Testing the Ability of Standard Molecular Dynamic Software Force Fields to Accurately Model the Structural Features of Intrinsically Disordered Proteins
MetadataShow full metadata
A main tenet regarding the study of proteins states the structure of a protein defines its function. That is, the structure of the protein imbues characteristics that allow it to perform a specific function in the body. Intrinsically disordered proteins (IDPs) and intrinsically disordered regions of proteins (IDRPs) are common in all biological systems. It is estimated that up to forty percent of the human proteome is composed of IDPs or IDRPs. These proteins and regions of proteins, respectively, are responsible for gene regulation, cellular control, and molecular signaling pathways. Many human diseases are caused by mutations in IDPs and IDRPs, thus one could rationalize the mutated proteins fail to adopt appropriate structures to perform their intended function. It is estimated that up to fifty percent of human cancers are caused by mutations in the human tumor suppressor protein p53 (p53). p53 has an N-terminal IDRP mapping to residue positions 1 through 93. This region of p53 is crucial for transcription and apoptotic pathways. However, traditional techniques used to elucidate high resolution structural features (HRSF) of structured proteins, such as NMR and x-ray crystallography, are not amenable to the study of IDPs and IDRPs structural features, thus they remain unresolved. Previously, we used the intrinsically disordered N-terminal region of the human tumor suppressor protein p53 (1-93) as a model system to study temperature and sequence effects on the structural features of IDPs using a combination of computational and laboratory experimental techniques. Using AMBER, a molecular dynamics suite software package, we now perform atomistic molecular dynamic simulations of the IDP p53(1-93) with the ff12SB force field and the ff99SB force field in explicit solvent in an attempt to recapitulate the experimental results from the temperature and sequence effects studies. Molecular dynamic (MD) simulations have the potential to help resolve the structural features of IDPs and IDRPs with atomic-scale resolution. Unfortunately, MD simulations are only as good as the force fields used. That is, if the force field parameters unrealistically express physical relationships between atoms in the simulation the results can end up being quite artificial and thus useless in elucidating molecular descriptions of protein interactions and consequently, protein structure. Here we analyze the structures produced from completed MD simulations and compare them to our experimental measurements. This analysis shows that the ff12SB protein force field did not accurately model the structural features of p53 (1-93). The ff99SB force field simulations have not yet completed.