Protein disorder reduced in Saccharomyces cerevisiae to survive heat shock.

Recent experiments established that a culture of Saccharomyces cerevisiae (baker's yeast) survives sudden high temperatures by specifically duplicating the entire chromosome III and two chromosomal fragments (from IV and XII). Heat shock proteins (HSPs) are not significantly over-abundant in the duplication. In contrast, we suggest a simple algorithm to " postdict " the experimental results: Find a small enough chromosome with minimal protein disorder and duplicate this region. This algorithm largely explains all observed duplications. In particular, all regions duplicated in the experiment reduced the overall content of protein disorder. The differential analysis of the functional makeup of the duplication remained inconclusive. Gene Ontology (GO) enrichment suggested over-representation in processes related to reproduction and nutrient uptake. Analyzing the protein-protein interaction network (PPI) revealed that few network-central proteins were duplicated. The predictive hypothesis hinges upon the concept of reducing proteins with long regions of disorder in order to become less sensitive to heat shock attack.

Resolving differences in substrate specificities between human and parasite phosphoribosyltransferases via analysis of functional groups of substrates and receptors.

We herein review experimental and theoretical approaches widely applied to delineation of the differences in substrate specificities between human and parasite phosphoribosyltransferases (PRTases), the latter of which are key targets for treatment of diseases caused by parasites. Standard Molecular Dynamics (MD) simulations have been applied to determine why the human PRTase prefers guanine over xanthine, whereas the Tritrichomonas foetus enzyme exhibits only a slight preference. We analyze this problem with the aid of standard MD simulations, as well as constant-pH MD simulations. Comparison of results of the two approaches reveals substantial differences, e.g. several Asp and Glu residues in the parasite enzyme, and one Glu residue in the human enzyme, are predicted to be permanently or frequently protonated during constant-pH simulations, whereas standard MD simulations assume that these residues are always ionized. Most interesting is the observation of a large conformational change, leading to tighter binding of the ligand, observed in constant-pH MD simulations of the parasite PRTase complexed with XMP, and lack of such a change in the human enzyme complexed with XMP.