Periodic Reporting for period 1 - MetEmbed (Addressing metalloenzymes for clean energy production with advanced embedding schemes andquantum mechanical methods)
Período documentado: 2018-08-01 hasta 2020-07-31
The MetEmbed project was carried out by Ph.D. Erik D. Hedegård (EDH) in the group of Professor Ulf Ryde (UR) at Division of Theoretical Chemistry, Lund University (LU).
MetEmbed concerned sustainable energy production, which is one of the greatest challenges for the modern society. Today, most energy production relies on non-renewable (fossil) sources. Biofuel is an advancing, sustainable alternative, but production costs are today too high. A current goal is to employ cellulose, which is both cheap and abundant. Unfortunately, cellulose is extremely stable, preventing efficient degradation.
A family of recently discovered enzymes, denoted lytic polysaccharide monooxygenases (LPMOs), has shown great potential degrading cellulose. The LPMOs contain a copper reaction center and this center oxidizes the cellulose surface, thereby boosting the degradation. However, the working mechanism of this oxidation is unknown. Progress towards industrial utilization of LPMOs will be greatly facilitated by elucidating this mechanism. Currently, the role of hydrogen peroxide is particular interesting as it seems to both drive the reaction, while it also can destroy the enzymes.
Crystallographic and spectroscopic methods alone have so far not been able to elucidate the mechanism of LPMOs. Thus, interdisciplinary studies using quantum mechanical (QM) methods are crucial to complement experimental investigations. Yet, current theoretical methods often fail for transition metals, which is caused by either 1) the underlying protein structure 2) the used electron correlation method or 3) the description of the protein electrostatics. By combining advanced multireference methods with an efficient method for large systems the “MetEmbed” will address these three points to understand the LPMO mechanism.
Using a method that combines quantum and classical mechanics (QM/MM), we have successfully elucidated a mechanism that shows 1) which key LPMO intermediates that can oxidize the substrate. 2) why hydrogen peroxide leads to the experimentally observed faster reactions. 3) How LPMOs themselves can generate hydrogen peroxide. 4) How the architecture of reaction center prevents oxidative damage when LPMOs react with hydrogen peroxide without substrate. 5) Finally, we have successfully shown evidence for that at least one of the key LPMO intermediates are multiconfigurational, and therefore specialized theoretical methods can be expected to be required. These results were disseminated over 10 papers (papers 1–10) in esteemed, peer-reviewed scientific journals. Some of these papers will be explicitly referred to below.
MetEmbed concerned sustainable energy production, which is one of the greatest challenges for the modern society. Today, most energy production relies on non-renewable (fossil) sources. Biofuel is an advancing, sustainable alternative, but production costs are today too high. A current goal is to employ cellulose, which is both cheap and abundant. Unfortunately, cellulose is extremely stable, preventing efficient degradation.
A family of recently discovered enzymes, denoted lytic polysaccharide monooxygenases (LPMOs), has shown great potential degrading cellulose. The LPMOs contain a copper reaction center and this center oxidizes the cellulose surface, thereby boosting the degradation. However, the working mechanism of this oxidation is unknown. Progress towards industrial utilization of LPMOs will be greatly facilitated by elucidating this mechanism. Currently, the role of hydrogen peroxide is particular interesting as it seems to both drive the reaction, while it also can destroy the enzymes.
Crystallographic and spectroscopic methods alone have so far not been able to elucidate the mechanism of LPMOs. Thus, interdisciplinary studies using quantum mechanical (QM) methods are crucial to complement experimental investigations. Yet, current theoretical methods often fail for transition metals, which is caused by either 1) the underlying protein structure 2) the used electron correlation method or 3) the description of the protein electrostatics. By combining advanced multireference methods with an efficient method for large systems the “MetEmbed” will address these three points to understand the LPMO mechanism.
Using a method that combines quantum and classical mechanics (QM/MM), we have successfully elucidated a mechanism that shows 1) which key LPMO intermediates that can oxidize the substrate. 2) why hydrogen peroxide leads to the experimentally observed faster reactions. 3) How LPMOs themselves can generate hydrogen peroxide. 4) How the architecture of reaction center prevents oxidative damage when LPMOs react with hydrogen peroxide without substrate. 5) Finally, we have successfully shown evidence for that at least one of the key LPMO intermediates are multiconfigurational, and therefore specialized theoretical methods can be expected to be required. These results were disseminated over 10 papers (papers 1–10) in esteemed, peer-reviewed scientific journals. Some of these papers will be explicitly referred to below.
The QM/MM methods allow us to optimize the structures involved in the oxidative mechanism accurately for the entire LPMO enzyme. We could thus map out the full mechanism together with activation and reaction energies. These calculations highlighted two key intermediates of particular interest. In addition, we have shown that the amino acids close to the copper reaction center can influence the reaction and activation energies.
In a parallel set of calculations we have shown that generation of the two key intermediates from hydrogen peroxide is faster than from oxygen. These results were used as basis for the remaining investigations and disseminated in form of a paper (paper 1) in the high-impact journal Chemical Science.
We next investigated the mechanism of hydrogen peroxide generation. Our proposed mechanism consists of adding two protons and two electrons to dissociate hydrogen peroxide from Cu(I). The first protonation is spontaneous, whereas the second protonation has a very low activation barrier, provided that the first electron has been added to the system. Then, hydrogen peroxide dissociates from Cu(I) with a barrier of only 4 kJ/mol, in contrast to hydrogen superoxide, as previously suggested (the latter has an dissociation energy barrier of 60 kJ/mol in our calculations). A prerequisite for this investigation was that we knew the protonation state of the copper reaction center, which unfortunately was not clarified when MetEmbed was started. Through quantum refinement of the crystal structure against both X-ray and neutron data (in collaboration with Dr. Esko Oksanen at European Spallation Source), we could establish the correct protonation state. These results were disseminated in form of a paper (paper 6) in Chemical Science.
Finally, we have provided a mechanism for how a conserved active site tyrosine can help protect the LPMOs against oxidative damage when reacting with hydrogen peroxide without substrate.
In a different direction, we have also used the accurate QM/MM structures from paper 1 as basis for employing a multiconfigurational method (CASPT2) to investigate the identified key intermediates. We concluded that at least one of the key intermediates indeed requires and accurate multiconfigurational electron-correlation method. The results from this calculations were disseminated in form of a paper (paper 7) in Dalton Transactions.
In a parallel set of calculations we have shown that generation of the two key intermediates from hydrogen peroxide is faster than from oxygen. These results were used as basis for the remaining investigations and disseminated in form of a paper (paper 1) in the high-impact journal Chemical Science.
We next investigated the mechanism of hydrogen peroxide generation. Our proposed mechanism consists of adding two protons and two electrons to dissociate hydrogen peroxide from Cu(I). The first protonation is spontaneous, whereas the second protonation has a very low activation barrier, provided that the first electron has been added to the system. Then, hydrogen peroxide dissociates from Cu(I) with a barrier of only 4 kJ/mol, in contrast to hydrogen superoxide, as previously suggested (the latter has an dissociation energy barrier of 60 kJ/mol in our calculations). A prerequisite for this investigation was that we knew the protonation state of the copper reaction center, which unfortunately was not clarified when MetEmbed was started. Through quantum refinement of the crystal structure against both X-ray and neutron data (in collaboration with Dr. Esko Oksanen at European Spallation Source), we could establish the correct protonation state. These results were disseminated in form of a paper (paper 6) in Chemical Science.
Finally, we have provided a mechanism for how a conserved active site tyrosine can help protect the LPMOs against oxidative damage when reacting with hydrogen peroxide without substrate.
In a different direction, we have also used the accurate QM/MM structures from paper 1 as basis for employing a multiconfigurational method (CASPT2) to investigate the identified key intermediates. We concluded that at least one of the key intermediates indeed requires and accurate multiconfigurational electron-correlation method. The results from this calculations were disseminated in form of a paper (paper 7) in Dalton Transactions.
We were the first to map out the LPMO mechanism using a method (QM/MM) that includes the entire LPMO enzyme. One of the two identified key intermediates was not previously considered. The favorable use of hydrogen peroxide was another important finding. This is both industrially important and has been an ongoing debate since it was experimentally reported that hydrogen peroxide can speed up the LPMO reaction (although it also leads to oxidative damage).
The MetEmbed project has also gained further insight in the role of hydrogen peroxide, e.g. how LPMOs form hydrogen peroxide in absence of substrate. We also identified the role of a tyrosine amino acid in close proximity of the active site. This role has been somewhat mysterious, but according to our findings the tyrosine protects the reaction center from oxidation, when reacting with hydrogen peroxide in absence of substrate.
The MetEmbed project has also gained further insight in the role of hydrogen peroxide, e.g. how LPMOs form hydrogen peroxide in absence of substrate. We also identified the role of a tyrosine amino acid in close proximity of the active site. This role has been somewhat mysterious, but according to our findings the tyrosine protects the reaction center from oxidation, when reacting with hydrogen peroxide in absence of substrate.