Protein conformation dictates prion strain


A new conformation-based immunoassay provides a smart and sensitive way of differentiating between strains of prions, the infectious particles believed to cause spongiform encephalopathy (pages 1157–1165 ).


We still do not know whether the transmissible agent causing spongiform encephalopathy—baptized 'prion' by Stanley Prusiner—is identical to PrPSc, the misfolded form of the normal cellular protein PrPC. The protein-only hypothesis of the transmission of spongiform encephalopathies1—which predicates that prions are devoid of nucleic acid and replicate by converting PrPC into a likeness of themselves—can accommodate most experimental and epidemiological findings. However, those still believing that the prion is a virus argue that infectivity comes in various 'flavors', each one stable upon serial propagation. This could be readily explained if the prion contains informational nucleic acid specifying its phenotypic traits. In this issue of Nature Medicine , Safar et al.2 provide significant new information hinting that protein conformation is the basis for the existence of different prion strains. In addition, they present a remarkably sensitive assay for detection of the pathological prion protein, PrPSc.

The problem of different prion strains has long been regarded as a formidable obstacle to the protein-only hypothesis, and remains vexing for those working in the field3,4. Strains are subspecies of the infectious agent capable of maintaining specific phenotypic profiles (such as incubation time of disease, lesion profiles in the central nervous system, and possibly even tropism of the agent for particular extracerebral cell types) when passaged from one experimental animal to another, and in several instances even when passaged between different species.

If one assumes that prions contain nucleic acid, it can be argued that strains are the result of genetic polymorphisms—but no evidence of nucleic acid has been found. On the other hand, if prions consist solely of PrPSc, the phenotypic traits of each strain must in some way be encrypted within the pathological prion protein. Because several distinct prion strains can be maintained in the same strain of inbred mouse, the primary structure of PrP cannot contribute to strain-specific properties. Therefore, it has been suggested that the various strains may be specified by distinct infectious conformers of PrPSc, each of which is capable of imparting its particular three-dimensional imprint onto host PrPC. It is perhaps understandable that the conformation theory has long been regarded by most virologists and protein chemists as a sort of intellectual acrobatics.

A method to tell prion strains apart was developed in Edinburgh and relies on a combination of measuring the time elapsing from inoculation to manifest disease in a panel of a half-dozen inbred mouse strains, and assessing tissue damage in a standardized list of brain subregions. Although many infectious diseases can be defined by PCR, DNA biochips and other marvels of automated molecular diagnostics, it is more than frustrating that strain typing of prions relies on a nineteenth-century technology that, though undoubtedly robust, is impossible to 'scale-up' and has readout times typically in the range of two years or more!

Apart from the formidable intellectual problems it raises—which go to the heart of the protein-only hypothesis—there is a very real reason why physicians should care about prion strain differences. All isolates of the agent of bovine spongiform encephalopathy (BSE, mad cow disease) fulfill the characteristics of a single, highly malignant strain of prion. Brain homogenates from cows with BSE produce a characteristic pattern of brain lesions in mice. This is identical to the pattern elicited by brain tissue from individuals who have died from new-variant Creutzfeldt-Jakob disease5 (nvCJD), which has caused the deaths of 27 young Britons and one Frenchman thus far6. Most disturbingly, there is grave concern that the BSE strain that seems to be transmissible to humans may have infected sheep, where it could produce a disease hardly distinguishable from scrapie (a long-known prion disease of sheep, probably innocuous to humans). But if its ominous strain-specific properties are maintained across the species barrier, sheep BSE may be a threat to human health. On the positive side, there is some evidence that sheep BSE, even if it really exists, may already have disappeared from flocks, given that contaminated meat-and-bone meal from cattle offal is no longer fed to sheep.

All of this calls for a reliable, sensitive and fast method for differentiating prion strains that ideally should be amenable to automation. The first hint that strain specificity may reside within the structure of PrPSc came from the work of the late Richard Marsh, who identified differences in the electrophoretic mobility patterns of PrPSc derived from two strains of mink prions7. Distinct patterns of PrPSc also exist in human Creutzfeldt-Jakob disease8, and John Collinge has shown that these patterns reflect unique combinations of glycoforms that are maintained upon transmission to mice9. Furthermore, he found that nvCJD produces the same glycoform pattern as BSE adding considerable momentum to the hypothesis that nvCJD prions and BSE prions are identical10. Also, Creutzfeldt-Jakob disease and fatal familial insomnia PrPSc give rise to two distinct and transmissible electrophoretic patterns11.

These findings provide strong evidence that molecular weight and glycoform analysis of protease-resistant PrPSc is a valid (and much faster) method for differentiating between prion strains. Although this is a promising approach for distinguishing scrapie from BSE in sheep, it is somewhat risky to rely exclusively on western blot patterns for identifying strains because the physical basis of strain differences is not understood. A third, independent assay would certainly be invaluable because it could be used to cross-check the reliability of the other two methods.

Thanks to the Safar study, a third method may now be available. The underlying idea is original and clever: if the molecular basis of strain diversity resides within the shape of misfolded prions, then their affinity for particular antibodies may be strain-specific. The investigators measured the differential binding of anti-PrP antibodies to native versus denatured prion protein, and found the ratio of the two measurements to be specific for each one of the prion strains analyzed. The format of the assay is that of time-resolved fluorescent ELISA using europium as a marker, which is probably the most sensitive secondary detection system currently available. In order to improve the detection threshold of the assay, Safar and colleagues introduced an initial step to precipitate PrPSc from raw material using phosphotungstate. The final sensitivity is rather impressive, and at present is only surpassed by bioassays in which transgenic indicator animals are inoculated with serial dilutions of infectious material.

This conformational typing methodology represents an important advance for the field. So, is the problem of prion strains resolved for good? Not yet. The goal of the Safar study was neither to clarify the specific PrPSc conformations that give rise to strain-specific traits nor to propose a mechanism by which specific pathological PrPSc conformations maintain their structural characteristics during amplification. Yet eventually both of these issues will need to be elucidated in order to be certain that conformational differences underlie the nature of individual strains. But the work of Safar and co-workers undoubtedly adds a powerful argument to the growing body of evidence that misfolding of PrP is intimately connected with the strain phenomenon.

On a more practical note, the Safar assay provides a third, independent tool to assess the prevalence of certain strains of prion important in human disease. Besides screening sheep with scrapie, it will be of utmost interest for public health to see whether nvCJD and BSE behave concordantly in this system, and whether the molecular imprint of BSE can be identified in those cases of human prion disease that by clinical or histopathological definition would not necessarily fall into the category of nvCJD.


  1. 1

    Prusiner, S.B. Novel proteinaceous infectious particles cause scrapie. Science 216, 136–144 ( 1982).

    CAS  Article  Google Scholar 

  2. 2

    Safar, J. et al. Evidence that eight prion strains possess PrPSc molecules with different conformations. Nature Med. 4, 1157–1165 (1998).

    CAS  Article  Google Scholar 

  3. 3

    Chesebro, B. BSE and prions: uncertainties about the agent. Science 279, 42–43 (1998).

    CAS  Article  Google Scholar 

  4. 4

    Farquhar, C.F., Somerville, R.A. & Bruce, M.E. Straining the prion hypothesis. Nature 391, 345–346 ( 1998).

    CAS  Article  Google Scholar 

  5. 5

    Bruce, M.E. et al. Transmissions to mice indicate that 'new variant' CJD is caused by the BSE agent. Nature 389, 498 –501 (1997).

    CAS  Article  Google Scholar 

  6. 6

    Will, R.G. et al. A new variant of Creutzfeldt-Jakob disease in the UK. Lancet 347, 921–925 ( 1996).

    CAS  Article  Google Scholar 

  7. 7

    Bessen, R.A. & Marsh, R.F. Biochemical and physical properties of the prion protein from two strains of the transmissible mink encephalopathy agent. J. Virol. 66, 2096– 2101 (1992).

    CAS  PubMed  PubMed Central  Google Scholar 

  8. 8

    Parchi, P. et al. Molecular basis of phenotypic variability in sporadic Creutzfeldt-Jakob disease. Ann. Neurol. 39, 767– 778 (1996).

    CAS  Article  Google Scholar 

  9. 9

    Collinge, J., Sidle, K.C., Meads, J., Ironside, J. & Hill, A.F. Molecular analysis of prion strain variation and the aetiology of 'new variant' CJD. Nature 383, 685–690 (1996).

    CAS  Article  Google Scholar 

  10. 10

    Hill, A.F. et al. The same prion strain causes vCJD and BSE. Nature 389, 448–450 ( 1997).

    CAS  Article  Google Scholar 

  11. 11

    Telling, G.C. et al. Evidence for the conformation of the pathologic isoform of the prion protein enciphering and propagating prion diversity. Science 274, 2079–2082 ( 1996).

    CAS  Article  Google Scholar 

Download references

Author information



Rights and permissions

Reprints and Permissions

About this article

Cite this article

Aguzzi, A. Protein conformation dictates prion strain. Nat Med 4, 1125–1126 (1998).

Download citation

Further reading


Quick links

Nature Briefing

Sign up for the Nature Briefing newsletter — what matters in science, free to your inbox daily.

Get the most important science stories of the day, free in your inbox. Sign up for Nature Briefing