factor of the EPR spectrum were
Ϫ1
N
di-Br-HB)Cl4]n (0.5 mg·mL ) and DMPO (50 mM) was ir-
radiated an EPR signal appeared immediately (Fig. 5, inset).
This EPR spectrum was characterized by three coupling con-
stants, which are due to the nitrogen and two hydrogen at-
oms at
constants (g
1.3 G) were in good agreement with the literature values
for DMPO–superoxide radical adduct (31). Control experi-
ments confirmed that [Al2(5,8-di-Br-HB)Cl4]n, oxygen and
light were all necessary for the production of EPR signal
shown in inset of Fig. 5 (Fig. 5, line B). The addition of
identical with those of commercial TEMPO (
␣
ϭ
16.3 G,
g
ϭ 2.0056) (29). In the absence of [Al2(5,8-di-Br-HB)Cl4]n,
oxygen or irradiation, no EPR signal could be detected (Fig.
3, line B). These data demonstrated that the formation of the
nitroxide radical was a photosensitization process.
␤ and ␥ positions. The g factor and the determined
N
H
H
ϭ
2.0056,
␣
ϭ
12.8 G,
␣
ϭ 10.5 G and ␣
In the presence of 5 mM DABCO (a well-known 1O2 scav-
enger) the EPR signal was suppressed completely (Fig. 3,
line C), which provided further evidence for generation of
1O2 by photosensitization of [Al2(5,8-di-Br-HB)Cl4]n. Fur-
thermore, the effect of deuterium solution on the yield of
TEMPO was also studied. It was found that the intensity of
the EPR signal increased by approximately four times when
DMSO was replaced by fully deuterated DMSO as solvent
␤
␥
ϭ
1
(Fig. 3, line D). This is because the lifetime of O2 in deu-
terated solvents is 10–15 times longer than that in nondeu-
terated solvents (30). These two powerful diagnostic tools
1
for O2 suggest that TEMPO is derived from the reaction of
1
TEMP with O2 generated by [Al2(5,8-di-Br-HB)Cl4]n.
Determination of 1O2 quantum yield. The 9,10-DPA
bleaching method was applied to determine the quantum
1
yield of O2 generated by [Al2(5,8-di-Br-HB)Cl4]n with HB
as the reference. The optical densities of [Al2(5,8-di-Br-
HB)Cl4]n and HB at 578 nm were adjusted to the same.
Figure 4 showed the rates of DPA-bleaching photosensitized
by [Al2(5,8-di-Br-HB)Cl4]n (line A) and HB (line B) as a
function of irradiation time in DMSO. Control experiments
indicated that no DPA bleaching occurred when photosen-
sitizers, oxygen or irradiation were absent (Fig. 4, line C).
The addition of DABCO (2 mM) inhibited DPA bleaching
(Fig. 4, line D). These confirmed that DPA-bleaching re-
sulted from the reaction of DPA with 1O2 formed by
Figure 5. Dependence of the DMPO-superoxide radical adduct in-
tensity on irradiation time derived from irradiation of an air-saturat-
ed DMSO solution containing [Al2(5,8-di-Br-HB)Cl4]n (0.5
mg·mLϪ1) and DMPO (50 mM) (line A). Line B: same as line A
except that [Al2(5,8-di-Br-HB)Cl4]n, oxygen or illumination was
omitted. Line C: same as line A except that SOD (30
added. Line D: same as line A except that [Al2(5,8-di-Br-HB)Cl4]n
␮
g·mLϪ1) was
1
[Al2(5,8-di-Br-HB)Cl4]n photosensitization. The O2-gener-
ating quantum yield by [Al2(5,8-di-Br-HB)Cl4]n in DMSO
was estimated to be 0.33, relative to HB (0.76).
•Ϫ
was replaced by HB (39 ␮M). Inset: the EPR signal of DMPO–O2
adduct generated by [Al2(5,8-di-Br-HB)Cl4]n under the conditions as
line A (a.u., arbitrary unit).

146 Jianghua Ma et al.
•Ϫ
Figure 6. The dependence of the relative intensity of DMPO–O2
•
Figure 7. (A) The EPR spectrum of DMPO– OH adduct resulting
adduct signal after 2 min irradiation on [Al2(5,8-di-Br-HB)Cl4]n con-
centration.
from irradiation of an air-saturated aqueous solution (pH 7.4) con-
taining [Al2(5,8-di-Br-HB)Cl4]n (0.5 mg·mLϪ1) and DMPO (50 mM).
(B) Same as in (A) except that ethanol (10%, vol/vol) was added
prior to illumination. (C) Same as in (A) except that DABCO (10
mM) was added. (D) Same as in (A) except that H2O was replaced
by D2O at pH 7.4. (E) Same as in (A) except that catalase (30
Ϫ
1
SOD (30
␮g·mL ) prior to illumination inhibited the EPR
signal intensity (Fig. 5, line C). These observations suggest-
ed the correct assignment of the EPR spectrum to the
␮
␮
g·mLϪ1) was added. (F) Same as in (A) except that SOD (25
g·mLϪ1) was added. (G) Same as in (A) except that DTPA (5 mM)
•Ϫ
DMPO–O2 adduct.
was added.
It was previously mentioned that the {[Al2(5,8-di-Br-
•Ϫ
HB)Cl4]n} signal appeared only in the presence of electron
donor reduced nicotinamide adenine dinucleotide (NADH)
•
DMPO– OH spin adduct (33). In the aqueous solution no
•Ϫ
(26), while the DMPO–O2 could also be generated in the
•
EPR signal of DMPO– OH was observed in the absence of
light or the complex. The observed DMPO–OH signal may
be due to the decomposition of the DMPO–O2 adduct or,
alternatively, direct trapping of OH radicals produced by
•Ϫ
absence of NADH, which suggested that the formation of O2
•
in the absence of NADH was not possible through Eq. 1.
•Ϫ
•
•Ϫ
{[Al2(5,8-di-Br-HB)Cl4]n}
ϩ O2
[Al2(5,8-di-Br-HB)Cl4]n photosensitization, or a combination
of both processes. In order to determine the source of the
•Ϫ
→
O2
ϩ
[Al2(5,8-di-Br-HB)Cl4]n
(1)
•
DMPO– OH adduct, ethanol (10%, vol/vol) was added into
The addition of DABCO (10 mM), which is commonly
the solution prior to irradiation. The EPR signal of the
1
•Ϫ
used to inhibit O2 signal (32), decreased the DMPO–O2
•
N
H
DMPO– CH(OH)CH3 adduct (
␣ ϭ 15.9 G, ␣ ϭ 23.0 G)
signal but did not make it disappear completely (data not
shown). Moreover, deuterated DMSO had an enhancing ef-
•
was observed (Fig. 7B) (33). This suggests that DMPO–OH
comes from the reaction between DMPO with OH generated
by [Al2(5,8-di-Br-HB)Cl4]n photosensitization.
The addition of catalase (30
duced the intensity of the DMPO–OH signal (Fig. 7E),
which indicated that OH generation by [Al2(5,8-di-Br-
HB)Cl4]n photosensitization was H2O2 relevant. The addition
•
•Ϫ
fect on the DMPO–O2 signal. The results indicated that 1O2
•Ϫ
was involved in the formation of O2 (see later).
Ϫ
1
␮g·mL ) significantly re-
We also studied the dependence of the ESR signal inten-
•
•Ϫ
sity of DMPO–O2 on the concentration of [Al2(5,8-di-Br-
•
•Ϫ
HB)Cl4]n (Fig. 6). It was found that the DMPO–O2 signal
decreased subsequently with the increase of photosensitizer
concentration. When the concentration increased to 50
Ϫ
1
of SOD (25
␮g·mL ) enhanced the signal intensity of
•
DMPO– OH (Fig. 7F). Moreover, the addition of DTPA (5
Ϫ
1
•Ϫ
mg·mL the DMPO–O2 signal disappeared completely.
This is probably because the autoquenching rate of triplet-
state photosensitizer increases with concentration, which in-
mM), which is a well-known chelator of iron preventing its
further reaction with H2O2 (34), inhibited DMPO–OH for-
mation (Fig. 7G). These findings support that the [Al2(5,8-
di-Br-HB)Cl4]n photosensitization can generate OH via the
Fenton–Haber–Weiss reaction after O2 dismutation (see lat-
er). When the concentration of catalase and DTPA increased
to 150 ␮g·mL and 15 mM, respectively, the DMPO– OH
signal could not disappear completely, implying that other
pathways may be involved in the formation of OH.
The addition of DABCO (10 mM) decreased the DMPO–
OH signal significantly (Fig. 8C). When the deuterated buff-
er was used as solvent the stronger signal of DMPO–OD
was observed (Fig. 8D). The spectrum of DMPO–OD is
similar to that of the DMPO– OH adduct because the split-
ting by deuterium is small and is not resolved within the line
width of the signal (35). The results suggest that O2 is in-
•
•Ϫ
dicates that the formation of O2 was certainly from the
•
excited state of [Al2(5,8-di-Br-HB)Cl4]n.
•Ϫ
To provide evidence for the effect of chemical modifica-
•Ϫ
•Ϫ
tions on the O2 production the O2 -generation efficiency
of [Al2(5,8-di-Br-HB)Cl4]n was compared with that of HB
Ϫ
1
•
•Ϫ
(Fig. 5, line D). It can be seen that O2 generation by
•
[Al2(5,8-di-Br-HB)Cl4]n is as efficient as that by HB.
•
•
Photogeneration of hydroxyl radical (OH)
•
•
As shown in Fig. 7(A), the irradiation of an air-saturated
aqueous solution (pH 7.4) containing [Al2(5,8-di-Br-
•
Ϫ1
HB)Cl4]n (0.5 mg·mL ) and DMPO (50 mM) led to a strong
1
quartet spectrum with an intensity ratio of 1:2:2:1. The EPR
•
spectrum was characterized by hyperfine coupling constants
volved in the formation of DMPO– OH.
Under acidic conditions (pH 4.6) irradiation of an air-sat-
␥
of
␣
ϭ ␣H
ϭ
14.9 G, showing that this signal was the
N

Photochemistry and Photobiology, 2001, 74(2) 147
It has been shown that the TEMPO signal generated by
Ϫ
1
[Al2(5,8-di-Br-HB)Cl4]n (0.5 mg·mL ) could be inhibited
completely by DABCO (5 mM). When the concentration of
•Ϫ
DABCO reached 10 mM, the DMPO–O2 signal could not
disappear, confirming that other pathways participated in the
•Ϫ
formation of O2
.
•Ϫ
We have found that the DMPO–O2 signal decreased with
the increase of photosensitizer concentration, which indicat-
•Ϫ
ed that O2 might originate from the excited state of
[Al2(5,8-di-Br-HB)Cl4]n (Eq. 4).
{[Al2(5,8-di-Br-HB)Cl4]n}*
ϩ O2
•Ϫ
•ϩ
→
O2
ϩ
{[Al2(5,8-di-Br-HB)Cl4]n}
(4)
•
1
•Ϫ
Figure 8. The dependence of DMPO– OH signal intensity as a func-
In addition to the formation of O2 and O2 the DMPO–
•
tion of irradiation time using 532 nm light in an air-saturated aque-
ous solution (pH 7.4) containing [Al2(5,8-di-Br-HB)Cl4]n (0.5
mg·mLϪ1) and DMPO (50 mM) (line A). Line B: Same as line A
OH adduct can be generated by [Al2(5,8-di-Br-HB)Cl4]n in
•
aqueous solution (Fig. 7). The formation of OH was super-
oxide-dependent by a rapid dismutation to H2O2 and O2 (Eq.
5). Furthermore, the inhibitory effect of catalase indicated
that the photogeneration of OH by [Al2(5,8-di-Br-HB)Cl4]n
was H2O2-dependent. The addition of SOD stimulated the
formation of the DMPO– OH adduct due to an accelerated
rate of hydrogen peroxide production by superoxide dis-
mutation (Eq. 5). These data indicated that the formation of
OH was involved in the Fenton–Haber–Weiss reaction (Eqs.
6 and 7). It was also found that the formation of OH could
except that was replaced by HB (39 ␮M).
•
urated aqueous solution of [Al2(5,8-di-Br-HB)Cl4]n (0.5
mg·mL ) and DMPO (50 mM) led to the EPR spectrum of
DMPO– OH adduct. The intensity of DMPO– OH adduct
signal did not change much from that under neutral medium
(pH 7.4) (data not shown).
In addition, we compared the efficiency of OH-generation
by [Al2(5,8-di-Br-HB)Cl4]n with that of HB (Fig. 8). The
result showed that the efficiency of OH generation by
[Al2(5,8-di-Br-HB)Cl4]n was approximately 10 times as high
as that by HB.
Ϫ
1
•
•
•
•
•
•
not be inhibited completely by catalase or DTPA; moreover,
1O2 quencher and deuterated solvent had some influence on
the formation of OH. Thus it can be concluded that O2
•
•
1
•
participates in the generation of OH through Eq. 8 (35).
Hϩ or SOD
DISCUSSION
•Ϫ
•Ϫ
•Ϫ
O2
ϩ
ϩ
O2
O2
→
→
→
H2O2
ϩ
O2
O2
˙OH
(5)
(6)
(7)
1
In this study we have demonstrated unequivocally that O2,
ϩ
ϩ
Fe3
Fe2
ϩ
•Ϫ
•
O2 and OH can be generated by irradiation of [Al2(5,8-di-
Br-HB)Cl4]n. Using the formation of TEMPO nitroxide rad-
ical and DPA-bleaching methods it was confirmed that a
type-II mechanism was involved in the photosensitization
Fe2
ϩ
H2O2
1O2
Fe3
ϩ
ϩ OH
ϩ
ϩ
Ϫ
DMPO
→
→
DMPO–1O2
˙OH DMPO–˙OH
1
process (Figs. 3 and 4), i.e. O2 was generated through en-
ergy transfer from the excited state of [Al2(5,8-di-Br-
HB)Cl4]n to ground-state molecular oxygen (Eq. 2).
ϩ
(8)
{[Al2(5,8-di-Br-HB)Cl4]n}*
ϩ O2
CONCLUSIONS
1O2
ϩ
[Al2(5,8-di-Br-HB)Cl4]n
(2)
[Al2(5,8-di-Br-HB)Cl4]n, with excellent water-solubility and
enhanced red absorption, is a novel potential phototherapeu-
tic agent. The EPR and spectrophotometric investigations
→
1
The quantum yield of O2 production was estimated as
0.33 in DMSO with HB as the reference (0.76). The result
demonstrated that the 1O2-generating quantum yield by
[Al2(5,8-di-Br-HB)Cl4]n photosensitization has been reduced
significantly because the 4,9-dihydorxyl-3,10-perylenequi-
nonoid moiety has been changed in [Al2(5,8-di-Br-HB)Cl4]n.
Previous studies indicated that the 4,9-dihydorxyl-3,10-per-
ylenequinonoid moiety was the essential structural require-
•Ϫ
demonstrate that superoxide anion radical (O2 ), especially
•
hydroxyl radical ( OH), can be produced by [Al2(5,8-di-Br-
HB)Cl4]n photosensitization in enhanced yields as compared
with those of HB. Singlet oxygen (1O2) is generated in a
1
ment for the generation of O2 by hypocrellin-like com-
pounds (4).
1
Taking into account the influence of O2 quencher and
•Ϫ
deuterated solvent on the formation of O2 we can assume
1
•Ϫ
that O2 participates in the generation of O2 through Eq. 3.
[Al2(5,8-di-Br-HB)Cl4]n
ϩ
1O2
•Ϫ
•ϩ
→
O2
ϩ
{[Al2(5,8-di-Br-HB)Cl4]n}
(3)
Scheme 1

148 Jianghua Ma et al.
Synthesis of 2-BA-2-DMHA and its phototoxicity to MGC803
cells. J. Photochem. Photobiol. B: Biol. 44, 21–28.
16. He, Y. Y., J. Y. An and L. J. Jiang (2000) Synthesis and EPR
investigations of new aminated hypocrellin derivatives. Free
Radic. Biol. Med. 28, 1642–1651.
17. Jiang, Y., J. Y. An and L. J. Jiang (1993) Photodynamic action
of brominated hypocrellin B. Chin. Sci. Bull. 38, 797–800.
18. Diwu, Z. J., C. L. Zhang and J. W. Lown (1992) Photosensiti-
zation with anticancer agents. 13. The production of singlet ox-
ygen by halogenated and metal ion-chelated perylenequinone.
J. Photochem. Photobiol. A: Chem. 66, 99–112.
decreased quantum yield due to the destruction of the 4,9-
dihydorxyl-3,10-perylenequinonoid moiety in [Al2(5,8-di-
Br-HB)Cl4]n. The generation pathways of these active oxy-
gen species from [Al2(5,8-di-Br-HB)Cl4]n are summarized in
Scheme 1. It is inferred that photosensitization by [Al2(5,8-
di-Br-HB)Cl4]n involves both type-I and type-II mechanisms,
in which the type-I reaction is greatly enhanced, although
type-II reaction is reduced significantly as compared with
HB.
19. An, J. Y., F. Tian, Y. Z. Hu and L. J. Jiang (1993) Complexation
of hypocrellins and bivalent metals and one-electron reduction
of the complexes with electron donor. Chin. J. Chem. 11, 532–
539.
Acknowledgements This research was supported by the 95 key pro-
ject of academia Sinica (KJ952-J1-523) and the National Science
Foundation of China (39830090).
20. Hu, Y. Z., J. Y. An and L. J. Jiang (1994) Studies on the che-
lation of hypocrellin A with aluminium ion and the photody-
namic action of the resulting complex. J. Photochem. Photobiol.
B: Biol. 22, 219–227.
21. Moan, J., Q. Peng, J. F. Evensen, K. Berg, A. Western and C.
Rimington (1987) Photosensitizing efficiencies, tumor and cel-
lular uptake of different photosensitizing drugs relevant for pho-
todynamic therapy of cancer. Photochem. Photobiol. 46, 713–
721.
22. Oenbrink, G., P. Ju¨rgenlimke and D. Gabel (1988) Accumula-
tion of porphyrins in cells: influence of hydrophobicity aggre-
gation and protein binding. Photochem. Photobiol. 48, 451–456.
23. Zhang, H. Y., Z. Y. Zhang, Q. Wan, D. H. Wang and M. F. Li
(1996) Protective effects of nonionic micelle on ground and
excited states of hypocrellin B and its brominated products.
Photogr. Sci. Photochem. 14, 138–143.
24. Pang, S. Z., J. F. Qing, J. C. Yue and J. Y. An (1994) Photo-
sensitizing damage to morphology of Hela cells by bromide of
hypocrellin B (5-Br-HB). Acta Biophys. Sin. 10, 651–656.
25. Zhao, H. X., Y. M. Xu and Z. Y. Zhang (1998) Raman spec-
troscopic study of DNA photodamage sensitized by hypocrellin
B and 5-brominated-hypocrellin B. Chin. Sci. Bull. 43, 1128–
1135.
26. Ma, J. H., J. Q. Zhao and L. J. Jiang (2000) Synthesis of a novel
hypocrellin B derivative and its photogeneration of semiquinone
anion radical. (In press)
27. Zhao, K. H. and L. J. Jiang (1989) Structural conversion of
hypocrellin A alkaline and neutral solution. Chin. J. Org. Chem.
9, 252–254.
28. Diwu, Z. J. and J. W. Lown (1992) Photosensitization with an-
ticancer agents. 12. Perylenequinonoid pigments, a novel type
of singlet oxygen sensitizer. J. Photochem. Photobiol. A: Chem.
64, 273–287.
29. Lion, Y., M. Delmelle and A. van de Vorst (1976) New method
of detecting singlet oxygen production. Nature 263, 442–443.
30. Kearns, D. R. (1979) Organic chemistry. A series of mono-
graphs. In Singlet Oxygen (Edited by H. H. Wasserman and R.
W. Murray), 40, 139–171. Academic Press, New York.
31. Lang, K., D. M. Wagnerova˘, P. Stopka and W. Damerau (1992)
Reduction of dioxygen to superoxide photosensitized by anthra-
quinone-2-sulphonate. J. Photochem. Photobiol. A: Chem. 67,
187–195.
REFERENCES
1. Dougherty, T. J. (1993) Photodynamic therapy. Photochem.
Photobiol. 58, 895–900.
2. Gomer, C. J. (1991) Preclinical examination of first and second-
generation photosensitizers used in photodynamic therapy. Pho-
tochem. Photobiol. 54, 1093–1107.
3. Grossweiner, L. I. (1994) The Science of Phototherapy. CRC
Press, Boca Raton. [Chap. 8]
4. Diwu, Z. J. (1995) Novel therapeutic and diagnostic applications
of hypocrellins and hypericins. Photochem. Photobiol. 61, 529–
539.
5. Jiang, L. J. (1990) The structures, properties, photochemical re-
actions and reaction mechanisms of hypocrellin (I)—the struc-
tures and properties of hypocrellin. Chin. Sci. Bull. 21, 1608–
1616.
6. Jiang, L. J. (1990) The structures, properties, photochemical re-
actions and reaction mechanisms of hypocrellin (II)—reactions
of hypocrellin. Chin. Sci. Bull. 21, 1681–1690.
7. Hu, Y. Z., J. Y. An and L. J. Jiang (1993) Studies of the sul-
fonated of hypocrellin A and the photodynamic actions of the
product. J. Photochem. Photobiol. B: Biol. 17, 195–201.
8. Hu, Y. Z. and L. J. Jiang (1996) Generation of semiquinone
radical anion and reactive oxygen (1O2, O2 and OH) during
the photosensitization of a water-soluble perylenequinone deriv-
ative. J. Photochem. Photobiol. B: Biol. 33, 51–59.
•Ϫ
•
9. He, Y. Y., J. Y. An, W. Zou and L. J. Jiang (1998) Photore-
actions of hypocrellin B with thiol compounds. J. Photochem.
Photobiol. B: Biol. 44, 45–52.
10. He, Y. Y., J. Y. An and L. J. Jiang (1999) Glycoconjugated
hypocrellin: synthesis of [(␤-D-glucosyl)ethylthiyl]hypocrellins
and photosensitized generation of singlet oxygen. Biochim. Bio-
phys. Acta 1472, 232–239.
11. Diwu, Z. J., C. L. Zhang and J. W. Lown (1993) Photosensiti-
zation with anticancer agents. 18. Perylenequinonoid pigments
as potential photodynamic therapeutic agents: preparation and
photodynamic properties of amino-substituted hypocrellin deriv-
atives. Anti-Cancer Drug Des. 8, 129–143.
12. He, Y. Y., J. Y. An and L. J. Jiang (1999) Photodynamic action
of a water-soluble hypocrellin derivative with enhanced absorp-
tivity in the phototherapeutic window II. Photosensitized gen-
eration of active oxygen species. J. Photochem. Photobiol. A:
Chem. 120, 191–199.
32. Lissi, E. A., M. V. Encinas, E. Lemp and M. A. Rubio (1993)
1
Singlet oxygen O2
(
⌬
g) bimolecular processes. Solvent and
compartmentalization effects. Chem. Rev. 93, 699–723.
33. Finkelstein, E., G. M. Rosen and E. J. Rauckman (1980) Spin
trapping of superoxide and hydroxyl radical: practical aspects.
Arch. Biochem. Biophys. 200, 1–16.
13. He, Y. Y., J. Y. An and L. J. Jiang (1998) EPR and spectro-
•Ϫ
photometric studies on free radicals (O2•Ϫ, Cysa-HB ) and sin-
glet oxygen (1O2) generated by irradiation of cysteamine sub-
stituted hypocrellin B. Int. J. Radiat. Biol. 74, 647–654.
34. Halliwell, B. and J. M. C. Gutteridge (1986) Oxygen free rad-
icals and iron in relation to biology and medicine: some prob-
lems and concepts. Arch. Biochem. Biophys. 246, 501–514.
35. Bilski, P., K. Reszka, M. Bilska and C. F. Chignell (1996) Ox-
idation of the spin trap 5,5-dimethyl-1-pyrroline N-oxide by sin-
glet oxygen in aqueous solution. J. Am. Chem. Soc. 118, 1330–
1338.
14. Song, Y. Z., J. Y. An and L. J. Jiang (1999) ESR evidence of
•Ϫ
the photogeneration of free radicals (GDHB , O2•Ϫ) and singlet
oxygen (1O2) by 15-deacetyl-13-glycine-substituted hypocrellin
B. Biochim. Biophys. Acta 1472, 307–313.
15. Zhang, W. G., M. Weng, S. Z. Pang, M. H. Zhang, H. Y. Yang,
H. X. Zhao and Z. Y. Zhang (1998) A novel photosensitizer, 2-
butylamino-2-demethoxy-hypocrellin A (2-BA-2-DMHA) 1.