FEBS Letters 582 (2008) 3657–3662

Spectrally silent light induced conformation change inphotosynthetic reaction centers

Laszlo Nagya,*, Peter Marotia, Masahide Terazimab

a Institute of Medical Physics and Biophysics, University of Szeged, 6720 Szeged, Rerrich B. ter. 1., Hungaryb Kyoto University, Graduate School of Science, Department of Chemistry, Kyoto 6068502, Japan

Received 17 September 2008; accepted 23 September 2008

Available online 7 October 2008

Edited by Richard Cogdell

Abstract Spectrally silent conformation change after photo-excitation of photosynthetic reaction centers isolated fromRhodobacter sphaeroides R-26 was observed by the opticalheterodyne transient grating technique. The signal showed spec-trally silent structural change in photosynthetic reaction centersfollowed by the primary P+BPh� charge separation and thischange remains even after the charge recombination. Withoutbound quinone to the RC, the conformation change relaxes withabout 28 ls lifetime. The presence of quinone at the primary qui-none (QA) site may suppress this conformation change. How-ever, a weak relaxation with 30–40 ls lifetime is still observedunder the presence of QA, which increases up to 40 ls as a func-tion of the occupancy of the secondary quinone (QB) site.� 2008 Federation of European Biochemical Societies. Pub-lished by Elsevier B.V. All rights reserved.

Keywords: Transient grating; Heterodyne detection; Reactioncenter

1. Introduction

There are several types of photosynthetic reaction centers

(RCs) in living organisms (PS-I and PS-II of plants and cyano-

bacteria, RCs of purple and green bacteria; see e.g. [1]) which

are developed to convert light energy into chemical potential in

a series of (photo)physical and (photo)chemical reactions.

The basic processes of the photosynthetic energy conversion

in all types of the RCs include (a) electron excitation of chlo-

rine pigments by light, (b) charge separation and stabilization

reactions between redox active cofactors bound to the protein,

(c) rearrangement of the dielectric medium and hydrogen bond

interactions (including protonation and deprotonation of spe-

cific amino acids), and (d) conformational movements within

the protein [including transition of (sub)states between dark

and light adapted forms].

Abbreviations: BPh, bacteriopheophytin; DEAE, diethylaminoethyl;LDAO, N,N-dimethyldodecylamine-N-oxide; LO, local oscillator;OHD, optical heterodyne detection; P, primary donor; QA, primaryquinone; QB, secondary quinone; R., Rhodobacter; RC, photosyntheticreaction center; TG, transient grating; TrL, transient lens; UQ-10,ubiquinone-10

*Corresponding author. Fax: + 36 62 544121.E-mail address: lnagy@sol.cc.u-szeged.hu (L. Nagy).

0014-5793/$34.00 � 2008 Federation of European Biochemical Societies. Pu

doi:10.1016/j.febslet.2008.09.048

There are several evidences that RCs are in different confor-

mation states in dark and light, e.g. [2–6]. Kinetic components

in transient absorption were observed at specific wavelength

[7,8], in capacitive potentiometry [9] and time resolved FTIR

spectroscopy [10,11] which are generally assigned to conforma-

tional transients related to QA�QB to QAQB

� electron transfer.

Recent crystallographic experiments did not show large qui-

none displacements at the QB site on the time scale of the sec-

ondary electron transfer [12,13] in agreement with FTIR

studies [14,15]. It seems reasonable to conclude that even larger

quinone movement is not necessarily accompanied with con-

siderable structural change [13].

Transient grating technique was successfully used for many

applications including thermodynamics and kinetics of CO

binding of myoglobin [16,17], the photocycle of the photoac-

tive yellow protein (PYP, [18]), determining diffusion coeffi-

cient of proteins and DNA [19] and conformational changes

of photosensor proteins [20–23]. In a recent publication we

have shown that this method can be successfully applied to

investigation of light induced charge transfer processes and

accompanied protein relaxation movements in bacterial reac-

tion centers [24]. Here, we give further evidences.

1.1. Principle of the OHD-TG measurement

The principle of the optical heterodyne detection (OHD) of

the transient grating (TG) signal was described previously [25–

28]. After photoexcitation of a sample with a grating pulsed

light, a signal field (ES(t)) is created by the diffraction of a

probe light. If it is interfered with a local oscillator (LO) field

(ELO) the observed light intensity is

IðtÞ ¼ ajELO þ EsðtÞj2 � ILO þ dnðtÞ cos D/IexIpr; ð1Þ

where a is an instrumental constant, dn(t) is the refractive in-

dex change, D/ phase difference between the local oscillator

(LO) and probe light fields and Iex, ILO, and Ipr are the inten-

sities of the excitation, LO, and probe light, respectively. Here

we assume that the refractive index change is the dominant

source of the signal, and the signal field is weak compared to

ELO (|ELO| >> |ES|). Since the local oscillator light intensity is

a constant, ILO provides constant background. The second

term of Eq. (1) represents the OHD-TG signal. The OHD-

TG signal intensity depends on the relative phase difference be-

tween the LO and signal fields, D/. The maximum intensity of

the thermal grating component of a reference sample is

achieved at D/ = 0.

Two main factors contribute to the refractive index change:

the thermal effect [thermal grating, dnth(t)] and a change in

blished by Elsevier B.V. All rights reserved.

3658 L. Nagy et al. / FEBS Letters 582 (2008) 3657–3662

chemical species by the reaction [species grating, dnspe(t)]. The

species grating component consists of the population and vol-

ume terms. Hence, we can monitor the temporal changes of the

energy, absorption change, and molecular volume by analyz-

ing the OHD-TG signal.

When the energy relaxation is faster than the diffusion pro-

cess, the temporal profile of dnth(t) is determined by the ther-

mal diffusion, and is given by

dnthðtÞ ¼ dn0th expð�Dthq2tÞ; ð2Þ

where Dth is the diffusion constant, q is the grating wavenum-

ber.

The kinetics of the species grating signal intensity, dnspe(t), is

given by the difference of the refractive index changes due to

the reactant (dnr) and product (dnp). If the back reaction from

the product to the reactant is comparable to the diffusion pro-

cess it is given by

dnspeðtÞ ¼ dn0p �

kk þ ðDr � DpÞq2

dn0r

� �expf�ðk

þ Dpq2tÞg þ dn0r ðDr � DpÞq2

k þ ðDr � DpÞq2expf�Drq2tg; ð3Þ

where Dr and Dp are diffusion coefficients of the reactant and

the product, respectively.

2. Materials and methods

2.1. Sample preparationsRhodobacter sphaeroides R-26 cells were grown photoheterotrophi-

cally under anaerobic conditions. RCs were prepared by detergent(LDAO, N,N-dimethyldodecylamine-N-oxide) solubilization followedby ammonium sulfate precipitation and diethylaminoethyl (DEAE)Sephacell anion exchange chromatography [29]. The primary (QA)and the secondary (QB) quinones were extracted out according toOkamura et al. [30]. RCs with different amount of bound quinoneswere prepared by addition of ubiquinone-10 (UQ-10) to the solution.The quinone/RC ratio was checked by kinetic absorption change mea-surement as described by Tandori et al. [29].

Fig. 1. The intensity of different components of the measured grating signaTG� indicate the original data measured with two opposite phases of the LOthe transient lens and grating components (gray curves), respectively, calculacurves are shown by the black lines. The best fitting parameters for the OHD-s) for the OHD-TG component and I = �0.97exp(�t/1.5 s) � 0.03exp(�t/0.11better comparison.

2.2. Transient grating measurementsAn excitation laser beam (from the second harmonics of a Nd:YAG

laser, k = 532 nm, s = 10 ns) and a cw IR beam (YAG, 1064 nm) weresplit by a transmission grating (optical mask) and the first order dif-fracted beams were combined again on the sample using a concavemirror. In order to avoid overexcitation of the sample, the energy ofthe excitation beam was reduced by neutral density filters (typically be-low 100 lJ). One of the cw IR beams was used for the probe beam ofthe TG signal. The other beam intensity was attenuated about 1/100 bya neutral density filter for the use of the local LO light and of theOHD-TG signal. The filter was slightly adjusted tilting by a computerand used for a fine adjustment of the phase of the LO field to the probefield by changing the optical path length inside. The LO light intensitywas detected by a photodiode. The LO light intensity could be changednot only by the interference between the signal and the LO light, butalso by the transient lens (TrL) signal created by the pump beam[31]. The OHD-TG signal was obtained by calculating the differencebetween minimum and maximum LO light intensities matching twoopposite phases after filter adjustments.

The spacing of the grating fringe, equivalently, the grating wave-number, was measured by the decay rate constant of the thermal grat-ing signal from a calorimetric standard sample (bromocresol purple).The signal was averaged and stored by a digital oscilloscope (Tektron-ics, TDS-520) [17]. In some experiments, absorption change was mon-itored by a He–Ne laser (590 nm) after the flash excitation. Theexperimental set up was described in our earlier work [24].

3. Results and discussion

After photoexcitation of the sample solutions, the detected

probe light intensity increased abruptly then decreased as a

function of time. The two components in the signal (TG and

TrL) can be separated based on the phase sensitive nature of

the OHD-TG signal. In contrast to TrL, the amplitude and

the sign of the OHD-TG signal depends on the phase differ-

ence between the LO and the signal (see D/ in Eq. (1)). The

OHD-TG signal changes the sign by phase difference of

180�. Hence, subtraction of the D/ = 180� signal from the

D/ = 0� signal provides OHD-TG without TrL contribution.

The curves TG+ and TG� in Fig. 1 show the probe light inten-

sities measured at D/ = 180� and D/ = 0�, respectively in QB

l at different LO light conditions for QB reconstituted RCs. TG+ andlight (D/ = 180� and D/ = 0�, respectively). TrL and OHD-TG indicateted from TG+ and TG� as described in the Section 1.1. The best fittedTG and TrL components are I = 0.97exp(�t/0.15 s) + 0.03exp(�t/0.055s) for the TrL component. Here the amplitudes are normalized to 1 for

Fig. 2. The intensity of the OHD-TG signal of RCs at different (bound quinone)/RC ratios as indicated. The signal of the reference sample is alsoshown. Measurement was done by the experimental setup outlined in Ohmori et al. [24] and the OHD-TG signal was calculated as described in Fig. 1.RCs were suspended in 10 mM Tris, pH 8.0, 0.01% LDAO, 100 mM NaCl. The optical density of the sample was 1 at the excitation wavelength of532 nm. The frequency of the excitation was 0.2 Hz for each sample, except for that of the reference (10 Hz).

L. Nagy et al. / FEBS Letters 582 (2008) 3657–3662 3659

reconstituted RCs. The OHD-TG signal was obtained by sub-

traction of these signals. Once, we obtained the OHD-TG sig-

nal, the TrL component may be calculated from the OHD-TG

and the observed signals.

The OHD-TG signals measured for the calorimetric refer-

ence and for RCs at different ratio of bound quinone to the

reaction center ([Q]/[RC]) are shown in Fig. 2. For the refer-

ence solution, the TG signal demonstrated monoexponential

decay of about 35 ls lifetime. Apparently, this is the thermal

grating signal and the decay rate was determined by Dthq2

(see Eq. (2)). For photosynthetic RCs typically two main

phases appeared after the abrupt initial increase of the refrac-

tive index. The fast phase (lifetime s1) decayed in the few

10�s ls time range and the slow phase (lifetime s2) in the milli-

second–second time scale:

ITGðtÞ ¼ afA1 expð�t=s1Þ þ A2 expð�t=s2Þg ð4Þ

In RCs of 8% Q/RC we obtained s1 = 28.1 ls, s2 = 110 ms and

the ratio of the amplitudes A1:A2 = 94.2:5.8. With increasing

[Q]/[RC], the fast component decreased and that of the slow

component increased. Hence, it is reasonable to consider that

the fast component represents the dynamics of RC without Q.

First, the slow component is discussed. We reported previ-

ously [24] that the lifetime of the milliseconds–seconds dynam-

ics depended on the grating wavenumber that indicated that

this kinetics represented the protein diffusion process. The sin-

gle exponential behavior of this component showed that the

diffusion coefficient of the reactant was almost the same as that

of the charge separated state. The rate constant of the grating

signal was expressed as Dq2 + k, where k was the rate constant

of the charge recombination: (100 ms)�1 and (1 s)�1 for

P+QA� fi PQA and P+QAQB

� fi PQAQB, respectively. From

these data, we calculated the diffusion coefficient

D = 4.5 Æ 10�11 m2/s and the hydrodynamic diameter

d = 10.2 nm (estimated from the Einstein–Stokes relationship)

for the QA reconstituted RC/LDAO micelles. These values

agreed fairly well with results of our earlier publication

(D = 3.8 Æ 10�11 m2/s and d = 11.4 nm), and other data ob-

tained e.g. by AFM [32] or dynamic light scattering [33] exper-

iments.

Since the time constant of the fast component did not de-

pend on the grating wavenumber, this phase is not due to

the thermal diffusion or protein diffusion processes, but it

should represent reaction dynamics of the photoexcited RC.

The thermal grating signal intensity was too weak to be ob-

served under these experimental conditions. The thermal grat-

ing could be seen only when the RC was overexcited by high

excitation rate and the absorbed laser energy was released as

heat (Fig. 3). After the photoexcitation, the P+BPh� charge

pair is formed in about 3 ps after the flash excitation and de-

cays via charge recombination in about 10 ns (see [34]). The

observed TG signal of 28 ls lifetime does not appear in the

transient absorption measurement. Hence, this component is

spectrally silent and represents conformation change induced

by PBPh fi P+BPh� charge separation. Interestingly, the

relaxation rate constant of this conformation change is small,

and the protein reserves its state even after the charge recom-

bination when the chromophores relaxed to their initial

(ground) states. As the amplitude of this component is large

for RC without Q sample, the 28 ls-signal should be due to

structural rearrangement of the protein in the absence of the

primary and secondary quinones. ‘‘Spectrally silent’’ compo-

nents are well known in many proteins [16,18,23], and there

are only few appropriate methods suitable to investigate them;

transient grating is one of these techniques.

The decrease of the relative amplitude of the 28 ls decay

component by addition of quinone to the solution suggests

that the presence of quinone at the QA site suppresses this con-

formational relaxation, or the conformation change is fixed

until the charge recombination occurs with the lifetimes of

100 ms for QA sample and of 1 s for QAQB sample.

However, we should note that a weak 30 ls-component was

observed even for QA and QAQB samples. This component

might be observed by other spectroscopic techniques. In fact,

0

2

4

6

8

10

12

14

1.E-07 1.E-06 1.E-05 1.E-04 1.E-03 1.E-02 1.E-01 1.E+00

Time (s)

ITG(r

el.)

NoQ: 26.0 μτ s

Ref:: 8.7 μτ s

QB+trb

QB highenergy

0.32

0.8

Fig. 3. The intensity of the OHD-TG signal of RCs at different (bound quinone)/RC ratios as indicated. Measurement was done as described inFig. 2 but the grating conditions were changed. This is indicated by the lifetime of the reference signal. The OHD-TG signal was calculated asdescribed in Fig. 1. ‘‘QB high energy’’ indicates reconstituted QB sample, but the excitation repetition rate increased to 10 Hz and the energy of thelaser beam was oversaturating.

3660 L. Nagy et al. / FEBS Letters 582 (2008) 3657–3662

several experimental techniques demonstrated two or more ki-

netic components in the 20–40 ls and 200–500 ls range which

were suggested to be related to the P+QA�QB to P+QAQB

�

electron transfer [7–9].

Additional increase in the lifetime of this decay component

can be seen, although its amplitude is small, if quinone is titrated

into the QB site (Fig. 4 and Table 1). Since this component is re-

lated to the presence of the secondary quinone, QB, it can be as-

signed to the interquinone electron transfer. To our surprise the

amplitude of this component is smallest for the fully reconsti-

tuted 2Q/RC sample and we did not see other kinetic compo-

nents, e.g. the components of few hundreds of microseconds,

described by the ‘‘conformational gating’’ model [3].

Interestingly, the presence of non-redox active inhibitor at

the QB site, like terbutryn, does not affect the 28 ls component.

0

10

20

30

40

50

60

0 0.5 1Q

τ(μs

)

A(r

el.)

Fig. 4. The lifetime of the submillisecond component of the transient gratirepresent the data calculated from first (Fig. 2) and second (Fig. 3) grating cThe insert shows the relative amplitude of the TG component as a function

Only the observed slow phase is speeded up which is under-

standable, since the life time of the absorption change is mod-

ified from sslow = 1.5 s to sfast = 0.11 s after the treatment with

this chemical. This fact indicates that ‘‘empty’’ quinone site is

responsible for this kinetic component in the microsecond time

scale, regardless the QB site is occupied by the redox active

UQ-10 or non-redox active terbutryn.

Fig. 5 shows that the primary quinone is sitting in a loop

formed by random structures at the end of D and E transmem-

brane and mde cytoplasmic helices. The strong packing of the

intramembrane helix structure does not support large confor-

mation movements, but the empty random loop can be very

flexible. We also indicate the amino acid residue TrpM252

which is in close contact with the BPh in the A electron trans-

port branch and QA and supports tunneling of the electron to

1.5 2 2.5/RC

0

4

8

12

16

0 0.5 1 1.5 2Q/RC

ng signal as a function of the Q/RC ratio. Filled circles and squaresonditions, respectively. Measurement was done as described in Fig. 1.of the Q/RC ratio.

Table 1Summary of kinetic parameters of the measured transient grating signal presented in Fig 2 and hydrodynamic data calculated for isolated RCs. A1

(%), A2 (%) and s1 (ls), s2 (ms) represent the amplitudes and life times of the main phases of the kinetic components of the ITG signal, respectively.sTrL (ms) is the life time of the thermal lens component of the signal. D is the diffusion coefficient, and d is the hydrodynamic diameter of the RCmicellar system.

Q/RC A1 (%) s1 (ls) A2 (%) s2 (ms) sTrL (ms) D 10�11 (m2/s) d (nm)

0a 100 34.70ab 100 15.00a 100 8.7

0.06 94.2 28.1 5.8 � 1100.32 1.2 28.7 28.8 52.4 1150.54 48.5 29.0 51.5 49.3 119(0.5)b (50.0)b (28.0)b (50.0)b (54.3)b (108)b 4.5c 10.2c

0.68 37.1 29.6 62.4 63.4 106 (3.8)b (11.4)b

0.81 23.7 32.4 75.4 64.8 1220.91 17.9 31.7 84.8 54.6 1111.00 6.7 40.8 92.6 52.6 123

1.2 6.7 39.5 92.6 70.9 182.31.3 8.2 2.3 92.6 71.7 285.82.0 1.8 46.1 98.2 158.1 1490 2.8 15.4

aCalorimetric reference sample at different grating conditions.bOhmori et al. (2008).cCalculated from the average of samples Q/RC < 1 as described in the text.

L. Nagy et al. / FEBS Letters 582 (2008) 3657–3662 3661

the primary acceptor. In the absence of QA and its interactions

with the surrounding amino acid residues it is reasonable to ac-

Fig. 5. The position of the primary quinone, QA, at the loop formedby helical and random structures of the protein close to the membranesurface. The BPh in the active ‘‘A’’ electron transport branch and thecharacteristic amino acids around the QA (HisM219, AlaM260,TrpM252) are also indicated. Picture was drawn by HyperChem usingthe crystal structure 1PCR [36] downloaded from the Protein DataBank, Brookhaven.

cept a large conformational movement in the empty QA pock-

et. Also, the contribution of relaxation of water molecules

(internal and external ones) should be taken into account.

Although the lifetime of this dynamics is in the range of the

triplet formation, the probability of this process is too small

in our experimental conditions (no secondary donor is present,

primary quinone is oxidized, low probability for double excita-

tion, oxygen concentration is saturating). The discussion of the

contribution of the triplet states to the TG signal would be

important, but it needs more investigation and cannot be the

scope of this paper.

The absence of the few hundreds of microsecond compo-

nent is rather surprising because it is well stated by other

measurements in the literature [3,7–9]. Considering the extre-

mely high sensitivity of the OHD-TG method, we can argue

for no large displacement of the secondary quinone induced

by the charge movements and the other volume changes

due to charge relaxation processes should be very small. Re-

cently Koepke et al. [35] studied crystal structures with QB in

proximal and distal positions and found that the quinone

movement between the two positions was not accompanied

with large conformational rearrangement. This result sup-

ports our observation.

4. Conclusions

The photoreaction of photosynthetic reaction center was

investigated by the OHD-TG technique. Here, for the first time

in the literature, we demonstrated that the PBPh fi P+BPh�

charge separation induced considerable structural change in

the protein that was relaxed much slower (28 ls) than the

P+BPh� fi PBPh charge recombination (10 ns). The confor-

mation change was suppressed by the presence of quinone at

the QA site.

Acknowledgments: This work was supported by the grant of JapanSociety for Promotion of Science, by the Hungarian Technology andScience Foundation (JP-12/2006) and NKTH-OTKA (K 67850) andalso by the MEXT Japan to MT (Nos. 15076204, 18205002).

3662 L. Nagy et al. / FEBS Letters 582 (2008) 3657–3662

References

[1] Allen, J.P. and Williams, J.C. (1998) Photosynthetic reactioncenters. FEBS Lett. 438, 5–9.

[2] Kleinfeld, D., Okamura, M.Y. and Feher, G. (1984) Electron-transfer kinetics in photosynthetic reaction centers cooled tocryogenic temperatures in the charge-separated state: evidence forlight-induced structural changes. Biochemistry 23, 5780–5786.

[3] Graige, M.S., Feher, G. and Okamura, M.Y. (1998) Conforma-tional gating of the electron-transfer reaction QA

�QB� – QAQB

�

in bacterial reaction centers of Rhodobacter sphaeroides deter-mined by a driving force assay. Proc. Natl. Acad. Sci. USA 95,11679–11684.

[4] Sebban, P., Maroti, P., Schiffer, M. and Hanson, D. (1995)Electrostatic dominoes: Long distance propagation of mutationaleffects in photosynthetic reaction centers of Rhodobacter capsul-atus. Biochemistry 34, 8390–8397.

[5] Stowell, M.H.B., McPhillips, T.M., Rees, D.C., Soltis, S.M.,Abresh, E. and Feher, G. (1997) Light induced structural changesin photosynthetic reaction center: implications for mechanism ofelectron–proton transfer. Science 276, 812–816.

[6] Fritzsch, G., Koepke, J., Diem, R., Kuglstatter, A. and Baciou, L.(2002) Charge separation induces conformational changes in thephotosynthetic reaction centre of purple bacteria. Acta Crystal-logr. D Biol. Crystallogr. 58, 1660–1663.

[7] Tiede, D.M., Vazquez, J., Cordova, J. and Marone, P.A. (1996)Time resolved electrochromism associated with the formation ofquinone anions in the Rhodobacter sphaeroides R26 reactioncenter. Biochemistry 35, 10763–10775.

[8] Nagy, L., Milano, F., Dorogi, M., Agostano, A., Laczko, G.,Szebenyi, K., Varo, X., Trotta, M. and Maroti, P. (2004) Protein/lipid interaction in the bacterial photosynthetic reaction center:phosphatidylcholine and phosphatidylglycerol modify the freeenergy levels of the quinones. Biochemistry 43, 12913–12923.

[9] Kozlova, M.A., Juhnkec, H.D., Cherepanov, D.A., Lancasterc,C.R.D. and Mulkidjanian, A.Y. (2008) Proton transfer in thephotosynthetic reaction center of Blastochloris viridis. FEBS Lett.582, 238–242.

[10] Hermes, S., Stachnik, J.M., Onidas, D., Remy, A., Hofman, E.and Gerwert, K. (2006) Proton uptake in the reaction centermutant L210DN from Rhodobacter sphaeroides via protonatedwater molecules. Biochemistry 45, 13741–13749.

[11] Remy, A. and Klaus, G. (2003) Coupling of light-induced electrontransfer to proton uptake in photosynthesis. Nat. Struct. Biol. 10,637–644.

[12] Baxter, R.H.G., Ponomarenko, N., Srajer, V., Pahl, R., Moffat,K. and Norris, J.R. (2004) Time-resolved crystallographic studiesof light-induced structural changes in the photosynthetic reactioncenter. Proc. Natl. Acad. Sci. USA 101, 5982–5987.

[13] Baxter, R.H.G., Seagle, B.-L., Ponomarenko, N. and Norris, J.R.(2005) Cryogenic structure of the photosynthetic reaction centerof Blastochloris viridis in the light and dark. Acta Cryst. D61,605–612.

[14] Breton, J. (2004) Absence of large-scale displacement of quinoneQB in bacterial photosynthetic reaction centers. Biochemistry 43,3318–3326.

[15] Breton, J. (2007) Steady state FTIR spectra of the photoreductionof QA and QB in Rhodobacter sphaeroides reaction centers provideevidence against the presence of a proposed transient electronacceptor X between the two quinones. Biochemistry 46, 4459–4465.

[16] Sakakura, M., Yamaguchi, S.Y., Hirota, N. and Terazima, M.(2001) Dynamics of structure and energy of horse carboxymyo-globin after photodissociation of the carbon monoxide. J. Am.Chem. Soc. 123, 4286–4294.

[17] Nishihara, Y., Sakakura, M., Kimura, Y. and Terazima, M.(2004) The escape process of carbon monoxide from myoglobin tosolution at physiological temperature. J. Am. Chem. Soc. 126,11877–11888.

[18] Khan, J.S., Imamoto, Y., Kataoka, M., Tokunaga, F. andTerazima, M. (2006) Time-resolved thermodynamics: heat capac-

ity change of transient species during photo-reaction of PYP. J.Am. Chem. Soc. 128, 1002–1008.

[19] Baden, N. and Terazima, M. (2004) A novel method formeasurement of diffusion coefficients of proteins and DNA insolution. Chem. Phys. Lett. 393, 539–545.

[20] Inoue, K., Sasaki, J., Morisaki, M., Tokunaga, F. and Terazima,M. (2004) Time-resolved detection of the sensory rhodopsin II–transducer interaction. Biophys. J. 87, 2587–2597.

[21] Hazra, P., Inoue, K., Laan, W., Hellingwerf, K.J. and Terazima,M. (2006) Tetramer formation kinetics in the signaling state ofAppA monitored by time-resolved diffusion. Biophys. J. 91, 654–661.

[22] Eitoku, T., Zarate, X., Kozhukh, G.V., Kim, J.I., Song, P.S. andTerazima, M. (2006) Time-resolved detection of conformationalchanges in oat phytochrome A: time-dependent diffusion. Bio-phys. J. 91, 3797–3804.

[23] Nakasone, Y., Ono, T., Ishii, A., Masuda, S. and Terazima, M.(2007) Transient dimerization and conformational change of aBLUF protein: YcgF. J. Am. Chem. Soc. 129, 7028–7035.

[24] Ohmori, H., Nagy, L., Dorogi, M. and Terazima, M. (2008)Charge stabilization in reaction center protein investigated byoptical heterodyne detected transient grating spectroscopy. Eur.Biophys. J. Biophys. Lett. 37, 1167–1174.

[25] Terazima, M. (1999) Optical heterodyne detected transientgrating for studies of photochemical reactions and solutiondynamics. Chem. Phys. Lett. 304, 343–349.

[26] Terazima, M. (1999) Optical heterodyne detected transientgrating for the separation of phase and amplitude gratings. J.Phys. Chem. A 103, 7401–7407.

[27] Rogers, J.A., Fuchs, M., Banet, M.J., Hanselman, J.B., Logan, R.and Nelson, K.A. (1997) Optical system for rapid materialscharacterization with the transient grating technique: applicationto nondestructive evaluation of thin films used in microelectron-ics. Appl. Phys. Lett. 71, 225–227.

[28] Goodno, G.D., Dadusc, G. and Miller, R.J.D. (1998) Ultrafastheterodynedetected transient-grating spectroscopy using diffrac-tive optics. J. Opt. Soc. Am. B 15, 1791–1794.

[29] Tandori, J., Nagy, L., Puskas, A., Droppa, M., Horvath, G. andMaroti, P. (1995) The IleL229fiMet mutation impairs thequinone binding to the QB-pocket in reaction centers of Rhodob-acter sphaeroides. Photos. Res. 45, 135–146.

[30] Okamura, M.Y., Isaacson, R.A. and Feher, G. (1975) Primaryacceptor in bacterial photosynthesis: obligatory role of ubiqui-none in photoactive reaction centers of Rhodopseudomonassphaeroides. Proc. Natl. Acad. Sci. USA 72, 3491–3495.

[31] Terazima, M. (1998) Transient lens spectroscopy in a fast timescale. Isr. J. Chem. 38, 143–157.

[32] Dorogi, M., Balint, Z., Miko, C., Vileno, B., Milas, M., Hernadi,K., Forro, L., Varo, Gy. and Nagy, L. (2006) Stabilization effectof single-walled carbon nanotubes on the functioning of photo-synthetic reaction centers. J. Phys. Chem. B 110, 21473–21479.

[33] Palazzo, G., Mallardi, A., Francia, F., Dezi, M., Venturoli, G.,Pierno, M., Vignatid, E. and Piazzad, R. (2004) Spontaneousemulsification of detergent solubilized reaction center: proteinconformational changes precede droplet growth. Phys. Chem.Chem. Phys. 6, 1439–1445.

[34] Tang, C.-K., Williams, J.C., Taguchi, A.K.W., Allen, J.P. andWoodbury, N.W.W. (1999) P+HA

� charge recombination reac-tion rate constant in Rhodobacter sphaeroides reaction centers isindependent of the P/P+ midpoint potential. Biochemistry 38,8794–8799.

[35] Koepke, J., Krammer, E.-M., Klingen, A.R., Sebban, P.,Ullmann, G.M. and Fritzsch, G. (2007) PH modulates thequinone position in the photosynthetic reaction center fromRhodobacter sphaeroides in the neutral and charge separatedstates. J. Mol. Biol. 371, 396–409.

[36] Ermler, U., Fritzsch, G., Buchanan, S.K. and Michel, H. (1994)Structure of the photosynthetic reaction centre from Rhodobactersphaeroides at 2.65 A resolution: cofactors and protein–cofactorinteractions. Structure 2, 925–936.