Horse Prion Protein NMR Structure and Comparisons with Related Variants of the Mouse Prion Protein

J. Mol. Biol. (2010) 400, 121–128

doi:10.1016/j.jmb.2010.04.066

Available online at www.sciencedirect.com

COMMUNICATION

Horse Prion Protein NMR Structure and Comparisons with Related Variants of the Mouse Prion Protein Daniel R. Pérez†, Fred F. Damberger† and Kurt Wüthrich⁎ Institute of Molecular Biology and Biophysics, ETH Zurich, Schafmattstrasse 20, CH-8093 Zurich, Switzerland Received 2 March 2010; received in revised form 30 April 2010; accepted 30 April 2010 Available online 8 May 2010

The NMR structure of the horse (Equus caballus) cellular prion protein at 25 °C exhibits the typical PrPC [cellular form of prion protein (PrP)] global architecture, but in contrast to most other mammalian PrPCs, it contains a well-structured loop connecting the β2 strand with the α2 helix. Comparison with designed variants of the mouse prion protein resulted in the identification of a single amino acid exchange within the loop, D167S, which correlates with the high structural order of this loop in the solution structure at 25 °C and is unique to the PrP sequences of equine species. The β2–α2 loop and the α3 helix form a protein surface epitope that has been proposed to be the recognition area for a hypothetical chaperone, “protein X,” which would promote conversion of PrPC into the disease-related scrapie form and thus mediate intermolecular interactions related to the transmission barrier for transmissible spongiform encephalopathies (TSEs) between different species. The present results are evaluated in light of recent indications from in vivo experiments that the local β2–α2 loop structure affects the susceptibility of transgenic mice to TSEs and the fact that there are no reports on TSE in horses. © 2010 Elsevier Ltd. All rights reserved.

Edited by M. F. Summers

Keywords: cellular horse prion protein; transmissible spongiform encephalopathy; NMR structure determination; protein structure; β2–α2 loop

Introduction Transmissible spongiform encephalopathies (TSEs) are a group of invariably fatal neurodegenerative diseases including kuru, Creutzfeldt–Jakob disease, fatal familial insomnia and the Gerstmann– Sträussler–Scheinker syndrome in humans, scrapie in sheep, bovine spongiform encephalopathy in cattle, and chronic wasting disease in deer and .1–3 There have been no reports of TSE in horses, which is remarkable considering that TSEs in other species tend to develop at advanced age and horses are often kept for a long lifetime. *Corresponding author. E-mail address: [email protected]. † D.R.P. and F.F.D. contributed equally to this work. Abbreviations used: PrP, prion protein; PrPC, cellular form of PrP; PrPSc, scrapie form of PrP; ecPrP, horse PrP; mPrP, mouse PrP; TSE, transmissible spongiform encephalopathy; NOE, nuclear Overhauser enhancement; NOESY, NOE spectroscopy; ePrP, elk PrP; bvPrP, bank vole PrP.

Although transmission of a TSE between different individuals of the same mammalian species can occur efficiently, the process of transmission between different species is either inefficient or completely ineffective.4 Considering that the “protein-only hypothesis” suggests a change in protein conformation as the possible cause of the onset of TSEs, the stringency of the species barrier may be related to species-specific variations in the PrPC [cellular form of prion protein (PrP)] structure.5 In this context, NMR structures of recombinant prion proteins from a variety of species have been determined.6–14 The relevance of this approach has been substantiated by the demonstration that recombinant PrP has the same fold as PrPC isolated from bovine brain, indicating that the lack of posttranslational modifications in PrPC expressed in Escherichia coli has at most limited local effects on the protein molecular architecture.15 The global architecture of recombinant PrPCs is nearly identical for different mammals, consisting of a flexibly disordered N-terminal polypeptide segment with about 100 residues and a C-terminal globular domain of about 100 residues, which

0022-2836/$ - see front matter © 2010 Elsevier Ltd. All rights reserved.

122 invariably contains three α-helices and a short antiparallel β-sheet. Within this conserved scaffold, the three-dimensional (3D) solution structures of PrPCs show local structure variations, most prominently in a loop of residues 165–172 (see Schätzl et al.16 for the numeration used) that connects two regular secondary structures, β2 and α2. Whereas in PrPC from most mammalian species this loop shows a poorly defined conformation at around 20 °C, it is well defined in PrPC of elk,12 bank vole13 and wallaby.14 Improved definition of the β2–α2 loop, when compared to the cellular form of mouse PrP (mPrP C ), has also been reported for an NMR structure determination of recombinant Syrian hamster PrP.8 In elk and bank vole PrPC, this

Fig. 1. Alignment of residues 163–175 and 215–227 of ecPrP (E. caballus, NCBI protein database (http://www. ncbi.nlm.nih.gov) accession code ABL86003), mPrP (Mus musculus, AAA39997), ePrP (Cervus elaphus nelsoni, AAB94788), bvPrP (Myodes glareolus, AAL57231), twPrP (Macropus eugenii, AAT68001), shPrP (Syrian hamster PrP) (Mesocricetus auratus, AAA37091), hPrP (human PrP) (Homo sapiens, AAA60182), and bPrP (bovine PrP) (Bos taurus, AAB25514). The complete ecPrP sequence of residues 163–175 and 215–227 is given. For the other proteins, only the amino acids that are different from ecPrP are indicated, with dots indicating the presence of identical amino acids. For the cloning of ecPrP, horse genomic DNA was purified from whole blood obtained from race horse Thoroughbred breed (Grosstierchirugie, Tierspital, University of Zurich) using silica gel-based separation (QIAamp DNA Blood MaxiKit10, Qiagen). The ecPrnp gene was amplified via PCR in one step with Taq DNA polymerase (Qiagen), using the two primers 5′-CGG ATC CGT GGT TGG GGG CCT CGG C-3′ and 5′-GGA ATT CTT AGC TCG CCC CTC TTT GTT GAA AAG CC-3′ to amplify the sequence corresponding to the globular domain fragment ecPrP(121–231). The mPrP variants mPrP[D167S](121–231) and mPrP[D167S,N173K](121– 231) were generated by mutagenesis from mPrP(121– 231), using the QuikChange site-directed mutagenesis kit (Stratagene) in accordance with the instructions of the manufacturer. For mPrP[D167S](121–231), the two primers 5′-CAG TGA GTC AGT ACA G-3′ and 5′-CTG TAC TGA CTC ACA CTG-3′ were used. The product of this first mutagenesis step was the starting point for generating mPrP[D167S,173K](121–231), using the two primers 5′-CAA CCA GAA AAA CTT CG-3′ and 5′-CGA AGT TTT TCT GGT TG-3′. Two additional variants with single amino acid exchanges in mPrP, mPrP[Q168E](121–231) and mPrP[N173K](121–231) were prepared by performing sitedirected mutagenesis of mPrP(121–231) in an analogous way as described for mPrP[D167S](121–231), using the forward primers 5′-CTA CAG GCC AGT GGA TGA GTA CAG CAA CCA G-3′ for mPrP[Q168E](121–231) and 5′CAA CCA GAA AAA CTT CG-3′ for mPrP[K173](121–231).

Horse Prion Protein Solution Structure

conformational difference has been traced to a single amino acid substitution within the loop, S170N,12,13 which is also found in Syrian hamster PrPC (Fig. 1), whereas in wallaby PrPC it correlates with a longrange interaction between residue 166 in the β2–α2 loop and residue 225 in the α3 helix.14 The β2–α2 loop and the spatially nearby C-terminal end of the α3 helix form a continuous surface epitope that has been proposed as the binding site for a hypothetical “protein X,” which would affect the conversion of PrP C into the disease-related “scrapie form,” PrPSc.17,18 The observation of local species-specific structure variations in this presumed functional surface epitope of cellular prion proteins is of special interest as a result of the findings of a study showing that these variations can affect the susceptibility of transgenic mice to TSEs.19 The transgenic mice used for this study express a variant mPrP with the two substitutions S170N and N174T, which are characteristic of elk PrPC, and lead to a well-defined β2–α2 loop conformation at 20 °C. These tg1020 mice developed a spontaneous TSE that was transmissible over several generations of tga 20 mice, and from these to wild-type mice, which both express wildtype mPrPC. The present study reports the NMR structure of the globular domain of horse (Equus caballus) PrPC [ecPrP(121–231)], which attracted our interest, because within the β2–α2 loop it shows two amino acid substitutions unique to equine species when compared to other mammals (Fig. 1). Since these sequence variations were found to associate with NMR data indicating the presence of a well-defined β2–α2 loop in the temperature range 15–25 °C, we then designed variants of mPrP to identify the amino acid exchanges between the two species that result in an ordered loop in the cellular form of horse PrP (ecPrPC). Cloning, expression and purification of ecPrP (121–231) and related designed variants of mPrP The ecPrnp gene was amplified via polymerase chain reaction from genomic DNA obtained from blood of a thoroughbred race horse (Grosstierchirugie, Tierspital, University of Zurich). The genes for variant mPrP(121–231) were prepared by sitedirected mutagenesis with the Quikchange kit (Stratagene) to replace amino acid residues in the β2–α2 loop of mPrP with the amino acids present in corresponding positions in horse PrP, that is, mPrP [D167S](121–231), mPrP[Q168E](121–231), mPrP [N173K](121–231), mPrP[D167S,Q168E](121–231), mPrP[D167S,N173K](121–231), and mPrP[Q168E, N173K](121–231) (Fig. 1). Uniformly 15 N- and 13 C,15N-labeled proteins were prepared by overexpression in Escherichia coli and purified as described.20,21 For the preparation of the NMR samples, the purified proteins were concentrated by nitrogen flux evaporation and dialyzed overnight against NMR buffer [90% H2O, 10% D2O, and 10 mM deuterated sodium acetate (sodium acetated4) at pH 4.5].

Horse Prion Protein Solution Structure

123

NMR structure determination of ecPrP(121–231) Nearly complete sequence-specific backbone and amino acid side-chain assignments were obtained for ecPrP(121–231), including those of residues 167– 171 and 175, for which no backbone 15N–1H crosspeaks are observable in mPrP(121–231) at 20 °C.12 Steady-state 15N{1H} nuclear Overhauser enhancement (NOE) data indicate that the polypeptide segment of residues 125–228 forms a globular domain and that the flanking peptide segments 121–124 and 229–231 undergo high-frequency motions (data not shown). Structure calculation based on using these resonance assignments for analysis of 3D 15N- and 13 C-resolved [1 H,1 H]NOESY data sets resulted in a high-quality structure for ecPrP(121–231), as indicated by the statistics for the final ATNOS/CANDID/DYANA cycle (Table 1). The globular domain of ecPrP residues 125–228 displays the same global architecture as that observed

for PrPC from other mammalian species,6–14 with three α-helices of residues 143–156, 173–194, and 200– 228, and a two-stranded antiparallel β-sheet of residues 128–131 and 161–164 (Fig. 2a and b). The last turn of helices α1 and α3 is in a 310 conformation, and a disulfide bridge connects Cys179 in the α2 helix to Cys214 in the α3 helix. The polypeptide segment of residues 165–172, which is referred to as the “β2–α2 loop” in this study, is structurally well defined at 25 °C (Fig. 2a) and includes a 310 helical turn formed by residues 165–169, with hydrogen bonds from Glu168 HN to Pro165 CfO and from Tyr169 HN to Val166 CfO. Screening of designed mPrP variants for a well-structured β2–α2 loop at 25 °C There are three amino acid substitutions within the β2–α2 loop of ecPrP when compared with mPrP, that is, D167S, Q168E, and N173K (Fig. 1). To correlate the formation of a well-defined β2–α2 loop

Table 1. Input for the structure calculations and characterization of the energy-minimized NMR structures of ecPrP (121–231), mPrP[D167S,N173K](121–231), and mPrP[D167S](121–231) at pH 4.5 and 25 °C Quantity

ecPrPa

mPrP[D167S,N173K]a

mPrP[D167S]a

3202 40 112 1.97 ± 0.47

2986 44 112 2.00 ± 0.44

2512 40 112 1.92 ± 0.44

30 ± 5 0.24 ± 0.09

36 ± 5 0.28 ± 0.26

42 ± 7 0.18 ± 0.10

0±0 1.20 ± 0.70

0±0 1.27 ± 1.90

0±1 1.40 ± 0.96

− 4589 ± 84 − 166 ± 17 − 5350 ± 79

− 4814 ± 78 − 302 ± 18 − 5509 ± 71

− 4832 ± 100 − 317 ± 20 − 54 ± 69

0.38 ± 0.06 0.80 ± 0.06 0.24 ± 0.05 0.69 ± 0.11

0.46 ± 0.09 0.80 ± 0.06 0.28 ± 0.12 0.65 ± 0.14

0.46 ± 0.10 0.86 ± 0.08 0.39 ± 0.13 0.77 ± 0.14

83 16 1 0

78 20 2 0

83 15 2 0

b

NOE upper-distance limits All atoms (125–228) All atoms (167–174)c Dihedral angle constraints Residual target function (Å2) Residual NOE violations Number ≥0.1 Å Maximum (Å) Residual dihedral angle violations Number ≥2.5° Maximum (deg) AMBER energies (kcal/mol) Total Van der Waals Electrostatic rmsd to the mean coordinates (Å)d bb (125–228) All heavy atoms (125–228) bb (165–172)c,d All heavy atoms (165–172)c Ramachandran plot statisticse Most favored regions (%) Additionally allowed regions (%) Generously allowed regions (%) Disallowed regions (%)

a Except for the two top entries, the average value for the 20 energy-minimized conformers with the lowest residual DYANA target function values and the standard deviation among them are given. b NMR experiments for the structure determinations were performed with uniformly 13C,15N-labeled protein samples. All proteins were expressed in E. coli and purified as described.20,21 NMR samples contained approximately 1 mM protein and 10 mM sodium acetated4 buffer in 90% H2O/10% D2O. NMR experiments were recorded on Bruker DRX500, DRX600 and Avance 900 spectrometers equipped with triply tunable room temperature or cryogenic probe heads (DRX500). Resonance assignments were obtained with 3D HNCA,22 3D 15 N-resolved 1H,1H total correlation spectroscopy (TOCSY),23 3D HCCH-TOCSY,24 2D Hβ(Cβ)(CγCδ)Hδ,25 3D 15N-resolved [1H,1H]-NOE spectroscopy (NOESY), and two 3D 13C-resolved [1H,1H]-NOESY data sets, the latter being recorded with the 13C carrier centered in the aliphatic and aromatic regions, respectively.23,26 The NOESY spectra were obtained with mixing times of 60 ms. NMR data were processed using the program PROSA27 or the software XWINNMR (Bruker, Karlsruhe, Germany). The programs XEASY28 and CARA29 (www.nmr.ch) were used for the spectral analysis. Automatic peak-picking and NOE assignment were performed using the stand-alone ATNOS/CANDID program package.30,31 Structure calculation was performed with DYANA,32 and the 20 conformers with the lowest target function values were energy-minimized in a water shell, using OPALp with the AMBER force field as described previously.33–35 The program MOLMOL was used to analyze the results of the structure calculations and to prepare drawings of the structures.36 Regular secondary-structure boundaries were determined with MOLMOL, using the method of Kabsch and Sander.37 c This polypeptide segment forms the loop linking the β2 strand with the α2 helix. d bb indicates the backbone atoms N, Cα and C′. The numbers in parentheses indicate the residues for which the rmsd was calculated. e As determined by PROCHECK.38

124

Horse Prion Protein Solution Structure

Fig. 2. Stereo views of the NMR structures of ecPrP(121–231) and two variant mouse PrPs, mPrP[D167S](121–231) and mPrP[D167S,N173K](121–231). (a) Backbone presentation for residues 125–228 of the bundle of 20 energy-minimized conformers of ecPrP(121–231) used to represent the solution structure. The conformers were aligned for best fit of the backbone N, Cα, and C′ atoms of residues 125–228, for an rmsd relative to the mean atom coordinates of 0.38 Å. The residue numbers at the start and end of the three helices and at the chain ends are indicated to aid orientation. Color code: gold for helices, green for β-strands, red for the single disulfide bond, blue for the β2–α2 loop of residues 165–172 (see the text), and gray for other nonregular secondary structures. (b) All-heavy-atom presentation of the energy-minimized conformer of ecPrP(121–231) from (a) with the lowest DYANA target function value. The backbone is represented by a gray spline function through the Cα positions. The side chains are colored according to their global heavy-atom displacements D39: cyan, D ≤ 0.6 Å; yellow, 0.6 Å b D ≤ 1.2 Å; red, D N 1.2. (c) mPrP[D167S](121–231). (d) mPrP[D167S, N173K](121–231). The same presentation as for (b) was prepared for (c) and (d) with the program MOLMOL.36

Horse Prion Protein Solution Structure

125

Fig. 3. 2D [15N,1H]-HSQC spectra of ecPrP(121–231), mPrP(121–231) and four variants of mPrP(121–231) recorded at a H frequency of 750 MHz and T = 25 °C. The uniformly 15N-labeled proteins were dissolved in 10 mM sodium acetate-d4 (pH 4.5) containing 10% D2O. The protein concentrations ranged from 0.6 to 3.2 mM, and the number of transients per t1 increment was adjusted from 2 to 26 so that each spectrum had approximately the same signal-to-noise ratio. (a) ecPrP (121–231). (b) mPrP[D167S](121–231). (c) mPrP[D167S,N173K](121–231). (d) mPrP(121–231).12 (e) mPrP[Q168E](121–231). (f) mPrP[N173K](121–231). Peak positions of residues 167–175 are highlighted, with blue circles indicating the positions of peaks visible in all six spectra, and red circles indicating the approximate peak positions for signals that are not observed in mPrP(121–231), mPrP[Q168E](121–231), and mPrP[N173K](121–231), as expected on the basis of studies with other variant mPrPs. Comparison of the spectra in (e) and (f) with (d) reveals that only the 15N–1H resonance of the altered residue has shifted significantly. The peak positions for the 15N–1H signals from E168 in (e) and K173 in (f) were taken from the observed positions of the corresponding signals in (a) and (c), respectively. In (a) to (c), the peaks from S170 and F175 are broadened, so that only weak contours are seen. 1

structure in ecPrP with the amino acid sequence, we prepared the mouse PrP variants mPrP[D167S](121– 231), mPrP[Q168E](121–231), mPrP[N173K](121– 231), mPrP[D167S,Q168E](121–231), and mPrP [D167S, N173K](121–231). The [15N,1H]-HSQC spectra at 20 °C of the proteins containing the substitution D167S included signals of residues 167–171 and 175 (Fig. 3a–c), whereas those mPrP variant proteins that do not include the substitution D167S were missing the signals of these residues (Fig. 3e and f), as was previously observed in mouse (Fig. 3d), human, cattle, and sheep PrP. This indicated that introducing the substitution D167S into mPrP might result in a well-defined structure of the β2–α2 loop at 25 °C, which was then further investigated by NMR structure determinations of these variant mPrPs. Structure determinations of mPrP[D167S](121–231) and mPrP[D167S,N173K](121–231) Resonance assignments for the two variant mPrPs were completed to the same extent as for ecPrP; in particular all resonances for the residues 167–171 and 175 could be assigned, resulting in similar statistics for the structure determinations, except that smaller numbers of NOE distance constraints were obtained because of somewhat lower protein concentrations in the NMR samples (Table 1). The globular domains of the mouse/horse hybrid PrPs

have the same overall fold as that of ecPrP(121–231) (Fig. 3c and d), and the bundles of 20 conformers are well defined, with rmsd values of 0.46 Å for the backbone N, Cα, and C′ atoms of residues 125–228 (Table 1). For both of the variant mouse PrPs, the β2–α2 loop is well defined at 25 °C (Table 1). Structure comparisons of ecPrP(121–231), mPrP[D167S](121–231), mPrP[D167S, N173K](121–231), and mPrP(121–231) Whereas the architecture of the globular domain of mPrP(121–231) is preserved in ecPrP(121–231), the β2–α2 loop comprising residues 165–172 is well defined in ecPrP(121–231) and structurally disordered in mPrP(121–231).6,12 Correspondingly, the N, Cα, and C′ coordinates for residues 165–172 in mPrP (121–231) have a local rmsd value of 0.60 Å, and in ecPrP(121–231), of 0.24 Å. In the two variant proteins mPrP[D167S](121–231) and mPrP[D167S,N173K] (121–231), the loop is also well defined, with rmsd values of 0.39 and 0.28 Å for the backbone N, Cα, and C′ atoms calculated for residues 165–172, respectively (Table 1), and it adopts the same conformation as that observed in ecPrP (Fig. 3b and c). The number of long-range upper-distance constraints for region 167–171, defined as constraints between atoms in residues separated by at least five sequence positions,40 is 44 in mPrP[D167S,N173K](121–231)

126 and 39 in mPrP[D167S](121–231), which is similar to the 40 upper-distance constraints found in ecPrP (121–231). In contrast, in mPrP(121–231) only two long-range distance constraints with residues 167– 172 were identified, involving Hɛ of Y218, and HN and Hβ2 of Q172, which explains the lack of precise structure definition of the β2–α2 loop in mPrP(121– 231) at 20 °C6,12 from the viewpoint of the structure determination procedure. Overall, these comparisons show that the single amino acid replacement of Asp167 in mPrP(121– 231) by Ser is sufficient for obtaining a well-defined conformation of the β2–α2 loop in the NMR structure at 25 °C. The other amino acid exchanges between mPrP and ecPrP appear to have at most subtle effects on the 3D structure, although they include the two loop positions 168 and 173 (Fig. 1). The less well defined loop structure in mPrP probably includes conformational exchange between two or multiple locally different conformers, which leads to the observed NMR line broadening. The observation that replacing a carboxylate group by an alcohol moiety in position 167 caused changes in the line broadening of 15N–1H cross-peaks of the β2–α2 loop would appear to indicate that the presence of a negative charge in this position might affect the frequency of segmental motions of the β2–α2 loop. Implications for TSE pathology Overall, the PrP amino acid sequences from different mammalian species show very high conservation,16,41,42 and all the available 3D structures show the same typical “PrPC fold” of the globular domain (Fig. 2).6–14 With the initial determination of the PrPC fold,6 it had been readily apparent that compared to the remainder of the

Horse Prion Protein Solution Structure

amino acid sequence, there is an outstandingly high frequency of amino acid exchanges among mammalian PrPs in the surface area formed by the β2–α2 loop and the α3 helix, which is solvent-exposed also in the glycosylated native form of PrPC.15,43 This surface epitope has independently been implicated in interactions with a hypothetical protein X, which has been proposed to mediate the conversion of PrPC to PrPSc.17,18 Subsequently, it has been found that elk and bank vole PrP (ePrP and bvPrP, respectively) contain a well-structured β2–α2 loop with structural definition comparable to that of the rest of the protein,12,13 which contrasts with the structurally disordered loop in mPrP.6,12 As mentioned in the Introduction, this local structure variation could be correlated with the amino acid substitution S170N (Fig. 1), and it attracted special interest because both elk and bank vole are highly susceptible to TSEs.44–47 A direct link between structural features of the β2–α2 loop in PrPC and TSEs then resulted from experiments with transgenic mice expressing a hybrid mouse/elk PrPC, which developed a spontaneous TSE19 (see also the Introduction). As a follow-up to the aforementioned experiments, it is of interest to investigate possible correlations between susceptibility to TSEs of elks and bank voles and either the well-defined β2–α2 loop conformation or the introduction of the asparagine side chain of residue 170. The present observations with ecPrP provide a contribution toward a better understanding of this presumed structure–function correlation: horse PrP contains not only a novel single amino acid substitution that results in a well-defined loop structure, but in contrast to elk and bank vole, there have been no cases of TSE reported in horses, in spite of their often very long lifespan. A similar situation was revealed by recent studies of tammar wallaby PrP (twPrP).

Fig. 4. Stereo views of the contact area between the β2–α2 loop and the α3 helix in the NMR structure of ecPrP(121– 231). The backbone segments 165–172 and 217–226 of ecPrP are represented by gray tubes showing a spline function fitted to the Cα coordinates, where the radius is proportional to the mean global displacement among the 20 energy-minimized conformers used to represent the NMR structure. Amino acid side chains are drawn as stick models and identified: Val166, yellow; Ser167, orange; Ser170, blue; Tyr218 and Tyr222, violet; Phe225, magenta.

Horse Prion Protein Solution Structure

There are no reports about TSEs in either tammar wallaby or equine species; in addition, twPrPC forms a well-defined β2–α2 loop. The high structural definition of the loop could be correlated with long-range effects due to the amino acid replacement Y225A in the α3 helix when compared to that of mPrP (Fig. 1). In this context, it is worth noting that relative to mPrP, horse PrP also includes the substitutions Y225F and Y226Q (Fig. 1). In the [15N,1H] correlation NMR spectra of the mPrP (121–231) variants carrying either the substitution Y225F or the substitution Y226A, the resonances of residues 167–171 and 175 were not observed at 20 °C, similar to that of mPrP(121–231) (Fig. 3d), indicating that the structural definition of the β2–α2 loop is not visibly affected by either of these substitutions (unpublished data and Ref. 14). The observations on the β2–α2 loop structure in horse and wallaby PrPC combined with the TSE history of these species thus appear to support the notion that the nature of the asparagine side chain in the residue position 170 plays a major role in determining susceptibility of bank voles and elks to TSEs. It should also be pointed out that the critical single amino acid exchange D167S in ecPrP is located in the contact area between the β2–α2 loop and the α3 helix (Fig. 4), whereas the crucial position 170 in ePrP and bvPrP is spatially separated from α3 and readily accessible on the protein surface.12,13 Data bank accession codes The resonance assignments of ecPrP(121–231) and the two variant mouse proteins mPrP[D167S](121– 231) and mPrP[D167S,N173K](121–231) have been deposited with the BioMagResBank‡ with entry codes 16720, 16722, and 16723, respectively. The structure coordinates have been deposited at the RCSB§ with accession codes 2KU4, 2KU5, and 2KU6, respectively.

Acknowledgements We thank Dr. Barbara Christen for helpful discussions and Dr. Pedro Serrano for help with structure refinement. Financial support by the Swiss National Science Foundation and the ETH Zurich through the National Center of Competence in Research (NCCR) “Structural Biology” is gratefully acknowledged.

References 1. Collinge, J. (2001). Prion diseases of humans and animals: their causes and molecular basis. Annu. Rev. Neurosci. 24, 519–550.

‡ http://www.bmrb.wisc.edu § http://www.pdb.org

127 2. Prusiner, S. B. (1998). Prions. Proc. Natl Acad. Sci. USA, 95, 13363–13383. 3. Weissmann, C., Enari, M., Klohn, P. C., Rossi, D. & Flechsig, E. (2002). Transmission of prions. Proc. Natl Acad. Sci. USA, 99, 16378–16383. 4. Moore, R. A., Vorberg, I. & Priola, S. A. (2005). Species barriers in prion diseases—brief review. Arch. Virol. Suppl. 19, 187–202. 5. Hill, A. F. & Collinge, J. (2002). Species-barrierindependent prion replication in apparently resistant species. APMIS, 110, 44–53. 6. Riek, R., Hornemann, S., Wider, G., Billeter, M., Glockshuber, R. & Wüthrich, K. (1996). NMR structure of the mouse prion protein domain PrP(121–231). Nature, 382, 180–182. 7. James, T. L., Liu, H., Ulyanov, N. B., Farr-Jones, S., Zhang, H., Donne, D. G. et al. (1997). Solution structure of a 142-residue recombinant prion protein corresponding to the infectious fragment of the scrapie isoform. Proc. Natl Acad. Sci. USA, 94, 10086–10091. 8. Liu, H., Farr-Jones, S., Ulyanov, N. B., Llinas, M., Marqusee, S., Groth, D. et al. (1999). Solution structure of Syrian hamster prion protein rPrP(90–231). Biochemistry, 38, 5362–5377. 9. Zahn, R., Liu, A., Lührs, T., Riek, R., von Schroetter, C., López-García, F. et al. (2000). NMR solution structure of the human prion protein. Proc. Natl Acad. Sci. USA, 97, 145–150. 10. López-García, F., Zahn, R., Riek, R. & Wüthrich, K. (2000). NMR structure of the bovine prion protein. Proc. Natl Acad. Sci. USA, 97, 8334–8339. 11. Lysek, D. A., Schorn, C., Nivon, L. G., Esteve-Moya, V., Christen, B., Calzolai, L. et al. (2005). Prion protein NMR structures of cats, dogs, pigs, and sheep. Proc. Natl Acad. Sci. USA, 102, 640–645. 12. Gossert, A. D., Bonjour, S., Lysek, D. A., Fiorito, F. & Wüthrich, K. (2005). Prion protein NMR structures of elk and of mouse/elk hybrids. Proc. Natl Acad. Sci. USA, 102, 646–650. 13. Christen, B., Pérez, D. R., Hornemann, S. & Wüthrich, K. (2008). NMR structure of the bank vole prion protein at 20 °C contains a structured loop of residues 165–171. J. Mol. Biol. 383, 306–312. 14. Christen, B., Hornemann, S., Damberger, F. F. & Wüthrich, K. (2009). Prion protein NMR structure from tammar wallaby (Macropus eugenii) shows that the beta 2-alpha 2 loop is modulated by long-range sequence effects. J. Mol. Biol. 389, 833–845. 15. Hornemann, S., Schorn, C. & Wüthrich, K. (2004). NMR structure of the bovine prion protein isolated from healthy calf brains. EMBO Rep. 5, 1159–1164. 16. Schätzl, H. M., Da Costa, M., Taylor, L., Cohen, F. E. & Prusiner, S. B. (1995). Prion protein gene variation among primates. J. Mol. Biol. 245, 362–374. 17. Kaneko, K., Zulianello, L., Scott, M., Cooper, C. M., Wallace, A. C., James, T. L. et al. (1997). Evidence for protein X binding to a discontinuous epitope on the cellular prion protein during scrapie prion propagation. Proc. Natl Acad. Sci. USA, 94, 10069–10074. 18. Telling, G. C., Scott, M., Mastrianni, J., Gabizon, R., Torchia, M., Cohen, F. E. et al. (1995). Prion propagation in mice expressing human and chimeric PrP transgenes implicates the interaction of cellular PrP with another protein. Cell, 83, 79–90. 19. Sigurdson, C. J., Nilsson, K. P., Hornemann, S., Heikenwälder, M., Manco, G., Schwarz, P. et al. (2009). De novo generation of a transmissible spongiform encephalopathy by mouse transgenesis. Proc. Natl Acad. Sci. USA, 106, 304–309.

128 20. Zahn, R., von Schroetter, C. & Wüthrich, K. (1997). Human prion proteins expressed in Escherichia coli and purified by high-affinity column refolding. FEBS Lett. 417, 400–404. 21. Lysek, D. A. & Wüthrich, K. (2004). Prion protein interaction with the C-terminal SH3 domain of Grb2 studied using NMR and optical spectroscopy. Biochemistry, 43, 10393–10399. 22. Kay, L. E., Ikura, M., Tschudin, R. & Bax, A. (1990). Three-dimensional triple-resonance NMR spectroscopy of isotopically enriched proteins. J. Magn. Reson. 89, 496–514. 23. Fesik, S. W. & Zuiderweg, E. R. P. (1988). Heteronuclear three-dimensional NMR spectroscopy—a strategy for the simplification of homonuclear twodimensional NMR spectra. J. Magn. Reson. 78, 588–593. 24. Bax, A., Clore, G. M. & Gronenborn, A. M. (1990). 1 H–1H correlation via isotropic mixing of 13C magnetization, a new three-dimensional approach for assigning 1H and 13C spectra of 13C-enriched proteins. J. Magn. Reson. 88, 425–431. 25. Yamazaki, T., Forman-Kay, J. D. & Kay, L. E. (1993). Two-dimensional NMR experiments for correlating 13 Cβ and 1Hδ/ɛ chemical shifts of aromatic residues in 13C-labeled proteins via scalar couplings. J. Am. Chem. Soc. 115, 11054–11055. 26. Marion, D., Kay, L. E., Sparks, S. W., Torchia, D. A. & Bax, A. (1989). Three-dimensional heteronuclear NMR of N-15-Labeled Proteins. J. Am. Chem. Soc. 111, 1515–1517. 27. Güntert, P., Dötsch, V., Wider, G. & Wüthrich, K. (1992). Processing of multidimensional NMR data with the new software PROSA. J. Biomol. NMR, 2, 619–629. 28. Bartels, C., Xia, T. H., Billeter, M., Güntert, P. & Wüthrich, K. (1995). The program XEASY for computer-supported NMR spectral-analysis of biological macromolecules. J. Biomol. NMR, 6, 1–10. 29. Keller, R. (2004). The Computer-Aided Resonance Assignment Tutorial CARA. Cantina Verlag, Goldau, Switzerland. 30. Herrmann, T., Güntert, P. & Wüthrich, K. (2002). Protein NMR structure determination with automated NOE-identification in the NOESY spectra using the new software ATNOS. J. Biomol. NMR, 24, 171–189. 31. Herrmann, T., Güntert, P. & Wüthrich, K. (2002). Protein NMR structure determination with automated NOE assignment using the new software CANDID and the torsion angle dynamics algorithm DYANA. J. Mol. Biol. 319, 209–227. 32. Güntert, P., Mumenthaler, C. & Wüthrich, K. (1997). Torsion angle dynamics for NMR structure calculation with the new program DYANA. J. Mol. Biol. 273, 283–298. 33. Luginbühl, P., Güntert, P., Billeter, M. & Wüthrich, K. (1996). The new program OPAL for molecular dynamics simulations and energy refinements of biological macromolecules. J. Biomol. NMR, 8, 136–146.

Horse Prion Protein Solution Structure 34. Koradi, R., Billeter, M. & Güntert, P. (2000). Pointcentered domain decomposition for parallel molecular dynamics simulation. Comput. Phys. Commun. 124, 139–147. 35. Cornell, W. D., Cieplak, P., Bayly, C. I., Gould, I. R., Merz, K. M., Jr., Ferguson, D. M. et al. (1995). A second generation force field for the simulation of proteins, nucleic acids, and organic molecules. J. Am. Chem. Soc. 117, 5179–5197. 36. Koradi, R., Billeter, M. & Wüthrich, K. (1996). MOLMOL: a program for display and analysis of macromolecular structures. J. Mol. Graphics, 14, 51–55. 37. Kabsch, W. & Sander, C. (1983). Dictionary of protein secondary structure: pattern recognition of hydrogenbonded and geometrical features. Biopolymers, 22, 2577–2637. 38. Morris, A. L., MacArthur, M. W., Hutchinson, E. G. & Thornton, J. M. (1992). Stereochemical quality of protein structure coordinates. Proteins, 12, 345–364. 39. Billeter, M., Kline, A. D., Braun, W., Huber, R. & Wüthrich, K. (1989). Comparison of the high-resolution structures of the alpha-amylase inhibitor tendamistat determined by nuclear magnetic resonance in solution and by X-ray diffraction in single crystals. J. Mol. Biol. 206, 677–687. 40. Wüthrich, K. (1986). NMR of Proteins and Nucleic Acids. Wiley, New York. 41. Wopfner, F., Weidenhöfer, G., Schneider, R., von Brunn, A., Gilch, S., Schwarz, T. F. et al. (1999). Analysis of 27 mammalian and 9 avian PrPs reveals high conservation of flexible regions of the prion protein. J. Mol. Biol. 289, 1163–1178. 42. van Rheede, T., Smolenaars, M. M., Madsen, O. & de Jong, W. W. (2003). Molecular evolution of the mammalian prion protein. Mol. Biol. Evol. 20, 111–121. 43. Billeter, M., Riek, R., Wider, G., Hornemann, S., Glockshuber, R. & Wüthrich, K. (1997). Prion protein NMR structure and species barrier for prion diseases. Proc. Natl Acad. Sci. USA, 94, 7281–7285. 44. Agrimi, U., Nonno, R., Dell'Omo, G., Di Bari, M. A., Conte, M., Chiappini, B. et al. (2008). Prion protein amino acid determinants of differential susceptibility and molecular feature of prion strains in mice and voles. PLoS Pathog. 4, e1000113. 45. Nonno, R., Di Bari, M. A., Cardone, F., Vaccari, G., Fazzi, P., Dell'Omo, G. et al. (2006). Efficient transmission and characterization of Creutzfeldt–Jakob disease strains in bank voles. PLoS Pathog. 2, e12. 46. Piening, N., Nonno, R., Di Bari, M., Walter, S., Windl, O., Agrimi, U. et al. (2006). Conversion efficiency of bank vole prion protein in vitro is determined by residues 155 and 170, but does not correlate with the high susceptibility of bank voles to sheep scrapie in vivo. J. Biol. Chem. 281, 9373–9384. 47. Williams, E. S. (2005). Chronic wasting disease. Vet. Pathol. 42, 530–549.