How might stress influence prion formation and inheritance?
Answer: Stress directly reduces proteins’ abilities to fold or to maintain their native folds. Consequently a prion domain may be able to access an alternative, nucleating conformation under stress. Additionally, stress is likely to change the dynamics between chaperones and prion proteins. Accumulation of misfolded proteins in the cell could titrate chaperones away from prion domains, or conversely, the stress-induced upregulation of chaperones may increase their activity on prions. Finally, stress-induced protein aggregates could directly cross-seed prion proteins by serving as an imperfect amyloid template.
Prions may accelerate evolution. Is it possible for organisms to evolve mechanisms for evolvability?
Answer: This is a difficult and provocative question. Kind of a chicken-and-egg scenario. For a thorough treatise, refer to: Kirschner M and Gerhart J. (1998) Evolvability. Proc. Natl. Acad. Sci. 95, 8420-8427
From a structural standpoint, consider the different steps that are involved with prion replication – association, conformational conversion, and fiber division.
A) How do the non-prion conformer and the prion conformer recognize and associate with one another? What are the attractive forces?
Answer: Amyloid is essentially a continuous ?-sheet, a structure determined by main chain hydrogen-bonding. In a ?-sheet, each residue forms hydrogen bonds with other residues that are distant in sequence but close in space, such that each strand in the sheet is continuously H-bonded to the strands both above and below it. At the ends of sheets, however, this pattern is broken. In globular proteins, the end strands typically satisfy their H-bonding potential by interacting with other, discontinuous segments of the protein. But in amyloid, the ends of the sheet have unresolved H-bonding potential and are thereby free to interact with other polypeptides. The side-chains also play an important role in these interactions. In solved amyloid structures, residues like glutamine and asparagine create additional networks of H-bonds, even forming continuous stacks (or “ladders”) running along the fiber axis. Hydrophobic side chains, too, can stack and create additional stabilizing contacts.
B) What are the characteristics of the non-prion conformer that allow templating to happen? What causes specificity?
Answer: There are two major considerations: sequence and structure. Considering the role of side chains discussed above, it is easy to imagine how a polypeptide’s sequence dictates specificity. The ends of an amyloid fiber display a region of the polypeptide’s linear sequence. An in-coming polypeptide can only maximize the number of homotypic stabilizing contacts (hydrophobic interactions, amide H-bonds, etc.) when its sequence is identical or very similar to that displayed on the ends of the fiber. Secondly, the protein molecule to be templated must have those residues exposed in order to interact with the fiber. This cannot happen if they are already involved in some other interactions in the soluble protein’s structure. Presumably, the highly flexible or natively unstructured character of prion domains (the regions of the protein that form amyloid) allows maximum accessibility of these residues. This facilitates amyloid nucleation and rapid templating of soluble subunits.
C) What variable property of the amyloid fibers themselves might be important for prion propagation?
Answer: Amyloids must fragment at a rate that is greater than the rate of cell division in order to be prions. Amyloids differ in their stabilities, or ability to be easily broken into smaller pieces. More “fragile” amyloids generally create stronger prions.
Aggregation generally inhibits protein activity. This can be a direct consequence of proteins failing to access their native, active conformations. Evidence for the mammalian prion protein, PrP, suggests that much of the protein undergoes extensive conformation rearrangements in the prion state. However, other known prions are modular and have a distinct region of the protein that is involved with aggregation, and another (non-prion) region that generally retains its native fold and activity. Why then do prions often cause loss-of-function phenotypes?
Answer: Aggregation inactivates modular prion proteins in (at least) two ways. First, aggregation can sterically impede binding of the prion proteins’ normal interaction partners. Secondly, prion polymers are very large complexes relative to the native soluble forms of these proteins. They have limited diffusibility and may be too bulky to access some regions of the cell. Thus, aggregation causes mislocalization of the prion protein. Transcription factors that are prions, for instance, are occluded from the nucleus in their prion state. It should be mentioned that cells appear also to have an active mechanism to confine prion aggregates to a specific compartment of the cytoplasm.
There are also examples of gain-of-function prions. How might prions cause increased protein activity?
- The prion conformation can directly increase the protein’s function. The prion conformation can serve as a scaffold for biochemical processes, like mRNA binding. One protein that is involved in the activation of neuronal synapses is thought to become active and bind certain mRNAs specifically when it is in its putative prion conformation (see Si et al. 2003). Similarly, the yeast prion protein Rnq1 has no known function in its non-prion form, but in the prion form it interacts with other prion proteins and “cross-seeds” them into their prion forms. Yet another example is the [Het-s] prion in the filamentous fungus Podospora anserina, whose sole known function is to prevent cytoplasmic fusion between genetically divergent colonies. It does so only when the protein is in its prion state, by forming a toxic protein complex with a related protein that is uniquely expressed in the non-prion containing colony.
- The prion aggregates can sequester that protein’s interaction partners. The normal, non-prion functions of prion proteins require their interacting with other proteins. Sup35, for instance, interacts with Sup45 to fulfill its translation termination function. When Sup35 is overexpressed in [PSI+] cells, the cells can no longer grow in part because the aggregated Sup35 has sequestered too much Sup45, which is essential. A protein’s aggregation can thus cause loss-of-function phenotypes for other proteins, which cannot be mimicked by reduction of the prion protein’s normal activity by, for instance, gene deletion.