Alternative mechanisms for O 2 release and O–O bond formation in the oxygen evolving complex of photosystem IIby Xichen Li, Per E. M. Siegbahn

Phys. Chem. Chem. Phys.


Physical and Theoretical Chemistry / Physics and Astronomy (all)



This journal is© the Owner Societies 2015 Phys. Chem. Chem. Phys.

Cite this:DOI: 10.1039/c5cp00138b

Alternative mechanisms for O2 release and O–O bond formation in the oxygen evolving complex of photosystem II†

Xichen Liab and Per E. M. Siegbahn*b

In a previous detailed study of all the steps of water oxidation in photosystem II, it was surprisingly found that O2 release is as critical for the rate as O–O bond formation. A new mechanism for O2 release has now been found, which can be described as an opening followed by a closing of the interior of the oxygen evolving complex. A transition state for peroxide rotation forming a superoxide radical, missed in the previous study, and a structural change around the outside manganese are two key steps in the new mechanism. However, O2 release may still remain rate-limiting. Additionally, for the step forming the

O–O bond, an alternative, experimentally suggested, mechanism was investigated. The new model calculations can rule out the precise use of that mechanism. However, a variant with a rotation of the ligands around the outer manganese by about 301 will give a low barrier, competitive with the old DFT mechanism. Both these mechanisms use an oxyl–oxo mechanism for O–O bond formation involving the same two manganese atoms and the central oxo group (O5).

I. Introduction

Water oxidation in nature is catalyzed by the oxygen evolving complex (OEC) in photosystem II (PSII), containing four manganese and one calcium connected by oxo bridges. For the understanding of this process computational modeling studies have played a major role. A theoretically determined structure of the OEC was suggested in 2008,1 where information from earlier low-resolution X-ray structures2,3 was used. This structure was essentially confirmed by a high-resolution structure in 2011,4 where the only difference was the positioning of Asp170. Very recently the same group obtained a structure of the S1 state using a free-electron laser,5 showing even better agreement with the corresponding DFT structure. An oxo–oxyl mechanism for O–O bond formation was suggested based on DFT calculations in 2006,6 which was slightly refined in 2009.7 Recent water exchange experiments using a W-band 17O-ELDOR detected NMR spectroscopy8 for the substrate oxygen positions in the S2 state have confirmed the most important parts of that mechanism. The quantum chemical structures have been confirmed by experiments for both the S2 and the S3 states. A comparison of the structures of the S2 state suggested by DFT modeling 9,10 and by spectroscopic experiments11 is shown in Fig. 1, and for the corresponding S3 structures 10,12,13 in Fig. 2. The only visible difference is a slight rotation of the His residue, but this has negligible energetic consequences and was therefore outside the goal of the DFT modeling.

Over the years, a variety of different mechanisms for O–O bond formation have been suggested. Most noteworthy of these is the attack of a water molecule, either free or bound to calcium, on an Mn(V)Qoxo or Mn(IV)–oxyl state, see for example.14,15 This mechanism was compared to the one above and was ruled out by model calculations in 2006,6 since the computed barrier was more than 10 kcal mol1 higher, which is much more than the uncertainty of the methods. More recently, spectroscopic experiments have given very strong arguments against the nucleophilic water attack mechanism.16 New information favoring this mechanism appears to be absent in the literature since a couple of years. The only alternative scenario still advocated in the literature is the use of the low-oxidation paradigm.17 This type of mechanism has been found very unfavorable in all investigations of the present type. Recently, a thorough investigation of the oxidation states of the OEC using a large number of spectroscopic techniques was presented.18 The analysis strongly argues against the low-oxidation paradigm.

The most complete mechanism for all the steps in water oxidation in PSII, including oxidations, proton release pathways,

O–O bond formation and O2 release was published a year ago. 10

A surprising feature of this mechanism was that the final step, involving O2 release and water binding, had the highest barrier.

This complicated step is composed of several parts. After O–O a College of Chemistry, Beijing Normal University, 100875, Beijing, China b Department of Organic Chemistry, Arrhenius Laboratory, Stockholm University,

SE-106 91, Stockholm, Sweden. E-mail:; Tel: +46-8-16-26-16 † Electronic supplementary information (ESI) available. See DOI: 10.1039/c5cp00138b

Received 9th January 2015,

Accepted 8th April 2015

DOI: 10.1039/c5cp00138b



Pu bl ish ed o n 10

A pr il 20 15 . D ow nl oa de d by

U ni ve rs ity o f C al ifo rn ia

Sa n

Fr an ci sc o on 1 7/ 04 /2 01 5 10 :0 5: 24 .

View Article Online

View Journal

Phys. Chem. Chem. Phys. This journal is© the Owner Societies 2015 bond formation leading to a bridging peroxide, O2 is first released in anendergonic step.This is followedbyaproton transfer fromawater bound to calcium, over an intermediate water, to a hydroxide bound to the outer manganese. In the next two substeps, the hydroxide on calciummoves into the cavity between themanganese centers, anda new water molecule becomes bound to calcium.10 The highest barrier was found for the proton transfer from the calcium bound water, with a value of 14.0 kcal mol1 counted from the resting S3 state. This barrier actuallymade this step rate-limiting for the entire water oxidation mechanism, since the barrier for O–O bond formation was only 11.3 kcal mol1. A large effort was spent investigating whether a concerted O2 release with water binding would be advantageous, but this was not successful.

The finding that the proton transfer step after O2 release should be rate-limiting remained quite surprising, and during the past year new investigations were made trying to find a better mechanism for this step. In the present article, a new mechanism for all steps of O2 release is presented.