Testing the Mutation Accumulation Theory of Aging Using Bioinformatic Tools

Abstract

Objective: An interplay of many variant mechanisms is thought to underlie aging or senescence. The Mutation Accumulation Theory proposes the accumulation of mutations in proteins to engender their aging phenotype. Testing whether random mutations lead to the aging phenotype was never done and is deemed infeasible. Bioinformatic algorithms provide an a-priori approach that allows testing; they employ prior knowledge of well-studied proteins to predict the changes brought on by mutations. Here, the Mutation Accumulation Theory of aging is tested using such bioinformatic tools. Methods: This is a simulation study, conducted 2017, September, using algorithms with Web accessibility. Three well-studied proteins implicated in aging were chosen: Collagen, Beta-amyloid Precursor Protein (β-APP) and Low-density-lipoprotein-receptor (LDL-receptor). Random mutations were introduced to their native coding sequences. Then, the mutated sequences were tested using three different prediction algorithms: SPpred for solubility, I-mutant for stability (delta-free energy), SNP and GO for pathogenicity. The new mutated phenotype was then correlated to the aging phenotype of the protein; decrease in solubility for Collagen and β-APP; and accelerated atherosclerosis for LDL-receptor. Results: 15 mutated variants for each protein (45 in total). For collagen and β-APP, the SPpred algorithm did not predict changes in solubility of the naked protein, but the I-mutant and SNP and GO definitely predicted changes that fit the aging phenotype. However, for LDL-receptors, none of the mutated variants when studied could account for the aging phenotype. Conclusion: for Collagen and β-APP, it is shown here that random mutations and their accumulation could explain the aging phenotype of both proteins; backing the Mutation Accumulation Theory for aging.