Purushottam Dixit

Thermodynamics of metal-protein interactions

The biological importance of metal-protein interactions can be appreciated by noting that nearly half the human genome codes for proteins that require a metal cofactor. Metal ions interact with protein amino acid residues with strengths rivaling covalent bonds, and as a result, they are exceptional agents in conferring stability and structure to proteins and polypeptides. In our quest to understand general principles of metal-induced stability of protein folds, and complementarily the selectivity of a fold for a particular metal, we have investigated two systems: (1) the metal selective site in the selectivity filter of the KcsA K+ channel, and (2) Zn(II) association and selectivity in a zinc-finger protein.

The KcsA K+ channel is an integral component of electrically active cells. This channel selects K+ ions over competing Na+ ions, and transports them across the cell membrane at near diffusion rate. A selectivity filter with ion binding sites S0 to S4 is responsible for the observed exclusion of Na+. We study the K+-over-Na+ equilibrium selectivity in the most selective S2 site of the filter. Paradoxically, the average binding energy of the ion with the site is much lower for Na+ than K+, and the difference is comparable to the difference in the bulk hydration free energy of the ion. Starting from the potential distribution theorem, we formulate a molecular theory to understand this puzzle. We find that although Na+ is better bound on average, its binding energy also experiences a large dispersion. We show that this feature is ultimately responsible for selectivity. Using simple thermodynamic arguments, we show how this greater dispersion in binding energy is related to properties of the ion-binding site.

The zinc finger peptide is part of the transcription factor machinery. Zn(II) is essential for the stability of the native fold of the peptide. The peptide selectively binds to Zn(II) over the competing Fe(II), Ni(II), Co(II), and Cd(II) ions in a site comprising two cysteine and two histidine residues in a tetrahedral arrangement. We show that the effective Hamiltonian of the metal-residue cluster in the protein is adequately represented by its Hamiltonian in vacuum plus a molecular field approximated by generalized harmonic restraints. Our model reproduces the thermodynamics of Zn(II) selectivity in quantitative agreement with experiments and helps elucidate the role of metal-residue chemistry and protein restraints in metal association in this protein. An analytical inspection of the protein field also suggests common characteristics among designed and natural zinc fingers. Possible utility of this approach in metalloprotein design is suggested.

Purushottam Dixit is a Doctoral candidate in Chemical and Biomolecular Engineering at Johns Hopkins University. For more information, please visit: http://shiva.che.jhu.edu/Purushottam/