Design of dinuclear manganese cofactors for bacterial reaction centers

Design of dinuclear manganese cofactors for bacterial reaction centers

BBABIO-47525; No. of pages: 9; 4C: 2, 3, 4, 5, 6, 7 Biochimica et Biophysica Acta xxx (2015) xxx–xxx Contents lists available at ScienceDirect Bioch...

2MB Sizes 0 Downloads 14 Views

BBABIO-47525; No. of pages: 9; 4C: 2, 3, 4, 5, 6, 7 Biochimica et Biophysica Acta xxx (2015) xxx–xxx

Contents lists available at ScienceDirect

Biochimica et Biophysica Acta journal homepage: www.elsevier.com/locate/bbabio

Review

Design of dinuclear manganese cofactors for bacterial reaction centers☆ Tien L. Olson, Eduardo Espiritu, Selvakumar Edwardraja, Chad R. Simmons, JoAnn C. Williams, Giovanna Ghirlanda, James P. Allen ⁎ School of Molecular Sciences, Arizona State University, Tempe, AZ 85287-1604, USA

a r t i c l e

i n f o

Article history: Received 25 June 2015 Accepted 14 September 2015 Available online xxxx Keywords: Photosynthesis Mn-binding proteins De novo protein design Photosystem II Four-helix bundles Electron transfer

a b s t r a c t A compelling target for the design of electron transfer proteins with novel cofactors is to create a model for the oxygen-evolving complex, a Mn4Ca cluster, of photosystem II. A mononuclear Mn cofactor can be added to the bacterial reaction center, but the addition of multiple metal centers is constrained by the native protein architecture. Alternatively, metal centers can be incorporated into artificial proteins. Designs for the addition of dinuclear metal centers to four-helix bundles resulted in three artificial proteins with ligands for one, two, or three dinuclear metal centers able to bind Mn. The three-dimensional structure determined by X-ray crystallography of one of the Mn-proteins confirmed the design features and revealed details concerning coordination of the Mn center. Electron transfer between these artificial Mn-proteins and bacterial reaction centers was investigated using optical spectroscopy. After formation of a light-induced, charge-separated state, the experiments showed that the Mn-proteins can donate an electron to the oxidized bacteriochlorophyll dimer of modified reaction centers, with the Mn-proteins having additional metal centers being more effective at this electron transfer reaction. Modeling of the structure of the Mn-protein docked to the reaction center showed that the artificial protein likely binds on the periplasmic surface similarly to cytochrome c2, the natural secondary donor. Combining reaction centers with exogenous artificial proteins provides the opportunity to create ligands and investigate the influence of inhomogeneous protein environments on multinuclear redox-active metal centers. This article is part of a Special Issue entitled Biodesign for Bioenergetics - the design and engineering of electron transfer cofactors, proteins and protein networks. © 2015 Elsevier B.V. All rights reserved.

1. Recreation of Mn clusters The incorporation of metal ions as cofactors in proteins greatly extends their attainable chemical functionality. For example, the Mn4Ca cluster of photosystem II enables the catalysis of the oxidation of water, one of the key reactions in biology. The Mn4Ca cluster has many unique features compared to other metal clusters found in metalloproteins and is highly conserved in organisms that use photosystem II for the conversion of light into chemical energy. However, our ability to capitalize on this activity is limited by many questions concerning the molecular mechanism of water oxidation and the challenges in replicating these features in artificial systems. Therefore an alternative strategy is to recreate specific aspects of the oxygen-

Abbreviations: DF, Due Ferri, or diiron protein design; P680, chlorophyll cofactor serving as the primary electron donor in photosystem II; P865, bacteriochlorophyll cofactor serving as the primary electron donor in reaction centers; QA, ubiquinone cofactor serving as an electron acceptor in reaction centers; YZ, tyrosine 161 of the D1 subunit of photosystem II. ☆ This article is part of a Special Issue entitled Biodesign for Bioenergetics — The design and engineering of electron transfer cofactors, proteins and protein networks. ⁎ Corresponding author. E-mail address: [email protected] (J.P. Allen).

evolving complex in a model system where we can control the protein environment and perform detailed spectroscopic studies. Water oxidation is a multi-electron process occurring in photosystem II, in which light excitation of the primary electron donor, P680, is followed by electron transfer to the primary and secondary quinones. The oxidized primary donor is then reduced by the Mn4Ca cluster by electron transfer through a redox-active tyrosine, YZ. After four sequential electron transfers, two water molecules are converted into molecular oxygen. Characterization of the properties of the Mn4Ca cofactor by a number of spectroscopic experiments helps explain its ability to oxidize water. The determination of the three-dimensional structure of photosystem II at 1.9 Å resolution provides a detailed view of the Mn4Ca cofactor (Fig. 1) [1,2]. Electron paramagnetic resonance, electron nuclear double resonance, electron spin envelope modulation, and X-ray absorption fine structure spectroscopies have been essential in characterizing Mn clusters, in particular the Mn4Ca cluster as it proceeds through the different oxidation states of the S cycle [3–7]. Fourier transform infrared spectroscopy has proven to be a sensitive tool for characterizing the coordination of the Mn4Ca cluster and bond changes resulting from alterations of the ligands by mutagenesis [6,8–11]. However, the factors that allow the system to oxidize water are not fully understood. For example, the oxidation/reduction midpoint potentials of the Mn4Ca cluster, YZ, and P680 are critical parameters to describe these electron

http://dx.doi.org/10.1016/j.bbabio.2015.09.003 0005-2728/© 2015 Elsevier B.V. All rights reserved.

Please cite this article as: T.L. Olson, et al., Design of dinuclear manganese cofactors for bacterial reaction centers, Biochim. Biophys. Acta (2015), http://dx.doi.org/10.1016/j.bbabio.2015.09.003

2

T.L. Olson et al. / Biochimica et Biophysica Acta xxx (2015) xxx–xxx

the reactions are initiated by light, identification of features that are triggered by light excitation facilitates the interpretation of spectra. Our ability to manipulate the energetics of the cofactors provides the possibility to determine the oxidation/reduction midpoint potential of metals acting as secondary donors, and consequently allows us to relate electron transfer rates to the free energy differences. Building on our previous success in binding metals to the reaction center, we are initiating an approach that extends the Mn cofactor complexity by interaction with protein domains containing dinuclear Mn clusters. 2. Bacterial reaction centers as a template for the design of Mn cofactors Fig. 1. Mn centers and associated amino acid ligands in photosystem II and modified reaction centers. The Mn4Ca cluster of photosystem II (Mn, purple spheres; oxygen, red spheres; Ca, green sphere) is shown with residues Asp 61, Asp 170, and Glu 189 of the D1 subunit core (cyan), His 332, Glu 333, His 337, Asp 342, and the carboxyl-terminal residue Ala 344 of the D1 subunit carboxyl-terminal region (green), and Glu 354 and Arg 357 of the CP43 subunit (orange), with numbering of Thermosynechococcus vulcanus (left) [1]. Residues Tyr 161 (YZ) and His 190 of the D1 subunit are also nearby. The binding site for the mononuclear Mn (purple sphere) in modified reaction centers includes residues Tyr 164, Glu 168, Glu 173, His 193, Asp 288 and Asp 292 of the M subunit (cyan) (right) [42].

transfer reactions, but their potentials have only been inferred based on the electron transfer rates because of the difficulties in directly measuring the potentials of highly-oxidizing cofactors [11–13]. The surrounding protein contributes to the structure of the Mn4Ca cluster, which is organized in a distorted cuboidal arrangement (Fig. 1). The cluster is bound through two sets of ligands involving the D1 subunit, including three from the core region and five from the carboxyl-terminal region, along with two ligands provided by the CP43 subunit. In addition, two pairs of water molecules are bound to the cluster. To study the influence of the protein environment on the mechanism of water oxidation, all of the protein ligands associated with the Mn4Ca cluster have been modified in photosystem II [8, 14–17]. Despite extensive characterization of these mutants, the mechanism remains debated, leaving questions to be addressed by a gain-offunction approach. The Mn4Ca cluster efficiently catalyzes water oxidation near neutral conditions with a very low overpotential and high turnover frequency. These capabilities have inspired the design of Mn-based catalysts for water oxidation resulting in a wealth of information on synthetic Mn compounds that have features characteristic of Mn-enzymes, with mononuclear or dinuclear cofactors, as found in Mn-superoxide dismutase and Mn-catalase, or more complex tetranuclear Mn sites [18–21]. The mononuclear and dinuclear Mn compounds have the advantage of relatively straightforward syntheses and can replicate certain features of the Mn4Ca cluster, including in some cases catalytic molecular oxygen formation, although they cannot duplicate the four oxidation steps. The tetranuclear Mn compounds are designed in a variety of arrangements, from linear to cubane, and have revealed features such as the range of couplings expected for the electronic states, but many do not access the higher valences found in the Mn4Ca cluster. Most of these compounds are not stable under neutral, aqueous conditions, and the synthetic ligands serve as highly limited models for a protein environment. The incorporation of Mn clusters in artificial proteins can build upon this knowledge of synthetic compounds to establish welldefined Mn arrangements in inhomogeneous protein environments that can be used to address the role of the protein environment of the Mn4Ca cluster in water oxidation. The design of novel Mn cofactors bound to the bacterial reaction center presents an opportunity for investigations using a proteinbased framework capable of intricate, light-induced oxidation/reduction reactions. The bacterial reaction center is a well-defined system for modification and offers several advantages for these studies. Because

Our work is directed towards understanding the factors that influence the electron and energy transfer properties of bacteriochlorophyll–protein complexes in photosynthetic bacteria. In Rhodobacter sphaeroides, reaction centers are embedded in the membrane and perform the primary photochemical conversion of light energy [22]. The bacterial reaction center and photosystem II are evolutionarily related and share common structural features, although bacterial reaction centers participate in a cyclic process rather than the Z scheme of oxygenic photosynthesis [23]. As in photosystem II, light excitation of the primary electron donor, the bacteriochlorophyll dimer P865, results in electron transfer to the primary and secondary quinone. However, reaction centers from R. sphaeroides have no Mn cluster, and the oxidized donor is reduced by a water-soluble cytochrome c2. The reaction center can be excited again resulting in a second electron transfer to the secondary quinone that is coupled with proton transfer to form a quinol. The cycle is completed by electron and proton transfers of the quinol and the oxidized cytochrome c2 through the membrane-bound cytochrome bc1 complex. The reaction center has proven to be a robust system for the characterization of electron transfer, and the influence of the protein environment has been extensively studied. One of the primary directions of mutagenesis studies has been the elucidation of the factors that control electron transfer between the cofactors [24,25]. For example, the reaction center cofactors are divided into symmetry-related A and B branches, with electron transfer occurring only through the A branch. Studies in which the protein environment has been manipulated have yielded a general model that the asymmetry of the initial electron transfer steps is primarily determined by energetics, with transfer along the A branch having much more favorable energy changes than those involving the B branch. Application of this model has yielded significant enhancement of B-side electron transfer by combining alterations of amino acid residues that block transfer along the A branch and improve the energetics for transfer along the B branch [26]. In addition to energetics, a key determinant of the initial electron transfer is the dynamical properties of the reaction center. The protein environment establishes an inhomogeneous network of interactions with the different cofactors, resulting in complex protein kinetics that control the initial rate of electron transfer [27–30]. Protein interactions and dynamics also establish the factors that control the initial electron transfer reactions in photosystem II, in particular electrostatic interactions that help establish the high midpoint potential of P680[31–35]. In addition to studies of the initial electron transfer, mutagenesis has been used to reveal key aspects of the properties of electron transfer on a slower timescale, namely electron transfer involving the quinones and cytochrome c2. One of the major common features of reaction centers and photosystem II is the role of the secondary quinone, which serves as a two-electron acceptor. The full reduction of the secondary quinone to quinol requires the transfer of two protons from solution. The proton transfer occurs through specific proton transfer pathways, including protonatable amino acid residues forming the binding site that are conserved between reaction centers and photosystem II [36–38]. Electrostatic interactions involving protonatable amino acid residues also control the binding of cytochrome c2 to the reaction center as part

Please cite this article as: T.L. Olson, et al., Design of dinuclear manganese cofactors for bacterial reaction centers, Biochim. Biophys. Acta (2015), http://dx.doi.org/10.1016/j.bbabio.2015.09.003

T.L. Olson et al. / Biochimica et Biophysica Acta xxx (2015) xxx–xxx

of the second-order electron transfer process, as changes of surface carboxylates of the reaction center and nearby residues result in large changes in the binding affinity [39,40]. In studying electron transfer in bacterial reaction centers, we have focused on reactions involving P865, which consists of two symmetryrelated bacteriochlorophyll a molecules. The properties of P865 have been altered by changing the interactions with nearby residues, resulting in a highly-oxidizing donor that serves as the foundation for adding functional capabilities [35]. Drawing on our extensive experience in systematically changing the protein environment by mutagenesis, we have characterized altered functional properties using variety of approaches such as vibrational, electron paramagnetic resonance, and transient optical spectroscopy, followed by modeling of the electronic structure of the cofactors and the electron transfer processes [23,41]. The adaptability of the reaction center in response to such alterations provides the ability to create complexes with new redox-active metal cofactors. Mn-binding sites have been introduced in the reaction center, with the M2 reaction center having a mononuclear Mn cofactor coordinated by Glu M168, Glu M173, His M193, and Asp M288 (Fig. 1) [42]. In addition to the similar location of the Mn-binding site with that of the Mn4Ca cluster of photosystem II, the site includes Glu M173, corresponding to Asp D1-170, and the pair Tyr M164 and His M193, corresponding to Tyr D1-161 (YZ) and His D1-190, which participate in the electron transfer reactions between the Mn4Ca cluster and P+ 680. The altered reaction centers have been shown to tightly bind a mononuclear Mn cofactor at the designed site with additional ligands provided by bound water molecules. The binding is associated with proton release and has a pronounced pH dependence, and the inclusion of synergistic anions facilities the binding at all pH values [43,44]. Although the bound Mn cofactor cannot oxidize water, the oxidation/reduction midpoint potential of 0.63 V provides the capacity for light-driven oxygen production from superoxide [45,46]. The Mn-binding site in the reaction center reflects overall structural similarities with photosystem II (Fig. 2). Although photosystem II has at least 20 different protein subunits and approximately 60 cofactors, it can be viewed as having a core composed of the D1 and D2 subunits, each containing five transmembrane helices, and the cofactors involved in the initial electron transfer reactions, similar to the reaction center. Structural alignment of the D1 and D2 subunits to the L and M subunits reveals a strong similarity in the transmembrane regions, allowing us to position the Mn site at a location homologous to the Mn4Ca cluster. Although the two sites have similar structural features, the binding site for the Mn4Ca cluster contains several additional ligands provided by the longer carboxyl-terminal regions of the D1 and D2 subunits, which each forms a loop-helix-loop motif not present in the reaction center. Based on this comparison, a way of gaining the ability to bind multinuclear Mn clusters and perform multi-electron reactions would be to modify the reaction center to include protein domains containing

3

additional metal ligands. Towards this end, a series of synthetic Mnproteins consisting of four-helix bundles that contain binding sites for dinuclear Mn clusters were designed and tested for their ability to bind to the reaction center and donate electrons to the oxidized bacteriochlorophyll dimer. 3. Artificial Mn-proteins The design of de novo synthetic proteins with binding sites for metal ions is an attractive means to characterize the influence of protein interactions on the properties of metal cofactors. While mutagenesis studies of metal-binding sites in proteins can be revealing, interpretation of such studies are often hampered by the complexity of native proteins. The use of de novo design provides the opportunity to investigate protein interactions with user-defined physical characteristics. The introduction of novel metal centers into proteins remains a challenge but can create new enzymatic reactions [47,48]. In particular, the design of four-helix bundles has been very productive as a platform for examining dinuclear metal cofactors [49,50] and for binding multiple cofactors capable of electron transfer [51,52]. One well-studied group of metal-binding artificial proteins is the DF (Due Ferri) family [reviewed in [49,53]]. The scaffold for the DF proteins is the four-helix bundle, which has a stable alpha-helical structure that can accommodate binding of hemes in the hydrophobic core [54]. In the initial DF1 protein, the Glu-Xxx-Xxx-His helical motif, found in a number of metalloproteins, was incorporated in the design of a model four-helix bundle with two-fold symmetry [55]. Structural analysis confirmed the designed homodimeric helix-loop-helix topology, with each metal ion in the dimetal cofactor being pentacoordinate with a carboxylate bridge [55–57]. Modeling pointed to the contributions of the protein scaffold, including the first and second shell ligands, in constraining the metal-binding site [58,59]. This design was later modified to incorporate a more accessible dimetal site (L13A-DF1) [56] and improved solubility (DF2)[60]. Additional variants on the DF peptide design include heterotetrameric (DFtet, four-chain) [61,62] and monomeric (DFsc, single chain) four-helix bundles [63], and analysis of the turns led to a change in the loop connecting the two helices (DF2t) [64–67]. In addition to Fe, the DF proteins also bind Mn, Co, Zn, Cu, and Ni [60]. X-ray or NMR structures have been reported for several of the DF proteins with different metals [55–57,64,68–71]. These types of protein frameworks enable the synthesis of redoxactive proteins that can perform enzymatic reactions, as has been found for the DF family and other de novo proteins with metal centers. Some proteins of the DF family were shown to have catalytic activity, particularly ferrooxidase reactions but also phenol oxidase, hydroquinone oxidase, and arylamine N-hydroxylation reactions [71–76]. The four-helix bundle structure can accommodate complex cofactor arrangements that are capable of dioxygen binding, superoxide and peroxide generation, electron transfer to cytochrome c, and potentially

Fig. 2. Structural comparison of reaction centers and photosystem II. (A) The similar core protein regions that surround the cofactors (red) contain the D1 subunit (cyan) and D2 subunit (yellow) of photosystem II (left) and the M subunit (cyan) and L subunit (yellow) of the reaction center (right). The Mn4Ca cluster and the mononuclear binding site in the modified reaction centers are located in corresponding positions in the proteins, with the environment of the Mn4Ca cluster having additional contributions from the CP43 subunit (orange). (B) Significant overlap in the five transmembrane helices is observed for the M subunit (cyan) and D1 subunit (wheat) (left) and the L subunit (yellow) and D2 subunit (wheat) (right), with the D1 and D2 subunits having additional residues forming a loop-helix-loop at the carboxyl-terminus (green).

Please cite this article as: T.L. Olson, et al., Design of dinuclear manganese cofactors for bacterial reaction centers, Biochim. Biophys. Acta (2015), http://dx.doi.org/10.1016/j.bbabio.2015.09.003

4

T.L. Olson et al. / Biochimica et Biophysica Acta xxx (2015) xxx–xxx

multiple electron transfer reactions [52]. A protein capable of internal electron transfer was created using bacterioferritin as the platform for protein design, with the heme being replaced with a photoactive chlorin, and the dinuclear Fe center replaced with a dinuclear Mn cluster [77–79]. Excitation of the chlorin using light was followed by electron transfer, resulting in oxidation of the Mn2 center with a redox-active tyrosine serving as an intermediate electron acceptor in a process that had characteristics related to those of the Mn4Ca cluster. Our focus in this paper is on four-helix bundles with bound Mn centers, but synthetic enzymes can also make use of a range of transition metal ions and other secondary structure arrangements as described in other articles of this special issue.

4. Design of four-helix bundles with additional dinuclear Mn centers We constructed a set of three Mn-proteins based on synthetic peptides that bind Mn. The template for the designed proteins was the Df2t homodimer protein structure, which contains a central dinuclear metal site (PDB ID: 1MFT) [64,76]. Each of the model structures was constructed using the Discovery Studio 2.0 modeling and simulation software (Accelrys Inc., San. Diego, CA, USA). Energy minimization on each modified structure was performed using the CHARMm force field [80]. Minimization was initially implemented using the steepest descent method, followed by the conjugate gradient method for better convergence to attain a minimum. The energy minimization cycles were carried out until the RMS gradient tolerance was less than 0.1 kcal mol−1 Å−1. The sequence of the P0 protein is that of the Df2t protein [64,65], with the substitution of Gly for Met at the amino-terminus and lack of Gly at the carboxyl-terminus (Fig. 3). The P0 protein design is for a four-helix bundle with a dinuclear Mn center in the middle region and metal ligands at Glu 11, Glu 41 and His 44. Additional designs incorporated new metal-binding sites added in the outer regions. Relative to the P0 protein, the P1 and P2 proteins have several substitutions, of Leu 4 to Glu, Tyr 18 to Glu, Leu 34 to Glu, Ile 37 to His, Leu 48 to Glu and Ile 51 to His, which potentially form two additional Mn-binding sites symmetrically positioned on either side of the central metal-binding site. For the P2 protein, the ligands of the central Mn-binding site have been removed with the substitutions of Glu 11 to Leu, Glu 41 to Leu, and His 44 to Ile. Thus the site present in the middle core region of P0 is retained in the P1 model but replaced with a hydrophobic core in the P2 model. With these alterations, the P2 and P1 proteins should form four-helix bundles containing two and three dinuclear Mn centers, respectively, compared to one dinuclear center in the P0 protein. Each of the artificial proteins was expressed in Escherichia coli and purified through affinity chromatography using a histidine tag followed by removal of the histidine tag with protease cleavage. The proteins were found to fold as four-helix bundles using circular dichroism experiments [81]. The metal binding of the expressed proteins was investigated using electron paramagnetic resonance spectroscopy [81]. Samples with bound Cu(II) showed spectra characteristic of classic type II Cu(II) centers, with the additional two metal centers in the P1 protein having a different environment than the metal center found in the P0 protein. The additional characterization of these artificial proteins using X-ray diffraction and optical spectroscopy is described below (Sections 5-7).

Fig. 3. Sequences of the P0, P1, and P2 proteins. Highlighted amino acid residues provide the ligands, for the central metal-binding site in the P0 and P1 proteins (cyan) and the additional ligands designed for the P1 and P2 proteins (green), and constitute the turn region (yellow) between the two helices in the monomer.

5. Structure of the P0 Mn-protein The P0 protein was amenable to structure determination by X-ray diffraction of the crystallized protein bound to Mn. After purification, the P0 protein was dialyzed against 15 mM HEPES, pH 7.5 for crystallization. The protein was set at a concentration of 140 μM for the protein monomer and incubated with MnCl2 at a ratio of 7 MnCl2:1 peptide monomer by rocking at 4 °C for 2 h. The Mn-bound protein was then concentrated to 1 mM (12 mg/mL). A broad screen of crystallization conditions using hanging drops showed crystals when 1,4-dioxane was present in the solutions. The crystallization conditions were refined, and large crystals with a size of approximately 0.2 mm in length were obtained when the reservoir contained 35% 1,4-dioxane, with the protein and reservoir being mixed in a 1:1 volume ratio. Diffraction data were initially collected with a Rigaku RU200 rotating anode, RAXIS IV imaging plate, and low temperature cooling system at the ASU Protein Crystallography facility. A data set with improved overall quality was collected at the Advanced Light Source synchrotron at Berkeley, CA, on beamline 8.2.2 with an ADSC Q315R detector at λ = 1.0 Å, yielding a complete data set with a resolution limit of 1.75 Å. All data were processed using the HKL2000 package [82]. Initial phases were determined using molecular replacement with the template for the designed proteins, the Df2t homodimer protein structure (PDB ID: 1MFT) [64]. A unique solution was found, and the structural model was rebuilt and refined using crystallographic packages including CCP4, PHENIX, COOT, and PyMOL [83–86]. The resulting model is in excellent agreement with the data, with Rwork and Rfree of 19.9% and 23.5%, respectively, and all values of the Ramachandran plot being in the favored region with no outliers. The diffraction and refinement data are summarized in Table 1. The three-dimensional structure of the P0 Mn-protein is consistent with the design features (Fig. 4). Electron density is excellent for all of the protein except for a few amino acid residues at the carboxylterminus and amino-terminus regions. Each subunit of the protein forms two long alpha helices, formed by Tyr 3 to Lys 24 and Glu 29 to Ile 51, with a short connecting loop composed of residues Ala 25 to Pro 28. Together the subunits form a four-helix bundle, related to each other by a two-fold rotation axis about the center of the bundle. At the middle of the bundle is the dinuclear Mn center (Fig. 4). The two Mn ions are 4.0 Å apart and are coordinated by ligands from two sets of symmetry-related side chains. The ligands include the Nε of His 44 to each Mn, the two terminal oxygen atoms of Glu 11 providing a bidentate ligand of each Mn, and the two terminal oxygen atoms of Glu 41 bridging the two Mn ions. In addition, a bridging water molecule is located 2.7 Å between the two Mn ions, and completing the metalbinding site are Ala 14, Tyr 18, Ile 37 from each subunit. The structure of the P0 protein closely resembles the structures of the DF family of proteins. The Mn–Mn distances range from 3.6 to 4.3 Å, with the shorter distances found when water or a dimethyl sulfoxide molecule was serving as a bridging ligand, in the four structures of DF proteins that have been determined with a dinuclear Mn center [56,57,69]. The distances and coordination of the Mn center in the P0 structure most closely resemble the structures of the DF proteins with the bridging molecules. In all cases, the two Mn ions are coordinated by four carboxylates and two imidazoles that are symmetrically positioned about the Mn center. 6. Electron transfer from Mn-proteins to reaction centers from R. sphaeroides In our previous studies (Section 2), we found that reaction centers could be modified to bind a mononuclear Mn center at a location analogous to that of the Mn4Ca cluster of photosystem II (Fig. 2). Upon illumination, the bound Mn center was capable of performing rapid electron transfer to the oxidized electron donor P865, provided that the oxidation/reduction midpoint potential of P865 was larger than the 0.63 V midpoint potential of the Mn center [45]. These results suggested

Please cite this article as: T.L. Olson, et al., Design of dinuclear manganese cofactors for bacterial reaction centers, Biochim. Biophys. Acta (2015), http://dx.doi.org/10.1016/j.bbabio.2015.09.003

T.L. Olson et al. / Biochimica et Biophysica Acta xxx (2015) xxx–xxx Table 1 Summary of X-ray diffraction data and refinement. Space group Resolution limit (Å) Unit cell lengths a (Å) b (Å) c (Å) Angles (°) Reflections Wavelength (Å) Total Unique Multiplicity I/σ (I) RMerge (%) Completeness (%) Refinement Rwork (%) Rfree (%) Number in asymmetric unit RMSD bond length (Å) RMSD bond angle (°) Ramachandran plots Favored (%) Outlier (%)

P212121 26.23–1.75 41.2 52.4 52.6 90, 90, 90 1.00 71,238 10,892 6.6 (5.4) 16.2 (1.4) 0.141 99.8 (99.4) 19.9 23.5 1 0.010 1.103 100 0

RMSD, root mean square deviation. Highest resolution shell is shown in parentheses.

that the artificial proteins with the bound Mn centers would also be able to reduce P+ 865 in highly-oxidizing reaction centers. To test this hypothesis, we measured the capability for electron transfer from the artificial Mn-proteins to high-potential reaction centers. These reaction centers have a P865/P+ 865 midpoint potential of 0.77 V as a result of three mutations near the bacteriochlorophyll dimer, Leu to His at L131, Leu to His at M160, and Phe to His at M197 [87]. Cells were grown and reaction centers were purified using procedures previously described [42]. Light-induced absorption spectra were measured using a Cary 6000i UV–Vis-NIR spectrophotometer (Agilent Technologies, CA, USA). Light excitation was from an Oriel tungsten lamp with an 860 nm interference filter, with a nonsaturating intensity. Light-minus-dark spectra were recorded as previously described [42,43]. Samples were prepared in 15 mM Ches, pH 9.4 and 0.05% Triton X-100 with the reaction centers poised at 1.5 μM. In all measurements, terbutryn at a concentration of 100 μM was added to block electron transfer from the primary to secondary quinone. The Mn-proteins were included at concentrations up to 240 μM for the protein monomer. Mn was added to the apo-proteins either prior to or after adding the proteins to the assay solution. In some cases, the

5

Mn-proteins were concentrated by centrifugation of the solution through Amicon Ultra 3 K filters (EMD Millipore, Massachusetts, USA). The ability of the Mn-protein to transfer an electron to the oxidized bacteriochlorophyll dimer, P+ 865, was tested using an optical assay that measures light-induced changes in the reaction center spectrum (Fig. 5). In the absence of the Mn-protein, the steady-state lightminus-dark spectrum of reaction centers exhibited the characteristic − features of the P+ 865 QA charge-separated state, with an absorption decrease, centered at 865 nm, in the absorption band associated with the bacteriochlorophyll dimer, as well as electrochromic shifts of the 800 nm and 760 nm bands of the bacteriochlorophyll monomers and bacteriopheophytins, respectively. The extent of bleaching of the dimer band provides a measure of the extent of oxidation of the dimer, which is at a maximum in the absence of an external donor. When the Mn-proteins were present, the bleaching of the dimer band decreased, with the extent of the changes being dependent on the concentration of the Mn-protein. The features associated with the presence of Q− A were still observed. Spectral changes were seen for all three Mnproteins, with the P1 protein producing the largest decreases, followed by the P2 and P0 proteins. Comparable changes were observed when the Mn was bound to the proteins before or after mixing with the reaction centers. The spectra are consistent with electron transfer from the Mnprotein to the oxidized P865 dimer of the reaction centers. Under these conditions, light excites P865, resulting in formation of the charge− separated state P+ 865 QA , which recombines in the absence of a secondary donor. Reduction of P+ 865 in the presence of the Mn-protein shows that the Mn center effectively competes with the chargerecombination reaction. The forward electron transfer reactions can be written as.   hv Mn2þ Mn2þ P ðP865 Q A ÞRC → Mn2þ Mn2þ P ðP865  Q A ÞRC    → Mn2þ Mn2þ P P 865 þ Q A − RC → Mn2þ Mn3þ P ðP 865 Q A − ÞRC

ð1Þ

where (Mn2+ Mn2+)P and (Mn2+ Mn3+)P are one of the dinuclear cofactors of the Mn-proteins in the reduced and oxidized state, respectively, (P865*QA)RC is the reaction center in the light-induced excited state, − and (P+ 865 QA )RC is the reaction center in the charge-separated state. The oxidized state of the Mn center is not optically active, but the spectra show the Q− A spectral signature. After illumination, the spectra recover to the ground state, showing that the process is fully reversible, rather than representing an irreversible oxidation/reduction reaction. In the absence of illumination, no spectral changes are observed, show− ing that the process is driven by the formation of the P+ 865 QA chargeseparated state. These measurements follow the same behavior as

Fig. 4. (A) Electron density map (2FOFC, 2σ) and (B) structural model of the P0 Mn-protein at 1.7 Å resolution. The design features of the expressed protein include a four-helix bundle with a dinuclear Mn cofactor (purple spheres) and coordination by carboxylates and imidazoles (colored by atom type) as well as a water molecule (red sphere).

Please cite this article as: T.L. Olson, et al., Design of dinuclear manganese cofactors for bacterial reaction centers, Biochim. Biophys. Acta (2015), http://dx.doi.org/10.1016/j.bbabio.2015.09.003

6

T.L. Olson et al. / Biochimica et Biophysica Acta xxx (2015) xxx–xxx

Fig. 5. Light-minus-dark difference spectra of reaction centers. (A) Samples are high-potential reaction centers with no Mn-protein added (black), with 15 μM MnP1 protein (cyan), and with 60 μM MnP1 protein (blue). (B) Samples are high-potential reaction centers with no Mn-protein added (black), 100 μM MnP0 protein (pink), 100 μM MnP2 protein (green), and 100 μM MnP1 protein (blue) (right). The decrease in the amount of bleaching of the dimer bacteriochlorophyll band centered at 865 nm in the presence of the Mn-proteins is consistent with electron transfer from the Mn to P+ 865.

when reaction centers are rapidly reduced after illumination by other secondary electron donors such as cytochrome c2. The amount of P+ 865 reduced by the Mn cofactor is dependent upon the free energy difference for this electron transfer reaction, which is proportional to the difference in the midpoint potentials of P865 and the Mn cofactor. In the light-induced spectra of reaction centers with the Mn-proteins (Fig. 5), an appreciable amount of dimer remains oxidized. This is likely because the free energy difference is relatively small due to the Mn2 +/Mn3 + midpoint potential of the Mn-protein being comparable to the 0.77 V value for the P865/P+ 865 midpoint potential of the reaction center. In experiments using wild-type reaction centers, the addition of the Mn-proteins did not significantly alter the spectral features, consistent with the Mn2+/Mn3+ midpoint potential of the Mn-proteins being much larger than the P865/P+ 865 midpoint potential of 0.51 V for wild type. An assumption of the experiments is that the Mn added to the samples is bound to the Mn-protein. Consistent with this requirement, Mn has been shown to tightly bind to the DF-type proteins with a dissociation constant of approximately 1 μM [60,75]. Based on the number of designed binding sites, the P0, P2, and P1 proteins should bind Mn up to stoichiometric amounts of 1:1, 2:1, and 3:1, respectively, relative to the protein monomer. Free MnCl2 can also reduce the oxidized dimer in high-potential reaction centers, with a concentration of approximately 100 μM required to appreciably bind to the reaction center and reduce the oxidized dimer [88]. As a control experiment to test for the amount of free Mn in the samples, we measured the solution that passed through the membrane during concentration of the Mn-protein. Metal analysis of the solution that passed through the membrane using an optical assay [89] yielded low concentrations of free Mn. The addition of this solution to the reaction centers did not alter the spectral features, showing that the effect on P+ 865 arises from the presence of the Mn-protein and not free Mn in the solution. The decrease in dimer bleaching resulting from the presence of the P0, P2, and P1 Mn-proteins, corresponding to an increase in the extent of electron transfer, was greatest for the P1 protein and least for the P0 protein, corresponding to the different Mn cofactors. The P0 protein contains a central dinuclear Mn center, while the P2 protein incorporates two outer dinuclear Mn-binding sites, and P1 has both the central and two outer dinuclear Mn centers. The relative effectiveness of the three Mn-proteins to serve as a secondary electron donor could arise from electron transfer being more likely to occur from proteins with a larger number of Mn cofactors or differences in parameters such as the midpoint potentials between the central and outer Mn centers or the distances between the Mn cofactors and P865 when the Mn-protein is bound to the reaction center. The Mn proteins constitute an experimental system to examine factors that control the electron

transfer, and such tests are currently under investigation. To gain insight into the distances associated with the Mn-protein when bound to the reaction center, we performed a docking study (Section 7). 7. Docking of the Mn-protein to the reaction center For efficient electron transfer, the distance between P865 and the Mn-center is expected to be between 10 and 20 Å [90–92], which is achieved by placing the Mn-protein on the periplasmic side of the reaction center with the helices approximately perpendicular to the transmembrane helices of the reaction center. The precise position and orientation of the Mn-protein when bound to the reaction center is dependent upon the many interactions between amino acid residues on the two surfaces. We have used a docking program to predict a possible arrangement of the complex of the P0 Mn-protein structure described above (Section 5) and the reaction center structure [40]. The program Rosetta docks the two structures by balancing various protein–protein interactions, including electrostatic, van der Waals, and hydrophobic interactions, as well as hydrogen bonds and solvation energies [93]. The program starts with an initial position of each protein and then generates and scores a thousand different rearrangements of the two proteins within a certain window of changes of the positions and angles relative to the initial configuration. A fundamental assumption is that the backbone conformations of the proteins do not change upon binding. The lowest energy models had the consistent feature of having the two-fold symmetry axis of the Mn-protein approximately perpendicular to the two-fold symmetry axis of the reaction center, as this arrangement allows several interactions between two helices of the P0 protein and the reaction center. In addition, the Mn-protein was approximately centered on the surface of the reaction center in the lowest energy structures. The docking solution provides a probable overall structure of the complex (Fig. 6). In a sense, the Mn-protein is behaving as does the natural secondary electron donor, cytochrome c2, and docking the two structures to maximize their interactions places the dinuclear Mn cofactor at a location similar to that of the heme iron of cytochrome c2 (Fig. 6). In R. sphaeroides, cytochrome c2 binds near the bacteriochlorophyll dimer and reduces the oxidized dimer following light-induced − formation of the P+ 865 Q charge-separated state [40]. The interactions that drive the binding of the cytochrome to the reaction center surface are well mapped, including the determination of the threedimensional structure of the complex [94,95]. An initial attraction through electrostatic interactions is followed by formation of the active complex site involving several different interactions [96]. The versatility of the site is shown by its accommodation of a range of cytochromes from different species, with site-specific mutations of charged residues

Please cite this article as: T.L. Olson, et al., Design of dinuclear manganese cofactors for bacterial reaction centers, Biochim. Biophys. Acta (2015), http://dx.doi.org/10.1016/j.bbabio.2015.09.003

T.L. Olson et al. / Biochimica et Biophysica Acta xxx (2015) xxx–xxx

7

Fig. 6. Structural comparison of reaction centers with a bound cytochrome c2 and with the P0 Mn-protein. The X-ray structure shows the cofactors (red), L (yellow) and M (cyan) subunits of the reaction center with the bound cytochrome c2 (orange) and heme (red) (PDB code 1L9B) (left) [94]. The docking model of the P0 Mn-protein (orange) and dinuclear Mn center (red spheres) shows the cofactors (red), L subunit (yellow) and M subunit (cyan) of the reaction center (right).

having altered dissociation constants but comparable electron transfer rates associated with the bound complex [39,97]. Similar features are evident when comparing the structure of reaction centers with a bound cytochrome c2 and the docking model with the P0 Mn-protein (Fig. 6). In both cases, the secondary donor is approximately centered on the periplasmic surface. For the bound cytochrome c2, the distance between the heme iron and P865 is 18 Å, and the closest distance between the heme and P865 is 11 Å. The relatively short distance between the exposed heme edge and P865 allows efficient electron transfer, with a measured rate of approximately 106 s−1. For the P0 protein, the distance between the Mn center and P865 is longer, at 17 Å, as the Mn cofactor is buried in the protein with no exposed surface. Based upon models of electron transfer [90–92], this increase in the distance to P865 is predicted to decrease the rate to approximately 103 s−1, assuming minor changes in factors such as the reorganization energy. The binding of the Mn-protein in the docking model involves many of the same residues associated with the binding of the cytochrome c2, in particular electrostatic interactions involving carboxylate side chains found on the surface of the reaction center play a role in both cases. 8. Summary Three different de novo four-helix bundles have been designed to bind different combinations of dinuclear Mn centers that are coordinated by carboxylates and imidazoles as found in proteins such as photosystem II. The artificial Mn-proteins are capable of transferring electrons to a natural system, bacterial reaction centers. This mixture of artificial and natural proteins provides the opportunity to examine the oxidation/reduction properties of dinuclear Mn cofactors using light to initiate the redox reactions. The use of de novo designed proteins represents a flexible system that allows protein interactions to be investigated from a well-controlled standpoint without the evolutionary biases normally associated with naturally-occurring proteins. Electron transfer from the dinuclear Mn center in the artificial Mn-protein to the light-induced oxidized bacteriochlorophyll dimer enables a new environment for creating Mn cofactors compared to previous designs for bound mononuclear Mn centers in modified reaction centers. However, water oxidation is a four-electron reaction and in order to move beyond a one-electron transfer, it would be necessary to bind the Mn-protein to the reaction center. Once bound, the Mn-protein should be capable of multiple electron transfers, depending

upon the midpoint potentials for the higher oxidation states, providing the opportunity to investigate how the properties of multinuclear Mn centers control different steps of multielectron transfer. Transparency document The Transparency document associated with this article can be found, in the online version. Acknowledgements This work was supported by the NSF (CHE 1505874) and as part of the Center for Bio-Inspired Solar Fuel Production, an Energy Frontier Research Center funded by the U.S. Department of Energy, Office of Science, Office of Basic Energy Sciences under Award Number DE-SC0001016. The coordinates and X-ray diffraction data have been submitted to the RCSB data bank (PDB ID:5C39). The Berkeley Center for Structural Biology is supported in part by the National Institutes of Health, National Institute of General Medical Sciences, and the Howard Hughes Medical Institute. The Advanced Light Source is supported by the Director, Office of Science, Office of Basic Energy Sciences, of the U.S. Department of Energy under Contract No. DE-AC02-05CH11231. References [1] Y. Umena, K. Kawakami, J.R. Shen, N. Kamiya, Crystal structure of oxygen-evolving photosystem II at a resolution of 1.9 Å, Nature 473 (2011) 55–60. [2] M. Suga, F. Akita, K. Hirata, G. Ueno, H. Murakami, Y. Nakajima, T. Shimizu, K. Yamashita, M. Yamamoto, H. Ago, J.R. Shen, Native structure of photosystem II at 1.95 Å resolution viewed by femtosecond X-ray pulses, Nature 517 (2015) 99–103. [3] N. Cox, D.A. Pantazis, F. Neese, W. Lubitz, Biological water oxidation, Acc. Chem. Res. 46 (2013) 1588–1596. [4] R.D. Britt, P.H. Oyala, One step closer to O2, Science 345 (2014) 736. [5] N. Cox, M. Retegan, F. Neese, D.A. Pantazis, A. Boussac, W. Lubitz, Electronic structure of the oxygen-evolving complex in photosystem II prior to O–O bond formation, Science 345 (2014) 804–808. [6] R. Pokhrel, G.W. Brudvig, Oxygen-evolving complex of photosystem II: correlating structure with spectroscopy, Phys. Chem. Chem. Phys. 16 (2014) 11812–11821. [7] J. Yano, V. Yachandra, Mn4Ca cluster in photosynthesis: where and how water is oxidized to dioxygen, Chem. Rev. 114 (2014) 4175–4205. [8] R.J. Debus, Protein ligation of the photosynthetic oxygen-evolving center, Coord. Chem. Rev. 252 (2008) 244–258. [9] R.J. Service, W. Hillier, R.J. Debus, Network of hydrogen bonds near the oxygenevolving Mn4CaO5 cluster of photosystem II probed with FTIR difference spectroscopy, Biochemistry 53 (2014) 1001–1017.

Please cite this article as: T.L. Olson, et al., Design of dinuclear manganese cofactors for bacterial reaction centers, Biochim. Biophys. Acta (2015), http://dx.doi.org/10.1016/j.bbabio.2015.09.003

8

T.L. Olson et al. / Biochimica et Biophysica Acta xxx (2015) xxx–xxx

[10] R.J. Debus, FTIR studies of metal ligands, networks of hydrogen bonds, and water molecules near the active site Mn4CaO5 cluster in photosystem II, Biochim. Biophys. Acta 1847 (2015) 19–34. [11] F. Rappaport, B.A. Diner, Primary photochemistry and energetics leading to the oxidation of the (Mn)4Ca cluster and to the evolution of molecular oxygen in photosystem II, Coord. Chem. Rev. 252 (2008) 259–272. [12] G. Renger, Light induced oxidative water splitting in photosynthesis: energetics, kinetics, and mechanism, J. Photochem. Photobiol. B 104 (2011) 35–43. [13] R.J. Pace, R. Stranger, S. Petrie, Why nature chose Mn for the water oxidase in photosystem II, Dalton Trans. 41 (2012) 7179–7189. [14] T.J. Wydrzynski, K. Satoh (Eds.), Photosystem II: the light-driven water:plastoquinone oxidoreductaseSpringer, Dordrecht, The Netherlands, 2005. [15] J. Barber, Photosystem II: the water-splitting enzyme of photosynthesis, Cold Spring Harb. Symp. Quant. Biol. 77 (2012) 295–307. [16] T. Cardona, A. Sedoud, N. Cox, A.W. Rutherford, Charge separation in photosystem II: a comparative and evolutionary overview, Biochim. Biophys. Acta 1817 (2012) 26–43. [17] H. Bao, P.L. Dilbeck, R.L. Burnap, Proton transport facilitating water-oxidation: the role of second sphere ligands surrounding the catalytic metal cluster, Photosynth. Res. 116 (2013) 215–229. [18] S. Mukhopadhyay, S.K. Mandal, S. Bhaduri, W.H. Armstrong, Manganese clusters with relevance to photosystem II, Chem. Rev. 104 (2004) 3981–4026. [19] J.P. McEvoy, G.W. Brudvig, Water-splitting chemistry of photosystem II, Chem. Rev. 106 (2006) 4455–4483. [20] C.S. Mullins, V.L. Pecoraro, Reflections on small molecule manganese models that seek to mimic photosynthetic water oxidation chemistry, Coord. Chem. Rev. 252 (2008) 416–443. [21] M.M. Najafpour, S.I. Allakhverdiev, Manganese compounds as water oxidizing catalysts for hydrogen production via water splitting: from manganese complexes to nano-sized manganese oxides, Int. J. Hydrog. Energy 37 (2012) 8753–8764. [22] C.N. Hunter, F. Daldal, M.C. Thurnauer, J.T. Beatty (Eds.), The purple phototrophic bacteriaSpringer, Dordrecht, The Netherlands, 2009. [23] J.P. Allen, J.C. Williams, The evolutionary pathway from anoxygenic to oxygenic photosynthesis examined by comparison of the properties of photosystem II and bacterial reaction centers, Photosynth. Res. 107 (2011) 59–69. [24] M.C. Wakeham, M.R. Jones, Rewiring photosynthesis: engineering wrong-way electron transfer in the purple bacterial reaction centre, Biochem. Soc. Trans. 33 (2005) 851–857. [25] J.C. Williams, J.P. Allen, Directed modification of reaction centers from purple bacteria, in: C.N. Hunter, F. Daldal, M.C. Thurnauer, J.T. Beatty (Eds.), The purple phototrophic bacteriaSpringer, Dordrecht, The Netherlands 2009, pp. 337–353. [26] L. Kressel, K.M. Faries, M.J. Wander, C.E. Zogzas, R.J. Mejdrich, D.K. Hanson, D. Holten, P.D. Laible, C. Kirmaier, High yield of secondary B-side electron transfer in mutant Rhodobacter capsulatus reaction centers, Biochim. Biophys. Acta 1837 (2014) 1892–1903. [27] A. Warshel, Z.T. Chu, W.W. Parson, Dispersed polaron simulations of electron transfer in photosynthetic reaction centers, Science 246 (1989) 112–116. [28] J.N. Gehlen, M. Marchi, D. Chandler, Dynamics affecting the primary charge transfer in photosynthesis, Science 263 (1994) 499–502. [29] H. Wang, S. Lin, J.P. Allen, J.C. Williams, S. Blankert, C. Laser, N.W. Woodbury, Protein dynamics control the kinetics of initial electron transfer in photosynthesis, Science 316 (2007) 747–750. [30] H. Wang, S. Lin, E. Katilius, C. Laser, J.P. Allen, J.C. Williams, N.W. Woodbury, Unusual temperature dependence of photosynthetic electron transfer due to protein dynamics, J. Phys. Chem. B 113 (2009) 818–824. [31] H. Ishikita, W. Saenger, J. Biesiadka, B. Loll, E.W. Knapp, How photosynthetic reaction centers control oxidation power in chlorophyll pairs P680, P700, and P870, Proc. Natl. Acad. Sci. U. S. A. 103 (2006) 9855–9860. [32] M. Szczepaniak, J. Sander, M. Nowaczyk, M.G. Müller, M. Rögner, A.R. Holzwarth, Charge separation, stabilization, and protein relaxation in photosystem II core particles with closed reaction center, Biophys. J. 96 (2009) 621–631. [33] E. Romero, I.H.M. van Stokkum, V.I. Novoderezhkin, J.P. Dekker, R. van Grondelle, Two different charge separation pathways in photosystem II, Biochemistry 49 (2010) 4300–4307. [34] K. Saito, T. Ishida, M. Sugiura, K. Kawakami, Y. Umena, N. Kamiya, J.R. Shen, H. Ishikita, Distribution of the cationic state over the chlorophyll pair of the photosystem II reaction center, J. Am. Chem. Soc. 133 (2011) 14379–14388. [35] J.P. Allen, J.C. Williams, Energetics of cofactors in photosynthetic complexes: relationship between protein-cofactor interactions and midpoint potentials, in: J. Golbeck, A. van der Est (Eds.), The Biophysics of PhotosynthesisSpringer, Dordrecht, The Netherlands 2014, pp. 275–295. [36] M.Y. Okamura, M.L. Paddock, M.S. Graige, G. Feher, Proton and electron transfer in bacterial reaction centers, Biochim. Biophys. Acta 1458 (2000) 148–163. [37] M.R. Gunner, J. Mao, Y. Song, J. Kim, Factors influencing the energetics of electron and proton transfers in proteins. What can be learned from calculations? Biochim. Biophys. Acta 1757 (2006) 942–968. [38] C.A. Wraight, Chance and design–proton transfer in water, channels and bioenergetic proteins, Biochim. Biophys. Acta 1757 (2006) 886–912. [39] M. Tetreault, S.H. Rongey, G. Feher, M.Y. Okamura, Interaction between cytochrome c2 and the photosynthetic reaction center from Rhodobacter sphaeroides: effects of charge-modifying mutations on binding and electron transfer, Biochemistry 40 (2001) 8452–8462. [40] H.L. Axelrod, M.Y. Okamura, The structure and function of the cytochrome c2: reaction center electron transfer complex from Rhodobacter sphaeroides, Photosynth. Res. 85 (2005) 101–114.

[41] J.P. Allen, J.C. Williams, Reaction centers from purple bacteria, in: P. Fromme (Ed.) Photosynthetic Protein Complexes: A Structural ApproachWiley-VCH, Weinheim, Germany 2008, pp. 275–293. [42] M. Thielges, G. Uyeda, A. Cámara-Artigas, L. Kálmán, J.C. Williams, J.P. Allen, Design of a redox-linked active metal site: manganese bound to bacterial reaction centers at a site resembling that of photosystem II, Biochemistry 44 (2005) 7389–7394. [43] L. Kálmán, M.C. Thielges, J.C. Williams, J.P. Allen, Proton release due to manganese binding and oxidation in modified bacterial reaction centers, Biochemistry 44 (2005) 13266–13273. [44] K. Tang, J.C. Williams, J.P. Allen, L. Kálmán, Effect of anions on the binding and oxidation of divalent manganese and iron in modified bacterial reaction centers, Biophys. J. 96 (2009) 3295–3304. [45] L. Kálmán, J.C. Williams, J.P. Allen, Energetics for oxidation of a bound manganese cofactor in modified bacterial reaction centers, Biochemistry 50 (2011) 3310–3320. [46] J.P. Allen, T.L. Olson, P. Oyala, W.J. Lee, A.A. Tufts, J.C. Williams, Light-driven oxygen production from superoxide by Mn-binding bacterial reaction centers, Proc. Natl. Acad. Sci. U. S. A. 109 (2012) 2314–2318. [47] A.F.A. Peacock, Incorporating metals into de novo proteins, Curr. Opin. Chem. Biol. 17 (2013) 934–939. [48] M.L. Zastrow, V.L. Pecoraro, Designing functional metalloproteins: from structural to catalytic metal sites, Coord. Chem. Rev. 257 (2013) 2565–2588. [49] J.R. Calhoun, F. Nastri, O. Maglio, V. Pavone, A. Lombardi, W.F. DeGrado, Artificial diiron proteins: from structure to function, Biopolymers 80 (2005) 264–278. [50] R.L. Koder, P.L. Dutton, Intelligent design: the de novo engineering of proteins with specified functions, Dalton Trans. 2006 (2006) 3045–3051. [51] T. Wydrzynski, W. Hillier, B. Conlan, Engineering model proteins for photosystem II function, Photosynth. Res. 94 (2007) 225–233. [52] T.A. Farid, G. Kodali, L.A. Solomon, B.R. Lichtenstein, M.M. Sheehan, B.A. Fry, C. Bialas, N.M. Ennist, J.A. Siedlecki, Z. Zhao, M.A. Stetz, K.G. Valentine, J.L.R. Anderson, A.J. Wand, B.M. Discher, C.C. Moser, P.L. Dutton, Elementary tetrahelical protein design for diverse oxidoreductase functions, Nat. Chem. Biol. 9 (2013) 826–833. [53] O. Maglio, F. Nastri, R. Torres Martin de Rosales, M. Faiella, V. Pavone, W.F. DeGrado, A. Lombardi, Diiron-containing metalloproteins: developing functional models, C. R. Chim. 10 (2007) 703–720. [54] B.R. Gibney, F. Rabanal, K.S. Reddy, P.L. Dutton, Effect of four helix bundle topology on heme binding and redox properties, Biochemistry 37 (1998) 4635–4643. [55] A. Lombardi, C.M. Summa, S. Geremia, L. Randaccio, V. Pavone, W.F. DeGrado, Retrostructural analysis of metalloproteins: application to the design of a minimal model for diiron proteins, Proc. Natl. Acad. Sci. U. S. A. 97 (2000) 6298–6305. [56] L. Di Costanzo, H. Wade, S. Geremia, L. Randaccio, V. Pavone, W.F. DeGrado, A. Lombardi, Toward the de novo design of a catalytically active helix bundle: a substrate-accessible carboxylate-bridged dinuclear metal center, J. Am. Chem. Soc. 123 (2001) 12749–12757. [57] S. Geremia, L. Di Costanzo, L. Randaccio, D.E. Engel, A. Lombardi, F. Nastri, W.F. DeGrado, Response of a designed metalloprotein to changes in metal ion coordination, exogenous ligands, and active site volume determined by X-ray crystallography, J. Am. Chem. Soc. 127 (2005) 17266–17276. [58] A. Magistrato, W.F. DeGrado, A. Laio, U. Rothlisberger, J. VandeVondele, M.L. Klein, Characterization of the dizinc analogue of the synthetic diiron protein DF1 using ab initio and hybrid quantum/classical molecular dynamics simulations, J. Phys. Chem. B 107 (2003) 4182–4188. [59] G.A. Papoian, W.F. DeGrado, M.L. Klein, Probing the configurational space of a metalloprotein core: an ab initio molecular dynamics study of Duo Ferro 1 binuclear Zn cofactor, J. Am. Chem. Soc. 125 (2003) 560–569. [60] A. Pasternak, J. Kaplan, J.D. Lear, W.F. DeGrado, Proton and metal ion-dependent assembly of a model diiron protein, Protein Sci. 10 (2001) 958–969. [61] C.M. Summa, M.M. Rosenblatt, J.K. Hong, J.D. Lear, W.F. DeGrado, Computational de novo design, and characterization of an A2B2 diiron protein, J. Mol. Biol. 321 (2002) 923–938. [62] E.N.G. Marsh, W.F. DeGrado, Noncovalent self-assembly of a heterotetrameric diiron protein, Proc. Natl. Acad. Sci. U. S. A. 99 (2002) 5150–5154. [63] J.R. Calhoun, H. Kono, S. Lahr, W. Wang, W.F. DeGrado, J.G. Saven, Computational design and characterization of a monomeric helical dinuclear metalloprotein, J. Mol. Biol. 334 (2003) 1101–1115. [64] S.J. Lahr, D.E. Engel, S.E. Stayrook, O. Maglio, B. North, S. Geremia, A. Lombardi, W.F. DeGrado, Analysis and design of turns in α-helical hairpins, J. Mol. Biol. 346 (2005) 1441–1454. [65] O. Maglio, F. Nastri, J.R. Calhoun, S. Lahr, H. Wade, V. Pavone, W.F. DeGrado, A. Lombardi, Artificial di-iron proteins: solution characterization of four helix bundles containing two distinct types of inter-helical loops, J. Biol. Inorg. Chem. 10 (2005) 539–549. [66] P.P. Wei, A.J. Skulan, H. Wade, W.F. DeGrado, E.I. Solomon, Spectroscopic and computational studies of the de novo designed protein DF2t: correlation to the biferrous active site of ribonucleotide reductase and factors that affect O2 reactivity, J. Am. Chem. Soc. 127 (2005) 16098–16106. [67] H. Wade, S.E. Stayrook, W.F. Degrado, The structure of a designed diiron(III) protein: implications for cofactor stabilization and catalysis, Angew. Chem. Int. Ed. Engl. 45 (2006) 4951–4954. [68] O. Maglio, F. Nastri, V. Pavone, A. Lombardi, W.F. DeGrado, Preorganization of molecular binding sites in designed diiron proteins, Proc. Natl. Acad. Sci. U. S. A. 100 (2003) 3772–3777. [69] W.F. DeGrado, L. Di Costanzo, S. Geremia, A. Lombardi, V. Pavone, L. Randaccio, Sliding helix and change of coordination geometry in a model di-MnII protein, Angew. Chem. Int. Ed. Engl. 42 (2003) 417–420.

Please cite this article as: T.L. Olson, et al., Design of dinuclear manganese cofactors for bacterial reaction centers, Biochim. Biophys. Acta (2015), http://dx.doi.org/10.1016/j.bbabio.2015.09.003

T.L. Olson et al. / Biochimica et Biophysica Acta xxx (2015) xxx–xxx [70] J.R. Calhoun, W. Liu, K. Spiegel, M. Dal Peraro, M.L. Klein, K.G. Valentine, A.J. Wand, W.F. DeGrado, Solution NMR structure of a designed metalloprotein and complementary molecular dynamics refinement, Structure 16 (2008) 210–215. [71] M. Faiella, C. Andreozzi, R. Torres Martin de Rosales, V. Pavone, O. Maglio, F. Nastri, W.F. DeGrado, A.A. Lombardi, An artificial di-iron oxo-protein with phenol oxidase activity, Nat. Chem. Biol. 5 (2009) 882–884. [72] J. Kaplan, W.F. DeGrado, De novo design of catalytic proteins, Proc. Natl. Acad. Sci. U. S. A. 101 (2004) 11566–11570. [73] J.R. Calhoun, C.B. Bell, T.J. Smith, T.J. Thamann, W.F. DeGrado, E.I. Solomon, Oxygen reactivity of the biferrous site in the de novo designed four helix bundle peptide DFsc: nature of the “intermediate” and reaction mechanism, J. Am. Chem. Soc. 130 (2008) 9188–9189. [74] C.B. Bell, J.R. Calhoun, E. Bobyr, P.P. Wei, B. Hedman, K.O. Hodgson, W.F. Degrado, E.I. Solomon, Spectroscopic definition of the biferrous and biferric sites in de novo designed four-helix bundle DFsc peptides: implications for O2 reactivity of binuclear non-heme iron enzymes, Biochemistry 48 (2009) 59–73. [75] R. Torres Martin de Rosales, M. Faiella, E. Farquhar, L. Que, C. Andreozzi, V. Pavone, O. Maglio, F. Nastri, A.A. Lombardi, Spectroscopic and metal-binding properties of DF3: an artificial protein able to accommodate different metal ions, J. Biol. Inorg. Chem. 15 (2010) 717–728. [76] A.J. Reig, M.M. Pires, R.A. Snyder, Y. Wu, H. Jo, D.W. Kulp, S.E. Butch, J.R. Calhoun, T. Szyperski, E.I. Solomon, W.F. DeGrado, Alteration of the oxygen-dependent reactivity of de novo Due Ferri proteins, Nat. Chem. 4 (2012) 900–906. [77] A. Dautant, J.B. Meyer, J. Yariv, G. Précigoux, R.M. Sweet, A.J. Kalb (Gilboa), F. Frolow, Structure of a monoclinic crystal form of cytochrome b1 (bacterioferritin) from E. coli, Acta Crystallogr. D54 (1998) 16–24. [78] B. Conlan, N. Cox, J.H. Su, W. Hillier, J. Messinger, W. Lubitz, P.L. Dutton, T. Wydrzynski, Photo-catalytic oxidation of a di-nuclear manganese centre in an engineered bacterioferritin ‘reaction centre, Biochim. Biophys. Acta 1787 (2009) 1112–1121. [79] K. Hingorani, R. Pace, S. Whitney, J.W. Murray, P. Smith, M.H. Cheah, T. Wydrzynski, W. Hillier, Photo-oxidation of tyrosine in a bio-engineered bacterioferritin ‘reaction centre’—a protein model for artificial photosynthesis, Biochim. Biophys. Acta 1837 (2014) 1821–1834. [80] F.A. Momany, R. Rone, Validation of the general purpose QUANTA 3.2/CHARMm force field, J. Comput. Chem. 13 (1992) 888–900. [81] M. Flores, T.L. Olson, D. Wang, S. Edwardraja, S. Shinde, J.C. Williams, G. Ghirlanda, J.P. Allen, The copper environment in artificial metalloproteins probed by EPR spectroscopy, J. Phys. Chem. B (2015) (in press). [82] Z. Otwinowski, W. Minor, Processing of X-ray diffraction data collected in oscillation mode, Methods Enzymol. 276 (1997) 307–326.

9

[83] Collaborative Computational Project, Number 4, the CCP4 suite: programs for protein crystallography, Acta Crystallogr. D50 (1994) 760–763. [84] P.D. Adams, R.W. Grosse-Kunstleve, L.W. Hung, T.R. Ioerger, A.J. McCoy, N.W. Moriarty, R.J. Read, J.C. Sacchettini, N.K. Sauter, T.C. Terwilliger, PHENIX: building new software for automated crystallographic structure determination, Acta Crystallogr. D58 (2002) 1948–1954. [85] W.L. DeLano, The PyMOL molecular graphics system, DeLano Scientific, San Carlos, CA, 2002. [86] P. Emsley, B. Lohkamp, W.G. Scott, K. Cowtan, Features and development of Coot, Acta Crystallogr. D66 (2010) 486–501. [87] X. Lin, H.A. Murchison, V. Nagarajan, W.W. Parson, J.P. Allen, J.C. Williams, Specific alteration of the oxidation potential of the electron donor in reaction centers from Rhodobacter sphaeroides, Proc. Natl. Acad. Sci. U. S. A. 91 (1994) 10265–10269. [88] L. Kálmán, R. LoBrutto, J.P. Allen, J.C. Williams, Manganese oxidation by modified reaction centers from Rhodobacter sphaeroides, Biochemistry 42 (2003) 11016–11022. [89] M. Högbom, U.B. Ericsson, R. Lam, H.M.A. Bakali, E. Kuznetsova, P. Nordlund, D.B. Zamble, A high throughput method for the detection of metalloproteins on a microgram scale, Mol. Cell. Proteomics 4 (2005) 827–834. [90] R.A. Marcus, N. Sutin, Electron transfers in chemistry and biology, Biochim. Biophys. Acta 811 (1985) 265–322. [91] D.N. Beratan, J.N. Onuchic, J.R. Winkler, H.B. Gray, Electron-tunneling pathways in proteins, Science 258 (1992) 1740–1741. [92] C.C. Moser, J.M. Keske, K. Warncke, R.S. Farid, P.L. Dutton, Nature of biological electron transfer, Nature 355 (1992) 796–802. [93] S. Lyskov, J.J. Gray, The RosettaDock server for local protein–protein docking, Nucleic Acids Res. 36 (2008) W233–W238. [94] H.L. Axelrod, E.C. Abresch, M.Y. Okamura, A.P. Yeh, D.C. Rees, G. Feher, X-ray structure determination of the cytochrome c2: reaction center electron transfer complex from Rhodobacter sphaeroides, J. Mol. Biol. 319 (2002) 501–515. [95] E.C. Abresch, X.M. Gong, M.L. Paddock, M.Y. Okamura, Electron transfer from cytochrome c2 to the reaction center: a transition state model for ionic strength effects due to neutral mutations, Biochemistry 48 (2009) 11390–11398. [96] O. Miyashita, J.N. Onuchic, M.Y. Okamura, Transition state and encounter complex for fast association of cytochrome c2 with bacterial reaction center, Proc. Natl. Acad. Sci. U. S. A. 101 (2004) 16174–16179. [97] D.M. Tiede, A.C. Vashishta, M.R. Gunner, Electron-transfer kinetics and electrostatic properties of the Rhodobacter sphaeroides reaction center and soluble c-cytochromes, Biochemistry 32 (1993) 4515–4531.

Please cite this article as: T.L. Olson, et al., Design of dinuclear manganese cofactors for bacterial reaction centers, Biochim. Biophys. Acta (2015), http://dx.doi.org/10.1016/j.bbabio.2015.09.003