Rapid Evolution of the Photosystem II Electronic Structure during Water Splitting

Photosynthetic water oxidation is a fundamental process that sustains the biosphere. A ${\mathrm{Mn}}_4\mathrm{Ca}$ cluster embedded in the photosystem II protein environment is responsible for the production of atmospheric oxygen. Here, time-resolved x-ray emission spectroscopy (XES) is used to observe the process of oxygen formation in real time. These experiments reveal that the oxygen evolution step, initiated by three sequential laser flashes, is accompanied by rapid (within $50\mu \mathrm{s}$) changes to the Mn $K\beta$ XES spectrum. However, no oxidation of the ${\mathrm{Mn}}_4\mathrm{Ca}$ core above the all-${\mathrm{Mn}}^{\mathrm{IV}}$ state is detected to precede $\mathrm{O}─\mathrm{O}$ bond formation, and the observed changes are therefore assigned to $\mathrm{O}─\mathrm{O}$ bond-formation dynamics. We propose that $\mathrm{O}─\mathrm{O}$ bond formation occurs prior to the transfer of the final (fourth) electron from the ${\mathrm{Mn}}_4\mathrm{Ca}$ cluster to the oxidized tyrosine ${\mathrm{Tyr}}_Z$ residue. This model resolves the kinetic limitations associated with $\mathrm{O}─\mathrm{O}$ bond formation and suggests an evolutionary adaptation to avoid releasing harmful peroxide species.

Popular Summary

Society is faced with multiple challenges to energy security including increased energy demand, the scarcity of carbon-based fuels, and climate disruption as a result of using those fuels. Finding new ways of generating clean energy is therefore an imperative. Since green plants already do this with unparalleled efficiency, there is significant interest in echoing photosynthesis in artificial devices. Despite intense efforts from the biophysics community to understand this process on a mechanistic level, the final step remains a matter of speculation. In this work, we use very bright, short laser and x-ray pulses to drive photosynthesis and track electronic transitions through the process.

During photosynthesis, green plants use sunlight to extract electrons and protons needed to generate fuel, creating molecular oxygen as a byproduct. The critical step in which two oxygen atoms bond to form ${\mathrm{O}}_{\mathrm{2}}$ is the part that is not understood. Our experiments strongly suggest that this bond forms prior to the transfer of the last electron. We further show that this new catalytic mechanism promotes rearrangement of atoms and the formation of new molecules with the smallest possible waste of energy.

By providing a blueprint for mimicking this biological process, these results will accelerate the development of artificial photosynthesis on a scale large enough to address society’s future energy needs.

Figure 1

(a) Current model of the Kok cycle. Depiction of sequential incident visible-light photon absorptions triggering electron or proton release [3]. The dashed region is based on previous analysis of the $S_3$-to-$S_0$ transition in which the $S_4$ state is proposed [3, 4]. (b) Electronic transitions, reflected in the $K\beta$ main lines, are influenced by the spin state of the Mn ion. (c) A spectral comparison of Mn oxides depicts the effect of oxidation state on the $K\beta$ emission lines. (d) Nanosecond laser pulses (1, 2, or 3) are used to advance the Kok cycle in the protein (Table S1 [5]). The pump-probe delay time $\mathrm{\Delta }t$, measured from the final laser flash to the center of the x-ray pulse, is set dependent on the desired $S$ state. X-ray fluorescence from the sample is reflected by ten flat analyzer crystals onto a 2D-position-sensitive detector. $K\beta$ emission spectra are extracted to form snapshots of the electronic structure in time. Smoothed emission spectra are presented for $2F$ (majority $S_3$) and two time points during the $S_3$-to-$S_0$ transition. (e) The proposed reaction scheme shows the early evolution of the OEC during the $S_3$-to-$S_0$ transition, providing an interpretation of spectroscopic results.

Figure 2

Analysis of the statistical significance for $S_1\to S_2\to S_3$ transitions serves as a proof of concept. (a) Spectral shifts occurring as a result of $0F\to 1F$ and $1F\to 2F$ transitions are characterized using 2D heat maps, illustrating the $p$ value (see Table S2 for $n$ [5]) for changes to first moments calculated over the ranges defined by the $x$ and $y$ axes, i.e., start and end energy, respectively. Contours appear in plots comparing $0F\to 1F$ data indicate a statistically significant difference. By contrast, limited low $p$-value regions are observed for the $1F\to 2F$ transition, suggesting a smaller change. The directionality and magnitude of spectral shifts are shown in the final column. These 2D heat maps, which graphically illustrate the change in first moments ($\mathrm{\Delta }\mathrm{FM}={\mathrm{FM}}_{\text{postflash}}-{\mathrm{FM}}_{\text{preflash}}$) calculated over the ranges defined by the $x$ and $y$ axes. $0F\to 1F$ and $1F\to 2F$ transitions are dominated by negative, or oxidative, shifts. An example of statistical noise is presented by randomly dividing a $0F$ data set and performing the same analysis, e.g., $0F\to 0F$. Relevant data from data sets 1–4 are merged to generate the final columns. Note that $0F$ and $1F$ data are not collected for data set 4, while data set 5 is collected independently following a beam-line upgrade and is therefore analyzed separately; see Fig. S4 [5]. (b) Wavelet-transform smoothed and background-subtracted emission spectra for merged $0F$, $1F$, and $2F$ data. The region (6.485–6.495 keV), over which the first moment is calculated, is highlighted, and a magnified inset as well as difference spectra are presented. Difference spectra are smoothed with a rolling average calculated over 14 points (approximately $\pm 0.7\mathrm{eV}$). (c) (Top) Average first moments from unprocessed (color) and processed (gray) spectra. Errors are presented as SEM with $n$ given in Table 3. Those moments with a statistically significant difference ($p<0.05$) from $0F$ data are indicated with an asterisk. (Bottom) Dot plot of first moments from raw data. Each dot represents a calculated first moment from a thread collected during the beam time corresponding to its color in the legend. Dashed lines represent the average first moment, while solid bars are the 95% confidence interval.

Figure 3

Rapid onset of spectral changes observed during the $S_3$-to-$S_0$ transition. (a) Spectral shifts occurring as a result of the $2F\to 3F$ transition are characterized using 2D heat maps colored by $p$ value (see Table S2 [5] for $n$), for changes to first moments calculated over the ranges defined by the $x$ and $y$ axes, i.e., start and end energy, respectively. Strong contours are consistently observed, indicative of a statistically significant spectral change. The directionality and magnitude of spectral shifts occurring with these transitions are also shown as 2D heat maps, which graphically illustrate the change in first moments ($\mathrm{\Delta }\mathrm{FM}={\mathrm{FM}}_{\text{postflash}}-{\mathrm{FM}}_{\text{preflash}}$) calculated over the ranges defined by the $x$ and $y$ axes. All $2F\to 3F$ comparisons yield primarily positive, or reductive, shifts. Determined $p$ values of approximately 0.02 (Table 3) for the $200\mu \mathrm{s}$ time point are further supported by the above heat maps; however, little information can be gleaned from these plots of data set 2, likely due to much lower x-ray shot count; see Table 2. Although the $500\mu \mathrm{s}$ time point is equally compelling, the traditional energy range (6.485–6.495 keV) selected for reporting first moments does not deliver statistical significance (Table 3). (b) Wavelet transform smoothed and background-subtracted emission spectra for merged $2F$, $3F_{50\mu \mathrm{s}}$, and $3F_{200\mu \mathrm{s}}$ data, calculated as in Fig. 2. (c) (Top) Average first moments from unprocessed (color) and processed (gray) spectra. Errors are presented as SEM with $n$ given in Table 3. Those moments with a statistically significant difference ($p<0.05$) from $2F$ data are indicated with an asterisk. (Bottom) Dot plot of first moments from raw data, presented as in Fig. 2. See Fig. S3 [5] for 2D plots of the 40 ms (majority $S_0$ state) time point.

Figure 4

Analysis of Mn $K\beta$ first moments. Placement of fully processed $2F$ and $3F$ data along a linear fit to a series of Mn oxide first moments empirically predicts the average spin state of Mn centers in the OEC. For reference, relevant Mn coordination complexes are placed along the line by using the nominal spins reported by Davis et al. and Jensen et al. [31, 46]. Note that $S_{4a}$ and $S_{4b}$ correspond to $3F$ states with $\mathrm{\Delta }t$ between the final laser flash and an x-ray pulse of $50$ and $200\mu \mathrm{s}$, respectively. Compounds 1–3 are formally mixed valence ${\mathrm{Mn}}^{\mathrm{III}}/{\mathrm{Mn}}^{\mathrm{IV}}$ complexes. 1 is a $\mathrm{di}\text{−}\mu \text{−}\mathrm{oxo}$ dimer, [${\mathrm{Mn}}_2{\mathrm{O}}_2{\mathrm{L}}_4^{\prime }$] ${({\mathrm{ClO}}_4)}_3$, while 2 and 3 are two examples from the Mn cubane family, ${\mathrm{Mn}}_4{\mathrm{O}}_4{\mathrm{L}}_6$. Compounds 4 and 5, by contrast, are mononuclear Mn complexes $[{\mathrm{Mn}}^{\mathrm{IV}}(\mathrm{OH}{)}_2({\mathrm{Me}}_2\mathrm{EBC}){]}^{2+}$ and $[{\mathrm{Mn}}^{\mathrm{IV}}(\mathrm{O})(\mathrm{OH}{)}^{-}({\mathrm{Me}}_2\mathrm{EBC}){]}^{+}$, respectively.