Skip to main content

Accelerating metal-directed assembly, recognition and catalysis with computational methods

Periodic Reporting for period 1 - MetAccembly (Accelerating metal-directed assembly, recognition and catalysis with computational methods)

Reporting period: 2015-05-01 to 2017-04-30

Transition metals (TMs) are found in the core of several phenomena such as catalysis, folding, assembly, and (bio)molecular recognition and are directly involved in a number of diseases from cancer to neurodegenerative disorders. The variety of oxidation and spin states displayed by TMs makes them a versatile tool to take part in a large number of biological processes. Metal ions stabilize permanent and transient protein-protein interactions in both natural and artificially designed proteins and are the drivers of metal-directed protein folding. The presence of metal ions or metal cofactors in the active site of enzymes remarkably enhances the diversity of functions displayed by these biomolecules.

Understanding how a substrate/drug binds to a biological receptor, how proteins assemble and interact with each other, how lipids and proteins aggregate in membranes, and how these events trigger or block a wide range of biochemical reactions is of paramount importance. The role of TMs in all of these events and its implications for diseases or (bio)catalyst and drug design is the goal of this project in the long-term. The aim of the current proposal is to design a computational approach to lay the foundations for the study of these phenomena.

Therefore, the main aim of the MetAccembly project is to design a computational approach that pave the way for studying the nature of these metal dependent processes at the atomic level and with a reasonable computational cost. Particularly in the last decade, molecular dynamics simulations of large biomolecules have undergone a step forward because of the increase in computational power and algorithms translating to longer and more accurate simulations going beyond the microsecond time scale. However, conventional MD simulations are associated with a high computational cost for describing biological events such as assembly, folding, or molecular recognition that take place in long time scales involving slow conformational changes that require milliseconds to be completed. There is a need for new techniques that speed up these processes. In this project, we focus our attention on accelerated molecular dynamics (aMD), a versatile enhanced sampling technique that speeds up molecular dynamics and does not rely on the a priori definition of reaction coordinates. This method has shown promising results in the field of protein dynamics and drug discovery.

In this project, we reformulated aMD with special focus on the fine-tuning of acceleration parameters and apply it to describe in detail the processes of assembly and biomolecular recognition. The new method has been assessed for a large number of different systems that involve protein folding, assembly of proteins and nanotubes, supramolecular host-guest systems, peptide assembly, and redox driven folding and recognition in metalloproteins.
The main aim of the MetAccembly project is to design a computational approach that pave the way for studying the nature of important processes such as assembly and biomolecular recognition at the atomic level and with a reasonable computational cost.

First, we designed and validated a novel computational protocol to study these important processes. We selectively tuned the parameters of aMD to enhance conformational sampling. This allowed us to efficiently explore the conformational landscape of four fast folding proteins (chignolin, Trp-cage, villin, and WW-domain). Starting from the extended conformation of the protein we were able to fold these proteins in less than 500 ns of simulation time with he new method (200 speed up compared to conventional molecular dynamics simulations). Thus, we identified a set of parameters that allow us to efficiently study the processes of self-assembly and (bio)molecular recognition. To validate the robustness of this novel computational approach, we assessed the performance of the finely tuned aMD parameters to study TM-induced protein folding and biomolecular recognition in small systems. In particular, we studied by means of aMD simulations the Zn(II) induced folding of N-terminal residues of the amyloid-β peptide, Aβ1-16, that encompass the metal binding region and the recognition process that take place in supramolecular host-guest systems (metalloporphyrin cages and cavitands).

Second, once the method has been proposed and properly assessed for small systems. We used the computational protocol to gain insight into the TMs-driven assembly and biomolecular recognition in more complex systems. First, We applied the computational protocol to study the process of assembly between a number of protein-metalloporphyrin nanohybrids and carbon nanotubes. Second, we explored the process of assembly of a peptide-based molecule that bears a photo- and electroactive π-extended tetrathiofulvalene (exTTF) unit at different temperatures. Third, we explored the Fe2S2 oxidation state driven conformational refolding in FeSII protein that is crucial for nitrogenase conformational protection.
The general aim of the MetAccembly project was to study metal-directed protein folding, molecular recognition and self-assembly using accelerated molecular dynamics. The first goal of this proposal was to develop a novel computational method based on accelerated molecular dynamics that can be used to gain insights into assembly, molecular recognition, and biocatalysis at a reasonable computational cost compared to other existing methods. This method has been developed and tested for a wide range of systems proving its robustness. The second research goal of the proposal was to combine state of the art techniques of electronic structure with enhanced sampling techniques to gain insight into the assembly, biomolecular recognition with applications in the fields of protein engineering, drug design, and supramolecular chemistry. With the help of accelerated molecular dynamics we have been able to study processes that were unaffordable with conventional molecular dynamics simulations.

In the framework of the MetAccembly project several collaborations have been established to study the processes of biomolecular recognition and assembly that will contribute to enhance the impact of the findings of this proposal. For example, we expect a high impact of the method in the field of computational design of novel enzymes. The field of computational enzyme design is still in its infancy and methods that accelerate molecular dynamics will be extremely useful to analyze the effects mutations on the conformational dynamics of the protein. This can help on the design of more effective and efficient enzymes that play important roles on the synthesis of pharmaceutical products. Another potential application of the computational approach proposed within the MetAccembly project is on the field of drug design. In particular, this method allows us study the process of ligand protein binding and unbinding events. Understanding how a substrate/drug binds to a biological receptor, how proteins assemble and interact with each other, how lipids and proteins aggregate in membranes, and how these events trigger or block a wide range of biochemical reactions is of paramount importance. This means that we can make an accurate prediction of the binding pose of a molecule in the active site of a protein can be obtained in a few nanoseconds of aMD simulation. This will be extremely interesting from the drug discovery perspective.
Summary of the Metaccembly Project