Hamster PrP

Fig. 2. Primary structure of hamster PrP (Stahl et al., 1993). The first 22 residues at the N-terminus are the signal sequence. PrPc is completely digested by proteinase K, whereas the N-terminal sequence of PrPSc to residue 89 (arrow, closed head) is digested. — CHO indicates the glycosylation sites at residues 181 and 197 Gpi the glycosylpho-sphatidylinositol anchor at 231 and the N-terminal octarepeats. In one case of human prion disease, a stop codon was found at 145 (arrow, open head) (Kitamoto et al., 1993). HI, H2, H3, and H4 denote the predicted a-helices (Huang et al, 1994, 1996), and A-C denote the a-helices and SI, S2 the /(-strands determined by solution NMR (James et al, 1997).

Fig. 5. Physical-chemical parameters as a function of residue number for hamster PrP (Inouye and Kirschner, 1998). The parameters (arbitrary scale) are charge at pH 7 hydrophobicity a-helix (solid), /8-strand (dashed) turn (solid), coil (dashed) a-helical (solid) and /8-strand amphiphilicity (dashed). The predicted helices (Huang et al., 1994) are labeled HI, H2, H3, and H4, and the NMR-observed helices and /8-strands are A-C and SI, S2, respectively (James et al., 1997).

MD simulations have been employed extensively to study the conformational dynamics of the wild-type PrP for a range of species. A wide variety of force fields, setups, and analyses has been used to this end. The majority of studies is performed on human, mouse, or hamster PrP. In all but a few contributions [53-55], simulations include only the protein portion, i.e., the unglycosylated form without membrane anchoring. These simulations therefore present the recombinant PrP (recPrP) that is used for many in vitro studies, usually produced in Escherichia coli. Often, only the structured part of recPrP is simulated. Although most initial MD studies could only capture local dynamics [40—431. some studies were able to shed light on the misfolding process over short timescales ( 10 ns) [56, 57]. More recent studies reporting multiple simulations of 50 ns or more [58, 59], have provided a more comprehensive view of the conformational dynamics. 

Fig. 3 Potential misfolding and aggregation of PrP. (a) Misfolding of D147N hamster PrP observed in simulation at strongly acidic pH and docking of three misfolded monomers into an initial aggregate [117]. (b) Spiral protofibril models (built up from six monomers each) for D147N hamster PrP (as shown in panel a), bovine PrP and human PrP [121]...

After our protofibril model was shown to be plausible, similar models were built for WT hamster, human, and bovine PrP [121]. Again, the initial monomer structure was obtained from MD simulation at the strongly acidic pH regime. In contrast to hamster PrP, human and bovine PrP showed a left-handed spiral formation (Fig. 3b). This difference and further subtle differences between the individual protofibril models may reflect strain differences and give clues to the origin of observed species barriers [122] transmission of prion disease between different species can be inefficient or even absent. Furthermore, the models may help to... 

Parchment OG, Essex JW (2000) Molecular dynamics of mouse and Syrian hamster PrP implications for activity. Proteins Struct Funct Genet 38 327... 

Before we discuss approaches for the synthesis of the PrP the overall structure should be explained. Mature PrP, a protein of about 240 amino acids, exhibits three domains (see [30] for primary sequence and posttranslational modifications of hamster PrP). The N-terminal domain is an intrinsically unstructured fragment with around 100 amino acids. It harbors five octarepeats whose function is unknown. The central core of PrP is structured with mainly a-helical elements. It is followed by the C-terminal fragment which is equipped with the GPI anchor. Further posttranslational modifications are two carbohydrate moieties which are connected to Asn residues at the second helix (a2) of the structured domain. Because infectivity has been associated with the central core (for a review see [31]), most chemical work has been focused on this region. 

Fig. 3. NMR structures of PrP-sen. The refined NMR structure of (A) recombinant mouse PrP-sen from amino acid residues 121-231 (Riek et al., 1996) and (B) recombinant hamster PrP-sen from amino acid residues 23-231 (Liu et al, 1999). The structure of full-length mouse PrP-sen has been derived (Riek et al., 1997), but residues from the signal peptide cleavage site to the structure illustrated (i.e., 23-120) exist in a flexible, random coil-like state as shown for the hamster PrP-sen structure in (B). S, P strand H, a helix. Figure prepared with the program MOLMOL (Koradi et al, 1996).

Fig. 4. PrP-res isolated from hamsters infected with different scrapie strains can determine the final conformation of the protease-resistant product. Radiolabeled hamster PrP-sen without the GPI anchor was mixed with hamster PrP-res isolated from the brains of hamsters infected with either hyper (Hy) or drowsy (Dy) scrapie in a cell-free conversion assay (Kocisko et al, 1994). The PrP-res was either pre-treated (+) or not (-) with proteinase K to remove any contaminating proteins (Bessen et al, 1995). The left side panel shows the radiolabeled PrP-sen in the reaction after a 2-day incubation and before digestion with proteinase K (-PK) and represents 10% of the total reaction. The right side panel shows the radiolabeled, protease-resistant product remaining after digestion of the reaction with proteinase K (+PK) and represents the remaining 90% of the reaction. Note the characteristic size shift of 1 to 2 kDa which is used to distinguish PrP-res from the hyper and drowsy scrapie strains (Bessen and Marsh, 1992a). There is no significant difference in the amount of protease-resistant product formed when PK pretreated or untreated PrP-res is used (53% vs. 48% for hyper PrP-res and 47% vs. 40% for drowsy PrP-res). Molecular mass markers are shown in kilodaltons on the right.

The first evidence to suggest that the primary sequence of PrP could profoundly influence TSE species barriers came from studies utilizing the strong barrier to infection that exists between hamsters and mice. Although mice are fully susceptible to mouse scrapie, they are resistant to infection with a particular strain of hamster scrapie. Mice infected with hamster scrapie do not become clinically ill within the lifetime of the animal even though it appears that infectivity can be sequestered in these animals (Race and Chesebro, 1998). However, transgenic mice that overexpressed hamster PrP-sen were fully susceptible to hamster scrapie (Scott et al, 1989). Subsequent studies demonstrated that the hamster/mouse species barrier could be crossed even when hamster PrP-sen expression was restricted to neurons (Race et al, 1995) or astrocytes (Raeber et al, 1997), but not T-cells or hepatocytes (Raeber et al,... 

Therefore, even though mouse and hamster PrP-sen are highly homologous and differ at only 16 amino acid positions (i.e., 94% homology), these differences are sufficient to significantly effect crossspecies transmission of TSE infectivity. 

The Syrian hamster PrP gene sequence was already known from partial purification of PrP 27-30, Edman iV-terminal sequencing, cloning of a cDNA, and subsequent similar studies on full length PrP and... 

For structural studies of PrP, recombinant proteins were expressed in Escherichia coli. The constructs used contain either the intact polypeptide chain of the mature form of natural PrP (Fig. 1), possibly with some additional, construct-related residues at either chain end, or fragments thereof. Presently it appears impractical to envisage three-dimensional structure determinations with mammalian prion proteins from natural sources. However, sufficient amounts of natural PrP have been isolated to enable qualitative comparative studies with the recombinant protein by optical spectroscopy. Overall, these experiments indicate close similarity between the natural and the corresponding recombinant prion protein. Thus, the circular dichroism (CD) spectrum of monomeric hamster PrP extracted from hamster brains into a micellar environment of 30 mM w-octyl-P-glucopyranoside at pH 7.5 is typical for... 

P-sheet conformations at acidic pH, caused by reduction of the disulfide bond, were first reported for reduced hamster PrP(90-231) (Mehlhorn et at., 1996) and reduced human PrP(91-231) (Jackson etal, 1999a, 1999b). 

Fig. 5. Cell-free PrP conversion reaction. The right panel is a phosphor-autoradiographic image of an SDS-PAGE gel showing only the radiolabeled hamster PrP molecules in the reaction with and without PK digestion after incuhation in the presence or absence of unlabeled PrP-res. The PrP-res (PrP ) was isolated from scrapie-infected hamster brain tissue. Details of the reaction can be found in (Kocisko et at., 1994).

Since the conversion reaction has been dissected kinetically into initial binding and conversion-to-PrP-res steps, tbe question arises as to which step is most important in controlling the sequence specificity of the conversion reaction (DebBurman et al, 1997 Horiucbi et al, 1999). Recent experiments using mouse and hamster PrP isoforms indicated that the binding of PrP-sen to heterologous PrP can occur much more efficiently than the subsequent conversion to PrP-res (Horiucbi et al,... 

Evidence that the two strain-associated conformations of hamster PrP-res could propagate themselves from the same hamster PrP-sen precursor was obtained using a cell-free conversion reaction (Fig. IB) (Bessen et al, 1995). In these studies, HY and DY PrP-res were each incubated with hamster PrP-sen, and the PrP-sen was converted to PK-resistant PrP products with the same l-2 kDa difference in molecular mass that distinguishes the PK-treated HY and DY PrP-res molecules. This finding suggested that the strain-specific forms of PrP-res are faithfully propagated through direct PrP-sen-PrP-res interactions both in vitro and in vivo as had been proposed earlier (Bolton and Bendheim, 1988 Prusiner, 1991 Bessen and Marsh, 1994). 

Initial studies with synthetic PrP peptides focused on the identification of protein domains that could be involved in the conformational conversion of PrP into disease-associated species. To test the hypothesis that this conversion involves the transition of a helices into (3 sheet, Gasset et al (1992) examined the physicochemical properties of peptides homologous to residues 109-122,129-141,178-191, and 202-218 of Syrian hamster PrP, which were predicted to correspond to a-helical regions (designated HI, H2, H3, and H4, respectively) by computational studies. At variance with the theoretical predictions, peptides HI, H3, and H4 exhibited extensive (3-sheet structure in the solid state and in aqueous solution and assembled into fibrils showing ultrastuctural and tinctorial properties of amyloid, as described in detail later. Multidimensional het-eronuclear NMR has revealed that the prediction of the two C-terminal... 

---