Background and significance
Metal ions exert a pivotal role across the fields of chemistry, biochemistry and material science. With their wide spectrum of coordination numbers, coordination geometries, thermodynamic and kinetic preferences for ligand atoms, and as well as redox activity, metals offer unique mechanisms of action often impossible to organic compounds, alone. In fact, metal-containing systems are exploited for multiple purposes. For example, metals are crucial for natural biological processes in proteins, while they are also of extreme importance for therapeutic applications, in particular for cancer treatment. Finally, metal-containing systems are relevant for technological applications that are poised to change our lifestyle, such as solar cells and new-generation batteries. It is therefore critical to understand, at the atomic level, how metals enable such a broad, yet fundamental, set of functions.
One key example of how metals operate in biochemistry are metalloproteins, which have been widely studied over the years by the computational community (Palermo G. et al., Acc Chem Res 2015, 48, 220). However, much remains to be done in order to fully clarify the mechanisms of recognition and selectivity used by metalloproteins. Moreover, some of the main issues of computational methods in treating metal ions has still to be overcome. For instance, the limitations of the current classical force fields (FFs) are well known, which still call for better ways to treat the metal-ligand interaction when running classical molecular dynamics simulations (Li P. & Merz K. M. Jr., Chem Rev 2017, 117, 1564). Equally, when metals interact with nucleic acids, like in the case of Mg ions and DNA, the FF accuracy is also of paramount importance. In these cases, mixed quantum-classical (QM/MM) MD simulations are required, although the quantum (QM) region, usually described using density functional theory (DFT) for its cost effective inclusion of correlation effects, is often plagued by the accuracy of the exchange correlation functionals (Roethlisberger U. & Carloni P. Lect Notes Phys 2006, 704, 449).
Other relevant examples are metal-drugs interactions, like in the case of well-known anticancer agents (Adhireksan Z. et al. Nat Commun 2017, 8, 14860). Besides traditional Pt based agents, organometallic and transition metal compounds based on Ru, Os, Rh and even Au, are showing increasing potential in the treatment of cancer. Tuning the physico-chemical properties of these metal complexes is a successful strategy for increasing the selectivity and reducing the toxicity of metal complexes. In this respect, predicting properties of metal complexes requires high-level computations, usually at QM level. Moreover, a QM/MM scheme is essential for realistically capture the mechanistic action of metal drugs at the level of the biological target (Palermo G. et al ChemMedChem 2016, 11, 1199).
Finally, the field of material science also greatly benefits of the employment of metal ions. Metals like Pb, Sn, Ti, Cs, Rb are the core of the revolution in the field of new materials for solar cells technology (Sygantseva O. et al. J Phys Chem Lett, 2017, 8, 1191), as well as for next-generation batteries (Zhong Y. et al. J Phys Chem C 2017, 121, 14222). In this respect, QM calculations are key for predicting the electronic structure properties, which are key in the design of more efficient materials. High accuracy in computing properties such as band gap, stability and polarization, relies on advances the development of novel functionals for DFT calculations accurately including all relevant interaction terms.
Our workshop will revolve around four specific goals, as described below:
1) Facing the challenges of computing metals
We will open the discussion on the challenges of computing metals at the classical and Quantum Mechanical (QM) levels. Each metal, with its own coordination geometry and electronic structure represents a challenge for molecular simulations, due to the limitations of the current FFs. We will focus on the development of site-specific force-field (FF) parameters, which is promising to overcome the limitations of current models. On the QM side, high-level methods are able to describe multiconfigurational states, reproducing the correct spin multiplicity and estimating possible charge-delocalization, resulting useful in the case of transition metal complexes. Density Functional Theory (DFT) protocols (spin projections, spin flipping) and post-HF methods will be under intense debate in this workshop. Modeling metals considering their realistic biological environment fundamentally relies on the use of a quantum mechanics/molecular mechanics (QMMM) scheme. We will discuss the development of polarizable QMMM schemes that can help in dealing with problems such as the “spill- out” effect, arising from the unphysical interaction between the electronic density and the classical charge at the boundary between the QM and MM partitioning.
2) Metal ions and biological function.
Here, three major topics will be discussed:
Recognition and binding. We will focus on studies on metal ion recognition and binding in proteins and nucleic acids. We will deepen the role of computations in deciphering how protein conformational plasticity allows the selection of the metal ion, as well as on the mechanisms underlying the metal-aided catalysis.
Steric vs electronic effects. We will discuss the role of QM-methods in deciphering how steric and electronic effects affect enzymatic function. QM-methods can (i) discern steric and electronic effects of metals during catalysis; (ii) provide information on which metal is most efficient/selective into the enzyme pocket; (iii) explain the structural/electronic basis for different catalytic efficiency of “similar” metals (i.e. Mg, Ca, Zn or Mn) within enzymes.
Electron transfer. Charge transfer has particular relevance for processes such as respiration and photosynthesis. Metalloproteins are key players, being able to fine-tuning their redox potential, and providing natural binding centers for the moving electrons. We will discuss on the role of QM in providing critical information – usually out of reach via experiments – on redox properties and electron transfer processes.
3) Metal ions for drug discovery
Recent advances in metal-based anticancer treatment have shown that combined computations and experiments allow characterizing the mechanism of action. We will discuss how QMMM schemes can be fully integrated with a variety of experiments, characterizing the action of metal centers at the level of proteins and DNA, elucidating the cytotoxicity mechanism and impact on cancer cell function.
4) Metal ions in material science.
QM calculations can predict electronic structure properties for the design of efficient materials. High accuracy in computations relies on the development of novel exchange correlation functionals including all relevant interactions, with attention to dispersion and spin-orbit coupling. All these points will be discussed in light of the applications for solar cell technology and next-generation batteries.