Photoassembly of the CaMn ₅ O ₅ Catalytic Core and Its Inorganic Mutants in Photosystem II

Abstract

“What I cannot create (build), I do not understand”—Richard Feynman The ability to take apart and reassemble atom by atom the elements of the water‐oxidizing complex (WOC) of photosynthetic enzymes and substitute each cofactor, while monitoring the consequences on photolytic charge separation and catalytic activity, has given deep insights into how photosynthetic water oxidation works, equal to any other method or technique. Here we summarize the literature covering the reconstitution of the inorganic core (Mn 4 CaO 5 ) of photosystem II water‐oxidizing complexes (PSII‐WOCs), called photoassembly, and the consequences of substitution of the inorganic cofactors with nonnative cofactors. We give the evidence that explains the universal utilization of manganese as the sole redox‐active element for O 2 evolution by all native PSII‐WOCs. The selection of Mn 2+ over other more abundant transition metals during uptake and photooxidation by apo‐WOC‐PSII is shown to be achieved by a combination of thermodynamic advantage (bicarbonate stabilization of Mn 3+ photoproduct) and kinetic advantage favoring slow charge recombination from its intermediates. Manganese is shown to be favored over other transition metals because it can operate over a wider range of turnover rates produced by solar fluxes from morning to midday to dusk. The speciation of aqueous manganese in cells in equilibrium with air and bicarbonate is described from thermodynamic data and used to explain how (bi)carbonate stabilizes Mn 2+ → Mn 3+ photooxidation and provides the core oxo bridges that build the Mn 4 CaO 5 cluster. The oxidation states of manganese obtained from photoassembly experiments are described and compared to spectroscopic results. The functional role of Ca 2+ in photoassembly and catalysis is revealed by substitution with Sr 2+ and Cd 2+ . These properties are advanced in order to understand how it is that evolution of only a single enzymatic blueprint for photosynthetic water oxidation has emerged, despite the access to every ecological niche on Earth and over 3 billion years of evolutionary history.

Why Did Nature Choose Manganese over Cobalt to Make Oxygen Photosynthetically on the Earth?

“…47 The exponential phase becomes biexponential if an optimal Ca/Mn ratio is not preserved (also reducing the yield) and is firstorder in the Ca 2+ concentration below a threshold based on earlier studies of spinach thylakoids and core complexes. 50,52,53 The same two-step rate law and comparable exponential lifetimes (less than 50% change) were observed in earlier work on isolated spinach thylakoid membranes 50,52,53 and PSII complexes 33,44,52,54,55 and cyanobacterial thylakoids, 54 in which the water-soluble PSII extrinsic subunits are removed. PSII microcrystals do not lose their inter-subunit contacts during metal ion removal.…”

“…Additionally, these studies allow a degree of protein association and dissociation, which is not possible in a microcrystal system; we have demonstrated that photoassembly is possible in well-defined, physically constrained PSII crystals with all protein subunits available, as detailed in 61 and using considerably different conditions than those shown by Cheah et al 40 Our results with cobalt provide the first functional replacement of manganese in any PSII preparation, in vivo or in vitro, although many elements have been attempted. 33,37,50 We selected cobalt for two reasons. First, Pourbaix and Latimer−Frost diagrams prove that cobalt can exist in only three oxidation states, Co 2+,3+,4+ in aqueous media, compared to six for manganese, Mn 2+,3+,4+,5+,6+,7+ , thereby limiting the number of redox transitions to 8 (=4× Co 2+ → Co 3+ → Co 4+ ), assuming that it replaces Mn 1:1 in PSII.…”

“…The consequences of this delay being absent are clear in Figure 5 as a far greater fraction of the photoinactivation experienced by Co-PSIIX is permanent and irreversible degradation of the WOC or the binding site within the PSII protein. It is known that binding and photooxidation of free Co 2+ occur in some centers, resulting in nonfunctional cobalt−oxo structures, 33,64 but as there is no free cobalt aside from any that may be lost from Co-WOC degradation or the breakdown of nonfunctional cobalt oxides formed during photoassembly, this is unlikely to comprise a major portion of the photoinactivation. Furthermore, as seen from comparing the oxygen yield in Figure 3 and the ICP-OES results over time in Table 1, the Co-WOC does not appear to be easily converted to these structures even with excess free cobalt available.…”

“…32−34 Both outcomes arise because the redox equilibrium with the oxidized tyrosine radical, WOC(Sr)Y Z + ↔ WOC-(Sr) + Y Z , shifts to the right, placing the hole on manganese and slowing charge recombination. 33 However, there is a trade-off as this gain is realized only at low light intensity (turnover rate), while at high light flux, the O 2 yield decreases by 40% due to fivefold slower O 2 release in the final step (S 4 → S 0 + O 2 ). 32 Although WOC(Sr) operates more efficiently than native WOC(Ca) at low light and higher temperatures (faster barrier crossing), the opposite is true at high light.…”

“…We previously investigated several electron acceptors that allow high O 2 flash yields from PSII microcrystals 45 and from apo-WOC-PSII particles. 33,50 As seen in Figure S2, the commonly used hydrophobic electron acceptor PPBQ is not ideal for supporting O 2 evolution activity in crystals. Titrations (Figure S2A) reveal that at concentrations that restore the largest steady-state yield of O 2 (≥400 μM), the oscillations in flash O 2 yield are severely damped (elevated miss parameter, α = 0.12 for the Kok cycle vs 0.04−0.058 for intact PSII crystals), stemming from reduction of the WOC (Mn n+ , Y z + , or P + ) by the accumulation of reduced forms of PPBQ (semiquinone or quinol).…”