g-Tensor Directions in the Protein Structural Frame of

Subscriber access provided by Kaohsiung Medical University

Article

The g-tensor directions in the protein structural frame of hyperthermophilic archaeal reduced Rieske-type ferredoxin explored by C pulsed EPR 13

Alexander T. Taguchi, Daijiro Ohmori, Sergei A. Dikanov, and Toshio Iwasaki Biochemistry, Just Accepted Manuscript • DOI: 10.1021/acs.biochem.8b00438 • Publication Date (Web): 11 Jun 2018 Downloaded from http://pubs.acs.org on June 16, 2018

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 27 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Biochemistry

The g-tensor directions in the protein structural frame of hyperthermophilic archaeal reduced Rieske-type ferredoxin explored by 13C pulsed EPR Alexander T. Taguchia,†,*, Daijiro Ohmorib, Sergei A. Dikanovc, and Toshio Iwasakia,* a

Department of Biochemistry and Molecular Biology, Nippon Medical School, Sendagi, Tokyo 113-8602, Japan b

c

Department of Chemistry, Juntendo University, Inzai-shi, Chiba 270-1695, Japan

Department of Veterinary Clinical Medicine, University of Illinois at Urbana-Champaign, Urbana, Illinois 61801, United States

KEYWORDS: pulsed EPR, g-tensor, [2Fe-2S] cluster, Rieske, ferredoxin. Corresponding contact author: Toshio Iwasaki, PhD. Department of Biochemistry and Molecular Biology, Nippon Medical School 1-1-5 Sendagi, Bunkyo-ku, Tokyo 113-8602, Japan. Tel: +81-3-3822-2131 ext 5237, Fax: +81-3-5685-3054, E-mail: [email protected]

ACS Paragon Plus Environment

1

Biochemistry 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 2 of 27

ABSTRACT. Interpretation of magnetic resonance data in the context of structural and chemical biology requires prior knowledge of the g-tensor directions for paramagnetic metallo-cofactors with respect to the protein structural frame. Access to this information is often limited by the strict requirement of suitable protein crystals for single-crystal electron paramagnetic resonance (EPR) measurements, or the reliance on protons (with ambiguous locations in crystal structures) near the paramagnetic metal site. Here we develop a novel pulsed EPR approach with selective 13

Cβ-cysteine labeling of model [2Fe-2S] proteins to help bypass these problems. Analysis of the

13

Cβ-cysteine hyperfine tensors reproduces the g-tensor of the Pseudomonas putida ISC-like

[2Fe-2S] ferredoxin (FdxB). Its application to the hyperthermophilic archaeal Rieske-type [2Fe2S] ferredoxin (ARF) from Sulfolobus solfataricus, for which the single-crystal EPR approach was not feasible, provided the best-fit gx-, gz-, and gy-tensor directions of the reduced cluster as nearly along Fe-Fe, S-S, and the cluster plane normal, respectively. These approximate principal directions of the reduced ARF g-tensor, solved by

13

C pulsed EPR, are less skewed from the

cluster molecular axes and are largely consistent with those previously determined by singlecrystal EPR for the cytochrome bc1-associated, reduced Rieske [2Fe-2S] center. This suggests the approximate g-tensor directions are conserved across the phylogenetically and functionally divergent Rieske-type [2Fe-2S] proteins.


2

Page 3 of 27 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Biochemistry

INTRODUCTION Proteins containing iron-sulfur clusters and their derivative cofactors play highly versatile roles in biospheres, ranging from electron transfer, substrate activation, detoxification, transcriptional regulation, to production of biofuels such as methane and hydrogen

1-5

. A deeper understanding

of how the protein matrix achieves proper assembly of the iron-sulfur cluster to modulate site specificity and reactivity would facilitate metabolic engineering and application of some of the most difficult chemical reactions in biology using these metalloenzymes. For biological iron-sulfur protein systems, the electronic structure and geometry of a cluster are primarily defined by the immediate ligand coordination environment, which can be further tuned by the pattern of hydrogen bonding with the backbone peptide matrix

1-3, 6-8

. All of these

through-bond interactions can be addressed by spectroscopic approaches, such as pulsed electron and nuclear magnetic resonance (EPR and NMR) techniques 9, 10. EPR in particular is the method of choice for probing the physicochemical parameters that quantitatively define how the unpaired electron spin density delocalizes into the protein framework with respect to the molecular atomic coordinates

2, 3, 10-14

. Currently, the most straightforward experimental approach for iron-sulfur

protein systems with slow electronic relaxation rates is selective 15N, 13C, or 2H isotope labeling of each amino acid type in the region local to the paramagnetic cluster-binding domain, regardless of whether pulsed EPR or NMR methods are employed

10, 12, 13, 15

. For this purpose,

we have engineered a set of Escherichia coli amino acid auxotroph expression host strains for heterologous overproduction of selectively

15

N- and/or

13

C-amino acid isotope labeled iron-

sulfur proteins 15, 16, and have applied these techniques to characterize the electronic structure of the reduced cluster site in the hyperthermophilic archaeal Rieske-type [2Fe-2S](His)2(Cys)2 ferredoxin (ARF) from Sulfolobus solfataricus P1 (DSM 1616) 17-19 as a tractable model protein


3

Biochemistry 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 4 of 27

using two-dimensional, four-pulse electron spin-echo envelope modulation (ESEEM) (also called hyperfine sublevel correlation, HYSCORE) spectroscopy 12, 13, 20. While modern pulsed EPR techniques in conjunction with selective amino acid isotope labeling can resolve and quantify hyperfine interactions between the unpaired electron spin and the magnetic nuclei of the input label, a deeper understanding of the enzymatic mechanism based on the precise interpretation of pulsed EPR data in the context of protein atomic coordinates greatly benefits from knowledge of the g-tensor orientation within the protein structural frame 14, 21

. Some applications of the defined g-tensor directions to biological redox systems include the

ability to analyze the separation distance and relative orientations between two paramagnetic centers by electron-electron dipolar interactions 22-24, to assign dynamic changes in orientation of the paramagnetic center to possible conformational movements from enzyme catalytic action 27

25-

, or in more specialized cases, to locate possible proton positions relative to the paramagnetic

center not readily available from moderate-resolution protein X-ray diffraction studies 28-30. Most ideally, the g-tensor orientation of a protein-bound paramagnetic center would be determined precisely by single-crystal EPR 21. This direct approach is, however, often difficult to apply due to the strict requirement for large single-crystals of proteins with appropriate molecular packing in the lattice. In practice, orientation-selective pulsed EPR techniques, such as orientationselective 1H electron nuclear double resonance (ENDOR) spectroscopy, which probes throughspace hyperfine interactions between the unpaired electron and the nuclei of the input label, can provide geometric information on the nuclei in the g-tensor reference frame

14, 28, 31

. When

combined with the atomic coordinates of the target protein, this information is in certain cases sufficient for a full determination of the g-tensor directions of a paramagnetic center within the


4

Page 5 of 27 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Biochemistry

protein structural frame, so as to describe the mechanistic and functional aspects in the protein atomic coordinates using the high-resolution magnetic resonance data. In this paper, we report a unique pulsed EPR approach for determining the best-fit g-tensor directions using the selectively

13

Cβ(Cys)-labeled ARF in a

15

N-protein background

13

. Because

ARF belongs to the low-potential Rieske-type ferredoxin typically involved in the putative multiple component oxygenase system, this sample contains only two 13Cβ atoms from the Cys42 and Cys61 ligand residues at the non-reduced Fe3+ site of the reduced [2Fe-2S](His)2(Cys)2 cluster12,

13, 18

, and does not possess any disulfide linkage conserved in the high-potential

cytochrome bc1/b6f-associated Rieske proteins

32, 33

. The atomic positions of the input

13

Cβ(Cys)

labels are clearly defined in the crystal structure (Fig. 1), unlike the rather ambiguous locations of the proton atoms near the cluster (not even accessible by NMR due to the paramagnetic relaxation enhancement Our

13

11, 34

) whose coordinates must be derived from molecular calculations.

C pulsed EPR methodology was first validated for the reduced [2Fe-2S](Cys)4 cluster in

the ISC-like [2Fe-2S](Cys)4 ferredoxin (called FdxB)35, 36 from Pseudomonas putida JCM 20004 (formerly Pseudomonas ovalis IAM 1002) where the g-tensor orientation was previously determined by orientation-selective 1H ENDOR spectroscopy 28. This approach was then used to determine the approximate g-tensor directions of the reduced Rieske-type [2Fe-2S](His)2(Cys)2 cluster in ARF, for which single-crystal EPR analysis is not feasible

37

(Fig. 1). This is the first

report where the approximate g-tensor directions are determined for the reduced cluster in a lowpotential Rieske-type ferredoxin, which is evolutionarily and functionally distant from the cytochrome bc1/b6f-associated, high-potential Rieske proteins 18.


5

Biochemistry 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 6 of 27

EXPERIMENTAL PROCEDURES Preliminary single-crystal EPR measurements. The archaeal Rieske-type ferredoxin (ARF) from Sulfolobus solfataricus P1 (DSM 1616) 28

Pseudomonas putida JCM 20004

18

and the ISC-like [2Fe-2S] ferredoxin from

were purified and crystallized as described previously. For

ARF, the crystals belong to the tetragonal space group P43212 (unit cell parameters a = 60.72, c = 83.31 Å), with eight protein molecules per crystallographic unit cell

37

. A single crystal was

placed randomly in a quartz EPR tube (without a crystal holder) and reduced with sodium dithionite in crystallization mother liquor solution under argon gas. Continuous-wave (CW) EPR measurements were performed using a JEOL X-band ES-FA300 spectrometer equipped with an ES-CT470 Heli-Tran cryostat system and a Scientific Instruments digital temperature indicator/controller (model 9700). For FdxB, the crystals belong to the hexagonal space group P6122 (unit cell parameters a = 87.58, c = 73.14 Å), with twelve protein molecules per crystallographic unit cell 35, and the average size of the thin needle-shaped crystals was too small for single-crystal EPR measurements. Preparation of selectively

13C

β (Cys)

labeled [2Fe-2S] protein samples in the

15N-protein

background and pulsed EPR measurements. For selective isotope labeling studies on target metalloenzymes, we have previously reported a collection of C43(DE3)- and BL21(DE3)derived E. coli auxotrophic expression host strains

15, 16

. Selectively

13

Cβ(Cys)-labeled protein

samples were prepared in a 15N protein background as described previously 13, using the cysteine auxotroph YM154 (having a deletion of the cysE gene from the chromosome of E. coli C43(DE3)) 16, CHL-15N (~97 atm%) medium (Chlorella Industry Co. Ltd., Fukuoka, Japan), and L-3-13C,14N(N/A)-cysteine labeled at the

13

Cβ position (Cambridge Isotope Laboratories, Inc.,

Andover, MA). The samples were reduced with sodium dithionite under a continuous flow of dry


6

Page 7 of 27 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Biochemistry

argon gas inside suprasil quartz EPR tubes (Wilmad) prior to sealing, and then rapidly frozen in liquid nitrogen. For a detailed analysis of 13Cβ(Cys) cross-ridges in the HYSCORE spectra of ARF and FdxB, it was essential to make the

13

Cβ(Cys)-labeled protein samples in a

background to prevent significant overlap of the

13

15

N-protein

C cross-ridges with signals from the

14

N

(natural abundance)-protein background 13. The pulsed EPR experiments were carried out at 10-20 K using an X-band Bruker ELEXSYS E580 spectrometer equipped with an Oxford CF 935 cryostat. Two-dimensional, four-pulse hyperfine sublevel correlation spectroscopy (HYSCORE, π/2-τ-π/2-t1-π-t2-π/2-τ-echo) was employed with appropriate phase-cycling schemes to eliminate unwanted features from the experimental echo envelopes. The intensity of the echo after the fourth pulse was measured with t1 and t2 varied and constant τ. The length of a π/2 pulse was nominally 16 ns and a π pulse 32 ns. HYSCORE data were collected in the form of 2D time-domain patterns containing 256×256 points in steps of 32 ns. Spectral processing of ESEEM patterns, including subtraction of the relaxation decay (fitting by 3rd degree polynomials), apodization (Hamming window), zero filling, and fast Fourier transformation (FT), was performed using the Bruker Win-EPR software. 13C

HYSCORE simulations and determination of the approximate g-tensor directions. All

HYSCORE spectral simulations were performed with EasySpin v5.0.9 in Matlab R2014b, as described previously (Fig. 2)

13

. For selectively

13

Cβ(Cys)-labeled [2Fe-2S]

protein samples, the magnetic interaction between the 13Cβ(Cys) nuclei from the input L-cysteine isotope labels and the S=1/2 electron spin from the reduced [2Fe-2S] cluster comes in the form of a hyperfine tensor, which arises from the local magnetic field induced at the nucleus by the electron spin. The hyperfine interaction is orientation dependent, and in the axial approximation is described by the two principal components A⊥ and A|| (corresponding to the perpendicular and


7

Biochemistry 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 8 of 27

parallel tensor orientations, respectively). For systems of lower symmetry the hyperfine tensor becomes rhombic (where A⊥ is expressed as two different perpendicular principal components), and is defined by the full set of rhombic components (AX, AY, and AZ) and Euler angles (α, β, and γ). In the EasySpin v5.0.9 formalism, the Euler angles are the series of rotations that bring the gtensor reference frame with principal components gmin (gx), gint (gy), and gmax (gz) into the hyperfine tensor eigenframe: α is the angle between gmin and the projection of A|| onto the gmin/int plane, β is the angle between A|| and gmax, and γ is the rotation about A|| describing the orientations of the perpendicular components of the rhombic hyperfine tensor (Fig. 3). The complete description of the analysis of the experimental

13

C HYSCORE spectra has been

reported in the Supporting Information of Ref 13. The g-tensor orientation relative to the [2Fe-2S] molecular structural frame can be determined under the assumption of a point-dipole approximation for the hyperfine tensor, which was previously demonstrated to be valid for the magnetic interaction between the

13

Cβ(Cys) nuclear

spin and the electron spin density on Fe3+ of the reduced cluster 13. In this case, A|| is expected to lie along the 13Cβ(Cys)-Fe3+ direction (Fig. 3), and the Euler angles α and β from the 13C hyperfine tensor describe a distribution of allowed g-tensor directions where only the angle between A|| and any individual g-tensor component is known. Euler angle γ is not considered here, due to a lack of an intuition on how the perpendicular components of the rhombic hyperfine tensor should be oriented in the real system. While the g-tensor directions cannot be solved from one axial hyperfine tensor alone, constraints from multiple hyperfine tensors can substantially narrow down the number of possible solutions. When combined with insight from previous g-tensor studies in similar systems

14, 21, 25-27, 31

, it was found that two

13

Cβ(Cys) hyperfine tensors are

sufficient to solve the g-tensor orientations in the present cases.


8

Page 9 of 27 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Biochemistry

RESULTS AND DISCUSSION Protein single-crystal EPR. Both S. solfataricus ARF 12, 37 and P. putida FdxB 28, 35 have been crystallized, allowing for high-resolution X-ray diffraction studies of the protein structures (Fig. 1A, B). However, these protein crystals were found to be unsuitable for single-crystal EPR analysis. For FdxB, the crystals belong to the hexagonal space group P6122 (unit cell parameters a = 87.58, c = 73.14 Å), with twelve protein molecules per crystallographic unit cell 35 (Fig. 1A), and the average size of these thin needle-shaped crystals was too small to obtain EPR spectra of sufficient quality. For ARF, the crystals belong to the tetragonal space group P43212 (unit cell parameters a = 60.72, c = 83.31 Å)

37

, and grew large enough to obtain single-crystal EPR spectra with

measurable changes in the EPR lineshape roughly as a function of the crystal orientation in the spectrometer (Fig. 1C, D). However, the presence of eight protein molecules per crystallographic unit cell

37

(Fig. 1B) has made the EPR spectra of the dithionite-reduced P43212 crystals (Fig.

1D) far too complex to directly trace and assign the g-values to individual paramagnetic sites, thereby inherently preventing a determination of the g-tensor orientation with respect to the crystallographic frame by this method. A significantly smaller number of protein molecules per crystallographic unit cell would allow for a more feasible single-crystal EPR analysis, but such single-crystals are not available at present. In this situation, the “orientation-selective” pulsed EPR approach with randomly oriented (i.e., non-crystalline) protein samples (Fig. 2) is a more practical alternative to determine the best-fit g-tensor directions in the molecular structural frame defined by the corresponding X-ray structural coordinates, as described below.


9

Biochemistry 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Figure 1.

Page 10 of 27

Local structures around the [2Fe-2S] cluster site (depicted by PyMOL v0.99

) (left) and the protein packing in the crystallographic symmetric unit (right) of FdxB (12 protein molecules

28, 35

) (A) and ARF (8 protein molecules

12, 37

) (B). Cys50 and Cys86 bound to the

non-reducible Fe3+ site of the adrenodoxin-like [2Fe-2S](Cys)4 cluster in FdxB (A), and Cys42 and Cys61 bound to the non-reducible Fe3+ site of the Rieske-type [2Fe-2S](His)2(Cys)2 cluster in ARF (B) are shown as dark gray sticks, cysteine β-carbons as dark gray spheres, Fe as brown spheres, and bridging S as dark yellow spheres. Also shown are the preliminary CW X-band EPR spectra at 15 K, 1 mW, of a typical powder sample of dithionite-reduced ARF (C) and a large single-crystal of ARF

37

placed for

arbitrary orientation and rotated with respect to the magnetic field roughly by steps of ~45° (D).

Orientation-selective 13C HYSCORE of reduced FdxB and ARF. In an orientation-selective pulsed EPR experiment, the hyperfine interactions of the paramagnetic center with the surrounding nuclei are measured as a function of the g-tensor orientation 14. This is achieved by


10

Page 11 of 27 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Biochemistry

taking measurements at different magnetic fields, thereby exciting only a subset of g-values out of the full g-tensor at a time

14

. We have recently reported the orientation-selective

13

C

HYSCORE analyses of selectively 13Cβ-cysteine labeled, reduced FdxB and ARF samples in the 15

N-protein background 13, prepared by using E. coli cysteine auxotrophic expression host strain

YM154

16

(Fig. 2). In short, for the reduced [2Fe-2S] cluster in FdxB with a complete cysteinyl

coordination (gmax (g||) = 2.020, gint = 1.936, gmin = 1.934 28), 13C HYSCORE spectra showed four 13

Cβ(Cys) cross-features resolved by optimized EasySpin simulations, with two anisotropically

elongated cross-ridges from the Cys86 (1c) and Cys50 (2c) ligands coordinated at the nonreduced Fe3+ site, and two additional components at low contour levels with weak anisotropy (3c and 4c) from Cys41 and Cys47 at the Fe2+ site 13 (Fig. 2A and Fig. S1A). At the Fe3+ site of the reduced cluster, the magnitude of the electron spin population at Cys50

13

Cβ is significantly

smaller than that at Cys86 13Cβ in FdxB 13. The 13C HYSCORE spectra of reduced ARF (gz = 2.022, gy = 1.901, gx = 1.804) showed only elongated

13

Cβ(Cys) cross-ridges that are shifted off the antidiagonal dashed line defined by να +

νβ = 2ν13C (Fig. 2B and Fig. S1B); these cross-features 1c and 2c, characteristic of

13

C tensors

with significant hyperfine anisotropy, are similar to those observed for the cross-ridges 1c and 2c in the FdxB spectra (Fig. 2A and Fig. S1A), but have unresolved individual contributions from the Cys42 and Cys61 ligands coordinated at the Fe3+ site (Fig. 2B and Fig. S1B), owing to the relatively similar magnitudes of the electron spin populations at Cys42

13

Cβ and Cys61

13

Cβ in

reduced ARF 13. The 2D pulsed EPR spectroscopy applied here has clearly resolved the cross-ridges from nuclei with different hyperfine anisotropies located at the Fe3+ or Fe2+ site, which would otherwise be masked by more intense ridges in the 1D spectra. Simulation analysis of these 13C


11

Biochemistry 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 12 of 27

HYSCORE spectra collected at multiple magnetic field positions spanning the full g-tensor for each protein therefore provides the principal values of the 13Cβ(Cys) tensors and Euler angles that define the directions of the hyperfine principal axes in the g-tensor reference frame, where ever possible.

Figure 2.

13

Cβ(Cys) HYSCORE spectra of FdxB measured at g|| (left), gint (middle left), and g⊥ (middle

right) (A), and those of ARF measured at gz (gmax) (left), gy (gint) (middle left), and gx (gmin) (middle right) (B), shown in 3D-stacked (top) and contour (bottom) representations, respectively, and schematic diagrams (right) of the powder 13C HYSCORE spectra of 13Cβ-Cys labeled, reduced FdxB (A) and ARF (B) in the 15N-protein background

13

for the simulated cross-ridge lineshapes (the schematic figures are

not drawn to scale, and the cross-ridge curvature is exaggerated for better clarity). The spectral simulations are shown in red overlaying the experimental 13C HYSCORE spectra in the (++) quadrant. In the contour representations, the dashed lines along the antidiagonal are defined by να + νβ = 2ν13C. In the optimized EasySpin simulations of the FdxB spectra (A), two

13

Cβ(Cys) tensors, 3c (a=2.8±0.1 MHz;


12

Page 13 of 27 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Biochemistry

T=0.5±0.1 MHz; δ=0; Euler angles α=0°, β=50°, γ=0°) and 4c (a=1.8±0.2 MHz; T=0.5±0.2 MHz; δ=0; Euler angles α=0°, β=-60°, γ=0°) have simple cross-peak shapes lying along the dashed antidiagonal line defined by να + νβ = 2ν13C, and two anisotropic

13

Cβ(Cys) tensors of the similar low rhombicity, 1c

(a=1.1±0.2 MHz; T=1.3±0.1 MHz; δ=0.2±0.1; Euler angles α=40±180°, β=81±12°, γ=33±62°) and 2c (a=-0.2±0.2 MHz; T=1.2±0.1 MHz; δ=0.2±0.2; Euler angles α=-30±180°, β=-19±11°, γ=82±180°) make up the more intensive “arc-shaped” cross-ridges shifted off of the antidiagonal in the (++) quadrant (right), where the principal values of the rhombic hyperfine tensor are [a – T(1 + δ), a – T(1 – δ), a + 2T] and the rhombicity parameter δ ranges from 0 to 1 corresponding to axial and rhombic tensors, respectively. 2C has a very small isotropic constant, thus being represented with only a single cross-ridge symmetric about the diagonal in the FdxB spectra, and points of cross-ridge overlap producing the 1C+2C peaks in the stacked spectra are indicated with orange circles (right). For ARF (B), the overall experimental spectra showing the intense “arc-shaped”

13

C cross-ridges off the antidiagonal line were

successfully reproduced by EasySpin simulations with two anisotropic

13

Cβ(Cys) tensors of similar low

rhombicities, 1c (a=0.8±0.2 MHz; T=1.3±0.1 MHz; δ=0.1±0.1; Euler angles α=-35±10°, β=-59±9°, γ=89±33°) and 2c (a=0.4±0.2 MHz; T=1.3±0.1 MHz; δ=0.2±0.1; Euler angles α=25±22°, β=53±6°, γ=0±14°), respectively, where the heavy overlap of 1C and 2C results in peaks (orange circles) that can only be identified as 1C+2C (right). Experimental parameters: magnetic fields = 341.5 mT (g||), 347.0 mT (gint), and 355.7 mT (g⊥), microwave frequency = 9.646 GHz, and temperature = 20 K for the FdxB spectra (A); magnetic fields = 344.0 mT (gz), 364.0 mT (gy), and 386.2 mT (gx), microwave frequency = 9.696 GHz, and temperature = 10 K for the ARF spectra (B). Data taken from Taguchi et al.13, and the details of the 13Cβ(Cys) cross-ridge lineshapes based on the simulations of the experimental 13C spectra are shown in Fig. S1 in the Supporting Information.


13

Biochemistry 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 14 of 27

Figure 3. Schematic model illustrating the direction of the parallel component (A||) of the

13

Cβ(Cys)

hyperfine tensor defined by the Euler angles α, β, and γ relative to the g-tensor reference frame of a protein-bound, reduced [2Fe-2S] cluster, in which α is the angle in the gmin-gint plane and γ is the rotation about A||. In this point-dipole model for the magnetic interaction between the 13Cβ(Cys) nucleus and the Fe3+ spin density, the hyperfine tensor is oriented with A|| along the 13Cβ(Cys)-Fe3+ direction (red arrow). For the reduced FdxB system, while the Euler angle α could not be determined with accuracy because the gtensor is strongly axial 36, the Euler angle β from the 13C HYSCORE spectral analysis is sufficient to give information on the spatial orientations of the largest parallel components (A||) of the

13

Cβ(Cys) hyperfine

tensors at the Fe3+ site. For the reduced ARF system, the precise g-tensor direction is not known, but previous work on the related Rieske-type protein system suggest that it should be similar to the idealized orientation shown above, with gmin (gx), gmax (gz), and gint (gy) along the molecular structural axes defined by the Fe atoms (purple spheres), the S atoms (white spheres), and the axis perpendicular to the cluster plane, respectively

21, 31

. Only the atoms of the [2Fe-2S] cluster core and one of two 13Cβ(Cys) (dark gray

sphere) at the Fe3+ site are shown in the figure for clarity.

The best-fit g-tensor directions of reduced FdxB defined by When orientation-selective

1

13C

β (Cys)

hyperfine tensors.

H ENDOR is performed at the magnetic field position

corresponding to g|| (gmax), only the select components of the proton hyperfine interactions at or nearly parallel to the g||-axis contribute to the resulting pulsed EPR spectrum

31

. Previously we


14

Page 15 of 27 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Biochemistry

have applied orientation-selective 1H ENDOR spectroscopy to determine the best-fit g|| (gmax)tensor direction of the reduced [2Fe-2S](Cys)4 cluster of FdxB

28

. Although electron density of

proton atoms in FdxB are not visible in the X-ray crystallographic structure, and consequently their locations remain hypothetical, the average g|| (gmax) direction is shown to be skewed from the molecular axes of the reduced cluster, forming an angle of ~27±4° with the normal of the [2Fe-2S] cluster plane

28

. This orientation-selective 1H ENDOR study also assigned the

innermost Fe2 site as the non-reduced iron retaining the Fe3+ valence with a high positive spin population, and the outermost Fe1 site as the reduced iron with a low negative spin density 28, 31. In the point-dipole approximation for the hyperfine interaction between the

13

Cβ(Cys) nuclear

spin and the electron spin density on the non-reduced Fe3+ site of the protein-bound, reduced [2Fe-2S] cluster (Fig. 3), which was previously demonstrated to be valid component (A||) of the

13

13

Cβ(Cys) hyperfine tensor is expected to lie along the

direction (indicated by the

13

, the largest 13

Cβ(Cys)-Fe3+

Cβ(Cys)-Fe3+ red arrow), and the Euler angles α and β from the

13

Cβ(Cys) hyperfine tensor are the primary source of information to describe a distribution of the

allowed directions for the g-tensor axes where only the angle between A|| and any individual gtensor axis is known. While Euler angle α could not be determined with accuracy in the reduced FdxB system (the simulated angles had errors of ±180°, because the g-tensor is strongly axial 36), the simulated Euler angle β from the experimental

13

C HYSCORE spectra (Fig. 2A and Fig.

S1A) gives information on the spatial orientations of the largest parallel components (A||) of the Cβ(Cys) hyperfine tensors at the Fe3+ site of FdxB as β = -19±11° for Cys50 and β = 81±12° for

13

Cys86 13. The range of angles for β agrees well with the expected values based on the best-fit g||tensor directions of the FdxB system previously determined by orientation-selective 1H ENDOR 28

(β = -11° for Cys50 and β = 82° for Cys86), such that the average g||-tensor direction points


15

Biochemistry 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

roughly towards Cys50

13

Page 16 of 27

Cβ from the Fe3+ site and is skewed from the normal of the [2Fe-2S]

plane (Fig. 4). As shown in Fig. 2 and Fig. S1, the HYSCORE signals from the two 13Cβ(Cys) around Fe3+ have the largest anisotropic hyperfine coupling constants (T) and these cross-ridge lines dominate the experimental

13

C spectra

13

. Although our previous

13

C HYSCORE analysis has accounted for

the experimentally determined hyperfine tensors reflecting interaction of all four 13Cβ(Cys) nuclei at both Fe3+ and Fe2+ sites, the Fe2+ coordinated cysteines (Cys41 and Cys47) are not considered in this study, because the Euler angles for these 13C tensors (3C and 4C) could not be accurately estimated by the spectral simulations, as described previously

13

(see Fig. S1 in the Supporting

Information).

Figure 4. The possible g-tensor directions of the reduced [2Fe-2S](Cys)4 cluster system of FdxB (gmax = 2.020, gint = 1.936, gmin = 1.934)

28

calculated from the

13

Cβ(Cys) Euler angles under the point-dipole

approximation for the magnetic interaction between the 13Cβ(Cys) nuclear spin and the electron spin density on Fe3+ (indicated by 13Cβ(Cys)-Fe3+ dashed lines; see Fig. 3), using the simulated 13Cβ(Cys) hyperfine tensors as constraints. The possible g|| (gmax) directions of FdxB from the present 13Cβ(Cys) HYSCORE simulation analysis, which point roughly toward Cys50 13Cβ(Cys), are shown as a red bundle, with the best-fit shown with an extended red rod, in wall-eye stereoview representations, demonstrating the successful


16

Page 17 of 27 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Biochemistry

reproduction of the average g|| direction of this system previously determined by 1H ENDOR (green rod) 28

. Cys50/86 13Cβ, Fe, and S atoms are indicated as dark gray, purple and white spheres, respectively.

Abdalla et al.

38

have reported the best-fit g|| (gmax) direction of the reduced form of a closely

related [2Fe-2S](Cys)4 ferredoxin called PuxB, determined by orientation-selective 1H ENDOR with different models of the spin density distribution by considering the spin density delocalized on the Fe ions, bridging sulfurs, and cysteine sulfurs. These authors concluded that the optimal g|| (gmax) direction did not vary significantly (within ±3 degrees of the mean value), regardless of a purely ionic model assuming the entire localization of the spin densities on the Fe ions, or a more sophisticated model with experimentally determined spin density delocalization parameters accounting for the (rescaled) spin density distribution over the Fe and S atoms from density functional theory (DFT) calculations

38

. Similarly, for FdxB, the expected skewing of the

13

Cβ(Cys) hyperfine tensors with a model accounting for this DFT-estimated spin delocalization

onto the cysteine sulfur ligands

38

is 10 degrees or less. Also, the estimated contribution to this

skewing of the Fe2+ site spin density is also less than 5 degrees when typical values reported for the Fe2+ spin density in [2Fe-2S](His)n(Cys)4-n proteins (n=0,2) 28, 31, 38, 39 are used in point dipole calculations assuming localization of the spin densities on the Fe ions. Because these contributions are all less than the errors of the calculated principal g-tensor directions reported in Table 1, we used the simple point dipole approximation for the magnetic interaction between the 13

Cβ(Cys) nucleus and the Fe3+ spin density in this study (Fig. 3). The g-tensor directions of the

protein-bound, reduced [2Fe-2S](Cys)4 clusters seem to be primarily determined by the ironcysteine bond interactions and the conformation of the cysteine ligands relative to the [2Fe-2S]


17

Biochemistry 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 18 of 27

core plane and are not grossly influenced by other protein structural factors surrounding the [2Fe-2S](Cys)4 cluster site. Orientation of the ARF g-tensor within the protein structural frame. Having successfully reproduced the approximate g||-tensor direction of the reduced FdxB system with the

13

Cβ(Cys)

HYSCORE approach, the same methodology was applied to the reduced Rieske-type [2Fe2S](His)2(Cys)2 cluster in ARF (Fig. 2B and Fig. S1B), for which the g-tensor directions are not known. In the previous study, the 13Cβ(Cys) hyperfine tensors of the two cysteine ligands (Cys42 and Cys61) at the non-reduced Fe3+ site of the reduced ARF cluster were identified and characterized, but their residue-specific assignments could not be determined because of the pseudo-two-fold symmetrical locations of the Rieske-type [2Fe-2S] cluster plane

13

13

Cβ(Cys42) and

13

Cβ(Cys61) atoms relative to the

(Fig. 1B). Therefore, two different g-tensor distributions

were calculated here for the two possible assignments of the Euler angles α = 25±22°, β = 53±6° and α = -35±10°, β = -59±9° (see Fig. 3) to Cys42/Cys61 (and vice versa). Although a self-consistent, unique determination of the approximate g-tensor directions is inherently not possible for the ARF system, the two resulting g-tensor distributions for the two possible assignments of the Euler angles to Cys42 and Cys61 are serendipitously well within error of each other (Fig. 5B, C). In the case of the residue-specific assignment of α = 25±22° and β = 53±6° to Cys42, and α = -35±10° and β = -59±9° to Cys61, we found that only one set of gtensors (Fig. 5B) is consistent with previous single-crystal EPR ENDOR

31

21

and orientation-selective 1H

studies on the cytochrome bc1-associated Rieske protein system, as well as EPR

studies on the cytochrome bc1 complex in oriented membrane multilayer fragments that indicated gx to point roughly along the molecular structural axis defined by the Fe-Fe direction of the reduced Rieske cluster

25-27

. An analysis of the reduced ARF system gave almost the same g-


18

Page 19 of 27 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Biochemistry

tensor orientation for the alternate assignment of α = 25±22° and β = 53±6° to Cys61, and α = 35±10° and β = -59±9° to Cys42 (Fig. 5C). Notably, the calculated g-tensors are insensitive to these two scenarios due to the similarity in magnitude of the

13

Cβ(Cys) Euler angles associated

with the pseudo-two-fold symmetry of the two cysteine ligand locations relative to the ARF cluster plane (see Fig. 1B). Both show strongly collinear g-tensor directions with respect to the molecular structural frame of the Rieske-type [2Fe-2S] cluster core (Fig. 5B,C). The extent of this skewing from the ARF cluster plane is less than that previously determined for the cytochrome bc1-associated, reduced Rieske [2Fe-2S] protein system by other methods 21, 31, with the gx, gz, and gy components of the reduced ARF system roughly along the molecular structural axes defined by the Fe-Fe direction, the S-S direction, and the axis perpendicular to the cluster plane, respectively (Table 1 and Fig. 5D), suggesting approximately conserved principal axes of the g-tensors for protein-bound, reduced Rieske [2Fe-2S] cluster systems in biology. Interestingly, the recent orientation-selective ENDOR study of the reduced mitoNEET-type [2Fe-2S](His)1(Cys)3 cluster system is reported to have the principal axes of the g-tensors similar to these reduced Rieske [2Fe-2S] cluster systems

40

. The best-fit direction cosines relating the

approximate g-tensor directions to the molecular structural frame of ARF, the skew (angular deviation) of each g-tensor component from its nearest molecular axis, and the maximum skew of the distribution of possible g-tensors from the nearest molecular axis are given in Table 1. One may postulate that the extent of g-tensor skewing in the protein-bound, reduced [2Fe-2S] clusters is correlated primarily to the conformational variations in the cysteine ligands at the nonreduced Fe3+ site of the reduced [2Fe-2S](His)n(Cys)4-n cluster (n=0,2) (Fig. 1), the local structural distortions around the Fe2+ site environment, and possibly to a lesser extent the pattern of peptide hydrogen bonding with the bridging and terminal sulfur atoms. By this logic the local


19

Biochemistry 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 20 of 27

structure and molecular interactions around the Fe3+ site environment of the reduced cluster would appear to be highly symmetrical 13 for the ARF spin system.

Figure 5. Determination of the allowed (colored bundles) and best-fit (long colored rods) g-tensor directions of the reduced Rieske-type [2Fe-2S] cluster system (gz = 2.022, gy = 1.901, gx = 1.804) relative to the protein structural frame of ARF in wall-eye stereoview representations, based on the simulation of the two axial

13

Cβ(Cys) hyperfine tensors (see Fig. 2B). In the point-dipole model for the magnetic

interaction between the 13Cβ(Cys) nuclear spin and the electron spin density on the non-reducible Fe3+ site of the reduced cluster, the parallel component (A||) of the 13Cβ(Cys) hyperfine tensor is expected to lie along the 13Cβ(Cys)-Fe3+ direction (indicated by the 13Cβ(Cys)-Fe3+ dashed lines), and the Euler angles α and β from the

13

Cβ(Cys) hyperfine tensor describe a distribution of the allowed g-tensor directions where only the

angle between A|| and any individual g-tensor component is known (see Fig. 3). For the individual case assignments of α = 25±22° and β = 53±6° to Cys42 13Cβ, and α = -35±10° and β = -59±9° to Cys61 13Cβ


20

Page 21 of 27 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Biochemistry

13

, the resulting set of all possible g-tensor orientations can be represented as a series of rings on a unit

sphere (A). In this representation, each point on the ring feature is one possible direction of a particular gtensor component extending from the non-reducible Fe3+ site of the reduced [2Fe-2S] cluster. The only allowed g-tensor orientations are those corresponding to the intersection points where the overlap of the corresponding magenta, blue, and red rings is found for α = 25±22° and β = 53±6° to Cys42 13Cβ (A, top), and for α = -35±10° and β = -59±9° to Cys61 13Cβ (A, bottom). For the assignment of α = 25±22° and β = 53±6° to Cys42, and α = -35±10° and β = -59±9° to Cys61 13, the number of possible solutions is reduced to a single localized bundle of g-tensors when additional constraints from previous g-tensor studies on Rieske protein systems 21, 25-27, 31 are taken into account (B). The g-tensor distribution was also calculated under these constraints for the alternate assignment of α = 25±22° and β = 53±6° to Cys61, and α = -35±10° and β = -59±9° to Cys42 13 (C). Although a self-consistent, unique determination of the best-fit g-tensor directions is inherently not possible with only the two 13Cβ(Cys42/61) axial hyperfine tensors available, the two resulting g-tensor distributions (red bundle, average gz (gmax) direction; blue bundle, average gy (gint) direction; and magenta bundle, average gx (gmin) direction) for the two possible assignments of the Euler angles to Cys42 and Cys61 are very similar to each other in the reduced ARF system (B, C), where the best-fit gx direction (magenta) is almost along the Fe-Fe direction and the bestfit gz direction (red) is almost along the S-S direction (D, depicted by PyMOL v0.99 ).

Table 1. Simulated g-tensor direction cosines and skew from the molecular structural frame ARF Scenario 1a

FdxB g-tensorc

Fe-Fe

gz (gmax)

0.4263

gy





gx





Skew Max skew a b

S-S

Perpd

0.1277 0.8957

Fe-Fe

S-S

ARF Scenario 2b

Perpd

Fe-Fe

S-S

Perpd

0.1316 0.9915 0.0506

0.0069 0.9997 0.0277



-0.2120 -0.0242 0.9771

0.0096 -0.0277 0.9996



0.9684 -0.1284 0.2070

1.0000 0.0047 -0.0098

d

Fe-g⊥

S-g⊥

Perp -g||

Fe-gx





26°





64°

d

Fe-gx

S-gz

Perpd-gy

S-gz

Perp -gy

14°

8°

12°

1°

2°

2°

35°

20°

37°

24°

20°

23°

Euler angle assignment of α = 25° and β = 53° to Cys42, and α = -35° and β = -59° to Cys61. Euler angle assignment of α = 25° and β = 53° to Cys61, and α = -35° and β = -59° to Cys42.


21

Biochemistry 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 22 of 27

c

Simulated g-values: gmax = 2.020, gint = 1.936, gmin = 1.934 for FdxB 28, and gz = 2.022, gy = 1.901, gx = 1.804 for ARF. d Perpendicular to the [2Fe-2S] plane.

CONCLUSIONS By using the selectively dimensional

13

13

Cβ(Cys) labeled [2Fe-2S] proteins, we have developed a two-

C pulsed EPR approach to obtain the approximate g-tensor directions of a

paramagnetic form of a biological iron-sulfur cluster system with respect to the protein structural frame. This approach is shown to be complementary to 1H ENDOR and applicable in cases where the single-crystal EPR approach is not feasible. Despite the relatively larger errors in our 13

C pulsed EPR approach compared with the orientation-selective 1H ENDOR method 14, it can

set additional constraints for narrowing down and/or independently verifying approximate gtensor directions, without knowledge of simulated (or hypothetical) proton locations at the immediate cluster site. With this method, we determined the best-fit g-tensor directions of the reduced Rieske-type [2Fe-2S] cluster in

13

Cβ(Cys)-labeled ARF in the

15

N protein background,

where gy points roughly perpendicular to the [2Fe-2S] cluster plane and gx is almost along the Fe-Fe vector (Table 1 and Fig. 5). This assignment, reported herein for the first time with a lowpotential type Rieske-type ferredoxin, is apparently less skewed from the molecular axes of the reduced cluster, but nearly consistent with three independent g-tensor direction analyses of the high-potential type and cytochrome bc1-associated Rieske [2Fe-2S] protein systems by singlecrystal EPR

21

, orientation-selective 1H ENDOR

31

, and EPR measurements on oriented

membrane multilayer fragments 25-27. There remain discrepancies between our best-fit orientation and those proposed earlier for other Rieske-type [2Fe-2S] protein systems by a ligand-field model with the gz direction perpendicular to the cluster plane 41, and by an earlier ENDOR study with the gx direction perpendicular to the cluster plane

42

. Thus, even the correct assignment of


22

Page 23 of 27 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Biochemistry

the g-tensor axes to the molecular structural axes of the protein-bound reduced cluster requires experimental verification, so as to link the high-resolution magnetic resonance data to the protein atomic coordinates and characterize the mechanistic and functional aspects. Although the origin and extent of the distortion and skewing of the g-tensor axes in biological iron-sulfur protein systems is not well-understood

14, 21, 28, 40

, a larger protein model (in C1 symmetry) of the cluster

site plus its surrounding region will likely help improve molecular quantum theoretical calculations and bridge the gap between high-resolution magnetic resonance data and a deeper understanding of the protein structure-electron transfer function relationships in redox biology.

Notes The authors declare no competing financial interest.

ACKNOWLEDGMENTS We thank Dr. Yoshiharu Miyajima-Nakano and Ms. Risako Fukazawa (Nippon Medical School) for their help with the 13Cβ(Cys)-labeled protein sample preparations, and Drs. Kazuya Hasegawa and Takashi Kumasaka (SPring-8/JASRI) for providing us the X-ray crystallographic coordinates of ARF.

ASSOCIATED CONTENT Supporting Information. The following file is available free of charge. Figure S1 (PDF)

AUTHOR INFORMATION


23

Biochemistry 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 24 of 27

*Corresponding authors: E-mail: [email protected]; [email protected] Present Address †Department of Chemistry, Massachusetts Institute of Technology, Cambridge, MA 02139, United States. Author Contributions The manuscript was written through contributions of all authors. All authors have given approval to the final version of the manuscript. Funding Sources This investigation was supported in part by the International Collaborations in Chemistry Grant from JSPS (T.I.) and NSF (CHE-1026541 to S.A.D.), the Chemical Sciences, Geosciences and Biosciences Division, Office of Basic Energy Sciences, Office of Sciences, U.S. DOE Grant DE-FG02-08ER15960 (S.A.D., pulsed EPR work), the JSPS Grants-in-aid 24659202 and 26670215 (T.I.), 26•04415 (T.I., A.T.T.), and the Nagase Science and Technology Foundation Research Grant (T.I.). A.T.T. thanks the JSPS Postdoctoral Fellowship for Foreign Researchers for support.

REFERENCES [1] Holm, R. H., Kennepohl, P., and Solomon, E. I. (1996) Structural and functional aspects of metal sites in biology, Chem. Rev. 96, 2239-2314. [2] Beinert, H., Holm, R. H., and Münck, E. (1997) Iron-sulfur clusters: Nature's modular, multipurpose structures., Science 277, 653-659. [3] Solomon, E. I., Xie, X., and Dey, A. (2008) Mixed valent sites in biological electron transfer, Chem. Soc. Rev. 37, 623-638. [4] Peters, J. W., and Broderick, J. B. (2012) Emerging paradigms for complex iron-sulfur cofactor assembly and insertion, Annu. Rev. Biochem. 81, 429-450. [5] Johnson, D. C., Dean, D. R., Smith, A. D., and Johnson, M. K. (2005) Structure, function, and formation of biological iron-sulfur clusters, Annu. Rev. Biochem. 74, 247-281.


24

Page 25 of 27 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Biochemistry

[6] Solomon, E. I., Gorelsky, S. I., and Dey, A. (2006) Metal-thiolate bonds in bioinorganic chemistry, J. Comput. Chem. 27, 1415-1428. [7] Hunsicker-Wang, L. M., Heine, A., Chen, Y., Luna, E. P., Todaro, T., Zhang, Y. M., Williams, P. A., McRee, D. E., Hirst, J., Stout, C. D., and Fee, J. A. (2003) Highresolution structure of the soluble, respiratory-type Rieske protein from Thermus thermophilus: analysis and comparison, Biochemistry 42, 7303-7317. [8] Ullmann, G. M., Noodleman, L., and Case, D. A. (2002) Density functional calculation of the pKa values and redox potentials in the bovine Rieske iron-sulfur protein, J. Biol. Inorg. Chem. 7, 632-639. [9] Hoffman, B. M. (2003) Electron-nuclear double resonance spectroscopy (and electron spinecho envelope modulation spectroscopy) in bioinorganic chemistry, Proc. Natl. Acad. Sci. U.S.A. 100, 3575-35778. [10] Machonkin, T. E., Westler, W. M., and Markley, J. L. (2005) Paramagnetic NMR spectroscopy and density functional calculations in the analysis of the geometric electronic structures of iron-sulfur proteins, Inorg. Chem. 44, 779-797. [11] Lin, I.-J., Gebel, E. B., Machonkin, T. E., Westler, W. M., and Markley, J. L. (2005) Changes in hydrogen-bond strengths explain reduction potentials in 10 rubredoxin variants, Proc. Natl. Acad. Sci. U.S.A. 102, 14581-14586. [12] Iwasaki, T., Fukazawa, R., Miyajima-Nakano, Y., Baldansuren, A., Matsushita, S., Lin, M. T., Gennis, R. B., Hasegawa, K., Kumasaka, T., and Dikanov, S. A. (2012) Dissection of hydrogen bond interaction network around an iron-sulfur cluster by site-specific isotope labeling of hyperthermophilic archaeal Rieske-type ferredoxin, J. Am. Chem. Soc. 134, 19731-19738. [13] Taguchi, A. T., Miyajima-Nakano, Y., Fukazawa, R., Lin, M. T., Baldansuren, A., Gennis, R. B., Hasegawa, K., Kumasaka, T., Dikanov, S. A., and Iwasaki, T. (2018) Unpaired electron spin density distribution across reduced [2Fe-2S] cluster ligands by 13Cβ-cysteine labeling, Inorg. Chem. 57, 741-746. [14] Kappl, R., Bracic, G., and Hüttermann, J. (2009) Probing structural and electronic parameters in randomly oriented metalloproteins by orientation-selective ENDOR spectroscopy, In High Resolution EPR: Applications to Metalloenzymes and Metals in Medicine (Hanson, G., and Berliner, L., Eds.), pp 63-103, Springer, New York. [15] Lin, M. T., Sperling, L. J., Frericks Schmidt, H. L., Tang, M., Samoilova, R. I., Kumasaka, T., Iwasaki, T., Dikanov, S. A., Rienstra, C. M., and Gennis, R. B. (2011) A rapid and robust method for selective isotope labeling of proteins, Methods 55, 370-378. [16] Lin, M. T., Fukazawa, R., Miyajima-Nakano, Y., Matsushita, S., Choi, S. K., Iwasaki, T., and Gennis, R. B. (2015) Escherichia coli auxotroph host strains for amino acid-selective isotope labeling of recombinant proteins, Methods Enzymol. 565, 45-66. [17] Cosper, N. J., Eby, D. M., Kounosu, A., Kurosawa, N., Neidle, E. L., Kurtz, D. M. Jr, Iwasaki, T., and Scott, R. A. (2002) Redox-dependent structural changes in archaeal and bacterial Rieske-type [2Fe-2S] clusters, Protein Sci. 11, 2969-2973. [18] Kounosu, A., Li, Z., Cosper, N. J., Shokes, J. E., Scott, R. A., Imai, T., Urushiyama, A., and Iwasaki, T. (2004) Engineering a three-cysteine, one-histidine ligand environment into a new hyperthermophilic archaeal Rieske-type [2Fe-2S] ferredoxin from Sulfolobus solfataricus, J. Biol. Chem. 279, 12519-12528. [19] Iwasaki, T., Kounosu, A., Kolling, D. R. J., Crofts, A. R., Dikanov, S. A., Jin, A., Imai, T., and Urushiyama, A. (2004) Characterization of the pH-dependent resonance Raman


25

Biochemistry 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 26 of 27

transitions of archaeal and bacterial Rieske [2Fe-2S] proteins, J. Am. Chem. Soc. 126, 4788-4789. [20] Iwasaki, T., Samoilova, R. I., Kounosu, A., and Dikanov, S. A. (2009) Two-dimensional pulsed electron spin resonance characterization of 15N-labeled archaeal Rieske-type ferredoxin, FEBS Lett. 583, 3467-3472. [21] Bowman, M. K., Berry, E. A., Roberts, A. G., and Kramer, D. M. (2004) Orientation of the g-tensor axes of the Rieske subunit in the cytochrome bc1 complex, Biochemistry 43, 430-436. [22] Bertrand, P., More, C., Guigliarelli, B., Fournel, A., Bennett, B., and Howes, B. J. (1994) Biological polynuclear clusters coupled by magnetic interactions: from the point dipole approximation to a local spin model, J. Am. Chem. Soc. 116, 3078-3086. [23] More, C., Asso, M., Roger, G., Guigliarelli, B., Caldeira, J., Moura, J., and Bertrand, P. (2005) Study of the spin-spin interactions between the metal centers of Desulfovibrio gigas aldehyde oxidoreductase: identification of the reducible sites of the [2Fe-2S]2+,1+ clusters, Biochemistry 44, 11628-11635. [24] Guigliarelli, B., and Bertrand, P. (1999) Application of EPR spectroscopy to the structural and functional study of iron-sulfur proteins, Adv. Inorg. Chem. 47, 421-497. [25] Brugna, M., Rodgers, S., Schricker, A., Montoya, G., Kazmeier, M., Nitschke, W., and Sinning, I. (2000) A spectroscopic method for observing the domain movement of the Rieske iron-sulfur protein, Proc. Natl. Acad. Sci. U.S.A. 97, 2069-2074. [26] Roberts, A. G., Bowman, M. K., and Kramer, D. M. (2002) Certain metal ions are inhibitors of cytochrome b6f complex "Rieske" iron-sulfur protein domain movements, Biochemistry 41, 4070-4079. [27] Schoepp, B., Brugna, M., Riedel, A., Nitschke, W., and Kramer, D. M. (1999) The Qo-site inhibitor DBMIB favors the proximal position of the chloroplast Rieske protein and induces a pK-shift of the redox-linked proton, FEBS Lett. 450, 245-250. [28] Iwasaki, T., Kappl, R., Bracic, G., Shimizu, N., Ohmori, D., and Kumasaka, T. (2011) ISClike [2Fe-2S] ferredoxin (FdxB) dimer from Pseudomonas putida JCM 20004: structural and electron-nuclear double resonance characterization, J. Biol. Inorg. Chem. 16, 923935. [29] Taguchi, A. T., O'Malley, P. J., Wraight, C. A., and Dikanov, S. A. (2015) Hydrogen bond network around the semiquinone of the secondary quinone acceptor QB in bacterial photosynthetic reaction center, J. Phys. Chem. B 119, 5805-5814. [30] Flores, M., Isaacson, R., Abresch, E. C., Calvo, R., Lubitz, W., and Feher, G. (2007) Protoncofactor interactions in bacterial reaction centers from Rhodobacter sphaeroides R-26: II. Geometry of the hydrogen bonds to the primary quinone formula by 1H and 2H ENDOR spectroscopy, Biophys. J. 92, 671-682. [31] Kappl, R., Ebelshäuser, M., Hannemann, F., Berhardt, R., and Hüttermann, J. (2006) Probing electronic and structural properties of the reduced [2Fe-2S] cluster by orientation-selective 1H ENDOR spectroscopy: adrenodoxin versus Rieske iron-sulfur protein, Appl. Magn. Reson. 30, 427-459. [32] Berry, E. A., Guergova-Kuras, M., Huang, L.-S., and Crofts, A. R. (2000) Structure and function of cytochrome bc complexes, Annu. Rev. Biochem. 69, 1005-1075. [33] Link, T. A. (1999) The structures of Rieske and Rieske-type proteins, Adv. Inorg. Chem. 47, 83-157.


26

Page 27 of 27 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Biochemistry

[34] Bertini, I., Luchinat, C., Parigi, G., and Pierattelli, R. (2005) NMR spectroscopy of paramagnetic metalloproteins, Chembiochem, 1536-1549. [35] Iwasaki, T., Ohmori, D., Shimizu, N., and Kumasaka, T. (2007) Crystallization and preliminary X-ray diffraction studies of the ISC-like [2Fe–2S] ferredoxin (FdxB) from Pseudomonas putida JCM 20004, Acta Cryst. Sect. F 63, 1014-1016. [36] Ohmori, D., Yamakura, F., and Suzuki, K. (1989) Purification and some properties of a 2Fe ferredoxin in Pseudomonas ovalis, Biochem. Int. 18, 573-580. [37] Kounosu, A., Hasegwa, K., Iwasaki, T., and Kumasaka, T. (2010) Crystallization and preliminary X-ray diffraction studies of hyperthermophilic archaeal Rieske-type ferredoxin (ARF) from Sulfolobus solfataricus P1, Acta Cryst. Sect. F 66, 842-845. [38] Abdalla, J. A. B., Bowen, A. M., Bell, S. G., Wong, L. L., Timmel, C. R., and Harmer, J. (2012) Characterisation of the paramagnetic [2Fe-2S]+ centre in palustrisredoxin-B (PuxB) from Rhodopseudomonas palustris CGA009: g-matrix determination and spin coupling analysis, Phys. Chem. Chem. Phys. 14, 6526-6537. [39] Canne, C., Ebelshäuser, M., Gay, E., Shergill, J. K., Cammack, R., Kappl, R., and Hüttermann, J. (2000) Probing magnetic properties of the reduced [2Fe-2S] cluster of the ferredoxin from Arthrospira platensis by 1H ENDOR spectroscopy, J. Biol. Inorg. Chem. 5, 514-526. [40] Dicus, M. M., Conlan, A., Nechushtai, R., Jennings, P. A., Paddock, M. L., Britt, R. D., and Stoll, S. (2010) Binding of histidine in the (Cys)3(His)1-coordinated [2Fe-2S] cluster of human mitoNEET, J. Am. Chem. Soc. 132, 2037-2049. [41] Bertrand, P., Guigliarelli, B., Gayda, J.-P., Beardwood, P., and Gibson, J. F. (1985) A ligand-field model to describe new class of 2Fe-2S clusters in proteins and their synthetic analogues, Biochim. Biophys. Acta 831, 261-266. [42] Gurbiel, R. J., Doan, P. E., Gassner, G. T., Macke, T. J., Case, D. A., Ohnishi, T., Fee, J. A., Ballou, D. P., and Hoffman, B. M. (1996) Active site structure of Rieske-type proteins: electron nuclear double resonance studies of isotopically labeled phthalate dioxygenase from Pseudomonas cepacia and Rieske protein from Rhodobacter capsulatus and molecular modeling studies of a Rieske center, Biochemistry 35, 7834-7845.

SYNOPSIS (TOC figure)


27

g-Tensor Directions in the Protein Structural Frame of

Recommend Documents