Functional Water Networks in Fully Hydrated Photosystem II

Abstract

Water channels and networks within photosystem II (PSII) of oxygenic photosynthesis are critical for enzyme structure and function. They control substrate delivery to the oxygen-evolving center and mediate proton transfer at both the oxidative and reductive endpoints. Current views on PSII hydration are derived from protein crystallography, but structural information may be compromised by sample dehydration and technical limitations. Here, we simulate the physiological hydration structure of a cyanobacterial PSII model following a thorough hydration procedure and large-scale unconstrained all-atom molecular dynamics enabled by massively parallel simulations. We show that crystallographic models of PSII are moderately to severely dehydrated and that this problem is particularly acute for models derived from X-ray free electron laser (XFEL) serial femtosecond crystallography. We present a fully hydrated representation of cyanobacterial PSII and map all water channels, both static and dynamic, associated with the electron donor and acceptor sides. Among them, we describe a series of transient channels and the attendant conformational gating role of protein components. On the acceptor side, we characterize a channel system that is absent from existing crystallographic models but is likely functionally important for the reduction of the terminal electron acceptor plastoquinone Q B . The results of the present work build a foundation for properly (re)evaluating crystallographic models and for eliciting new insights into PSII structure and function.

Water Networks in Photosystem II Using Crystalline Molecular Dynamics Simulations and Room-Temperature XFEL Serial Crystallography

“…Moreover, decreasing the detergent and water content in PS II crystals is a specific strategy that has been used to increase the resolution of the diffraction data, 72,73 which tends to increase� not decrease�the number of ordered waters in the crystal structure, as shown in Table 1 by Sirohiwal and Pantazis. 59 We also note that the S-state advancement shown in the XFEL data by Kern et al proves there is no sample dehydration that would inactivate PS II and release Mn(II). 4 Our findings therefore do not support the claim that the crystal structures are dehydrated compared to the MD simulations.…”

“…In addition, the number of waters observed in the previous crystal structures ,,,,− also differs due to the resolution and measurement temperature (cryogenic vs room temperature). Moreover, decreasing the detergent and water content in PS II crystals is a specific strategy that has been used to increase the resolution of the diffraction data, , which tends to increasenot decreasethe number of ordered waters in the crystal structure, as shown in Table 1 by Sirohiwal and Pantazis . We also note that the S-state advancement shown in the XFEL data by Kern et al proves there is no sample dehydration that would inactivate PS II and release Mn(II) .…”

“…For example, there are 249 strong peaks in the MD density within a 20 Å radius of the OEC (coordinates available at ), compared to 161 waters in this region of the crystal structure. A recent unrestrained solution-state MD study by Sirohiwal and Pantazis of a single PS II monomer found 182 MD waters in this same region, using a different analysis method based on counting waters close to the protein every 105 ps. The increased number of peaks in our simulation compared to that of Sirohiwal and Pantazis is consistent with our previous finding that protein restraints can increase the number of strong peaks in the water density .…”

Lys264 of the D2 Protein Performs a Dual Role in Photosystem II Modifying Assembly and Electron Transfer through the Quinone–Iron Acceptor Complex

“…High resolution structures of PS II have also confirmed the involvement of D2-Lys264 in the hydrogen bond network of the environment around the NHI and bicarbonate (Figure ). , In addition, molecular dynamics simulations have implicated D2-Lys264 in a gating role for the access of water to the NHI via specific channels and in the involvement of a putative triple salt bridge between D1-Glu244, D2-Lys264, and D2-Glu242 in stabilizing incoming waters through hydrogen bonding. ,, Given the apparent importance of D2-Lys264 in the architecture of the hydrogen bond network, we investigated its role in vivo by replacing it with either Ala (removing the side chain) or Glu (reversing the charge).…”

Reaction Center Excitation in Photosystem II: From Multiscale Modeling to Functional Principles

“…These techniques are typically applied at cryogenic temperatures and rarely inform on conformational dynamics that are crucial for a complete understanding of structure and function. Sample treatment can also lead to nonphysiological changes . Molecular dynamics (MD) calculations conducted under simulated “physiological” conditions offer a way to obtain advanced understanding of protein dynamics and allow for conformational sampling across a wide range of time and length scales, providing insights into functional states that may not be experimentally accessible.…”