The solid state photo-CIDNP effect
Daviso, E.
Citation
Daviso, E. (2008, November 18). The solid state photo-CIDNP effect. Retrieved from https://hdl.handle.net/1887/13264
Version: Corrected Publisher’s Version
License: Licence agreement concerning inclusion of doctoral thesis in the Institutional Repository of the University of Leiden
Downloaded from: https://hdl.handle.net/1887/13264
Note: To cite this publication please use the final published version (if applicable).
Chapter 4
ELECTROIC STRUCTURE OF THE PRIMARY ELECTRO
DOOR OF RHODOBACTER SPHAEROIDES AT ATOMIC RESOLUTIO
4.1 ITRODUCTIO
The essential steps in photosynthesis, photon absorption and ET, occur in the RC.
Simple purple photosynthetic bacteria possess only a single RC and perform anoxygenic photosynthesis. In RC from the purple bacterium Rb. sphaeroides, P consists of two symmetrically arranged BChl cofactors labeled PL and PM coordinatively bound by His- L168 and His-M202, respectively (Figure. 4.1) (Komiya et al., 1988; Camara-Artigas et al., 2002). The other cofactors are two accessory BChl and two bacteriopheophytins a that are arranged in a nearly C2 symmetry.
Despite the symmetrical arrangement, the electron pathway is entirely unidirectional occurring along the L branch (for review see Hoff and Deisenhofer, 1997).
Using photo-CIDNP MAS NMR (Zysmilich and McDermott, 1994; for review see Matysik and Jeschke, 2003; Daviso et al., 2008a), it has been shown that the symmetry between both cofactors is already broken in the dark electronic ground state (Schulten et al., 2002). The ratio of electron spin densities between the two cofactors PL and PM in the radical cation state has been determined to be ~2:1 using ENDOR (Lendzian et al., 1993;
Huber, 1997) and photo-CIDNP MAS NMR (Prakash et al., 2005a). The present work aims to zoom into the electronic structure of the special pair at the atomic level by mapping the electron densities in the dark ground state and the electron spin densities in the light-induced radical cation state. The latter is achieved by a newly developed time- resolved photo-CIDNP MAS NMR technique that gives access to transient polarization related to Fermi contact spin densities and avoids relaxation processes that may contribute to spectra in steady-state conditions.
4 . 1
Figure 4.1: Cofactors of the reaction center of Rhodobacter sphaeroides R26 [PDB entry 1AIJ]. (A) Donor bacteriochlorophylls (PM and PL) forming the special pair are labeled and highlighted. (B) Bacteriochlorophyll a (BChl) with carbon numbering according to IUPAC number.
Figure 4.2: Kinetics and spin dynamics of electron transfer in quinone-removed RCs of Rb. sphaeroides R26. From the photochemically excited state of the primary donor, P*, an electron is transferred to the primary acceptor Φ, a bacteriopheophytin cofactor. This radical pair is initially in pure singlet state
1(P+•Φ−•). Electron back-transfer leads to the electronic ground-state via two decay channels. In time- resolved experiments, the transient nuclear polarization of the singlet state decay channel of the radical pair can be detected (red arrow).
Kinetics and spin dynamics in high magnetic fields of a single photo-cycle in quinone-removed RCs of Rb. sphaeroides R26 are presented in Figure 4.2. From the photochemically excited primary donor P*, an electron is transferred to the primary acceptor Φ. This light-induced spin-correlated radical pair is born in a pure non-stationary state and oscillates between the S and the T0 by hf coherent evolution. With the discovery of the solid-state photo-CIDNP effect by Zysmilich and McDermott (1994) in frozen and quinone-blocked bacterial RCs of Rb. sphaeroides R26 by 15N MAS-NMR under
continuous illumination (Zysmilich and McDermott, 1994), it was shown that the electron polarization is transferred to nuclei. Due to development of time-resolved photo-CIDNP MAS NMR (Daviso et al., 2008b), TNP that is generated selectively from the S by a spin sorting mechanism similar to the RPM in which diffusion is neglected can now be detected (see Chapter 3 for full explanation). Based on that progress, we applied time- resolved ns-flash photo-CIDNP MAS NMR to obtain the electronic structure of the special pair at atomic resolution. The magnitudes and the orientations of the 13C-electron hyperfine interactions have been estimated.
4.2 MATERIALS AD METHODS
4.2.1 Sample preparation
The RCs from Rb. sphaeroides R26 were isolated by the procedure of Feher and Okamura (1978). The removal of QA has been done by incubating the RCs at a concentration of 0.6 µM in 4% LDAO, 10 mM o-phenanthroline, 10 mM Tris buffer, pH 8.0, for 6 h at 26 °C, followed by washing with 0.5 M NaCl in 10 mM Tris buffer, pH 8.0, containing 0.025% LDAO and 1 mM EDTA (Okamura et al., 1975). Approximately 5 mg of the RC protein complex embedded in LDAO micelles was used for NMR measurements.
4.2.2 MAS MR experiments
NMR experiments were performed with an Avance DRX-200 NMR spectrometer equipped with a MAS probe (Bruker, Karlsruhe, Germany). The sample was loaded into a clear 4-mm sapphire rotor and inserted into the MAS probe and it was frozen slowly at a low spinning frequency of 800 Hz to ensure a homogeneous sample distribution against the rotor wall (Fischer et al., 1992). The light and dark spectra were collected with a spin echo pulse sequence with the CYCLOPS phase cycle of the (π/2) pulse and detection under TPPM carbon-proton decoupling (Bennett et al., 1995) at a temperature of 223 K.
The optimum length of the (π/2) carbon pulse, determined on uniformly 13C labeled tyrosine, is ~4 µs under our experimental conditions using a rf power of ~250 W.
A pulsed nanosecond-flash laser provides sufficient radiation intensity for time resolved photo-CIDNP MAS NMR studies and does not decrease the time-resolution which can be obtained in NMR experiments.
4 . 1
Figure 4.3: Example of a fitted spectrum of RCs (A) Residual spectrum obtained from B-C. (B) Fitting result of spectrum C. (C) Photo-CIDNP MAS NMR spectrum of Rb. sphaeroides R26 in which laser light excitation was followed by NMR detection with a delay of 0 µs.
The laser is operating with repetition rates between 1 and 4 Hz. Using 1064-nm flashes of a Nd:YAG laser (SpectraPhysics Quanta-Ray INDI 40-10, Irvine CA, USA), upon frequency-doubling with a second harmonic generator, 532-nm laser flashes with pulse lengths of 6-8 ns and an energy between 20 to 270 mJ are produced.
Time resolved photo-CIDNP MAS NMR data were acquired using a presaturation pulse sequence to erase polarization and coherence from previous scans as described in Daviso et al. (2008b). The used delay times between light excitation and NMR detection are 0 µs and 400 µs. The rotational frequency for MAS was 8 kHz in all experiments. A cycle delay of 133 ms is used. All the 13C MAS NMR spectra were referenced to the
13COOH response of solid tyrosine•HCl at 172.1 ppm.
4.2.3 Spectral fitting
The fitting of the spectrum collected using time resolved MAS NMR photo- CIDNP of pure RCs of Rb. sphaeroides R26 has been performed using Igor Pro 6.01 (Lake Oswego, Oregon). The peaks are Lorentzian with a FWHH between 40 and 60 Hz as experimentally observed using the 4-ALA labeling pattern of Rb. sphaeroides WT (Daviso et al., 2008b). A typical characteristic fitting result is shown in Figure 4.3.
4.3 RESULTS AD DISCUSSIO
4.3.1 Time resolved ns-flash photo-CID P MAS MR The singlet-born SCRP undergoes two processes:
(i) ISC driven by hfi with nuclei. As the two electron spins are coupled with each other by dipole-dipole and exchange interactions, the symmetry between α and β nuclear spin states is broken by aiso. Further evolution of the spin system leads to TSM, which transfers the electron polarization to nuclei on the nanosecond timescale by the action of the ∆A (Jeschke, 1997).
(ii) Back-ET to the special pair can occur in the S and in the T0 where the different lifetimes of the two decay channels break the symmetry between the S and the T0. The symmetry breaking by ∆A also leads to polarization transfer from the electron spin pair to nuclei. This DD mechanism (Polenova and McDermott, 1999) as well as the TSM contribute to the photo-CIDNP buildup by unbalancing the ratio of α to β nuclear spins in the two decay channels (Prakash et al., 2005a). Both mechanisms require ∆A and have been observed in steady-state experiments with continuous illumination. Under these conditions, polarization generation due to the aiso, which is the basis of the RPM in liquids (Closs and Closs, 1969; Kaptein and Oosterhoff, 1969), is canceled in the ground state, as it leads to contributions with the same magnitude and opposite sign in the two decay channels.
In time-resolved photo-CIDNP MAS NMR experiments, the light pulse for excitation (~8 ns) and the NMR pulse for detection (~4 µs) are much shorter than the lifetime of the T of R26 (~100 µs). Thus transient nuclear polarization of the S can be detected in the ground state, while nuclear spins in the triplet branch are still totally invisible by NMR (Figure 4.2). In transient experiments, nuclear polarization is generated by secular interactions (Azz) without a requirement for nuclear spin level mixing. Thus effects due to isotropic coupling between electrons and nuclei become observable, similar to RPM-based CIDNP in liquids. The sign rules are the same as for cage products from a singlet-born pair in RPM-based CIDNP (Kaptein, 1971b). Therefore, the time-resolved experiment provides information on both the aiso and ∆A, i.e. on local total electron spin densities. Figure 4.4 shows the 13C MAS NMR spectra of RCs of Rb. sphaeroides R26 collected in the dark (Trace A) and in a time-resolved experiment where light excitation was followed by NMR detection with a delay of 0 µs (Trace B) and of 400 µs (Trace C).
4 . 1
Figure 4.4: 13C photo-CIDNP MAS NMR spectra of natural abundance RCs of Rb. sphaeroides R26 obtained (A) in the dark, (B) immediately after the ns-flash and (C) 400 µs after the light flash using 532- nm 8-ns flashes. All the spectra have been collected at a magnetic field of 4.7 T and a temperature of 233 K. The spectrum D shows the simulated transient nuclear polarization occurring in the singlet state of the radical pair and projected out in the diamagnetic ground state.
Trace A shows no signal, demonstrating that all signals in Traces B and C are due to the solid-state photo-CIDNP effect. Trace B presents the transient nuclear polarization occurring in the ground state from the S only. According to both theoretical considerations and numerical simulations, this polarization arising from the S is roughly proportional to the aiso and thus to s spin density on the carbon atoms (Figure 4.5B-B’ and Figure 4.5C- C’). This transient spectrum shows new polarization patterns which have not been observed previously under continuous illumination. For example in the aliphatic region (between 0 and 55 ppm) positive (20-35 ppm) and negative (45-55 ppm) signals appear and in the aromatic region a negative feature at 120-135 ppm emerges. Trace C, representing the full photo-CIDNP buildup after one photo-cycle, is within the noise level
identical to the spectrum obtained in the steady state under continuous illumination (Prakash et al., 2006). Theoretical considerations and numerical simulations suggest that the polarization arising at the end of the photo-cycle, i.e., from the S and T0, is roughly proportional to ∆A2 and thus to the square of p spin density on the carbon atoms (Diller et al., 2007b).
4.3.2 Electron densities in the ground state of P
Electron densities in the electronic ground state of the special pair are obtained by the difference of the 13C chemical shifts between photo-CIDNP MAS NMR data of the special pair and a BChl molecule in solution in acetone (Abraham and Rowan, 1991;
Prakash, 2006). In Figure 4.5, the yellow spheres represent upfield shifts and are caused by an increase of local electron density, while the orange spheres are due to downfield shifts. It appears that the electron density distribution is especially increased in the overlapping region of the cofactors formed by the two pyrrole rings I, while the peripheral regions are emptied. This distribution enhances interaction between the two moieties of the special pair, thus lowering excitation energy and making this pair a better energy sink for excitation transfer from light harvesting pigments.
The prevalence of upfield shifts of the donor indicates that the electron density in the ground state is increased compared to a BChl molecule in acetone solution suggesting that the special pair carries partial negative charge induced by the protein assumable via hydrogen interactions. Hence, our observation will allow to discuss previous results on RCs from WT and mutants of Rb. sphaeroides in more detail (for discussions see Liu et al., 1994; Muegge et al., 1996). The chemical shifts are significantly different between the two parts of the special pair (Figure 4.5A-A’, Tables 4.1 and 4.2) as proposed earlier by Schulten et al. (2002). In particular, electron density is concentrated in the pyrrole ring I of PL. The upfield shifts of up to 14 ppm cannot be explained solely by ring current shifts. On the other hand, several significant downfield shifts in a range of 7-10 ppm occur, especially at the C-5 and C-20 methine carbons of both PL and PM. The asymmetry in favor of PL and the high electron density of both pyrrole rings I may be stabilized by the hydrogen bond interaction with His-L168 to the acetyl group C-31 of PL. The existence of this hydrogen bond (Tiede et al., 1988; Mattioli et al., 1991; Ermler et al., 1994) and its crucial role for the electronic structure has been proposed earlier (Rautter et al., 1995;
Hughes et al., 2001).
4 . 1
4.3.3 Electron spin densities in the radical cation state
Figure 4.5B-B’ shows the map of 13C photo-CIDNP intensities of P in the radical cation state obtained from the fitted spectrum collected using a delay time of 0 µs between light excitation and NMR detection. The size of the spheres is proportional to the magnitude of the aiso (Figure 4.5C-C’). The red spheres represent positive (absorptive) nuclear signals associated with an excess of α nuclear spin states, while blue spheres are associated with the negative (emissive) signals due to excess of β nuclear spin states.
According to RPM sign rules, the absorptive signals are associated with positive and the emissive signals with negative aiso.
IUPAC BChla number
CS (ppm) PM
CS (ppm) PM
this work
CS (ppm) BChla acetone
∆δ=δss−δacetone
(ppm)
Deconvoluted normalized
areas
Computed normalized
aiso
71 22.6 21.6c 1.0 0.4 0.6
81 29.5 29.0c 0.5 0.1 0.6
17 47.3a 47.4 50.4a −3.0 −0.1 −0.2
18 48.0 48.0c 0.0 −0.1 −0.2
8 53.0a 53.0 55.6a −2.6 −0.1 −0.2
5 101.6a 101.5 99.6a 1.9 −0.2 −0.4
20 102.0 96.3a 5.7 −0.2 −0.3
15 106.8a 106.6 109.7a −3.1 −0.3 −0.4
3 130.2a 129.7 137.4a −7.7 −0.1 −0.2
13 131.0a 132.8 130.3a 2.5 −0.2 −0.2
1 148.2a 148.5 150.8a −2.3 0.3 0.2
11 150.3a 149.8 149.4a 0.4 0.1 0.2
16 150.9b 151.0 152.2a −1.2 0.2 0.6
9 158.7b 154.1 158.5a −4.4 0.2 0.5
19 162.5a 163.1 167.1a −4.0 0.4 0.5
6 166.8a 166.5 167.4a −0.9 0.2 0.5
a (Prakash, 2006)
b (Prakash et al., 2006)
c (Abraham and Rowan, 1991)
Table 4.1: Selected NMR 13C chemical shifts (ppm) assigned to PM, experimental electron densities in the ground state, experimental and computed isotropic electron spin densities of the light induced cation radical state. The experimental data were normalized by setting the intensity of the strongest signal to 1, while the aiso were scaled according to the largest value obtained by DFT calculations (see Appendix A).
The size of the spheres has been normalized on the most intense signal arising from C-9 of PL (161 ppm). According to this analysis based on ns-flash photo-CIDNP, the ratio of the total electron spin densities between both cofactors is (68 ±4) : (32 ±4) in favor of PL (Tables 4.1 and 4.2). The electron spin density is localized mainly on the periphery of the special pair, while the center appears to be emptied. Interestingly, electron spin density is shifted towards the primary acceptor in the active L branch.
IUPAC BChla number
CS (ppm) PL
CS (ppm) PL this work
CS (ppm) BChla acetone
∆δ=δss−δacetone
(ppm)
Deconvoluted normalized
areas
Computed normalized
aiso
71 24.1 21.6c 2.5 0.5 0.7
81 31.5 29.0c 2.5 0.4 0.8
17 49.7a 49.7 50.4a −0.7 −0.2 −0.3
18 50.6 48.0c 2.6 −0.2 −0.2
8 55.4a 55.0 55.6a −0.6 −0.1 −0.2
10 99.6a 98.9 99.6a −0.7 −0.3 −0.5
5 103.2 96.3a 6.9 −0.2 −0.5
20 106.1 96.6a 9.5 −0.4 −0.5
3 127.6a 127.4 137.4a −10.0 −0.1 −0.3
13 131.1a 130.8 130.3a 0.5 −0.2 −0.3
4 136.8b 136.4 150.2a −13.8 0.1 0.6
1 143.4a 143.5 150.8a −7.3 0.3 0.3
16 145.3b 145.3 152.2a −6.9 0.7 1.0
11 154.2a 153.3 149.4a 3.9 0.3 0.3
14 158.0a 157.9 160.8a −2.9 0.3 0.5
19 159.7a 160.1 167.1a −7.1 0.8 0.9
9 161.0a 160.8 158.5a 2.3 1.0 0.7
6 164.6a 164.1 167.4a −3.3 0.5 0.8
a (Prakash, 2006)
b (Prakash et al., 2006)
c (Abraham and Rowan, 1991)
Table 4.2: Selected NMR 13C chemical shifts (ppm) assigned to PL, experimental electron densities in the ground state, experimental and computed isotropic electron spin densities of the light induced cation radical state. The experimental data were normalized by setting the intensity of the strongest signal to 1, while the aiso were scaled according to the largest value obtained by DFT calculations (see Appendix A).
4 . 1
Figure 4.5: Structure of the primary donor P [PDB entry 1AIJ]. (A-A’) Two different viewpoints of relative electron densities in the dark state derived from chemical shift differences with respect to monomeric chlorophyll in acetone. (B-B’) Two different viewpoints of normalized fitted photo-CIDNP intensities observed from transient nuclear polarization at a timescale of 0 µs between light excitation and NMR detection in RCs of Rb. sphaeroides R26. (C-C’) Two different viewpoints of normalized aiso
in P•+ derived from density functional theory computations considering PL and PM coordinated to His- L173 and His-M 202, respectively. To compare the experimental data with the theory, both were normalized by setting the cumulative volume of the red and blue spheres to a fixed value for optimal visualization.
Particularly high electron spin density is observed on pyrrole rings II and IV of both PL and PM. This result is in contrast to the electron density maximum in the ground state that is localized in the center, mainly on ring I of PL. The relative shift of electron density may be related to a breaking or weakening of the hydrogen bond between the C-31
acetyl of PL and His-L168 in the radical cation state (Rautter et al., 1995; Hughes et al., 2001). Such electric polarization effect may change the dielectric constant of the protein matrix and may help to prevent the back electron transfer.
Figure 4.5C-C’ presents the map of normalized aiso, which have been obtained by DFT calculations (see Appendix A) considering PL and PM as well as their axial histidine ligands. Side chain conformations of the macrocycles as observed in the crystal structure have been taken into account. As indicated by the blue-red color spheres, the alternating phase pattern found by calculations matches exactly that one observed spectroscopically (Figure 4.5B-B’). The calculated global distribution of electron spin density between PL
and PM, however, is slightly more symmetric than experimentally observed. Our calculations suggest an electron spin density distribution between PL and PM equal to 60:40 compared to the ratio 68:32 found experimentally. This difference may be explained by interactions with the rest of the matrix. In any case, the main part of the asymmetry is an intrinsic property of the special pair complex with its axial histidines and caused by geometric differences conserved by the matrix. Interestingly, the matrix has been shown to be extremely rigid and without structural heterogeneities (Fischer et al., 1992; Shochat et al., 1995). As already assumed by Diller et al. (2007a) on the basis of photo-CIDNP MAS NMR on photosystem II, cofactor-matrix units of functional relevance are conserved in evolution. That principle may explain Shelnutt’s observation on porphins that cofactor geometries are related to their function (Shelnutt et al., 1998). Such interpretation sheds new light on the origin of the break of symmetry of the electron spin density in the special pair. That would imply that the matrix effect is predominantly due to local conformational conservation and neither caused by far-reaching Coulomb interaction of charged amino acids nor by overlap with other cofactors.
Hence, the combination of ns-flash photo-CIDNP MAS NMR and DFT computations allows to resolve the electronic structure of primary donors in photosynthetic RCs at the atomic scale.
