Computational studies of signalling proteins
Identifying structural changes in proteins is important, because it reveals how the enzymes function. We use time-resolved X-ray scattering to characterize structural change in light sensitive proteins. To structurally interpret these data it is essential to use computational methods. The systems under study are photoreceptor proteins such as cryptochromes and LOV-domain proteins. Analyzing the experimental diffuse X-ray scattering requires extensive molecular dynamics studies with long simulation times to be able to sample the light-driven conformational changes [Takala et al., Nature 2014, doi: 10.1038/nature13310]. Recently we have also implemented a method to use experimental X-ray scattering data to directly guide molecular dynamics simulations in Gromacs [Björling, … Westenhoff, J. Chem. Theo. Comput. 2015, doi: 10.1021/ct5009735]. We now plan to extend the calculations to fit experimental scattering data on cryptochromes and different LOV-domain proteins. We have collected X-ray scattering data sets for these proteins. In the past year we have used a medium SNIC allocation to model the structural changes in a LOV-domain protein (YF1; [Berntsson et al, submitted; Berntsson et al., submitted]). We have also investigated monomeric phytochrome using this resource [Takala et al., Struct. Dyn. 2016, doi: 10.1063/1.4961911]. Obtaining a new allocation for computational time is essential for us to be able to continue this successful line of research. Our results will help to identify the structural photocycle of photosensor proteins, which control the light response in bacteria, plants and animals. The simulations add a atomic scale resolution to the low-resolution experimental results and with this enhance the chemical relevance considerably.