Membrane proteins typically constitute 30% of the protein coding genes in all organisms, and are targeted by a majority of the drugs developed. Drug development, and understanding of the molecular mechanisms involved in diseases, requires knowledge of the three-dimensional structure of mammalian membrane proteins. This is very challenging to obtain through experiments, and today only 124 unique membrane protein structures exist, compared to over 40.000 water-soluble protein structures. Only three of these solved membrane protein structures are human proteins. However, inferring structure knowledge from bacterial homologues of human membrane proteins has proven very successful. In our research, we combine three-dimensional structure modelling and molecular dynamics (MD) simulations to gain insights in the structure and mechanism. Our structural modeling of a human  neurotransmitter transporter protein, based on the structure of a bacterial homologue in the major facilitator super-family (MFS) of membrane transporter proteins. The model is used to explain observations from mutagenesis- and transport inhibition experiments. Because of low sequence similarity between the modelled protein and its template, a fundamental question is whether or not homology modeling is suitable. Besides describing the structure of the model, and showing that its surface properties resemble those of integral membrane proteins, we have carried out molecular dynamics simulations on both the model and its template to assess the stability of the model.

References

Jonas Almqvist, Yafei Huang, Aatto Laaksonen, DaNengWang and Sven Hovmöller,
Modelling the docking and inhibition of human vesicular glutamate transporters,
Protein Science, 16(9) 1819-1829 (2007).