Replication Mechanisms.
Prions are the autocatalytically replicating proteins (Bieschke et al 2004) responsible for BSE, CJD, Scrapie and GSS (Prusiner 1998). These diseases have genetic, sporadic and infective features, suggesting an interesting pathophysiology, which any molecular mechanism must explain. The prion protein is very stable resisting high temperatures, formaldehyde and UV light. The infective agent PrP-Sc differs from the host-produced protein PrPc (present on the plasma membrane of neurons) only in conformation. PrP-Sc is more structurally stable, being rich in Beta-sheets (Wille et al 2002), and having a lower free energy of formation. It is thought that a high activation energy limits the rate of conversion of PrPc into PrPSc (Cohen and Prusiner 1998, see figure 0aa ), so that PrPc is much more common even though PrPSc is thermodynamically more stable. The conversion is thought to be due to the induction by PrPSc of the conversion of PrPc alpha-helices to PrPSc beta sheets (See figure below. Left PrPc, Right PrPSc).
Cohen and Prusiner suggest that if the effect of single point mutations in autosomally recessive sporadic disease were to reduce the activation energy, the all-or-none property of prion disease could be explained, whereas they suggest it is unlikely that a point mutation to the prion protein could alter the free energy difference between PrPc and PrPSc sufficiently to produce this all-or-none effect.
Figure 0aa. From Cohen and Prusiner 1998.
The incubation period for prion diseases is in the order of years. The infective dose depends on the method of inoculation (large doses into the intracerebral space being more effective than ingestion. Prions are interesting for our purposes since they are an example of a 'simple' naturally occurring self-replicating entity. Of-course, prion self-replication depends on nucleic acid self-replication since translation of the PrPc protein must occur in each organism if it to be infected with PrP-Sc.
Several models exist to explain the mechanism of prion replication. Several mechanisms have been suggested. The first was that PrP exists in several stable, noninterconverting, monomeric conformational isomers.
Prusiner's autocatalytic mechanism for Prion Replication is a standard model of an autocatalytic cycle (See Figure 0a). Eigen is skeptical that such a mechanism can work because the rate enhancement provided by the cycle for the conversion of C to Sc would have to be very great, to account for the fact that spontaneous disease is so uncommon (i.e. a rate of direct conversion of c-> sc less than the metabolic decay rate of Sc), whereas disease is possible with an incubation period in the order of years, if Sc is introduced externally. For example, k c-> sc must be no greater than 10^-22 s^-1 if less than one molecule of C is to be transformed into Sc per year, given nanomolar concentrations of C. This is required if spontaneous disease is to be rare. However, since one molecule of Sc is sufficient to start an autocatalytic explosion, even if the turnover period of the cycle was once per year, then Sc would reach the same concentration as C within 34 years. However, the autocatalytic period cannot be this long since infective processes occur at a much greater rate. Therefore, the ratio of autocatalytic to direct production of B, kt/(k c-> sc), must be very high, Eigen calculates 10^15. Eigen considers a catalytic enhancement of greater than 10^15 unlikely given a non-cooperative autocatalytic mechanism. Therefore he rejects the idea of non-cooperative autocatalytic prion formation based on the kinetic implausibility of such a high rate enhancement. This simple mechanism with realistic rate enhancements would result in either total absence of disease, or spontaneous appearance of disease.
Figure 0a. Prusiner's autocatalytic model of Prion replication.
Instead he suggests a cooperative version of Prusiner's mechanism: Cooperative autocatalytic mechanism (Eigen 1996) (See figure 0b). Again the rate of spontaneous B (i.e. PrP-Sc) formation must be low. Catalysis involving only one B molecule must be slow, so that in the absence of infection, [B] cannot increase easily, i.e. linear autocatalysis must be suppress
ed. Coorperativity imposes a threshold on B, above which autocatalytic growth occurs, and below which the metabolic decay of B can control further growth.
Figure 0b. From Eigen (1996). A view of a cooperative mechanism of PrP-Sc production. Here PrP-Sc (B) reacts with PrPc (A) forming AB, and BB, but BB does not immediately decay into 2B, instead a further 2 Bs are incorporated producing BBBB which then decays into 4B. Several prion proteins are required to catalyse the formation of a prion protein.
Cohen's Heterodimer Mechanism (Cohen et al 1994) has it that the infectious agent PrP-Sc is a dimer. PrPc exists in equilibrium with PrP*-Protein X complex. If this binds an exogenous PrPSc dimer, the resulting complex causes a conformational change in PrP*, (lowering the activation energy barrier that normally exists) converting it into another PrPSc. Protein X can no longer be bound and so is released to bind another PrPc. A new PrPSc monomer has been formed.
Figure 0bb.
Lansbury's Plaque Formation or Nucleated Polymerisation Mechanism is an alternative mechanism shown below (Figure 0c,i and ii). Lansbury suggests that PrP-Sc and PrPc differ primarily in their quaternary structures, and that the conformational differences are a result of this. This is in contrast to any model in which a unimolecular conformational change is rate limiting for PrP-Sc formation. PrP-Sc is a less stable conformation which is stabilized by intermolecular (allosteric) interactions with other PrP-Sc proteins. Nucleation, i.e. the formation of the oligomer of Bj (PrP-Sc) is rate limiting. Lansbury claims that mutations that cause genetic prion disease result from alterations to this nucleation rate, in a manner analogous to how a point mutation causes sickle-cell disease by haemoglobin fibril formation. The bottleneck up to Bj, serves to limit the rate of spontaneous disease, and the rapid growth after Bj accounts for the rapid growth in the disease phase. Bj growth is autocatalytic if the aggregates undergo fission, producing more small seeds. Fission is more likely with linear crystals. Infective particles will be those greater than or equal to Bj in size. Lansbury's own figure (ii) looks more plausible than Eigen's reworking of it (i). since it includes the fission steps and hints at a structural basis underlying the kinetics of the slow and fast steps.
Figure 0c i. From Eigen (1996). According to Lansbury, A molecules exist in equilibrium with B molecules. B molecules aggregate slowly, until a threshold size whereupon they aggregate much faster, growing indefinitely.
Figure 0c ii. From Lansbury (1995).
Figure 0d. From Lansbury (1995). Mouse PrP-Sc (Green) can take up Hamster PrPc (yellow) but more slowly than it can take up mouse PrPc (green). The host does not inherit the prion configuration or sequence of the infective agent.
Lansbury also provides a mechanism to explain the phenomena of interspecies infection. The incubation period of interspecies infection is longer than intraspecies infection, however, the resulting PrP-Sc strain in the infected species is that of the host, with the seed PrP-Sc only present in trace amounts. Mouse to Hamster incubation time is initially long, but shortens in subsequent hamster to hamster transfers. This is explained by the mechanism in figure 0d, showing heterologous nucleation.
Finally, Lansbury's mechanism explains the fact that for a PrP-Sc with identical sequence, several strains can be isolated with distinct incubation periods and pathology. Lansbury suggests this may be due to the possibility of alternate packing structures of aggregates. Infection in this case can result in the host inheriting the PrPSc* strain if the packing structure of the infective PrP-Sc determines the structure of the host polymer (see figure 0e.).
Figure 0e. From Lansbury (1995).
Another model of the nucleated polymerisation mechanism was developed by Poshel et al. and is shown in the diagram below (Fig 1). Here there are two kinds of entity, PrP-c monomers (produced by the host and present on the surfaces of neural tissues), and unbranched PrP-Sc polymers. Typically these are observed to be of apx 200 monomeric units in length. (Jeffrey et al 1995) (Prusiner 1999). These are known to grow in one dimension in both directions (Scheibel et al 2001). The prion polymers are the replicating entities, which grow by incorporating PrPc at a rate Beta. Initial formation of PrP-Sc oligomers is assumed to be so slow, and it is ignored. Polymers fragment into smaller polymers at a rate proportional to their length. Below a certain length, n, they are unstable and decay back into monomers (Lansbury and Caughey, 1995), because they cannot form a stable B-helix (Willie et al 2002). Thus only the infective case in which some PrP-Sc > n in length already exists in the system is considered. ODE and stochastic modeling is carried out, demonstrating the existence of a steady state distribution of PrP-Sc polymer lengths, dependent on a, b and Beta, not Lambda, or d. Aggregation of oligomeric components is ignored. Unfortunately, no experimental data exists of the length distribution of PrP-Sc in infected animals and so the model predictions cannot yet be confirmed. Also, it is not clear which PrP-Sc properties correlate with disease severity, i.e. PrP-Sc total mass, or PrP-Sc length distribution.
Figure 1. Poschel Model of Prion Replication Kinetics. PrPc exists as monomers, whereas PrPSc exists as polymers that incorporate PrPc at a rate Beta. PrPc is produced at rate Lambda and lost at rate d. PrPSc polymers are lost at rate a, and break into smaller fragments in proportion to b(i-1) where i is the length of the polymer. Below a threshold length n, polymers degrade 'instantaneously' into monomers.
V. Vitagliano and G. DÏErrico (2001) were interested in the possibility of multistationary state kinetics in autocatalytic systems when maintained far-from equilibrium. For example, the system may grow exponentially, or may collapse, depending on the threshold concentration of food available. They use the kinetic model shown in Figure 2 to address how infection dynamics in prion diseases may be explained by this multistationary kinetics. There are two stable state's for [PrP-Sc]. One high and one low. Which is entered depends on the initial concentrations of PrP-Sc and PrPc (see figure 3). Although there is a steady state for [PrPSc] as also argued by Poschel. Vitagliano et al claim there is not always a steady state of C (the clinical effect) after infection. This is due to the fact that in their model C depends on the ratio of [PrPSc] and k0. In the model, C grows linearly if [Z] >k0 and collapses to zero if [Z] < k0.
Figure 2. V. Vitagliano and G. DÏErrico (2001) model of prion replication kinetics. A molecules are metabolised into the PrPc protein (Y, shown in blue). Spontaneous formation of PrPSc (X, shown in orange) occurs at rate k1. Z then replicates by a third order autocatalytic reaction requiring collision between 2 Z's and one Y, releasing 3 Zs, at rate k2. Z is also converted into C which can be degraded to P, as can Y. C represents the damaging effect of Z on cells.
Figure 3. (From V. Vitagliano and G. DÏErrico., 2001) Below trajectories 3 and 6, the stable steady state S1 is reached. Above 3 and 6, the stable state S3 is reached. S2 is a saddle point.
The authors justify the 3rd order autocatalytic reaction since "according to the literature, more than one pathogenic molecule is required to convert one healthy molecule". However, this does not necessarily justify a third order reaction, since two serial second order reactions could also explain the need for more than one pathogenic molecule. They further claim that "From a mathematical point of view third order autocatalysis is the simplest step leading to a multistationary kinetics." Is this actually true, or is a second order reaction sufficient to show multistationary kinetics? The system of differential equations used by the authors was modeled using mathematica (Download .nb file here).
Prion Strains and Evolution.
The capacity for different strains of PrP-Sc prions to utilize PrPc of other strains is of considerable importance, since this may determine whether prions infecting one species will be able to infect other species. Lansbury provides one explanation for the mechanism of cross-strain seeding in the nucleated polymerisation model.
Selective Pressure for Prion Evolution.
Positive Selective Pressure for PrPc.
At the clinical level, work by Sakaguchi on mice in which the prion producing gene has been removed shows that old aged mice knockouts suffer from greater degenerative neural symptoms, "They developed an odd, uncoordinated gait. Their back legs began to tremble as they tried to walk, they took short tentative steps and were unable to keep a straight path. As they got older, their coordination became progressively worse, and they frequently collapsed. By ninety weeks, the mice had deteriorated even further. Hardly able to stand, most had developed spasmodic arching of their backs." (Nature 11 April 1996). These are similar to the symptoms of CJD suggesting that loss of PrPc rather that gain of PrPSc may be responsible for the premature death of Purkinjie cells observed in these mice.
Positive Selective Pressure for PrP-Sc.
At the molecular level, Tompa et al (2001) have proposed that the prion protein originated as a transmembrane protein that underwent a mutation to become a topologically frustrated extracellular protein (one with many different low free energy stable configurations). Based on the fact that in lower eucaryotes, prion proteins act as autocatalytic molecular switches mediating epigenetically heritable traits transmitted between cells during mating, they suggest there may have been positive selection pressure for the prion producing mutation to have evolved. The capacity for self-replication of a lower level unit of selection (prion) could confer advantages to the higher unit (yeast). The yeast prions are
1. PSI+ , a prion form of the translation termination factor Sup35p.
2. URE3, a prion form of the transcriptional activator Ure2p, normally responsible for nitrogen catabolism repression.
If prions are to be epigeneticly inherited in yeast, their replication must be able to keep up with cell division. It is thought this may be accomplished by them growing in size by aggregation, dividing with the help of cellular chaperones, and being distributed to the daughter cells by passive diffusion. Osherovich et al have examined which parts of the prion protein are responsible for these functions.
Aggregation Function: Mutation of the glutamine/asparagine-rich (Q/N-rich) amino-terminal (N) domains of these proteins interfears with their ability to aggregate. "Aggregation and specificity are dictated by the NYN repeat (residues 70–100) of New1p and by the Q/N-rich amino terminal region (residues 1–57) of Sup35p." (Osherovich et al 2004). In vitro, Q/N-rich domains form self-seeding, Beta-sheet-rich amyloid fibrils, that show preferential aggregation with prions manufactured in the same yeast species.
Division: Hsp104p is a molecular chaperone essential for prion division. Many Q/N rich proteins capable of forming amyloid like aggregates have been found, but only a few are capable of propogating these aggregates over generations. If the Lansbury theory were true, one hypothesis would be that non-transmissible Q/N rich protein aggregates may be too stable to undergo fission to produce further infective seed particles. Indeed, Osherovich et al found that distinct sequence elements are responsible for aggregation by incorperation of monomers, and for generation of new seed particles. "In New1p, the NYN repeat alone can aggregate and induce [NU þ ] but requires an adjacent oligopeptide repeat to form a minimal heritable New1p prion, [NU þ ]mini. Similarly, in Sup35p, the Q/N-rich amino terminal region mediates aggregation whereas most of the oligopeptide repeats are needed for the inheritance of [PSI þ ] propagons." (ibid, p 448). They propose that "oligopeptide repeats turn nonheritable aggregates into prions by facilitating chaperone-mediated division."
The proposed mechanism of replication is shown in figure 4 below.
Figure 4. From Osherovich et al 2004.
Prion Variation and Inheritance.
Whereas RNA and DNA have the capacity for unlimited heredity, i.e. any sequence of RNA or DNA is capable of self-replication (if appropriate enzymes are present), it is not the case that any sequence of protein is capable of self-replication. Even the former claim for the 'inherent' self-replicating ability of RNA and DNA must be qualified since in fact it has so far not been possible to obtain sequence independent non-enzymatic RNA template replication.
Bibliography.
Jan Bieschke, Petra Weber, Nikolaus Sarafoff, Michael Beekes, Armin Giese, and Hans Kretzschmar. Autocatalytic self-propagation of misfolded prion protein. (2004) PNAS,101(33) , 12207–12211. Download pdf.
Cohen, F. E., K. M. Pan, Z. Huang, M. Baldwin, R. J. Fletterick, and S. B. Prusiner. 1994. Structural clues to prion replication. Science. 264:530–531.
Cohen, F.E. Prusiner, S.B. Pathologic Conformations of prion proteins. (1998) Annu. Rev. Biochem. 67, 793-819. Download pdf.
Eigen, M. 1996. Prionics or the kinetic basis of prion diseases. Biophys. Chem. 63:A1–A18. Download pdf.
Jarrett, J. T., and P. T. Lansbury. 1993. Seeding ‘‘one-dimensional crystallization’’ of amyloid: a pathogenic mechanism in Alzheimer’s
disease and scrapie? Cell. 73:1055–1058.
Jeffrey, M., I. A. Goodbrand, and C. M. Goodsir. 1995. Pathology of the transmissible spongiform encephalopathies with special emphasis on ultrastructure. Micron. 26:277–298.
Lansbury, P. T., and B. Caughey. 1995. The chemistry of scrapie infection: implications of the ‘‘ice 9’’ metaphor. Chem. Biol. 2:1–5. Download pdf.
J. Masel, V. A. A. Jansen and M. A. Nowak, (1999) Biophys. Chem., 77, 139.
G. Nicolis and I. Prigogine, Self-Organization in Nonequilibrium Systems, Wiley, New York, 1977. [Multistationary kinetics]
Osherovich, L.Z. Cox, B.S. Tuite, M.F. Weissman, J.S. Dissection and Design of Yeast Prions. (2004) PloS Biology, 2(4), 442-451. Download pdf.
I. Prigogine, Introduction to T hermodynamics of Irreversible Processes, Wiley-Interscience, New York, 1967. [Multistationary kinetics]
Poschel, T. Brilliantov, N.V. Frommel, C. Kinetics of Prion Growth. (2003) Biophysical Journal, 85, 3460-3474. Download pdf.
Prusiner, S.B. Prions, (1998) PNAS, 95, 13363-13383. Download pdf.
Prusiner, S. B. (editor). 1999b. Prion Biology and Diseases. Cold Spring Harbor Laboratory, New York. [Templated Assembly Model.]
Serio, T. R., A. G. Gashikar, A. S. Kowal, G. J. Sawicki, J. J. Moslehi, L. Serpell, M. F. Arnsdorf, and S. L. Lindquist. 2000. Nucleated
conformational conversion and replication of conformational information by a prion determinant. Science. 289:1317–1321. [ Conformational Conversion Model.] Download pdf.
Shorter J, Lindquist S (2005). "Prions as adaptive conduits of memory and inheritance". Nat Rev Genet 6 (6): 435-50. Download pdf.
Tompa, P. Tusnady, G.E. Cserzo, M. Simon, I. Prion Protein: Evolution caught en route. (2001) PNAS,98(8), 4431-4436 Download pdf.
V. Vitagliano and G. DÏErrico. A simple kinetic model to describe the progression of prion disease (2001) Phys. Chem. Chem. Phys. 3, 4547-4550. Download pdf.