The physical basis of how prion conformations determine strain phenotypes

Abstract

Particles of identical prion proteins can produce different phenotypes or 'strains' in vivo. This paradox is usually explained by differences in prion conformation. Tanaka et al. present an analytical model that describes how the interplay of various physical parameters of prions and prion particles in yeast leads to the emergence of a particular prion strain. Their experiments reveal that the strongest phenotype is in fact created by a slow growing particle with increased brittleness that promotes prion division. The tendency of prion particles to break up and generate new seeds for further growth may be a key factor in the large physiological impact of both infectious (prion) and non-infectious amyloids on their hosts. An analytical model describes how the interplay of various physical parameters of prions and prion particles in yeast leads to the emergence of a particular prion strain. The ability of prion particles to divide and generate new seeds for further growth turns out to be a key determinant of their physiological impact. A principle that has emerged from studies of protein aggregation is that proteins typically can misfold into a range of different aggregated forms. Moreover, the phenotypic and pathological consequences of protein aggregation depend critically on the specific misfolded form1,2. A striking example of this is the prion strain phenomenon, in which prion particles composed of the same protein cause distinct heritable states3. Accumulating evidence from yeast prions such as [PSI+] and mammalian prions argues that differences in the prion conformation underlie prion strain variants3,4,5,6,7. Nonetheless, it remains poorly understood why changes in the conformation of misfolded proteins alter their physiological effects. Here we present and experimentally validate an analytical model describing how [PSI+] strain phenotypes arise from the dynamic interaction among the effects of prion dilution, competition for a limited pool of soluble protein, and conformation-dependent differences in prion growth and division rates. Analysis of three distinct prion conformations of yeast Sup35 (the [PSI+] protein determinant) and their in vivo phenotypes reveals that the Sup35 amyloid causing the strongest phenotype surprisingly shows the slowest growth. This slow growth, however, is more than compensated for by an increased brittleness that promotes prion division. The propensity of aggregates to undergo breakage, thereby generating new seeds, probably represents a key determinant of their physiological impact for both infectious (prion) and non-infectious amyloids.