Prion amyloids, the infectious protein agent of mad cow disease (bovine spongiform encephalopathy, or BSE), and other transmissible spongiform encephalopathies, are protein aggregates with a flair for self-promotion. These abnormally folded, spontaneously propagating polymers enter cells and coax soluble cellular prion protein (PrP) to adopt their toxic conformation. The result is a widespread amyloidosis and neurodegenerative disease that leaves the brain with a characteristic sponge-like pathology. Two papers, published this week in Cell, describe how researchers used new techniques for propagating PrP fibers in vitro to illuminate the role of both amino acid sequence and protein conformation in prion infectivity. The results not only explain how infectious prions can break the species barrier, but also raise questions about the structure of non-infectious amyloids, as found in Alzheimer disease.

The likelihood that prion protein from one species will successfully infect another, as happened with the bovine-to-human transmission of BSE, depends on the amino acid similarities between each species’ PrP—the closer the match, the more likely that the infectious prion can whip the host protein into pathological shape. But amino acid sequence does not tell the whole story, because prions with identical amino acid sequences can adopt different conformations, and these different-shaped prion strains may coexist in a single host species. So what’s more important, primary sequence or secondary structure?

Both studies, one from Jonathan Weissman and colleagues at the University of California in San Francisco, and the other from Eric Jones and Witold Surewicz at Case Western Reserve University, Cleveland, Ohio, show that proteins with an identical primary amino acid sequence can be induced to form distinctly different fiber conformations. But in each case, it is the shape of the protein, not its amino acid sequence per se, that dictates whether it will jump the species barrier and catalyze fibril formation in a different host.

Jones and Surewicz’s work extends a study they published last year (Vanik at al., 2004) showing that soluble recombinant fragments of human, mouse, or hamster PrP are easily cajoled into forming amyloid fibrils if they are seeded with preformed amyloid fibrils in vitro. Using this assay, the researchers established species barriers for the seeding reaction; mouse prion could seed human and vice versa, while a hamster prion induced polymerization of mouse protein, but not human.

In the new paper, Jones and Surewicz describe the physical characteristics of these fibrils. Spectroscopic and microscopic techniques show that human and mouse PrP fibrils take on a similar string-of-beads shape, while hamster PrP yields a different conformation—smooth fibers. But their most intriguing results come from the cross-seeding experiments. When they used a small amount of hamster amyloid to seed mouse prion, the resulting fibrils, even though made almost entirely of mouse protein, had the appearance and physical properties of hamster fibrils. In addition, the new hamster-type mouse PrP fibrils had lost the ability to seed human PrP polymerization, proving that by cross-species seeding, the investigators had generated a new strain of mouse prion that inherited the secondary structure of its hamster template.

The species barrier-busting ability of the new mouse strain remains theoretical, since none of the recombinant prion fibers has been proven to be infectious in animals yet. But in the second paper, first author Motomasa Tanaka and the UCSF group get around that limitation by using yeast. In yeast, a transmissible, prion-like conformation of the protein Sup35 suppresses translation. As with mammalian prions, a strong species barrier normally prevents transmission from one strain of yeast to another. For example, Sup35 from Candida albicans (Ca) can only infect C. albicans and not Saccharomyces cerevisiae (Sc) and vice versa.

Is this barrier also based on secondary structure? Apparently so. The researchers showed that a truncated recombinant Sup35 from Sc formed different fibril conformations at different temperatures. One conformation in particular, formed at 4 degrees C, displayed the unusual ability to seed the polymerization of Ca Sup35. The fibrils generated from this seeding could themselves seed both Sc and Ca monomers, and could infect both strains of yeast Sup35. The species cross was all the more dramatic given that the Ca and Sc Sup35 proteins are fairly divergent—unlike the closely related mammalian PrPs used by Jones and colleagues, the Sup35s share only 40 percent amino acid similarity in the prion domain. Biophysical measurements on the fibrils supported the idea that attaining a permissive three-dimensional shape was the important determinant of successful infection.

Both studies, then, point to a model where the amino acid sequence of a PrP proscribes a set of possible conformations, and a species barrier can be breached when the sets of possible shapes for two PrP sequences overlap. This shape-shifting mechanism for new strain production provides a plausible explanation for the emergence of variant Cruetzfeldt-Jakob disease after infection of humans with the BSE prion. Moreover, the specter of the production of new prion strains in nature by repeated cross-species infection should raise a flag as prion diseases keep popping up among both domestic and wild animals.

These results may also have application to other amyloid disease, even where the protein aggregates are non-infectious, as in AD. In these cases, the California group speculates, different amyloid conformations adopted by the pathogenic protein could determine its ability to recruit other, heterologous proteins, possibly modulating the toxic effects of accumulating misfolded protein aggregates.—Pat McCaffrey

Pat McCaffrey is a freelance science writer in Newton, Massachusetts.


  1. Why are prion proteins able to propagate in one species and not another? What happens during the rare events when strains of prion proteins are able to breach the species barrier? These critical questions, elegantly addressed by the recent papers from Surewicz and Weissman, may have structural answers with parallels to Alzheimer disease (AD).

    One vexing problem in the field of prion biology is that strain-dependent differences at the amino acid level do not fully account for transmission specificity. The hypothesis that prion propagation is the result of specific conformational states influenced by, but not dependent on, primary amino acid sequence was tested in two separate sets of in vitro experiments. These groups characterized specific biophysical and structural properties of different strains of prion protein and related them to transmission specificity. Taken together, the results suggest that cross-species transmission is actually a propagation of protein conformation occurring between proteins that differ in primary sequence. Sequence may ultimately determine the range of conformations prion proteins can adopt, but the ability of the prion to propagate resides in the protein fold.

    Although an equivalent infectious mechanism has never been established for AD, several parallels exist at the protein level between amyloid-β and prion proteins. Both proteins can undergo a transition from a soluble monomeric state to a polymerized amyloid fibril via a nucleation-dependent polymerization reaction. In addition, both proteins exhibit conformation-specific biological activities.

    To determine how structural changes in prion proteins relate to transmission specificity, changes in protein structure were monitored using several complementary biophysical techniques by both groups and related to cross-species transmission. Some of the most striking evidence presented by Jones and Surewicz for the transmission of protein conformation occurring between different strains of prion proteins came from analysis of the fine morphological detail of the amyloid fibrils by atomic force microscopy (AFM). Changes in transmission specificity could be directly correlated with distinct fibril morphologies. It is tempting to speculate that a similar correlation exists for amyloid-β fibrils, also shown to exhibit similar smooth, beaded, and helical morphologies by AFM (Stine et al., 1996). Amyloid-β conformation has been shown to be strongly dependent on primary sequence and solution conditions (Stine et al., 2003), and these different protein conformations have distinct biological activities (Dahlgren et al., 2002). Changes in amyloid-β primary sequence associated with familial AD also influence both amyloid fibril assembly and fibril toxicity compared to wild-type fibrils (Dahlgren et al., 2002), paralleling the structural and functional differences observed with the “species-mimicking” point mutations in PrP23-144.

    Changes in prion fibril morphology parallel changes in secondary structure as measured by Fourier transform infrared (FTIR) spectroscopy by Jones and Surewicz. Very similar conclusions are made by the Weissman group working with different strains of prion protein. Their results also relate prion transmission specificity to structural changes measured by FTIR, thermostability, and fine conformational detail by site-directed spin labeling-electron paramagnetic resonance (SDSL-EPR). (Although results from negative stain electron microscopy (EM) hint at a correlation between fibril morphology and transmission specificity). Both of these studies benefit from a robust functional measure for prion activity. Unfortunately for amyloid-β, a similar direct functional system currently does not exist and most assays depend on biological endpoints such as behavior testing, neurotoxicity, glial inflammation, or changes in LTP.

    Identical protein folds can result from polypeptide chains that differ at the amino acid level and, conversely, identical primary sequences can assemble into distinct secondary, tertiary, and quaternary structures. The suggestion that for prion proteins (and possibly other amyloidogenic proteins), primary sequence only serves to provide boundaries for a pool of possible conformational states provides new insight into the perplexing question of amyloid structure-function. Strain-dependent prion propagation and cross-species prion transmission appear to have a structural foundation: the transmission of an infectious protein conformation.


    . The nanometer-scale structure of amyloid-beta visualized by atomic force microscopy. J Protein Chem. 1996 Feb;15(2):193-203. PubMed.

    . In vitro characterization of conditions for amyloid-beta peptide oligomerization and fibrillogenesis. J Biol Chem. 2003 Mar 28;278(13):11612-22. PubMed.

    . Oligomeric and fibrillar species of amyloid-beta peptides differentially affect neuronal viability. J Biol Chem. 2002 Aug 30;277(35):32046-53. PubMed.

Make a Comment

To make a comment you must login or register.


Paper Citations

  1. . Molecular basis of barriers for interspecies transmissibility of mammalian prions. Mol Cell. 2004 Apr 9;14(1):139-45. PubMed.

Further Reading


  1. . Mechanism of prion propagation: amyloid growth occurs by monomer addition. PLoS Biol. 2004 Oct;2(10):e321. PubMed.
  2. . Conformational variations in an infectious protein determine prion strain differences. Nature. 2004 Mar 18;428(6980):323-8. PubMed.

Primary Papers

  1. . Fibril conformation as the basis of species- and strain-dependent seeding specificity of mammalian prion amyloids. Cell. 2005 Apr 8;121(1):63-72. PubMed.
  2. . Mechanism of cross-species prion transmission: an infectious conformation compatible with two highly divergent yeast prion proteins. Cell. 2005 Apr 8;121(1):49-62. PubMed.