Rationalizing the Geometries of the Water Oxidising Complex in the Atomic Resolution, Nominal S3 State Crystal Structures of Photosystem II.

Three atomic resolution crystal structures of Photosystem II, in the double flashed, nominal S3 intermediate state of its Mn4 Ca Water Oxidising Complex (WOC), have now been presented, at 2.25, 2.35 and 2.08 Šresolution. Although very similar overall, the S3 structures differ within the WOC catalytic site. The 2.25 Šstructure contains only one oxy species (O5) in the WOC cavity, weakly associated with Mn centres, similar to that in the earlier 1.95 ŠS1 structure. The 2.35 Šstructure shows two such species (O5, O6), with the Mn centres and O5 positioned as in the 2.25 Šstructure and O5-O6 separation of ∼1.5 Å. In the latest S3 variant, two oxy species are also seen (O5, Ox), with the Ox group appearing only in S3 , closely ligating one Mn, with O5-Ox separation <2.1 Å. The O5 and O6/Ox groups were proposed to be substrate water derived species. Recently, Petrie et al. (Chem. Phys. Chem., 2017) presented large scale Quantum Chemical modelling of the 2.25 Šstructure, quantitatively explaining all significant features within the WOC region. This, as in our earlier studies, assumed a 'low' Mn oxidation paradigm (mean S1 Mn oxidation level of +3.0, Petrie et al., Angew. Chem. Int. Ed., 2015), rather than a 'high' oxidation model (mean S1 oxidation level of +3.5). In 2018 we showed (Chem. Phys. Chem., 2018) this oxidation state assumption predicted two energetically close S3 structural forms, one with the metal centres and O5 (as OH- ) positioned as in the 2.25 Šstructure, and the other with the metals similarly placed, but with O5 (as H2 O) located in the O6 position of the 2.35 Šstructure. The 2.35 Štwo flashed structure was likely a crystal superposition of two such forms. Here we show, by similar computational analysis, that the latest 2.08 ŠS3 structure is also a likely superposition of forms, but with O5 (as OH- ) occupying either the O5 or Ox positions in the WOC cavity. This highlights a remarkable structural 'lability' of the WOC centre in the S3 state, which is likely catalytically relevant to its water splitting function.

Explaining the Different Geometries of the Water Oxidising Complex in the Nominal S3 State Crystal Structures of Photosystem II at 2.25†Šand 2.35†Å.

Recently two atomic resolution crystal structures of Photosystem II, in the double flashed, nominal S3 intermediate state of its Mn4 Ca water oxidising complex (WOC), have been presented (Young et al., Nature 2016, 540, 453; Suga et al., Nature 2017, 543, 131). These structures are at 2.25 Šand 2.35 Šresolution, respectively. Although highly similar in most respects, the structures differ in a key region within the WOC catalytic site. In the 2.25 Šstructure, one oxy species (O5) is observed within the WOC cavity, weakly associated with the Mn centres, similar to that seen earlier in the 1.95 ŠXRD structure of the S1 intermediate (Suga et al., Nature, 2015, 517, 99). In the 2.35 Šstructure, two such species are seen (O5, O6), with the Mn centres and O5 positioned as in the 2.25 Šstructure and an O5-O6 separation of ∼1.5 Å, consistent with peroxo formation. This suggests O5 and O6 are substrate water derived species in this double flashed form. Recently we have presented (Petrie, et al., Chem. Phys. Chem., 2017) a large scale (220 atom) quantum chemical model of the Young et al. 2.25 Šstructure, which quantitatively explains all significant features within the WOC region of that structure, particularly the positions of the metal centres and O5 group. Critical to this was our assumption of a 'low' Mn oxidation paradigm (mean S1 Mn oxidation level of +3.0, Petrie et al., Angew. Chem. Int. Ed., 2015), rather than a 'high' oxidation model (mean S1 oxidation level of +3.5), widely assumed in the literature. Here we show that our same oxidation state model predicts two classes of energetically close S3 structural forms, analogous to the S1 state, one with the metal centres and O5 positioned as in the 2.25 Šstructure, and the other with the metals similarly placed, but with O5 located in the O6 position of the 2.35 Šstructure. We show that the Suga et al. 2.35 Šstructure is likely a superposition of two such forms, one from each class, which is consistent with reported atomic occupancies for that structure and the relative total energies we calculate for the two structural forms.

What Mn Kß spectroscopy reveals concerning the oxidation states of the Mn cluster in photosystem II.

The oxygen evolving complex, (OEC) in Photosystem II contains a Mn4Ca cluster and catalyses oxidation of water to molecular oxygen and protons, the most energetically demanding reaction in nature. The catalytic mechanism remains unresolved and the precise Mn oxidation levels through which the cluster cycles during functional turnover are controversial. Two proposals for these redox levels exist; the 'high' and 'low' oxidation state paradigms, which differ systematically by two oxidation equivalents throughout the redox accumulating catalytic S state cycle (states S0…S3). Presently the 'high' paradigm is more favored. For S1 the assumed mean redox levels of Mn are 3.5 (high) and 3.0 (low) respectively. Mn K region X-ray spectroscopy has been extensively used to examine the OEC Mn oxidation levels, with Kß emission spectroscopy increasingly the method of choice. Here we review the results from application of this and closely related techniques to PS II, building on our earlier examination of these and other data on the OEC oxidation states (Pace et al., Dalton Trans., 2012, 41, 11145). We compare the most recent results with a range of earlier Mn Kß experiments on the photosystem and related model Mn systems. New analyses of these data are given, highlighting certain key spectral considerations which appear not to have been sufficiently appreciated earlier. These show that the recent and earlier PS II Kß results have a natural internal consistency, leading to the strong conclusion that the low paradigm oxidation state assignment for the functional OEC is favoured.

Rationalizing the 2.25†ŠResolution Crystal Structure of the Water Oxidising Complex of Photosystem†II in the S3 State.

Quantum chemical calculations are described which rationalize the recent X-ray diffraction (XRD) structure at 2.25 Šof the Mn4 Ca water oxidising complex (WOC) of photosystem II (PSII) in the S3 intermediate state. The new S3 XRD structure shows remarkable similarity to earlier atomic resolution (1.9, 1.95 Å) WOC structures in the dark stable S1 state and is inconsistent with most current proposals, from computational chemistry and other sources, regarding the Mn oxidation state levels in the WOC cluster and the nature of water substrate binding, particularly in S3 . This mirrors earlier failures to rationalise the WOC geometry in the 1.9 and 1.95 ŠS1 XRD structures, assuming "high" paradigm Mn oxidation models. However, we recently showed that a lower Mn oxidation assumption closely reproduces the S1 XRD structures, computationally. This same "low" Mn oxidation model, now computationally applied in S3 , not only reproduces the latest 2.25 ŠXRD structure but also rationalises a number of other important, experimental features of the WOC, including the metal-metal distances inferred from EXAFS studies as well as earlier S3 state XRD structures of lower resolution (4-5 Å). As found previously for S1 , the WOC in the S3 state is computationally revealed to be structurally variable, consistent with some EXAFS and lower-resolution XRD data. This is a direct consequence of at least two MnIII ions being present in all metastable S states.

What computational chemistry and magnetic resonance reveal concerning the oxygen evolving centre in Photosystem II.

Density Functional Theory (DFT) computational studies of the Mn4/Ca Oxygen Evolving Complex (OEC) region of Photosystem II in the paramagnetic S2 and S3 states of the water oxdizing catalytic cycle are described. These build upon recent advances in computationally understanding the detailed S1 state OEC geometries, revealed by the recent high resolution Photosystem II crystal structures of Shen et al., at 1.90Å and 1.95Å (Petrie et al., 2015, Angew. Chem. Int. Ed., 54, 7120). The models feature a 'Low Oxidation Paradigm' assumption for the mean Mn oxidation states in the functional enzyme, with the mean oxidation levels being 3.0, 3.25 and 3.5 in S1, S2 and S3, respectively. These calculations are used to infer magnetic exchange interactions within the coupled OEC cluster, particularly in the Electron Paramagnetic Resonance (EPR)-visible S2 and S3 states. Detailed computational estimates of the intrinsic magnitudes and molecular orientations of the 55Mn hyperfine tensors in the S2 state are presented. These parameters, together with the resultant spin projected hyperfine values are compared with recent appropriate experimental EPR data (Continuous Wave (CW), Electron-Nuclear Double Resonance (ENDOR) and ELDOR (Electron-Electron Double Resonance)-Detected Nuclear Magnetic Resonance (EDNMR)) from the OEC. It is found that an effective Coupled Dimer magnetic organization of the four Mn in the OEC cluster in the S2 and S3 states is able to quantitatively rationalize the observed 55Mn hyperfine data. This is consistent with structures we propose to represent the likely state of the OEC in the catalytically active form of the enzyme.

Deprotonation of Water/Hydroxo Ligands in Clusters Mimicking the Water Oxidizing Complex of PSII and Its Effect on the Vibrational Frequencies of Ligated Carboxylate Groups.

The IR absorptions of several first-shell carboxylate ligands of the water oxidizing complex (WOC) have been experimentally shown to be unaffected by oxidation state changes in the WOC during its catalytic cycle. Several model clusters that mimic the Mn4O5Ca core of the WOC in the S1 state, with electronic configurations that correspond to both the so-called "high" and "low" oxidation paradigms, were investigated. Deprotonation at W2, W1, or O3 sites was found to strongly reduce carboxylate ligand frequency shifts on oxidation of the metal cluster. The frequency shifts were smallest in neutrally charged clusters where the initial mean Mn oxidation state was +3, with W2 as an hydroxide and O5 a water. Deprotonation also reduced and balanced the oxidation energy of all clusters in successive oxidations.

Resolving the Differences Between the 1.9†Šand 1.95†ŠCrystal Structures of Photosystem†II: A Single Proton Relocation Defines Two Tautomeric Forms of the Water-Oxidizing Complex.

Great progress has been made in characterizing the water-oxidizing complex (WOC) in photosystem II (PSII) with the publication of a 1.9 Šresolution X-ray diffraction (XRD) and recently a 1.95 ŠX-ray free-electron laser (XFEL) structure. However, these achievements are under threat because of perceived conflicts with other experimental data. For the earlier 1.9 Šstructure, lack of agreement with extended X-ray absorption fine structure (EXAFS) data led to the notion that the WOC suffered from X-ray photoreduction. In the recent 1.95 Šstructure, Mn photoreduction is not an issue, but poor agreement with computational models which adopt the 'high' oxidation state paradigm, has again resulted in criticism of the structure on the basis of contamination with lower S states of the WOC. Here we use DFT modeling to show that the distinct WOC geometries in the 1.9 and 1.95 Šstructures can be straightforwardly accounted for when the Mn oxidation states are consistent with the 'low' oxidation state paradigm. Remarkably, our calculations show that the two structures are tautomers, related by a single proton relocation.

Rationalising the geometric variation between the A and B monomers in the 1.9†Šcrystal structure of photosystem†II.

Density functional theory calculations are reported on a set of models of the water-oxidising complex (WOC) of photosystem II (PSII), exploring structural features revealed in the most recent (1.9 Šresolution) X-ray crystallographic studies of PSII. Crucially, we find that the variation in the Mn-Mn distances seen between the A and B monomers of this crystal structure can be entirely accounted for, in the low oxidation state (LOS) paradigm, by consideration of the interplay between two hydrogen-bonding interactions involving proximate amino acid residues with the oxo bridges of the WOC, that is, His337 with O3 (which leads to a general elongation in the Mn-Mn distances between Mn1, Mn2 and Mn3) and Arg357 with O2 (which results in a specific elongation of the Mn2-Mn3 distance).

Ab Initio modeling of the effect of oxidation coupled with HnO deprotonation on carboxylate ligands in Mn/Ca clusters.

Oxidation of some manganese complexes containing both carboxylate and water/hydroxo ligands does not result in changes to the carboxylate stretching frequencies. The water oxidizing complex of photosystem II is one motivating example. On the basis of electronic structure theory calculations, we here suggest that the deprotonation of water or hydroxo ligands minimizes changes in the vibrational frequencies of coligating carboxylates, rendering the carboxylate modes "invisible" in FTIR difference spectroscopy. This deprotonation of water/hydroxo ligands was also found to balance the redox potentials of the Mn(II)/Mn(III) and Mn(III)/Mn(IV) couples, allowing the possibility for successive manganese oxidations at a relatively constant redox potential.

Electronic structure of the oxygen evolving complex in photosystem II, as revealed by 55Mn Davies ENDOR studies at 2.5 K.

We report the first (55)Mn pulsed ENDOR studies on the S2 state multiline spin ½ centre of the oxygen evolving complex (OEC) in Photosystem II (PS II), at temperatures below 4.2 K. These were performed on highly active samples of spinach PS II core complexes, developed previously in our laboratories for photosystem spectroscopic use, at temperatures down to 2.5 K. Under these conditions, relaxation effects which have previously hindered observation of most of the manganese ENDOR resonances from the OEC coupled Mn cluster are suppressed. (55)Mn ENDOR hyperfine couplings ranging from ∼50 to ∼680 MHz are now seen on the S2 state multiline EPR signal. These, together with complementary high resolution X-band CW EPR measurements and detailed simulations, reveal that at least two and probably three Mn hyperfine couplings with large anisotropy are seen, indicating that three Mn(III) ions are likely present in the functional S2 state of the enzyme. This suggests a low oxidation state paradigm for the OEC (mean Mn oxidation level 3.0 in the S1 state) and unexpected Mn exchange coupling in the S2 state, with two Mn ions nearly magnetically silent. Our results rationalize a number of previous ligand ESEEM/ENDOR studies and labelled water exchange experiments on the S2 state of the photosystem, in a common picture which is closely consistent with recent photo-assembly (Kolling et al., Biophys. J. 2012, 103, 313-322) and large scale computational studies on the OEC (Gatt et al., Angew. Chem., Int. Ed. 2012, 51, 12025-12028, Kurashige et al. Nat. Chem. 2013, 5, 660-666).

What does the Sr-substituted 2.1 Å resolution crystal structure of photosystem II reveal about the water oxidation mechanism?

A density functional study of the Sr-substituted photosystem II water oxidising complex demonstrates that its recent X-ray crystal structure is consistent with a (Mn(III))4 oxidation state pattern, and with a Sr-bound hydroxide ion. The Sr-water-hydroxide interactions rationalize differences in the exchange rates of substrate water and kinetics of dioxygen bond formation relative to the Ca-containing structure.

Computational design of a carbon nanotube fluorofullerene biosensor.

Carbon nanotubes offer exciting opportunities for devising highly-sensitive detectors of specific molecules in biology and the environment. Detection limits as low as 10(-11) M have already been achieved using nanotube-based sensors. We propose the design of a biosensor comprised of functionalized carbon nanotube pores embedded in a silicon-nitride or other membrane, fluorofullerene-Fragment antigen-binding (Fab fragment) conjugates, and polymer beads with complementary Fab fragments. We show by using molecular and stochastic dynamics that conduction through the (9, 9) exohydrogenated carbon nanotubes is 20 times larger than through the Ion Channel Switch ICS(TM) biosensor, and fluorofullerenes block the nanotube entrance with a dissociation constant as low as 37 pM. Under normal operating conditions and in the absence of analyte, fluorofullerenes block the nanotube pores and the polymer beads float around in the reservoir. When analyte is injected into the reservoir the Fab fragments attached to the fluorofullerene and polymer bead crosslink to the analyte. The drag of the much larger polymer bead then acts to pull the fluorofullerene from the nanotube entrance, thereby allowing the flow of monovalent cations across the membrane. Assuming a tight seal is formed between the two reservoirs, such a biosensor would be able to detect one channel opening and thus one molecule of analyte making it a highly sensitive detection design.

What are the oxidation states of manganese required to catalyze photosynthetic water oxidation?

Photosynthetic O(2) production from water is catalyzed by a cluster of four manganese ions and a tyrosine residue that comprise the redox-active components of the water-oxidizing complex (WOC) of photosystem II (PSII) in all known oxygenic phototrophs. Knowledge of the oxidation states is indispensable for understanding the fundamental principles of catalysis by PSII and the catalytic mechanism of the WOC. Previous spectroscopic studies and redox titrations predicted the net oxidation state of the S(0) state to be (Mn(III))(3)Mn(IV). We have refined a previously developed photoassembly procedure that directly determines the number of oxidizing equivalents needed to assemble the Mn(4)Ca core of WOC during photoassembly, starting from free Mn(II) and the Mn-depleted apo-WOC complex. This experiment entails counting the number of light flashes required to produce the first O(2) molecules during photoassembly. Unlike spectroscopic methods, this process does not require reference to synthetic model complexes. We find the number of photoassembly intermediates required to reach the lowest oxidation state of the WOC, S(0), to be three, indicating a net oxidation state three equivalents above four Mn(II), formally (Mn(III))(3)Mn(II), whereas the O(2) releasing state, S(4), corresponds formally to (Mn(IV))(3)Mn(III). The results from this study have major implications for proposed mechanisms of photosynthetic water oxidation.

What spectroscopy reveals concerning the Mn oxidation levels in the oxygen evolving complex of photosystem II: X-ray to near infra-red.

Photosystem II (PS II), found in oxygenic photosynthetic organisms, catalyses the most energetically demanding reaction in nature, the oxidation of water to molecular oxygen and protons. The water oxidase in PS II contains a Mn(4)Ca cluster (oxygen evolving complex, OEC), whose catalytic mechanism has been extensively investigated but is still unresolved. In particular the precise Mn oxidation levels through which the cluster cycles during functional turnover are still contentious. In this, the first of several planned parts, we examine a broad range of published data relating to this question, while considering the recent atomic resolution PS II crystal structure of Umena et al. (Nature, 2011, 473, 55). Results from X-ray, UV-Vis and NIR spectroscopies are considered, using an approach that is mainly empirical, by comparison with published data from known model systems, but with some reliance on computational or other theoretical considerations. The intention is to survey the extent to which these data yield a consistent picture of the Mn oxidation states in functional PS II - in particular, to test their consistency with two current proposals for the mean redox levels of the OEC during turnover; the so called 'high' and 'low' oxidation state paradigms. These systematically differ by two oxidation equivalents throughout the redox accumulating catalytic S state cycle (states S(0)···S(3)). In summary, we find that the data, in total, substantially favor the low oxidation proposal, particularly as a result of the new analyses we present. The low oxidation state scheme is able to resolve a number of previously 'anomalous' results in the observed UV-Visible S state turnover spectral differences and in the resonant inelastic X-ray spectroscopy (RIXS) of the Mn pre-edge region of the S(1) and S(2) states. Further, the low oxidation paradigm is able to provide a 'natural' explanation for the known sensitivity of the OEC Mn cluster to cryogenic near infra-red (NIR) induced turnover to alternative spin/redox states in S(2) and S(3).

Modelling the metal atom positions of the Photosystem II water oxidising complex: a density functional theory appraisal of the 1.9 Å resolution crystal structure.

Density functional theory (DFT) calculations are reported for a set of model compounds intended to represent the structure of the Photosystem II (PSII) water oxidising complex (WOC) as determined by the recent 1.9 Å resolution single crystal X-ray diffraction (XRD) study of Umena et al. In contrast with several other theoretical studies addressing this structure, we find that it is not necessary to invoke photoreduction of the crystalline sample below the S(1)'resting state' in order to rationalise the observed WOC geometry. Our results are consistent with crystallised PSII in the S(1) state, with S(1) corresponding to either (Mn(III))(4) or (Mn(III))(2)(Mn(IV))(2) as required by the two competing paradigms for the WOC oxidation state pattern. Of these two paradigms, the 'low-oxidation-state' paradigm provides a better match for the crystal structure, with the comparatively long Mn(2)-Mn(3) distance in particular proving difficult to reconcile with the 'high-oxidation-state' model. Best agreement with the set of metal-metal distances is obtained with a S(1) model featuring µ-O, µ-OH bridging between Mn(3) and Mn(4) and deprotonation of one water ligand on Mn(4). Theoretical modelling of the 1.9 Å structure is an important step in assessing the validity of this recent crystal structure, with implications for our understanding of the mechanism of water oxidation by PSII.

Why nature chose Mn for the water oxidase in Photosystem II.

Nature performs a vital but uniquely energetic reaction within Photosystem II (PS II), resulting in the oxidation of two water molecules to yield O(2) and bio-energetic electrons, as reducing equivalents. Almost all life on earth ultimately depends on this chemistry, which occurs with remarkable efficiency within a tetramanganese and calcium cluster in the photosystem. The thermodynamic constraints for the operation of this water oxidising Mn(4)/Ca cluster within PS II are discussed. These are then examined in the light of the known redox chemistry of hydrated Mn-oxo systems and relevant model compounds. It is shown that the latest high resolution crystal structure of cyanobacterial PS II suggests an organization of the tetra-nuclear Mn cluster that naturally accommodates the stringent requirements for successive redox potential constancy with increasing total oxidation state, which the enzyme function imposes. This involves one region of the Mn(4)/Ca cluster being dominantly involved with substrate water binding, while a separate, single Mn is principally responsible for the redox accumulation function. Recent high level computational chemical investigations by the authors strongly support this, with a computed pattern of Mn oxidation states throughout the catalytic cycle being completely consistent with this interpretation. Strategies to design synthetic, bio-mimetic constructs utilising this approach for efficient electrolytic generation of hydrogen fuel within Artificial Photosynthesis are briefly discussed.

The interaction of His337 with the Mn4Ca cluster of photosystem II.

The most recent XRD studies of Photosystem II (PS II) reveal that the His337 residue is sufficiently close to the Mn(4)Ca core of the Water Oxidising Complex (WOC) to engage in H-bonding interactions with the µ(3)-oxo bridge connecting Mn(1), Mn(2) and Mn(3). Such interactions may account for the lengthening of the Mn-Mn distances observed in the most recent and highest resolution (1.9 Å) crystal structure of PS II compared to earlier, lower-resolution (2.9 Å or greater) XRD structures and EXAFS studies on functional PS II. Density functional theory is used to examine the influence on Mn-Mn distances of H-bonding interactions, mediated by the proximate His337 residue, which may lead to either partial or complete protonation of the µ(3)-oxo bridge on models of the WOC. Calculations were performed on a set of minimal-complexity models (in which WOC-ligating amino acid residues are represented as formate and imidazole ligands), and also on extended models in which a 13-peptide sequence (from His332 to Ala344) is treated explicitly. These calculations demonstrate that while the 2.9 Å structure is best described by models in which the µ(3)-oxo bridge is neither protonated nor involved in significant H-bonding, the 1.9 Å XRD structure is better reproduced by models in which the µ(3)-oxo bridge undergoes H-bonding interactions with the His337 residue leading to expansion of the 'close' Mn-Mn distances well known from EXAFS studies at ∼ 2.7 Å. Furthermore, full µ(3)-oxo-bridge protonation remains a distinct possibility during the process of water oxidation, as evidenced by the lengthening of the Mn-Mn vectors observed in EXAFS studies of the higher oxidation states of PS II. In this context, the Mn-Mn distances calculated in the protonated µ(3)-oxo bridge structures, particularly for the peptide extended models, are in close agreement with the EXAFS data.

Application of computational chemistry to understanding the structure and mechanism of the Mn catalytic site in photosystem II--a review.

Applications of Density Functional Theory (DFT) computational techniques to studies of the molecular structure and mechanism of the oxygen evolving, water oxidising Mn(4)/Ca catalytic site in Photosystem II are reviewed. We summarise results from the earlier studies (pre 2000) but concentrate mainly on those developments which have occurred since publication of several PS II crystal structures of progressively increasing resolution, starting in 2003. The work of all computational groups actively involved in PS II studies is examined, in the light of direct PS II structural information from X-ray diffraction crystallography and EXAFS on the metals in the catalytic site. We further address the consistency of the various computational models with results from a range of spectroscopic studies on the PS II site, in all of those functionally intermediate states (S-states) amenable to study. Experimental data considered include Mn K-edge XANES studies, hyperfine coupling of Mn nuclei and various ligand nuclei (including those from substrate water) seen by several EPR techniques applied to the net spin half intermediates, S(0) and S(2), at low temperatures. Finally we consider proposed catalytic mechanisms for the O-O bond formation step, from two groups, in the light of the available experimental evidence bearing on this process, which we also summarise.

Hydration preferences for Mn4Ca cluster models of photosystem II: location of potential substrate-water binding sites.

Density functional theory calculations are reported on a set of three model structures of the Mn(4)Ca cluster in the water-oxidizing complex of Photosystem II (PSII), which share the structural formula [CaMn(4)C(9)H(10)N(2)O(16)](q+)·(H(2)O)(n) (q=-1, 0, 1, 2, 3; n=0-7). In these calculations we have explored the preferred hydration sites of the Mn(4)Ca cluster across five overall oxidation states (S(0) to S(4)) and all feasible magnetic-coupling arrangements to identify the most likely substrate-water binding sites. We have also explored charge-compensated structures in which the overall charge on the cluster is maintained at q=0 or +1, which is consistent with the experimental data on sequential proton loss in the real system. The three model structures have skeletal arrangements that are strongly reminiscent, in their relative metal-atom positions, of the 2.9-, 3.7-, and 3.5 Å-resolution crystal structures, respectively, whereas the charge states encompassed in our study correspond to an assignment of (Mn(III))(3)Mn(II) for S(0) and up to (Mn(IV))(3)Mn(III) for S(4). The three models differ principally in terms of the spatial relationship between one Mn (Mn(4)) and a generally robust Mn(3)Ca tetrahedron that contains Mn(1), Mn(2), and Mn(3). Oxidation-state distributions across the four manganese atoms, in most of the explored charge states, are dependent on details of the cluster geometry, on the extent of assumed hydration of the clusters, and in some instances on the imposed magnetic-coupling between adjacent Mn atoms. The strongest water-binding sites are generally those on Mn(4) and Ca. However, one structure type displays a high-affinity binding site between Ca and Mn(3), the S-state-dependent binding-energy pattern of which is most consistent with the substrate water-exchange kinetics observed in functional PSII. This structure type also permits another water molecule to access the cluster in a manner consistent with the substrate-water interaction with the Mn cluster, seen in electron spin-echo envelope modulation (ESEEM) studies of the functional enzyme in the S(0) and S(2) states. It also rationalizes the significant differences in hydrogen-bonding interactions of the substrate water observed in the FTIR measurements of the S(1) and S(2) states. We suggest that these two water-binding sites, which are molecularly close, model the actual substrate-binding sites in the enzyme.