Karthik Arumugam
Structure prediction of P1-type ATPases and molecular dynamic simulation on their Metal Binding Domains.
Published on 12 November 2009
Body text 1
Thesis presented November 12, 2009
Abstract:
P1-type ATPases are a special kind of ATP driven pumps which transport soft heavy metals (Cu+, Zn2+, Pb2+, Cd2+) across the cell membranes. Their complete structure is, in general, unknown. We are interested in the structure and dynamics of the transmembrane part of the Cadmium ATPase (CadA) and the metal binding domains of the human Copper transporting Menkes ATPase. Sequence similarity and hydropathic analyses, completed by experiments have shown that soft metal pumps are constituted of 8 transmembrane regions (TMs), compared to 10 for SERCA, the calcium pump for which the 3D structure is known. In collaboration with the biochemists in the laboratory, and using standard programs like MODELLER, CHARMM, XPLOR, AMD together with our own programs, we have predicted the structure of the TMs of CadA. We have built several models of the TM bundle corresponding to several topologies, calculated all atom coordinates with a procedure similar to structure determination from NMR experiments and refined these coordinates using Molecular Dynamics (MD) simulations in an implicit membrane. The Adaptive MD (AMD) program has been used to interactively check the models for bad loop positioning or ''knots'' in the structure. Another interesting feature of P1-type ATPase is the presence of one to six N-terminal conserved GxTCxxC metal binding domain(s) (MBDS). In the case of the Menkes ATPase, there are 6 MBDs, each of them being able to bind Cu+. The structure of each MBD separately is known from NMR spectroscopy but the structure of the assembly is unknown. Using MD simulations, we have studied the dynamics of each MBD in the presence or absence of metal with the final goal of understanding how the metal is transferred from the metallochaperone which binds the metal when it enters the cell to the MBD and why the presence of 6 MBDs is needed for the correct functioning of the pump. These studies have used recent work of researchers in the team on the parameterization of metal ions for molecular mechanics force fields and MD studies on metallochaperones. We show that fast approximate
in silico models of metal ions can help understand metal binding and transport in metalloproteins or ATPases, which are known to be major pharmaceutical targets.
Keywords:
P1-type ATPases, MBDs, Molecular Dynamic simulation, CadA
Download this thesis.
Top page