Deciphering Copper Coordination in the Mammalian Prion Protein Amyloidogenic Domain

Journal Pre-proof Deciphering copper coordination in the mammalian prion protein amyloidogenic domain Giulia Salzano, Martha Brennich, Giordano Mancini, Thanh Hoa Tran, Giuseppe Legname, Paola D’Angelo, Gabriele Giachin PII:

S0006-3495(19)34425-X

DOI:

https://doi.org/10.1016/j.bpj.2019.12.025

Reference:

BPJ 10237

To appear in:

Biophysical Journal

Received Date: 31 May 2019 Accepted Date: 23 December 2019

Please cite this article as: Salzano G, Brennich M, Mancini G, Tran TH, Legname G, D’Angelo P, Giachin G, Deciphering copper coordination in the mammalian prion protein amyloidogenic domain, Biophysical Journal (2020), doi: https://doi.org/10.1016/j.bpj.2019.12.025. This is a PDF file of an article that has undergone enhancements after acceptance, such as the addition of a cover page and metadata, and formatting for readability, but it is not yet the definitive version of record. This version will undergo additional copyediting, typesetting and review before it is published in its final form, but we are providing this version to give early visibility of the article. Please note that, during the production process, errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain. © 2020 Biophysical Society.

1

Deciphering copper coordination in the mammalian prion protein amyloidogenic domain

2 3

Giulia Salzano1, Martha Brennich2, Giordano Mancini3,4, Thanh Hoa Tran1, Giuseppe

4

Legname1,5, Paola D’Angelo6* and Gabriele Giachin7*

5 6

1

7

Trieste, Italy

8

2

European Molecular Biology Laboratory (EMBL), Grenoble Outstation, Grenoble, France

9

3

Scuola Normale Superiore, Pisa, Italy

10

4

Istituto Nazionale di Fisica Nucleare (INFN), Pisa, Italy

11

5

ELETTRA-Sincrotrone Trieste S.C.p.A, Trieste, Italy

12

6

Department of Chemistry, Sapienza University of Rome, Rome, Italy

13

7

European Synchrotron Radiation Facility (ESRF), Grenoble, France

Department of Neuroscience, Scuola Internazionale Superiore di Studi Avanzati (SISSA),

14 15

*

16

([email protected]).

Correspondence

should

be

addressed

to

17

1

P.D.

([email protected])

and

G.G.

18

ABSTRACT

19

Prions are pathological isoforms of the cellular prion protein (PrPC) responsible for transmissible

20

spongiform encephalopathies (TSE). PrPC interacts with copper, Cu(II), through octarepeat and

21

non-octarepeat (non-OR) binding sites. The molecular details of Cu(II) coordination within the

22

non-OR region are not well characterized yet. By means of small angle X-ray scattering (SAXS)

23

and X-ray absorption spectroscopic methods (XAS) we have investigated the effect of Cu(II) on

24

prion protein folding and its coordination geometries when bound to the non-OR region of

25

recombinant prion proteins (recPrP) from mammalian species considered resistant or susceptible

26

to TSE. As prion resistant model, we used ovine recPrP (OvPrP) carrying the protective

27

polymorphism at residues A136, R154 and R171 (ARR); while as TSE-susceptible models we

28

employed OvPrP with V136, R154 and Q171 (VRQ) polymorphism and Bank vole recPrP

29

(BvPrP). Our analysis reveals that Cu(II) affects the structural plasticity of the non-OR region

30

leading to a more compacted conformation. We then identified two Cu(II) coordination

31

geometries: in the type-1 coordination observed in OvPrP ARR the metal is coordinated by four

32

residues; conversely, the type-2 coordination is present in OvPrP VRQ and BvPrP, where Cu(II)

33

is coordinated by three residues and by one water molecule, making the non-OR region more

34

exposed to the solvent. These changes in copper coordination affect the recPrP amyloid

35

aggregation. This study may provide new insights into the molecular mechanisms governing the

36

resistance or susceptibility of certain species to TSE.

37 38

STATEMENT OF SIGNIFICANCE

39

The C-terminal domain of the prion protein from residues
90 to 230 is involved in the

40

pathological conversion to prions (PrPSc). Copper binds to this segment through the fifth copper

41

binding site but its role in the structural changes leading to PrPSc is unclear. By means of

42

synchrotron-based biophysical techniques we have demonstrated that copper affects the

43

structural plasticity of the protein C-terminal domain leading to a more compacted conformation.

44

We have identified two copper coordination geometries: type-I, or “closed”, for prion resistant

45

PrPC and type-2, or “open”, for prion susceptible PrPC. We provide novel arguments showing

46

that changes in the fifth copper binding site conformation dictate the lower or higher

47

susceptibility of PrPC to prion conversion.

48

2

49

INTRODUCTION

50

Prion diseases or transmissible spongiform encephalopathies (TSE) are neurodegenerative

51

disorders affecting humans and animals, and are caused by the misfolding of the α-helical form

52

of the physiological cellular prion protein (PrPC) into a β-sheet enriched isoform called prion or

53

PrPSc [1]. TSE are rare disorders that can be sporadic, genetic or infectious. Animal TSE include

54

scrapie in sheep and goats, chronic wasting disease in cervids and bovine spongiform

55

encephalopathy in cattle [2].

56

The PrPC structure consists in a C-terminal domain (residues 128 to 231, in human numbering)

57

containing three α-helices and two short anti-parallel β-sheets. The N-terminal moiety (residues

58

23-127) is unstructured [3] and features an octapeptide-repeat region (OR) (residues 61-91)

59

composed by four octapeptides, each carrying histidine residues able to coordinate one copper

60

ion, Cu(II) [4]. Cu(II) can also bind with high affinity (i.e. the affinity constant is of order 108-

61

1010 M-1 but varies with the methods) at two additional histidine residues, H96 and H111, with

62

coordination from the imidazole rings and nearby backbone amides and are located in a segment

63

(residues 90-111) called “fifth” or non-OR copper binding site [5-8] (Figure 1 A). Adjacent to

64

the non-OR region is the palindromic motif of sequence “AGAAAAGA” (residues 113-120)

65

known to be able to initiate prion conversion [9]. Although PrPC and PrPSc share identical

66

primary sequence, they have distinct physicochemical and structural features: PrPC exists as

67

detergent-soluble α-helical monomer, whereas PrPSc forms detergent-insoluble amyloidal

68

aggregates that are structured as a 4-rung β-solenoid [10]. Following treatment with proteinase-K

69

PrPSc typically generates a partially protease-resistant core, which is N-terminally truncated at

70

around residue 78 [11] and it is sufficient to support prion replication [12].

71

The functional implications of Cu(II)-binding to PrPC are not unequivocal. Compelling evidences

72

propose a Cu(II)-mediated neuroprotective role for PrPC as modulator of synaptic plasticity and

73

S-nitrosylation [13]; some others point out the role of Cu(II) as promoter or attenuator of prion

74

conversion [14-16]. The proximity of the non-OR region to the palindromic motif suggests a link

75

between Cu(II) binding and prion conversion [17]. We have reported that pathogenic mutations

76

affect Cu(II) coordination in the non-OR region and this altered coordination promotes prion

77

conversion in vitro and in cell models [18, 19]. The involvement of H96 and H111 in Cu(II)

78

binding shows that the metal occupancy plays a role in determining the conformation of this

79

section inducing local conformational changes and possible interactions with the C-terminal

80

domains [14]. Given the importance of the C-terminal region for prion propagation and the 3

81

controversial role of Cu(II) as attenuator or facilitator of TSE, further studies on the Cu(II)-non-

82

OR interactions are of pivotal importance to clarify the structural and functional consequences of

83

Cu(II) binding to the PrPC amyloidogenic domain.

84

Here, we set out to investigate the Cu(II) structural effects and its coordination geometries when

85

bound to the non-OR region of recombinant prion proteins (recPrP) from mammalian species

86

considered susceptible or resistant to TSE by means of small angle X-ray scattering (SAXS),

87

extended X-ray adsorption fine structure (EXAFS), X-ray absorption near edge structure

88

(XANES) spectroscopies and Thioflavin T (ThT)-fluorescence fibrillization assays. As TSE-

89

resistant model, we used the C-terminal form of ovine recPrP carrying the protective A136, R154

90

and R171 (OvPrP ARR) polymorphism; while as TSE-susceptible recPrP models we employed

91

OvPrP with polymorphism at residues V136, R154 and Q171 (OvPrP VRQ) and Bank vole

92

recPrP (BvPrP) (Figure S1) [20, 21].

93

Our results reveal that Cu(II) affects the structural plasticity of the non-OR region leading to a

94

more compacted conformation of recPrP. We also observe that the Cu(II) coordination changes

95

in the recPrP of TSE-resistant and susceptible species. These changes in copper coordination

96

may have important physiological implications, providing new insights into the molecular

97

mechanisms governing the resistance or susceptibility of certain species to TSE.

98 99

MATERIALS AND METHODS

100

Protein expression and purification

101

The pET-11a plasmid (Novagen) encoding for BvPrP (residues 90-231), OvPrP VRQ (residues

102

94-234, carrying the TSE-susceptible V136, R154 and Q171 polymorphism) and OvPrP ARR

103

(residues 94-234, carrying the TSE-resistant A136, R154 and R171 polymorphism) were

104

purchased from Genewiz (Germany, GmbH). All the recPrP were expressed, purified and

105

refolded according to our previous protocols [18].

106 107

SEC-SAXS measurements

108

All the experiments were performed at the ESRF bioSAXS beamline BM29 in Grenoble, France

109

[22]. Given the sensitivity for batch SAXS mode for even small amounts of soluble aggregates

110

we used SEC-SAXS approaches to measure SAXS data on monodisperse samples. A volume of

111

250 µL of protein per each BvPrP sample (apo and copper-loaded) at 12 mg/mL was loaded on a

112

GE Superdex 75 10/300 GL column. Due to the limited expression and purification yields, a 4

113

volume of 45 µL protein per each OvPrP ARR and OvPrP VRQ (apo and copper-loaded)

114

samples at 7 mg/mL was loaded on a GE Superdex 200 5/150 GL column. All the recPrP have

115

been purified using HPLC system (DGU-20A5R, Shimadzu) attached directly to the sample-inlet

116

valve of the sample changer [23]. All the apo samples were measured in Buffer A (25 mM

117

Sodium Acetate, 250 mM NaCl, pH 5.5) at 20 °C. Cu(II) loading on recPrP samples was

118

achieved by dialysis (Spectra/Por 3.5 kDa MWCO membrane) against Buffer A containing

119

CuSO4 at 1:1 Cu(II): recPrP molar ratio at 4 °C for 12 hours and then against Buffer B (25 mM

120

Sodium Acetate, 250 mM NaCl, 1 µM of CuSO4, pH 5.5) to remove the excess of metal at 4 °C

121

for 4 hours. After centrifugation (30 minutes, 16,000 g at 4 °C), Cu(II)-samples were then

122

measured in Buffer B at 20 °C. The columns were equilibrated with 3 column volumes to obtain

123

a stable background signal. All SAXS data were collected at a wavelength of 0.99 Å using a

124

sample-to-detector (PILATUS 1 M, DECTRIS) distance of 2.81  m. The scattering of pure water

125

was used to calibrate the intensity to absolute units [24].

126 127

SAXS data analysis and modelling

128

Data reduction was performed automatically using the EDNA pipeline [25]. Frames in regions of

129

stable Rg were selected and averaged using PRIMUS [26] to yield a single averaged frame

130

corresponding to the scattering of individual SEC species. All parameters for SAXS analysis are

131

described in Table S1 according to SAXS community guidelines [27]. Analysis of the overall

132

parameters was carried out by PRIMUS from ATSAS 2.8.4 package [26] and by ScÅtter 3.0

133

software [28]. The pair distance distribution function, P(r), and maximum diameter of the

134

particle (Dmax) were calculated in GNOM using indirect Fourier Transform (FT) method [29].

135

Protein molecular masses were estimated using both Porod volume [26] and scattering mass

136

contrast [27] methods. For low-resolution structural models, Ensemble Optimization Method

137

(EOM) analysis was conducted using 3-dimensional (3D) models of BvPrP (amino acids 136-

138

226, PDB id 2K56) and OvPrP (amino acids 133-226, PDB id 1TQC) as rigid bodies and the rest

139

of the protein (corresponding to the N-terminal unfolded region) was modeled by EOM and

140

represented as edged lines [30]. The missing residue connecting the rigid bodies and the EOM

141

models were added with MODELLER [31]. An initial random pool of 10,000 models was

142

generated in RanCh (version 2.1) [32]; final ensembles were selected from the starting pool

143

using a genetic algorithm implemented in GAJOE (version 2.1) [32]. The quality of EOM fits

144

over modelled q-range of experimental SAXS curves were assessed with CORMAP analysis [26]

145

and with error-weighted residual difference plots of [Iexp(q) -Icomp(q)]/σexp(q), corresponding to 5

146

the difference between the experimental and computed intensities weighted by the experimental

147

uncertainty [27]. Eight independent EOM internal replica were carried out per each apo and

148

Cu(II)-recPrP SAXS curves and the distributions of selected Rg and Dmax ensembles were

149

averaged. EOM parameters are detailed in Table S1, section e. All the plots and 3D recPrP

150

models were generated using OrginPro 9.0 and UCSF Chimera software, respectively. SAXS

151

data were deposited into SASBDB data bank (Table S1, section f).

152 153

XAS spectra measurements

154

Protein samples were concentrated at
1 mM in Buffer C (25 mM Sodium Acetate, pH 5.5)

155

using Amicon Ultra 10 kDa cut-off filter units. The recPrP samples were first dialyzed

156

(Spectra/Por® 3.5 kDa MWCO membrane) against Buffer D (25 mM Sodium Acetate, 1 mM of

157

CuSO4, pH 5.5) for 4 hours (for copper loading we used a CuSO4:recPrP molar ratio of 1:1) and

158

then against Buffer E (25 mM Sodium Acetate, 1 µM of CuSO4, pH 5.5) for 4 hours to remove

159

the excess of unbound metal. Sample monodispersity was assessed by SEC using a GE Superdex

160

200 Increase 10/300 GL column equilibrated in Buffer E. Cu(II) loading on concentrated recPrP

161

was performed the day before XAS measurements at ESRF BM30A sample preparation

162

laboratory and then a sample volume of 50 µL per each Cu(II)-recPrP was immediately flash

163

frozen with liquid nitrogen in the sample holder.

164

X-ray absorption spectra were recorded at the ESRF BM30B (FAME) beamline [33]. The

165

spectra were collected at the Cu K-edge in fluorescence mode using a solid state 30-element Ge

166

detector, with sample orientation at 45° to incident beam. The X-ray photon beam was vertically

167

focused by a Ni−Pt mirror, dynamically and sagittally focused in the horizontal size. The

168

monochromator was equipped with a Si(111) double crystal in which the second crystal was

169

elastically bent to a cylindrical cross section. The energy resolution at the Cu K-edge is 0.5 eV.

170

The spectra were calibrated by assigning the first inflection point of the Cu foil spectrum to 8981

171

eV. All the spectra were collected at 10 K. For Cu(II)-samples, photo reduction is typically

172

observed and thus the beam was moved to different spots of the sample at each scan. During

173

collection, data were continuously monitored in order to insure sample homogeneity across the

174

multiple spots collected from different sample-holder cells. The following samples were

175

measured: Cu(II)-OvPrP ARR, Cu(II)-OvPrP VRQ and Cu(II)-BvPrP. For each sample, 12

176

spectra were recorded with a 1 s/point collection statistic and averaged. The collection time was

177

25 min for each spectrum.

178 6

179

XAS data analysis

180

The analysis of the XAS spectra was carried out starting from the coordination models reported

181

in the literature for the HuPrP wild-type (WT) and HuPrP Q212P proteins [5, 18, 19], and

182

considering the amino acid sequences of the species (Figure S1). In particular, the analysis of

183

the OvPrP ARR resistant specie has been carried out considering the coordination with a

184

nitrogen atom of H114, with two oxygen atoms of S98 that chelates the Cu(II) ion forming a ring

185

in the equatorial plane [34, 35], with an oxygen atom of Q101 and a sulfur atom of M112. The

186

XAS spectra of the more susceptible OvPrP VRQ and BvPrP species have been analyzed using

187

the same model as HuPrP Q212P where the Cu(II) ion is coordinated by H111, Q98, M109 and a

188

water molecule [18].

189

The analysis of the XANES data was carried out with the MXAN code. The X-ray

190

photoabsorption cross-section was calculated using the full multiple scattering (MS) scheme in

191

the framework of the muffin-tin (MT) approximation for the shape of the potential [36]. The real

192

part of the complex exchange and correlation energy was calculated by using the HL energy-

193

dependent potential. Inelastic processes were accounted by convolution with a broadening

194

Lorentzian function having an energy-dependent width of the form Γ(Ε)=Γc+Γmfp(Ε). The

195

constant part Γc accounts for the core-hole lifetime and it was fixed to the tabulated value of 1.55

196

eV FWHM, while the energy-dependent term Γmfp(Ε) represents all the intrinsic and extrinsic

197

inelastic processes. Atomic MT radii were obtained using the Norman criterion including

198

overlap between MT spheres that is optimized by fitting the ovlp parameter (only one parameter

199

for the entire set of MT radii). Least-squares fits of the XANES experimental data were

200

performed by minimizing the Rsq function defined as:

201

Rsq

1 = mε 2

∑ [(y i =1

)ε ]

2

m

th i

−y

exp i

−1 i

202

where m is the number of experimental points, yith and yiexp are the theoretical and experimental

203

values of the absorption cross-section; the error εi is constant and equal to 0.5% of the

204

experimental jump. Minimization of the Rsq function was performed in the space of four

205

structural parameters for the OvPrP ARR protein: the Cu−NHis distance, the Cu-O distances of

206

the S98 residue, the Cu-O distance of Q101 and the Cu-S distance of M112. The whole histidine

207

molecule was included and it was treated as perfectly rigid body. The Cu(II) ion forms a chelate

208

with S98 and a single Cu-O distance was minimized while the geometry of the serine molecule

209

was kept fixed. Additional geometries not including the M112 residue and including a water 7

210

molecule were also tested. In the case of the OvPrP VRQ and BvPrP species, different models

211

were tried out to get the best-fit model. A k-weight fitting procedure was followed in all the

212

XANES analysis.

213 214

Monitoring the kinetics of fibril formation in vitro

215

To monitor the formation of ThT-positive fibrils, we used OvPrP (ARR and VRQ) and BvPrP

216

stored in 25 mM Sodium Acetate, 5 M GndHCl, pH 5.5 at approximately 0.8 mg/mL. ThT is a

217

small molecule that gives strong fluorescence upon binding to amyloids. To induce protein

218

polymerization we added a preformed seed to the reaction [37]. Sheep brain homogenate (20%

219

w/v, VRQ/VRQ genotype) was kindly provided by Prof. Olivier Andreoletti (École Nationale

220

Vétérinaire de Toulouse, France) and used for phosphortungstic acid (PTA) precipitation. PTA

221

precipitation was carried out adding 500 µL of PBS buffer containing 4% sarkosyl, protease

222

inhibitor (Roche) and 0.5% PTA, with constant shaking at 350 rpm, at 37 °C for 1 hour and

223

centrifuged at 16,000 g for 30 minutes at room temperature (RT). The pellet was washed with

224

500 µL washing buffer (PBS, 2% Sarkosyl protease inhibitor) and centrifuged at 16,000 g for 30

225

min at RT. After that, the pellet was resuspended in 150 µL of sterile distilled water and stored at

226

−80 °C until use. One µL of resuspended pellet was diluted into 200 µL of water and 10 µL of

227

the diluted sample was added to each well (96-well plates, BD Falcon) containing 9.6 µM of

228

recPrP, 9.6 µM CuSO4 (i.e. 1:1 recPrP:CuSO4 molar ratio) 10 µM ThT, 25 mM Sodium Acetate

229

pH 5.5, 2 M GndHCl and one 3-mm glass bead (Sigma-Aldrich) in a final reaction volume of

230

200 µL. The plate was continuous shaking at 37 °C on Spectramax M5 (Molecular Device) plate

231

reader. The kinetics of fibril formation was monitored for 65 hours by reading fluorescence

232

intensity every 10 min at 444 nm excitation and 485 nm emission. The lag phase (i.e. the region

233

before the transition zone) was estimated based on 10% of the ThT-fluorescence increase over

234

the baseline as previously reported [38].

235 236

RESULTS

237

Mammalian prion proteins undergo compactness changes in presence of copper

238

The structural consequences of Cu(II) on PrPC were previously investigated by SAXS on full-

239

length murine recPrP (MoPrP) reporting a global compaction of the protein upon metal binding

240

to the eight tandem repeat region [14]. Here, we investigated by SAXS the Cu(II) role on the

241

structures of recPrP carrying only the non-OR region as metal binding site. High affinity Cu(II) 8

242

binding to the C-terminal MoPrP (residues 90-231) has been previously observed at both neutral

243

and acidic pH conditions (affinity constants of the order of 108 and 106 M-1, respectively) using

244

biophysical methods [8], suggesting that our copper loading at 1:1 molar ratio is sufficient to

245

generate Cu(II)-chelated proteins.

246

The Rg, I(0) and UV traces as functions of frames show that proteins were well separated in

247

individual peaks (Figure S2) and monodispersed after Cu(II) loading (Figure S3). Data frames

248

under each of the main elution peaks (for which the Rg values were the same within error and

249

statistically indistinguishable as assessed using CORMAP) were selected and averaged for

250

further analysis. Primary data analysis from scattering curves showed that both apo and Cu(II)-

251

recPrP are similar with an elongated and flexible shape (Figure 1, B-D) as previously observed

252

for the apo and Cu(II)-loaded full-length MoPrP [14]. Similar conclusions regarding flexibility

253

can be drawn from Kratky plots (Figure S4). The Rg of recPrP determined by Guinier analysis

254

showed small differences between apo and Cu(II)-bound proteins (Figure 1, B-D insets). In

255

particular, the calculated Rg of apo BvPrP (2.42 nm), OvPrP ARR (2.31 nm) and OvPrP VRQ

256

(2.31 nm) are similar within the standard deviation to the Rg of the same Cu(II)-bound recPrP

257

(2.36 nm, 2.25 nm and 2.26 nm, respectively) (Table S1, section c). For comparison, the Rg of

258

BvPrP(119-231) and OvPrP(119-231) calculated from the solution NMR structures [39, 40] are

259

1.65 and 1.69 nm, respectively; however, these structures lack atomic coordinates for residues 90

260

to 118.

261

Overall, the molecular dimensions observed for the apo recPrP are in agreement with previous

262

SEC-SAXS studies on MoPrP(89-230) [41]. Distance distribution function, P(r) analysis,

263

revealed reduction in the Dmax values from ~9.4 nm for apo recPrP to ~8.6 nm for Cu(II)-recPrP.

264

For all the proteins, the Rg and I(0)-based mass values were in agreement with the expected

265

monomeric recPrP(90-231) molecular weight (i.e. ~16 kDa, Table S1, section c).

266

The N-terminal region (residues
90-127) of recPrP is largely flexible in solution. Hence, we

267

analyzed the data using EOM, which gives useful information, such as Rg and Dmax distributions,

268

in case of proteins with flexible domains. The EOM analysis was performed on SAXS curves

269

where the q-range region was fixed from 0.1 to 3.5 nm-1 and using available high-resolution

270

C-terminal OvPrP and BvPrP structures (see Table S1, section e). We found that these entry

271

parameters yielded good quality fits as visible in the log I(q) versus q plots of the EOM curves

272

overlaid with the experimental data and in the error-weighted residual difference plots (Figure 2,

273

A-C). 9

274

Size distributions (Rg and Dmax) of apo versus Cu(II)-recPrP provided qualitative assessment on

275

the structural effect of metal to protein compactness through direct comparison of the

276

multimodal distributions of the selected ensembles and the pools (Figure 2, D-F and Figure S5,

277

for Rg and Dmax plots, respectively). In all the apo-proteins, bimodal Rg and Dmax ensemble

278

distributions were observed with subpopulations corresponding to compact and extended

279

conformers. The first and most populated one corresponds to ensembles of Rg
2.2 nm for

280

OvPrP (ARR and VRQ) and of Rg
2.4 nm for BvPrP, while the second fractions reflect to very

281

extended conformational states. A similar bimodal distribution was observed in other

282

intrinsically disordered proteins [42, 43].

283

Interestingly, bimodality is perturbed with different extent after Cu(II) binding to recPrP. In

284

Cu(II)-OvPrP ARR the multi-Gaussian fit on the Rg distribution shows the absence of extended

285

conformers at high Rg values and the presence of multiple subpopulations with differently

286

compacted cores (Figure 2 D). In Cu(II)-OvPrP VRQ the bimodal size distribution is present,

287

but the binding to the metal increases the population of conformers consisting of compact

288

structures, which display a shift in size compared to the apo form (from Rg
2.2 to
2.1 nm)

289

and reduces the fraction of the extended one (Figure 2 E). Size distribution of Cu(II)-BvPrP

290

indicates a prediction of multiple protein conformations. Shifts in size parameters are present in

291

both the highly represented compact (from Rg
2.3 to
2 nm) and less represented extended

292

(from Rg
4 to
3.5 nm) conformers; notably, an intermediate population is present as visible in

293

the shoulder at Rg
2.5 nm (Figure 2 F). The effect of copper in the recPrP compaction is

294

visible also in Figure 2 G where the frequency-weighted Rg and Dmax average values of the

295

representative 3D structures generated by EOM are represented.

296

Previous studies interpreted the reduction in the Rg and Dmax, leading to the global compactness

297

of full-length Cu(II)-MoPrP, as result of a decrease in the flexibility of the N-terminal region,

298

which exhibits interaction with the C-terminal globular domain upon metal binding [14]. Here,

299

we quantitatively measured the protein flexibility using two metrics, Rflex and Rσ, available in the

300

EOM analysis, thus complementing the low-resolution structural descriptors above. Using Rflex

301

metric, the selected ensemble distributions can be numerically compared to that of the pool, the

302

latter representing a reference for flexibility. Both apo and Cu(II)-recPrP are flexible systems,

303

but the protein shows a decrease of predicted flexibility in the presence of copper (Table S1,

304

section e).

10

305

Illustrative structures belonging to apo and Cu(II) OvPrP VRQ structural models generated by

306

EOM are displayed in Figure 2 H and I showing extended N-terminal domain conformations in

307

the apo form and more compact structures in the Cu(II)-chelated protein, respectively.

308 309

Copper coordination in the non-OR region of the TSE resistant OvPrP ARR

310

The Cu(II) coordination structure in the non-OR binding site of HuPrP WT was assessed by

311

EXAFS in previous investigations [5, 18] where the Cu(II) ion was found to be coordinated by

312

two histidine residues (H96 and H111), by Q98 and by a sulfur atom belonging either to M109 or

313

M112. The EXAFS experimental spectrum of HuPrP WT (data from previous studies) extracted

314

with a three-segmented cubic spline shows a typical feature around k = 5 Å-1 (Figure S6 A)

315

while the FT spectrum is characterized by a first-shell peak centered at 1.5 Å and additional high

316

intensity outer shell peaks in the distance range between 2 and 4 Å (Figure S6 C) that are

317

indicative of two histidine residues coordinated to the metal ion [5, 18, 19]. FT of all the EXAFS

318

spectra were calculated using a Gaussian window in the k range 2.4-10.1 Å-1. Note that a direct

319

comparison of the FT features among different spectra can be carried out only using the same

320

filter and the same k range as the FT shapes is strongly dependent on these parameters.

321

The H99 residue (corresponding to H96 in HuPrP) is present in the OvPrP ARR amino acid

322

sequence but the EXAFS spectrum of this specie is different from that of HuPrP WT showing

323

markedly different features in the k range around 5 Å-1, that is sensitive to the His ligands

324

(Figure S6 B). Additionally, the intensity of the second peak of the FT at 2.2 Å of OvPrP ARR

325

is lower as compared to HuPrP WT (Figure S6 D) as only one histidine ligand coordinates the

326

Cu(II) center, but the overall shape and intensity of the FT higher distances peaks indicate that an

327

additional amino acid is chelated to Cu(II) ion [44].

328

By comparison of the amino acid sequences of HuPrP and OvPrP (Figure S1) in the OvPrP

329

region between residues 94 and 101, two serine (S98 and S100) are present and both can chelate

330

Cu(II) [44]. Here, we assigned S98 as the ligand of Cu(II) ion. Starting from these observations a

331

quantitative analysis of the XANES data of OvPrP ARR has been carried out using a

332

coordination model around the Cu(II) ion comprising S98, Q101 and H114 in the equatorial

333

plane. In particular, a coordination model has been assumed where S98 is chelated to Cu(II) ion

334

through two oxygen atoms forming a 6-fold ring [44], two second shell Cu-C contributions are

335

present at a distance of about 2.86 Å and one Cu-S contribution is located at 3.25 Å that has been

336

assigned to M112. Note that during the acquisition of the XAS spectra photo reduction of Cu(II)

337

to Cu(I) occurs and for this reason the S/N ratio of the EXAFS data at k values larger than 10 Å-1 11

338

is low. Due to the limited k-range available the quantitative analysis of the EXFAS spectra would

339

provide a determination of the structural parameters affected by a big error. Conversely, XANES

340

is less affected by the experimental noise and the quantitative analysis of this region of the

341

spectra allowed us to obtain a reliable determination of the coordination structure around the

342

Cu(II) ion.

343

We applied the MXAN theoretical procedure to analyze the XANES data, and Figure 3 A

344

displays the best fitting theoretical XANES spectrum of OvPrP ARR superimposed on the

345

experimental data. A very good agreement was obtained between the XANES experimental

346

spectrum and theoretical best-fit data and the structural parameters obtained from the MXAN

347

optimization are listed in Table 1.

348

To gain a definite proof of the involvement of M112 in Cu(II) coordination an additional MXAN

349

analysis has been carried out using a coordination model not including the M112 residue. The

350

best fit results are shown in Figure 3 B and in this case a worse agreement between the

351

theoretical and experimental XANES data has been obtained as testified by the higher value of

352

the Rsq parameter. Moreover, a model where the M112 was replaced by a water molecule was

353

also tested allowing the water molecule to rotate and the Cu-OH2 distance to vary. In this case

354

the best-fit calculated XANES spectrum resulted in a worse agreement with the experimental

355

data (Rsq = 2.3) despite the opening of a new degree of freedom in the minimization procedure.

356

This finding rules out the presence of a water molecule in the Cu(II) coordination shell in OvPrP

357

ARR.

358

In conclusion for OvPrP ARR both EXAFS and XANES reveal the existence of a distorted five-

359

fold geometry of the copper center and we denoted this Cu(II) coordination geometry as type-1.

360 361

Copper coordination in the non-OR region of prion susceptible species

362

In previous works we highlighted that HuPrP mutations (i.e. P102L and Q212P) associated with

363

genetic forms of prion diseases induce a modification of Cu(II) coordination geometries when

364

bound to the non-OR binding site [18, 19]. In HuPrP Q212P, the non-OR binding site becomes

365

less structured because H111, Q98, M109 and a water molecule bind to the metal. Both EXAFS

366

and FT spectra of HuPrP Q212P are different from those of OvPrP ARR here analyzed (Figure

367

S7 A, D, data from previous studies). In particular, the HuPrP Q212P FT is characterized by

368

outer shell peaks in the range between 2 and 4 Å with lower amplitude and this shows that only

369

one histidine coordinates the Cu(II) ion. 12

370

In the present work the copper coordination of two susceptible mammalian species, namely

371

OvPrP VRQ and BvPrP, has been investigated by means of XAS. The EXAFS and FT

372

experimental spectra of these systems are similar to those of HuPrP Q212P and the low intensity

373

of the second peak of the FT suggests that only one histidine coordinates the Cu(II) ion (Figure

374

S7 B, C and E, F). Notably, in the case of TSE-susceptible mammalian species here investigated

375

the intensity of the FT peak at 2.2 Å is lower than that of the TSE-resistant species, thus

376

indicating that no other amino acid residues are chelated to the Cu(II) ion and the only MS

377

contributions are those associated with H114 (in OvPrP) or H111 (in BvPrP) as in the case of

378

HuPrP Q212P. The XANES data have been analyzed considering a fourfold coordination around

379

the Cu(II) ion with His114 (in OvPrP) or H111 (in BvPrP), two O/N and one water molecule as

380

previously determined for HuPrP Q212P [18]. In addition, the copper center interacts with the

381

sulfur atom of M112 (in OvPrP) or M109 (in BvPrP) at longer distance. The results of the

382

MXAN minimization procedure are shown in Figure 3 C, D for OvPrP VRQ and BvPrP,

383

respectively. The excellent agreement between the experimental and theoretical XANES spectra

384

confirms the close similarity of the Cu(II) geometry between the susceptible mammalian species

385

and HuPrP Q212P.

386

The refined structural parameters obtained from the MXAN minimization are listed in Table 1

387

and they are almost identical for OvPrP VRQ and BvPrP. Also in this case we tested an

388

additional model where the water molecules was not considered but the agreement between the

389

experimental and theoretical XANES spectra was quite poor (Rsq = 2.4) thus confirming the

390

presence of a water molecule in the proximity of the Cu(II) ion.

391

From the XAS analysis of the prion susceptible species a less structured coordination of the

392

Cu(II) ion has been obtained and this coordination geometry is denoted as type-2.

393 394

Copper accelerates the formation of amyloid fibrils in TSE-susceptible recPrP

395

The amyloid seeding assay is currently considered a sensitive approach to detect amyloid fibrils

396

from recPrP in partially denaturing conditions by monitoring ThT fluorescence. We monitored

397

and quantitatively described the kinetics of seeded-fibril formation of Cu(II)-OvPrP ARR,

398

Cu(II)-OvPrP VRQ and Cu(II)-BvPrP. Seeding the reaction with purified brain-derived ovine

399

PrPSc reproducibly accelerated the formation of amyloid fibrils of recPrP. Cu(II)-OvPrP VRQ

400

and Cu(II)-BvPrP displayed significant faster aggregation kinetics (lag phases 5.3 ± 0.7 hours

401

and 11.9 ± 2.2 hours, respectively) than Cu(II)-OvPrP ARR (43.8 ± 0.9 hours) (Figure 4 A, B).

402 13

403

DISCUSSION

404

Different findings suggest a protective role for copper when bound to the OR region since the

405

metal inhibits prion aggregation [45, 46]. The OR region exhibits high reduction potential for the

406

Cu(II)/Cu(I) couple and can initiate reactive oxygen species-mediated β-cleavage of PrPC at

407

around residue 90 [47, 48]. This may generate the N-terminally truncated form of PrPC that takes

408

part in the amyloid core during prion conversion [10]. The N-terminal PrPC cleavage may be also

409

mediated by the acidic environment and enzymes found in recycling endosomal compartments

410

[49, 50]. Given the known importance of the C-terminal region from residues
90 to 231 for

411

prion formation, different studies have addressed the question whether the Cu(II)-bound non-OR

412

region plays a role in prion conversion. Transgenic mice, TgPrP(H95G), with an amino acid

413

replacement at residue H95 (homolog to H96 in human numbering) had shorter disease

414

progression than WT control mice and classical clinical signs of TSE [51]. We observed that

415

alteration of Cu(II) coordination due to H95Y mutation causes spontaneous PrPSc-like formation

416

in neuronal cultured cells [19]. At that time, we proposed a model whereby HuPrP coordinating

417

copper with two histidine residues (H96 and H111) in the non-OR region is more resistant to

418

prion conversion compared to the protein coordinating Cu(II) with one histidine. However, little

419

in the way of structural information exists regarding the Cu(II)-mediated non-OR stabilization

420

and inter-domain contacts due to the intrinsic flexible nature of the segment from residues
90

421

to 127.

422

Here, we provide new insights into the Cu(II) structural consequences when the metal is bound

423

to the non-OR region, particularly with regard to recPrP from of well-known animal species

424

considered to be TSE-resistant or susceptible. Using combined SAXS and XAS approaches we

425

found that Cu(II) promotes structural compactness of recPrP upon metal binding and it displays

426

different coordination geometries when bound to TSE-resistant or susceptible recPrP. All SAXS

427

and XAS measurements were carried out at acidic pH where the non-OR region can still

428

coordinate one Cu(II) ion [8, 15]. The capability for Cu(II) binding at acidic pH indicates that the

429

metal could be maintained during the cycling of the protein in the acidic endosomal

430

compartments, which are considered an important site for PrPSc conversion and accumulation,

431

and subsequent propagation via endosomal-derived exosomes [52-54]. Additionally, our

432

previous experience with XAS spectroscopy leads us to the conclusion that remarkable

433

differences in the Cu(II) coordination geometries are present only at acidic conditions and not at

14

434

pH 7.0 [18, 19]. These data have been compared with previous measurements on Cu(II)

435

coordination geometries in the non-OR region of HuPrP WT and Q212P.

436

We applied the SEC-SAXS method that allows one to collect and interpret SAXS data on

437

experimentally difficult aggregation-prone protein system, such as truncated recPrP, and to

438

minimize radiation damage thanks to the continuous flow. Overall the Rg and Dmax values of apo

439

recPrP calculated from Guinier and P(r) analysis are slightly larger than Cu(II)-recPrP, although

440

higher resolution experiments are necessary to confirm these observations. To dissect the

441

different conformational recPrP states we used EOM approaches that provided a semi-

442

quantitative assessment on the structural effect of Cu(II) to protein compactness through direct

443

comparison of the size distributions of the apo versus Cu(II)-recPrP conformers. Our EOM data

444

analysis indicates that copper induces a global folding compaction in all the recPrP and promotes

445

multiple Cu(II)-protein ensemble conformations.

446

It should be emphasized that the EOM strategies result only in the prediction of the optimal

447

conformational distribution, therefore they are affected by intrinsic limitations such as the low-

448

resolution of the SAXS technique. Nonetheless, the resulting Cu(II)-recPrP structural models

449

generate insights into the biological function of flexible systems. The prediction of significant

450

protein conformational distributions reflects the ability of the recPrP N-terminal domain to

451

generate an equilibrium ensemble of domain orientations. Notably, Cu(II)-OvPrP ARR displays

452

a significantly different size distribution than VRQ and BvPrP conformers, which clearly cluster

453

in mainly two separated groups (one compact and one extended).

454

We reasoned that the tendency of the TSE-resistant OvPrP ARR to form a more uniform

455

population without extended structures may account for the reduced fibrillization propensity

456

observed in the amyloid seeding assay. The Cu(II)-mediated stabilization of native OvPrP ARR

457

conformers may render the protein less vulnerable to restructuring and subsequent misfolding.

458

Conversely, Cu(II)-OvPrP VRQ and Cu(II)-BvPrP are predicted to be present also in extended

459

conformers that may account for the faster aggregation propensity (Figure 4 A-B). Interestingly,

460

the fibrilization curves follow a double nucleation mechanism in which fibrils can also be

461

initiated by secondary nucleation catalyzed by the initial population of fibrils. This observation

462

leads us to hypothesize that intermediate species are present in the very early stage of the

463

aggregation curve, as also observed by other groups [55, 56].

464

Then, we expanded our understanding on Cu(II) coordination in the non-OR region of different

465

mammalian recPrP. The XAS results showed that in Cu(II)-OvPrP ARR the non-OR region is

466

structured while the Cu(II)-OvPrP VRQ and Cu(II)-BvPrP are characterized by a more flexible 15

467

non-OR binding site where solvent molecules can enter. The same Cu(II) coordination was

468

previously found in HuPrP Q212P and P102L mutants [19].

469

Based on our results we proposed two models of Cu(II) coordination in the non-OR region

470

(Figure 4 C). The type-1 coordination displays a “closed” conformation, which can be

471

associated with TSE-resistant species possibly because of higher stability of the non-OR region.

472

In type-2 coordination a water molecule enters the coordination shell, thus leading to a less

473

structured and solvent exposed non-OR region. We believe that this more opened conformation

474

of the non-OR region in the type-2 renders the overall N-terminal segment more flexible and

475

prone to structural rearrangements linked to prion conversion.

476

Several amino acid substitutions are present in the recPrP here investigated. Based on previous

477

structural studies, they typically affect the rigidity of some loops through stabilizing hydrogen

478

bonds without affecting global folding of the structured domain. In fact in BvPrP the presence of

479

N170 determines the rigidity of the β2-α2 loop [39], which is correlated with a higher

480

susceptibility to TSE transmission [57, 58]. In sheep, the different prion susceptibility is dictated

481

by amino acid variations in the C-terminal globular domain (Figure S1). The X-ray crystal and

482

NMR structures of the OvPrP ARR and VRQ folded domains reveal only minor differences

483

showing that the A136 and R171 polymorphisms lead to ablation of hydrogen bonds located at

484

the protein surface [40, 59]. This suggests that the mechanisms of TSE susceptibility and

485

resistance are unlikely due only to mutation-induced structural changes but require the

486

contribution of other cellular components [59] such as, for instance, copper ions as mediator of

487

tertiary contacts between the N- and C-terminal domains.

488

Structural studies on pathological point mutations have provided new clues on the early

489

structural rearrangements occurring in some epitopes in the structured domain [60], but also

490

revealed that amino acid variations have structural effects on the N-terminal part, affecting both

491

Cu(II) coordination in the non-OR region [18] and long-range interactions between the Cu(II)-

492

OR segment and the structured domain [61]. Tertiary interactions between this flexible segment

493

and the globular C-terminal domain occur when PrPC binds to Cu(II) or Zn(II) through the OR

494

region [14, 61]. The structural correlation between the N- and the C-terminal domains appears to

495

be mediated by the non-OR region which acts as “anchor” of the two halves of the protein [62].

496

Previous studies identified by NMR and EPR methods an interacting surface made by the Cu(II)-

497

OR and a negatively charged pocket in the C-terminal domain comprising residues located

498

nearby the β1-α1 loop and the α2-α3 loop region [63]. Similarly, the addition of Cu(II) to the non-

499

OR of truncated MoPrP caused significant broadening of peaks corresponding to residues close 16

500

to the β1-α1 loop and helix α1 regions [14]. A recent investigation using NMR, MS/MS and

501

molecular dynamics approaches has identified on Cu(II)-full length MoPrP the interacting

502

surfaces involved in docking between N-terminal polybasic domains (i.e. the segment from

503

residues 23 to 31, the OR and non-OR regions) and a C-terminal negatively charged epitope

504

coincident with the α1-β2-α2 loop region as well as the end of α3 [62].

505

Our SAXS and XAS observations suggest that Cu(II) in type-1 coordination stabilizes the non-

506

OR region and increases the protein compactness; this may favor more stable long-range

507

interaction contacts between the
90–127 segment and the C-terminal pocket. Conversely,

508

Cu(II) in type-2 coordination renders the non-OR region more flexible and the N-terminal moiety

509

more extended possibly due to less frequent inter-domain contacts. Alterations of this inter-

510

domain contacts may have relevant physiological implications for TSE progression since this

511

Cu(II)-promoted non-OR interaction with the C-terminal domain renders PrPC more stable to

512

PrPSc conversion. This is consistent with observations showing that an antibody, POM1, able to

513

abolish the N- to C-terminus interaction upon binding to the negatively charged pocket causes

514

acute neurotoxicity in mice and cultured cerebellar brain slices [64]. Our hypothesis awaits

515

confirmation from high-resolution structural studies but it provides a rationale for a platform of

516

experiments aimed at elucidating the atomic mechanisms of Cu(II)-mediated interdomain

517

interaction in TSE-resistant and susceptible prion proteins.

518 519

CONCLUSIONS

520

We showed that Cu(II) bound to the non-OR region of the prion protein mediates physiologically

521

relevant structural changes in the N-terminal moiety. We proposed two novel Cu(II) coordination

522

geometries in the non-OR copper binding site of animal species considered resistant or

523

susceptible to TSE. Changes in the Cu(II) coordination geometry from type-1 to type-2 might

524

dictate the lower or higher susceptibility to TSE observed in different animals. Considering the

525

role of the non-OR and palindromic regions in supporting prion conversion, this study proposes a

526

novel structural mechanism responsible for prion susceptibility in different mammalian species.

527 528

Author contributions

529

G.G., P.D. and G.L. conceived the project. G.G. and P.D. jointly supervised this work and wrote

530

the manuscript. G.S., G.G., T.H.T., G.M. and P.D. carried out XAS data collection at ESRF and

531

analysed the data. G.G. and M.B. carried out SAXS data collection at ESRF and G.G. analysed 17

532

the data. G.S. and T.H.T. provided the recombinant protein samples. G.S. carried out and

533

analysed the amyloid seeding assays. All authors read and approved the final manuscript.

534 535

Acknowledgments

536

Authors acknowledge Prof. Olivier Andreoletti for kindly providing the ovine brain homogenate,

537

Natali Nakic for brain homogenate preparation, Andrea Raspadori and Paola Zago for their

538

contributions to recPrP production. Authors thank to the anonymous reviewers for the valuable

539

comments, which helped us to improve the quality of XAS and SAXS data analysis. We are

540

thankful to the ESRF Structural Biology group for the SAXS beamtimes at BM29 and to Isabelle

541

Kieffer (BM30B) for assistance during the XAS beamtime.

542 543

Funding sources

544

This work was supported by the ESRF in-house Research Program (to G.G), by University of

545

Rome “La Sapienza” (Progetto Ateneo 2016, n. RM116154C8D7E4D0, to P.D.) and by SISSA

546

intramural funding (to G.S., T.H.T. and G.L.). This work was also supported by a fellowship

547

from the HEaD “Higher Education and Development” project (funded by the Friuli Venezia

548

Giulia autonomous Region, Italy, through the Operational Program of the European Social Fund

549

2014/2020; grant n. FP1619889004) (to G.S.).

550 551

Conflict of interest

552

The authors declare that they have no conflicts of interest with contests of this article.

553

18

554

FIGURES AND TABLE LEGEND

555

Table 1. Structural parameters derived from the XANES analysis. Structural parameters

556

derived from the XANES analysis of the Cu K-edge spectra of Cu(II)-OvPrP ARR, Cu(II)-

557

OvPrP VRQ, and Cu(II)-BvPrP. N is the coordination number and R is the distance. Statistical

558

errors are reported in parenthesis.

559 560

Figure 1. SAXS measurements on apo and Cu(II)-recPrP. (A) Cartoon representation of the

561

C-terminal OvPrP VRQ with the non-OR copper binding site. The structure of the non-OR

562

copper binding site from residues 99 to 114 is shown in blue; in ball and stick the residues

563

coordinating a copper ion (S98, Q101, M112 and H114), the copper ion is shown as a green

564

sphere. The palindromic region is shown with an enlarged ribbon. (B-D) SAXS curves of BvPrP,

565

OvPrP ARR and OvPrP VRQ, respectively. Black and light blue dots represent the apo and

566

Cu(II)-recPrP SAXS curves, respectively, with in red the GNOM fitting. Insets show the Guinier

567

fits (yellow dots).

568 569

Figure 2. Characterization of the flexibility of apo and Cu(II)-recPrP using EOM. (A-C)

570

I(q) versus q experimental SAXS profiles (black dots for apo proteins and cyan dots for copper-

571

loaded proteins) with the EOM fit models (continuum red lines for apo proteins and red dotted

572

lines for copper-loaded proteins with the corresponding χ2 values) of the recPrP. The curves are

573

shifted by an arbitrary offset for better comparison. The lower plots show the residuals defined as

574

[Iexp(q) - Icomp(q)]/σexp(q) corresponding to the difference between the experimental and the

575

computed intensities weighted by the experimental uncertainty. (D-F) Size distributions (Rg) of

576

OvPrP ARR, OvPrP VRQ and BvPrP, respectively, providing qualitative assessment through

577

direct comparison of the distributions of the ensembles [grey and cyan columns for apo and

578

Cu(II)-protein, respectively] and the pools [black and cyan lines for apo and Cu(II)-protein,

579

respectively]. The ensemble distributions of Cu(II)-recPrP were fitted with a multi-Gaussian

580

function to better highlight the subpopulations (red fitting line). (G) Frequency-weighted size

581

average values of the protein structures obtained by EOM modelling. In the upper and lower

582

panels the Dmax and Rg descriptors, respectively, obtained for apo and Cu(II)-recPrP. (H-I)

583

Illustrative 3D structures obtained after EOM modelling describing the apo and Cu(II)-OvPrP

584

VRQ, respectively. Representative EOM descriptors (i.e. frequency-weighted size average

585

values, Rflex and Rσ metrics) are included in the panels. The 3D OvPrP structure features a C19

586

terminal globular domain (from residues 133 to 226) here used as rigid body and the unfolded N-

587

terminal segment (from residues 94 to 133) modelled by EOM. The VRQ polymorphisms are

588

highlighted in red. A representative N-terminal conformer is shown as edged black line with the

589

non-OR Cu(II) binding site coloured in violet, while the other frames are visualized in

590

transparency mode.

591 592

Figure 3. Fit of the Cu K-edge XANES spectra. (A) Cu(II)-OvPrP ARR and minimization

593

carried out without M112 as ligand (B)). Fit of the Cu K-edge XANES spectra of Cu(II)-OvPrP

594

VRQ (C) and Cu(II)-BvPrP (D)).

595 596

Figure 4. Seeded-fibrillization reactions and the type-1, type-2 non-OR Cu(II) coordination

597

models. (A) Kinetics of ThT-positive Cu(II)-OvPrP VRQ (violet coloured dots), Cu(II)-BvPrP

598

(green dots) and Cu(II)-OvPrP ARR (blue dots) aggregates. (B) Mean value of the lag phases of

599

the three Cu(II)-recPrP (in hours, bars denote the standard error,

600

type-2 non-OR Cu(II) coordination proposed models for prion resistant (OvPrP ARR) and

601

susceptible (OvPrP VRQ and BvPrP) mammalian species (here using the OvPrP numbering).

602

Blue spheres identify nitrogen atoms, red spheres are oxygen atoms and yellow spheres encode

603

for sulfur atoms. Gray and white spheres represent carbon and hydrogen atoms, respectively. In

604

brackets, the putative amino acid residues coordinating copper on the basis of quantitative

605

analysis of XANES data.

606

20





P<0.01). (C) The type-1 and

607

Table 1 Cu(II)-OvPrP ARR

Cu(II)-OvPrP VRQ

Cu(II)-BvPrP

N

R (Å)

N

R (Å)

N

R (Å)

1 NHis

1.93(6)

1 NHis

1.91(5)

1 NHis

1.91(5)

2 OSer

2.04(4)

2 O/N

2.02(3)

2 O/N

1.98(4)

1 OQ98

2.05(9)

1 Owater

2.34(8)

1 Owater

2.33(9)

1S

3.19(9)

1S

3.19(0)

1S

3.16(9)

608 609

21

610

References

611 612 613 614 615 616 617 618 619 620 621 622 623 624 625 626 627 628 629 630 631 632 633 634 635 636 637 638 639 640 641 642 643 644 645 646 647 648 649 650 651 652 653 654 655 656 657 658 659

1. 2. 3. 4.

5. 6.

7. 8.

9.

10. 11. 12. 13. 14. 15. 16.

17. 18. 19. 20.

21. 22. 23.

Colby, D.W. and S.B. Prusiner, Prions. Cold Spring Harb Perspect Biol, 2011. 3(1): p. a006833. Imran, M. and S. Mahmood, An overview of human prion diseases. Virol J, 2011. 8: p. 559. Zahn, R., et al., NMR solution structure of the human prion protein. Proc Natl Acad Sci U S A, 2000. 97(1): p. 145-50. Walter, E.D., M. Chattopadhyay, and G.L. Millhauser, The affinity of copper binding to the prion protein octarepeat domain: evidence for negative cooperativity. Biochemistry, 2006. 45(43): p. 13083-92. Hasnain, S.S., et al., XAFS study of the high-affinity copper-binding site of human PrP(91-231) and its low-resolution structure in solution. J Mol Biol, 2001. 311(3): p. 467-73. Jones, C.E., et al., Preferential Cu2+ coordination by His96 and His111 induces beta-sheet formation in the unstructured amyloidogenic region of the prion protein. J Biol Chem, 2004. 279(31): p. 32018-27. Younan, N.D., et al., Copper(II)-Induced Secondary Structure Changes and Reduced Folding Stability of the Prion Protein. Journal of Molecular Biology, 2011. 410(3): p. 369-382. Thompsett, A.R., et al., High Affinity Binding between Copper and Full-length Prion Protein Identified by Two Different Techniques. Journal of Biological Chemistry, 2005. 280(52): p. 42750-42758. Kuwata, K., et al., NMR-detected hydrogen exchange and molecular dynamics simulations provide structural insight into fibril formation of prion protein fragment 106-126. Proc Natl Acad Sci U S A, 2003. 100(25): p. 14790-14795. Spagnolli, G., et al., Full atomistic model of prion structure and conversion. PLoS Pathog, 2019. 15(7): p. e1007864. Parchi, P., et al., Genetic influence on the structural variations of the abnormal prion protein. Proceedings of the National Academy of Sciences, 2000. 97(18): p. 10168-10172. Legname, G., et al., Synthetic mammalian prions. Science, 2004. 305(5684): p. 673-6. Gasperini, L., et al., Prion protein and copper cooperatively protect neurons by modulating NMDA receptor through S-nitrosylation. Antioxid Redox Signal, 2015. 22(9): p. 772-84. Thakur, A.K., et al., Copper alters aggregation behavior of prion protein and induces novel interactions between its N- and C-terminal regions. J Biol Chem, 2011. 286(44): p. 38533-45. Wells, M.A., et al., A reassessment of copper(II) binding in the full-length prion protein. Biochem J, 2006. 399(3): p. 435-44. Wong, E., A.M. Thackray, and R. Bujdoso, Copper induces increased beta-sheet content in the scrapie-susceptible ovine prion protein PrPVRQ compared with the resistant allelic variant PrPARR. Biochem J, 2004. 380(Pt 1): p. 273-82. Migliorini, C., et al., Copper-induced structural propensities of the amyloidogenic region of human prion protein. J Biol Inorg Chem, 2014. 19(4-5): p. 635-45. D'Angelo, P., et al., Effects of the pathological Q212P mutation on human prion protein nonoctarepeat copper-binding site. Biochemistry, 2012. 51(31): p. 6068-79. Giachin, G., et al., The non-octarepeat copper binding site of the prion protein is a key regulator of prion conversion. Sci Rep, 2015. 5: p. 15253. Goldmann, W., et al., PrP genotype and agent effects in scrapie: change in allelic interaction with different isolates of agent in sheep, a natural host of scrapie. J Gen Virol, 1994. 75 ( Pt 5): p. 989-95. Nonno, R., et al., Efficient Transmission and Characterization of Creutzfeldt–Jakob Disease Strains in Bank Voles. PLOS Pathogens, 2006. 2(2): p. e12. Pernot, P., et al., Upgraded ESRF BM29 beamline for SAXS on macromolecules in solution. J Synchrotron Radiat, 2013. 20(Pt 4): p. 660-4. Brennich, M.E., A.R. Round, and S. Hutin, Online Size-exclusion and Ion-exchange Chromatography on a SAXS Beamline. J Vis Exp, 2017(119).

22

660 661 662 663 664 665 666 667 668 669 670 671 672 673 674 675 676 677 678 679 680 681 682 683 684 685 686 687 688 689 690 691 692 693 694 695 696 697 698 699 700 701 702 703 704 705 706 707 708 709 710 711 712

24.

25. 26. 27.

28. 29. 30. 31. 32. 33.

34.

35.

36. 37. 38. 39. 40. 41. 42. 43.

44. 45. 46. 47.

Orthaber, D., A. Bergmann, and O. Glatter, SAXS experiments on absolute scale with Kratky systems using water as a secondary standard. Journal of Applied Crystallography, 2000. 33(2): p. 218-225. Brennich, M.E., et al., Online data analysis at the ESRF bioSAXS beamline, BM29. Journal of Applied Crystallography, 2016. 49(1): p. 203-212. Franke, D., et al., ATSAS 2.8: a comprehensive data analysis suite for small-angle scattering from macromolecular solutions. J Appl Crystallogr, 2017. 50(Pt 4): p. 1212-1225. Trewhella, J., et al., 2017 publication guidelines for structural modelling of small-angle scattering data from biomolecules in solution: an update. Acta Crystallogr D Struct Biol, 2017. 73(Pt 9): p. 710-728. Rambo, R.P. and J.A. Tainer, Accurate assessment of mass, models and resolution by small-angle scattering. Nature, 2013. 496: p. 477. Svergun, D., Determination of the regularization parameter in indirect-transform methods using perceptual criteria. Journal of Applied Crystallography, 1992. 25(4): p. 495-503. Bernado, P., et al., Structural characterization of flexible proteins using small-angle X-ray scattering. Journal of the American Chemical Society, 2007. 129(17): p. 5656-5664. Eswar, N., et al., Comparative Protein Structure Modeling Using Modeller. 2006. 15(1): p. 5.6.15.6.30. Tria, G., et al., Advanced ensemble modelling of flexible macromolecules using X-ray solution scattering. Iucrj, 2015. 2: p. 207-217. Proux, O., et al., FAME: a new beamline for x-ray absorption investigations of very-diluted systems of environmental, material and biological interests. Physica Scripta, 2005. 2005(T115): p. 970. Filipponi, A. and A. Di Cicco, X-ray-absorption spectroscopy and n-body distribution functions in condensed matter. II. Data analysis and applications. Phys Rev B Condens Matter, 1995. 52(21): p. 15135-15149. Filipponi, A., A. Di Cicco, and C.R. Natoli, X-ray-absorption spectroscopy and n-body distribution functions in condensed matter. I. Theory. Phys Rev B Condens Matter, 1995. 52(21): p. 15122-15134. Benfatto, M. and S. Della Longa, Geometrical fitting of experimental XANES spectra by a full multiple-scattering procedure. Journal of Synchrotron Radiation, 2001. 8(4): p. 1087-1094. Colby, D.W., et al., Prion detection by an amyloid seeding assay. 2007. 104(52): p. 20914-20919. Polano, M., et al., Structural Insights into Alternate Aggregated Prion Protein Forms. Journal of Molecular Biology, 2009. 393(5): p. 1033-1042. Christen, B., et al., NMR structure of the bank vole prion protein at 20 degrees C contains a structured loop of residues 165-171. J Mol Biol, 2008. 383(2): p. 306-12. Lysek, D.A., et al., Prion protein NMR structures of cats, dogs, pigs, and sheep. Proc Natl Acad Sci U S A, 2005. 102(3): p. 640-5. Carter, L., et al., Prion Protein-Antibody Complexes Characterized by Chromatography-Coupled Small-Angle X-Ray Scattering. Biophys J, 2015. 109(4): p. 793-805. Boze, H., et al., Proline-Rich Salivary Proteins Have Extended Conformations. Biophysical Journal, 2010. 99(2): p. 656-665. Paz, A., et al., Biophysical Characterization of the Unstructured Cytoplasmic Domain of the Human Neuronal Adhesion Protein Neuroligin 3. Biophysical Journal, 2008. 95(4): p. 19281944. D'Angelo, P., et al., X-ray Absorption Study of Copper(II)-Glycinate Complexes in Aqueous Solution. J. Phys. Chem. B, 1998. 102: p. 3114-3122. Bocharova, O.V., et al., Copper(II) inhibits in vitro conversion of prion protein into amyloid fibrils. Biochemistry, 2005. 44(18): p. 6776-87. Orem, N.R., et al., Copper (II) ions potently inhibit purified PrPres amplification. J Neurochem, 2006. 96(5): p. 1409-15. McMahon, H.E., et al., Cleavage of the amino terminus of the prion protein by reactive oxygen species. J Biol Chem, 2001. 276(3): p. 2286-91. 23

713 714 715 716 717 718 719 720 721 722 723 724 725 726 727 728 729 730 731 732 733 734 735 736 737 738 739 740 741 742 743 744 745 746 747 748 749 750 751 752 753

48.

49.

50.

51. 52. 53. 54.

55. 56. 57. 58. 59.

60. 61. 62.

63.

64.

Yamamoto, N. and K. Kuwata, Difference in redox behaviors between copper-binding octarepeat and nonoctarepeat sites in prion protein. Journal of Biological Inorganic Chemistry, 2009. 14(8): p. 1209-1218. Jeffrey, M., S. Martin, and L. Gonzalez, Cell-associated variants of disease-specific prion protein immunolabelling are found in different sources of sheep transmissible spongiform encephalopathy. J Gen Virol, 2003. 84(Pt 4): p. 1033-45. Caughey, B., et al., N-terminal truncation of the scrapie-associated form of PrP by lysosomal protease(s): implications regarding the site of conversion of PrP to the protease-resistant state. J Virol, 1991. 65(12): p. 6597-603. Eigenbrod, S., et al., Substitutions of PrP N-terminal histidine residues modulate scrapie disease pathogenesis and incubation time in transgenic mice. PLoS One, 2017. 12(12): p. e0188989. Yim, Y.-I., et al., The multivesicular body is the major internal site of prion conversion. Journal of Cell Science, 2015. 128(7): p. 1434. Marijanovic, Z., et al., Identification of an Intracellular Site of Prion Conversion. PLOS Pathogens, 2009. 5(5): p. e1000426. Bartoletti-Stella, A., et al., Analysis of RNA Expression Profiles Identifies Dysregulated Vesicle Trafficking Pathways in Creutzfeldt-Jakob Disease. Molecular Neurobiology, 2019. 56(7): p. 5009-5024. Ferrone, F.A., M. Ivanova, and R. Jasuja, Heterogeneous Nucleation and Crowding in Sickle Hemoglobin: An Analytic Approach. Biophysical Journal, 2002. 82(1): p. 399-406. Moda, F., et al., Synthetic prions with novel strain-specified properties. PLOS Pathogens, 2016. 11(12): p. e1005354. Sigurdson, C.J., et al., A molecular switch controls interspecies prion disease transmission in mice. J Clin Invest, 2010. 120(7): p. 2590-9. Rossetti, G., et al., Structural facets of disease-linked human prion protein mutants: a molecular dynamic study. Proteins, 2010. 78(16): p. 3270-80. Eghiaian, F., et al., Insight into the PrPC-->PrPSc conversion from the structures of antibodybound ovine prion scrapie-susceptibility variants. Proc Natl Acad Sci U S A, 2004. 101(28): p. 10254-9. Giachin, G., et al., Probing early misfolding events in prion protein mutants by NMR spectroscopy. Molecules, 2013. 18(8): p. 9451-76. Spevacek, A.R., et al., Zinc drives a tertiary fold in the prion protein with familial disease mutation sites at the interface. Structure, 2013. 21(2): p. 236-46. McDonald, A.J., et al., Altered Domain Structure of the Prion Protein Caused by Cu2+ Binding and Functionally Relevant Mutations: Analysis by Cross-Linking, MS/MS, and NMR. Structure, 2019. Evans, E.G., et al., Interaction between Prion Protein's Copper-Bound Octarepeat Domain and a Charged C-Terminal Pocket Suggests a Mechanism for N-Terminal Regulation. Structure, 2016. 24(7): p. 1057-67. Sonati, T., et al., The toxicity of antiprion antibodies is mediated by the flexible tail of the prion protein. Nature, 2013. 501(7465): p. 102-6.

754

24