Project Summary: Metalloenzymes play various important biological roles and therefore are major targets for biomedical research. They also drive further development of computational methodologies that can strike the proper balance of accuracy and sampling efficiency. Encouraged by progress made in the last funding period, we continue to develop hybrid quantum mechanical/molecular mechanical (QM/MM) methods to un- derstand the catalytic mechanism of metalloenzymes that play major roles in key biological processes such as phosphoryl transfers and DNA replication. We will conduct extensive comparison of kinetic isotope effect (KIE) and combinatorial mutation effects with experiments to calibrate our methodologies. The specific aims are: 1. Further develop an approximate Density Functional method (DFTB3) for transition metal ions in biological applications. This involves: (i). improving the description of polarization and charge transfer of metal-ligand interactions for charged ligands, guided by the Natural Bonding Orbital analysis; (ii). establishing high quality benchmark dataset for metal-ligand interactions using highly correlated QM methods such as Density Matrix Renormalization Group with Canonical Transform theory; (iii). including explicit on-site d - d interactions at the orbital rather than population level in the framework of ligand-field theory. 2. Enhance mechanistic understand- ing in the roles of metal ions in phosphoryl transfer enzymes. Through a combination of QM/MM free energy and KIE calculations, we will: (i). explain why is the phosphoryl transfer transition state in phosphatase-1, but not in alkaline phosphatase, substantially modified relative to solution, despite their generally similar bimetallic active sites; establish whether the difference is dictated by the identity of the metal ions (Zn2+ vs. Mn2+), the distance between them or the distribution of charges/dipoles in the active site; (ii). quantify the catalytic con- tribution of the third “transient” Mg2+ to DNA polymerase ⌘ identified in recent time-resolved crystallography studies, and establish the impact of this ion on the mechanism of 3'OH activation. 3. Integrate DFTB3/MM and DFT/MM methodologies to provide a mechanistic understanding of co-operativity associated with various “catalytic modules” identified in alkaline phosphatase through combinatorial mutation of key motifs in the active site. The broad range of catalytic activities of these mutants, which span ten orders of magnitude in kcat/Km, provides an unprecedented opportunity to test and calibrate QM/MM methods. In the long run, our efforts will help establish “best-practice” QM/MM protocols that are able to aid rational design of metalloenzymes and understand their evolution.

Narrative: The computational methodologies we develop will be applicable to a broad set of metalloen- zymes of biomedical relevance. In particular, we target fundamental mechanistic problems in enzymes that catalyze phosphoryl transfers and DNA synthesis, since mutations in these enzymes are implicated in numer- ous human diseases such as cancer. Although our project does not focus on design of drugs, the mechanistic insights obtained in our work will broaden strategies that can be used to target these enzymes.