Metalloenzymes are important but challenging targets for computational studies because a reliable and computationally efficient potential function for the metal site is not straightforward to obtain. The challenges are particularly severe for metalloenzymes that feature structurally flexible active sites and/or dynamical metal coordination spheres, for which good examples include metalloenzymes with open active sites and thus promiscuous catalytic activities and proteins whose metal coordination sphere undergoes major redox/pH-coupled changes. Here we propose to develop novel QM/MM methods based on the latest formulation of density functional tight binding (DFTB3) approach to strike the proper balance of computational efficiency and accuracy for metalloenzymes.
The specific aims are: 1. Parameterize DFTB3 for closed-shell systems (Mg, Zn, P and S) to facilitate the study of reactivity in flexible metalloenzymes that employ Mg and Zn as co-factors. Refine a novel DFTB3-MM interaction Hamiltonian and a multi-level QM/MM free energy framework that judiciously combines DFTB3/MM and high-level QM/MM potentials. The methods will be carefully benchmarked with both small molecule and metalloenzyme systems. 2. As an application to closed-shell metalloenzymes, dissect the catalytic mechanism of AP superfamily enzymes with both cognate and non-cognate substrates. Test the hypothesis that the remarkable catalytic promiscuity observed for the AP superfamily members is due to the significant structural flexibilities of the bi-metallic active site, which is able to recognize trasition states of distinct nature for different substrates. Test the computational results by comparing relevant observables (linear free energy relations, kinetic isotope effects, mutation and thio-effects) to experiments. 3. Introduce orbital angular momentum dependence of the Hubbard parameters into spin-polarized DFTB approach for the description of open-shell species. We will focus on the parameterization for copper and the method will be systematically tested and refined by comparing structural and energetic (e.g., reduction potential and pKa) properties of small molecules and copper proteins. Apply the approach to resolve several outstanding questions regarding the pH dependence of redox properties in blue copper proteins, setting the stage for applications to more complex problems, such as the binding of copper to proteins implicated in neurodegenerative diseases.
The computational methods we develop will greatly facilitate mechanistic studies of metalloenzymes/metalloproteins implicated in key biological processes such as phosphoryl transfers, electron transfers and bioenergy transduction, which are often perturbed in devastating human conditions such as cancer and neurodegenerative diseases. Therefore, our research will have major long-term impacts on our ability to design novel strategies that combat these diseases.
|Cui, Qiang; Elstner, Marcus (2014) Density functional tight binding: values of semi-empirical methods in an ab initio era. Phys Chem Chem Phys 16:14368-77|
|Goyal, Puja; Qian, Hu-Jun; Irle, Stephan et al. (2014) Molecular simulation of water and hydration effects in different environments: challenges and developments for DFTB based models. J Phys Chem B 118:11007-27|