Effect of mutations in SUMO protein on interaction with proteins involved in SUMOylation

Sumoylation is a reversible post-translational modification process, that affects the structure and subcellular localization of a wide variety of proteins. Plenty of proteins are controlled by SUMOylation, accounting for effects on many aspects of cell survival like control of gene expression, DNA replication and repair, mRNA processing and export, cytoskeleton assembly and signaling at the plasma membrane. Among these are a lot of proteins play a role in human diseases, including cancer, Huntington's, Alzheimer's, and Parkinson's diseases, spinocerebellar ataxia 1, familial dilated cardiomyopathy and amyotrophic lateral sclerosis. It is similar to ubiquitylation with respect to the reaction scheme and enzyme classes used, but rather than conjugation by ubiquitin, SUMOylation involves the addition of SUMO molecule to the lysine residue. Its yeast homolog SMT3 in nearly 50 % identical to SUMO1 and hence can act as a model to understand SUMO1 function. SMT3 is about 100 amino acid long protein in yeast that is conjugated to lysine residues of target proteins. Availability of nearly 20 PDB structure of SMT3 and complexes involving SMT3 (with Ulp1, PCNA, UBC9, etc.) on RCSB further aids this study. We start our study with yeast homolog SMT3 to facilitate easy validation of our results with in vivo studies and numerous other existing reports on SMT3. In this study we attempt to understand structure-function correlation of SMT3 by analyzing interface in the various complex it is involved. We intend to understand the effect of mutation in SMT3 on interaction with its partners involved in SUMOylation computationally by considering specific residues in SMT3 that are involved in interaction interface and create a mutated complex. We will relax the mutated structure and calculate the change in free energy caused by point mutations in the complex. 5 A + B --> AB ΔG (wild-type) A' + B --> A'B ΔG' (mutant) ΔΔG = ΔG' - ΔG This ΔΔG thus obtained can act as a measure of lethality of particular mutation. With the data acquired from our study, we may not be able to say that all interaction affecting mutations can be predicted computationally by Rosetta commons but mutations predicted by Rosetta are all affecting the interactio