Advancing Prion Science: Guidance for the National Prion Research Program: Interim Report (2003)

This chapter addresses diagnostics for transmissible spongiform encephalopathies (TSEs), the fundamental focus of this interim report. In the case of animals, the ability to diagnose or detect an infection drives food safety interventions, which can prevent the introduction of tainted food into the food chain and offset economic damage to the food production industry. In the case of people, the detection of infection can avert the introduction of potentially infectious blood into the blood supply system and can be used to direct appropriate treatment. In addition, the use of diagnostic tests for mass screening has epidemiological utility since it can reveal the extent and distribution of a prion-related disease within a herd or population and can be used to monitor the effects of animal and public health intervention strategies.

Most infectious diseases, such as malaria, tuberculosis, hepatitis, and human immunodeficiency virus, can be diagnosed by conventional methods. This is not the case with TSEs. A prion cannot be identified by direct visualization under a microscope, cultivation in a laboratory, detection of specific antibodies or antigens by standard immunology methods, or detection of its nucleic acid by molecular methods such as polymerase chain reaction (PCR). It consists of host protein with an altered conformation such that the body does not recognize it as foreign and does not produce antibodies against it. It also lacks DNA or RNA, so it cannot be identified by PCR or other nucleic acid-based tests. These factors make detecting the agent very difficult.

TSE agents have other peculiarities that stymie detection. They are insoluble, distributed unevenly in body tissues, and found in a limited

set of tissues by currently available tests. PrP^Sc (the protease-resistant protein associated with prion disease) is neurotropic, so ultimately, it affects cells of the nervous system tissues. However, where and how PrP^Sc progresses through the body prior to its final assault on the nervous system is largely unclear, complicating the ability to locate and detect it.

The similarities between host PrPC (the protease-sensitive cellular protein) and PrP^Sc pose a fundamental problem. Since it is normal to find PrP^C in healthy individuals, detection tests must differentiate between the two proteins. The strategy so far has been to mix the test material with the proteinase K (PK) enzyme, which digests normal prion protein but only a portion of the abnormal protein. Then, various techniques, described below, detect the residual PrP^Sc after digestion. Since this process inadvertently reduces the small amount of original PrP^Sc captured, this approach is inherently less sensitive than methods that do not rely on PK digestion.¹

The fact that only small amounts of the infectious prion may be available for detection in accessible living tissues such as blood, urine, and cerebrospinal fluid (CSF) challenges the diagnostician to develop a sufficiently sensitive test. In addition, diagnostic tests must be of sufficient specificity to differentiate between normal and abnormal prion proteins and, for some purposes, to discriminate between one or more strains of PrP^Sc-a challenge resulting from basic deficiencies in understanding of prion strain diversity and the nature of strain variation.

This then introduces the ultimate objective of a prion detection test: find a single infectious unit while avoiding a falsely positive test result. Reaching this objective will be a complicated task, because a single infectious unit is a variable measure rather than a static one. For example, the size of a single infectious unit of prions injected intracerebrally would be different than the size of a single infectious unit of prions given parenterally, intravenously, or orally. The smallest amount of prions needed to cause an infection may also vary by the strain of prion involved, by the physical composition of aggregated prion molecules, by the type of source tissue, and by the genetic susceptibility of the host animal or person.

The quest for antemortem diagnostics will play a fundamental role in controlling the spread of TSEs, yet current tests are largely

unvalidated and not readily available. Although much effort has been made to improve the means of TSE detection, the ultimate objective is far from being achieved. This chapter describes the clinical and laboratory diagnostic tests in use as well as experimental ones under development. In Chapter 4, the committee recommends new approaches that are more likely to yield effective antemortem diagnostics in the foreseeable future.

In general, diagnoses of prion diseases by clinical description or ancillary clinical tests are not specific enough to confirm a specific prion disease. However, in important circumstances, they give the clinician some clues that may help to support or question the diagnosis of a prion disease.

Differentiation of prion disease from other neurodegenerative diseases and differentiation among different prion strains on clinical grounds are problematic because affected individuals exhibit similar symptoms. Clinical diagnostic criteria have nevertheless been established for sporadic Creutzfeldt-Jakob disease (sCJD) and variant Creutzfeldt-Jakob disease (vCJD) (Will et al., 2000). Some general clinical differences distinguish sCJD from vCJD (Table 3-1). For example, vCJD, contrary to Creutzfeldt-Jakob disease (CJD), occurs in patients generally younger than 40 years old; often presents with early psychiatric and sensory neurological symptoms; and has a longer duration of illness prior to death, usually longer than a year (WHO, 2001a). Spencer and colleagues recently reviewed the early psychiatric manifestations of vCJD (Spencer et al., 2002). They have described the clinical characteristics of the first 100 vCJD patients identified and concluded that "the combination of a psychiatric disorder with affective or psychotic features and persistent pain, dysarthria, gait ataxia, or sensory symptoms should at least raise the suspicion of variant Creutzfeldt-Jakob disease, particularly if this is combined with any suggestion of cognitive impairment" (Spencer et al., 2002, p. 1482). Despite these differences between vCJD and CJD, they are not sufficient to establish a definitive diagnosis.

Ancillary clinical testing typically supplements the medical workup for vCJD or CJD. The most helpful noninvasive tests have been electroencephalography (EEG), neuroimaging, examination of CSF, and more recently, tonsillar biopsy and prion strain identification by immunoblotting. Evaluation of tissue obtained by brain biopsy establishes or excludes the diagnosis of TSE in almost all cases, but

brain biopsy is highly invasive and is limited to cases in which a treatable condition must be excluded.

In typical cases of sCJD (Table 3-2), the electroencephalographs of more than 80 percent of patients show distinctive changes (Parchi et al., 1999). The tracing shows biphasic and triphasic periodic complexes in the clinical course (Parchi et al., 1999). They are evident more than 90 percent of the time with repeated tracings (Chiofalo et al., 1980). These

periodic complexes are observed less frequently in patients with the other subtypes of sCJD and in familial CJD (Gambetti et al., 1999; Parchi et al., 1999) and have never been found in patients with vCJD, although nonspecific slow-wave abnormalities can be seen.

Neuroimaging by computed tomography (CT) and magnetic resonance imaging (MRI) can be useful, especially to rule out non-prion-related neurological diseases. The CT result is usually normal, although in patients with a protracted clinical course the CT scan may show atrophy (WHO, 1998). This finding may be absent and is nonspecific. The MRI scan may also show atrophic changes in patients with late-course disease. When patients are evaluated by T2 MRI, proton-density-weighted MRI, or fluid-attenuated-inversion-recovery MRI, there is an increased signal in the basal ganglia about 80 percent of the time (WHO, 1998). MRI also can be used to help differentiate vCJD from sCJD because the posterior pulvinar region of the thalamus shows a hyperintense signal in patients with vCJD. This pulvinar sign is present in 90 percent of patients with vCJD and is more than 95 percent specific in selected cases, making it the best in vivo test for the diagnosis of vCJD (WHO, 2001b).

CSF contains a protein called 14-3-3 that is normally present in CSF and whose levels are elevated in both patients with sCJD and patients with vCJD, especially those with the typical sCJD subtype (Zerr et al., 2000). This test was 53 percent sensitive when it was used to diagnose vCJD (WHO, 2001a). The 14-3-3 protein has the same electrophoretic pattern in patients with vCJD and sCJD, so it cannot be used to differentiate the two diseases (WHO, 2001a). Other conditions, such as viral encephalitis and recent stroke, can also cause elevations in the levels of this protein (Johnson and Gibbs, 1998; WHO, 1998).

More recently, tonsil biopsy has been used for the presumptive identification of vCJD. Immunohistochemical testing for the prion protein in these tissues has demonstrated that the protein is present in patients with vCJD but not in patients with sCJD (Hill et al., 1999; WHO, 2001a). The postulated reasons for this difference include a strain effect, a species-barrier effect, or the oral route of

exposure in vCJD (Hill et al., 1999). There have been too few case series to determine the sensitivity or specificity of this ancillary test.

Although controversial, tests of tonsil and appendix lymphoid tissues are being used to screen large, asymptomatic populations for TSEs. The largest study to date was conducted in the United Kingdom (Hilton et al., 2002). Between 1995 and 1999, Hilton and colleagues tested 8,318 tonsil and appendix tissue samples from 10- to 50-year-old individuals and found one appendix tissue sample that tested positive for PrP^Sc. From that they estimated the prevalence of vCJD to be 120 per million in the United Kingdom, with a 95 percent confidence limit of 0.5 to 900 cases (Hilton et al., 2002).

It remains unknown how early PrP^Sc accumulates in human tonsils or the human appendix before the onset of symptoms or whether all positive individuals will inevitably progress to the fatal central nervous system (CNS) disease. Nevertheless, studies of sheep naturally infected with the scrapie agent and of mice experimentally infected with it have demonstrated that PrP^Sc is detectable in lymphoid tissues long before clinical signs of neurological disease appear.

In atypical cases of CJD, brain biopsy with histological examination for spongiform changes, immunocytochemical staining, and Western blotting for PrP^Sc, as well as analysis of the PrP gene, is diagnostic in virtually all cases. This approach is seldom needed to diagnose patients with a typical clinical course and consistent findings of classical sCJD by EEG, MRI, and CSF analysis, however. Histological examination of brain tissue should be performed for all patients with possible and probable cases of TSE, as well as for all individuals with questionable neurodegenerative diseases at autopsy, so that a new phenotype of prion disease is not missed.

In both sCJD and vCJD, histology typically reveals the spongiform appearance of the CNS tissues. However, amyloid plaque formations with the characteristic morphology known as florid plaques are seen in all patients with vCJD, whereas kuru plaques (without the characteristics of the florid plaques) are observed only in patients with the sCJD subtype methionine/valine 2 (M/V 2), which accounts for about 10 percent of all cases of sCJD (Johnson and Gibbs, 1998; Parchi et al., 1999).

Additional diagnostic precision has been made possible by the introduction of PrP^Sc isotypes on the basis of the mobility of the PrP^Sc fragment, which is PK-resistant, after gel electrophoresis and Western blotting (Collinge et al., 1996; Monari et al., 1994; Parchi et al., 1997). According to a widely used typing method, there are two major types (or strains) of PrP^Sc in all forms of CJD and fatal insomnia including sCJD, iatrogenic CJD, vCJD, fatal familial insomnia (FFI), and sporadic familial insomnia (sFI) (Parchi et al., 1997).

PrP^Sc type 1 migrates to 21 kilodaltons (kDa) on gels after treatment with PK and deglycosylation; PrP^Sc type 2 migrates to19 kDa under the same conditions. The different gel mobilities of the two PrP^Sc types are due to the different sites of PrP^Sc cleavage by PK, resulting in PK-resistant fragments of different sizes. These two types codistribute with distinct disease phenotypes and are conserved upon transmission to receptive animals. Therefore, they fulfill the definition of prion strains, and this strongly indicates that they have distinct conformations. Additional subtypes of PrP^Sc have been distinguished on the basis of the ratios of the three PrP^Sc glycoforms (Parchi et al., 1997) and on the basis of the profile generated by two-dimensional gel electrophoresis (Pan et al., 2001).

In addition to the PrP^Sc type, the phenotype of human prion diseases is also influenced by the genotype at codon 129 of the PRNP, the site of a common M/V polymorphism. A classification of sporadic prion diseases has been generated on the basis of the combination of the genotype at codon 129 and the PrP^Sc type (Parchi et al., 1996, 1999). This classification includes five subtypes of sCJD and sFI. Each of these subtypes has distinct clinical and pathological features.

Despite these advances, clinical methods remain supportive rather than diagnostic. As in virtually all other disease conditions, the diagnosis is most reliable when it is obtained by combining information from the clinical examination, ancillary clinical tests, and laboratory tests. Yet, even when all this information is combined, the present diagnostic tools lack sufficient sensitivity and specificity. The need to develop tests to improve the early diagnosis of human prion disease and to more reliably detect presymptomatic infections in human and animals is a major priority.

The first method used to confirm the diagnosis of a TSE is postmortem neuropathogical examination of brain tissue from an animal or a human, and this method remains the "gold standard." The World Health Organization position is that "a definitive diagnosis of CJD including nvCJD [new variant CJD] is established only by neuropathological examination" (WHO, 1998, p. 13). Tissue is collected, preserved in formalin, sectioned, stained, and then examined with a light microscope, which is used to look for the characteristic pathological abnormalities on histological examination. This is generally augmented with immunohistochemical staining of the tissue, which uses a PrP antibody-tagged stain that affixes onto PrP. The stain would be abnormally dark or dense in areas where an abnormal amount of PrP was present. Electron microscopy can also be used to observe fibrils, called the scrapie-associated fibrils, in fresh postmortem tissue (Merz et al., 1983) as well as in autolytic tissue.

Three standardized commercial screening tests (by Prionics, Enfer, and Commissariat à l'Energie Atomique [CEA]) have been approved by the European Commission for use in the direct and rapid detection of PrP^Sc (Moynagh and Schimmel, 1999). They were developed in Europe and are primarily used there. To date the U.S. Food and Drug Administration (FDA) has not received a request from any of the European companies that manufacture these tests to approve them for human use in the United States, nor has any company based in the United States submitted any TSE screening test to FDA for approval for human use (Personal communication, D.M. Asher, FDA, July 18, 2002). However, the U.S. Department of Agriculture's Center for Veterinary Biologics has approved the use of one test produced by Bio-Rad Laboratories for the detection of chronic wasting disease (CWD) in mule deer.

The rapid test most widely used to screen for bovine spongiform encephalopathy (BSE) in Europe is a rapid Western blotting test produced by Prionics AG in Switzerland called Prionics Check Western. Test kits are available for the diagnosis of both scrapie in sheep and BSE in cattle. The test uses gel electrophoresis with a

specific antibody against PrP after the PrP^C in homogenized brain tissue is broken down by proteinase K.

The two other tests use slightly different mechanisms for detection. The Enfer test from Ireland is an enzyme-linked immunosorbent assay (ELISA). After digestion with PK, antibody is bound to the residual PrPSc. This material is men bound to a second antibody-enzyme complex that binds to the original PrP^Sc-antibody complex, activating an enzyme reaction that colors the substrate when PrP^Sc is present. The CEA test from France is also an ELISA, but in the initial step after digestion, it uses two different antibodies to bind to different epitopes of PrP^Sc. This is the most sensitive of the three approved tests (Table 3- 3).

The European Commission's laboratory has performed a second round of testing on five more candidate assays, all of which look promising (Schimmel et al., 2002). The results were under review as of November 2002.

The approved rapid tests are generally used to diagnose TSEs in animals after death. Postmortem, the level of accumulation of PrP^Sc has reached its peak and PrP^Sc is most concentrated in brain tissue. Testing for CWD by immunohistochemistry and ELISA of both CNS and peripheral lymphoid tissue samples can provide positive results fairly

early in the incubation period (Sigurdson et al., 1999). In BSE, however, the level of PrP^Sc in those tissues is too low to detect by immunohistochemistry and ELISA until later in the incubation period. Consequently, these approved rapid tests are sufficiently sensitive for the detection of BSE only when they are used to evaluate clinically sick cattle or cattle that are apparently healthy but late in the incubation period. This has given impetus to the development of newer, more sensitive test systems. The following sections review that progress.

Animal bioassays have been used extensively in TSE research and diagnostic testing. Like all tests, animal bioassays have limitations. Two striking limitations are the length of time that it takes to obtain results and the species barrier effect. Since the end-point measurement is neurodegenerative disease and death of the test animal, and since the incubation period from the time of infection to the time of death is measured in months and years, this method is very time-consuming. Yet animal bioassays remain the most sensitive assay available for the detection of prions in potentially infectious material, even though animal bioassays do not directly detect PrP^Sc.

The animals first used to successfully demonstrate infectivity were goats infected with sheep scrapie (Cuille and Chelle, 1939). Goats were used in experiments to study sheep scrapie because the goats became infected more consistently than sheep did (Pattison, 1966). Sheep were also used to demonstrate how resistant the scrapie agent was to formalin inactivation (Pattison and Millson, 1961).

A breakthrough in the pace of TSE research occurred when investigators successfully infected mice with the scrapie agent by intracranial inoculation (Chandler, 1961). Mice incubated the scrapie agent for only 4 or 5 months before clinical signs of scrapie became apparent (Chandler, 1961), many months less than the amount of time required for the appearance of clinical signs in sheep and goats. Later, the successful use of Syrian hamsters reduced the incubation period to illness even further to 70 days (Marsh and Kimberlin, 1975). Further enhancements to the mouse model produced inbred strains that helped elucidate the role of the mouse Prnp gene in susceptibility, incubation times, and prion transmissibility. Understanding of the effect of Prnp on the molecular and biochemical mechanisms of PrP improved with the introduction of mutant, transgenic, and PrP-deficient (knockout) strains of mice. These engineered murine models helped to "define the biochemical and genetic basis of the 'species barrier,' demonstrated the

inverse relationship between the level of PrP^C expression and the incubation time, established the de novo synthesis of prion infectivity from mutant PrP, and revealed the molecular basis of prion strains" (Prusiner et al., 1999, p. 116).

Although mice are the predominant animal model used in bioassays for TSE research, nonhuman primates have been used in the past and continue to have importance. The reason for this is that the species-barrier effect is reduced when the prion being tested is more similar in composition to the host animal's prion protein. Therefore, because the gene that produces PrP in nonhuman primates is more similar to the human PRNP gene than Prnp is, nonhuman primates are excellent candidates for the study of human prion disease and represent a more authentic surrogate than rodents for such study. Yet, the cost and scarcity of nonhuman primates, the complexity of their PrP genotype, and the long incubation period associated with TSE infection in them make their use limited to selected studies. When the use of nonhuman primates is not feasible, transgenic mice that express human PRNP may be the best available assays for the study of human TSE.

No cell culture assay system for the identification of PrP^Sc has been approved. A number of investigators have used in vitro cell culture systems to learn more about the biology of prions. The obvious advantage of a cell culture system would be to significantly shorten the time to detection of an observed end-point effect such as cell death following infection with PrP^Sc. In addition, cell cultures are simpler models with fewer biological interactions than whole-body animal systems. This makes it easier to interpret the molecular and biological effects due to any specific variable being studied. The space and personnel needed to maintain a cell culture system are significantly less than those needed to maintain an animal colony for laboratory studies.

Scientific investigators have successfully used some cell culture systems in prion research. One cell type that has been used rather extensively is the N2a mouse neuroblastoma cell. Both sheep and human prions have been propagated in this cell system after the agent was first passaged through mice (Kingsbury et al., 1984; Race et al., 1987). Other cells reported to have been used in cell culture systems include the GT-1 cell line, which is derived from mouse hypothalamic neurons and which has been used successfully to study the scrapie agent (Schatzl et al., 1997). The PC 12 cell line, which is derived from

rat pheochromocytoma cells, also has been used to study mouse prions (Rubenstein, 1984; Prusiner et al., 1999).

The main shortcomings of existing cell culture systems have been that they do not replicate large amounts of PrP^Sc, the efficiency of infection is low, and the factors that influence susceptibility to infection are poorly understood. These problems diminish the usefulness of cell culture systems for the detection of PrP^Sc.

Various strategies have been adopted to increase the sensitivities of tests used to detect PrP^Sc. These include concentrating PrP^Sc within a given test sample, amplifying the initial amount of prions present in a sample, developing antibody tags that preferentially bind to various conformations of prion protein, electrophoretic separation techniques, and special spectroscopic methods (Table 3-4). In most cases, the test protocols combine many of these strategies.

Physical techniques such as centrifugation and chemical techniques such as those that use sodium phosphotungstate (Na PTA) can concentrate PrP^Sc in a test sample. Safar and colleagues reported that the use of Na PTA resulted in selective precipitation of the oligomers and polymers of PrP^Sc and PrP 27-30, the PK-resistant fragment of PrP^Sc, but not PrP^C (Safar et al., 1998). Other agents, including plasminogen (Fischer et al., 2000), procadherin-2, immobilized metal ion affinity chromatography (IMAC), wheat germ agglutinin, heparin, and various antibodies, have been used to selectively bind to PrP^Sc and thus concentrate the abnormal protein for further characterization (Harris, 2002).

A novel in vitro approach introduced by Saborio, Soto and colleagues involves the cyclic amplification of PrP^Sc by sonication (Saborio et al., 2001; Soto et al., 2002). PrP^Sc in a test sample is incubated with an excess of normal prion protein such that PrP^C converts to PrP^Sc and aggregates into complexes. These complexes are periodically subjected to sonication, which breaks them up and turns them into several new templates for the further conversion of PrP^C to PrP^Sc. This technique is called protein misfolding cyclic amplification, and in the laboratory of Saborio and colleagues, the amount of PrP^Sc in the original sample was found to represent only 3 percent of the

ultimate amount generated. Therefore, the test generated an approximately 30-fold increase in the amount of PrP^Sc.

Another innovative approach has been to identify selective sites on the prion protein that are not visible to an antibody when the protein is in its PrP^Sc conformation, but that are visible to the antibody when the protein is in its native PrP^C configuration or when it is chemically denatured. In general, the less denaturable the protein is, the greater the amount of β-sheet structure present in the protein compared with the amount of α-helical structure present (Safar et al., 1998). By this technique, guanidine hydrochloride was used to denature PrP. When the ELISA feature of this test was used, the test could quantify the amount of PrP^Sc present by determining the ratio of the amount of native PrP to the amount of denatured PrP. Safar and colleagues used this conformation-dependent immunoassay not only to detect PrP^Sc with a notable degree of sensitivity, but also to characterize eight different strains of PrP^Sc. The assay for strain characterization was based on calculation of the ratio of the amount of denatured PrP to the amount of native PrP as a function of the amount of PrP^Sc before and after limited digestion with PK (Safar et al., 1998).

At least one group of investigators has reported on the use of capillary electrophoresis to detect PrP^Sc (Schmerr et al., 1997). This method has been used to analyze other proteins in the past (Tsuji, 1994) and was further adapted to detect PrP^Sc. The technique involves ultracentrifugation and PK digestion of a sample containing PrP followed by resuspension in a sodium dodecyl sulfate-buffered gel. The migration times were calculated by using Ferguson plots, which in turn were used to estimate the molecular weights of the proteins in the test material. The investigators compared the sensitivity of this method to that of the Western blot method and claimed a 100-fold improvement in sensitivity for the detection of PrP^Sc.

Another approach to improve the sensitivity and specificity of TSE diagnostics is to use newer, more advanced biotechnology tools such as fluorescent correlation spectroscopy (FCS). One group of investigators tagged PrP-specific antibodies with florescent dyes designed to bind to

any PrP complexes within CSF (Bieschke et al., 2000). They measured the bound complexes using FCS, which was further modified by using a dual-colored florescence intensity distribution analysis system and confocal microscopy with a scanner. This method incorporates a technology involving the scanning of intensely florescent targets, which improves both the sensitivity and the specificity of a test. The sensitivity alone is 20 times better than that of the Western blotting test (Bieschke et al., 2000).

Other spectroscopic devices and techniques have been developed to improve PrP detection. An example is multispectral ultraviolet fluorescence spectroscopy (MUFS) (Rubenstein et al., 1998). This technique excites a test sample by exposing it to monochromatic light at specific wavelengths. The resulting ultraviolet fluorescence from that exposure is then captured and plotted. Rubenstein and colleagues successfully applied this method to PrP. They showed that PK-treated hamster brain had spectral signatures different from those of untreated hamster brain. They also demonstrated that the spectral signals from PK-treated PrP^Sc proteins of two different species, the mouse and the hamster, were sufficiently intense and distinctive that the two proteins could be differentiated by least-squares analysis, which quantifies the orthogonal difference in the signals. They Concluded that MUFS has great promise as a rapid, sensitive, and specific tool for the direct detection of PrP^Sc as well as for the differentiation of disparate prion strains (Rubenstein et al., 1998).

A recently reported spectroscopic approach to the identification of prion-infected hosts used Fourier transform infrared spectroscopy, in combination with a highly sophisticated automated computer-assisted pattern recognition program referred to as artificial neural networks, to detect disease-associated differences in patterns of small molecules in serum (Schmitt et al., 2002). By this method the investigators correctly differentiated between blood taken from Syrian hamsters with terminal infections and blood from healthy control hamsters. They reported a sensitivity of 97 percent and a specificity of 100 percent. The predictive value was 100 percent for a positive test result and 98 percent for a negative test result. The investigators indicated that the test needed to be assessed with species other than hamsters, and they cautioned that

the differences observed between the scrapie-infected animals and the controls may not be specific for detection of the scrapie agent. It is noteworthy that Fourier transform infrared spectroscopy does not involve PK digestion.

Despite recent improvements in the sensitivities of diagnostic tests for TSE, they are not sensitive enough for antemortem screening of asymptomatic animals and humans, nor are they adequately specific. False-negative and false-positive results still occur too frequently.

False-positive tests for the detection of TSE in human populations would result in individuals being erroneously informed that they have an incurable, fatal disease.

The impact of a false-positive test result used on livestock in countries reporting BSE would be the disposal of perfectly good meat. However, in countries where a false-positive test result represents a sentinel BSE case, the economic, political, and societal consequences of that incorrect result would be monumental. The impact of a false-negative test result might allow a contaminated beef product to enter the food chain. A false-negative result by a test used to diagnose scrapie or CWD not only might allow an animal to escape detection but also might allow horizontal transmission of the infectious agent.

Even if they were adequate, many of these newer tests for TSE are not available for general diagnostic use or for screening. They are being used exclusively in research laboratories. Their utility for commercial applications still requires validation and scaling for high-throughput testing.

The larger issue here is that investigators have focused on relatively few strategies for prion detection. They have relied heavily on PK digestion of PrP^C, on a small number of antibodies, and on a few model systems. The result of this narrow focus is today's limited set of experimental approaches and reagents. Circumstances beg for fresh ideas that leverage a broader array of new technologies.

Bieschke J, Giese A, Schulz-Schaeffer W, Zerr I, Poser S, Eigen M, Kretzschmar H. 2000. Ultrasensitive detection of pathological prion protein aggregates by dual-color scanning for intensely fluorescent targets. Proceedings of the National Academy of Sciences of the United States of America 97(10):5468-5473.

Chandler RL. 1961. Encephalopathy in mice produced by inoculation with scrapie brain material. Lancet i:1378-1379.

Chiofalo N, Fuentes A, Galvez S. 1980. Serial EEG findings in 27 cases of Creutzfeldt-Jakob disease. Archives of Neurology 37(3):143-145.

Collinge J, Sidle KC, Meads J, Ironside J, Hill AF. 1996. Molecular analysis of prion strain variation and the aetiology of 'new variant' CJD. Nature 383(6602):685-690.

Cuille J, Chelle PL. 1939. Experimental transmission of trembling to the goat [translation from French]. Comptes Rendus de l'Académie des Sciences 208:1058-1060.

Fischer MB, Roeckl C, Parizek P, Schwarz HP, Aguzzi A. 2000. Binding of disease-associated prion protein to plasminogen. Nature 408(6811):479-483.

Gambetti P, Petersen RB, Parchi P, Chen SG, Capellari S, Goldfarb L, Gabizon R, Montagna P, Lugaresi E, Piccardo P, Bernardino G. 1999. Inherited prion diseases. Prusiner S, Editor. Prion Biology and Diseases. Cold Spring Harbor, NY: Cold Spring Harbor Laboratory Press. Pp. 509-583.

Harris D. 2002. Conventional methods for PrPSc detection. Presentation to the IOM Committee on Transmissible Spongiform Encephalopathies: Assessment of Relevant Science, Meeting II. The National Academies, Washington, D.C.

Hill AF, Butterworth RJ, Joiner S, Jackson G, Rossor MN, Thomas DJ, Frosh A, Tolley N, Bell JE, Spencer M, King A, Al-Sarraj S, Ironside JW, Lantos PL, Collinge J. 1999. Investigation of variant Creutzfeldt-Jakob disease and other human prion diseases with tonsil biopsy samples. Lancet 353(9148):183-189.

Hilton DA, Ghani AC, Conyers L, Edwards P, McCardle L, Penney M, Ritchie D, Ironside JW. 2002. Accumulation of prion protein in tonsil and appendix: review of tissue samples. BMJ 325(7365):633-634.

Ingrosso L, Vetrugno V, Cardone F, Pocchiari M. 2002. Molecular diagnostics of transmissible spongiform encephalopathies. Trends in Molecular Medicine 8(6):273- 280.

Johnson RT, Gibbs CJ Jr. 1998. Creutzfeldt-Jakob disease and related transmissible spongiform encephalopathies. New England Journal of Medicine 339(27):1994- 2004.

Kingsbury DT, Smeltzer D, Bockman J. 1984 Purification and properties of the K. Fu. isolate of the agent of Creutzfeldt-Jakob disease. In Abstracts from the Proceedings of the 6th International Congress of Virology, Sendai, Japan, abstr. W47-6, p. 70.

Marsh RF, Kimberlin RH. 1975. Comparison of scrapie and transmissible mink encephalopathy in hamsters. II. Clinical signs, pathology, and pathogenesis. Journal of Infectious Diseases 131(2):104-110.

Merz PA, Somerville RA, Wisniewski HM, Manuelidis L, Manuelidis EE. 1983. Scrapie-associated fibrils in Creutzfeldt-Jakob disease. Nature 306(5942):474-476.

Monari L, Chen SG, Brown P, Parchi P, Petersen RB, Mikol J, Gray F, Cortelli P, Montagna P, Ghetti B, Goldfarb LG, Gajdusek DC, Lugaresi E, Gambetti P, Autilio-Gambetti L. 1994. Fatal familial insomnia and familial Creutzfeldt-Jakob disease: different prion proteins determined by a DNA polymorphism. Proceedings of the National Academy of Sciences of the United States of America 91(7):2839-2842.

Moynagh J, Schimmel H. 1999. Tests for BSE evaluated. Bovine spongiform encephalopathy. Nature 400(6740):105.

Pan T, Colucci M, Wong BS, Li R, Liu T, Petersen RB, Chen S, Gambetti P, Sy MS. 2001. Novel differences between two human prion strains revealed by two-dimensional gel electrophoresis. Journal of Biological Chemistry 276(40):37284- 37288.

Parchi P, Castellani R, Capellari S, Ghetti B, Young K, Chen SG, Farlow M, Dickson DW, Sima AA, Trojanowski JQ, Petersen RB, Gambetti P. 1996. Molecular basis

of phenotypic variability in sporadic Creutzfeldt-Jakob disease. Annals of Neurology 39(6):767-768.

Parchi P, Capellari S, Chen SG, Petersen RB, Gambetti P, Kopp N, Brown P, Kitamoto T, Tateishi J, Giese A, Kretzschmar H. 1997. Typing prion isoforms. Nature 386(6622):232-234.

Parchi P, Giese A, Capellari S, Brown P, Schulz-Schaeffer W, Windl O, Zerr I, Budka H, Kopp N, Piccardo P, Poser S, Rojiani A, Streichemberger N, Julien J, Vital C, Ghetti B, Gambetti P, Kretzschmar H. 1999. Classification of sporadic Creutzfeldt-Jakob disease based on molecular and phenotypic analysis of 300 subjects. Annals of Neurology 46(2):224-233.

Pattison IH. 1966. The relative susceptibility of sheep, goats and mice to two types of the goat scrapie agent. Research in Veterinary Science 7(2):207-212.

Pattison IH, Millson GC. 1961. Scrapie produced experimentally in goats with special reference to the clinical syndrome. Journal of Comparative Pathology 71:101-108.

Prusiner SB, Tremblay P, Safar J, Torchia M, DeArmond SJ. 1999. Bioassays of prions. Prusiner SB, Editor. Prion Biology and Diseases. Cold Spring Harbor, NY: Cold Spring Harbor Laboratory Press. Pp. 113-145.

Race RE, Fadness LH, Chesebro B. 1987. Characterization of scrapie infection in mouse neuroblastoma cells. Journal of General Virology 68 (Pt 5):1391-1399.

Rubenstein R, Gray PC, Wehlburg CM, Wagner JS, Tisone GC. 1998. Detection and discrimination of PrPSc by multi-spectral ultraviolet fluorescence. Biochemical and Biophysical Research Communications 246(1):100-106.

Saborio GP, Permanne B, Soto C. 2001. Sensitive detection of pathological prion protein by cyclic amplification of protein misfolding. Nature 411(6839):810-813.

Safar J, Wille H, Itri V, Groth D, Serban H, Torchia M, Cohen FE, Prusiner SB. 1998. Eight prion strains have PrP(Sc) molecules with different conformations. Nature Medicine 4(10):1157-1165.

Schatzl HM, Laszlo L, Holtzman DM, Tatzelt J, DeArmond SJ, Weiner RI, Mobley WC, Prusiner SB. 1997. A hypothalamic neuronal cell line persistently infected with scrapie prions exhibits apoptosis. Journal of Virology 71(11):8821-8831.

Schimmel H, Catalani P, Le Guern L, Prokisch J, Philipp W, Trapmann S, Zeleny R, Moynagh J. 2002. The Evaluation of Five Rapid Tests for the Diagnosis of Transmissible Spongiform Encephalopathy in Bovines (2nd Study). Geel, Belgium: European Commission.

Schmerr MJ, Jenny A, Cutlip RC. 1997. Use of capillary sodium dodecyl sulfate gel electrophoresis to detect the prion protein extracted from scrapie-infected sheep. Journal of Chromatography B: Biomedical Science Applications 697(1-2):223-229.

Schmitt J, Beekes M, Brauer A, Udelhoven T, Lasch P, Naumann D. 2002. Identification of scrapie infection from blood serum by Fourier transform infrared spectroscopy. Analytical Chemistry 74(15):3865-3868.

Sigurdson CJ, Williams ES, Miller MW, Spraker TR, O'Rourke KI, Hoover EA. 1999. Oral transmission and early lymphoid tropism of chronic wasting disease PrP^res in mule deer fawns (Odocoileus hemionus). Journal of General Virology 80 (Pt 10):2757-2764.

Soto C, Saborio GP, Anderes L. 2002. Cyclic amplification of protein misfolding: application to prion-related disorders and beyond. Trends in Neurosciences 25 (8):390- 394.

Spencer MD, Knight RS, Will RG. 2002. First hundred cases of variant Creutzfeldt-Jakob disease: retrospective case note review of early psychiatric and neurological features. BMJ 324(7352):1479-1482.

Tsuji K. 1994. Sodium dodecyl sulfate polyacrylamide gel- and replaceable polymer-filled capillary electrophoresis for molecular mass determination of proteins of pharmaceutical interest. Journal of Chromatography B: Biomedical Science Applications 662(2):291-299.

WHO (World Health Organization). 1998. Global Surveillance, Diagnosis and Therapy of Human Transmissible Spongiform Encephalopathies: Report of a WHO Consultation. Geneva: World Health Organization.

WHO. 2001a. The Revision of the Surveillance Case Definition for Variant Creutzfeldt-Jakob Disease (VCJD): Report of a WHO Consultation. Edinburgh, United Kingdom: World Health Organization.

WHO. 2001b. Working Group on International Reference Materials for Diagnosis and Study of Transmissible Spongiform Encephalopathies (TSEs): Third Meeting. WHO Blood Safety and Clinical Technology. Geneva: World Health Organization.

Will RG, Zeidler M, Stewart GE, Macleod MA, Ironside JW, Cousens SN, Mackenzie J, Estibeiro K, Green AJ, Knight RS. 2000. Diagnosis of new variant Creutzfeldt-Jakob disease. Annals of Neurology 47(5):575-582.

Zerr I, Pocchiari M, Collins S, Brandel JP, de Pedro Cuesta J, Knight RS, Bernheimer H, Cardone F, Delasnerie-Laupretre N, Cuadrado Corrales N, Ladogana A, Bodemer M, Fletcher A, Awan T, Ruiz Bremon A, Budka H, Laplanche JL, Will RG, Poser S. 2000. Analysis of EEG and CSF 14-3-3 proteins as aids to the diagnosis of Creutzfeldt-Jakob disease. Neurology 55(6):811-815.