Environ. Sci. Technol. 2006, 40, 1497-1503
Fate of Prions in Soil: Adsorption and Extraction by Electroelution of Recombinant Ovine Prion Protein from Montmorillonite and Natural Soils P E G G Y R I G O U , * ,† H U M A N R E Z A E I , † JEANNE GROSCLAUDE,† S I O B H AÄ N S T A U N T O N , ‡ A N D H E R V EÄ Q U I Q U A M P O I X ‡ Virologie et Immunologie Mole´culaires, INRA, F-78352 Jouy-en-Josas, France, and UMR Rhizosphe`re et Symbiose, INRA-ENSAM, 2 Place Pierre Viala, 34060 Montpellier Cedex 01, France
Prions, the infectious agents thought to be responsible for transmissible spongiform encephalopathies, may contaminate soils and have been reported to persist there for years. We have studied the adsorption and desorption of a model recombinant prion protein on montmorillonite and natural soil samples in order to elucidate mechanisms of prion retention in soils. Clay minerals, such as montmorillonite, are known to be strong adsorbents for organic molecules, including proteins. Montmorillonite was found to have a large and selective adsorption capacity for both the normal and the aggregated prion protein. Adsorption occurred mainly via the N-terminal domain of the protein. Incubation with standard buffers and detergents did not desorb the full length protein from montmorillonite, emphasizing the largely irreversible trapping of prion protein by this soil constituent. An original electroelution method was developed to extract prion protein from both montmorillonite and natural soil samples, allowing quantification when coupled with rapid prion detection tests. This easy-to-perform method produced concentrated prion protein extracts and allowed detection of protein at levels as low as 0.2 ppb in natural soils.
Introduction The spread of transmissible spongiform encephalopathies (TSE) is a matter of considerable environmental concern. Various hypotheses exist on the nature of the infectious agent or “prion” responsible for TSE, but the most widely accepted supposed agent is the “protein only” (the proteinaceous infectious particle that is a modified form of the host-encoded prion protein). The degradation of prions in the environment is poorly understood, and there is no direct evidence of the precise role of soil in the persistence or the transmission of infectivity. It has been reported that healthy animals bred in places previously frequented by infected animals developed the disease, but the source of infection was not determined, although it is generally suspected to be related to grazing and direct contact with infected biological material (1-5). * Corresponding author phone: +33 134 65 26 28; fax: +33 134 65 26 21; e-mail:
[email protected]. † INRA. ‡ INRA-ENSAM. 10.1021/es0516965 CCC: $33.50 Published on Web 02/01/2006
2006 American Chemical Society
Soil has been shown to retain the infectious agent for years, in a form experimentally transmissible to laboratory animals, although there was very little dissemination. However the niche that allowed this persistence was not identified (6). Under natural conditions, the most likely way that the infectious agent could enter the environment is through the soil from the decay of infected animal carcasses (with the accumulation of prion in nervous system and lymphoid tissues through the disease), excreta from infected animals, or infected placenta remaining on the ground after whelping (1, 7, 8). Agricultural and industrial practices, the uncontrolled incineration of scrapie-contaminated tissues, may contribute to prion dissemination in the environment (2, 9). Although there are established standard conditions for safe handling, transportation, and storage of infected meat and bone meal (10), accidental spillage during transportation or inappropriate storage may occur, as well as the spreading of effluents of slaughterhouses, rendering plants, and the gelatin industry. The US Environmental Protection Agency (EPA) stated that at present there has been little evidence of prion-contaminated manure, but that the risk of prion transmission from animals to biosolids can increase with the presence of small amounts of neural tissues or placenta from slaughterhouses (11). Proteins are macromolecules with large affinities for water-solid interfaces, since they are flexible polymeric chains with lateral groups having contrasting physicochemical properties: hydrophilic or hydrophobic, negatively, neutrally, or positively charged lateral chains. As a result, proteins have strong and complex interactions with the extensive mineral and organomineral surfaces in soil (1216). The complexity of the various types of interactions involved makes the prediction of the interaction of a given protein with the different components of soil surfaces extremely difficult. For example, the β-glucosidase of Aspergillus niger does not adsorb on the surface of montmorillonite above pH 6 (17), nor does bovine serum albumin above pH 6.5 (18). Acid phosphatases of ectomycorrhizal fungi have particularly diversified behavior in the presence of soil clays, from total adsorption, observed for Suillus mediterraneensis, to total repulsion, for Pisolithus tinctorius (19). The interaction of prion protein with soil fractions has been currently investigated, as reported at recent conferences (20). On other solid surfaces, the prion protein binds exceptionally strongly to plastic or metal surfaces encountered in medical and chirurgical devices (21-24), which has led to specific provisions for prion decontamination (25). The prion protein is thus expected to display a unique behavior on soil organic or mineral solid surfaces. The key event in the pathogenesis due to the prion is the conversion of the R-helix-rich host prion protein (PrPC or PrPsens) into a pathogenic isoform (PrPSc or PrPres) characterized by its insolubility, its higher β-sheet content, and its protease resistance (PrPres) (26). Adsorption of the prion protein onto solid components might explain trapping and the persistence of infectivity in soil. Detection of prion protein bound to soil mineral surfaces is a prerequisite to the monitoring and control of environmental contamination. Direct immunochemical detection of proteins adsorbed to mineral or organic material is impeded by the strong nonspecific adsorption of the antibodies to the matrix (27). No EPA-established standard method exists for the extraction and detection of the prion pathogen in soil (28). We believe that a better understanding of the adsorption and desorption of the prion protein on specified solid components such as clay is required in order to develop VOL. 40, NO. 5, 2006 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
1497
TABLE 1. Characteristics of Reagents Investigated for Possible Desorption of RecPrP Adsorbed onto Montmorillonite reagent names
SDS (2%)
Laemmli buffera 4×
urea (8 M)
reagent type
detergent
reductor detergent
chaotropic solution buffer solution
reagent properties denatures the reduces S-S bridges denatures the protein protein a
PBS (pH 7.3)
FCS (10%) mixture of proteins
changes the ionic possible strength of the competition for protein environment adsorption sites
Tris, SDS, and DTT.
reliable methods for the extraction of trapped prion protein and the detection of contamination. We first investigated the interaction of a model prion protein (recombinant purified prion protein, recPrP) in a simple system, namely montmorillonite, a model clay mineral, revealing a unique and selective trapping capacity for prion protein. This adsorption was quasi-irreversible, and other proteins did not compete efficiently. We further developed an original technique of electroelution that allowed simultaneous extraction, concentration, and quantitative detection of recPrP in montmorillonite as well as natural soil samples.
Experimental Procedures 1. Model Protein. The recombinant full-length ovine prion protein (recPrP) was used as a model (29). NMR experiments indicate that the three-dimensional structure and stability of the recombinant bovine prion protein expressed in Escherichia coli and the cellular form of the bovine prion protein isolated from healthy calf brains are essentially identical (30). Full-length recPrP is structured in two domains: (i) the N-terminal part bears the octapeptides repeat that can bind transition metals and presents a number of positively charged side chains of amino acid residues and (ii) the C-terminal part is negatively charged and plays a major role in the conversion from the R- to the β-form. The C-terminal fragment used in this study was produced following the protocol described by Eghiaian et al. (31). Conversion of R-recPrP to β-sheeted oligomers can be induced by heat, at an acidic pH, to obtain similar structural properties to the pathological protein (32-34). Throughout this study, monomeric recPrP was a substitute for normal PrP and β-sheeted oligomer for the pathological protein. 2. Clay Mineral Montmorillonite. The clay mineral montmorillonite (Mt) is a common constituent of the mineral phase of soils with a very large adsorption capacity due to its large negatively charged surface area. In aqueous suspension the mineral occurs as submicron-sized particles. An aqueous colloidal suspension (20 g/L) of the clay-sized fraction (
