Nucleation of protein aggregation kinetics as a basis for genotype-phenotype correlations in polyglutamine diseases

Recent studies of inherited neurodegenerative disorders have suggested a linkage between the propensity toward aggregation of mutant protein and disease onset. This is particularly apparent for polyglutamine (polyQ) diseases caused by expansion of CAG-trinucleotide repeats. However, a quantitative framework for relating aggregation kinetics with molecular mechanisms of neurodegeneration initiation is lacking. Here, using the repeat-length-dependent age-of-onset in polyQ diseases, we derived a mathematical model based on nucleation of aggregation kinetics to describe genotype-phenotype correlations, and validated the model using both in vitro data and clinical data. Instead of describing polyQ aggregation kinetics with a derivative equation, our model divided age-of-onset (equivalent to the time required for aggregation) into two processes: nucleation lag time (a first-order exponential function of CAG-repeat length) and elongation time. With the exception of spinocerebellar ataxia (SCA) 3, the relation between CAG-repeat length and age-of-onset in all examined polyQ diseases, including Huntington's disease, dentatorubral-pallidoluysian atrophy and SCA1, -2, -6 and -7, could be well explained by three parameters derived from linear regression analysis based on the nucleated growth polymerization model. These parameters composed of probability of nucleation at an individual repeat, a protein concentration associated factor, and elongation time predict the overall features of neurodegeneration initiation, including constant risk for cell death, toxic polyQ species, main pathological subcellular site and the contribution of cellular factors. Our model also presents an alternative therapeutic strategy according to the distinct subcellular loci by the finding that nuclear localization of soluble mutant protein monomers itself has great impact on disease onset.