Cellular Prion Proteins of Mammalian Species Display an Intrinsic Partial Proteinase K Resistance

BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS ARTICLE NO.

253, 693–702 (1998)

RC989838

Cellular Prion Proteins of Mammalian Species Display an Intrinsic Partial Proteinase K Resistance A. Buschmann,* T. Kuczius,* W. Bodemer,† and M. H. Groschup*,1 *Federal Research Centre for Virus Diseases of Animals, P.O. Box 1149, 72001 Tu¨bingen, Germany; and †Department of Neurology, Universita¨tsklinikum Go¨ttingen, 37077 Go¨ttingen, Germany

Received November 6, 1998

Prion diseases are characterized by the intraneuronal accumulation of a pathological isoform (PrPSc) of host-encoded prion protein (PrPC). While PrPSc displays a partial resistance, PrPC is easily degraded by this enzyme. As it turned out in our experiments, PrPC of six species is initially degraded to an intermediate fragment of 25-28 kDa prior to complete proteolysis which was solely detected by antibodies binding to epitopes carboxy-terminally of amino acid 144 of PrPC. The intermediate fragment thus lacked the aminoterminus of PrPC. These findinds are well in line with the putative structure of PrPC: the amino-terminus consists of a highly flexible and thus more proteinase K sensitive tail while the carboxy-terminus is folded into possibly more resistant a-helices and b-sheets. We observed significant differences in the PK sensitivities of PrPC from six different species and from three ovine PrP alleles, while no remarkable variation was seen in PrPC from six regions of an ovine brain. This indicates that variations in the sequence of PrP may alter its three-dimensional structure and consequently change its sensitivity towards proteolytic enzymes. © 1998 Academic Press

In transmissible spongiform encephalopathies syn. prion diseases, the conversion of prion protein (PrPC) into its conformational isoform (PrPSc) seems to be a crucial event during pathogenesis and generation of the infectious agent (1). Hence, the detection of PrPSc accumulation in neuronal and other tissues of diseased individuals is used as a reliable diagnostic marker for prion diseases (2). Two physicochemical properties are essential for the definition of PrPSc in comparison to PrPC in all systems: a) PrPSc exhibits a decreased solubility in detergent solution and b) a core fragment of PrPSc displays an unusually high resistance to proteinase K digestion. The decreased solubility can be 1 To whom correspondence should be addressed. Fax: 49-7071-967303. E-mail: [email protected].

revealed electronmicroscopically by detecting so-called scrapie-associated fibrils or prion rods (3–5) or physicochemically by hydrophobic interaction chromatography (6). The partial proteinase K resistance, i.e. cleavage of 62 N-terminal amino acids which leaves behind the proteinase K resistant core fragment of approx. 141 amino acids, can be demonstrated in immunoblot by a shift in the molecular mass of PrPSc from 33-35 kDa to 27-30 kDa. In contrast, PrPC is completely proteolyzed under conditions that do not lead to hydrolysis of the PrPSc core fragment of all known prion strains (7, 8). Proteinase K is a serine protease with the molecular mass of 27 kDa from the mould Tritirachium album Limber, of which the three-dimensional structure as well as the mode of action are well established (9 –11). It acts an endopeptidase, which unlike the digestive mammalian proteolytic enzymes, exhibits a marked unselectivity towards peptide substrates. It splits peptide bonds of amino acids preferably at hydrophobic side chains. As revealed recently, the amino acid backbone of PrPC is composed of two b-sheets (amino acids 127-130, 160-163) and three a-helices (amino acids 143-153, 178-192, 199-216) which are interconnected by loops (12,13). In contrast, the amino-terminus of the protein (22–121) forms a long and highly flexible tail which might assist in PrP conversion by template assisted formation of b-sheets (14). It has also been proposed that PrPSc is composed of four b-sheet and two carboxyterminal a-helical domains (15, 16). NMR structure analysis of recombinant murine (residues 121-231) and hamster PrP (residues 90-230) revealed that single amino acid mutations may lead to altered electrostatic surface potentials of the protein, resulting in changed long-range electrostatic interactions and short-range hydrogen-bonding with other proteins (13, 17). Amino acid mutations in PrP found in familial CJD are assumed to lead to a destabilization of the PrPC structure and thus can spontaneously lead to a PrPC conversion into PrPSc (18 –20). Similarly, a variety of alleles have been found in the prion gene of sheep, some of which

693

0006-291X/98 $25.00 Copyright © 1998 by Academic Press All rights of reproduction in any form reserved.

Vol. 253, No. 3, 1998

BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS

seem to be linked to the length of the scrapie incubation time, i.e. codons encoding amino acids 136 {valine (V) 7 alanine (A)}, 154 {arginine (R) 7 histidine (H)}and 171 {glutamine (Q) 7 R 7 H} (21). Despite all knowledge about the proteinase K resistance of PrPSc, only little is known about the proteinase K sensitivity of PrPC of prion susceptible species. In this current study we were able to demonstrate that even for bovine, ovine, murine, human, mink and hamster PrPC intermediate proteinase K degradation products can be generated, if only mild proteolytic conditions are applied. Interestingly, the molecular mass of these degradation intermediates was reminiscent to those of the proteinase K resistant PrPSc fragments. MATERIALS AND METHODS Preparation of brain homogenates. Twenty two brain samples from healthy cattle (4 animals with 6 octarepeat allele), sheep (3 animals with homozygous PrPARQ alleles ((i.e. PrP with alanine, arginine and glutamine at the positions 136, 154 and 171), 3 animals with heterozygous PrPARQ/PrPARR alleles, and 2 animals with homozygous PrPARR alleles), man (6 individuals), mouse (3 animals with s7s7-sinc genotype), syrian golden hamster (3 animals) and mink (1 animal) were used. Brain samples were homogenized in nine parts of 0,32 M sucrose solution containing Nonidet P-40 (0.5%) and deoxycholic acid (0.5%) (Sigma, Deisenhofen, Germany) and by forcing them through 27g syringe needles and intensive ultrasonification. Residual cell debris were removed by a short centrifugation step (1000 g/5 min) and the supernatants collected for further studies. Total protein contents of all homogenates were determined using the “Total protein” kit (Sigma) and homogenates adjusted to total protein concentrations of 50 mg/ml. Validation of samples and proteinase K stock solution. In order to reveal impacts of the initial homogenization step on the subsequent analysis triplicate homogenates were prepared from a bovine and an ovine (homozygous PrPARQ allele) brain. In order to reveal differences in proteinase K sensitivity of PrPC originating from various brain regions, homogenates of brain stem, cerebellum, frontal cortex, diencephalon, hypothalamus and medulla oblongata (C1-region) of an PrPARQ allelic sheep were analysed. For the determination of the enzyme stability, proteinase K stock solution (0,1 mg/ml A. dest.) was exposed to repeated freezing/ thawing cycles (0, 1, 2, 5, 10 or 20 times) prior to brain homogenate digestion. In another experiment, the stock solution was kept at room temperature for 0, 1, 2, 4 or 6 hours prior to protein digestion. Residual enzyme activity in respect to PrPC degradation was determined as described above using a standard homogenate of ovine brain (PrPARQ allele). Proteinase K digestion experiments. Homogenate volumes (200 ml) corresponding to 10 mg total protein were added to 40 ml proteinase K (Boehringer Mannheim, Mannheim, Germany) stock solution (0.1 mg/ml) and incubated for 5, 10, 20, 30 and 60 minutes at 37°C. Care was taken that all solutions were prewarmed to this tempera-

ture prior to use. The final proteinase K concentration corresponded to 20 mg/ml. Samples were taken at the described time points and the reactions immediately stopped by addition of 20 ml of 100 mM phenylmethylsulfonylfluoride (PMSF) (Sigma). As proteinase K might not immediately be inactivated by PMSF, withdrawn samples were immediately frozen at 220°C. In order to minimize experimental artefacts, digestions were performed at least three times with each homogenate and the digestion products analysed by SDSpolyacrylamide gel electrophoresis (PAGE) and immunoblot, respectively (see below). Quantification of the proteinase K resistance of PrPC. Homogenates were run on SDS-PAGE gels (13%) and separated compounds electrophoretically transferred onto immobilon membranes (Millipore, Bedford, MA, United States). PrPC bands were visualized immunochemically by incubation with poly- or monoclonal antibodies followed by corresponding anti-IgG antibodies covalently linked to horseraddish peroxidase. Banding reactions were developed by use of an enhanced chemiluminescence detection system (Amersham, Little Chalfont, United Kingdom). In order to level out deviations in SDS-PAGE and immunoblot runs each sample was run and quantified at least three times. The following well referenced and/or characterized antibodies were used for PrPC detection: polyclonal rabbit antibodies designated Ra18/4, Ra 32/12 and Ra 38/16 (22) and the monoclonal antibodies (mab) l42 (23), mab 3F4 (24), mab 3B5 (25) and mab 6H4 (15). Banding signals were subsequently quantified using the photoimager technique by which photon emissions are determined electronically. The PrPC signal of a non-digested sample was set as 100% signal value. Deglycosylation experiments using PNGase F. In order to explore whether the glycosylation of PrP is retained in the proteolysis intermediate, a bovine brain homogenate was incubated with PNGase F which removes all high mannose, hybrid, and complex type oligosachharides from N-linked glycoproteins. Prior to digestion, samples were incubated for 5 minutes at 100°C in denaturation buffer (0,5% SDS, 1% b-mercaptoethanol in 50 mM Tris buffer, pH 7,4). Incubation with PNGase F was performed in 50 mM sodium phosphate and 1% Nonidet-P 40 overnight at 37°C. Enzyme-linked immunosorband assay (ELISA). In order to clarify whether the investigated homogenates had different effects on the PK activity and thus mimicked differences in PK sensitivities we established an ELISA test. In this experiment, residual Ovalbumin was specifically detected after incubation with proteinase K diluted in different brain homogenate samples. 5 mg ovalbumin per well were coated onto microtiter plates (Maxi Sorb, Nunc, Wiesbaden) for 2 hours at 37°C. Into each cavity a 1: 20 dilution of brain homogenate (equivalent to 5 mg of total protein) supplemented with 20 mg/ml proteinase K was added and plates incubated for 30 minutes at 37°C. As controls proteinase K dilutions in sucrose solution (i.e. without brain homogenate) were used. At least two brain homogenates per species were investigated in three assays respectively and each ELISA included duplicate determinations respectively. Non-degraded ovalbumin in cavities was subsequently determined by using a monoclonal antibody to this protein (Sigma, Deisenhofen) which was applied for 1.5 h at 37°C. Following thorough washing plates were incubated another hour at 37°C with a caprine anti-mouse-IgG antibody

FIG. 1. (A) Graphic illustration of the degradation of PrPC of sheep (PrPARQ allele), cattle, mink and man by proteinase K (20 mg/ml at 37°C). Standardized brain homogenates were proteolyzed for 0-60 minutes and residual PrPC antigen amounts determined by immunoblotting using mab l42 and the enhanced chemiluminescence system. Light emission signals were recorded by photo-imager technique. Bars depict arithmetrical means and hooks on top standard errors (SEM) above 1%. (B) Immunoblots showing that partial proteinase K degradation of PrPC of four mammalian species leads to the generation of an intermediate fragment of molecular mass of approximately 25-28 kDa. Immunoblots of proteinase K digested brain homogenates (experimental procedure as described under (A)) incubated with mab l42. Bands were visualized by exposure of autoradiographic films. 694

Vol. 253, No. 3, 1998

BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS

695

Vol. 253, No. 3, 1998

BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS

covalently linked to horseraddish peroxidase. After further intensive washing steps residual ovalbumin antigenicity was visualized by using ortho-phenyldiamine substrate into chromogen conversion and the optical density measured at 492 nm. Prior to the proteinase K assays optimal combinations of ovalbumin and mab to ovalbumin concentrations were determined. Results obtained when using brain homogenates of the different species or sucrose solution as diluent were eventually compared with another by statistical means. Statistical data analysis. Immunoblot and ELISA data were analysed, respectively, using the Kruskal-Wallis test (Statistica 5.1, StatSoft Inc., Tulsa, OK, United States) which allows comparison of multiple data sets composed of single measurements.

RESULTS The proteinase K sensitivity of ovine, bovine, human, murine, mink and hamster PrPC was analysed by serial digestion experiments on standardized brain homogenates (in respect to total protein content). PrPC compounds and degradation products were immunoblotted and visualized by chemiluminescence. Residual antigens were quantified electronically by photoimager technique. Impact of Sample Preparation and Enzyme Handling The initial homogenization step had no impact on the subsequent analysis, as three parallel homogenates of a bovine and another three of a sheep brain gave comparable results (data not shown). Moreover, no significant reduction of the proteolytic activity of a proteinase K stock solution was observed on ovine PrPC after repeated freezing/thawing cycles or prolonged storage at room temperature. The Degradation of PrPC by Proteinase K Is a Two Step Process PrPC was proteolyzed using experimental conditions commonly cited for demonstration of PrPSc in the literature, i.e. 20 mg proteinase K/ml (at 37°C) and incubation times of various length (5, 10, 30 and 60 min). Immunoblots were developed using monoclonal antibody mab l42 which is directed to an epitope located in the vicinity of amino acids 144. These experiments revealed that the antigenicity of PrPC of all species was completely diminished after 60 minutes of exposure to proteinase K (Fig. 1a). However, about 30% residual banding signals were visible in all species after 5 minutes of proteinase K digestion and even retained for up to 30 minutes of exposure. In all species, residual antigens did not constitute uncleaved PrPC, as the 35-38 kDa band representing diglycosylated PrPC completely disappeared on immunoblots after as early as five minutes of incubation. Thus mature PrPC was highly protease susceptible. Instead, PrPC was only partially degraded by proteinase K to a well defined intermediate fragment of mo-

lecular mass of approximately 25-28 kDa (Fig. 1b). If the lanes were intentionally overloaded with PrPC degradation products, another two smaller fragments became faintly visible which resembled analoges of the intermediate fragment in the mono- and nonglycosylated form (data not shown). Characterization of the 25-28 kDa Intermediate Fragment In order to clearly define the composition of partially degraded human and bovine PrPC we used sets of seven antibodies to PrP for which the epitopes are known (epitopes were calculated according to amino acid sequence of ovine PrPC, if not otherwise indicated): For detecting human PrPC, polyclonal rabbit antibodies Ra 18/4 (directed to aa 40-56), Ra 32/12 (directed to aa 176-185) and Ra 38/16 (directed to aa 224-244) as well as monoclonal antibodies mab 3F4 (epitope aa 109-112), mab l42 (directed to an epitope in the vicinity of amino acid 144) and mab 6H4 (directed to aa 144-152) were employed. For detecting bovine PrPC, antibodies Ra 18/4, Ra 2/7 (directed to aa 98113), Ra 32/12 and Ra 38/16 as well as mab 3B5 (directed to aa 57-72), mab l42 and mab 6H4 were used. Antibodies binding to epitopes located aminoterminally of position 113 failed to detect the intermediate fragment of both bovine and human PrPC, while the carboxy-terminus from position 144 initially resisted the the proteolytic activity of proteinase K (Fig. 2). This result was subsequently reconfirmed for ovine, murine, hamster and mink PrPC by using two suitable antibodies (Table 1) to amino-terminal and carboxyterminal epitopes (Fig. 3). Deglycosylation experiments revealed that the 25-28 kDa fragment is gylcosylated, as it is reduced to a molecular mass of approximately 16-19 kDa by PNGase F. Shifts in molecular mass resembled in size (approx. 8-9 kDa) those induced by PNGase F on nonproteinase K treated PrPC (data not shown). Again, this deglycosylated fragment was only detectable by using antibodies binding to carboxy-terminal epitopes of PrPC. Comparison of the Proteinase K Sensitivity of PrPC from Different Species and of Different Ovine PrPC Alleles Albeit proteinase K sensitivities are difficult to compare when using PrPC derived from different species and different detection antibodies, the degradation kinetic obtained for ovine, bovine, human, murine and mink PrPC proved to be significantly different (p , 0.001), i.e. mink PrPC resisted proteinase K much better than bovine PrPC. Differences were less pronounced, however, when using another detection antibody that reacts with all investigated species.

696

Vol. 253, No. 3, 1998

BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS

FIG. 2. Schematic illustration depicting the epitopes of the antibodies used for probing the proteinase K digestion intermediate of human and bovine PrPC. Amino-carboxy-terminal signal sequence and aminoterminal restricted fragment marked in grey colour PrPC . Antibodies binding amino-terminally of amino acid 113 fail to detect the fragment (light bars) whereas it is detected by all antibodies binding to epitopes that are located carboxy-terminally of aminoacid 144 (black bars).

Significant differences (p , 0.05) were observed in the proteinase K sensitivity of PrPC of sheep carrying different PrP alleles: PrPARQ/PrPARQ (i.e. PrPC in homozygous form with alanine, arginine and glutamine at the positions 136, 154 and 171, respectively) proved to be substantially more resistant towards proteinase K digestion than PrPARQ/PrPARR and PrPARR/PrPARR (Fig. 4). In order to exclude artefacts due to endogenous proteinase inhibitors, the proteolytic degradation of ovalbumin by proteinase K was monitored when using homogenates of different species as diluents. All investigated homogenates gave similar (p.0,1) results in this experiment (Table 2). Comparison of the Proteinase K Sensitivity of PrPC from Different Brain Regions and of Different Animals Sheep brain homogenates(PrPARQ allele in homozygous form) from brain stem, cerebellum, frontal cortex, diencephalon, hypothalamus and medulla oblongata

TABLE 1

Antibodies Used for the Detection of Bovine, Ovine, Human, Mink, Murine and Hamster PrPc Species

N-terminal

C-terminal

Cattle

mab 3B5 aa 57-72 mab 3B5 aa 57-72 mab 3F4 aa 109-112 Ra 18/4 aa 40-56 mab 3B5 aa 57-72 mab 3F4 aa 109-112

mab l42 aa 145-163 mab l42 aa 145-163 mab l42 aa 145-163 mab l42 aa 145-163 mab 6H4 aa 144-152 mab 6H4 aa 144-152

Sheep Man Mink Mouse Hamster

were proteinase K digested under the described conditions. No differences in the proteinase K sensitivity of PrPC originating from these brain localizations were observed (Fig. 5). Moreover, no differences in the proteinase K sensitivities of PrPC were found in different individuals of a given species, i.e. in sheep (5 animals), cow (4 animals), mouse (3 animals), hamster (3 animals) and man (6 individuals). DISCUSSION Results of this study indicate that PrPC of cattle, sheep, man, mink, mouse and hamster are initially cleaved by proteinase K to a 25-28 kDa intermediate fragment prior to complete proteolysis. This fragment basically represents the carboxy-terminus of the protein as it was exclusively detected by antibodies binding to epitopes at or carboxy-terminally to amino acid 144 of PrPC. In contrast, antibodies directed to epitopes at or amino-terminally to amino acid 112 did not react with it. According to NMR structure analysis, the carboxy-terminal half of PrPC (amino acids 122-213) is folded into three a-helices and two b-sheets with interconnecting loops, thus resembling almost a globule-like structure of 30-50 nm diameter (12, 13). It is assumed that the amino-terminus (22–121) extends from this well defined structure as a long and highly flexible tail (14). Proteinase K is a promiscuous enzyme that should degrade all areas of PrPC equally well and independantly from their amino acid composition. Only structural features such as a-helices and b-sheets can interfere with enzyme/substrate interactions. The two step kinetic of the PrPC degradation is therefore in line with the assumed structural model for this protein. It is well feasible that the unfolded amino-terminus displays a intrinsicly higher proteolytic sensitivity as the carboxy-terminus which is degraded only after prolonged exposure to proteinase K. One possible explanation for this two step kinetik is that the proteolytic activity of PK stops somewhere

697

Vol. 253, No. 3, 1998

BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS

FIG. 3. Different binding affinities of amino-terminal and carboxy-terminal antibodies could be demonstrated for all investigated species. Western blot analysis of proteinase K digested brain homogenates were incubated with amino-terminal or carboxy-terminal antibodies (antibodies are given below the immunoblots). Graphic illustration of mean values of more than 8 immunoblots that were quantified using photo-imager. Bars depict arithmetrical means and hooks on top standard errors (SEM) above 1%.

between residues 112 and 144, probably due to sterical reasons. A cleavage site in this region of PrP has been proposed for chicken PrP earlier (26). As chicken and

mammalian PrP sequences are almost identical at these residues, it is well feasible that this cleavage site also exists in mammalian PrP. After cleavage, the

698

Vol. 253, No. 3, 1998

BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS

FIG. 3—Continued

amino-terminal fragment is proteolyzed much faster than the carboxy-terminal fragment. Other authors proposed that a carboxy-terminal fragment of PrPC already exists in untreated human brain homogenate due to endogenous proteolysis, which might result of the metabolic turnover of PrPC. Radiosequencing of

this fragment revealed that it starts at residue 111 or 112 of human PrP (27). The PK digestion intermediate fragment contains carbohydrate moieties as the molecular mass shifted from 25-28 kDa to 16-19 kDa after PNGase F treatment. The shift in moleculas mass is in the same range

699

Vol. 253, No. 3, 1998

BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS

FIG. 4. Differences in the partial proteinase K resistance of the intermediate fragment of ovine cellular PrPARQ/PrPARQ, PrPARQ/PrPARR and PrPARR/PrPARR. Standardized brain homogenates were proteinase K digested (20 mg/ml at 37°C for 5, 10, 20 or 30 min, respectively) and retained PrPC antigens visualized using mab l42, enhanced chemiluminescence and autoradiographic films. Retained antigens were also quantified using the photo-imager technique. Bars depict arithmetrical means and hooks on top standard errors (SEM) above 1%.

as after PNGase F treatment of mature PrPC. This accords with the proposed two step kinetic of PrPC after treatment with proteinase K, as both glycosylation sites are located within the carboxy-terminal half and are thus not affected by cleavage of the aminoterminus of the protein. Experimental conditions used in this current study (i.e. 20 mg proteinase K/ml and incubation times ranging from 0-60 minutes) included those frequently applied in experimental as well as diagnostic protocols. Analytical experiments on the proteinase K sensitivity of PrPSc described in literature are performed using a TABLE 2

Mean Values of Residual Ovalbumin Detection in an ELISA after Incubation with Proteinase K and Brain Homogenates of Different Species Species

Mean value (MV)

Standard error of the mean (SEM)

Cattle Sheep Man Mouse Mink Hamster All species

18.8 28.6 29.0 35.5 41.3 40.7 32.3

7.9 5.7 11.4 8.0 14.4 9.1

Note. Figures represent percentage of increased OD values in comparison to incubation with proteinase K alone.

variety of experimental protocols with enzyme concentrations and exposure times from 3.3 mg proteinase K/ml for 10 minutes (28, 29) up to 100 mg proteinase K/ml for 60 minutes (30). For diagnostic purposes prion rods are purified by differential ultracentrifugation and subsequently exposed to enzyme concentrations between 20 and 50 mg/ml for 30-60 minutes (15, 31). Accordingly, results obtained in our study indicate that signals obtained by using simplified diagnostic protocols employing proteinase K treated “quick and dirty” homogenate preparations should be interpreted with caution, if the proteolytic efficacy is not monitored adequately. When using crude tissue homogenates, only exposure times longer than 30 minutes and enzyme concentrations of 20 mg proteinase K/ml and greater, reliably warrant the discrimination between PrPSc and PrPC on grounds of their relative proteolytic stability. Consequently, the intermediate PrPC compound may easily be misinterpreted as being the proteinase K resistant core fragment of PrPSc due to its molecular mass and retained antigenicity. Our data also revealed statistically significant differences in the proteinase K sensitivities of PrPC of the six investigated species. However, differences in the observed proteinase K sensitivities of PrPC may partly result from differences in the relative PrPC expression levels, PrPC antigenicities and corresponding antibody affinities. The latter applies particularly for the only scarce detection of the intermediate fragment of mu-

700

Vol. 253, No. 3, 1998

BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS

FIG. 5. Proteinase K sensitivity of PrPC from six regions of ovine brain (brain stem, cerebellum, frontal cortex, diencephalon, hypothalamus and medulla oblongata). Samples were digested for 0, 5 and 10 minutes with 20 mg proteinase K/ml and analysed in western blot using mab l42. Results were quantified using photo-imager technique. Bars depict arithmetrical means and hooks on top standard errors (SEM) above 1%.

rine PrPC as long as mab 6H4 is used which possesses only limited affinity to murine and hamster PrPC. Antibody specific effects should be negligible, however, when analysing PK sensitivities of bovine, ovine, mink and human PrPC, which were all detected by the same monoclonal antibody, namely mab l42. Species specific and antibody dependant artefacts can be excluded in the analysis of sheep harboring different PrP alleles. In our digestion experiments, PrPARR/PrPARR and PrPARQ/PrPARR displayed a significantly higher sensitivity to proteinase K than PrPARQ/ PrPARQ. The ovine PrP alleles may well differ in the three-dimensional structure of PrPC and/or the electrostatic surface potential leading to changed PK binding affinities and proteolytic activities. This observation coincides with the scrapie susceptibility of sheep carrying PrPARQ/PrPARQ alleles that are substantially more susceptible to scrapie infection and disease than those carrying PrPARQ/PrPARR or PrPARR/PrPARR. However, a possible connection of these two observations cannot be proved at present. It has been postulated that arginine at position 171 (R171) in the ovine sequence as well as lysine at position 219 (K219) in the human sequence result in a resistance to TSE infection due to blockage of the Protein X binding site (17). Protein X is supposed to act as a chaperon in PrPSc formation (26). This was confirmed by in-vitro conversion experiments in cell cultures expressing PrPC with point mutations at these sites. Therefore R171 in the ovine sequence and K219 in the human sequence seem

to act as dominant negative factors leading to resistance to TSE infection (32). Apart from these species and allelic effects there were no differences found in the proteinase K sensitivities of PrPC derived from six ovine brain areas (brain stem, medulla oblongata, frontal cortex, diencephalon, cerebellum, hypothalamus) or of PrPC from individuals of given species/PrP alleles. This corresponds with results obtained in PrPSc digestion experiments where no differences in the PK resistance of PrPSc derived from different murine brain areas or from different individuals of an experimental group was observed (33). As these samples harbor the same amino acid sequences, these results are in line with the proposed effect of amino acid variation on the protein structure and on protein binding affinities. Taken together, we have shown that proteinase K degrades PrPC by first removing the amino-terminus between residues 113 and 144 before proteolysing the carboxy-terminus. The region between residues 100 and 130 has been proposed earlier to play a crucial role in PrPSc formation, as this region forms an a-helix in PrPC and is folded into b-sheets in PrPSc (34). This two step process correlates well with the known threedimensional structure of PrPC and it is reminescent of the similarly two stepped complete PrPSc degradation which, however, requires either higher proteinase K concentrations or prolonged exposure times. Further studies will be performed to clarify the structural

701

Vol. 253, No. 3, 1998

BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS

mechanisms behind the proteinase K resistance of PrPC and PrPSc. 16.

ACKNOWLEDGMENTS Annett Bu¨nten and Maria Mo¨ller are acknowledged for their excellent technical assistance. We are gateful to Dr. Christoph Staubach for his advice on the statistical analysis and Torres Sweeney for her helpful suggestion for preparing this manuscript. This work was supported in parts by grants from the German “Bundesministerium fu¨r Erna¨hrung, Landwirtschaft und Forsten,” the German “Bundesministerium fu¨r Bildung, Wissenschaft und Technologie” as well as by grants from the EU commission.

17.

18. 19.

REFERENCES

20.

1. Prusiner, S. B. (1982) Lancet 216, 136 –144. 2. Stack, M. J., Aldrich, A. M., and Davis, L. A. (1997) J. Comp. Pathol. 116, 181–189. 3. Merz, P. A., Somerville, R. A., Wisniewski, H. M., and Iqbal, K. (1981) Acta Neuropathol. 54, 63–74. 4. Scott, A. C., Wells, G. A., Stack, M. J., White, H., and Dawson, M. (1990) Vet. Microbiol. 23, 295–304. 5. Prusiner, S. B., McKinley, M. P., Bowman, K. A., Bolton, D. C., Bendheim, P. E., Groth, D. F., and Glenner, G. G. (1983) Cell 35(Pt 1), 349 –58. 6. Lehmann, S., and Harris, D. A. (1995) J. Biol. Chem. 270, 24589 –24597. 7. Oesch, B., Westaway, D., Walchli, M., McKinley, M. P., Kent, S. B., Aebersold, R., Barry, R. A., Tempst, P., Teplow, D. B., Hood, L. E., and et al (1985) Cell 40, 735–746. 8. Prusiner, S. B., Scott, M. R., DeArmond, S. J., and Cohen, F. E. (1998) Cell 93, 337–348. 9. Pahler, A., Banerjee, A., Dattagupta, J. K., Fujiwara, T., Lindner, K., Pal, G. P., Suck, D., Weber, G., and Saenger, W. (1984) EMBO J. 3, 1311–1314. 10. Kraus, E. and Femfert, U. (1976) Hoppe Seylers. Z. Physiol. Chem. 357, 937–947. 11. Kraus, E., Kiltz, H. H., and Femfert, U. F. (1976) Hoppe Seylers. Z. Physiol. Chem. 357, 233–237. 12. Riek, R., Hornemann, S., Wider, G., Billeter, M., Glockshuber, R., and Wuthrich, K. (1996) Nature 382, 180 –182. 13. James, T. L., Liu, H., Ulyanov, N. B., Farr-Jones, S., Zhang, H., Donne, D. G., Kaneko, K., Groth, D., Mehlhorn, I., Prusiner, S. B., and Cohen, F. E. (1997) Proc. Natl. Acad. Sci. USA 94, 10086 –10091. 14. Donne, D. G., Viles, J. H., Groth, D., Mehlhorn, I., James, T. L., Cohen, F. E., Prusiner, S. B., Wright, P. E., and Dyson, H. J. (1997) Proc. Natl. Acad. Sci. USA 94, 13452–13457. 15. Korth, C., Stierli, B., Streit, P., Moser, M., Schaller, O., Fischer, R., Schulz-Schaeffer, W., Kretzschmar, H., Raeber, A., Braun,

21. 22. 23. 24.

25. 26. 27.

28. 29. 30.

31. 32.

33. 34.

702

U., Ehrensperger, F., Hornemann, S., Glockshuber, R., Riek, R., Billeter, M., Wuthrich, K., and Oesch, B. (1997) Nature 390, 74 –77. Peretz, D., Williamson, R. A., Matsunaga, Y., Serban, H., Pinilla, C., Bastidas, R. B., Rozenshteyn, R., James, T. L., Houghten, R. A., Cohen, F. E., Prusiner, S. B., and Burton, D. R. (1997) J. Mol. Biol. 273, 614 – 622. Billeter, M., Riek, R., Wider, G., Hornemann, S., Glockshuber, R., and Wuthrich, K. (1997) Proc. Natl. Acad. Sci. USA 94, 7281–7285. Duyckaerts, C., Sazdovitch, V., Seilhean, D., Delaere, P., and Hauw, J. J. (1993) J. Neural Transm. Suppl. 107–118. Schatzl, H. M., Da Costa, M., Taylor, L., Cohen, F. E., and Prusiner, S. B. (1995) J. Mol. Biol. 245, 362–374. Palmer, M. S., Dryden, A. J., Hughes, J. T., and Collinge, J. (1991) Nature 352, 340 –342. Dawson, M., Hoinville, L. J., Hosie, B. D., and Hunter, N. (1998) Vet. Rec. 623– 625. Groschup, M. H., Weiland, F., Straub, O. C., and Pfaff, E. (1996) Neurobiol. Dis. 3, 191–195. Harmeyer, S., Pfaff, E., and Groschup, M. H. (1998) J. Gen. Virol. 79, 937–945. Kascsak, R. J., Rubenstein, R., Merz, P. A., Tonna-DeMasi, M., Fersko, R., Carp, R. I., Wisniewski, H. M., and Diringer, H. (1987) J. Virol. 61, 3688 –3693. Krasemann, S., Groschup, M. H., Harmeyer, S., Hunsmann, G., and Bodemer, W. (1996) Mol. Med. 2, 725–734. Harris, D. A., Huber, M. T., van Dijken, P., Shyng, S. L., Chait, B. T., and Wang, R. (1993) Biochemistry 32, 1009 –1016. Chen, S. G., Teplow, D. B., Parchi, P., Teller, J. K., Gambetti, P., and Autilio-Gambetti, L. (1995) J. Biol. Chem. 270, 19173– 19180. Lehmann, S., and Harris, D. A. (1996) Proc. Natl. Acad. Sci. USA 93, 5610 –5614. Lehmann, S. and Harris, D. A. (1996) J. Biol. Chem. 271, 1633– 1637. Raymond, G. J., Hope, J., Kocisko, D. A., Priola, S. A., Raymond, L. D., Bossers, A., Ironside, J., Will, R. G., Chen, S. G., Petersen, R. B., Gambetti, P., Rubenstein, R., Smits, M. A., Lansbury, P. T., Jr., and Caughey, B. (1997) Nature 388, 285–288. Collinge, J., Sidle, K. C., Meads, J., Ironside, J., and Hill, A. F. (1996) Nature 383, 685– 690. Kaneko, K., Zulianello, L., Scott, M., Cooper, C. M., Wallace, A. C., James, T. L., Cohen, F. E., and Prusiner, S. B. (1997) Proc. Natl. Acad. Sci. USA 94, 10069 –10074. Kuczius, T., Haist, I., and Groschup, M. H. (1998) J. Infec. Dis. 178, 693– 699. Gasset, M., Baldwin, M. A., Lloyd, D. H., Gabriel, J. M., Holtzman, D. M., Cohen, F., Fletterick, R., and Prusiner, S. B. (1992) Proc. Natl. Acad. Sci. USA 89, 10940 –10944.