Effects of noncovalently bound quinones on the ground and triplet states of zinc chlorins in solution and bound to de novo synthesized peptides

Article abstract of DOI:10.1039/b612056c

The Q_y absorption band of two chlorophyll derivatives, zinc chlorin e6 (ZnCe6) and zinc pheophorbide a (ZnPheida), in aqueous solution is bathochromically shifted on addition of quinones, e.g., 1,4-benzoquinone (BQ), with a corresponding shift of the fluorescence band. This is due to a complex formation of zinc chlorins induced by BQs and subsequent rearrangement. The time-resolved absorption spectra after laser pulse excitation show triplet quenching of the pigments by BQ and other quinones via electron transfer. The effects of electron transfer to noncovalently bound BQs were also studied with de novo synthesized peptides, into which ZnCe6 and ZnPheida were incorporated as model systems for the primary steps of photosynthetic reaction centers. Whereas the photophysical properties are similar to those of the unbound zinc chlorins, no BQ-mediated complex formation was observed. the Owner Societies.

Full text of DOI:10.1039/b612056c

View Article Online / Journal Homepage / Table of Contents for this issue
PAPER
www.rsc.org/pccp | Physical Chemistry Chemical Physics
Effects of noncovalently bound quinones on the ground and triplet states
of zinc chlorins in solution and bound to de novo synthesized peptidesw
Anke Mennenga, Wolfgang Gartner, Wolfgang Lubitz and Helmut Gorner*
¨ ¨
Received 21st August 2006, Accepted 12th October 2006
First published as an Advance Article on the web 30th October 2006
DOI: 10.1039/b612056c
The Q absorption band of two chlorophyll derivatives, zinc chlorin e6 (ZnCe6) and zinc
y
pheophorbide a (ZnPheida), in aqueous solution is bathochromically shifted on addition of
quinones, e.g., 1,4-benzoquinone (BQ), with a corresponding shift of the fluorescence band. This
is due to a complex formation of zinc chlorins induced by BQs and subsequent rearrangement.
The time-resolved absorption spectra after laser pulse excitation show triplet quenching of the
pigments by BQ and other quinones via electron transfer. The effects of electron transfer to
noncovalently bound BQs were also studied with de novo synthesized peptides, into which ZnCe6
and ZnPheida were incorporated as model systems for the primary steps of photosynthetic
reaction centers. Whereas the photophysical properties are similar to those of the unbound zinc
chlorins, no BQ-mediated complex formation was observed.
1
Introduction
Chlorin e6 (Ce6) remains monomeric when an organic
2
0
solvent is replaced by water, whereas pheophorbide a
A major contribution to the remarkably high efficiency of
photosynthesis is light-harvesting by the (bacterio)chlorophyll
antenna system that transfers the light energy to the so-called
reaction center before the electron is passed via chlorophyll
derivatives to quinones. During these processes in the reaction
center, chlorophylls serve as electron donors and as acceptors
after electronic excitation. The initial photoproducts are radi-
cal ion pairs. Many of the primary processes in photosynthesis
23
Pheida) exists in a monomer–dimer equilibrium. The low
(
quantum yield of singlet molecular oxygen production (FD)
for Pheida in aqueous solution, with respect to the value for
the monomer in methanol, has been attributed to a low F
ISC
value due to the dimerization in water. For a detailed
16
analysis of the ground state interactions and electron transfer
processes of the excited chlorins to noncovalently attached
quinones, the zinc containing derivatives of Pheida and Ce6
were used in this work. Routinely, zinc complexes are prefer-
entially used since no changes in the spectroscopic properties
are to be expected by the exchange of magnesium into zinc; in
addition, the zinc complexes show, amongst other experimen-
tal advantages, also a higher stability. The solubility of the
pigments is of great importance when mimicking the primary
photochemistry of pheophytin/quinone-type reaction centers.
Therefore, the synthetic zinc chlorins in aqueous solution
1
have been studied in great detail, but still a number of open
questions remain unsolved, in particular many aspects of the
photophysical reactions of chlorophyll molecules. A major
drawback for these studies is the strong electronic interaction
of the many chlorophyll molecules in the antenna that makes it
practically impossible to address a single entity in this large
aggregated complex.
Since the photoprocesses of supramolecular aggregates in-
volving chlorin moieties play a key role for photosynthesis in
green plants and photosynthetic bacteria, many efforts have
been made to describe the properties of chlorophyll assem-
6
should include hydrophilic groups.
Besides their photophysical behaviour, the interaction of
chlorophylls with other molecules that serve as electron ac-
ceptors has been intensively studied. Scheme 1 shows the
energy diagram of a chlorin (C) quinone (Q) pair. The excita-
tion of an electron from the ground state of the chlorin as
1
–6
blies.
The self-assembly of synthetic zinc chlorins in non-
polar or aqueous microheterogeneous media to an artificial
supramolecular light-harvesting device has been intensively
investigated, see ref. 6 for a review. For chlorophyll a
7
–14
15–27
(
Chla)
and related pigments
many photophysical and
photochemical properties have been reported. Furthermore,
the intermediates in the photochemical reactions of Chla are
well characterized. It is known that the quantum yield of
intersystem crossing (FISC) of aggregated Chla is small, and
monomeric Chla in organic solvents shows both fluorescence
1
1,12
and ISC in substantial quantum yields.
Max-Planck-Institut fu¨r Bioanorganische Chemie, D-45413 Mu¨lheim
an der Ruhr, Germany. E-mail: goerner@mpi-muelheim.mpg.de; Fax:
+49(0)2083063951; Tel: +49(0)2083063593
w Dedicated to Professor Kurt Schaffner on the occasion of his 75th
birthday.
Scheme 1
5
444 | Phys. Chem. Chem. Phys., 2006, 8, 5444–5453
This journal is ꢀc the Owner Societies 2006

View Article Online
1
electron donor leads to the first excited state ( C*), which is
A useful approach is the investigation of simple model
systems with the aim of obtaining information about general
mechanisms leading to radical pairs. Incorporation of metal-
lochlorins into de novo synthesized proteins provides the
possibility to study the photochemical reaction of monomeric
chlorins, and upon addition of an electron acceptor, to mimic
electron transfer proteins found in photosynthesis and res-
piration.
characterized by its electronic energy and by different vibra-
tional states of the molecule. The excited chlorin can revert to
the ground state in several ways, through internal conversion
or fluorescence. Another pathway is ISC leading to the lowest
3
triplet state ( C*). The quinone as electron acceptor quenches
1
3
the C* or C* states by formation of the correlated radical
1
ꢁ
+
ꢁ
À
3
ꢁ
+
ꢁ
À
pairs: (C ..Q ) and (C ..Q ), where the former is only
involved at high BQ concentrations. A direct electron transfer
3
from the C* state to a quinone can be observed by transient
Much effort has been made to mimic the function of
34–39
reaction centers in peptide models.
Related to the present
absorption spectroscopy, whereby the rate constant for the
+
triplet decay and the appearance of the radical cation C
work, de novo designed four-helix bundle proteins with me-
talloporphyrin cofactors were characterized and anthraqui-
none-2-sulfonate was used for photoinduced electron transfer
ꢁ
should be the same. In particular the Chla/BQ system was
3
5a
studied in non-aqueous solution by flash photolysis and
–12
electron spin resonance spectroscopy.
from peptide-bound zinc protoporphyrin IX. Photoinduced
7
electron transfer from ZnCe6 in aqueous solution to free
Photoprocesses of other tetrapyrroles and related chromo-
phores with quinones in solution have been examined in great
phenyl-p-benzoquinone (phBQ) and covalently bound UQ
3
0
7,38
was observed by EPR.
The reaction from excited ZnCe6 to
2
8–33
detail.
These investigations have revealed a broad varia-
phBQ was reported to be faster and to have a higher yield in
3
6
tion of interactions and reactivities: preferentially the system
the presence of a peptide maquette. When UQ
attached via a cysteine to a protein as ZnCe6-cytochrome b562
electron transfer takes place. Static quenching was also sug-
0
is covalently
of zinc meso-tetrakis(sulfonatophenyl)porphyrin and 2,3-di-
,
0
methoxy-5-methyl-1,4-benzoquinone (UQ ) undergoes elec-
3
7
tron transfer in the triplet state and the excited singlet
f
gested based on the shorter fluorescence lifetime (t ). This
2
9
state. Both the excited singlet and triplet states are also
raises the question of the specific features of synthetic proteins
and their contributions to the reactivity of the incorporated
cofactors.
quenched by BQ for tetraphenylporphyrin in dichloro-
3
0
methane. In contrast, any detailed knowledge concerning
photoinduced intermolecular electron transfer from chlorins
in aqueous solution is still scarce.
In this study, we investigated the ground state interactions
of ZnCe6 and ZnPheida and some related pigments with
Fig. 1 Schematic representation of the structures of the peptides (a) M1, (b) M2 and (c) ZnCe6, (d) BQ, (e) phBQ and (f) ZnPheida.
This journal is ꢀc the Owner Societies 2006
Phys. Chem. Chem. Phys., 2006, 8, 5444–5453 | 5445

View Article Online
quinones in solution (see Fig. 1) and after binding to two
peptide maquettes (M1 and M2) as well as their triplet state
reactions. There are two major questions to be asked: do
synthetic proteins affect the electron transfer from the zinc
chlorin excited states to BQs? What is the nature of the
complex formed between zinc chlorin and BQ in aqueous
solution? We have addressed these questions by measuring
spectral and kinetic data, using BQ, phBQ and methyl violo-
2.2 Peptide synthesis
Maquette M1: The concept for this peptide is based on a
prototype peptide design known as ‘‘H10H24’’ which is mod-
elled from known sequences of the cytochrome bc
1
respiratory
complexes. It was originally synthesized to investigate heme
8
3
binding peptides. The sequence of the a-helical peptide is
identical to that used by Fahnenschmidt et al.: C-G-G-G-
35b
E-L-W-K-L-H-E-E-L-L-K-K-F-E-E-L-L-K-L-H-E-E-R-K-K-
K-L. The a-helical peptide was synthesized as a 31-amino
acid stretch that was subsequently homodimerized through
the N-terminal cysteine in 0.05 M ammonium acetate
at pH 7.6 with 2% DMSO to form a disulfide linked 62-mer
peptide.
2
+
gen (MV ) as electron acceptors. We show that the quench-
ing rate constant of zinc chlorins by BQs depends on several
factors, albeit the mechanism is the same either in solution or
bound into the four-helix bundle.
Maquette M2: Design of this peptide is based on the work
39
of Sharp et al. with the following sequence: L-K-K-L-R-E-E-
2
Experimental
2
+
The compounds phBQ (Acros), MV (Aldrich), Ce6 (Fron-
tier Scientific) and the solvents (Merck) were used as commer-
cially received or purified by sublimation (BQ) or distillation
A-L-K-L-L-E-E-F-K-K-L-L-E-E-H-L-K-W-L-E-G-G-G-G-
G-G-G-G-E-L-L-K-L-H-E-E-L-L-K-K-C-E-E-L-L-K-L-A-
E-E-R-L-K-K-L. The histidine as the axial ligand for the
cofactor is located in the centre of the hydrophobic core
inside the formed four-helix bundle maquettes. The ligand
position was also verified by 2D-Pulse-ESEEM spectro-
(
methanol). Water from a Millipore apparatus was applied
throughout. All measurements were performed under dimmed
light conditions at room temperature unless otherwise
indicated.
4
0
scopy.
The peptides M1 and M2 were synthesized on an automated
peptide synthesizer (Advanced ChemTech, 348O) using Fmoc-
protected amino acids (Novabiochem and Iris Biotech) on a
PAL-PEG-PS resin (Applied Biosystems). N-terminal acetyla-
tion of the peptides was performed with 5% acetic anhydride
in dimethylformamide prior to cleavage. Purification of the
crude peptides was performed by RP-HPLC (Vydac Protein &
Peptide), C18 column, 250 Â 22 mm, aqueous/acetonitrile
2
.1 Pigment preparation
Chla was isolated from Spirulina platensis by methanol ex-
traction and purified by reverse phase high performance liquid
chromatography (RP-HPLC) using a Nucleosil-5-C18 column
(
Macherey & Nagel), 125 Â 8 mm. Methanol was used as
À1
elution solvent (flow rate 1.5 ml min ). Pheida was prepared
by treatment of 0.14 mol Chla with pure trifluoroacetic acid
for 30 min. After evaporation of the acid, Pheida was purified
gradients containing 0.1% trifluoroacetic acid, flow rate
À1
1
0 ml min . The molecular mass of the purified peptides
À1
by RP-HPLC (methanol, flow rate: 1.5 ml min ).
was verified by MALDI-TOF-MS, M1: calcd 7607 Da, found
609 Da; M2: calcd 7179 Da, found 7181 Da. A yield of 8% of
the maquette M2 was determined.
ZnPheida was synthesized from Pheida by dissolving the
latter in pure acetic acid and addition of a 250-fold molar
excess of zinc acetate, 20-fold molar excess of sodium acetate,
and catalytic amounts (10–50 mg) of sodium ascorbate. This
reaction mixture was stirred for 1 h. The product was purified
by RP-HPLC, Nucleosil-5-C18, methanol/0.5 M ammonium
7
2
.3 Metallochlorin incorporation into peptides
The synthetic metallopeptides were prepared by slow addition
of an excess of the metallochlorin solution, ZnCe6 in glycine
buffer, ZnPheida dissolved in acetone, to the peptide with
gentle stirring for 30 min at 4 1C in the dark. The pig-
ment–peptide complex was separated from unbound pigment
by using a Pharmacia PD-10 column, which was equilibrated
with 50 mM potassium phosphate buffer (100 mM KCl) at
pH 7.3.
À1
acetate (8 : 1, v/v), flow rate 1.5 ml min , followed by
desalination with diethyl ether and water. The solution was
dried over sodium sulfate and the organic phase removed by
evaporation. Since the pigment is insoluble in water, it was
dissolved in acetone prior to preparation of aqueous solutions
for the spectroscopic measurements. The amount of acetone in
each experiment did not exceed 6%.
ZnCe6 was synthesized from Ce6 by addition of molar
equivalents of zinc acetate to a buffered aqueous solution
2
.4 Solution molecular weight determination
¨
(
25 mM glycine, 50 mM NaCl, pH 10) and stirring for
0 min at 4 1C in argon atmosphere. The reaction was
monitored by UV-Vis spectroscopy, probing a red shift of
the Soret band (401 - 412 nm) and a blue shift of the Q
A FPLC system (AKTA basic) equipped with a Superdex 75
HR10/30 column (all Amersham Biosciences) equilibrated at a
3
À1
flow rate of 0.5 ml min with 150 mM NaCl and 50 mM
y
phosphate buffer (pH 7.3) was used for gel-filtration chroma-
tography with detection wavelengths of 260 and 280 nm.
Column calibration was performed with a low molecular
weight gel filtration calibration kit (Amersham Biosciences)
and aprotinin (Roth). The different zinc chlorin peptide scaf-
folds elute with retention times consistent with a dimer of the
complex, i.e., four-helix bundles, with less than 8% of the
material comprising higher oligomerization states.
band (655 - 633 nm). In addition, ZnCe6 was separated from
metal free Ce6 by RP-HPLC, Kromasil-5-C18 (MZ-Analy-
sentechnik), 125 Â 4.6 mm, 50 mM triethylammonium acetate
À1
(
pH 8)/methanol (1 : 2, v/v, flow rate 0.8 ml min ) with
a purity of 94.5%. The pigments were stored under argon at
À80 1C until use. All preparations were performed under
green light.
5
446 | Phys. Chem. Chem. Phys., 2006, 8, 5444–5453
This journal is ꢀc the Owner Societies 2006

View Article Online
Fig. 2 Absorption and fluorescence spectra (insets) of (a) 8 mM ZnCe6 and (b) 9 mM ZnPheida in 0.05 M phosphate buffer at 24 1C, pH 7.3,
(dotted line) and incorporated in the peptide M2 (solid line).
2
.5 UV-Vis spectroscopy
up to 10 mJ, lexc = 410–710 nm. The set-up contains a 150-W
Xenon lamp and a home-made pulser, a monochromator with
stepper motor, a photomultiplier (Hamamatsu R955) and two
transient digitizers (Tektronix, 7912AD and 390AD). Chla in
The steady-state absorption spectra were recorded in 1 cm
cuvettes on spectrophotometers (either Varian Cary 5 or HP
8
453). The molar absorption coefficients in aqueous solution
4
1
7
4
À1
À1
23
ethanol was used as reference with FISC = 0.5. The experi-
mental error of t , kox, k
are e6 = 4.5 Â 10 M cm for Pheida,
67
e
= 4.7 Â 10
663
4
À1
À1
À1 À1
T
q
is Æ10%, that of FISC is Æ15% for
M
cm for ZnPheida and e633 = 4.0 Â 10 M cm for
values larger than 0.2 and Æ30% otherwise. The solutions
were freshly prepared and immediately photolyzed, especially
at pH 10 and for zinc chlorin/quinone systems at pH 7.3.
When complex formation is involved, the T–T absorption at
ZnCe6. The pigment concentrations were 5–10 mM in 0.01–50
mM phosphate buffer (100 mM KCl, pH 7.3) or in 25 mM
glycine buffer (50 mM NaCl, pH 10).
4
50–600 nm is lower and the bleaching is broader. The low
2
.6 Fluorescence spectroscopy
temperature measurements were carried out in aqueous buffer
and glycerol (2 : 3, v/v) using cooled nitrogen.
The fluorescence spectra were recorded on a Varian Cary-
Eclipse spectrofluorimeter. The quantum yields of fluorescence
were obtained from the emission areas with optically matched
f
solutions (lexc = 418 nm) using Chla in acetone with F =
0
3
Results
7
.24 as reference. The fluorescence decay kinetics were deter-
mined by a fluorimeter (Edinburgh Instr. F900) with a time
3.1 Steady state absorption
resolution of 0.2 ns, lexc = 385 nm. The experimental error of
The incorporation of the zinc chlorins into the four-helix
bundles and the steady state interaction of zinc chlorins (in
solution and in peptide-bound form) with quinones was
investigated by UV-VIS absorption spectroscopy. The max-
ima of the ground state absorption spectrum of ZnCe6 in
aqueous solution in the presence of phosphate buffer (pH 7.3)
t
f
is Æ0.2 ns, that of F is Æ10% for values larger than 0.1 and
f
Æ20% otherwise. The absorption and fluorescence measure-
ments refer to air-saturated solutions unless otherwise indi-
cated.
2
.7 Transient absorption spectroscopy
y
are at 412 and at 633 nm (Soret and Q band, respectively).
The time-resolved measurements were carried out with a
tunable laser (Opotek, Vibrant 355II) pumped from a Nd-
YAG and using an OPO, pulse width: 5 ns, energy per pulse:
The bands are red-shifted to 421 and 642 nm for ZnCe6
incorporated in maquette M1 and 419 and 640 nm for ZnCe6
bound to maquette M2 (Fig. 2a). Analogous shifts were found
a
Table 1 Absorption and fluorescence properties of chlorins
b
c
d
Compound
Quinone
l/nm
l
f
/nm
F
f
f
t /ns
Ce6
ZnCe6
None or BQ
None
BQ
401, 655
412, 633
412, 673
412, 673
421, 642
419, 640
662
643
684
684
647
647
669
0.32
2.1
phBQ
ZnCe6-M1
ZnCe6-M2
ZnPheida
None or BQ
None or BQ
None
BQ
phBQ
0.29
0.29
0.03
2.2
2.2
3.4
e
425 , 663
425 , 740
e
e
425 , 738
ZnPheida-M1
ZnPheida-M2
a
None or BQ
None or BQ
437, 669
433, 668
674
677
0.02
0.02
c
3.1
3.1
b
d
In air-saturated aqueous solution at 24 1C and 0.01 M phosphate buffer, pH 7.3. Typical concentration 10–200 mM.
lexc = 420 nm. Values
e
refer to the main component (70–90%); same values in argon-saturated solution. Broad signal due to aggregation.
This journal is ꢀc the Owner Societies 2006
Phys. Chem. Chem. Phys., 2006, 8, 5444–5453 | 5447

View Article Online
comes from the BQ concentration on the kinetics of the first
step (Fig. 4).
Most interestingly, virtually no spectral changes were regis-
tered for the metal-free chlorins, Ce6 and Pheida, under the
same experimental conditions. The course of the reaction
changes drastically at alkaline pH (B10). Under these condi-
3
6
tions, applied also by Razeghifard and Wydrzynski on the
same system, a competing side reaction interferes with the
previously described quinone-mediated complex formation:
the added quinone rapidly converts into the hydroquinone—
this process being observable spectroscopically—and prevents
complex formation. This thermal reaction can be stopped
when air is removed, and the complex with identical spectral
properties as described above was observed, indicating that the
chlorin/BQ solution requires a less alkaline environment or the
absence of oxygen.
Fig. 3 Absorption spectra of 4 mM ZnCe6 in 0.01 M phosphate
buffer at 24 1C, pH 7.3, directly upon addition of 20 mM BQ (solid
line) and after 200 and 600 s (dashed and dotted line); inset: corre-
sponding fluorescence spectra (lexc = 418 nm) prior to mixing with 20
mM BQ and after 150 and 1000 s (solid, dashed and dotted lines),
respectively.
For ZnPheida in phosphate buffer at pH 7.3 a two-step
pattern was also found and an even larger shift from 663 to
7
40 nm upon addition of a forty-fold excess of BQ (not shown)
with a decrease of the Q
zinc chlorins the total Q
y
band at 663 nm. For BQ and both
absorbance decreases initially and
for ZnPheida in solution and ZnPheida bound to the peptides
M1 and M2 (Fig. 2b and Table 1), e.g., 663 to 669/668 nm,
respectively. These effects were taken as evidence for the
ligation of the pigment to the peptide (also according to
y
reappears at a later time and at a longer wavelength. The
pronounced bathochromic shifts are reminiscent to coordina-
1
3
tion of the metal by water.
3
4–39
literature data),
which is assumed to form a four-helix
In order to investigate whether a strong ligation to the zinc
chlorins affects the shifts of the absorption bands, imidazole as
4
a commonly used ligand for porphyrins was applied. Upon
bundle as indicated by gel-filtration.
1
Addition of a five-fold excess of BQ and stirring for a short-
time leads to a decrease of the absorption of the Q band of
y
addition of imidazole to ZnPheida, a new absorption band at
ZnCe6 and creates an additional band at 673 nm with ‘‘quasi’’-
6
85 nm due to ligation of the imidazole to the ZnPheida is
isosbestic points at 422 and 646 nm (Fig. 3). The steady state
detectable, whereas the band at 663 nm decreases (not shown).
The smaller shift with respect to those of quinones indicates a
different type of interaction. Finally, it should be pointed out
that the shifts observed upon addition of a quinone are in
contrast to the results found for the four synthetic metallopep-
tides (ZnCe6-M1, ZnCe6-M2, ZnPheida-M1, ZnPheida-M2),
for which no spectral changes on addition of BQ were
observed under the same experimental conditions (Table 1).
spectrum is fully developed within 1 h. The decrease of the Q
y
band absorption is faster when the BQ concentration is
increased, see kinetics in Fig. 4a and b for a five- and a
forty-fold excess of BQ, respectively. The five-fold excess of
À1
BQ leads to a rate constant of decay (k
À
) of k
and to a rate constant of grow-in (k ) of k
À
= 0.008 s
À1
= 0.0001 s of
+
+
the absorption bands, whereas a forty-fold excess of BQ gave
À1
À1
k = 0.03 s and k+ = 0.002 s . A potential influence of the
À
buffer components (suggested salt effect) was tested by repeat-
ing the experiment in distilled water, which gave rise to the
same spectral features, although the reaction rate is larger
under buffered conditions. Each of the parameters pH
3
.2 Fluorescence spectroscopy
Fluorescence as well as steady-state absorption spectroscopy,
yields valuable information on the ligation of the zinc chlorins
into the four-helix bundle peptides and the interaction of the
quinone to the pigments. The fluorescence spectra of ZnCe6
and of ZnPheida in aqueous solution in the presence of
(
between 6 and 8), concentration of zinc chlorin and buffer
influences the reaction. The contribution of these parameters
to the kinetics was not separated. The only significant effect
Fig. 4 Kinetic traces of 4 mM ZnCe6 in 0.01 M phosphate buffer at 24 1C, pH 7.3, upon addition of a (a) low BQ- (20 mM) and (b) high
BQ-concentration (170 mM) at characteristic wavelengths of 410 (&), 630 (K), 675 (D) and 440 (.) nm.
5
448 | Phys. Chem. Chem. Phys., 2006, 8, 5444–5453
This journal is ꢀc the Owner Societies 2006

View Article Online
0
.1–10 mM phosphate buffer at pH 7.3 have emission maxima
at l = 643 and 669 nm, respectively (Fig. 2a and b, insets). The
f
fluorescence excitation spectra are in accord with the absorp-
tion data. The synthetic metallopeptides show emission max-
ima at l = 647 nm for ZnCe6, bound to maquette M1 as well
f
as to maquette M2 and 674/677 nm for ZnPheida-M1/-M2
systems, respectively. These bathochromic shifts of the fluor-
escence bands confirm the ligation of the pigments into the de
novo synthesized peptides.
f
The fluorescence quantum yield is F = 0.29–0.32 for
ZnCe6 in the absence or presence of M1/M2 and is much
smaller for ZnPheida and the ZnPheida-M1/-M2 systems.
This result provides evidence of the presence of monomeric
ZnCe6, whereas ZnPheida aggregates in aqueous solution.
The major fluorescence decay component corresponds to
t = 2–3 ns (Table 1). Virtually no change in t was observed
f
f
Fig. 5 Transient absorption spectra of 5 mM ZnCe6 in argon-
saturated 0.01 M phosphate buffer at 24 1C, pH 7.3, (a), (b) solid
line: at the end of the 420 nm pulse; (a) no additive (broken line: after
in the absence or presence of air as well as upon addition of ca.
.05 mM BQ, indicating no fluorescence quenching by the
quinone.
In accordance with the results of the absorption spectro-
0
1
ms) and (b) in the presence of BQ (10 mM, broken line: after 0.1 ms);
insets: kinetic traces for 400 and 460 nm as indicated. Note that the
results refer to conditions, where the proposed complex was not
formed in measurable yield.
scopy, the intensity of the emission maximum of ZnCe6
decreases in the presence of BQ and a new band at 684 nm
appears (Fig. 3). The BQ-induced steady state spectrum is not
specific for BQ, as we also found similar effects with phBQ.
The fluorescence emission spectra of either the metal-free
chlorins or the synthetic metallopeptides are essentially un-
changed upon addition of BQ, this is similar to the results
from absorption spectroscopy. It is noteworthy that the
ground state absorption and fluorescence spectra of chloro-
phylls are as yet not known to be affected by BQs.
1
0 ms (ethanol). For the hydrophobic compound Pheida in
ethanol t = 0.3 ms was determined. It should also be noted
T
that the triplet state of Chla is not observable in aqueous
14
solution due to aggregation, whereas it can be detected for
the chlorins used here. It can thus be concluded that the
hydrophobic ZnPheida component is not completely present
in aggregated form, but adopts a monomer–dimer equilibrium
2
3
as found for Pheida in aqueous solution.
The incorporation of the zinc chlorins into the maquettes
extended the triplet lifetime significantly. For the ZnCe6-M1
system in argon- and air-saturated phosphate buffer at pH 7.3,
3
.3 Triplet state
In addition to steady state measurements, transient absorption
spectroscopy was performed. It provides the possibility to
investigate the excited states of the pigments in a time-resolved
manner. The triplet state of the pigments can be characterized
and quenching processes with quinones, indicating electron
transfer and formation of radical pairs, can be investigated.
The results refer to 0–2 min. after mixing, i.e., under condi-
tions where the complex between a zinc chlorin and BQ is
formed to less than 10%. The transient spectrum of ZnCe6 in
phosphate buffer shows two bleaching signals at 400 and 635
nm which refer to the bleached ground state absorption bands
and transient absorption everywhere else (Fig. 5a). The spectra
of ZnPheida (Fig. 6a), the peptide-bound pigments (Fig. 7a
and 8a), and Ce6 show identical features upon excitation
except for the bleaching bands which are in accordance with
their absorption maxima. The transient species appears within
the duration of the laser pulse and decays by first-order
kinetics under argon and at low laser intensities. The lifetime
t = 1 ms and 50 ms, respectively. The values for the other
T
three metallopeptides are of comparable magnitude to that of
T
(t ) decreases in air and yields a rate constant for quenching
9
À1 À1
s (Table 2). In
by oxygen of kox = (0.1–1.5) Â 10 M
water, kox is lower and t is longer than in a water-free
T
Fig. 6 Transient absorption spectra of 5 mM ZnPheida in argon-
saturated 0.01 M phosphate buffer at 24 1C, pH 7.3, solid line: at the
end of the 420 nm pulse, broken line: after 0.1 m; in the presence of (a)
environment. The transient species that is formed during the
pulse is assigned to the lowest triplet state of the chlorin. This
behaviour is similar to that of the monomeric triplet state of
Chla in, e.g., acetone or ethanol, where the bleaching maxima
2+
BQ (10 mM, no complex formation) and (b) MV
kinetic traces at (a) 410 and 450 nm and (b) 370, 400, and 480 nm as
(10 mM); insets:
are at 420 and 635 nm and t = 30 Æ 3 ms (acetone) and 90 Æ
indicated.
T
This journal is ꢀc the Owner Societies 2006
Phys. Chem. Chem. Phys., 2006, 8, 5444–5453 | 5449

View Article Online
a
Table 2 Triplet properties of chlorins
b
c
À1 À1
9
d
/M
À1 À1
9
Compound
Solvent
T
t /ms
F
ISC
k
ox/M
s
 10
k
q
s
 10
Ce6
H
H
EtOH
2
O/MeCN
O/buffer
0.05
0.10
0.5
3
2
0.10
1.50
0.10
o0.03
0.5
0.3
e
ZnCe6
H
2
H
2
H
2
H
2
O/buffer
0.40
0.10 5.0
3 (3)
f
O/glycerol
O/buffer
O/buffer
h
h
h
ZnCe6-M2
ZnPheida
1.50 [1.0]
0.08
0.03
0.2
o0.1
0.03
0.10
1.50
0.03
0.7 [0.8]
3 (3) 2
e
g
EtOH
O/buffer
ZnPheida-M2
H
2
0.80 [0.5]
o0.1
0.7
a
b
In argon-saturated solution at 24 1C (except for kox and À50 1C) and 0.01 M phosphate buffer, pH 7.3 (ZnCe6, pH 10), lexc = 420 nm.
H
For BQ. Values refer to phBQ.
2
O/
c
17 d
e
MeCN (1 : 1, v/v) and H O/glycerol (2 : 3, v/v). Reference of FISC = 0.5 for Chla in ethanol.
2
f
g
2+ h
.
At À50 1C. Value refers to MV
Values refer to M1.
the ZnCe6-M1 complex (Table 2 and Fig. 7a and 8a). The T–T
lifetime (initial decay component) becomes shorter. The triplet
lifetime for ZnCe6 and ZnPheida, each bound to the peptide
M2, is shortened from 1 ms to 50 ms upon addition of BQ
(Fig. 7 and 8).
absorption at 480 nm (DA ) under optically matched condi-
rel
tions was taken as a measure of the FISC value, which is
essentially the same for Ce6 in water or ethanol and for ZnCe6
and ZnCe6-M1/-M2 systems in water, but much smaller
for ZnPheida and ZnPheida-M1/-M2 complexes in water
The transient observed after triplet decay (Fig. 5b) is due to
ꢁ
+
the radical cation C with maximum at 400 nm, whereas the
ꢁ
À
(Table 2). This result is consistent with the observed low
radical anion Q
absorption coefficient.
contribution of the corresponding radical cation, MV
is virtually not detected due to its low
4
2–44
quantum yield of fluorescence for ZnPheida and indicates a high
degree of aggregation. The absorbance of the triplet state of
ZnPheida in solution and of the peptide-bound ZnPheida sys-
tems is much lower than those of ZnCe6 and ZnCe6-M1/-M2.
In order to search for a spectral
2
+
,
+
ꢁ
having a higher molar absorption coefficient in its MV
À
ꢁ
32,45
form at 380 and 420 nm than that of Q at 430 nm,
was used as alternative electron acceptor. The stronger ab-
ꢁ
+
sorption of the radical MV with maxima at 380 and 420 nm
3
.4 Radical ions
after 100 ms is evident in Fig. 6b.
Temperature dependent measurements in a mixture of aqu-
The rate constant of the triplet decay in either air- or argon-
saturated aqueous solution shows a linear dependence on the
T
eous buffer and glycerol (2 : 3, v/v) show that 1/t of ZnCe6
and ZnPheida decrease at lower temperatures due to less
BQ concentration. The slope of the plot of 1/t
the rate constant of quenching by electron transfer, is k =
T
vs. [BQ], i.e.,
q
9
À1 À1
efficient self-quenching caused by an increase in solvent visc-
3
 10
M
s
for ZnCe6 and ZnPheida. Examples are
9
À1 À1
osity. On the other hand, 1/t
at a given temperature. Generally, k
BQ]) decreases with increasing 1/T (not shown). The radical
T
increases upon addition of BQ
shown in Fig. 9. Lower values of k = (0.7–0.8) Â 10 M
s
q
q
(i.e., the slope of 1/t vs.
T
were found for the four zinc chlorin maquettes (Table 2). The
triplet yield of the zinc chlorin maquettes remains unchanged
upon addition of a forty-fold excess of BQ, whereas the
[
yield also decreases when the temperature is lowered. At
Fig. 7 Transient absorption spectra of ZnCe6-M2 (6 mM) in argon-
saturated 0.01 M phosphate buffer at 24 1C, pH 7.3, (a), (b) solid line:
at the end of the 420 nm pulse; (a) no additive (broken line: after 1 ms)
and (b) in the presence of BQ (10 mM, broken line: after 0.1 ms); insets:
kinetic traces at (a) 410 and 460 nm and (b) 360 and 450 nm as
indicated.
Fig. 8 Transient absorption spectra of ZnPheida-M2 (6 mM) in
argon-saturated 0.01 M phosphate buffer at 24 1C, pH 7.3, (a), (b)
solid line: at the end of the 420 nm pulse; (a) no additive (broken line:
at 1 ms) and (b) in the presence of BQ (10 mM, broken line: at 0.1 ms);
insets: kinetic traces at (a) 410 and 460 nm and (b) 350 and 480 nm as
indicated.
5
450 | Phys. Chem. Chem. Phys., 2006, 8, 5444–5453
This journal is ꢀc the Owner Societies 2006

View Article Online
Pheida in methanol/heavy water (1 : 20, v/v) solution (heavy
water in that case being applied for better singlet oxygen
detection) a quantum yield of singlet oxygen production of
1
6
FD r 0.02 was found, i.e., aggregation suppresses intersys-
tem crossing. The yield, however, increases to F = 0.7, when
D
Triton X-100 is added, favouring the monomeric form of
1
6
Pheida with high FISC
monomeric Chla decreases upon addition of water and a
.
Similar to Pheida the fraction of
1
4
triplet of Chla aggregates was not detected. Besides an
addition of a surfactant, the aggregation of Chla in aqueous
solution can also be suppressed by addition of de novo pep-
1
9
tides. In contrast, we did not find an increase of F and FISC
f
for ZnPheida upon incorporation into the maquettes (Tables 1
and 2). Currently, this observation cannot be explained satis-
factorily. However, we can exclude aggregation as a cause for
these low values, since the spectral properties (bathochromic
shifts in absorbance and fluorescence) indicate that the pig-
ments are monomeric and bound within the maquettes, and no
excess of free chlorin is present.
T
Fig. 9 Plots of 1/t vs. [BQ] for ZnCe6/BQ (J), ZnPheida/BQ (n),
ZnCe6-M2/BQ (K) and ZnPheida-M2/BQ (m) in aqueous solution,
pH 7.3 and for Chla/BQ in EtOH at 24 1C (&).
À50 1C, for example, the triplet lifetime is ca. 5 ms for ZnCe6
and electron transfer was no longer observable. The observed
decrease in rate constant of the quenching process at lower
temperatures can be explained by an activated process. As the
viscosity changes by several orders of magnitude in a small
temperature range of e.g., À30 to À60 1C, this is essentially a
viscosity-induced activation barrier.
4
.2 Complex formation between zinc chlorins and BQs
The ground state absorptions of metal-free chlorins and the
fluorescence excitation spectra above ca. 400 nm are similar in
the presence and absence of BQs. In contrast, ZnCe6 and
ZnPheida show a bathochromic shift upon BQ addition
(
Table 1), which is not observed for maquette incorporated
zinc chlorins. A specific interaction between the zinc-contain-
ing pigment and a quinone should therefore be considered.
These findings may be compared to examples from the litera-
ture where porphyrins containing appropriate hydroxyphenol
4
Discussion
4
.1 Aggregation behaviour
The observed formation of a new red-shifted band in either the
ground state absorption or fluorescence spectra of zinc chlor-
ins in the presence of BQ (Fig. 2 and 3) is reminiscent to
changes in the aggregation–disaggregation behaviour. From
studies of chlorophyll derivatives in aqueous solution (without
quinone addition) the following aggregation pattern has been
proposed: a water molecule coordinates via its oxygen atom to
2
9–31
substituents and BQs show specific interactions.
These
ubiquinone analogues and functional porphyrins interact via
multiple hydrogen bonds, but no metal ion is required, in
contrast to the chlorin cases presented here.
The novel stable band is proposed to be due to complex
formation between BQ and ZnCe6. It is suggested to be
formed in two steps, as indicated by the dependence of the
rate constant k for the decay of the Soret and Q absorption
2
+
the Mg
atom of one chlorophyll macrocycle and forms a
À
y
hydrogen bond with the keto CQO function of another
bands on the BQ concentration, and the rate constant k
+
for
6
chlorophyll molecule: MgÁ Á ÁO(H)HÁ Á ÁOQC. This complexa-
the increase of the new band (Fig. 2 and 3). The new generated
absorption bands of ZnPheida at 738 nm induced by BQ and
at 685 nm mediated by imidazole, give evidence for two
different types of complexes. Although the structure of the
complex is not known, a quinone-assisted conversion of the
chlorin monomer into an aggregate and a water-assisted
conversion into another aggregate are considered to be in
accordance with the two-step kinetics (Fig. 4) and the pro-
tion can also occur with zinc as central metal. For chlorosomal
chlorophylls and their more stable zinc analogues, a self-
aggregation was found to be driven by intermolecular inter-
1
action among the 3 -hydroxyl, the central metal and the
2
–6
1
3-keto-group.
High concentrations, low temperatures
and dilution by nonpolar solvents shift this equilibrium to-
ward larger self-assembled species. For Ce6 it was also found
that a decrease of the pH to 2.7 leads to an increase in
hydrophobicity, which is related to the protonation of the
1
3
nounced red-shifts in absorption and fluorescence (Fig. 3
and Table 1).
2
4
compound. The variability of the tendency to aggregate,
which is visible through a red-shift of the absorption bands, is
therefore related to changes in the chemical structure. How-
ever, none of these descriptions apply to the present systems,
4
.3 Electron transfer from the chlorin triplet state to BQ
Dynamic quenching by BQs in polar solvents has been studied
9–12
1
because neither a 3 -hydroxyl group is present in the chemical
for Chla.
For the presented chlorin–quinone systems,
structure of the chlorins, nor were measurements carried out
below pH 7.3.
reactions (1)–(4) in Scheme 1 show reversible photooxidation
via the excited singlet state, ISC, electron transfer and charge
+
ꢁ
ꢁ
À 7–12
We propose (see also ref. 20 and 23) that aggregation does
not take place for Ce6, whereas for Pheida in aqueous solution
a monomer–dimer equilibrium describes the situation. For
recombination of the two ions, C
3
and Q . All other
decay processes of C* in the given system, even quenching by
traces of oxygen, play no significant role due to the dominance
This journal is ꢀc the Owner Societies 2006
Phys. Chem. Chem. Phys., 2006, 8, 5444–5453 | 5451

View Article Online
of reaction (2). The radicals, i.e., pigment cation and quinone
porated in peptides to noncovalently bound quinones were
investigated. The aim was to study the influence of peptides on
the electron transfer process from excited free and bound zinc
chlorins to quinones. ZnCe6 and ZnPheida show a steady-
state interaction with quinones which is evident by a new
absorption band at longer wavelengths. In contrast, this
complex formation induced by the quinone was not found
for the metal free compounds (Ce6 and Pheida) and is also
absent for the four different metallopeptides (ZnPheida-M1,
ZnPheida-M2, ZnCe6-M1, ZnCe6-M2).
anion, have already been characterized by different
4
2–44
groups.
At pH 4 4 in aqueous solution (as in our
ꢁ
À
experiments) the equilibrium between Q and the conjugated
ꢁ
45
acid (semiquinone radical HQ ) lies on the anion side. The
3
radical pair, (C
ꢁ
+
ꢁ
À
Q
), which is the direct result of the
quenching process, can separate into the chlorin cation radi-
ꢁ
+
cal, C , and the quinone anion radical, reaction (3). In
ꢁ
À
ꢁ
+
principle, a reaction of the separated Q and C to different
8
products is possible, reaction (5). Product formation is in
competition with charge recombination, reaction (4), but its
ZnCe6 in solution as well as bound to the peptides shows
substantial quantum yields of fluorescence and intersystem
crossing, whereas both values are lower for the ZnPheida
systems. This suggests that ZnCe6 shows no aggregation,
but ZnPheida has a higher tendency to aggregate. Never-
theless, both pigments could be studied by fluorescence and
transient absorption spectroscopy, which is not feasible for
Chla in aqueous solution.
occurrence is unlikely at ambient temperature. A typical
9
rate constant for back electron transfer is k
À1 À1 8–11
3
The transient absorption spectra of C* and C are over-
3
lapping and partly similar. For Chla* and Chla the molar
4
= 2 Â 10
M
s .
ꢁ
+
ꢁ
+
4
absorption coefficients are e = 3.5 Â 10 and e = 3.5 Â
4
60
42–44
400
4
0
À1
À1
1
M
cm , respectively.
For chlorins in aqueous
solution the T–T absorption and radical cation properties
4
The triplet state is quenched by BQs via electron transfer
and the observed radical cations of the zinc chlorins essentially
decay by ‘‘reverse’’ electron transfer from the quinone radical
anion. In comparison to unbound zinc chlorins, the pig-
ment–peptide complexes as a model system for a reaction
center show much longer triplet lifetimes and four times lower
rate constants of quenching by quinones, indicating shielding
of the pigments by the peptides.
À1
are not documented, thus values of emax = 4 Â 10
M
À1
3
ꢁ
+
cm
were taken for C* and C . The spectra of the
ꢁ
ꢁ
À
semiquinone radical, HQ and the radical anion, Q are only
slightly shifted in aqueous solution. They are characterized by
two peaks around 320 and 430 nm with molar absorption
4
À1
À1 45
3
coefficient for parent BQ: e420 = 0.6 Â 10
M
cm
.
Hence, in comparison with the absorption coefficients of C*
ꢁ
+
ꢁ
À
and C the transient absorption of Q is masked by that of
This work describes some of the difficulties and the notable
complexity on the way to synthesize an artificial reaction
center. The necessary conditions are that the pigments are
ligated by the peptides to prevent complex formation with the
acceptor molecule, but the peptides have to be as small as
possible in order to keep the distance between the donor and
acceptor molecule small enough to allow efficient electron
transfer. As a major finding we demonstrate that in model
systems, similar to nature, a covalent attachment of a quinone
to the peptide is not essential.
ꢁ
+
C
and therefore not detectable in our transient spectra.
Addition of up to a forty fold excess of quinone to the zinc
1
chlorins does not influence the fluorescence lifetime. Thus C*
of zinc chlorins does not participate in the electron transfer as
1
2
already concluded for a related case. This is in contrast to
native photosystems, where the electron transfer occurs from
the singlet excited state.
A significant effect on photoinduced electron transfer from
ZnCe6 in aqueous solution to phBQ was already observed by
3
6
EPR for a peptide with a sequence similar to M2. Under our
conditions, the yield of the radical cation formation is essen-
tially unchanged (see section 3.4), but the efficiency, as ex-
Acknowledgements
9
À1 À1
8
We thank the Deutsche Forschungsgemeinschaft (SFB 663,
subproject A7) for financial support, M. van Gastel for helpful
discussions, and I. Heise, M. Trinoga, G. Koc-Weier,
N. Dickmann and L. J. Currell for technical assistance.
pressed by k
q
= 3 Â 10 M
s
for the zinc chlorins in
8
À1 À1
aqueous solution vs. k
q
= 7 Â 10 and 8 Â 10 M
s for the
zinc chlorins bound to the peptides M2 and M1 (Table 2), is
about four times reduced. The smaller electron transfer rate is
due to shielding of the zinc chlorin by the four-helix bundle.
Nevertheless, the observed quenching rate constants are in
quite good agreement with the literature values for pigment
References
1
R. E. Blankenship, in Molecular Mechanisms of Photosynthesis,
Blackwell Science, Oxford, 2002.
9
–12
quenching by BQ.
The triplet state of zinc protoporphyrin
IX in a synthetic four-helix bundle protein in aqueous solution
9
2 (a) J. Chiefari, K. Griebenow, N. Griebenow, T. S. Balaban, A. R.
Holzwarth and K. Schaffner, J. Phys. Chem., 1995, 99, 1357; (b) T.
S. Balaban, A. R. Holzwarth, K. Schaffner, G.-J. Boender and H.
J. M. de Groot, Biochemistry, 1995, 34, 15259; (c) T. Mizoguchi, S.
Sakamoto, Y. Koyama, K. Ogura and F. Inagaki, Photochem.
Photobiol., 1998, 67, 239.
À1 À1 35b
is also quenched by AQS with k = 0.5 Â 10 M
s . The
q
k values for triplet quenching by BQ of the chlorins (Table 2),
q
Chla and zinc tetraphenylporphyrin are close to the diffusion-
2
6,28
controlled limit.
Significant differences with respect to
3
H. Tamiaki, M. Amakawa, A. R. Holzwarth and K. Schaffner,
Photosynth. Res., 2002, 71, 59.
H. Tamiaki, Photochem. Photobiol. Sci., 2005, 4, 675.
T. Miyatake, H. Tamiaki, M. Fujiwara and T. Matsushita, Bioorg.
Med. Chem., 2004, 12, 2173.
T. S. Balaban, H. Tamiaki and A. R. Holzwarth, Top. Curr.
Chem., 2005, 258, 1.
´
C. Bohne, A. Faljoni-Arario and G. Cilento, Photochem. Photo-
ligation, absorption and fluorescence behaviour between the
peptides M1 and M2 were not observed.
4
5
5
Conclusion
6
7
Light-induced electron transfer processes from zinc chlorins
ZnCe6 and ZnPheida) in aqueous solution as well as incor-
(
biol., 1988, 48, 341.
5
452 | Phys. Chem. Chem. Phys., 2006, 8, 5444–5453
This journal is ꢀc the Owner Societies 2006

View Article Online
8
9
A. K. Chibisov, Photochem. Photobiol., 1969, 10, 331.
J. Kelly and G. Porter, Proc. R. Soc. London, Ser. A, 1970, 319,
Kadish, J. Am. Chem. Soc., 2001, 123, 10676; (e) P. P. Mishra, J.
Bhatnagar and A. Datta, Chem. Phys. Lett., 2004, 386, 158; (f) T.
K. Mukherjee, P. P. Mishra and A. Datta, Chem. Phys. Lett., 2005,
407, 119.
319.
1
1
0 B. J. Hales and J. R. Bolton, J. Am. Chem. Soc., 1972, 94, 3314.
1 (a) P. G. Bowers and G. Porter, Proc. R. Soc. London, Ser. A, 1967,
28 G. J. Kavarnos and N. J. Turro, Chem. Rev., 1986, 86, 401.
29 (a) T. Buranda, N. Soice, S. Lin, R. Larsen and M. Ondrias, J.
Phys. Chem., 1996, 100, 18868; (b) T. Buranda, M. Enlow, J.
Griener, N. Soice and M. Ondrias, J. Phys. Chem. B, 1998, 102,
9081.
296, 435; (b) R. G. Brown, A. Harriman and L. Harris, J. Chem.
Soc., Faraday Trans. 2, 1978, 1193; (c) F. Castelli, G. Cheddar, F.
Rizzuto and G. Tollin, Photochem. Photobiol., 1979, 29, 153; (d) R.
H. Clarke, D. R. Hobart and W. R. Leenstra, J. Am. Chem. Soc.,
1
979, 101, 2416.
30 K. Okamoto, K. Ohkubo, K. M. Kadish and S. Fukuzumi, J.
Phys. Chem. A, 2004, 108, 10405.
1
1
2 F. Castelli, in Laser Application in Chemistry, ed. K. L. Kompa and
J. Wanner, NATO Series B, Plenum Press, New York, 1984, vol.
1
3 (a) K. Uehara, Y. Hioki and M. Mimuro, Photochem. Photobiol.,
1
7
4 A. K. Chibisov, T. D. Slavnova and H. Go
Photobiol., B, 2003, 72, 11.
5 (a) H. Scheer and J. J. Katz, J. Am. Chem. Soc., 1978, 100, 561; (b)
J. J. Katz, Spectrum, 1994, 7, 1; (c) G. W. Kenner, S. W.
McCombie and K. M. Smith, J. Chem. Soc., Perkin Trans. 2,
31 (a) Y. Aoyama, M. Asakawa, Y. Matsui and H. Ogoshi, J. Am.
Chem. Soc., 1991, 113, 6233; (b) T. Hayashi, T. Miyahara, N.
Koide, Y. Kato, H. Masuda and H. Ogoshi, J. Am. Chem. Soc.,
1997, 119, 7281; (c) H. Ogoshi and T. Mizutani, Curr. Opin. Chem.
Biol., 1999, 3, 736; (d) F. D’Souza and G. R. Deviprasad, J. Org.
Chem., 2001, 66, 4601; (e) A. Tsuda, C. Fukumoto and T. Oshima,
J. Am. Chem. Soc., 2003, 125, 5811; (f) J. R. Darwent, P. Douglas,
A. Harriman, G. Porter and M. C. Richoux, Coord. Chem. Rev.,
1982, 44, 83.
32 T. Watanabe and K. Honda, J. Phys. Chem., 1982, 86, 2617.
33 H. A. Montejano, M. Gervaldo and S. G. Bertolotti, Dyes Pigm.,
2005, 64, 117.
34 N. De Jonge, H. K. Rau and W. Haehnel, Z. Phys. Chem., 1999,
213, 175.
05, pp. 225.
993, 58, 127; (b) R. Vladkova, Photochem. Photobiol., 2000, 71,
1.
1
¨
rner, J. Photochem.
1
1973, 2517; (d) H. Scheer, R. J. Porra and J. M. Anderson,
Photochem. Photobiol., 1989, 50, 403; (e) T. Suzuki and Y. Shioi,
Photosynth. Res., 2002, 74, 217.
6 C. Tanielian, C. Wolff and M. Esch, J. Phys. Chem., 1996, 100,
1
1
1
1
2
2
2
2
2
2
6
555.
35 (a) M. Fahnenschmidt, R. Bittl, E. Schlodder, W. Haehnel and W.
Lubitz, Phys. Chem. Chem. Phys., 2001, 3, 4082; (b) M. Fah-
nenschmidt, R. Bittl, H. K. Rau, W. Haehnel and W. Lubitz,
Chem. Phys. Lett., 2000, 323, 329.
36 M. R. Razeghifard and T. Wydrzynski, Biochemistry, 2003, 42,
1024.
37 S. Hay, B. B. Wallace, T. A. Smith, K. P. Ghiggino and
T. Wydrzynski, Proc. Natl. Acad. Sci. U. S. A., 2004, 101,
17675.
38 D. E. Robertson, R. S. Farid, C. C. Moser, J. L. Urbauer, S. E.
Mulholland, R. Pidikiti, J. D. Lear, A. J. Wand, W. F. DeGrado
and P. L. Dutton, Nature, 1994, 368, 425.
7 M. Jabben, N. A. Garcia, S. E. Braslavsky and K. Schaffner,
Photochem. Photobiol., 1986, 43, 127.
8 A. Agostiano, L. Catucci, G. Colafemmina and H. Scheer, J. Phys.
Chem. B, 2002, 106, 1446.
9 A. Dudkowiak, C. Nakamura, T. Arai and J. Miyake, J. Photo-
chem. Photobiol., B, 1998, 45, 43.
0 B. Roeder and H. Wabnitz, J. Photochem. Photobiol., B, 1987, 1,
1
03.
1 J. D. Spikes and J. C. Bommer, J. Photochem. Photobiol., B, 1993,
7, 135.
1
2 S. V. Kuznetsov, M. Bazin and R. Santus, J. Photochem. Photo-
biol., A, 1997, 103, 57.
3 I. Eichwurzel, H. Stiel and B. Roeder, J. Photochem. Photobiol., B,
39 R. E. Sharp, J. R. Diers, D. F. Bocian and P. L. Dutton, J. Am.
Chem. Soc., 1998, 120, 7103.
¨
¨
40 A. Mennenga, PhD Thesis, Heinrich-Heine-Universitat, Dussel-
2
000, 54, 194.
ˇ
4 B. Cunderlı
Photobiol., B, 1999, 53, 81.
5 C. Bruckner, J. R. McCarthy, H. W. Daniell, Z. D. Pendon, R. P.
Ilagan, T. M. Francis, L. Ren, R. R. Birge and H. A. Frank, Chem.
Phys., 2003, 294, 285.
6 M. Ebersole, P. R. Levstein and H. van Willigen, J. Phys. Chem.,
kova, L. Gangeskar and J. Moan, J. Photochem.
´ ´
dorf, 2007.
41 W. R. Scheidt and D. M. Chipman, J. Am. Chem. Soc., 1986, 108,
1163.
42 (a) T. Watanabe and K. Honda, J. Am. Chem. Soc., 1980, 102, 370;
(b) J. Kiwi and M. Gratzel, J. Phys. Chem., 1980, 84, 1503.
¨
43 J.-P. Chauvet, R. Viovy, R. Santus and E. J. Land, J. Phys. Chem.,
1981, 85, 3449.
44 H. Levanon and P. Neta, J. Phys. Chem., 1982, 86, 4532.
45 (a) V. A. Roginsky, L. M. Pisarenko, W. Bors, C. Michel and M.
Saran, J. Chem. Soc., Faraday Trans., 1998, 94, 1835; (b) S.
Steenken and P. Neta, in The Chemistry of Phenols, ed. Z.
Rappoport, Wiley, New York, 2003, p. 1107.
¨
2
2
1
992, 96, 9310.
7 (a) G. Porter, Proc. R. Soc. London, Ser. A, 1978, 362, 281; (b) W.
E. Ford and G. Tollin, Photochem. Photobiol., 1983, 38, 441; (c) P.
R. Levstein and H. van Willigen, Chem. Phys. Lett., 1991, 187, 415;
(d) S. Fukuzumi, K. Ohkubo, H. Imahori, J. Shao, Z. Ou, G.
Zheng, Y. Chen, R. K. Pandey, M. Fujitsuka, O. Ito and K. M.
This journal is ꢀc the Owner Societies 2006
Phys. Chem. Chem. Phys., 2006, 8, 5444–5453 | 5453