Hydrogen Bond Network around the Semiquinone ... - ACS Publications

Article pubs.acs.org/JPCB

Hydrogen Bond Network around the Semiquinone of the Secondary Quinone Acceptor QB in Bacterial Photosynthetic Reaction Centers Alexander T. Taguchi,#,‡,∥ Patrick J. O’Malley,*,⊥ Colin A. Wraight,†,#,§ and Sergei A. Dikanov*,‡ #

Center for Biophysics and Computational Biology, University of Illinois at UrbanaChampaign, Urbana, Illinois 61801, United States ‡ Department of Veterinary Clinical Medicine, University of Illinois at UrbanaChampaign, Urbana, Illinois 61801, United States ⊥ School of Chemistry, The University of Manchester, Manchester M13 9PL, U.K. § Department of Biochemistry, University of Illinois at UrbanaChampaign, Urbana, Illinois 61801, United States S Supporting Information *

ABSTRACT: By utilizing a combined pulsed EPR and DFT approach, the highresolution structure of the QB site semiquinone (SQB) was determined. The development of such a technique is crucial toward an understanding of proteinbound semiquinones on the structural level, as (i) membrane protein crystallography typically results in low resolution structures, and (ii) obtaining protein crystals in the semiquinone form is rarely feasible. The SQB hydrogen bond network was investigated with Q- (∼34 GHz) and X-band (∼9.7 GHz) pulsed EPR spectroscopy on fully deuterated reactions centers from Rhodobacter sphaeroides. Simulations in the SQB g-tensor reference frame provided the principal values and directions of the H-bond proton hyperfine tensors. Three protons were detected, one with an anisotropic tensor component, T = 4.6 MHz, assigned to the histidine NδH of His-L190, and two others with similar anisotropic constants T = 3.2 and 3.0 MHz assigned to the peptide NpH of Gly-L225 and Ile-L224, respectively. Despite the strong similarity in the peptide couplings, all hyperfine tensors were resolved in the Q-band ENDOR spectra. The Euler angles describing the series of rotations that bring the hyperfine tensors into the SQB g-tensor reference frame were obtained by least-squares fitting of the spectral simulations to the ENDOR data. These Euler angles show the locations of the hydrogen bonded protons with respect to the semiquinone. Our geometry optimized model of SQB used in previous DFT work is in strong agreement with the angular constraints from the spectral simulations, providing the foundation for future joint pulsed EPR and DFT semiquinone structural determinations in other proteins.

■

INTRODUCTION Quinones are ubiquitous to living systems, serving key roles as electron transfer intermediates in photosynthesis and respiration. Despite their importance in biology, the functional influence of the quinone binding sites in proteins remains poorly understood. This is due primarily to a lack of detailed information on the binding site interatomic level interactions undergone by both the oxidized quinone and its one-electron reduced semiquinone form. Proteins are able to tune the chemical properties of quinones to a great extent through hydrogen bond donation to the two carbonyl oxygen atoms by neighboring protein residues. As such, the complete characterization of this interaction is essential to an understanding of how the protein environment regulates quinone function. Some of the most studied quinone binding sites are those present in bacterial reaction centers. The photosynthetic reaction center (RC) catalyzes light-activated charge separation, which is one of the primary events in photosynthesis. In RCs from the purple bacterium, Rhodobacter sphaeroides, an electron is transported across the membrane to two acceptor quinones QA and QB, which are both ubiquinone-10 (UQ10) molecules. Despite their chemical identity, QA and QB are functionally very © XXXX American Chemical Society

different, due to differences in the protein environments of the quinone sites and the different conformations adopted by the 2methoxy groups for the quinones.1−9 QA is a tightly bound prosthetic group which, upon light-induced reduction, forms the anionic semiquinone state (SQA). QA is only capable of one-electron redox chemistry and quickly transfers its electron to the QB site, which is placed symmetrically on the opposite side of a nonheme high spin Fe. The resulting QB semiquinone (SQB) can undergo a second reduction by SQA involving the uptake of two protons.1,2 The product of this proton-coupled electron transfer is quinol, which dissociates from the RC to be replaced by another quinone molecule. The versatility of ubiquinone to adopt a wide range of redox properties is partly achieved by finely tuned interactions with the protein environment. In particular, hydrogen bond donation by neighboring amino acid residues to the O1 and O4 atoms can alter the redox potential. The extent and/or strength of hydrogen bond donation is in general directly related to the Received: February 26, 2015 Revised: April 16, 2015

A

DOI: 10.1021/acs.jpcb.5b03434 J. Phys. Chem. B XXXX, XXX, XXX−XXX

Article

The Journal of Physical Chemistry B quinone electron affinity and hence the redox potential of the quinone. The plethora of X-ray structures for the RC now available have contributed greatly to an understanding of the possible locations of these hydrogen bond interactions for the oxidized ubiquinones, but the absence of proton atom positions leaves the full picture of the hydrogen bond network uncertain. No direct information is obtained from X-ray structures for possible changes in hydrogen bond interactions caused by generation of the semiquinone form on reduction. For no protein, therefore, has the relationship between the structural and functional properties of quinone binding sites been fully characterized or understood. The binding conformations of the oxidized and reduced forms of QA have been studied with great success using highresolution EPR methods. Earlier time-resolved and pulsed EPR works showed very little differences in the QA and SQA binding conformations.10−12 This was supported by a comparison of high frequency ENDOR and PELDOR measurements on samples either frozen in the dark and then illuminated, or frozen under illumination.13 The structure of the SQA hydrogen bond network was then later solved by a complete characterization of the H-bond hyperfine tensors by Q-band ENDOR, which was in good agreement with the quinone conformation found in crystal structures.14 In contrast to QA, the binding conformations of the oxidized and reduced forms of QB have not been as thoroughly investigated. Existing RC crystal structures show several different proposed locations of QB, including binding at both distal and proximal sites relative to the central Fe-(His)4 complex. However, the proximal binding of QB seen in earlier structures is the same as what is observed in the illuminated crystal structures, which likely trap SQB in the active position (Figure 1). In crystal structures of the proximally bound QB, the

Further support of the proximal binding seen in crystal structures for SQB comes from FTIR studies, where no largescale movement of QB was detected upon photoreduction, indicating that at room temperature QB is predominantly bound proximally in both the oxidized and semiquinone states.17 However, low temperature kinetic measurements on the RC showed that a structural change does accompany lightinduced charge separation; primary electron transfer from QA to QB is inactive in RCs frozen in the dark, but was found to be active for RCs frozen under illumination.18 The first interquinone electron transfer is also not electron transfer rate limited, but rather involves a conformational gating step, the origin of which remains unclear.19 An analysis of the electronic structure and bound conformation of SQB by highresolution pulsed EPR methods is an important step toward understanding structural changes that take place within the RC upon charge separation. In this work, we perform Q-band 1H ENDOR (electron nuclear double resonance) and X-band 1H HYSCORE (hyperfine sublevel correlation spectroscopy) on fully deuterated RCs in an H2O solvent to characterize the hyperfine interaction (hfi) between SQB and the protons of its hydrogen bonds from His-L190 NδH, Gly-L225 NpH, and Ile-L224 NpH. The principal values of the 1H hfi tensors and their orientations with respect to the SQB g-tensor reference frame are obtained by least-squares fitting of the simulations to the orientationselective Q-band 1H ENDOR spectra. The resulting Euler angles defining the relative orientations of the tensors are found to be in agreement with our DFT model of SQB reported previously,20 opening the door to future structural studies of the hydrogen bond networks for different quinone sites.

■

EXPERIMENTAL SECTION Sample Preparation. Reaction centers used in this study were isolated from a strain of Rb. sphaeroides expressing RCs with a histidine-tag on the M subunit.21 Full deuteration of the RCs was achieved by successive inoculations of the cells into media where 50%, 90%, and finally 100% of the water and carbon source were substituted with D2O and deuterated succinic acid, respectively (both obtained from Cambridge Isotopes). The cells were also grown in uniformly 15N-labeled media (using 15N ammonium sulfate obtained from Cambridge Isotopes) to avoid possible complications from the deeper electron spin echo envelope modulation of 14N. In order to remove the broad signal arising from the magnetic coupling of the semiquinone with the high spin Fe2+, Fe was biochemically replaced with diamagnetic Zn2+ according to the procedures outlined by Utschig et al. 22 A 3−4 fold excess of ferrocytochrome c (to quickly rereduce the primary donor after charge separation) and an excess of fully deuterated UQ10 (isolated from Rb. sphaeroides strain R26 grown on deuterated media23) were added. SQB was generated by exposing the sample to a single 532 nm Nd:YAG laser pulse, after which the sample was promptly frozen in liquid nitrogen. The absence of SQA contamination in the SQB sample was controlled through the lack of the characteristic SQA 15N cross-peaks in the (+,−) quadrant of the HYSCORE spectra.24 ESEEM and ENDOR Experiments. The instrumentation, pulse sequences, and spectral processing for X-band onedimensional (1D) and two-dimensional (2D) four-pulse ESEEM and HYSCORE25 (π/2−τ−π/2−t1−π−t2−π/2−τ− echo) were as described previously.26 Q-band measurements were carried out on an Oxford CF 935 cryostat equipped with

Figure 1. Interaction of SQB with its potential hydrogen bond donors His-L190 Nδ, Gly-L225 Np, and Ile-L224 Np in the illuminated crystal structure 1DV3. Ser-L223 OH was previously found not to form a strong H-bond with the semiquinone.15 The principal axes of the gtensor are generally assumed to be closely associated with the semiquinone molecular axes labeled as X, Y, and Z. The approximate g-tensor directions are gX along the line connecting the two carbonyl oxygen atoms that carry a large fraction of the spin density, gZ perpendicular to the quinone ring plane, and gY perpendicular to both other principal axes. The principal values of the SQB g-tensor are gX = 2.00626, gY = 2.00527, and gZ = 2.00213.16

possible hydrogen bond donors are the histidine Nδ of HisL190, the peptide nitrogens (Np) from Gly-L225 and Ile-L224, and the Ser-L223 hydroxyl OH. However, we previously concluded Ser-L223 OH is not strongly H-bonded to SQB, leaving the histidine and peptide nitrogens as the remaining hydrogen bond donors (Figure 1).15 B


Article

The Journal of Physical Chemistry B

structure of SQB (Figure S1, Supporting Information) were used for the assignment of the correct hfi tensors for NδH of His-L190, NpH of Gly-L225, and NpH of Ile-L224. In all cases, one of the eight hfi tensor orientations was in excellent agreement with the DFT model, whereas all others failed. The orientations of the hfi tensor principal axes with respect to the g-tensor reference frame are described with Euler angles (α, β, and γ) as defined by EasySpin. For X-band simulations, the excitation bandwidth was assumed fully excitatory, so ideal strong pulses were used. However, in the case of Q-band, a careful determination of the effective excitation bandwidth at each field position is important to accurately simulate the orientation selective spectra. This is done by considering the two major contributions to the excitation bandwidth as outlined by EasySpin: the broadening of the EPR spectrum and the selectivity of the microwave pulses. The EPR broadening was determined by simulating the Q-band continuous wave (CW) spectrum (Figure 2). The

an EN 5107D2 resonator. Pulsed ENDOR spectra were acquired using the Davies (π−t−π/2−τ−π−τ−echo) sequence with a radiofrequency π-pulse inserted during time interval t. The specifics of these experiments are described both in the text and in detail elsewhere.27 Powder 1H ESEEM and ENDOR Spectra. The highresolution pulsed EPR techniques ESEEM and ENDOR make use of the paramagnetic properties of the SQ intermediate to probe its interactions with nearby magnetic nuclei of the protein, the aqueous solvent, and the quinone molecule itself. The isotropic and anisotropic hyperfine interactions with magnetic nuclei such as 1H and 15N can provide detailed information on the electronic structure and geometry of the SQ in its binding pocket.24,28 For a hyperfine coupled 1H nucleus with nuclear spin I = 1/2, there are only two transitions with frequencies να and νβ, corresponding to the two different spin states mS = ±1/2 of the SQ electron spin in a constant applied magnetic field. The values of these frequencies depend on the vector sum of the applied magnetic field and local magnetic field induced at the nucleus by the isotropic and anisotropic hyperfine interactions with the electron spin. For a powder spectrum, the frequencies of the να and νβ transitions span the range between να(β) ⊥ = |ν1H ± A⊥ /2|

and

να(β) || = |ν1H ± A /2|

(1)

which correspond to the perpendicular and parallel orientations of the axial hyperfine tensor. ν1H is the Zeeman frequency of 1H in the applied magnetic field, and A⊥ = |a − T| and A|| = |a + 2T| (where a and T are the isotropic and anisotropic hyperfine coupling constants, respectively). The full axial hfi tensor has principal components (a − T, a − T, a + 2T). The principal values for a rhombic hfi tensor are defined as (a − T(1 + δ), a − T(1 − δ), a + 2T) where δ is the rhombic parameter (which ranges in value from 0 to 1). Powder HYSCORE and ENDOR spectra of I = 1/2 nuclei reveal, in the form of cross-ridges or Pake patterns, respectively, the interdependence of να and νβ at a given orientation. In this work we use X- and Q-band pulsed EPR with microwave frequencies ∼9.7 and ∼34 GHz, respectively. The X-band EPR spectrum of the SQ in frozen solutions is a single line with unresolved g-tensor anisotropy. The spectral width is comparable to the excitation width by microwave pulses, so in the X-band experiment, pulses can be considered as giving a complete excitation of the EPR spectrum. Therefore, at this microwave frequency the ESEEM and ENDOR powder spectra exhibit nuclear frequencies from all orientations of the applied magnetic field relative to the hfi tensor principal axes. On the other hand, at Q-band the principal components of the SQ gtensor are resolved, allowing for orientation selective measurements by exciting only one section of the EPR spectrum at a time. The combined knowledge from nonselective (X-band) and selective (Q-band) methods can provide an accurate estimate of the principal values and directions of the 1H hfi tensors. Spectral Simulations. HYSCORE and ENDOR simulations were performed in Matlab R2013b with EasySpin package v4.5.5.29 Simulations for each hydrogen bonded proton under an approximation of an axial hfi result in eight possible tensor orientations with respect to the g-tensor reference frame that are, in principle, magnetically indistinguishable in the ESEEM and ENDOR experiments. However, the correct hfi tensor can be assigned from knowledge of the SQB structural coordinates and g-tensor orientation. DFT calculations on the minimized

Figure 2. Field swept two-pulse echo (blue), CW EPR spectrum (red), and CW EPR simulation (dashed black) of SQB in fully deuterated RCs in an H2O solvent at Q-band in derivative mode. Yellow circles mark the field positions used for Davies ENDOR measurements. In the case of the field swept two-pulse echo, relaxation rates were found to change with field position, so the CW spectrum was simulated instead. Experimental parameters for the field swept two-pulse echo: π/2-pulse length 160 ns, time between first and second pulses τ = 820 ns, microwave frequency 34.168 GHz, and temperature 80 K. Experimental parameters for the CW spectrum: modulation amplitude 0.2 mT, microwave frequency 34.146 GHz, and temperature 90 K.

excitation bandwidth of the pulses was approximated by multiplying the inverse of the initial microwave π-pulse by two for Davies ENDOR. Additionally, the characteristic suppression of small couplings in Davies ENDOR was taken into account by multiplying the ENDOR simulations with an upside-down Lorentzian function with a width (full width at half-maximum) of 1/(2tp) = 1.6 MHz, where tp is the length of the first microwave π-pulse (320 ns).30 Note that use of a longer pulse length for tp narrows the Lorentzian function, thereby reducing the suppression of weaker couplings. All other parameters were the same as those used in the experiments. DFT Calculations. The DFT calculations were performed as previously described using the B3LYP functional.26 The EPR-II basis set was used for all atoms except Zn where 631g(d) was employed (Figures S1 and S2). All analyses were performed using the ORCA electronic structure program.31

■

RESULTS X-band HYSCORE. We previously reported the X-band proton HYSCORE spectrum for SQB, in which up to three cross-ridges corresponding to hydrogen bonding interactions C


Article


Figure 3. Comparison of the experimental (top) and simulated (bottom) X-band 1H HYSCORE spectra for SQB in fully deuterated RCs shown in contour representations. Spectra were accumulated at τ = 136 (left), 200 (middle), and 400 (right) ns. The dashed lines are defined by |ν1 + ν2| = 2ν1H. The 2H,15N region of the HYSCORE spectrum accumulated at τ = 200 ns is provided in Supporting Information (Figure S3). Experimental parameters: magnetic field 343.8 mT, microwave frequency 9.634 GHz, and temperature 90 K.

were identified.26 However, peaks from nonexchangeable protons congested the spectrum, preventing a straightforward analysis (especially for Ile-L224 NpH). In this work, samples are made using a fully deuterated protein background in an H2O solvent, so that only exchangeable protons contribute to the spectra. HYSCORE spectra were measured at τ = 136, 200, and 400 ns (Figure 3), and show significant peak resolution enhancement compared with the previous work.26 The spectra comprise at least three ridges 1H, 2H, and 3H with extended lineshapes typical of hydrogen bonding interactions. The narrowness of the cross-ridges extending along the antidiagonal suggests axial anisotropic hfi tensors may apply for the interacting protons. The ideal cross-peak shape in this case is an arc-type ridge between the points (να⊥, νβ⊥) and (να||, νβ||) located on the |να ± νβ| = 2ν1H lines (where ν1H is the proton Larmor frequency at the applied magnetic field). The shape of the ridge is described by the general equation να = (Qνβ2 + G)1/2, where Q and G are coefficients that are functions of a, T and ν1H.32 This line shape transforms into a straight line segment in (να)2 vs (νβ)2 coordinates. It should be noted, however, that HYSCORE intensity at points (να⊥, νβ⊥) and (να||, νβ||), corresponding to orientations of the magnetic field along the A⊥ and A|| principal directions of the hfi tensor, is equal to zero and is significantly suppressed at orientations around the principal directions.33 Therefore, in HYSCORE spectra, only the central part of the cross-ridge, which corresponds to orientations of the magnetic field substantially different from the principal directions, will possess observable intensity.33 This means that in real spectra the cross-peak borders do not cross the |ν1 ± ν2| = 2ν1H lines. However, the crossing points (να⊥, νβ⊥) and (να||, νβ||) can still be determined through linear fitting of the observable parts in (ν1)2 vs (ν2)2 coordinates.32,33 Comparison of the individual widths for ridges 1H, 2H, and 3H indicates that the width of 2H is larger than 1H and 3H. This suggests that 2H is the result of an overlap from two closely located, narrower ridges. Linear fitting of the ridges 1H and 3H

Figure 4. Analysis of the SQB HYSCORE spectrum measured at τ = 136 ns in (ν1)2 vs (ν2)2 coordinates. The full view of the spectrum (A) and a zoomed in view of the lower right peaks (B) are presented. Linear regressions are shown for cross-ridges 1H−2H (red) and 2H−3H (green). The curved black line is defined by |ν1 + ν2| = 2ν1H. Fits for the cross-ridges on the opposite side of the diagonal, as well as HYSCORE spectra taken at τ = 200 and 400 ns, are provided in Supporting Information (Figure S4). Experimental parameters: magnetic field 343.8 mT, microwave frequency 9.634 GHz, and temperature 90 K.

D


Article

The Journal of Physical Chemistry B shows that both regression lines cross through 2H located on the opposite side relative to the diagonal, thus confirming the formulated assumption. The two intersection points of the linear fits 1H−2H and 2H−3H with the curved line defined by |ν1 + ν2| = 2ν1H provide the two coordinates (να⊥2, νβ⊥2) and (να||2, νβ||2), corresponding to the perpendicular (a − T) and parallel (a + 2T) principal values of the hfi tensor, respectively. Uncertainty in the assignment of these coordinates to the perpendicular and parallel orientations produces two possible solution sets of a and T, where only the relative signs of the coupling constants are known. Example linear regressions in (ν1)2 vs (ν2)2 coordinates for 1H−2H and 2H−3H are shown in Figure 4 (fits for all cross-ridges of the HYSCORE spectra measured at τ = 136, 200, and 400 ns are provided in Figure S4). The average values from the fits are summarized in Table 1. Spectra simulated with these parameters reproduced well the

spectra in (ν1)2 vs (ν2)2 coordinates. The shifts in these peaks from 2ν1H are well described by the equation34

Table 1. Proton Hyperfine Constants from Linear Regression of the τ = 136, 200, and 400 ns HYSCORE Spectra in (να)2 vs (νβ)2 Coordinates

Q-band ENDOR. While X-band HYSCORE allows for an estimation of the hfi tensor principal values, the principal directions with respect to the g-tensor axes are left unknown. This information can be obtained by performing orientation selective experiments on the semiquinone at higher microwave frequencies. Davies ENDOR measurements were performed on SQB at Q-band (∼34 GHz), a frequency 3−4 times higher than X-band. The resulting spectra reveal the orientation dependence of the proton hfi in the g-tensor reference frame. Q-band Davies ENDOR was acquired at 14 evenly spaced field positions spanning gX, gY, and gZ of the field swept twopulse echo (Figure 2). The resulting orientation selective Davies 1H ENDOR spectra in first derivative mode (Figure 6)

set 1H−2H 2H−3H

a (MHz)a

T (MHz)a

−0.5 −4.4 0.1 −3.0

4.8 4.8 2.9 2.9

± ± ± ±

0.1 0.1 0.2 0.2

± ± ± ±

Δ = 9T 2/16ν1H

(2)

where Δ is the peak shift (in MHz) from 2ν1H. Equation 2 allows for an independent estimate of the anisotropic constant T for 1H−2H and 2H−3H (Table 2). The estimated values for T from this analysis are in good agreement with the values listed in Table 1. Table 2. Anisotropic Hyperfine Constants Determined from the Peak Shift of the Sum Combination Harmonics

0.1 0.1 0.2 0.2

a

Only the relative signs of a and T can be determined from this analysis; the absolute signs are undetermined.

τ-dependence of the intensity distribution for all three crossridges, and indicate the preferred choice of the isotropic constant for use in the ENDOR simulations described later. X-Band 1D Four-Pulse ESEEM. An alternative method for obtaining information on the anisotropic hfi with the exchangeable protons is from 1D four-pulse ESEEM spectra, as shown in Figure 5. These spectra contain peaks

set

Δ (MHz)

T (MHz)

1H−2H 2H−3H

0.9 ± 0.1 0.3 ± 0.1

4.8 ± 0.2 2.8 ± 0.4

Figure 6. Q-band 1H Davies ENDOR spectrum of SQB in fully deuterated RCs. Traces were taken at 14 field positions from 1216.7 mT (bottom trace, gX) to 1219.3 mT (top trace, gZ) in steps of 0.2 mT. The experimental data is shown in blue and is overlaid by the simulations in red. Experimental parameters: π/2-pulse length 160 ns, time between first and second pulses τ = 820 ns, RF π-pulse length 16 μs, microwave frequency 34.167 GHz, and temperature 80 K.

Figure 5. 1D four-pulse ESEEM spectra in stacked representation. The time between the first and second pulses τ was increased from 100 to 328 ns in steps of 12 ns. Experimental parameters: magnetic field 343.8 mT, microwave frequency 9.634 GHz, and temperature 90 K.

corresponding to the sum combination harmonics (να + νβ) of the two basic nuclear frequencies from opposite spin manifolds. The intensities of these peaks depend upon the time τ chosen between the first two pulses of the ESEEM experiment, so the experiment was performed at multiple values of τ so as to not miss any transitions. In Figure 5, the double proton Zeeman frequency (2ν1H = 29.2 MHz) is indicated, along with two well-resolved lines shifted to higher frequencies. These peaks correspond to crossridges 1H−2H and 2H−3H from the analysis of the HYSCORE

are rich with features, despite the deuteration of all nonexchangeable protons in the sample preparation. The major peaks appear symmetrically about ν1H = 51.8−51.9 MHz with average hyperfine splittings ∼5.5 and ∼3.5 MHz. These intense features correspond to orientations of the external magnetic field in-plane with the perpendicular components of the hfi tensor. For the more weakly coupled feature (A⊥ ≈ 3.5 MHz), the peak splits at orientations approaching gZ. This suggests this spectral line is actually the overlap of two different protons, which become resolved near gZ from differences in the E


Article

The Journal of Physical Chemistry B Table 3. SQB Proton Hyperfine Simulation Parameters and Comparison with DFT Calculationsa a (MHz)

T (MHz)

δ

euler angles (α, β, γ)c

His-L190

ENDOR HYSCORE DFTb

−0.6 −0.4 −0.2

4.5 4.7 5.2

0.1 0.1 0.0

[72 ± 14°, −73 ± 4°, −35 ± 14°] − [93°, −71°, −21°]

Gly-L225

ENDOR HYSCORE DFTb

−0.2 −0.2 −1.0

3.2 3.2 4.3

0.1 0.1 0.1

[0 ± 22°, 57 ± 6°, 49 ± 13°] − [21°, 50°, 56°]

Ile-L224

ENDOR HYSCORE DFTb

−0.2 −0.1 −0.1

3.0 3.0 2.5

0.0 0.0 0.1

[0 ± 180°, 64 ± 7°, −34 ± 14°] − [−26°, 64°, −29°]

a Principal values of the rhombic hfi tensor: a − T(1 + δ), a − T(1 − δ), a + 2T; δ ranges from 0 to 1 corresponding to axial and rhombic tensors, respectively. The errors for these coupling constants in the ENDOR and HYSCORE simulations are all ±0.1. bDFT calculations on the minimized structure of SQB (Figure S1). cThe errors for the Euler angles were determined by first deciding upon a least-squares value at which the simulation parameters no longer gave a visually acceptable fit of the ENDOR spectra. The Euler angles were then deviated away from the optimum solution one at a time until this least-squares value was reached, providing a consistent estimate of the error.

and 1H ENDOR in H2O and D2O spectra of SQB in wild-type and L223 mutant RCs, and a consideration of how the spin density distribution should change from removing an Hbond.15 Upon mutation of L223 from serine to alanine (L223SA), we did not observe any convincing differences in the HYSCORE and ENDOR spectra and no change in the 5′ ring methyl proton coupling. This methyl coupling acts as a sensitive probe of the semiquinone spin density distribution.28 As hydrogen bonding significantly alters the spin density distribution of semiquinones, the lack of any difference in the 1 H ENDOR spectra led us to conclude that the Ser-L223 Hbond is not present in the SQB state. The results were further supported by the agreement of our DFT calculations with the experimental ring methyl coupling only when the Ser-L223 Hbond was removed from the model. However, Paddock et al. concluded that the Ser-L223 Hbond is present in the SQB state, based on direct measurements of the H-bond proton hyperfine couplings by Q-band ENDOR on a quintuple mutant RC designed to facilitate direct electron transfer to QB along the B-branch without passing through QA.35 By comparing the ENDOR spectra for quintuple mutant RCs with and without the additional L223SA mutation, they observed a disappearance of a line of low intensity in the L223SA sample attributed to the Ser-L223 H-bond. On the basis of their results, they proposed a mechanism whereby the Ser-L223 H-bond is absent in the neutral quinone state of QB, but then forms in the presence of the SQB radical. With the higher resolution of the H-bond hyperfine couplings afforded by the full deuteration of the protein, we are now able to test the assignments of the H-bond protons made by Paddock et al. HYSCORE has the unique advantage over ENDOR in that cross-ridges are spread out into twodimensions, which resolves anisotropic couplings and makes the assignments of A⊥ and A|| to a single hfi tensor simple. Severe overlap of the Pake patterns in a one-dimensional ENDOR spectrum, on the other hand, can often make this assignment extremely difficult, if not impossible. Nevertheless, Paddock et al. assigned A⊥ and A|| for three unique H-bond proton tensors, one of which was concluded to be from the SerL223 H-bond.35 In Figure 7, the expected locations of the cross-ridges based on the peak assignments made by Paddock et al. are shown as dashed lines in the square-frequency representation. Two of the

orientation dependence of their hfi tensors. The patterns within the range ∼51−53 MHz belong to very weakly coupled exchangeable protons which are not within the focus of the current work. X- and Q-band HYSCORE and ENDOR Simulations. Simulations were performed under the initial assumption that only two hydrogen bonding interactions contribute to the spectra. However, these simulations failed to reproduce the relative intensities of the major features in the Q-band ENDOR spectra with A⊥ ∼5.5 and ∼3.5 MHz, and could not account for the peak splitting of the more weakly coupled feature near gZ. In order to reproduce the ENDOR spectra, a third hydrogen bonding interaction (with A⊥ ≈ 3.5 MHz) needed to be introduced. The Q-band ENDOR spectra were converted into derivative mode to increase the apparent resolution and sensitivity of the simulations. The simulation parameters were first adjusted iteratively by hand, and ultimately fine-tuned using a Nelder− Mead style simplex search algorithm for least-squares minimization. However, significant overlap of the peaks responsible for the split feature with A⊥ ∼ 3.5 MHz, and difficulty in assigning the low intensity parallel components of the hfi tensors (A||), resulted in ambiguity in the simulations. Therefore, the X-band HYSCORE spectra, in which the correlation between the parallel and perpendicular principal values of the hfi tensor is made clear by cross-ridges that are spread out into two-dimensions, were simulated alongside the Q-band ENDOR spectra (Figure 3). The perpendicular components of the hfi tensors (a − T(1 − δ) and a − T(1 + δ)) and the Euler angles (α, β, and γ) are well-defined by the Q-band ENDOR spectra, and were therefore held fixed in all of the simulations. Only the values of the parallel components (a + 2T) were allowed to vary separately in the X-band HYSCORE and Q-band ENDOR simulations. The results of the optimized simulations are shown in Table 3. The simulated values of a and T are in reasonable agreement with the values listed in Tables 1 and 2.

■

DISCUSSION

Serine L223 Hydrogen Bond. Whether or not Ser-L223 OH acts as an H-bond donor in the SQB state remains a matter of debate. Previously, we concluded that this hydrogen bond is not present, based on comparisons of X-band 1H,14N ESEEM F


Article


lie along the O−H bond, β and γ provide a direct means of locating the proton with respect to the carbonyl oxygen. This assumption about the 1H tensor orientation is valid to a good approximation when the dominant contribution to the anisotropic hfi comes from the magnetic dipolar interaction between the unpaired π-electron spin density on the carbonyl oxygen and its hydrogen bonded proton,14 and is confirmed by our DFT calculations which show only a slight deviation from this. In contrast to β and γ, the Euler angle α describes the directions of the perpendicular components of the hfi tensor (a − T(1 − δ) and a − T(1 + δ)), and because of the very low values of δ found for the hydrogen bonding interactions, is not considered in this discussion. The method for determining the proton locations described above relies upon a simple point-dipole approximation for the magnetic interaction between the unpaired π-electron spin density on the carbonyl oxygen and its hydrogen bonded proton. DFT provides a more realistic model of the unpaired electron spin density distribution over the quinone molecule, resulting in a more accurate prediction of the anisotropic hfi tensors.15,26 Our calculations on the minimized structure of SQB were found to be in strong agreement with the ENDOR Euler angles listed in Table 3. The DFT model used here is therefore a good representation of the real SQB conformation in the RC. A comparison of our SQB model overlaying the 1DV3 coordinates is provided in Figure 8 with H-bond lengths indicated.

Figure 7. Overlay of the expected cross-ridges from the Q-band ENDOR analysis of SQB by Paddock et al.35 on our HYSCORE spectrum measured at τ = 136 ns in (ν1)2 vs (ν2)2 coordinates. The full view of the spectrum (A) and a zoomed in view of the lower right peaks (B) are presented. The expected cross-ridge locations for the protons previously assigned as 1 and 1′, 2 and 2′, and 3 and 3′35 are shown as dashed lines in red, magenta, and green, respectively. The curved black line is defined by |ν1 + ν2| = 2ν1H. Experimental parameters: magnetic field 343.8 mT, microwave frequency 9.634 GHz, and temperature 90 K.

dashed lines (red and green) match the linear regression analysis in Figure 4 and Table 1 reasonably well, whereas the proton corresponding to the magenta dashed line does not overlap significantly with the experimentally observed HYSCORE intensity. This magenta dashed line, in fact, corresponds to the proton previously assigned to the Ser-L223 H-bond. This demonstrates that the peak assignments made by Paddock et al. in samples using the quintuple mutant are inconsistent with our spectra obtained from wild-type RCs. The spectral differences attributed to this proton were very small, and could have been a consequence of slight geometric rearrangements that come along with the L223SA mutation. H-Bond Environment of SQB. Both in our previous work15 and the current work, we have eliminated the hydroxyl group of Ser-L223 as a significant hydrogen bond donor to SQB, leaving only His-L190 NδH, Gly-L225 NpH, and Ile-L224 NpH as the remaining options from crystal structures. The assignments of the 1H hfi tensors to these residues in this work were made based on the best fit with our DFT calculations (Table 3). Generally, good agreement between the experimental and calculated values is observed. The Euler angles in Table 3 provide insight into the locations of the hydrogen bonded protons. Specifically, β and γ represent the out-of-plane and in-plane angles, respectively, that describe the orientation of the A|| component with respect to the SQB gtensor reference frame (see Figure S5 for a pictorial representation of the Euler angles). When A|| is assumed to

Figure 8. Comparison of our geometry optimized model of SQB (solid) with the original 1DV3 coordinates (transparent). Hydrogen bond distances are shown in units of Ångstroms.

Comparative Analysis of ENDOR and DFT Work on SQA and SQB. The orientation-selective SQA proton ENDOR spectra of fully deuterated RCs in an H2O solvent were reported and simulated previously.14 In the present work, we report the corresponding ENDOR spectra for SQB, and present DFT calculations for both semiquinones, allowing for a more detailed analysis of the two systems. A comparison of previous ENDOR and HYSCORE work on the SQA H-bond hfi tensors with our DFT calculations is shown in Table 4. Our calculated G


Article

The Journal of Physical Chemistry B Table 4. SQA Experimental Proton Hyperfine Tensors and Comparison with DFT Calculationsa a (MHz)

T (MHz)

δ

euler angles (α, β, γ)

His-M219

ENDORb HYSCOREc DFTd

−1.3 −1.4 0.6

5.2 5.4 5.3

0.0 − 0.0

[−, −50 ± 2°, −13 ± 2°] − [86°, −65°, 0°]

Ala-M260

ENDORb HYSCOREc DFTd

−0.2 −0.2 −1.5

4.6 5.1 4.6

0.0 − 0.0

[−, 63 ± 2°, 16 ± 8°] − [-66°, 50°, 27°]

a Principal values of the rhombic hfi tensor: a − T(1 + δ), a − T(1 − δ), a + 2T; δ ranges from 0 to 1 corresponding to axial and rhombic tensors, respectively. The errors for these coupling constants in the ENDOR and HYSCORE simulations are all ±0.1. bENDOR simulation parameters reported by Flores et al.14 Only the axial Euler angles β and γ were reported, and are shown recalculated for the g-tensor definition used in this work (Figure S2). cHYSCORE parameters reported by Martin et at.26 dDFT calculations on the minimized structure of SQA (Figure S2).

The Euler angles determined from the Q-band ENDOR simulations provide high-resolution information on the hydrogen bonds that can serve as structural constraints for investigating the SQB bound conformation. DFT calculations show that while the g-tensor for SQA is nearly collinear with the molecular axes, the g-tensor for SQB is rotated about 10° around gZ. Given the errors from the ENDOR simulations, we cannot assess whether or not the calculated skewed g-tensor for SQB is supported experimentally at this time. The methods described here have shown themselves to be a valuable alternative to single crystal EPR measurements for obtaining detailed structural information on the H-bond network of quinones, and in particular semiquinones, in a protein binding site. The study opens the way for similar future studies of other quinone sites, and will lead to a greater overall understanding of the influence of the protein environment on quinone function in biology.

Euler angles do not overlap with the error bars from the previous work, although this may be due to a difference in the criterion used to determine the errors (our assigned errors for SQB are significantly larger; see Table 3). In general, reasonable agreement is observed between DFT and the previous work on SQA. One point of interest is the small difference found in the SQA and SQB g-tensor orientations from DFT. The calculations show that the SQA g-tensor is essentially aligned with the semiquinone molecular axes (Figure S2). This is in support of the underlying assumption of a collinear g-tensor orientation in the previous analysis of the SQA ENDOR spectra.14 On the other hand, our calculations for SQB show a small rotation of the g-tensor by about 10° around gZ away from a collinear alignment with the molecular axes (Figure S1). If one corrects the DFT Euler angles in Table 3 to a g-tensor aligned with the molecular axes like in Figure 1, α and β remain essentially the same, but γ becomes −31°, 46°, and −39° for His-L190, GlyL225, and Ile-L224, respectively. These new angles for γ are well-within error of the values estimated from the Q-band ENDOR simulations, so we cannot rule out the possibility that the rotated g-tensor calculated for SQB is due to a limitation of the DFT method or of the model used. Upon closer inspection of the molecular models used to generate the g-tensors (Figures S1 and S2), gX points in the direction of the histidine hydrogen bond, which is the strongest H-bond donor in both cases. It is quite possible that for SQA and SQB, the gX symmetry axis of the g-tensor is influenced by the presence of this histidine. For SQB this hydrogen bond direction is displaced more from the CO axis direction which may have caused the rotated tensor. We are currently planning additional DFT work to investigate this matter further.

■

ASSOCIATED CONTENT

S Supporting Information *

Figure S1, QM Layer of our SQB model showing the g-tensor from DFT calculations; Figure S2, QM Layer of our SQA model showing the g-tensor from DFT calculations; Figure S3, 2H,15N region of the HYSCORE spectrum; Figure S4, linear regressions in (ν1)2 vs (ν2)2 coordinates for τ = 136, 200, and 400 ns HYSCORE spectra; Figure S5, pictorial representation of the Euler angles α, β, and γ. The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acs.jpcb.5b03434.

■

AUTHOR INFORMATION

Corresponding Authors

■

*(S.A.D.) E-mail: [email protected]. Telephone: (217) 3002209. *(P.J.O.) E-mail: patrick.o’[email protected]. Telephone: 00441612004536.

CONCLUSION Using fully deuterated RCs, the spectral resolution of the SQB hydrogen bonding interactions in the X-band HYSCORE spectra was significantly improved upon over previous work,26 allowing for a full characterization of the hydrogen bond network. Orientation selective Q-band ENDOR measurements revealed three proton couplings that exhibit strong hyperfine anisotropy. These proton lines were assigned to His-L190 NδH, Gly-L225 NpH, and Ile-L224 NpH, after eliminating Ser-L223 OH as a potential H-bond donor. The linear regression analysis of the HYSCORE spectra in (ν1)2 vs (ν2)2 coordinates, the peak shift analysis of the four-pulse ESEEM spectra, and the HYSCORE and ENDOR simulations are all generally in good agreement.

Present Address ∥

(A.T.T.) Department of Biochemistry and Molecular Biology, Nippon Medical School, Sendagi, Bunkyo-ku, Tokyo 113− 8602, Japan. Author Contributions

The manuscript was written through contributions of all authors. All authors have given approval to the final version of the manuscript. Funding

This research was supported by the DE-FG02−08ER15960 Grant from Chemical Sciences, Geosciences and Biosciences H


Article


from Photosynthetic Bacteria Rhodobacter sphaeroides Remains in the Same Orientation Upon Light-Induced Reduction. J. Phys. Chem. B 2010, 114, 16894−16901. (14) Flores, M.; Isaacson, R.; Abresch, E.; Calvo, R.; Lubitz, W.; Feher, G. Protein-Cofactor Interactions in Bacterial Reaction Centers from Rhodobacter sphaeroides R-26: II. Geometry of the Hydrogen Bonds to the Primary Quinone QA.‑ by 1H and 2H ENDOR Spectroscopy. Biophys. J. 2007, 92, 671−682. (15) Martin, E.; Baldansuren, A.; Lin, T. J.; Samoilova, R. I.; Wraight, C. A.; Dikanov, S. A.; O’Malley, P. J. Hydrogen Bonding between the QB Site Ubisemiquinone and Ser-L223 in the Bacterial Reaction Center: A Combined Spectroscopic and Computational Perspective. Biochemistry 2012, 51, 9086−9093. (16) Isaacson, R. A.; Lendzian, F.; Abresch, E. C.; Lubitz, W.; Feher, G. Electronic Structure of QA− in Reaction Centers from Rhodobacter sphaeroides. I. Electron Paramagnetic Resonance in Single Crystals. Biophys. J. 1995, 69, 311−322. (17) Nabedryk, E.; Breton, J. Coupling of Electron Transfer to Proton Uptake at the QB Site of the Bacterial Reaction Center: A Perspective from FTIR Difference Spectroscopy. Biochim. Biophys. Acta 2008, 1777, 1229−1248. (18) Kleinfeld, D.; Okamura, M. Y.; Feher, G. Electron-Transfer Kinetics in Photosynthetic Reaction Centers Cooled to Cryogenic Temperatures in the Charge-Separated State: Evidence for LightInduced Structural Changes. Biochemistry 1984, 23, 5780−5786. (19) Graige, M. S.; Feher, G.; Okamura, M. Y. Conformational Gating of the Electron Transfer Reaction QA−.QB → QAQB−. in Bacterial Reaction Centers of Rhodobacter sphaeroides Determined by a Driving Force Assay. Proc. Natl. Acad. Sci. U.S.A. 1998, 95, 11679− 11684. (20) Taguchi, A. T.; O’Malley, P. J.; Wraight, C. A.; Dikanov, S. A. Nuclear Hyperfine and Quadrupole Tensor Characterization of the Nitrogen Hydrogen Bond Donors to the Semiquinone of the QB Site in Bacterial Reaction Centers: A Combined X- and S-band 14,15N ESEEM and DFT Study. J. Phys. Chem. B 2014, 118, 1501−1509. (21) Goldsmith, J. O.; Boxer, S. G. Rapid Isolation of Bacterial Photosynthetic Reaction Center with an Engineered Polyhistidine Tag. Biochim. Biophys. Acta 1996, 1276, 171−175. (22) Utschig, L. M.; Greenfield, S. R.; Tang, J.; Laible, P. D.; Thurnauer, M. C. Influence of Iron-Removal Procedures on Sequential Electron Transfer in Photosynthetic Bacterial Reaction Centers Studied by Transient EPR Spectroscopy. Biochemistry 1997, 36, 8548−8558. (23) Takamiya, K.-I.; Takamiya, A. Light-Induced Reactions of Ubiquinone in Photosynthetic Bacterium, Chromatium D III. Oxidation-Reduction State of Ubiquinone in Intact Cells of Chromatium D. Plant & Cell Physiol. 1969, 10, 363−373. (24) Taguchi, A. T.; O’Malley, P. J.; Wraight, C. A.; Dikanov, S. A. Hyperfine and Nuclear Quadrupole Tensors of Nitrogen Donors in the QA Site of Bacterial Reaction Centers: Correlation of the Histidine Nδ Tensors with Hydrogen Bond Strength. J. Phys. Chem. B 2014, 118, 9225−9237. (25) Höfer, P.; Grupp, A.; Nebenführ, H.; Mehring, M. Hyperfine Sublevel Correlation (HYSCORE) Spectrocopy: A 2D ESR Investigation of the Squaric Acid Radical. Chem. Phys. Lett. 1986, 132, 279−282. (26) Martin, E.; Samoilova, R. I.; Narasimhulu, K. V.; Lin, T. J.; O’Malley, P. J.; Wraight, C. A.; Dikanov, S. A. Hydrogen Bonding and Spin Density Distribution in the QB Semiquinone of Bacterial Reaction Centers and Comparison with the QA Site. J. Am. Chem. Soc. 2011, 133, 5525−5537. (27) Schweiger, A.; Jeschke, G. Principles of Pulse Electron Paramagnetic Resonance. Oxford University Press: Oxford, U.K., 2001. (28) Dikanov, S. A. Resolving Protein-Semiquinone Interactions by Two-Dimensional ESEEM Spectroscopy. Electron Paramagn. Reson. 2013, 23, 103−179. (29) Stoll, S.; Britt, R. D. General and Efficient Simulation of Pulse EPR Spectra. Phys. Chem. Chem. Phys. 2009, 11, 6614−6625.

Division, Office of Basic Energy Sciences, Office of Sciences, US DOE (S.A.D.), NSF Grant MCB-0818121 (C.A.W.), and NCRR/NIH Grants S10-RR15878 and S10-RR025438 for pulsed EPR instrumentation. P.J.O. acknowledges the use of computer resources granted by the EPSRC UK national service for computational chemistry software (NSCCS). A.T.T. gratefully acknowledges support as a NIH trainee of the Molecular Biophysics Training Program (5T32-GM008276). Notes

The authors declare no competing financial interest. † (C.A.W.) We are deeply saddened to report that Professor Colin A. Wraight passed away on July 10, 2014.

■

REFERENCES

(1) Wraight, C. A. Proton and Electron Transfer in the Acceptor Quinone Complex of Photosynthetic Reaction Centers from Rhodobacter Sphaeroides. Front. Biosci. 2004, 9, 309−37. (2) Okamura, M. Y.; Paddock, M. L.; Graige, M. S.; Feher, G. Proton and Electron Transfer in Bacterial Reaction Centers. Biochim. Biophys. Acta 2000, 1458, 148−63. (3) Lancaster, C. R. D. The Role of Electrostatics in ProtonConducting Membrane Protein Complexes. FEBS Lett. 2003, 545, 52−60. (4) Lancaster, C. R. D. Ubiquinone Reduction and Protonation in Photosynthetic Reaction Centres from Rhodopseudomonas viridisXray Structures and their Functional Implications. Biochim. Biophys. Acta 1998, 1365, 143−150. (5) Xu, Q.; Gunner, M. R. Exploring the Energy Profile of the QA− to QB Electron Transfer Reaction in Bacterial Photosynthetic Reaction Centers: pH Dependence of the Conformational Gating Step. Biochemistry 2002, 41, 2694−2701. (6) Stowell, M. H. B.; McPhillips, T. M.; Rees, D. C.; Soltis, S. M.; Abresch, E.; Feher, G. Light-Induced Structural Changes in Photosynthetic Reaction Center: Implications for Mechanism of ElectronProton Transfer. Science 1997, 276, 812−816. (7) Wraight, C. A.; Gunner, M. R. The Acceptor Quinones of Purple Photosynthetic Bacteria - Structure and Spectroscopy. In The Purple Phototrophic Bacteria; Hunter, C. N., Daldal, F., Thurnauer, M. C., Beatty, J. T., Eds.; Springer Science + Business Media B.V.: Amsterdam, 2009; pp 379−405. (8) Taguchi, A. T.; Mattis, A. J.; O’Malley, P. J.; Dikanov, S. A.; Wraight, C. A. Tuning Cofactor Redox Potentials: The 2-Methoxy Dihedral Angle Generates a Redox Potential Difference of >160 mV Between the Primary (QA) and Secondary (QB) Quinones of the Bacterial Photosynthetic Reaction Center. Biochemistry 2013, 52, 7164−7166. (9) De Almeida, W.; Taguchi, A. T.; Dikanov, S. A.; Wraight, C. A.; O’Malley, P. J. The 2-Methoxy Group Orientation Regulates the Redox Potential Difference Between the Primary (QA) and Secondary (QB) Quinones of Type II Bacterial Photosynthetic Reaction Centers. J. Phys. Chem. Lett. 2014, 5, 2506−2509. (10) Borovykh, I. V.; Dzuba, S. A.; Proskuryakov, I. I.; Gast, P.; Hoff, A. J. Light-Induced Structural Changes in Photosynthetic Reaction Centres Studied by ESEEM of Spin-Correlated D+QA‑ Radical Pairs. Biochim. Biophys. Acta 1998, 1363, 182−186. (11) Zech, S. G.; Bittl, R.; Gardiner, A. T.; Lubitz, W. Transient and Pulsed EPR Spectroscopy on the Radical Pair State P+.865QA−. to Study Light-Induced Changes in Bacterial Reaction Centers. Appl. Magn. Reson. 1997, 13, 517−529. (12) Schnegg, A.; Dubinskii, A. A.; Fuchs, M. R.; Grishin, Y. A.; Kirilina, E. P.; Lubitz, W.; Plato, M.; Savitsky, A.; Möbius, K. HighField EPR, ENDOR and ELDOR on Bacterial Photosynthetic Reaction Centers. Appl. Magn. Reson. 2007, 31, 59−98. (13) Flores, M.; Savitsky, A.; Paddock, M. L.; Abresch, E. A.; Dubinskii, A. A.; Okamura, M. Y.; Lubitz, W.; Möbius, K. ElectronNuclear and Electron-Electron Double Resonance Spectroscopies Show that the Primary Quinone Acceptor QA in Reaction Centers I


Article

The Journal of Physical Chemistry B (30) Gemperle, C.; Schweiger, A. Pulsed Electron-Nuclear Double Resonance Methodology. Chem. Rev. 1991, 91, 1481−1505. (31) Neese, F. The ORCA Program System. Wiley Interdiscip. Rev.: Comput. Mol. Sci. 2012, 2, 73−78. (32) Dikanov, S. A.; Bowman, M. K. Cross-Peak Lineshape of TwoDimensional ESEEM Spectra in Disordered S = 1/2, I = 1/2 Spin Systems. J. Magn. Reson. 1995, 116, 125−128. (33) Dikanov, S. A.; Tyryshkin, A. M.; Bowman, M. K. Intensity of Cross-Peaks in Hyscore Spectra of S = 1/2, I = 1/2 Spin Systems. J. Magn. Reson. 2000, 144, 228−242. (34) Reijerse, E. J.; Dikanov, S. A. Electron Spin Echo Envelope Modulation Spectroscopy on Orientationally Disordered Systems: Line Shape Singularities in S=1/2, I=1/2 Spin Systems. J. Chem. Phys. 1991, 95, 836−845. (35) Paddock, M. L.; Flores, M.; Isaacson, R.; Chang, C.; Abresch, E. C.; Okamura, M. Y. ENDOR Spectroscopy Reveals Light Induced Movement of the H-Bond from Ser-L223 Upon Forming the Semiquinone (QB−.) in Reaction Centers from Rhodobacter sphaeroides. Biochemistry 2007, 46, 8234−8243.

J


Hydrogen Bond Network around the Semiquinone ... - ACS Publications

Recommend Documents