-
8/13/2019 Stuart Multipulse spectroscopy on the wild-type and YM210W Bacterial Reaction Centre uncovers a new intermedi
1/5
-
8/13/2019 Stuart Multipulse spectroscopy on the wild-type and YM210W Bacterial Reaction Centre uncovers a new intermedi
2/5
2. Materials and methods
2.1. Sample preparation
Reaction centres were isolated from Rhodobacter sphaeroides
R26.1 according to procedures described in Ref. [21]. The
YM210W mutant of BRC, with a long-lived P* state, is constructed
using site-specific mutagenesis. The detergent purified membrane-bound, antenna deficient, reaction centres were dissolved in
10 mM Tris (pH 8.1) and circulated through a shaking 1-mm sam-
ple cell, to minimize the amount of RCs doubly excited. A 10 mM
ascorbate solution was added to the sample to re-reduce the spe-
cial pair[8,22,23]. Steady state spectra were recorded using a Per-
kin Elmer Lambda 40 UV/VIS spectrograph.
2.2. Femtosecond pumpdumpprobe spectroscopy
Transient laser spectra were recorded on a spectrometer, oper-
ating at a 1 kHz repetition rate. A seed-pulse from a diode-pumped
oscillator (Coherent Vitesse, 800 nm, 80 MHz) is amplified to 2.5 W
by using a Nd:YLF high-power pump-laser (Coherent Evolution-20,
527 nm). The Ti:Sapphire based amplifier (Coherent Legend-USP)
incorporates CPA, chirped pulse amplification, and a stretcher/
compressor combination to deliver sub-50 fs pulses, with a centre
wavelength of 800 nm and a bandwidth of 25 nm at the FWHM.
The pulse to pulse energy is 2.5 mJ (0.025 mJ). For control of the
time of arrival of the pump and dump pulse, two 60-cm long delay
stages (Newport IMS-6000) are placed in the beam paths. In the
single pass mode, this can delay the pulse up till 3.7 ns, reflecting
our time window.
The probe pulse, a white light continuum, is generated by focus-
ing a small fraction of the fundamental 800-nm pulses into a later-
ally rotating sapphire plate. These white light pulses are 300 fs
due to group velocity dispersion, introduced by the sapphire plate.
The pump and dump pulses are generated by a commercial OPA
(Coherent OperA), pumped by the Legend-USP. The pulses are
100 fs long and 10 nm wide at FWHM.Reflective optics is used to steer and overlap all pulses into a
rapidly shaking sample cuvette of 1 mm thickness. The probe pulse
was then spectrally dispersed by a spectrograph (Oriel) onto a
homebuilt photo-diode array detector. This 256-pixel array is read
out and the data is fed into a computer recording and calculating
the varying optical densities.
2.3. Experimental setup
In pumpdumpprobe spectroscopy, the molecule of subject is
excited by the dump pulse. While the system evolves into a differ-
ent state, a second pulse resonant with the stimulated emission of
the evolving excited state is applied, followed by the probe pulse.
In the case of BRC, the pump excites B to B*. This transfers its en-
ergy to P*, which is then dumped by a resonant dump pulse of
940 nm. A schematic PDP picture is shown in Fig. 1(seeSupple-
mentary material). In the wild-type experiment, dump pulses have
been applied at 1 and 2 ps delay after the pump pulse. This corre-
lates with the timescale on which the P
*
state evolves into chargeseparated species P+B or P+H. For the YM210W mutant, where P*
decays bi-exponentially with constants of 12 and 70 ps, dumping
pulses have been applied at 2, 4, 6, 8, and 10 ps delay. Varying
the dump-delay could show a gradual in or decrease in dump effi-
ciency, or could result in different intermediate states.
2.4. Data analysis
Global analysis [24] was simultaneously performed on the PP
and the PDP signals to see differences purely resulting from the
dump pulse. In practice, a direct comparison of the normal
pumpprobe measurements and the pumpdumpprobe mea-
surements is executed. Differences herein can only be introduced
by the dump pulse. New dynamics that arise from the dump pulse
can then be analysed and described. Secondly, target analysis of
the PP and PDP transient signals was performed. Several kinetic
models based on previously gathered data have been applied
[15,19]. The instrument response function is taken into account
using a GAUSSIAN[25]and dump-induced artefacts can be removed
by subtracting dumpprobe (DP) signals from the PDP.
3. Results
3.1. Steady state measurements
Absorption spectra of the two samples shown inFig. 2are in full
agreement with previously reported spectra. [10,19]. The three
band structure arises from the Qy transitions of the various pig-
ments present in the BRC. The blue-most band has been attributed
to the bacteriopheophytins HA and HB, the central band to the
accessory BA and BB bacteriochlorophyll band, and the red-most
band to the low-energy exciton transition of the special pair. The
slight difference in the relative amplitudes of the P, B and H bands
Fig. 1. The pump introduces an excited state (B *), its dynamics evolve unhindered
until the next pulse. The dump pulse will, when chosen to be resonant with the P *,
dump a percentage of this excited state population back to the ground-state. We
propose in this paper that the P*
state exists in an equilibrium with its chargetransfer (CT) state and when it is being dumped, this equilibrium is shortly shifted.
Fig. 2. The steady state spectrum of WT BRC (red) and the YM210W mutant (black).
(For interpretation of the references to colour in this figure legend, the reader isreferred to the web version of this article.)
T.A. Cohen Stuart, R. van Grondelle / Chemical Physics Letters 474 (2009) 352356 353
-
8/13/2019 Stuart Multipulse spectroscopy on the wild-type and YM210W Bacterial Reaction Centre uncovers a new intermedi
3/5
of the WT and YM210W is a result of the mutation modifying the
electronic landscape around P, B and H[8].
3.2. Transient spectroscopy
PDP measurements were performed with excitation pulses cen-
tred at 795 nm, with a 100 nJ pulse to pulse energy. The de-excita-
tion or dump pulse is centred at 940 nm (5 nm width band-passfilter) and also has a 100 nJ energy. The pump and dump pulse
are parallel polarized to the magic angle, 54.7 relative to the
probe. The spotsize on the sample is a 100 lm. Steady state mea-
surements were taken before and after transient measurements
to ensure that no molecular breakdown is instigated by the excita-
tion and dump pulses. Using this power, non-linear effects are not
expected. The transient changes were recorded into four different
data sets. Recording PP, PDP and DP signals simultaneously, allows
a subtraction of the PDP signals from the PP signals. This results in
DDOD spectra that show dump-dependent dynamics. Two spectra
recorded for YM210W are depicted inFig. 3. The black spectrum
with a dump pulse present at 10 ps and the red spectrum without
this dump pulse. In this case the spectra are recorded just 1 ps after
the dump pulse. A dump pulse fired 10 ps after the pump is able to
remove up to 17% of the excited state if detected at 870 nm, the
maximum of the P* excited state. This is also clear from Fig. 3B,
where a time trace at 870 nm is shown during the measurement.
This efficiency appears reasonable, but it has not been studied in
detail. Note that in the spectrum shown inFig. 3A, before and after
the dump pulse, small but significant absorption changes can also
be observed in the region 740810 nm, most likely reflecting a
small amount of species being generated upon dumping. If we lookat the structure of the differences between dumped and not
dumped, we can more clearly observe the effect of the dump pulse
in this system. Time-gated DDOD spectra reveal interesting spec-
tral dynamics in the first ps following the dump pulse (see
Fig. 4). These early spectra are remarkably similar; irrespective of
the dump-delay chosen. In the initial spectra inFig. 4clear features
are observed around 810 nm which indicate the formation of P +, as
reflected by the BMband shift. Interesting is also the first spectrum
of the wild-type. The positive feature around 870 nm reflects the
loss of P* due to the dump. Since in the WT the amount of P* is
much lower, at 300 fs, this feature is overlapped by P* loss. This
behaviour is not seen in the YM210W mutant; however the larger
amount of P* still present at the dump moment may explain that
difference. At longer times, the DDOD spectra converge into a loss
of P+H. This is simply due to the fact that we have removed P* that
would have normally donated its energy to form P+H.
3.3. Global analysis
The species generated by the dump pulse in the mutant is
investigated by global analysis, using a well-established kinetic
model for BRC dynamics[15]with the possibility of an intermedi-
ate state being populated.
Besides the expected species that arise from this analysis (P*, B*,
P+H, etc.), a new state must be introduced in the YM210W mu-
tant. The model is shown in Fig. 5, with a preliminary SADS of
the intermediate species inFig. 6. Using this model, we find that
19% of the P* has been dumped and only a small part of that frac-
tion arrives at the intermediate state. The lifetimes found for thestates present are comparable to earlier literature, except for the
P+H lifetime. This lifetime is obscured by the dump-effect.
In Fig. 6 the intermediate species is compared to P+H. The
Fig. 3. Pumpdump experiment for the YM210W BRC. (A) The dump pulse arrives
10 ps after the pump and results in a 17% loss of the excited state population. (B)
The DDOD time trace at the excited state maximum (870 nm) shows the loss ofpopulation upon dumping.
Fig. 4. Time-gated DDOD spectra for the WT and YM210W as obtained by transient
multi-pulse spectroscopy. At T = 0 the dump pulse arrives. The reaction centre
shows a strong spectral feature around 800 nm reminiscent of the formation P+, andthe structure around 780 nm possibly represents involvement of a pheophytine.
354 T.A. Cohen Stuart, R. van Grondelle/ Chemical Physics Letters 474 (2009) 352356
-
8/13/2019 Stuart Multipulse spectroscopy on the wild-type and YM210W Bacterial Reaction Centre uncovers a new intermedi
4/5
time-gated spectra fromFig. 4show a positive feature arising due
to the pump pulse. The spectrum that arises from the target anal-
ysis has its sign inverted compared to these time-gated spectra.
This is due to the fact that in our analysis method the dump pulse
is interpreted as removing population instead of generating it.
The species generated in the mutant very clearly shows the for-mation of P+ represented by the sharp band shift around 810 nm.
At 870 nm the strong P* loss is present and at lower wavelengths
some population present at 740 is reminiscent of a H band.
Although we have analysed the effect of the dump pulse applied
at 2, 4, 6, 8 and 10 ps, every experiment shows the same interme-
diate species arising shortly after the dump, in comparable quanti-
ties. This is against expectation, and puts forth a different picture
for the interaction between the excited state and the dump pulse.
The lifetime of this intermediate has been found to be comparablewith the P* lifetime (70 ps).
4. Discussion
The main goal of this study was to characterize the state P* and
to identify possible dynamics happening in the P* state while it
develops into P+B. In the wild-type BRC this process is very fast
and multi-exponential (3 and 12 ps). The precise mechanism of
charge separation has not been fully understood, in particular the
idea that an internal P+P charge transfer state exists before charge
separation takes place needs further investigation. Also the under-
lying cause for the multi-exponentiality is not well known. It has
been proposed that the charge separation may somehow be facili-
tated by a favourable protein conformation [26]. Then the fast
phase would reflect favourable conformations while the slow
phase is a manifestation of the unfavourable ones. In that case with
a second dump pulse it must be possible to interrogate the nature
of the excited state as a function of time after excitation and in par-
ticular to investigate the ability to produce useful product states
once P* has been formed. To try and identify these P* dynamics
we have performed multi-pulse spectroscopy. We have used WT-
BRCs that display fast charge separation and BRCs of the
YM210W mutant which is more than a factor of 10 slower. After
initiation of the photoreaction by a pump pulse of 795 nm, the evo-
lution of P* is manipulated by a 940 nm dump pulse. The time-
gated spectra show that both in WT and in YM210 BRCs the dump
pulse successfully removes about 20% of P* excited state popula-
tion and for a large part recovers to the ground state. In addition,in YM210W a new state is formed upon dumping P* that clearly
resembles a charge-separated state involving P+. Note that the
time-gatedDDOD spectra for the WT and the YM210W look very
similar, but we have not been able to retrieve a clear intermediate
for the WT. Apparently it is difficult to separate P+ formation from
P* dumping. In previous pumpdumpprobe studies, hot ground
states are regularly found[25], but in this case the target analysis
does not indicate formation of such a state. Also, although ex-
pected, varying the dump-time does not provide coherent informa-
tion on the charge-separation process. In the mutant, the state
appearing upon dumping shows the same signs of P+ around
805 nm, but now combined with a much larger bleaching of the
P-band at 870 nm. Because the mutant has a higher P* state stabil-
ity, this state is much easier to dump and thus a higher P* loss rel-ative to P+ is expected. TheYM210W has P+ present while P* is
developing into P+B, but significant bleaching of B is not observed.
A possible remaining candidate for the intermediate state is a P+/P
charge separated pair combined with P* loss. From our present data
however we cannot conclude that undoubtedly. In conclusion, by
pumpdumpprobe experiments on the BRC we have identified a
new charge-separated state, possibly P+/P that can only be pro-
duced following P* formation by a first pulse and dumping the ex-
cited state by a second laser pulse. We speculate that this state is
P+/P, the internal charge transfer state of the special pair. The life-
time connected to this intermediate seems to be of the order of the
slow phase of P*, namely around 70 ps. The fact that the slow P* and
P+/P lifetimes appear similar strengthens the belief that P* and P+/
P
co-exist. The dump pulse could shift this equilibrium shortlyand therefore reveal P+/P.
Fig. 5. The model used in global analysis of YM210W. The intermediate does not
contribute to the end product P+Q, and decays independently to the ground state.
The blue arrow depicts the pump pulse, and the red arrows depict the effect of the
dump pulse. Both the slow and the fast fraction of the P * population can be dumped
and they both produce the intermediate. From B* there is a 15% direct P+B
formation, 20% formation of fast P* and the rest, 65%, transfers into the slower P*.
(For interpretation of the references to colour in this figure legend, the reader is
referred to the web version of this article.)
Fig. 6. Species Associated Decay Species (SADS) as obtained from global analysis.
The red spectrum is the intermediate as generated by the dump in the YM210W
mutant. For comparison the steady state spectrum is shown in blue. The P+H
spectrum arising from the target analysis is also shown.
T.A. Cohen Stuart, R. van Grondelle / Chemical Physics Letters 474 (2009) 352356 355
-
8/13/2019 Stuart Multipulse spectroscopy on the wild-type and YM210W Bacterial Reaction Centre uncovers a new intermedi
5/5
Acknowledgements
This research was supported by The Netherlands Organization
for Scientific Research through the Dutch Foundation for Earth
and Life Sciences (Investment Grant No. 834.01.002). T.A.C.S. re-
ceived support from NWO-CW (Grant No. 700.53.307). We grate-
fully acknowledge I.H.M. van Stokkum for extensive
computational data analysis, J. Thieme for technical support andM.R. Jones for supplying the samples.
Appendix A. Supplementary material
Supplementary data associated with this article can be found, in
the online version, atdoi:10.1016/j.cplett.2009.04.081.
References
[1] W. Zinth, J. Wachtveitl, ChemPhysChem 6 (2005) 871.[2] D. Larsen, I.H.M. van Stokkum, M. Vengris, M.A. van der Horst, F.L. De Weerd,
K.J. Hellingwerf, R. van Grondelle, Biophys. J. 87 (2004) 1858.[3] M.H. Vos, J. Breton, J.-L. Martin, J. Phys. Chem. B 101 (1997) 9820.[4] J. Fayer, D.C. Brune, M.S. Davis, A. Forman, L.P. Spaulding, Proc. Natl. Acad. Sci.
USA 12 (1975) 4956.
[5] W. Holzapfel, U. Finkele, D. Oesterhelt, H. Scheer, H.U. Stilz, W. Zinth, Proc.Natl. Acad. Sci. USA 87 (1990) 5168.
[6] P. Hamm, W. Zinth, J. Phys. Chem. (1995) 13537.[7] G. Haran, K. Wynne, C.C. Moser, P.L. Dutton, R.M. Hochstrasser, J. Phys. Chem.
100 (1996) 5562.
[8] M.R. Jones, M. Heer-Dawson, T.A. Mattioli, C.N. Hunter, B. Robert, FEBS Lett.339 (1994) 18.
[9] D.C. Arnett, C.C. Moser, P.L. Dutton, N.F. Scherer, J. Phys. Chem. B 103 (1999)2014.
[10] M.E. Van Brederode, F. van Mourik, I.H.M. Van Stokkum, M.R. Jones, R. vanGrondelle, Proc. Natl. Acad. Sci. USA 96 (1999) 2054.
[11] D. Frolov, M.C. Wakeham, E.G. Andrizhiyevskaya, M.R. Jones, R. Van Grondelle,Biochim. Biophys. Acta 1707 (2005) 189.
[12] X.J. Jordanides, G.D. Scholes, W.A. Shapley, J.R. Reimers, G.R. Fleming, J. Phys.Chem. B 108 (2004) 1753.
[13] V.I. Novoderezhkin, A.G. Yakovlev, R. van Grondelle, V.A. Shuvalov, J. Phys.Chem. B 108 (2004) 7445.
[14] B.A. King, T.B. McAnaney, A. de Winter, S.G. Boxer, J. Phys. Chem. B 104 (2000)8895.
[15] I.H.M. Van Stokkum, L.M.P. Beekman, M.R. Jones, M.E. Van Brederode, R. vanGrondelle, Biochemistry 36 (1997) 11360.
[16] M. Du et al., Proc. Natl. Acad. Sci. USA 89 (1992) 8517.[17] G. Raszewski, W. Saenger, T. Renger, Biophys. J. 88 (2005) 986.[18] H. Lee, Y.-C. Cheng, G.R. Fleming, Science 306 (2007) 1462.[19] M.E. Van Brederode, M.R. Jones, R. van Grondelle, Chem. Phys. Lett. 268 (1997)
143.[20] F. Gai, C. McDonald, P.A. Anfinrud, J. Am. Chem. Soc. 119 (1997) 6201.[21] M.R. Jones, R.W. Visschers, R. van Grondelle, C.N. Hunter, Biochemistry 31
(1992) 4458.[22] L.M.P. Beekman et al., J. Phys. Chem. 100 (1996) 7256.[23] J.A. Potter, P.K. Fyfe, D. Frolov, M. Wakeham, R. van Grondelle, B. Robert, M.R.
Jones, J. Biol. Chem. 280 (2005) 27155.[24] I.H.M. van Stokkum, D.S. Larsen, R. van Grondelle, Biochim. Biophys. Acta 1657
(2004) 82.[25] M. Vengris, I.H.M. van Stokkum, X. He, A.F. Bell, P.J. Tonge, R. van Grondelle, D.
Larsen, J. Phys. Chem. A 108 (2004) 4587.[26] N.P. Pawlowicz, R. van Grondelle, I.H.M. van Stokkum, J. Breton, M.R. Jones,
M.L. Groot, Biophys. J. 95 (2008) 1268.
356 T.A. Cohen Stuart, R. van Grondelle/ Chemical Physics Letters 474 (2009) 352356
http://dx.doi.org/10.1016/j.cplett.2009.04.081http://dx.doi.org/10.1016/j.cplett.2009.04.081