Investigation of Water Oxidizing Catalysis for Renewable Energy

Lead Research Organisation: Imperial College London
Department Name: Life Sciences


The burning of fossil fuels releases CO2 which is almost certainly responsible for anthropogenic climate change. Therefore, we must find alternative 'carbon-neutral' sources of energy as a matter of urgency. By far the largest potential source of renewable energy is sunlight. Harnessing this energy is one of the great challenges that our civilization faces, but using it is problematic. Existing silicon solar cells are expensive and inefficient, and do not produce fuel. Plants have hit on the perfect solution; they use the energy of sunlight to oxidise water (H2O), liberating O2, protons and electrons. The electrons and protons are used to fix carbon dioxide as organic sugars, which may then be used for biosynthesis or as fuel for respiration. The total process is known as photosynthesis. Plants can be grown to generate so-called 'biofuels' such as biodiesel, but this is inefficient, and competes with food production. What is needed is an artificial photosynthetic system, that, like plants, converts sunlight, water and CO2 into fuel, but is cheap, efficient and can be deployed over large areas. The proposed project is to create a vital component of a future solar energy conversion system. There are many components to the photosynthetic apparatus, but the main one of interest to this project is an enzyme called photosystem II (PSII). PSII is responsible for the light-driven water splitting reaction of photosynthesis. At its core, PSII has a cluster of one calcium and four manganese ions, which catalyse the water splitting reaction. This cluster is known as the oxygen evolving centre (OEC). The precise structure of the OEC and the mechanism of its action are still unknown, but both of these must be understood if a synthetic light-driven water oxidase is to be constructed. Building such a system is a vital prerequisite for the efficient large scale use of solar energy. The OEC in PSII is difficult to study, as PSII is a large complex containing many protein molecules and cofactors as well as the OEC. Therefore I propose to use small proteins as scaffolds for manganese ions, and so construct an OEC analogue that is uncoupled from the PSII enzyme and can be studied much more easily, and is a realistic prototype for future devices. Apart from PSII, there are many known enzymes which contain two manganese ions at their active sites, but PSII is unique in having four manganese ions at one site. I would like to take one of these simpler manganese enzymes and engineer it to bind more manganese ions, to mimic PSII. This can be accomplished by recombinant DNA technology. A DNA molecule with a sequence encoding the designed enzyme is constructed and then introduced into a harmless bacterium. The bacterium is then induced to produce the modified enzyme, which is then extracted and purified for further study. This technique has the advantage that DNA molecules are easy to manipulate, and specific sequences can be produced quickly and cheaply, allowing many designs of enzyme to be tried in a short time. Having produced a modified protein molecule that binds multiple manganese ions, the three dimensional structure can be determined. The protein will be probed for enzyme activity similar to that of PSII. I will try to catalyse the oxidation of water or other substrates using powerful oxidants as a substitute for light. In plants, chlorophyll is used as the main photosensitive pigment, but chlorophyll is usually unstable in artificial systems. Instead I will couple stable synthetic pigments to the protein and try to generate oxidative reactions using light. The results of these experiments can be related to the three dimensional structure of the enzyme and then used to to inform modifications in the design of the engineered proteins, which will then be subjected to further rounds of experimentation and design. This 'evolution by artificial selection', can be iterated until the desired goal of a soluble water oxidase is realised.

Technical Summary

Solar energy is the only feasible source of renewable energy that can replace fossil fuels. To develop sunlight as a viable source of energy it will be necessary to imitate the light reactions of photosynthesis. Photosystem II (PSII) is the only known system capable of using visible light to oxidise water to oxygen, releasing reducing equivalents, which can then be used to fix CO2 and generate fuel. The reaction is catalysed at a cluster of manganese ions and other associated cofactors. Despite crystal structures of PSII at up to 3.0A and characterization of the complex by other biophysical techniques, the structure of the cluster is still under debate. I propose to create a multiple manganese cluster, in imitation of PSII, using existing and de novo proteins as scaffolds into which the cluster is assembled. Such a cluster will be much easier to study than the OEC of PSII, which is part of a monolithic membrane protein complex. The synthetic clusters will be characterised by X-ray crystallography and EXAFS to confirm the location and incorporation of manganese ions. The protein design will be driven by knowledge of the rotameric properties of amino acids and proteins and statistical information from the protein structure and small molecule databases on the binding modes of manganese ions. A strong candidate protein for the initial design is the C. trachomatis ribonucleotide reductase. This protein has a Mn(IV)-Fe(III) site, which has the only known biological high valency manganese outside PSII. Therefore it is an ideal substrate into which more Mn sites can be introduced, having an existing Mn(IV) and a four-helix bundle architecture into which amino acid changes can be made. The engineered proteins will be expressed in E. coli. After a manganese cluster has been produced, I will attempt to generate oxidative chemistry using both chemical oxidants such as potassium peroxymonosulfate, and reactions driven by pigments, such as ruthenium polypyridines.


10 25 50

publication icon
MacDonald JT (2016) Synthetic beta-solenoid proteins with the fragment-free computational design of a beta-hairpin extension. in Proceedings of the National Academy of Sciences of the United States of America

publication icon
MacKellar D (2016) Streptomyces thermoautotrophicus does not fix nitrogen. in Scientific reports

Description Variant Proteins Constructed.

The original plan was to use chlamydial ribonucleotide reductase as a scaffold protein, as it is the only known protein other than PSII to have a high-valent manganese. Unfortunately, this protein proved very hard to express solubly in E. coli and the designed mutants were even

I then tried to generate the equivalent mutations in variants of the "Due Ferri" maquette, a synthetic four-helix bundle protein with a ferritin-like dimetal centre. These were expressed in E. coli, but were seemingly non-functional and could not be crystallised. Maquette proteins are in fact difficult to crystallize as they are very flexible, and in retrospect, despite their simplicity, probably not a good choice for protein design requiring precise atom placement.

Attempts were made to use non-metalloproteins as scaffolds, but inserting a large metal site proved to be bad for the protein solubility. Finally, I moved to E. coli bacterioferritin (BFR) as a
scaffold. This protein is a four-helix bundle, with a dimetal centre like ribonucleotide reductase. The protein is a strong dimer, with a heme molecule between the two monomers. Twelve monomers assemble to form a capsid. This protein proved a good template for design. BFR was
tolerant of many mutations to the active site, and still folded and was soluble and bound heme. All the various mutants crystallised, and the positions of the side chains in the crystal structures was usually close to the designed positions.

The manganese cluster of PSII has Mn(III) and Mn(IV) oxidation states. These are generated in vivo by the oxidative action of the photosystem itself. In my system it was necessary to develop a
protocol to chemically oxidise the manganese. This was done, using manganese(II) chloride as the starting material and periodic acid as the oxidant. The protein was incubated with Mn(II) ions, then excess removed by dialysis, then oxidised with periodic acid, which was then removed by dialysis. Using this procedure, mutant ferritins were shown to be capable of binding high-valent manganese. The manganese could be detected by changes to the absorption spectrum of the protein, and by EPR and cyclic voltammetry. Unfortunately, the oxidation procedure seemed to damage the protein so that it would no longer crystallize, so it has not been possible to investigate the structure in finer detail.

I was able to prepare ferritin with the heme replaced with a chlorin. The heme is removed with an organic solvent and chlorin added. This pigment has sufficient potential for water oxidation, so
should a catalytic cluster be developed, it may be used to investigate light-activated electron transfer.
Exploitation Route The Protein scaffold and associated design tools may be used as
templates for protein design projects by other academic labs. These
could have any number of applications, including biotechnology and
Sectors Energy,Environment

Description An invited talk on improving photosynthesis and regulation to the department of Biology in Munich, Germany 
Form Of Engagement Activity A talk or presentation
Part Of Official Scheme? No
Geographic Reach International
Primary Audience Professional Practitioners
Results and Impact An invited talk on improving photosynthesis and regulation to the department of Biology in Munich, Germany
Year(s) Of Engagement Activity 2017
Description Imperial Fringe, walking in the air 
Form Of Engagement Activity Participation in an activity, workshop or similar
Part Of Official Scheme? No
Geographic Reach International
Primary Audience Public/other audiences
Results and Impact A demonstration of photosynthesis to visitors to Imperial College
Year(s) Of Engagement Activity 2017
Description School Visit (Watford) 
Form Of Engagement Activity A talk or presentation
Part Of Official Scheme? No
Geographic Reach Local
Primary Audience Schools
Results and Impact A talk to a 6th form students on my scientific work.
Year(s) Of Engagement Activity 2019
Description York conference repetitive proteins 
Form Of Engagement Activity A talk or presentation
Part Of Official Scheme? No
Geographic Reach National
Primary Audience Professional Practitioners
Results and Impact Synthetic Beta Solenoid Scaffolds", talk at Biochemical Society Conference "Repetitive Non-Globular Proteins: Nature to Nanotechnology" York, Apr 2015.
Year(s) Of Engagement Activity 2015