Biochemical Studies of Prion Protein-Lipid Interaction

Home , Prion

BIOCHEMICAL STUDIES OF INTERACTIONS

BETWEEN PRION PROTEIN AND LIPIDS

DISSERTATION

Presented in Partial Fulfillment of the Requirements for

the Degree Doctor of Philosophy in the Graduate

School of The Ohio State University

Fei Wang, M. S.

The Ohio State University

2008

Dissertation Committee:

Dr. Jiyan Ma, Adviser Approved by

Dr. Mark Parthun

Dr. Jeff Kuret ______

Dr. Charles Bell Adviser, The Ohio State Biochemistry Graduate Program

ABSTRACT

Prion diseases, also know as Transmissible Spongiform Encephalopathies

(TSEs), are a group of invariably fatal neurodegenerative disorders that occur in

a wide variety of mammals including humans. According to the “Protein-Only”

hypothesis, prion is the transmissible agent, which is mainly composed of the

abnormal isoform (PrPSc) of the normal cellular prion protein (PrPC). PrPSc converts PrPC into likeness of itself, resulting in neurodegeneration. Despite

extensive research, the precise mechanism underlying the PrP conversion process in prion diseases still remains unclear.

Previous studies using biophysical methods revealed that PrP interacts

with lipids, which leads to a conformational change of PrP. Yet, whether

Proteinase K (PK) resistance, a classic biochemical feature of PrPSc, is associated with those altered PrP conformations induced by lipid interaction remains unclear. Main parts of this thesis were dedicated to elucidate how PrP and lipid interacts with each other and whether PrP-lipid interaction is sufficient to convert PrP to the classic PK-resistant conformation in the absence of denaturing treatments. Recombinant PrP expressed in E. Coli, which is considered structurally equivalent to PrPC, is monitored for lipid-induced conformational change in this study.

In this thesis, I, by virtue of biochemical techniques, found that full-length

ii recombinant PrP (rPrP) preferentially binds to anionic lipids. Under an environment reminiscent of physiological conditions, the lipid interaction causes rPrP conformational change, converting from a mainly α-helical structure to a

high β-sheet conformation featuring PrPSc-like PK resistance.

In the subsequent study on characterization of rPrP-lipid interaction, I

found that the lipid induced rPrP conversion requires both hydrophobic rPrP-lipid

interaction and the localization of anionic charges on the surface of lipid vesicles,

suggesting the involvement of highly conserved middle region of PrP that consists of a cluster of positively charged lysine residues followed by a hydrophobic domain.

To determine the relevance of rPrP-lipid interaction to prion biology, I

characterized rPrP-lipid interaction using rPrP with pathogenic mutations and

factors known to alter PrPSc propagation or stability. I found that, besides

phospholipids, arachidonic acid also supports the generation of C-terminal PK-

resistant rPrP. In addition, lipid oxidation, metal ions and RNA affect lipid-induced

rPrP conversion in a manner similar to their effects on the pathogenic PrPSc.

Collectively, these findings of lipid-induced PrP conversion under

physiological conditions provide strong support for the relevance of lipid-PrP

interaction to the pathogenic changes in prion disease. Further investigation of

this process together with the relationship between lipid-induced PrPSc-like

iii conformation and prion infectivity may help us to elucidate the pathogenic mechanism of prion disease.

iv Dedicated to

My Parents

and

My Wife

v ACKNOWLEDGEMENTS

I wish to thank my adviser, Dr. Jiyan Ma, for his intellectual support, encouragement, enthusiasm, which made this thesis possible, and for his patience in correcting both my stylistic and scientific errors.

I would like to thank Dr. Xinhe Wang for advising and discussing experimental details to me.

I am grateful to all my dissertation committee members, Dr. Mark Parthun,

Dr. Jeff Kuret, and Dr. Charles Bell, for their supports and efforts to this thesis.

I also wish to thank those who helped me with all the lipid-related problems, especially Warren Erdahl, Gregory Steinbaugh, and Dr. Douglas

Pfeiffer.

I also wish to thank Dr. Changwen Jin for providing the CD data of mouse recombinant prion protein and Dr. Man-sun Sy for providing the recombinant human prion protein samples.

I owe my loving thanks to my wife, Xi E Sun, for her caring and supporting me all the way during my study and research abroad.

Lastly, and most importantly, I wish to thank my parents, Zhiliang Wang and Baozhen Duan. They bore me, raised me, taught me, supported me, and loved me. To them I dedicate this thesis.

vi VITA

1995-1999 ….……………………………………………. B. S. Nankai University, Tianjin, China

1999-2002 ………………………………………..……… M. S. Nankai University, Tianjin, China

2002-Present …………………………………………. Graduate Research Associate The Ohio State University

FIELDS OF STUDY

Major Field: The Ohio State Biochemistry Program

vii TABLE OF CONTENTS

Page

Abstract …………………………………………………………………………………………………………... ii

Dedication ………………………………………………………………………………………………………. v

Acknowledgements ……………………………………………………………………………………... vi

Vita ……………………………………………………………………………………………………………………. vii

Fields of Study ………………………………………………………………………………………………. vii

List of Tables …………………………………………………………………………………………………. xiii

List of Figures ….……………………………………………………………………………………………. xiv

List of Abbreviations ……………………………………………………………………………………. xvii

Chapters

1 Introduction …………………………………...………………………………………………………… 1

1.1 Prion diseases ……………………………………………….………………………………… 1

Human prion diseases ……………………………………………………………………… 2 Scrapie and Bovine Spongiform Encephalopathy………………………… 6

Chronic Wasting Disease …………………………………………………………………. 8 1.2 The infectious agent in prion disease ……………………………………….. 9

History of prion research …………………………………………………………………... 9 Purification of Prions …………………………………………………………………………. 11

viii The “Protein-only” hypothesis ………………………………………………………….. 13 Prion strains and species barrier …………………………………………………….. 14 1.3 The cellular prion protein, PrPC …………………………………………………… 16

The prion protein gene, Prnp …………………………………………………………… 16 Cell biology of PrPC…………………………………………………………………………….. 16

Structure of PrPC ……………………………………………………………...………………… 19

Biological functions of PrPC……………………………………………….……………… 20 1.4 Conversion of PrPC to PrPSc ………………………………………………………….. 23

Comparison between PrPC and PrPSc …………………………………………….. 23 Studies on PrP conversion ……………………………………………………………….. 26

2 Aims of thesis …………………………………………………………………………………………. 30

3 Materials and Methods …………………………………………………………………………. 31

3.1 Construction of expression plasmids ……………………………………………… 31 3.2 Expression and purification of recombinant proteins …………………… 32

3.3 Isolation of membranes from cultured murine neuroblastoma N2A

cells ………………………………………………………………………………………………………. 38 3.4 Extraction of lipids from N2A cells or mouse brain tissues …………. 38 3.5 Preparation of lipid vesicles by sonication …………………………………….. 39

3.6 Preparation of lipid vesicles by extrusion ………………………………………. 39

3.7 Two-dimensional thin-layer chromatography (TLC) analysis of compositions of lipids extracted from N2A cells …………………………… 41

ix 3.8 Gradient assays …………………………………………………………………………………. 42 3.9 PrP-lipid incubation ……………………………………………………………………………. 44

3.10 PK digestion ……………………………………………………………………………………... 45

3.11 Sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) ……………………………………………………………………………………….. 45 3.12 Detection of prion protein by coomassie brilliant blue staining … 47

3.13 Detection of prion protein by immunoblot ……………………………………. 47

3.14 Phosphorus analysis ……………………………………………………………………….. 48 3.15 Transmission electron microscopy ……………………………………………….. 50

4 Results ………………………………………………………………………………………………………. 51

4.1 Introduction ……………………………………………………………………………………….. 51

4.2 Lipid interaction converts recombinant prion protein to a protease resistant conformation under physiological conditions ………………………………………………………………………………………… 56 Binding of recombinant PrP to lipids involves both electrostatic and hydrophobic interactions ………………………………………………… 56

rPrP undergoes conformational changes upon binding lipids and

gains the Proteinase K resistance ………………………………………… 64 The lipid-induced C-terminal PK-resistance of rPrP formed under different mechanism from N-terminal and full-length PK-resistance ……………………………………………………………………………. 67 4.3 Characterization of lipid-rPrP interaction …………………………………. 74

Both charge and lipid headgroup structure influence rPrP

x conformational change …………………………………………………………….. 74 The hydrophobic domain of PrP is essential for generation of C-

terminal PK-resistant rPrP ………………………………………………………. 81

Adjacent localization of anionic charges and the hydrophobic lipid core is necessary for inducing C-terminal PK-resistant rPrP …………………………………………………………………………………………….. 85 Aggregation of PrP stabilizes the lipid-induced PK-resistant rPrP conformation ……………………………………………………………………………… 87

Lipid-induced PK resistance of rPrP does not correlate with fiber formation ……………………………………………………………………………………. 93 4.4 The relevance of rPrP-lipid interaction to prion biology ………. 95

Arachidonic acid induces rPrP conversion …………………………………….. 95 Effects of lipid oxidation, copper, iron, and RNA on formation of

C-terminal PK-resistant rPrP ………………………………………………….. 98

Middle region localized P105L mutation and P129 polymorphism affect rPrP-lipid interaction ……………………………………………………… 104

5 Discussions and conclusions …………………………………………………………….. 114

Summary of thesis ……………………………………………………………………………... 114 Relevance of studying rPrP-lipid interaction to the pathogenesis

in prion disease ………………………………………………………………………… 115

Different interactions between rPrP and lipids result in different PK-resistant rPrP species ………………………………………………………. 117 The C-terminal PK-resistant rPrP conformation results from

xi lipid-induced rPrP conformational change …………………………… 118 Relevance of lipid-PrP interaction to in vivo prion biology …………... 120

Aggregation, but not formation of amyloid fibers, stabilizes the

C-terminal PK-resistance ………………………………………………………... 122 Conclusion and future directions ……………………………………………………... 123

References ……………………………………………………………………………………………………….. 125

xii LIST OF TABLES

Page

Chapter 1

Table 1.1 Comparison between PrPC and PrPSc ……………………………... 24

Chapter 4

Table 4.1 Lipids used in binding assay ……………………………………………. 62

xiii LIST OF FIGURES

Page

Chapter 1

Figure 1.1 Spongiform change in sCJD ……………………………………………….. 5

Figure 1.2 Biogenesis of mouse PrPC …………………………………………………... 18

Figure 1.3 Cartoon of the three-dimensional structure of the intact human prion protein, hPrP (23–230) ……………………….. 22 Figure 1.4 Trimeric model of PrP27–30 built by superimposing three monomeric models …………………………………………… 25

Chapter 4

Figure 4.1 Interactions between rPrP and lipid vesicles ……………………. 58

Figure 4.2 Thin layer chromatography (TLC) analyses of lipids extracted from N2A cells ……………………………………………. 60 Figure 4.3 Interactions between rPrP lipids with different charges ….. 63

Figure 4.4 Conformational change and PK-resistance of rPrP induced by anionic lipid interaction ………………………….. 66 Figure 4.5 Effect of salt at physiological concentration on rPrP conformation ………………………………………………………………... 68 Figure 4.6 Identity of PK-resistant rPrP ………………………………………………... 69

Figure 4.7 Characterizing the PK-resistant PrP conformation ………….. 72

xiv Figure 4.8 Negative charges on lipid vesicles affect rPrP binding and conformational change ………………………………………. 75 Figure 4.9 Freeze fracture EM analysis of lipid vesicles prepared by sonication ……………………………………………………………………... 76 Figure 4.10 PK digestion of rPrP incubated with 600 nm diameter liposomes composed of POPC and POPG at a 7:3 molar ratio …………………………………………………………………….. 77 Figure 4.11 Lipid headgroup structures affect rPrP conformation ……… 79

Figure 4.12 Different lipid compositions affect rPrP conformation ……... 80

Figure 4.13 The C-terminal PK-resistant rPrP fragment is associated with lipid hydrophobically …………………………………………... 82 Figure 4.14 The hydrophobic domain of PrP is essential for generation of C-terminal PK-resistance ………………….. 84 Figure 4.15 Adjacent location of anionic charge and the hydrophobic lipid core are necessary for generation of C- terminal PK-resistance ………………………………………………. 86 Figure 4.16 Aggregation stabilizing the C-terminal PK-resistant rPrP conformation ………………………………………………………………... 88 Figure 4.17 C-terminal PK-resistant rPrP conformation that formed with a lower lipid:rPrP ratio ……………………………………….. 92 Figure 4.18 Morphology of rPrP aggregates ………………………………………….. 94

Figure 4.19 Arachidonic acid (AA) induced rPrP conversion ……………… 97

Figure 4.20 Effects of lipid oxidation generation of C-terminal PK- resistance ……………………………………………………………………... 99 Figure 4.21 Effects metal ions on generation of C-terminal PK-

xv resistance …………………………………………………………………….. 101 Figure 4.22 Effects of RNA on generation of C-terminal PK- resistance …………………………………………………………………….. 103 Figure 4.23 Illustrations of wild type recombinant human PrP, mutants ΔKKPRK and P105L, and polymorphism at codon 129 ………………………………………………………………... 105 Figure 4.24 Middle region of PrP affects the formation of C-terminal PK-resistance ………………………………………………………………. 107 Figure 4.25 The 129 polymorphism influences rPrP-lipid interaction … 110

Supplemental The same blot in Figure 4.14B was reprobed with 8H4 Figure 4.1 antibody ………………………………………………………………………... 112 Supplemental The same blot in Figure 4.15 was reprobed with 8B4 Figure 4.2 antibody ………………………………………………………………………... 113

xvi LIST OF ABBREVIATIONS

TSE transmissible spongiform encephalopathy

CJD Creutezfeldt-Jakob disease vCJD variant Creutezfeldt-Jakob disease

GSS Gerstmann-Sträussler-Sheinker disease

FFI fatal familiar insomnia

BSE bovine spongiform encephalopathy

CWD chronic wasting disease

PrP prion protein rPrP recombinant prion protein cyPrP cytosolic PrP

PrPC cellular isoform of the prion protein

PrPSc scrapie isoform of the prion protein

TEV tobacco etch virus

PK proteinase K

PI-PLC phosphatidylinositol-specific phospholipase C

GPI glycosylphosphatidylinositol

ER endoplasmic reticulum

PNS post-nuclear supernatant

CD circular dichroism

xvii NMR nuclear magnetic resonance

FTIR fourier transform infrared

PMCA protein misfolding cyclic amplification

TLC thin-layer chromatography

TEM transmission electron microscopy

BME β- mercaptoethanol

EDTA ethylenediaminetetraacetic acid

SDS sodium dodecyl sulfate

GdmCl guanidinium chloride

SDS-PAGE sodium dodecyl sulfate-polyacrylamide gel electrophoresis

POPG 1-palmitoryl-2-oleoylphosphatidylglycerol

POPS 1-palmitoryl-2-oleoylphosphatidylserine

POPC 1-palmitoryl-2-oleoylphosphatidylcholine

POPA 1-palmitoryl-2-oleoylphosphatidic acid

DOTAP 1,2-dioleoyl-3-trimethylammonium propane

AA arachidonic acid

xviii CHAPTER 1

INTRODUCTION

Prion diseases, also know as Transmissible Spongiform Encephalopathies

(TSEs), are a group of invariably fatal neurodegenerative disorders that occur in

a wide variety of mammals [1-3]. According to the “Protein-Only” hypothesis,

prion is the infectious agent, which is mainly composed of the abnormal isoform,

PrPSc, of the normal cellular prion protein, PrPC. PrPSc converts PrPC into a likeness of itself, resulting in neurodegeneration [1-5]. Supporting this hypothesis,

PrPC purified from wild-type hamster brain, in the presence of co-purified lipid

molecules and synthetic polyanion molecules, has been demonstrated to be

sufficient to form PK-resistant, infectious PrP molecules de novo [5]. However,

the precise mechanism for PrP conversion in prion diseases is not fully

understood.

1.1 Prion diseases

Prion diseases can be manifested as sporadic, inherited (familial) and

infectious (acquired) disorders, including Kuru, Creutezfeldt-Jakob disease (CJD),

Gerstmann-Sträussler-Sheinker disease (GSS), and Fatal Familiar Insomnia (FFI)

1 in humans, Scrapie in sheep, Bovine Spongiform Encephalopathy (BSE) in cattle,

and Chronic Wasting Disease (CWD) in deer and elk, all of which involve

conformational change of the mammalian prion protein (PrP) [1-3].

1.1.1 Human prion diseases

In 1921, German neurologists, Hans Gerhard Creutzfeldt and Alfons Maria

Jakob, reported the first cases of human prion disease, which later became

known as “Creutzfeldt-Jakob” diseases (CJD) [6]. These cases were later

demonstrated as familial CJD, caused by a mutation in the prion protein gene,

Prnp [7, 8]. Thereafter, more and more inherited prion diseases, including different forms of familial CJD, classical and variant forms of Gerstmann-

Sträussler-Scheinker syndrome (GSS), and fatal familial insomnia (FFI) were discovered and linked to pathogenic mutations and insertions in Prnp gene [9].

Until now, more than 30 mutations in PrP have been linked to inherited prion disease in human [10].

Kuru, an endemic form of CJD, transmitted in the Fore people in Papua

New Guinea through ritual cannibalism [11, 12], was the first identified infectious

form of human prion disease [13, 14]. Iatrogenic acquired (infectious) form of

CJD (iCJD) was first reported in 1974 in a corneal transplant recipient [15]. Later,

more iatrogenic cases have been found resulting from dura mater grafts,

2 administration of pituitary hormone derived from deceased individuals with

unrecognized prion disease, or the use of contaminated instruments in the neurosurgical operations [16]. The emergence of a new variant form of CJD

(vCJD) in the United Kingdom in 1996 [17] drew a substantial attention from the public, largely due to its potential etiological relation to the Bovine Spongiform

Encephalopathy [18-21] that burst out in the UK since 1986. Recently, the transmission of CJD via blood transfusion has been confirmed in several cases, which quickly became another public focus [22].

The most prevalent form of human prion disease is the sporadic form of

CJD (sCJD), which composes of about 85% of human prion disease (the

inherited form accounts for about 15% and the acquired (infectious) form

accounts for less than 1%). Despite of its prevalence, the sCJD appears

worldwide with an incidence of approximately 1 case per million people per year

[23]. The etiology of sCJD is still unknown, which is not associated with any endogenous (mutations in Prnp gene) or exogenous factors (chemicals, et al.)

For human prion diseases, the clinical features include rapidly progressive

dementia, myoclonus, visual or cerebellar signs, pyramidal/extrapyramidal signs,

and akinetic mutism [24]. And neuropathological hallmarks such as spongiform

degeneration, astrocytic gliosis, neuronal loss, synaptic alterations, and

accumulation of PrPSc characterize the brain damage caused by prion diseases

3 (Figure 1.1) [25-27]. The pathological alterations may vary with disease subtype

(sporadic, familial, or acquired), the nature of the Prng gene defect (in familial cases), the codon 129 polymorphism (MM, MV, or VV), the PrPSc strain type, and the duration of illness [28].

Figure 1.1: Spongiform change in sCJD: occipital cortex showing confluent spongiform change in deep cortical layers (hematoxylin and eosin (H&E)) [28].

5 1.1.2 Scrapie and Bovine Spongiform Encephalopathy

Scrapie is the prototype of prion disease that was found in ovines and caprines. Affected sheep would develop pruritus that induces rubbing and scratch (scraping) resulting in wool loss. And about 250 years ago, scrape was already recognized as a contagious disease in sheep [29]. Scrapie in sheep has an incubation period of 2 to 5 years, with an average age of 2.5 years for onset of clinical symptoms [30].

In 1984, Bovine Spongiform Encephalopathy (BSE) was first recognized as a neurodegenerative disease in cattle and was specifically diagnosed in 1987

[31]. Since 1986, approximate 200,000 BSE cases have been confirmed in the

United Kingdom and a lot more (between 800,000 to 1.2 Million) cattle, which are estimated to be infected with BSE, have been slaughtered before the onset of clinical symptoms [32, 33]. Till 2004, BSE has been identified in up to 24 countries, including United States, most of which distribute in Europe [31].

Clinical features of BSE in cattle can be manifested by abnormal gait and stumbling, loss of body weight, hyperresponsiveness to stimuli, aggression, tremors [34], and the average age for the onset of clinical signs is 3.5 years, with an incubation period varying from 2 to 8 years [31]. Epidemic studies of BSE indicate a “common source”, which is most likely the practice of feeding cattle with rendered meat and bone meal (MBM) contaminated with bovine brain and

6 spinal cord from infected animals. Semen, chemicals, autosomal inheritance, biologics, and pharmaceuticals have been ruled out as the common source [35-

37]. After banning of feeding ruminants with meat and bone meal from ruminants in 1988 in UK, the BSE epidemic peaked in 1992 and, since then, has declined significantly [33].

Etiology studies of BSE led to several hypotheses about the origin of this disease. One theory postulates that the origin of BSE already existed as undetected sporadic cases for a long time before 1980s, and spread in a large scale coincidently with changes in processing meat and bone meal in most rendering factories in UK [36, 37]. An alternative theory favors BSE originated from Scrapie-infected meat and bone meal, which were recycled for feeding cattle and eventually adapted and resulted in current BSE. Another theory hypothesizes that BSE appeared as a new prion disease in 1980s due to the practice of low-temperature rendering in industry and introducing meat and bone meal into the cattle feed. The fourth one suggests that a new Scrapie strain arose and readily transmitted to cattle [31]. While the origin of BSE remains obscure, more and more evidence supported the linkage between human vCJD and BSE since the first case reported in UK in 1996 [18-21]. To date, only about

200 cases of vCJD have been identified worldwide and the incidence seems to decline steadily after reaching the peak in 2001 [10], However, since more

7 individuals have been exposed to BSE and these individuals may carry the

infectivity without clinical symptoms, further monitoring and more research are

crucial to address the public concern.

1.1.3 Chronic Wasting Disease

Chronic wasting disease (CWD) was first recognized as a fatal wasting

disease of captive mule deer in Colorado in 1967 [38] and has been identified as

the only prion disease found in free-ranging species, including mule deer, white

tail deer, and elk. To date, CWD has been found in 12 states in US and 2

provinces in Canada in both farm-raised and wild cervid populations [31]. The clinical symptoms of CWD include behavioral changes, progressive weight loss, ataxia, and emaciation (wasting) ultimately leading to death [38, 39]. The incubation period of CWD in elk ranges from 2 to 8 years, and after onset of

clinical symptoms, animals usually survive from 5-12 months before death [40].

Neuropathological characteristics can be manifested as neuronal degeneration

and spongiform changes, intracytoplasmic vacuoles in neurons, and astrocytic

hypertrophy and hyperplasic with occurrence of amyloidal plaques [41]. Like BSE,

the origin of CWD remains unknown, while several hypotheses have been

proposed. One theory suggests that CWD arises as a sporadic prion disease and

spreads through horizontal transmission supported by studies on captive deer

8 and elk [38, 42, 43]. Another theory favors the transmission of Scrapie from

sheep to cervids, evidenced by experimental inoculation of Scrapie into elk

resulting in a disease indistinguishable from CWD [44]. Although transgenic

mouse studies have suggested a species barrier between cervids and humans

[45, 46], in vitro conversion of human PrP into a protease-resistant form by CWD-

associated protease-resistant PrP [47], despite of its inefficiency, still arouse public concerns along with the possibility that CWD could cross human species

barrier given the linkage between BSE and vCJD.

1.2 The infectious agent in prion disease

1.2.1 History of prion research

Scrapie, the prototypic spongiform encephalopathy affecting sheep and

goat, was first reported in 1732 [31], although it is suggested that it had already

been present in northern Europe and Austro-Hungary before the beginning of the

18th century. In a German literature published in 1759, scrapie was recognized as

an infectious disease with detailed clinical description, which pulled up the curtain

of recorded history of scrapie. In 1898, scientists first reported neuronal

vacuolation as the characteristic neuropathological change in brains of scrapie

affected sheep. In the next year, the same group of scientists reported failed

attempts to transmit scrapie to healthy sheep by inoculation of brain and

9 transfusion of blood from diseased animals. Those negative results were largely

due to a short observation period while the incubation time could be

extraordinarily long. In 1936, under their persistent endeavor, Cuille and Chelle,

two French veterinarians, successfully transmitted scrapie to two healthy sheep by intraocular inoculation of brain or spinal cord tissue from an affected animal

[29]. In 1961, the success in transmitting scrapie to laboratory mice by Chandler relieved scientists from laborious work with sheep and goats [48].

After first report of CJD cases in early 1920s, Kuru, another human

transmissible spongiform encephalopathy, found among aboriginals in Papua

New Guinea, was reported by Gajdusek and Zigas in 1957 [11]. However, the first attempts to transmit Kuru to primates failed for the same reason as the earlier trials of transmitting Scrapie among sheep: observation period was not long enough. However, the similarities between Kuru and Scrapie, first pointed

out by American veterinarian William Hadlow [49], indicated a very long

incubation time for Kuru. Following this remarkable insight along with Igor

Klatzo’s suggestion that Kuru resembles CJD [50], Gajdusek and Gibbs

succeeded in transmitting Kuru and CJD to chimpanzees, respectively, in

consecutive years [13, 14, 51].

Due to it’s long incubation period, Scrapie was believed to be caused by

so-called “slow virus”, termed by Biorn Sigurdsson in 1954 [52], for a long time

10 until Tikvah Alper and her colleagues showed for the first time that scrapie

infectivity survived from inactivation by UV and ionizing radiation at the dose that would have demolished any nucleic acids-based pathogens [53-56]. This stunning observation led to several theories about the nature of the infectious agent, from membranes, membrane bound ligands, lipid-protein-polysaccharide complexes, to pure proteins [57-62]. Eventually, ineffective treatments to the infectivity of prion agent by harsh organic solvents [63-65], non-denaturing detergents, nucleases, proteases, glycosidases [66-68], or even steam autoclave

at 120°C [69] brought on the concept of “Prion”, defined as proteinaceous

infectious particles, by Stanley Prusiner in 1982 [70].

1.2.2 Purification of Prions

Stanley Prusiner initiated his work on CJD as he was beginning his

residency in neurology in 1972. Following forerunners’ suggestion that the

infectious scrapie agent might solely consist of proteins, Prusiner focused his

research on purifying the scrapie infectivity by virtue of differential centrifugation

[71]. Through characterization of the sedimentation properties of scrapie

infectivity in brain homogenates of scrapie affected hamster, Prusiner and co-

workers partially purified the infectious agent of prion diseases [72] and further

enriched the infectious agent about 400 folds, which allowed them to identify a

11 single, partially protease-resistant protein species of 27-30kDa generated by

limited Proteinase K (PK) digestion [73, 74]. PrP27-30, designated from its

molecular mass of 27-30kDa, was further confirmed to be only present in the

infected brain tissues but not in parallel preparations from non-infected animals,

and sequenced for its N-terminus amino acid sequence by Edman degradation

[7]. Soon after the determination of the N-terminal sequence of PrP27-30, the

gene encoding prion protein was cloned and found in the host animals [8] and

the mRNA of PrP was found to be present at similar levels in both normal healthy and scrapie infected animals [75]. However, the cellular form of prion protein, termed PrPC, was demonstrated to be sensitive to protease digestion and soluble in mild detergents, unlike PrP27-30, which is insoluble in non-denaturing

detergents and represents the N-terminus truncated, protease resistant core of a

larger molecule, designated PrPSc [8, 76, 77].

Due to its insolubility in mild detergents and tendency to form fibril, PrPSc is always found to be highly aggregated when extracted from infected brains, and, because of which, the high-resolution tertiary structure of PrPSc is still unavailable.

However, the Fourier Transform Infrared (FTIR) spectroscopic studies showed

that PrPSc, in sharp contrast to PrPC, has high content of β-sheet [78-80], though

PrPSc is covalently indistinguishable from PrPC [80, 81]. The high β-sheet feature

is consistent with its highly aggregated and protease-resistant state. Hence, the

12 Proteinase K resistance of PrPSc is traditionally linked to the presence of prion

infectivity [82, 83], making it a good marker for disease, although in some

experimental setups, PK-resistance cannot be detected in samples that contain

prion infectivity [84, 85].

1.2.3 The “Protein-only” hypothesis

Based on the initial findings of Alper [53, 54], Griffith proposed the

“protein-only” hypothesis for the first time in 1967 [57]. In 1982, Stanley Prusiner

enunciated the “Protein-Only” hypothesis by terming the concept of “Prion”,

derived from proteinaceous and infectious, after finding that scrapie infectivity

could be reduced by procedures hydrolyzing or modifying proteins but was

resistant to measures affecting nucleic acids [70]. According to the “protein-only”

hypothesis, prion is the transmissible particle, which is solely composed of the

abnormal isoform, PrPSc, of the normal cellular prion protein, PrPC, and PrPSc

C converts PrPP into a likeness of itself [1]. Since then, efforts in finding nucleic

acids in the infectious agent of prion diseases have never been stopped [86-91].

However, more and more results from those studies showed to favor the “protein-

only” hypothesis.

Despite of being the major component in the highly purified infectious

preparation, PrPSc is not the sole ingredient; about 1.5% lipids are co-purified

13 with PrPSc [92]. Even though they are not believed to represent the infectivity,

lipids have been demonstrated to be involved in the infectivity of scrapie; 1)

Inactivation of prion infectivity by UV-irradiation became more efficient in oxygen-

saturated water, conditions that target lipids [93]; 2) The prion infectivity could be

increased by about two orders of magnitude by dispersion of purified infectious

agent in liposomes or detergent-lipid-protein-complex [94]; and, 3) the

membrane-associated PrPSc infects cultured cells with higher efficiency than

detergent-purified PrPSc [95].

1.2.4 Prion strains and species barrier

Two intriguing features, prion strains and species barrier, in transmission

of prion disease make “protein only” hypothesis encounter difficulty in interpreting

those phenomena.

Prion strains are variations of prions, which reside in the same host species that exhibit distinct disease phenotypes, including different incubation times, histopathological damage patterns, and distribution of prion deposits in the brain [24]. Some biochemical traits can also be utilized to distinguish prion strains, such as the stability of PrPSc under denaturing conditions, patterns of PrPSc glycosylation and size of protease-resistant core of PrPSc [24]. Explanation

corresponding to “protein only” hypothesis suggests that diversity of prion strains

14 comes from the flexible PrPSc conformations, each of which could induce the identical host normal prion protein to various conformations with distinct properties and cause diseases. However, the final proof that PrPSc conformation variations represent the biological basis of prion strains is still missing.

Species barrier is usually observed in the transmission of prion disease from species A to species B. In species B, the transmission efficiency of prion A is much less than that in the species A, marked by a prolonged incubation time

[96]. Both the difference between primary structures of donor and recipient prion protein and the type or strain of the donor prions account for the barrier between different species. Wild type mice possess the barrier to the transmission of hamster prion, while the transgenic mice expressing hamster PrP overcome the barrier and are highly susceptible to hamster prion [97]. In contrast to the strong species barrier, the BSE prion can efficiently transmit the disease to a variety of species without losing its biological properties, even through an intermediate species [98]. Supporting this notion, classic CJD prions can hardly transmit to wild type mice, while the transmission of vCJD prions to wild type mice is much more efficient [20].

15 1.3 The cellular prion protein, PrPC

1.3.1 The prion protein gene, Prnp

Determination of N-terminus sequence of PrP27-30 by Edman

degradation allowed the identification of the prion protein gene, designated as

Prnp, which is a single copy chromosomal gene [7, 8]. The human prion protein gene contains three exons, but the entire protein-coding region consists in the second exon [99]. To date, more and more PrP genes have been identified and shown to be highly conserved in a wide variety of mammalian species with more than 90% sequence identity [100, 101].

1.3.2 Cell biology of PrPC

The cellular prion protein is constitutively and primarily expressed in the

neurons of the central nervous system [102] and, with a lower level, in several

peripheral tissues, such as heart, skeletal muscle, liver and the lymphoreticular

system [103, 104]. Mouse PrP is composed of 254 amino acids, containing an N-

terminal signal sequence (AA 1-22) leading it to the secretory pathway, a series of five Proline- and Glycine-rich octapeptide repeats (AA 51-90), a central

hydrophobic region (AA 111-134) that is highly conserved, and a C-terminal

signal peptide (AA 231-254) for a glycosylphosphatidylinositol (GPI) anchor

addition (Figure 1.2).

16 The N-terminal signal peptide directs the protein into the lumen of endoplasmic reticulum (ER), where it is cleaved off the primary translation

product. While in the ER, PrP is further N-glycosylated at Asparagine residues

180 and 196, and the protein’s single disulfide bond is formed then by the only

two Cysteine residues 178 and 213. Upon attachment of a GPI anchor to Ser230,

the C-terminal peptide is split off the prion protein (Figure 1.2). PrP then follows the secretory pathway via Golgi apparatus and secretory vesicles to the cell surface [105], where it is attached to the cell membrane out leaflet by its GPI- anchor that targets PrP to so called raft membrane subdomains [106], which are membrane regions rich in cholesterol, sphingolipids and saturated phospholipids

[107-109]. The prion protein at cell surface readily undergoes endocytosis and recycling [110, 111]. Therefore, beside its distribution on cell surface, PrPSc is

also present in late endosomes or lysosomes demonstrated by

immunofluorescence and electronic microscopic studies [112, 113]. Since PrPC and PrPSc share subcellular localization in some cellular compartments, it has

been proposed that conversion from PrPC to PrPSc might occur in late

endosomes or lysosomes [114-116].

Figure 1.2: Biogenesis of mouse PrPC.

18 1.3.3 Structure of PrPC

Studies on PrP structure were performed with recombinant PrP expressed

in E. coli. Purified recombinant PrP is a monomer that harbors a single disulfide

bond and shows a circular dichroism (CD) spectrum of a high α-helical

conformation. The high degree of similarity in secondary structure of recombinant

PrP and native PrPC purified from brain tissue, which contains 42% α-helix and

only 3% β-sheet [76], suggests that the recombinant PrP most likely resembles

the PrPC isoform. Further spectroscopic properties of recombinant PrP, are also

in complete agreement with those of the cellular isoform of the prion protein,

PrPC, indicating that recombinant PrP adopts the same three-dimensional

structure as PrPC [80, 117, 118].

To date, the solution structures, studied by nuclear magnetic resonance

spectroscopy (NMR), of murine, hamster, human and bovine recombinant PrPs have been published (Figure 1.3) [119-122]. In accordance with the high sequence identity of the DNA level of about 90% among all mammalian prion protein sequences [100], the structures of all prion proteins are very similar,

except some local differences. All proteins consist of a highly flexible, unstructured N-terminal domain (residues 23-124) and a globular, folded, C- terminal domain (residues 125-231), which contains three α-helices and a short

anti-parallel β-sheet. The helices 2 and 3 are linked by a single disulfide bond. In

19 the flexible N-terminal fragment, there are five octapeptide repeats rich in Glycine

and Proline residues. This domain is believed to be responsible for binding

copper ions due to the coordination among the histidine residues in these

repeats [123].

The mutations found in inherited prion diseases were proposed to

decrease the thermodynamic stabilities of PrPC, facilitating its conversion to

PrPSc and increase the stability of PrPSc [124-126]. Thermodynamic

measurements on the corresponding protein variants, however, detected a

destabilizing effect only on a few – but not all – of these mutations [127, 128].

The NMR structure of disease-associated E200K mutant was recently solved and

showed no structural differences between the mutant and wild-type PrP. The only

difference was the surface charge distribution, which may imply that an

abnormality in its interaction with auxiliary charged molecules might be the cause

of disease [129].

1.3.4 Biological functions of PrPC

Attempts to understand the biological function of PrPC resulted in mice

with disruption in the opening reading frame of Prnp, designated Prnp0/0, which developed normally compared to wild type animals [130]. The absence of obvious phenotype in PrP knockout mice leaves the physiological function of PrP

20 obscure, although a wide body of studies suggest PrP’s roles in numerous cellular processes, such as anti-apoptotic function, protection against oxidative stress, formation and function of synapse, neurite outgrowth and maintenance of white matter [10].

Figure 1.3: Cartoon of the three-dimensional structure of the intact human prion protein, hPrP(23–230). The helices are orange, the β-strands cyan, the segments with nonregular secondary structure within the C-terminal domain yellow, and the flexibly disordered “tail” of residues 23–121 is represented by yellow dots [24].

22 1.4 Conversion of PrPC to PrPSc

1.4.1 Comparison between PrPC and PrPSc

PrPC and PrPSc, encoded by the same gene, Prnp, are the isoforms of the

prion protein [131]. Although, they share the same primary structure and probably the same posttranslational modification [81, 132, 133], PrPC and PrPSc can be readily differentiated according to the remarkable difference at the level of secondary structures. According to the FTIR and CD spectroscopy studies, PrPC contains 42% α-helix and only 3% β-sheet [76], while PrPSc is composed of about

30% α-helix and 45% β-sheet [80]. NMR study of recombinant prion proteins

confirmed the structural properties of PrPC, while, due to its tendency to

aggregate, the tertiary structure of PrPSc has not been determined. Recent

studies, by virtue of electron crystallography technique, proposed that PrPSc is

composed of parallel left-handed β-helical trimers (Figure 1.4) [134, 135]. In

addition to the biophysical contrasts, PrPC and PrPSc can also be distinguished

based on their biochemical properties. PrPC is soluble in mild detergents and

sensitive to Proteinase K (PK) digestion. On the other hand, PrPSc is highly

aggregated and remains insoluble in nondenaturing detergents, and possesses a

partial resistance to PK digestion (Table 1.1) [76].

According to the “Protein-Only” hypothesis, prion consists principally or entirely of PrPSc that acts as a template to promote the conversion of PrPC to

23 PrPC PrPSc

42% α-helix, 3% β-sheet 30% α-helix, 45% β-sheet

Soluble in mild detergents Insoluble in mild detergents

PK sensitive Partial PK resistant

Table 1.1: Comparison between PrPC and PrPSc

Figure 1.4: Trimeric model of PrP27–30 built by superimposing three monomeric models [135].

25 PrPSc and this conversion involves only conformational change [1]. Although a

growing body of evidence supports this hypothesis [2-5], until the wildly accepted proof that the biosynthetic PrP could be converted to the infectious form under

defined conditions is obtained, the existence of other components required for

infectivity can not be excluded.

1.4.2 Studies on PrP conversion

Prion-seeded conversion

In vitro, both purified PrPC and crude brain homogenates have been

employed to investigate the mechanism of prion protein conformational change in

prion diseases [136, 137]. In a cell-free system, PrPSc-like conformation was

generated with very low efficiency when PrPSc and purified PrPC were mixed

together [136]. Most recently, Caughey and coworkers demonstrated that

recombinant hamster PrP could be converted to protease resistant PrP (PrPres), when seeded with diseased hamster brain homogenate, by using a technique termed Protein Misfolding Cyclic Amplification (PMCA). However, whether prion infectivity was associated with the newly formed PrPres molecules was unclear

[138]. Whereas, by using PMCA, both PrPSc and the prion infectivity were

successfully propagated from the crude hamster brain homogenates after

repeated cycles of sonication in the presence of the anionic detergent sodium

26 dodecyl sulfate (SDS) [137, 139]. When purified PrPC was used as substrate in

PMCA amplifications, decreased amplification efficiency was observed [140].

The significantly higher efficiency in crude brain homogenates than in purified preparations indicates that additional facilitating factors might be required for efficient PrP conversion. Indeed, several studies have suggested that specific

RNA molecules [141], proteoglycans [142], and/or lipid membranes [95] might play an important role in conversion of PrPC into PrPSc.

Further studies of the cell free conversion assay revealed an important

role of lipid in PrP conversion [143]. Both GPI-anchored PrPC and GPI-anchor-

deficient PrPC could be reconstituted into raft-like lipid vesicles, indicating a GPI- anchor independent association between GPI-anchor-deficient PrPC and lipids

[143]. More importantly, when mixed with exogenous PrPSc, cell free conversion only occurred with GPI-anchor-deficient PrPC. For GPI-anchored PrPC, it had to

be released from lipid vesicles by treatment with phosphatidylinositol-specific phospholipase C (PI-PLC) to allow it to convert [143]. These results suggest GPI- anchor-independent PrP-lipid interaction plays an important role in PrP conversion.

De novo formation of PrPSc

In vitro conversion assays have successfully reproduced the PrPSc-like

27 molecules and prion infectivity, mimicking the infectious cases of prion diseases

[137, 139]. In the absence of seed PrPSc molecules, how PrPSc and prion infectivity spontaneously initiate in sporadic and inherited prion diseases is largely unknown. Many studies have shown the formation of PrP amyloid fibrils in vitro under various partial denaturing conditions, e.g. acidic environment [144,

145], 5 M urea [145, 146], denaturing detergent [147], or high temperature [148].

However, little PK-resistance has been associated with these PrP amyloidogenic conformations and, more importantly, no infectivity has been generated either.

Most recently, Supattapone and colleagues startlingly generated infectious prion de novo, showing that the minimal components required to produce both protease resistant PrP and prion infectivity are PrPC, co-purified lipids, and single-stranded polyanions [5]. Although this brilliant work strongly supports the

“protein-only” hypothesis, it is still being challenged by arguing that the infectivity may be generated by the component(s) co-purified with PrPC from brain tissues.

The widely accepted proof for prion hypothesis, that de novo generation of prion by bacteria expressed recombinant PrP (rPrP), has not been achieved.

Because E. Coli expressed rPrP is purified through 6M guanidine hydrochloride treatment [149], any co-purified molecules will be eliminated. So use of E. Coli expressed rPrP will offer a clean system to study de novo PrP conversion.

The above finding that de novo generation of infectious prion in vitro

28 involves co-purified lipids strongly indicates the important role of lipid for PrP conversion, which is also in accordance with several previous studies. One in vitro study showed that recombinant PrP was able to bind negatively charged phosphatidylserine vesicles and then change its conformational structure and stability [150]. In another study, recombinant PrP (90-231) was also found to bind to variety of lipid vesicles, resulting in secondary structure change [151]. No protease resistance has been reported with the lipid interacting, conformationally changed PrP in any of these studies.

These findings support the relevance of the in vitro investigation of interaction between E. Coli expressed recombinant PrP and purely synthesized lipids to the de novo formation of PrPSc.

29 CHAPTER 2

AIMS OF THESIS

The primary aim of this thesis was to investigate the role of lipids in the

formation and propagation of prion by virtue of biochemical approaches.

Implications of these studies in the pathogenesis of prion disease were

discussed.

In the first part of this thesis, conformational changes of recombinant PrP

(rPrP) upon binding lipids were explored and formation of PrPSc-like molecules

was confirmed. Under physiologically relevant conditions, mouse rPrP was

demonstrated to be able to bind anionic phospholipids and undergo

conformational change, converting from a mainly α-helical, Proteinase-K (PK)

sensitive structure to a high β-sheet, C-terminal PK-resistant conformation.

The second part further investigated the properties of interaction between

rPrP and lipids. Both the highly conserved middle region of PrP and the hydrophobic core of lipids were found to be required for the lipid induced PrP

conversion.

The third part investigated the relevance of rPrP-lipid interaction to prion

biology. Our results showed that lipid oxidation, metal ions and RNA affect lipid-

induced rPrP conversion in a manner similar to their effects on PrPSc.

30 CHAPTER 3

MATERIALS AND METHODS

3.1 Construction of expression plasmids

The coding sequence of mouse PrP(23-230) was amplified by PCR using primers 23BamHI (5’ AAAGGATCCAAAAAGCGGCCAAGGCCTGAA 3’) and

230RHindIII (5’ CCCAAGCTTAAGATCTTCTCCCGTCGTAATA 3’) and ligated into pPROEX-HTb (Invitrogen) expression vector using BamHI and HindIII restriction sites, producing the recombinant expression vector pPROEX-HTb- mPrP(23-230). Another pair of primers 23F (5’ GAAAACCTGTATTTTCA-

GGGCAAAAAGCGGCCAAAG 3’) and 23R (5’ CCAGGCTTTGGCCGCTTTTT-

GCCCTGAAAATACAAGTTT 3’) were used to remove a DNA sequence coding 3 amino acids in front of the mouse PrP(23-230) sequence and only one amino acid, glycine, remained at the N-terminus of recombinant PrP(23-230) protein after tobacco etch virus (TEV) protease cleavage of the 6-Histidine linker.

The pPROEX-HTb-mPrPΔ111-131 expression vector was generated by

ligating coding sequences for mPrP(23-110) and mPrP(132-230) together into

the pPROEX-HTb expression vector. mPrP(23-110) and mPrP(132-230) coding

sequences were amplified by PCR with primers 23BamHI (5’ AAAGGA-

31 TCCAAAAAGCGGCCAAGGCCTGAA 3’) and DELN (5’ ACGTGCATGCTTGA-

GGTTGGTTTTTGG 3’) and primers DELC (5’ ACATGCATGCCATGAGCAGG-

CCCATGATCC 3’) and 230RHindIII (5’ CCCAAGCTTAAGATCTTCTCCCGT-

CGTAATA 3’), respectively. mPrP(23-110) coding sequence was cloned into

vector to produce pBlueScript(SK)-mPrP(23-110) vector. Amplified mPrP(132-

230) coding sequence was then cloned into pBlueScript(SK)-mPrP(23-110)

vector using HindIII and SphI restriction sites on both mPrP(132-230) coding

sequence and vector, generating pBlueScript(SK)-mPrPΔ111-131 vector. mPrPΔ111-

131 coding sequence was then cleaved off pBlueScript(SK)-mPrPΔ111-131 vector using restriction enzymes BamHI and HindIII and ligated into BamHI and HindIII digested pPROEX-HTb vector to form pPROEX-HTb-mPrPΔ111-131 expression

vector. The N-terminal DNA sequence for 3 amino acids was removed as

described above.

3.2 Expression and purification of recombinant proteins

Recombinant mouse PrPs were purified as previously described [149].

Buffer used in rPrP purification:

Buffer A: 10 mM Tris-HCl, 100 mM NaPO4, pH 8.0

Buffer B: 6 M GdmCl, 10 mM Tris-HCl, 100 mM NaPO4, 10 mM BME, pH

8.0

32 Buffer C: 10 mM Tris-HCl, 100 mM NaPO4, 50 mM Imidazole, pH 8.0

Buffer D: 10 mM Tris-HCl, 100 mM NaPO4, 500 mM Imidazole, pH 5.8

Buffer E: 10 mM NaPO4, pH 5.8

Buffer F: 10 mM NaPO4, pH 6.5

10x TEV reaction buffer: 0.5 M Tris-HCl (pH 8.0), 5 mM EDTA

Purification procedures are described below.

Step 1 1% E. Coli BL21 (DE3) bearing plasmid encoding recombinant mouse

PrP is inoculate into 150 mL LB liquid media with 100 μg/mL Ampicillin

and 34 μg/mL Chloramphenicol and incubated overnight with agitation at

250 rpm, 37°C.

Step 2 Next morning, 3% cultured E. Coli media is inoculated into 1000 mL LB

liquid media with 100 μg/mL Ampicillin and 34 μg/mL Chloramphenicol

and incubated at 250 rpm, 37°C till OD590=0.5-0.6. Then 1 mM IPTG is

added into LB media and cell culture is incubated at 250rpm, 37°C for 5

more hours.

Step 3 E. Coli cells are harvested by centrifugation at 5,000 rpm for 10 minutes

at 4°C and resuspended in buffer A. Resuspended cells are broken

through 4 rounds of sonication (on ice) with output=6 and duty

cycle=80% for 2 minutes/round. Cell lysates are rested on ice for 10

33 minutes between each round.

Step 4 Inclusion bodies in cell lysates are collected through centrifugation at

10,000 rpm for 30 minutes at 4°C and resuspended in 75 mL of buffer B.

Resuspended inclusion bodies are sonicated in buffer B till completely

solubilized and centrifuged at 10,000 rpm for 30 minutes at 4°C. Keep

the supernatant.

Step 5 Pre-equilibrate 30 mL of Nickel-nitrilotriacetic acid (Ni-NTA superflow)

agarose resin (Qiagen) with buffer B. Mix resin and supernatant from

step 4 and stir for 30 minutes at room temperature.

Step 6 Pour the resin into a column and wash the column with 120 mL buffer B.

Step 7 Apply 200 mL gradient of buffer B to Buffer A to the column.

Step 8 Wash the column with 75 mL buffer C to remove the impurities devoid of

the histidine tag.

Step 9 Elute the recombinant protein with buffer D.

Step 10 Combine the eluted fractions and dialyze against buffer E for 1.5 hours

(refresh buffer E every 30 minutes) and afterwards against water for

another 1.5 hours (refresh water every 30 minutes) using SnakeSkin

Pleated Dialysis Tubing (10,000 MWCO, Pierce).

Step 11 Stir the dialyzed protein solution in a beaker and apply 10x TEV reaction

buffer drop by drop and afterwards the recombinant TEV protease the

34 same way.

Step 12 Centrifuge the mixture at 2,000rpm for 10 minutes and incubate at 30°C

till 100% cleavage (Check the cleaving percentage by SDS-PAGE).

Step 13 Load the reaction mixture onto 30 mL of CM sepharose (CCF100,

Sigma) pre-equilibrated with buffer F and elute with 200 mL gradient of 0-

500mM NaCl.

Step 14 Collect the fractions and dialyze against water for 3 hours (refresh water

every 30 minutes) using SnakeSkin Pleated Dialysis Tubing (10,000

MWCO, Pierce).

Step 15 Aliquot 1 mL/tube of purified protein into 1.5 mL eppendorf tubes and

keep tubes in –80°C freezer.

Step 16 Check protein purity by SDS-PAGE.

Concentrations of PrP purified in this study were OD280=0.25-0.36,

reflecting different batches of purifications. The molar concentrations were

calculated using the Є280 molar extinction coefficients of 63,370 and 62,005 for mouse PrP(23-230) and mouse PrPΔ111-131, respectively, according to the

ExPASy Protein Server of the Swiss Institute of Bioinformatics

(http://us.expasy.org/tools/protparam.html).

Recombinant TEV used above was also expressed and purified using

buffers below.

35 Buffers used in rTEV purification:

Buffer G: 20 mM NaPO4, 100 mM NaCl, 1 mM BME, pH 7.5

Buffer H: 20 mM NaPO4, 100 mM NaCl, 1 mM BME, 40 mM Imidazole, pH

7.5

Buffer I: 20 mM NaPO4, 100 mM NaCl, 1 mM BME, 200 mM Imidazole, pH

7.5

Buffer J: 25 mM NaPO4, 200 mM NaCl, 2 mM EDTA, 1 mM BME, pH 8.0

Purification procedures are described below.

Step 1 1% E. Coli BL21 (DE3) bearing plasmid encoding recombinant TEV

protease is inoculated into 150 mL LB liquid media with 100 μg/mL

Ampicillin and 34 μg/mL Chloramphenicol and incubated overnight with

agitation at 250 rpm, 37°C.

Step 2 Next morning, 3% cultured E. Coli media is inoculated into 1000 mL LB

liquid media with100 μg/mL Ampicillin and 34 μg/mL Chloramphenicol and

incubate at 250 rpm, 37°C till OD590=1.0-1.5. 1mM IPTG is added into the

media and incubated at 250 rpm, 37°C for 5 more hours.

Step 3 E. Coli cells is harvested by centrifugation at 5,000 rpm for 10 minutes

and resuspended in 100 mL buffer G. Resuspended cells are broken

through 4 rounds* of sonication (on ice) with output=6 and duty cycle=70%

36 for 2 minutes/round. Cell lysates are rested on ice for 10 minutes between

each round). (*After each round of sonication, 10 μL of sample is taken and

centrifuged at top speed for 10 minutes to check the size and color of the

pellet. When pellet becomes very small and yellow colored, the pellet is

mostly composed of lipid membranes and almost 100% of the cells are

broken. This measure is to prevent oversonication, in which the pellet will

become large again due to denaturation of rTEV).

Step 4 The lysates are centrifuged at 10,000 rpm for 15 minutes at 4°C and the

supernatant is kept.

Step 5 Pre-equilibrate 30 mL of Nickel-nitrilotriacetic acid (Ni-NTA superflow)

agarose resin (Qiagen) with buffer G. Mix resin and supernatant from step

4 and stir for 3 hours at 4°C.

Step 6 Pour the resin into a column and wash the column with 500 mL buffer H.

Step 7 Elute rTEV with buffer I (During elution, incubate buffer I with resin for

proper time. Slow binding, slow elution).

Step 8 Add 2 mM EDTA and 1mM BME to the eluted fractions to keep rTEV

activity.

Step 9 Dialyze the fractions against buffer J for 1.5 hours (refresh buffer J every

30 minutes) using SnakeSkin Pleated Dialysis Tubing (10,000 MWCO,

Pierce).

37 Step 10 After dialysis, add 10% Glycerol into the purified TEV protease. Aliquot

rTEV into tubes and keep tubes in –80°C freezer.

Step 11 Check protein purity via SDS-PAGE.

The wild type and mutant recombinant human PrPs were obtained from Dr.

Man-sun Sy’s laboratory in Case Western Reserve University.

3.3 Isolation of membranes from cultured murine neuroblastoma N2A cells

Membranes of N2A cells were isolated following previously described

protocol with slight modification [152]. Cultured N2A cells were harvested and

collected at 4°C and then applied to homogenization at 4°C. The homogenate

was centrifuged at 2,000 rpm for 10 minutes at 4°C to get post-nuclear

supernatant (PNS). Then the PNS was applied to high-speed centrifugation at

100,000 g for 1 hour at 4°C. After centrifugation, supernatant was discarded and

membrane pellet was resuspended in proper buffer solutions.

3.4 Extraction of lipids from N2A cells or mouse brain tissues

One gram of N2A cells or mouse brain tissues were homogenized in 10 mL of methanol for 1 minute, followed by addition of 20 mL of chloroform to the

mixture and homogenization for an additional 2 minutes. The homogenate was

38 filtered, and the solid residues were resuspended in chloroform/methanol (2:1, v/v,

30 mL) and homogenized for 3 minutes. The extraction step was repeated once,

and filtrates were combined and washed with 0.25 volume of 0.88% KCl in water

followed by a wash with 0.25 volume of water/methanol (1:1). The purified lipids

in the bottom layer were collected, flushed with argon and kept at -20°C.

3.5 Preparation of lipid vesicles by sonication

Lipids in chloroform, including the lipids extracted from N2A cells, mouse brain and the lipids purchased from Avanti Polar Lipids, Inc., were dried under a stream of nitrogen at 42°C and then hydrated in 20 mM Tris-HCl buffer (pH 7.4)

to reach a final concentration of 2.5 mg/mL unless indicated. Hydrated lipids

were vortexed and sonicated in a cup-hold sonicator (Misonix Inc., model

XL2020) until clear. The lipid vesicles were kept under argon at 4°C.

3.6 Preparation of lipid vesicles by extrusion

to reach a final concentration of 2.5 mg/mL unless indicated. Freeze/thaw

extrusion procedures are described below.

39 Step 1 Vortex the hydrated lipids until all of the lipid film is suspended.

Step 2 Heat the tube under warm water stream from tap.

Step 3 Vortex and mix the hydrated lipids well (Large multilamellar vesicles (MLV)

are formed at this step).

Step 4 Crush dry ice and place it in a glass beaker.

Step 5 Carefully add methanol to the beaker until the level of the methanol is 1-2

inches from the top of the beaker

Step 6 With the MLV suspension tube capped, plunge the tube into the dry ice-

methanol mixture for 5 seconds.

Step 7 Vortex the tube quickly to allow a lipid film to form on the wall of the tube.

Step 8 Repeat steps 6 and 7 until all of the lipid is frozen as a film on the wall of

the tube.

Step 9 Place the tube in dry ice-methanol mixture for 1 minute.

Step 10 Thaw the frozen lipid film by warming the tube gently under cool water

from tap.

Step 11 Put the tube under warm water for 30 seconds

Step 12 Repeat steps 6-14 five more times

Step 13 Two 0.6 micron polycarbonate filters are placed shinny side up in the

Mini-Extruder (Avanti Polar Lipids, Inc.).

Step 14 Load the sample into one of the gas-tight syringes and carefully place

40 into one end of the Mini-Extruder.

Step 15 Place the empty gas-tight syringe into the other end of the Mini-Extruder.

Make sure the empty syringe plunger is set to zero; the syringe will fill

automatically as the lipid is extruded through the membrane.

Step 16 Gently push the plunger of the filled syringe until the lipid solution is

completely transferred to the alternate syringe.

Step 17 Gently push the plunger of the alternate syringe to transfer the solution

back to the original syringe.

Step 18 Repeat steps 9 and 10 nine more times (total of 20 passes through

membrane). In general, the more passes though the membrane, the more

homogenous the lipid solution becomes.

Step 19 The final extrusion should fill the alternate syringe. This is to reduce the

chances of contamination with larger particles or foreign material.

Step 20 Remove the filled syringe from the extruder and inject the lipid solution

into a clean glass tube.

Step 21 Flush the tube with argon and keep the tube at 4°C

3.7 Two-dimensional thin-layer chromatography (TLC) analysis of compositions of lipids extracted from N2A cells

Pour the first dimensional solvent (chloroform/methanol/concentrated

41 NH4OH (28%) (65/35/5)) into a TLC tank with lid to a depth of just less than 1cm.

Measure 1 cm from the bottom of a TLC plate and then turn the plate 90°

clockwise and measure 1 cm from the bottom of the plate. Rotate the plate 90°

counter-clockwise back and mark a spot as the origin at the intersection of two

lines. Load 100 μL of purified lipids in chloroform/methanol (2:1) onto the origin spot. The TLC plate was then placed in the first dimensional tank with the origin spot at bottom. Leave the plate undisturbed until the solvent runs to about 1 cm below the top of the plate. Then turn the plate 90° clockwise and place the plate into a second tank with the second dimensional solvent (with chloroform- acetone-methanol-acetic acid-water (100/40/20/20/10)) with a depth of just less than 1cm. Leave the plate undisturbed until the solvent runs to about 1 cm below the top of the plate. The plate was taken out of the tank and dried under nitrogen for 10 minutes. At the end, place the plate into a third tank filled with iodine vapor to visualize the separated lipid spots.

3.8 Gradient assays

For the discontinuous density gradient floatation assay, rPrP and lipid

vesicles (or membrane vesicles) were mixed together for 10 minutes at ambient

temperature. Then the rPrP-vesicle mixture was mixed with stock iodixanol

solution (OptiPrep, Axis-Shield PoC, AS) to generate 1 mL of 36% iodixanol

42 solution, and the mixture was loaded at the bottom of an ultracentrifuge tube. 1 mL of 31% iodixanol solution and 0.4 mL of 5% iodixanol solution were loaded on top of the sample sequentially. The gradient was centrifuged at 200,000 g for 3 hours at 4°C in a Sorvall micro-ultracentrifuge. Two-hundred-microliter fractions

(unless indicated) were collected from the top to the bottom, and proteins in all fractions were either precipitated by adding 200 µl of 40% trichloroacetic acid or directly applied to a 14% SDS-PAGE with part of the fraction. Precipitated proteins were sonicated in SDS-PAGE sample buffer and subjected to electrophoresis on SDS-PAGE. Proteins on SDS-PAGE were detected either by staining with coomassie brilliant blue G-250 staining solution or by immunoblot analysis. For high salt and pH extraction, 0.5 M NaHCO3 (pH 11.0), 1.5 M KCl or

1.5 M KCl+10 mM NaOH solution was added to rPrP-vesicle mixtures prior to the gradient analysis. For salt competition of lipid binding, rPrP was pre-equilibrated with various concentrations of KCl for 5 minutes at room temperature.

Immediately before mixing, KCl of same concentrations was added to lipid vesicles. The mixed rPrP-vesicle solutions were incubated at room temperature for 10 minutes prior to the iodixanol density gradient assay.

For the sucrose density gradient, a 2.4 mL portion of 5, 10, 15, 20, 30, 40, and 50% sucrose solutions prepared in PBS were loaded into centrifugation tubes to form a zonal gradient. rPrP-lipid samples were loaded on top of the

43 gradient and centrifuged at 100,000 g for 4 hours at 4°C. A total of 12 fractions

(200 µL/fraction) were collected from the top of the gradient and proteins were precipitated by the addition of 200 µL of 40% trichloroacetic acid. Precipitated

proteins were sonicated in SDS-PAGE sample buffer and subjected to

electrophoresis on SDS-PAGE followed by immunoblot analysis.

3.9 PrP lipid incubation

Only soluble rPrP after a 1-hour 100,000 g centrifugation was used for

analyses. For rPrP and lipid incubation, 300μL of rPrP (OD280 = 0.25) was mixed

with 100 μL of lipid vesicles (lipid concentration of 2.5 mg/mL except indicated).

Total mouse liver RNA was isolated with RNazol reagent (Tel-Test, Inc.)

according to the instruction provided by manufacturer. RNA (50 μg/mL), CuCl2

(10 μM), FeCl2 (10 μM) or FeCl3 (10 μM) were either added into rPrP before

mixing with lipids or added into PrP-lipid mixtures as indicated. PrP and lipid

mixtures were flushed with argon and incubated at 37°C for indicated time

periods. For preparation of oxidized Arachidonic acid, 100 μL Arachidonic acid (1

mg/mL) was incubated with 10 μM FeCl3 at 37°C for 24 hours. For agitation with

Triton X-100 treatment, the rPrP and lipid mixtures were incubated at 37°C for

the indicated times. After the incubation, 0.5% Triton X-100 was added and the

mixture was agitated at 37°C in an Eppendorf thermal mixer at 800 rpm for 16

44 hours unless indicated.

3.10 Proteinase K digestion

For all PK digestions, 10 μL of incubated samples were subjected to PK

digestion at 37°C for 30 minutes with the indicated PK:rPrP molar ratios. The

reaction was stopped by adding 5 mM phenylmethanesulfonyl fluoride (PMSF)

and keeping the mixture on ice for 10 minutes. One-tenth of PK-digested samples were subject to SDS-PAGE and immunoblot analyses with indicated antibodies.

3.11 Sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-

PAGE)

The procedures for preparing SDS-PAGE are described below.

Step 1 Assemble the gel casting system.

Step 2 Prepare the 14% separating gel solution by mixing 3.5 mL 40%

Acrylamide/Bis (29:1), 2.5 mL 1.5 M Tris-HCl (pH 8.8), 100 μL 10% SDS,

50 μL 10% ammonium persulfate (APS), 5 μL N,N N’ N’

tetramethylethylene diamine (TEMED), and 3.85 mL distilled deionized

water (ddH2O) together and degassing the mixture.

Step 3 Add the separating gel solution into the gel casting slot and overlay the

45 separating gel solution with 1 mL 75% ethanol.

Step 4 Allow the gel to polymerize for 45 minutes to 1 hour. Pour the ethanol

overlaying the gel and drain the excess ethanol with strips of filter paper.

Step 5 Prepare the 4% stacking gel solution by mixing 250 μL 40%

Acrylamide/Bis (29:1), 630 μL 0.5 M Tris-HCl (pH 6.8), 25 μL 10% SDS,

12.5 μL 10% APS, 2.5 μL TEMED, and 1.59 mL ddH2O together and

degassing the mixture.

Step 6 Overlay the separating gel with the stacking gel solution and place a

comb into the solution.

Step 7 Allow the gel to polymerize for 15 minutes and then remove the comb.

Step 8 Place the gel into buffer chamber and running gel buffer (3.0 g Tris, 14.4

g Glycine and 1 g SDS in 1 L ddH2O) is added into both the lower buffer

chamber and upper buffer chamber.

Step 8 Add 1 X sample (loading) buffer (65 mM Tris (pH 6.8), 5% SDS, 3% β-

mercaptoethanol (BME), 10% glycerol and trace Bromphenol Blue) into

samples and boil for 10 minutes.

Step 9 Load the sample into the wells on the gel.

Step 10 Run the gel at a constant current of 30 mA.

Step 11 Stop the electrophoresis when the tracker dye is ~ 1 cm above the end

of the glass plates.

46 3.12 Detection of prion protein by coomassie brilliant blue staining

Remove the gel from the buffer chamber and soak the gel in the

coomassie brilliant blue staining solution (2.5 g coomassie brilliant blue G-250,

100 mL acetic acid, 450 mL methanol and 450 mL ddH2O) for 1 hour at room

temperature on an orbital shaker. Then destain the gel in the destainnig solution

(100 mL acetic acid, 450 mL methanol and 450 mL ddH2O) for 1 hour at room temperature on an orbital shaker to visualize the protein bands.

3.13 Detection of prion protein by immunoblot

Remove the gel from the buffer chamber and soak the gel in transfer

buffer (3.0 g Tris, 14.4 g Glycine, 200 mL Methanol and 800 mL ddH2O). Cut the

Polyvinylidene Fluoride (PVDF) membrane with the same size as the gel. Soak

the PVDF membrane in Methanol for 15 seconds and then in ddH2O for 10

minutes. Set up the transferring unit and transfer the protein bands from gel to

the PVDF membrane at a constant current of 35 mA overnight.

Next morning, remove the PVDF membrane and block non-specific

binding sites by immersing the membrane in 5% milk (w/v), 0.1% Tween-20 in tirs-buffered saline buffer (TBST) for 1 hour at room temperature on an orbital shaker. Dilute the primary antibody at the ratio of 1:5000 in 2% milk (w/v) in

TBST. Incubate the membrane with the diluted primary antibody for 1 hour at

47 room temperature on an orbital shaker. Rinse the membrane for 3 x 5 minutes with 2% milk (w/v) in TBST. Diluted the secondary antibody at the ratio of 1:5000 in 2% milk (w/v) in TBST. Incubate the membrane with the diluted secondary antibody for 1 hour at room temperature on an orbital shaker. Rinse the membrane for 3 x 10 minutes with TBST. Remove the ECL Plus western blotting detection reagents (GE Health) from storage at 2-8°C and allow to equilibrate to room temperature before opening. Mix detection solutions A and B with a ratio of

40:1. Drain the excess TBST from the washed membrane and place the protein side up on a protecting sheet. Pipette the mixed detection reagent on to the membrane. Incubate the membrane and reagent for 5 minutes at room temperature. Drain excess detection reagent from the membrane and put the membrane on to a new protecting sheet and gently smooth out any air bubbles.

Place the membrane covered by the protecting sheet, with the protein side up, in an x-ray file cassette. Place a sheet of autoradiography film on top of the membrane. Close the cassette and expose for proper period of time. Develop the film to visualize the protein bands.

3.14 Phosphorus analysis

Mix 300 μL rPrP (OD280=0.25) with 100 μL POPG (2.5 mg/mL) at room temperature for 10 minutes. Apply 100 μL of the mixture to iodixanol gradient

48 assay. The procedures are described below.

-- Digestion

Step 1 Add 200 μL of the mixture into a 10 mL calibrated digestion tube.

Step 2 Add 200 μL buffer used in the mixture into two tubes as buffer blanks.

Step 3 Add 800 μL ddH2O into the sample and blank tubes.

Step 4 Add 1 mL KH2PO4 standard solutions at different concentrations (0, 10,

50, 100, 200, 400 μM) into 6 new tubes, respectively.

Step 5 Add 0.5 mL 70% Perchloric Acid to each tube.

Step 6 Vortex and mix the solutions thoroughly.

Step 7 Place tubes in 180°C heater block in a hood.

Step 8 Digest for 90 minutes.

-- Color development

Step 1 Remove tubes from the heater block and let them cool

Step 2 Add 2.3 mL Ammonium Molybdate solution (2.2 g Ammonium Molybdate,

7 mL concentrated H2SO4, 493 mL ddH2O) and vortex thoroughly.

Step 3 Add 0.1 mL ANSA (1-amino-2naphthol-4-sulfonic acid) solution (0.5 g 1-

amino-2naphthol-4-sulfonic acid, 30 g NaHSO3, 6 g Na2SO3, 250 mL

ddH2O), vortex thoroughly. 49 Step 4 Place tubes in boiling water for 15 minutes.

Step 5 Remove tubes and place them in cool water to bring to room temperature.

Step 6 Add ddH2O to bring the volume to 10.0 mL.

Step 7 Vortex tubes 10 times in short bursts to get uniform mixing.

-- Spectroscopy

Step 1 Set wavelength to 820 nm on a Beckman DU-40 Spectrophotometer.

Step 2 Record the readings of standards and samples.

Step 3 Subtract buffer blank from samples and water blank from standards.

Step 4 Curve fit standards using a quadratic equation fit.

Step 5 Calculate phosphorus concentrations in samples from curve fit equation.

3.15 Transmission electron microscopy

Twenty microliters of indicated samples was mixed with 2% glutaraldehyde and kept at room temperature for 5 minutes. Copper precoated

EM grids (Electron Microscopy Sciences) were layered on top of the sample drops for 5 minutes. The sample-containing grids were washed with distilled water, followed by staining with one drop of uranyl acetate (20 mg/mL) for 2 min at room temperature. Samples were analyzed on a Philips CM12 transmission electron microscopy at 80 kV with various magnifications.

50 CHAPTER 4

RESULTS

4.1 Introduction

The “protein only” hypothesis proposes that prion is the infectious agent,

which is composed of PrPSc, an abnormal, aggregated isoform of the normal host

cellular PrPC. Compelling evidence supports that the conversion from the normal

PrPC conformation to the pathogenic PrPSc isoform is intimately associated with the pathogenesis of prion disease [1, 153-155]. However, the mechanism underlying the conversion of α-helix-rich, protease sensitive PrPC to β-sheeted,

protease resistant PrPSc still remains unclear.

Several studies have investigated the conversion from PrPC to PrPSc in vitro [148, 156, 157]. A recent report showed that the anionic lipid bicelles converted α-helix-rich recombinant PrP to a β-sheet conformation with a 65°C heating step [148]. Addition of nonionic detergent to PrP and lipid bicelle mixtures induced amyloid fiber formation, and PrP in the amyloid fibrillar state has a PK- resistant core similar to PrPSc. This study clearly revealed a profound effect of

anionic lipids on PrP conformation, yet the requirement of the 65°C heating step for PrP conversion leaves the physiological relevance uncertain. In other studies,

51 α-helix-rich PrP was successfully converted to a β-sheet amyloidogenic

conformation with very limited protease resistance [156, 157]. Similarly, partial

denaturing treatments such as acidic environment or 5 M urea were employed to

facilitate the conversion process, presumably due to the energy barrier between

the two conformational states [146].

Since PrPSc forms in vivo under physiological conditions, some cellular

factors have to alleviate the energy barrier to facilitate PrP conversion. Several

candidates, including nucleic acids [141, 158], metal ions [159-161], sulfated

glycosaminoglycans [142, 162, 163], and lipids [153, 164], have been identified

as the putative facilitating factors for PrP conversion. Lipid is a plausible

candidate based on the fact that the GPI (glycosylphosphatidylinositol)-anchored

PrPC is on the surface of lipid membranes. More importantly, recent advances in protein folding field revealed that the interfacial region of lipid membrane is capable of inducing secondary protein structures [165, 166]. As mentioned in the

introduction, many experimental findings support the involvement of lipid in PrP conversion. Lipids have been identified in highly purified “prion rod” [92] and re- incorporation of “prion rod” into lipid vesicles results in a higher prion infectivity

[94]. In cultured cells, membrane-bound PrPSc was found to have a higher

infectivity compared to detergent extracted PrPSc [95]. Changing membrane lipid composition in prion-infected cells altered the yield of PrPSc [106, 167]. The

52 success of de novo formation of infectious prion with purified PrPC and co- purified lipid molecules [5] also supports the important role of lipid for PrP conversion. In addition, the resistance of PrPSc, but not PrPC, to

phosphatidylinositol-phospholipase C (PI-PLC) digestion is well documented [168,

169], which indicates that PrPSc, but not PrPC, has a GPI-anchor independent

interaction with lipid membranes. Supporting this notion, the GPI-anchor

independent PrP-lipid interaction has been found to be essential for PrP

conversion in the cell free assay [143].

Several biophysical studies have explored the GPI anchor-independent

recombinant PrP-lipid binding in vitro under native conditions. Their results

indicated that PrP-lipid interaction could induce conformational changes of

recombinant PrP [150, 151, 170, 171]. One study showed that, upon binding to

lipid vesicles, the N-terminal unstructured part of PrP becomes partially ordered

while the C-terminal structured region is destabilized [150]. A PrP fragment

(residues 90-231, PrP(90-231)) was found to interact with a variety of lipid

vesicles, resulting in secondary structure changes [151, 170, 171]. These results

suggested that the interaction with lipid membranes is capable of altering PrP

conformation under native conditions. In particular, the interaction with anionic

lipids induces α-helix-rich recombinant PrP(90-231) to gain β-sheet content,

which is similar to the PrP conformational change during prion disease [79, 80,

53 172, 173]. Very recently, an in vitro study showed that membrane anchored native PrP can be reconstituted to raft-like lipid vesicles through its GPI anchor, and high concentration of membrane anchored PrP induces PrP refolding, forming intermolecular β-sheet structures [174]. All these findings revealed that lipid interaction causes PrP conformational change, resulting in an increased content of β-sheets. Although the authentic PrPSc does have an increased β-

sheet content, the C-terminal Proteinase K (PK) resistance is a more widely used

and more stringent standard for the PrPSc conformation [1, 153-155]. Yet, no PK

resistance was reported in any of those studies.

Thus far, the exact role of lipid in PrP conversion remains unclear; which

is largely due to the difficulties in experimental methodologies. Bacteria

expressed recombinant PrP (rPrP) offers a practical approach to address this

difficult question. Besides the ease to obtain ample quantity, the guanidine

hydrochloride treatment during rPrP purification [149] eliminates any co-purified

lipid molecules and offers a clean system to study rPrP-lipid interaction. In

addition, rPrP amyloid fibrils were found to cause neurogedeneration and

generate prion infectivity in mice overexpressing the same PrP species [175].

Prion infectivity was successfully propagated in the PrPΔGPI transgenic mice, in

which the PrPΔGPI is very similar to rPrP, without the GPI anchor and mostly unglycosylated [176]. Moreover, the seeding ability and morphology of rPrP(23-

54 144) amyloid fiber provide an explanation for the peculiar prion strains and

species barrier [177, 178]. These findings support the relevance of investigating

rPrP conversion to in vivo prion biology.

In this study, I used rPrP and lipids of various compositions to characterize

whether rPrP-lipid interaction can induce a PrP conformation with the classic C- terminal PK-resistant pattern under physiological conditions. The answer to this question is crucial for the relevance of lipid-induced PrP conformational change to prion pathogenesis. Here, I showed that rPrP binds to anionic lipids via both electrostatic and hydrophobic interactions and the anionic lipid-rPrP interaction converts α-helical and PK-sensitive rPrP into a conformation with increased β–

sheet content and PK-resistance under native conditions. Similar to the

pathogenic PrPSc, the anionic lipid induced rPrP has a high PK-resistance and a

15 kDa C-terminal PK-resistant fragment that can be stabilized by aggregation.

The fact that lipid-induced PrP conversion occurs under an environment

reminiscent of physiological condition supports the relevance of previously

reported PrP-lipid interaction [148, 150, 151, 170, 174, 179] to the pathogenesis

of prion disease.

Subsequent characterization of lipid-PrP interaction showed that the lipid-

induced rPrP conversion depends on structure and/or arrangement of lipid

headgroups on the surface of lipid vesicles. More importantly, I showed here that

55 both the highly conserved middle region of PrP, which consists of a cluster of

positively charged lysine residues followed by a hydrophobic domain, and the

adjacent localization of anionic charges and hydrophobic lipid core of the lipid

vesicles are required for generation of C-terminal PK-resistant rPrP. This

conclusion was supported by the study of interaction between lipid and rPrP

carrying pathogenic mutations. In studies of factors known to alter PrPSc propagation or stability, I showed that lipid oxidation, metal ions and RNA affect lipid-induced recombinant PrP conversion in a manner similar to their effects on

PrPSc, which provide additional supporting evidence for the relevance of PrP-lipid

interaction to the in vivo PrP biology.

4.2 Lipid interaction converts recombinant prion protein to a protease

resistant conformation under physiological conditions

4.2.1 Binding of recombinant PrP to lipids involves both electrostatic and

hydrophobic interactions

High levels of cytosolic PrP (cyPrP) in cultured cells have been shown to

be able to form the PrPSc-like aggregates, which are resistant to Proteinase K

(PK) digestion, in the presence of mild detergents [180, 181]. Previous findings in

our lab showed that cyPrP bound to brain lipid membrane and full-length

56 recombinant mouse PrP(23-230) (rPrP) was able to bind to total lipids isolated from mouse brain [182]. Those findings led to the hypothesis that lipid-PrP interaction is involved in the de novo PrP conversion process. To determine whether rPrP binds to lipids from other sources, I first investigated the interaction between rPrP and lipid membranes isolated from cultured murine neuroblastoma

N2A cells using the iodixanol density gradient assay (Figure 4.1). In this assay, rPrP was mixed with N2A membranes for 5 minutes and then the rPrP- membrane mixture was loaded to the bottom of the centrifuge tube with iodixanol concentration adjusted to 36%. On top of the sample, 31% and 5%of iodixanol solutions were added sequentially (Figure 4.1A). After centrifugation at

200,000 g for 3 hours at 4ºC, all fractions from top to bottom were collected and subject to SDS-PAGE analysis. In the rPrP only control sample, all rPrP remained at the bottom fractions demonstrating rPrP itself would not move to the top of the gradient (Figure 4.1B, top two panels). When rPrP-membrane mixture was applied to the gradient, majority of rPrP migrated to the top fractions indicating binding between rPrP and N2A lipid membranes (Figure 4.1B, second panel). To differentiate whether rPrP binds to lipids or other components in N2A membranes, I repeated the same gradient assay using total lipids extracted from

N2A cells, and rPrP still migrated to the top of the gradient (Figure 4.1B, bottom panel). The results of this analysis confirmed the interaction between rPrP and

Figure 4.1: Interactions between rPrP and lipid vesicles. (A) The iodixanol density gradient. (B) rPrP alone, mixed with lipid membranes isolated from N2A cells or total lipids extracted from N2A cell membranes. (C) rPrP was mixed with lipids extracted from N2A cells, with 1.5 M KCl or 1.5 M KCl and 10 mM NaOH extraction. (D) rPrP was first equilibrated with 1.5 M KCl and then mixed with

N2A lipids with KCl the concentration maintained at 1.5 M. PrP in the first two panels of B was detected by coomassie brilliant blue staining. PrP in the bottom panel of B and in C and D was detected by immunoblot analysis with the POM1 monoclonal antibody [183].

58 lipids. To determine the physical nature of rPrP-lipid interaction, I performed the

gradient assay with high salt and/or high pH extraction (Figure 4.1C). First, rPrP

was mixed with total lipids extracted from N2A cells at room temperature for 10 minutes and then 1.5 M KCl was applied to the rPrP-lipid mixture for another 10 minutes prior to the gradient analysis. The result (Figure 4.1C, top panel) showed that majority of the rPrP migrated to the top fractions after the high salt extraction, suggesting a hydrophobic interaction between rPrP and lipids. This notion was supported by experiment using harsher extraction treatment with high salt plus high pH. The rPrP-lipid mixture was extracted with a buffer containing

1.5 M KCl+10 mM NaOH and after centrifugation, the majority of rPrP resisted the extraction and migrated to the top fractions (Figure 4.1C, bottom panel).

Interestingly, when rPrP was first equilibrated with 1.5 M KCl and then

mixed with N2A total lipids, the binding was largely eliminated (Figure 4.1D). The

dramatic inhibitory effect observed in Figure 4.1D suggested that an initial

electrostatic contact is likely involved in the early binding step, which allows the

hydrophobic domain of rPrP to insert into lipid bilayers. This explanation would

be in line with previous spectroscopic result suggesting that the PrP-lipid

interaction involves a binding step followed by the insertion into lipid bilayer [151].

To analyze what types of lipids are present in N2A membranes, I performed two-

dimensional TLC assay with N2A total lipids. Phosphatidylcholine (PC) and

Figure 4.2: Thin layer chromatography (TLC) analyses of lipids extracted

from N2A cells. Two-dimensional TLC of lipids extracted from N2A cell

membranes with chloroform-methanol-concentrated (28%) NH4OH (65/35/5) in

the first dimension (vertical dimension) followed by drying for 10 minutes in

nitrogen and then development in second dimension (from right to left) with chloroform-acetone-methanol-acetic acid-water (100/40/20/20/10). O, origin; PS,

phosphatidylserine; PC, phosphatidylcholine; PE, phosphatidylethanolamine;

FFA, free fatty acid; NL, neutral lipids; X, unidentified compounds.

60 phosphatidylethanolamine (PE) (both zwitterionic), phosphatidylserine (PS)

(negatively charged), other neutral lipids (NL), free fatty acids (FFA) and some other unidentified compounds (X) were identified to comprise the lipid content of

N2A membranes (Figure 4.2).

Individual phospholipids carrying different charges (Table 4.1) were also tested for their abilities of binding rPrP (Figure 4.3A). Phospholipids with palmitoyl and oleyl chains were used here because phospholipids with monounsaturated fatty acyl chain have lower transition temperatures than those with saturated fatty acyl chains and are much more resistant to lipid peroxidation than those with polyunsaturated fatty acyl chains. Under the experimental condition, almost 100% of rPrP bound to anionic POPG or POPS. In contrast, there was no binding of rPrP to zwitterionic POPC or cationic DOTAP (Figure

4.3A). To confirm that rPrP migrated to the top fractions is indeed due to its binding to lipids, I repeated the gradient assay for rPrP-POPG mixture and applied each fraction to the phosphorus analysis, which quantitates the concentration of phosphorus in tested samples. Since POPG is a phospholipid and no other source of phosphorus is present in the rPrP-POPG mixture, the amount of phosphorus present in each fraction were determined. My result showed that the distribution pattern of phosphorus was perfectly matched with that of rPrP detected by coomassie brilliant blue staining in this case (Figure 4.3B),

POPG 1-palmitoryl-2-oleoylphosphatidylglycerol

POPS 1-palmitoryl-2-oleoylphosphatidylserine

POPC 1-palmitoryl-2-oleoylphosphatidylcholine

POPA 1-palmitoryl-2-oleoylphosphatidic acid

DOTAP 1,2-dioleoyl-3-trimethylammonium propane

Brain Sulfatide HSO4-3-Galβ1-1’ceramid

Table 4.1: Lipids used in binding assay

Figure 4.3: Interactions between rPrP and lipids with different charges. (A)

rPrP was mixed with lipid vesicles composed of 2.5 mg/mL POPG, POPS, POPC,

or DOTAP. PrP was detected by immunoblot analysis with the POM1 monoclonal antibody. (B) Phosphorus analysis of fractions collected in rPrP+POPG gradient assay. The gradient assay result (solid arrow) was incorporated into the chart for the comparison purpose. rPrP was detected by coomassie brilliant blue staining.

63 confirming that rPrP migration to the top is due to its binding to anionic lipids. The specificity of rPrP binding to anionic lipids supported the possibility that the PrP- lipid interaction is initiated by electrostatic interaction.

4.2.2 rPrP undergoes conformational changes upon binding lipids and gains the Proteinase K resistance

To investigate whether rPrP undergoes conformational change due to

interaction with lipid, far-UV circular dichroism (CD) spectra were collected and

analyzed (Figure 4.4) (data were obtained from our collaborative laboratory).

The spectrum of rPrP alone is consistent with a conformation enriched with α- helices (Figure 4.4A) [118]. After incubation with POPG at 37°C for 1 hour, the

CD spectrum of rPrP showed an increase in negative ellipticity with a minimum peak around 216 nm (Figure 4.4A), suggesting an increase in β-sheet content.

These results are consistent with previous reports [151, 171].

The effect of longer incubation with lipid vesicles was tested and CD data

showed that a 24- or 48-hour incubation with POPG at 37°C resulted in additional changes in spectra (Figure 4.4A), which may suggest further conformational alterations. As a control, rPrP was incubated with zwitterionic POPC, and no characteristic conformational change was detected even after a prolonged incubation (Figure 4.4B). To determine whether the increase in the β-sheet

64 content of rPrP is accompanied by the developing of Proteinase K (PK)

resistance, PK digestion was performed with PK:rPrP molar ratios of 1:16, 1:7.5,

and 1:3. Two PK-resistant bands around 15 and 14.5 kDa, together with low

amount of full-length rPrP, were detected after a 24-hour incubation with anionic

POPG (Figure 4.4C). In contrast, rPrP incubated with either buffer alone or

zwitterionic POPC remained PK-sensitive. Interestingly, the interaction with

POPG also dramatically stabilized rPrP at 37°C (Figure 4.4C, compare samples

without PK digestion).

To determine the effect of salt at physiological concentration on lipid-rPrP interaction, 150 mM NaCl was added to the reaction mixture. Further changes in the CD spectra were observed (Figure 4.5A). Surprisingly, after merely 1-hour incubation, the addition of 150mM NaCl significantly enhanced production of the

15 kDa PK-resistant band, eliminated the 14.5 kDa band, but hardly affected the amount of full-length PK-resistant rPrP (Figure 4.5B). In contrast, rPrP remained

PK-sensitive after incubation with either salt alone or salt with POPC (Figure

4.5B), indicating that the anionic POPG-rPrP interaction induced rPrP

conformational change is the major determinant for the PK-resistance. Since

rPrP in the undigested control sample was 10% of that used in PK digestion and

the amount of 15 kDa PK-resistant rPrP fragment is similar to the undigested

control (Figure 4.5B, -PK), I estimated that around 10% of lipid-interacting rPrP

Figure 4.4: Conformational change and PK-resistance of rPrP induced by

anionic lipid interaction. (A) Far-UV CD spectra for rPrP or rPrP incubated with

POPG for the indicated time. (B) Far-UV CD spectra for rPrP or rPrP incubated with POPC for the indicated time. (C) PK digestion of rPrP after incubation at

37°C for the indicated time with buffer, POPG, or POPC. The PK:rPrP molar ratios were 1:16, 1:7.5, and 1:3. The unit of θ is degrees cm-1 mol-1 L.

66 gained PK-resistance despite an almost 100% binding of rPrP to anionic POPG

(Figure 4.3A).

PrPSc has a highly PK-resistant C-terminal fragment, while its N-terminus

remain PK-sensitive [1]. The specific C-terminal PK resistance differentiates

between nonspecific PrP aggregation and the PrPSc conformation. To determine

whether the 15 kDa PK-resistant rPrP band was at the C-terminus, characteristic

of PrPSc PK-resistance, I probed it with a panel of antibodies. All the antibodies

recognizing C-terminal epitopes of PrP detected the 15 kDa band (Figure 4.6,

solid arrow), which was not recognized by the N- terminal epitope-specific 8B4

antibody [184, 185], indicating that the 15 kDa band is similar to the PK-resistant

core of PrPSc. However, an additional PK-resistant band around 13.5kDa was

detected by the 8B4 antibody, while not recognized by any of other antibodies

(Figure 4.6, empty arrow).

4.2.3 The lipid-induced C-terminal PK-resistance of rPrP formed under

different mechanism from N-terminal and full-length PK-resistance

In addition to the C-terminal and N-terminal PK-resistant rPrP fragments,

some full-length PK-resistant rPrP was also detected after PK digestion (see figures above), which raised the question whether it was resulted from the insufficient PK digestion. To rule out this possibility, lipid-interacting rPrP was digested

Figure 4.5: Effect of salt at physiological concentration on rPrP

conformation. (A) Far-UV CD spectra for rPrP incubated with POPG for the

indicated time with or without 150 mM NaCl. The unit of θ is degrees cm-1 mol-1 L.

(B) PK digestion of rPrP after incubation at 37°C for the indicated time with buffer,

POPG, or POPC in the presence of 150 mM NaCl. The PK:rPrP molar ratios were 1:16, 1:7.5, and 1:3.

Figure 4.6 Identity of PK-resistant rPrP. rPrP was incubated with C9G1

(POPC:POPG mass ratio is 9:1) vesicles in the presence of 150 mM NaCl and subjected to PK digestion with PK:rPrP molar ratios of 1:7.5. Antibodies recognizing the indicated epitopes were used to detect the PK-resistant PrP fragment.

69 with increased amounts of PK and the full-length PK-resistant rPrP, though with

very low amount in this case, endured digestion with 0.75 mg/mL PK (Figure

4.7A). While both C-terminal (Figure 4.7A, POM1 antibody) and N-terminal

(Figure 4.7A, 8B4 antibody) of the lipid-induced rPrP showed a remarkably high

PK-resistance with the PK:rPrP molar ratio as high as 150:1 (equal to 3.75

mg/mL PK).

The concurrent appearance of 3 types of PK-resistant rPrP species (C-

terminal, N-terminal and full-length) led to the speculation whether they formed

from the same mechanism. Using various lipid combinations, I detected the PK-

resistant rPrP fragments after incubation for 1 hour (Figure 4.7B, lanes 3-8) or

24 hours (Figure 4.7B, lanes 9-14) with both POM1 (C-terminal) and 8B4 (N-

terminal) antibodies (Figure 4.7B). My results showed that the amounts of both

N-terminal and full-length PK-resistant rPrP fragments varied with different lipid

compositions. POPG/POPC bound rPrP produced more full-length PK-resistant

rPrP (Figure 4.7B, left panel, lane 5) than rPrP bound to POPS/POPC (Figure

4.7B, left panel, lane 6), but much less than rPrP mixed with lipids extracted N2A cells or mouse brain tissues (Figure 4.7B, left panel, lane 7 and 8). On the other hand, rPrP bound to N2A lipids or mouse brain lipids hardly generates N-terminal

PK-resistant bands (Figure 4.7B, right panel, lane 5 and 6), while POPG/POPC or POPS/POPC bound rPrP produced much more N-terminal PK-resistant

70 fragments (Figure 4.7B, right panel, lane 7 and 8). After 24-hour incubation,

those discrepancies became even more distinctive (Figure 4.7B, 24 hour, left

and right panels). Under all these conditions, however, similar level of C-terminal

PK-resistance was associated with rPrP interacting with different lipids (Figure

4.7B, POM1 antibody), indicating that these PK-resistant fragments were not

necessary resulted from the same mechanism. The smear of the PK resistant

band after 24-hour incubation with mixed lipids (Figure 4.7B, left panel, lane 13

and 14) likely represented lipid oxidation during the incubation.

As described above, about 10% of lipid-interacting rPrP gained the C-

terminal PK-resistance. Therefore, it is likely that both N-terminal and full-length PK-

resistant rPrPs represent other potion of rPrP that interacts with lipid differently than

those that produce the C-terminal PK-resistant fragment. This interpretation is consistent with our observation that the amount of N-terminal and full-length PK- resistant rPrP varied under experimental conditions presented in this thesis. In those cases, however, the amount of C-terminal PK-resistant rPrP remained relatively constant. Because of the similarity of the C-terminal PK-resistant rPrP fragment to

Sc the PK-resistant core of PrPP in both size and pattern, following the generation of C-

terminal PK-resistant rPrP is a plausible approach to study the role of lipid-PrP

interaction and it’s role in the pathogenesis of prion disease. Thus, in the rest of my

thesis study, I focused my efforts on studying C-terminal PK-resistant rPrP.

Figure 4.7: Characterizing the PK-resistant PrP conformation. (A) rPrP was incubated with POPG vesicles for 1 h at 37°C in the presence of 150 mM NaCl and then subjected to PK digestion with PK:rPrP molar ratios of 1:16, 1:7.5, 1:3,

72 1:1.5, 3:1, 15:1, 30:1, 75:1, 150:1, and 300:1. PrP was detected by immunoblot

analysis with POM1 or 8B4 antibody as indicated. (B) Lane 1 contained

undigested PrP and lane 2 protein standards. Lanes 3-8 contained PrP incubated

with different lipids for 1 hour and subjected to PK digestion with a PK:PrP molar ratio of 1:16. The following lipids were used: POPC (lane 3), POPG (lane 4),

C9G1 (lane 5), POPS and POPC (lane 6), N2A lipids (lane 7), and mouse brain

lipids (lane 8). Lanes 9-14 were the same as lanes 3-8 but with 24-hour

incubation. PrP was detected by immunoblot analyses with the POM1 or 8B4

antibody as indicated.

73 4.3 Characterization of lipid-rPrP interaction

4.3.1 Both charge and lipid headgroup structure influence rPrP

conformational change

The requirement of the anionic POPG raised the question that whether the negative charges on lipid vesicles are important for rPrP conversion. Using zwitterionic

POPC mixing with decreasing amounts of POPG, I observed a reduced binding of rPrP to lipid vesicles (Figure 4.8A). When the POPC:POPG ratio was at 9:1

(designated as C9G1), the majority of rPrP remained lipid-bound and migrated to the top of the gradient. The appearance of rPrP in the second fraction (Figure

4.8A) is likely due to the aggregation of rPrP-bound lipid vesicles since visible aggregates were observed under this condition. When the POPC:POPG ratio reached 99:1 (C99G1), the majority of rPrP remained in the bottom fractions of the gradient (Figure 4.8A), indicating that they were not lipid bound. Similarly, the intensity of the 15 kDa C-terminal PK-resistant band decreased with reduced

POPG content (Figure 4.8B), suggesting that the rPrP conformational change was closely associated with its interaction with anionic POPG.

The morphology of lipid vesicles was monitored by freeze fracture electron

microscopy. As expected from lipid vesicles prepared by sonication, POPC and

C9G1 formed small liposomes (Figure 4.9). Fewer lipid vesicles were observed

Figure 4.8: Negative charges on lipid vesicles affect rPrP binding and conformational change. (A) rPrP was mixed with lipid vesicles composed of

POPC and POPG at the indicated ratios. The concentration of total lipids was 2.5 mg/mL, except for the top panel (1:1*) where it was 5 mg/mL. The presence of rPrP was detected by coomassie brilliant blue staining. (B) PK digestion of rPrP incubated at 37°C with indicated lipid vesicles in the presence of 150 mM NaCl for 1 hour.

Figure 4.9: Freeze fracture EM analysis of lipid vesicles prepared by sonication. The bars correspond to 100 nm.

with anionic POPG, presumably due to the repulsion by its negative charge resulting in micelle formation instead of liposome. These results suggested that the charge-dependent lipid binding and rPrP conformational change appeared to be independent of the shape of lipid vesicles. This conclusion was further supported by preparing large unilamellar vesicles (POPC:POPG 7:3, 600nM in diameter) by extrusion, which induced the similar 15 kDa C-terminal PK-resistant rPrP band (Figure 4.10).

Figure 4.10: PK digestion of rPrP incubated with 600 nm diameter liposomes composed of POPC and POPG at a 7:3 molar ratio. The incubation was carried out at 37°C for 24 hours in the presence of 150 mM NaCl.

The PK:rPrP molar ratios were 1:16, 1:7.5, and 1:3 for both panels B and C, and

PrP was detected by immunoblot analysis with the POM1 antibody

77 Next, I investigated whether rPrP conformational change is solely

governed by negative charges or by both negative charge and lipid structure. I

compared POPS and POPG, two anionic phospholipids. Under the same

condition, much less PK-resistant rPrP was detected after incubation with anionic

POPS (Figure 4.11). When the same amount of POPS was simply mixed with

zwitterionic POPG at a 1:1 ratio, the conversion of rPrP to the C-terminal PK-

resistant conformation was at least as efficient as with POPG (Figure 4.11). It is

important to point out that, at the same condition, the binding of rPrP to either

POPS or POPG was almost identical (Figure 4.3A). In addition, the fatty acyl

chains were exactly the same between POPS and POPG. Therefore, these

results indicated that the lipid headgroup structure affects the resulting rPrP

conformation.

This interpretation was further supported by using anionic POPA and

brain-enriched glycolipid sulfatide (Table 4.1). Although the binding of rPrP to both anionic lipids is similar to it’s binding to POPG (Figure 4.12A, and

compared to Figure 4.3A), no PK-resistant rPrP was detected after incubation

with either POPA or sulfatide alone (Figure 4.12B). When mixed with zwitterionic

POPC at a 1:1 ratio, a strong PK-resistant PrP band was observed with sulfatide,

POPG and POPS, but very little with POPA/POPC mixture (Figure 4.12B).

Collectively, these results indicated that simply binding to anionic lipid does not

Figure 4.11: Lipid headgroup structures affect rPrP conformation. rPrP was incubated with POPS (2.5 mg/mL), POPS and POPC (5 mg/mL), or POPG (2.5 mg/mL) and subjected to PK digestion with PK:rPrP molar ratios of 1:16, 1:7.5, and 1:3.

Figure 4.12: Different lipid compositions affect rPrP conformation. (A)

Iodixanol density gradient assay for rPrP mixed with anionic POPA and sulfatide.

(B) rPrP was incubated with various lipids as indicated with the total lipid concentration kept at 2.5 mg/mL before mixing with rPrP, and subjected to PK digestion with a PK:rPrP molar ratio of 1:16. PrP was detected by immunoblot analysis with the POM1 antibody.

80 result in producing the specific C-terminal PK-resistant rPrP. The conversion of

rPrP to the C-terminal PK-resistant conformation depends on not only the

negative charges of the lipid vesicles but also the structure and/or arrangement

of lipid headgroups on the surface of lipid vesicles.

4.3.2 The hydrophobic domain of PrP is essential for generation of C-

terminal PK-resistant rPrP

The binding between rPrP and lipid initiates with electrostatic interaction,

which allows rPrP to interact with the hydrophobic core of lipids (Figure 4.3). Yet, it remains unclear whether the hydrophobic interaction is essential for generating

C-terminal PK-resistance of rPrP. To address this question, I performed PK

digestion of POPG interacted rPrP first and then applied it to the iodixanol

density gradient. The result showed that the C-terminal PK-resistant fragment migrated to the top fractions, indicating binding between the 15 kDa rPrP fragment and POPG. When PK-digested POPG-rPrP was first extracted by a

high salt and high pH solution and then applied to the iodixanol gradient, the C-

terminal PK-resistant fragment migrated to the top fractions as well (Figure 4.13,

bottom panel). These results indicate that the C-terminal PK-resistant part of rPrP is indeed lipid bound and interacts with lipid hydrophobically.

Next, I generated an rPrP mutant in which the hydrophobic domain

Figure 4.13: The C-terminal PK-resistant rPrP fragment is associated with

lipid hydrophobically. Mouse rPrP was incubated with POPG for 1 hour at 37oC,

PK digested, and separated by the iodixanol density gradient with or without prior extraction with an alkaline solution of 1.5 M KCl and 10 mM NaOH. Fractions were collected from top to the bottom of the gradient as indicated and rPrP was detected with POM1 antibody.

82 (Figure 1.2) was deleted (designated as rPrPΔ111-131). Using iodixanol gradient

assay, I found that rPrPΔ111-131 alone remained at the bottom of the gradient,

suggesting that the deletion of the hydrophobic domain did not drastically alter the density of rPrP (Figure 4.14A, top panel). When rPrPΔ111-131 was mixed with

total lipids extracted from N2A cells, it bound to the lipids and migrated to the top

fractions (Figure 4.14A, 2nd panel). In contrast to wild-type rPrP-lipid interaction

(Figure 4.1C), the rPrPΔ111-131-lipid binding was almost completely disrupted by

the extraction with a high salt and high pH solution (Figure 4.14A, bottom panel), suggesting a lack of hydrophobic interaction. More importantly, when rPrPΔ111-131 was mixed with a panel of anionic lipids, no C-terminal PK-resistant PrP fragment was detected (Figure 4.14B). To rule out the possibility that this result is due to the loss of antibody recognizing epitope, I probed the blot with 8H4 antibody that recognizes a further C-terminal epitope (amino acid 175-195) and the result showed that the lack of PK-resistant fragment is not due to the loss of epitope

(Supplemental Figure 4.1). Together, these results suggest that the hydrophobic interaction between rPrP and lipids is essential for the generation of

C-terminal PK-resistant rPrP. Furthermore, the lack of specific C-terminal PK- resistant fragment of rPrPΔ111-131 excludes the possibility that the C-terminal PK- resistance of lipid bound rPrP is a result of non-specific lipid vesicle protection.

Figure 4.14: The hydrophobic domain of PrP is essential for generation of

C-terminal PK-resistance. (A) Iodixanol density gradient separation of mouse

alone, rPrPΔ111-131 + total lipids isolated from N2A neuroblastoma cells (N2A

lipids), rPrPΔ111-131 + N2A lipids extracted with 1.5M KCl or 1.5M KCl plus 10mM

NaOH solution as indicated. (B) Mouse rPrP or rPrPΔ111-131 were incubated with

various lipids as indicated for 1 hour and subjected to PK digestion. Solid

arrowhead points to the 15 kDa PK-resistant band generated from wild-type rPrP.

Empty arrow points at the expected position of PK-resistant band of rPrPΔ111-131.

PrP was detected by immunoblot analysis with POM1 antibody.

84 4.3.3 Adjacent localization of anionic charges and the hydrophobic lipid core is necessary for inducing C-terminal PK-resistant rPrP

The requirement of the hydrophobic rPrP-lipid interaction raised the question that what kind of relationship between anionic charge and hydrophobic lipid core is required for generating C-terminal PK-resistance. To address this question, I used various combinations of anionic RNA, zwitterionic PAPC, and cationic DOTAP. RNA contains anionic charges but without hydrophobic region.

Zwitterionic PAPC contains hydrophobic fatty acyl chains but net charge is zero.

DOTAP contains hydrophobic fatty acyl chains allowing it to be incorporated into lipid vesicles. Because of its cationic charge, DOTAP does not bind rPrP (Figure

4.3A), but it will bind with anionic RNA. To ensure that anionic charges on RNA were not completely neutralized, the amount of DOTAP was only 5% (w/w) of total lipids used in these analyses. Lipid vesicles consisted of single or two lipids were prepared first, followed by adding anionic RNA, and rPrP was added at last.

In samples without anionic RNA (Figure 4.15, left panel) and in sample with RNA but without lipid (Figure 4.15, right panel, lane 1), no 15 kDa PK- resistant band was detected, indicating that both anionic charge and hydrophobic lipid are necessary for the formation of C-terminal PK-resistant rPrP. In samples incubated with both anionic RNA and lipids (Figure 4.15, right panel, lanes 2-4), the presence of zwitterionic PAPC or 5% DOTAP alone did not lead to the

Figure 4.15: Adjacent location of anionic charge and the hydrophobic lipid core are necessary for generation of C-terminal PK-resistance. Lipid vesicles consisted of zwitterionic PAPC (2.5 mg/ml), cationic DOTAP (0.125 mg/ml), or PAPC (2.375 mg/ml) plus DOTAP (0.125 mg/ml) were prepared respectively. In samples where RNA was added, 50 μg/ml of RNA was added to the lipid vesicles. Mouse rPrP was added at last and the incubation was carried out at 37oC for 1 hour followed by PK digestion. PrP was detected by immunoblot analysis with POM1 antibody.

86 formation of C-terminal PK-resistant rPrP (Figure 4.15, right panel, lanes 2 and 3).

Only when rPrP was incubated both zwitterionic PAPC and cationic DOTAP, the

15 kDa C-terminal PK-resistant band was detected (Figure 4.15, right panel,

lane 4), which suggests that the cationic DOTAP brings the anionic RNA to the vicinity of PAPC vesicles and then an rPrP conversion is allowed. Therefore, an adjacent localization of anionic charge and the hydrophobic lipid core appears to

be necessary for the formation of C-terminal PK-resistant PrP.

4.3.4 Aggregation of PrP stabilizes the lipid-induced PK-resistant PrP

conformation

Disrupting lipid vesicles with 0.5% Triton X-100 completely abolished the

PK resistance of rPrP (Figure 4.16A), which suggested that the binding of rPrP to lipid vesicles might be required for the stability of the altered rPrP conformation.

Since authentic PrPSc is highly aggregated, the aggregation of lipid-interacting rPrP may stabilize the lipid-induced C-terminal PK-resistant conformation when lipid vesicles are disrupted.

To increase the likelihood that rPrP would aggregate, the lipid rPrP

mixtures were agitated at 37°C for 16 hours in the presence of Triton X-100. The

aggregation status of rPrP was monitored by a sucrose gradient analysis [186].

In this gradient, the majority of the monomeric rPrP remained at the top, even

Figure 4.16: Aggregation stabilizing the C-terminal PK-resistant rPrP

conformation. (A) rPrP was mixed with POPC and POPG in the presence of

150 mM NaCl and incubated at 37°C for 1 h. PK digestion was carried out with a

PK:rPrP molar ratio of 1:16 in the presence or absence of 0.5% Triton X-100. PrP

was detected by immunoblot analysis with the polyclonal M20 antibody. (B) rPrP

with indicated treatments was loaded on top of the sucrose gradient, and the

presence of rPrP was detected with immunoblot analysis with the POM1 antibody.

88 For panels with Triton X-100 treatment, rPrP and lipid mixtures were incubated at

37°C for 1 h. Then, 0.5% Triton X-100 (final concentration) was added and agitated at 37°C for 16 h. (C) rPrP was incubated with buffer (lane 1), POPC

(lane 2), POPG (lane 3), POPC and POPG at a 1:1 ratio (lane 4), and C9G1

(lane 5) for 1 h at 37°C in the presence of 150 mM NaCl. After incubation, 0.5%

Triton X-100 (final concentration) was added and samples were agitated for 16 h at 37°C at 800 rpm. PK digestion was performed with a PK:rPrP ratio of 1:16.

PrP was detected by immunoblot analyses with POM1 and 8B4 antibodies.

89 after a 16-hour agitation in the presence of Triton X-100 (Figure 4.16B, top and second panels). After binding to POPG vesicles, rPrP spread at the middle and bottom fractions of the gradient, likely due to the migration of POPG vesicles in the gradient (Figure 4.16B, third panel). When Triton X-100 was included, the majority of rPrP appeared at the bottom of the gradient (Figure 4.16B, bottom panel), indicating the disruption of POPG vesicles and aggregation of rPrP. The reappearance of a small amount of rPrP in the top fractions supported a thorough detergent extraction.

To determine whether C-terminal PK-resistance remained in aggregated rPrP molecules, rPrP was incubated with various lipid combinations followed by agitation in the presence of Trition X-100 and then subjected to PK digestion.

The PK:rPrP molar ratio was 1:16, which was much higher than the regular ratio of 1:50. In control samples incubated in buffer alone or with POPC (Figure 4.16C, lanes 1 and 2), rPrP remained PK-sensitive. However, in samples incubated with lipid combinations that were able to induce the C-terminal PK-resistant rPrP, the

15 kDa fragment was detected (Figure 4.16C, POM1, lanes 3-5). On the contrary, the N- terminal PK-resistant fragment was not detected under these conditions (Figure 4.16B, 8B4). Together, these results indicated that the aggregation of rPrP molecules is able to stabilize the anionic lipid-induced PK- resistant conformation.

90 An alternative approach is to decrease the lipid:rPrP ratio, which will

improve the opportunity for rPrP molecules to interact with each other on the

surface of lipid vesicles and, thereby, stabilize the anionic lipid-induced PK-

resistant conformation. To test this possibility, I decreased the amount of POPG

to 10% of the amount used in above assays (designated as POPG10%). As

controls, rPrP/POPC and rPrP/POPG mixtures were subjected to the same treatment.

All rPrP/lipid mixtures were incubated at 37°C for 6 days. After the

incubation, part of the mixture was agitated for additional 16 hours at 37°C in the

presence or absence of Triton X-100 (Figure 4.17A). Without Triton, a weaker 15

kDa PK-resistant rPrP band was detected with POPG10% (Figure 4.17A, -Triton

X-100, POM1; compare lanes 2 and 3), which is consistent with the notion that the rPrP conformational change is dependent on the amount of anionic lipids.

Interestingly, the N-terminal PK-resistant band was not detected in rPrP incubated with POPG10% (Figure 4.17A, -Triton X-100, 8B4), suggesting that its appearance requires a high POPG:rPrP ratio. In the presence of Triton (Figure

4.17A, +Triton X-100), only the 15 kDa C-terminal PK-resistant fragment was detected, supporting the possibility that the aggregation of rPrP molecules stabilize the C-terminal PK-resistant conformation.

Another part of the mixture was incubated at 37°C for an additional 16

hour without agitation. Without Triton, the result was similar to that described

Figure 4.17: C-terminal PK-resistant rPrP conformation that formed with a lower lipid:rPrP ratio. Recombinant PrP was incubated with POPC (lane 1),

POPG (lane 2), or POPG10% (lane 3) for 6 days at 37°C in the presence of 150 mM NaCl. (A) After incubation, samples were agitated at 800 rpm for 16 h in the presence or absence of 0.5% Triton X-100. PK digestion was carried out at 37°C for 30 min with a PK:rPrP molar ratio of 1:16. PrP was detected with the POM1 or

8B4 antibody as indicated. (B) Same as panel A except samples were incubated without agitation.

92 above, more C-terminal PK-resistant rPrP with POPG and no N-terminal resistant

band with POPG10% (Figure 4.17B, -Triton X-100). However, in samples

incubated with Triton, more 15 kDa PK-resistant rPrP was detected in rPrP

incubated with POPG10% (Figure 4.17B, +Triton X-100, POM1; compare lanes 2 and 3).

This result indicated that a lower lipid:rPrP ratio increased the level of rPrP

aggregation on the surface of lipid vesicles, which stabilized the C-terminal PK-

resistant conformation even when lipid vesicles were disrupted by Trition X-100.

In all these assays, the rPrP incubated with zwitterionic POPC remained PK-

sensitive (lane 1 in all panels of Figure 4.17A and B), supporting the requirement of

anionic lipid interaction for rPrP conversion. Together, these results suggested

that the anionic lipid-induced C-terminal PK-resistant rPrP conformation, but not

the N-terminal PK-resistant rPrP form, could be stabilized by the aggregation of

rPrP molecules.

4.3.5 Lipid-induced PK resistance of rPrP does not correlate with fiber formation

Electronic microscopic (EM) analysis was performed to monitor the

morphology of rPrP incubated under various conditions (Figure 4.18). In the

presence of 150mM NaCl, rPrP incubated with POPG formed a “wormlike”

structure with a diameter of ~40-50nm (Figure 4.18A), which was not observed

Figure 4.18: Morphology of rPrP aggregates. (A) Recombinant PrP was

incubated with POPG at 37°C for 24 h in the presence of 150 mM NaCl. (B)

Same as panel A, just replace POPG with POPC. (C) Same as panel A, 1-month incubation. (D) Recombinant PrP incubated with POPG10% (the POPG concentration before mixing with rPrP was 0.25 mg/mL) at 37°C for 1 month in the presence of 150 mM NaCl. (E) Recombinant PrP was incubated with C9G1 at 37°C for 17 days in the presence of 150 mM NaCl. (F) The PrP/C9G1 mixture from panel C was agitated for 16 h in the presence of 0.5% Triton X-100. Arrows point to smaller aggregates. The bar corresponds to 50 nm.

94 in control sample containing rPrP with POPC (Figure 4.18B). However, these wormlike structures did not grow into amyloid fibers even after a very long period

(1 month) of incubation (Figure 4.18C). Rarely, amyloid fiber-like structures with a width of ~10nm could be found with samples incubated with POPG10% (Figure 4.18D).

When rPrP was incubated with C9G1 vesicles, large network-like structures were observed (Figure 4.18E). After agitation in the presence of Triton X-100, the large network-like structures disappeared and only amorphous aggregates were observed (Figure 4.18F). Since C-terminal PK-resistant rPrP were detected under all these conditions, these results suggested that amyloid fiber formation was not required for the anionic lipid-induced C-terminal PK-resistant rPrP conformation.

4.4 The relevance of rPrP-lipid interaction to prion biology

4.4.1 Arachidonic acid induces rPrP conversion

rPrP was showed above to preferentially bind to anionic phospholipids, which led a significant portion of rPrP to gain the characteristic C-terminal PK resistance. It still remains unclear whether this effect is specific for phospholipids, or fatty acid that also has a hydrophobic core and an anionic charge can induce

PrP conversion as well. Since arachidonic acid (AA) isomers were co-purified

95 with brain PrPC [5], I investigated the interaction between rPrP and AA and found

that AA alone was able to convert some rPrP molecules to PK-resistant

conformation, with a 15 kDa C-terminal PK-resistant fragment recognized by

POM1 antibody (Figure 4.19, top panel). Using the N-terminal specific 8B4

antibody, I found that the 15 kDa fragment was not recognized by the 8B4

antibody (Figure 4.19, bottom panel), confirming it is the C-terminal fragment.

Similar to several phospholipids used above, when AA was allowed to formed

lipid vesicles with zwitterionic phospholipid POPC, the efficiency of rPrP

conversion significantly increased (Figure 4.19, top panel). Since POPC does

not bind rPrP (Figure 4.3) or influence rPrP conformation (Figures 4.4 and 4.5),

this result is in accordance with the above conclusion that the arrangement of

anionic charged groups on the surface of lipid vesicles are important in

determining the resulting rPrP conformation.

Interestingly, the 13.5 kDa N-terminal PK-resistant fragment recognized

by 8B4 antibody, which was present in all phospholipids-induced rPrP conversion

(Figure 4.7), was not detected when rPrP incubated with either AA alone or AA plus POPC (Figure 4.19, bottom panel). Only when anionic phospholipid POPG was added, the N-terminal PK-resistant fragment was detected. Since POPG binds to rPrP and generates both the 15 kDa C-terminal and 13.5kDa N-terminal fragments (Figure 4.7), the detection of 13.5kDa N-terminal PK-resistant fragment

Figure 4.19: Arachidonic acid (AA) induced rPrP conversion Mouse rPrP

was incubated with AA with rPrP:AA molar ratios of 1:15, 1:30 and 1:150, respectively. The incubation was carried out at 37oC for 1 hour with or without

POPC or POPG as indicated, and PK digestion was performed afterwards. After

POM1 antibody detection (top panel), the blot was washed at room temperature

and immunoblot analysis with 8B4 antibody was performed on next day (lower

panel).

97 likely resulted from rPrP interaction with POPG instead of AA. This result

supports that the C-terminal PK-resistant rPrP is a result from rPrP

conformational change instead of forming transmembrane PrP in lipid vesicles.

4.4.2 Effects of lipid oxidation, copper, iron, and RNA on formation of C- terminal PK-resistant rPrP

Compared to POPG that contains one monounsaturated and one

saturated fatty acyl chain, the four double bonds of AA make it much more

susceptible to oxidation. To determine the effects of lipid oxidation on rPrP

conversion, I extended the incubation time from 1 hour to 24 or 48 hours at 37°C

(Figure 4.20, lane 1 and 2). When rPrP was incubated with AA alone or AA plus

POPC for 24 hours, the PK-resistant rPrP bands appeared to be a smear (Figure

4.20, 24 hr, lane 1 and 2). Since the smear migrated slower than 15 kDa, it is

likely due to the heterogeneity of lipid oxidation products instead of oxidation

caused rPrP fragmentation. After, 48 hours of incubation, no PK-resistant rPrP

could be detected in the sample incubated with AA alone and PK-resistant bands

became more smeared in samples incubated with AA plus POPC (Figure 4.20,

48hr, land 1 and 2). These results suggest that oxidized lipids cannot maintain

PK-resistant conformation of rPrP and this conclusion is supported by the fact

that AA failed to induce PK-resistant rPrP after it first reacted with oxidation

Figure 4.20: Effects of lipid oxidation generation of C-terminal PK-

resistance. PK digestion of mouse rPrP incubated with AA (lane 1), AA+POPC

(lane 2), oxidized AA (lane 3) or oxidized AA+POPC (lane 4) with rPrP:AA molar

ratios of 1:250 at 37°C for indicated time. Oxidized AA was obtained after AA was incubated 10μM FeCl2 for 24 hours. PrP was detected by immunoblot analysis with the POM1 antibody.

99 catalyst FeCl2 for 24 hours (Figure 4.20, lane 3 and 4).

Metal ions such as copper and iron have been shown to significantly

influence the PrPSc stability and/or propagation. Copper inhibits in vitro PrPSc- templated conversion [187], while iron enhanced the generation and stability of

PrPSc in diseased brain homogenates and in prion infected cells [161]. Interestingly,

I found that the POPG induced rPrP conversion was inhibited by copper, but was significantly enhanced in the presence of either ferrous or ferric ions (Figure

4.21A). To rule out the possibility that the reduced C-terminal PK-resistance in the presence of copper is caused by reduced lipid-rPrP interaction due to rPrP binding to copper, I performed iodixanol gradient assay using the undigested samples in Figure 4.21A. And I found that in the presence of copper, ferrous or ferric ions, the binding of rPrP to POPG is almost identical to control samples without those ions (Figure 4.21B). The similar result was observed when extraction with 1.5 M KCl+10 mM NaOH was applied to those samples prior to the gradient assay (Figure 4.21C). Since both copper and iron are redox cycling agents, the opposite effects observed here clearly indicated that they are not simply the result of lipid oxidation. Instead, these results indicated that the influence of these metal ions on rPrP is a more likely cause for the observed effects, which is in accordance with their effects on PrPSc in vivo [161, 188].

RNA has also been shown to significantly affect the propagation of PrPSc

100

Figure 4.21: Effects metal ions on generation of C-terminal PK-resistance.

(A) PK digestion of rPrP incubated with POPG at 37oC for 2 hours in the

101 presence or absence of copper or iron. The sequence of adding each component

is: lane 1, rPrP + POPG; lane 2, rPrP + POPG + 10mM CuCl2; lane 3, rPrP +

10mM CuCl2 + POPG; lane 4, rPrP + POPG + 10mM FeCl2; lane 5, rPrP + 10mM

FeCl2 + POPG; lane 6, rPrP + POPG + 10mM FeCl3; lane 7, rPrP + 10mM FeCl3

+ POPG. (B) Iodixanol density gradient assay for undigested samples in lanes 1,

3, 5 and 7. (C) Iodixanol density gradient assay for undigested samples in lanes

1, 3, 5 and 7 extracted with 1.5 M KCl+10 mM NaOH. PrP was detected by

immunoblot analysis with the POM1 antibody.

102

Figure 4.22: Effects of RNA on generation of C-terminal PK-resistance. The

15 kDa PK-resistant band of rPrP incubated with POPG in the presence of RNA for indicated times. The film was under exposed to ensure that the exposure was within linear range to reveal the differences. PrP was detected by immunoblot analysis with the POM1 antibody.

103 [5, 141]. To determine the effect of RNA on rPrP-lipid interaction, I carried out the incubation of rPrP and POPG in the presence and absence of RNA. In contrast to copper and iron, presence of RNA in the reaction did not significantly affect the formation of PK-resistant rPrP (Figure 4.22, 1 hour). Yet, after 24, 48, and 72 hours of incubation at 37°C, the 15 kDa PK-resistant rPrP band remained much stronger in the presence of RNA (Figure 4.22), suggesting that RNA enhances the stability of C-terminal PK-resistant rPrP conformation.

4.4.3 Middle region localized P105L mutation and P129 polymorphism affect rPrP-lipid interaction

The requirement of anionic charges in the vicinity of hydrophobic lipid core indicates that rPrP-lipid interaction likely involves the highly conserved middle region of PrP, which contains a positively charged region with four lysine residues and a hydrophobic domain. Interestingly, the first three lysine residues in the charged region were separated by two proline residues and replacing either proline with leucine is a cause of inherited prion disease, GSS [189]. Since proline is conformationally restrained, the replacement of proline with lecucine would alter the presentation of positively charged lysine residues, which could subsequently affect the interaction between rPrP and anionic lipids.

This hypothesis was tested by comparing three recombinant human PrPs

104

Figure 4.23: Illustrations of wild type recombinant human PrP, mutants

ΔKKPRK and P105L, and polymorphism at codon 129.

105 (rhPrP), which are wild-type rhPrP, an rhPrP mutant without the very N-terminal

cluster of positively charged amino acids (amino acids 23-27, designated as

ΔKKRPK), and another rhPrP mutant with 105 proline replaced by a leucine

(P105L) (Figure 2.23). All three rhPrPs bound to anionic lipids at physiological

salt concentration (Figure 4.24A). To compare their affinities for anionic lipids, I

performed a salt competition analysis. First, I established salt concentrations that

competes the binding of wild-type rhPrP to anionic POPG. As shown in Figure

4.24B, majority of the rhPrP bound to POPG in the presence of 500mM KCl,

while a portion of rhPrP failed to bind when KCl concentration increased (Figure

4.24B).

As expected, a significant portion of ΔKKRPK failed to bind POPG in the

presence of 500mM KCl (Figure 4.24C, top panel), likely resulting from its

reduced positive charges. Interestingly, the P105L mutant, which retains all the

charged amino acids, had a reduced POPG binding ability as well (Figure 4.24C,

bottom panel). Similar results were obtained when the same experiment was

repeated with total lipids isolated from mouse brain. The binding of rhPrP to

mouse brain lipids was lower and a portion of rhPrP appeared in the bottom

fractions in the presence of 500mM KCl (Figure 4.24D, top panel). Under the

same condition, both ΔKKRPK and P105L failed to bind mouse brain lipids

completely (Figure 4.24D). Together, these results suggest that both ΔKKRPK

106 Figure 4.24: Middle region of PrP affects the formation of C-terminal PK- resistance. (A) Iodixanol density gradient assay for human wild type or mutant rhPrP (ΔKKRPK and P105L) mixed with anionic POPG in the presence of

150mM KCl. (B) Salt competition assay for rhPrP. The rhPrP and POPG were pre-equilibrated with KCl at indicated concentrations and then mixed together.

After 10-minute incubation at room temperature, rhPrP-lipid mixtures were subjected to iodixanol density gradient separation. (C) Salt competition assay for

ΔKKRPK and P105L. Wild type or mutant rhPrPs were allowed to bind to POPG in the presence of 500 mM KCl, and then subjected to the iodixanol density gradient separation. (D) Same as B except total lipids isolated from mouse brain

(MBL) were used. (E) Human wild type or mutant rhPrP (ΔKKRPK and P105L) were incubated with POPG for 1 hour and then subjected to PK digestion. PrP was detected by immunoblot analysis with the POM1 antibody.

107

108 and P105L have a reduced anionic lipid binding capability.

Surprisingly, ΔKKRPK and P105L mutations have quite different effects on generating C-terminal PK-resistance. Incubated with POPG for 1 hour at

physiological salt concentration (150mM NaCl), ΔKKRPK mutant generated

similar amount of 15 kDa PK-resistant fragment as that of wild-type (Figure

4.24E, +PK, the equal amount of rPrP input was verified by immunoblot analysis

of samples without PK digestion, Figure 4.24E, -PK). However, much less 15 kDa

PK-resistant fragment was detected with P105L mutant under the same condition

(Figure 4.24E, +PK). Together with lipid binding data, these results suggest that

both positively charged regions at N-terminal and middle of PrP contribute to lipid

binding, but the interaction between the middle positively charged region and

anionic lipids contribute most to generating C-terminal PK-resistance. This

conclusion is in line with the finding that the formation of C-terminal PK-resistant

rPrP requires an adjacent localization of anionic charges and hydrophobic lipid

core.

The requirement of middle region in rPrP-lipid interaction was also tested

using rhPrP with methionine (M) or valine (V) at 129th residue (designated as

129M and 129V respectively) (Figure 4.23). The 129 methionine and valine polymorphism, which is in the middle of hydrophobic domain, has a significant influence on the pathogenesis of prion disease [190]. I found that the binding of

109

Figure 4.25: The 129 polymorphism influences rPrP-lipid interaction. Top panels: Human rPrP 129M or 129V was allowed to bind to total lipids extracted from mouse brain tissues in the presence of 500mM KCl and separated by the iodixanol density gradient. Bottom panels: The human rPrP 129M + MBL or 129V

+ MBL complex was extracted with a 0.5M NaCHO3, pH11 buffer and separated by iodixanol density gradient. PrP was detected by immunoblot analysis with the

POM1 antibody

110 129M or 129V to mixed lipids isolated from mouse brain was similar (Figure 4.25, top panels). However, when rhPrP-lipid complexes were extracted by 0.5M

NaHCO3 (pH 11.0), part of 129V was extracted from the complex (Figure 4.25,

bottom panels) while majority of 129M remained lipid bound. These results

suggest that the hydrophobic domain located 129 polymorphism does not affect

the binding of rPrP to anionic lipids, but alters the hydrophobic strength of rPrP-

lipid interaction.

111

Supplemental Figure 4.1: The same blot in Figure 4.14B was reprobed with

8H4 antibody.

112

Supplemental Figure 4.2: The same blot in Figure 4.15 was reprobed with 8B4 antibody.

113 CHAPTER 5

DISCUSSIONS AND CONCLUSIONS

Summary of thesis

PrP-lipid interaction and the subsequent PrP conformational changes have been implicated in the pathogenesis of prion disease [148, 150, 151, 170,

171]. This study provides novel insights into the interaction between PrP and lipids. Here I showed that rPrP interacts with anionic lipids and the interaction increases the β-sheet content of rPrP. In addition, this study also revealed that, under physiological conditions, the binding of rPrP to certain anionic lipid vesicles is sufficient to convert a significant portion of α–helix-rich rPrP to a characteristic

C-terminal PK-resistant conformation, reminiscent of that of PrPSc. The amount of

negative charges and the lipid headgroup structures of lipid vesicles significantly

affect the resulting rPrP conformation.

For the first time, the adjacent location of anionic charges and the

hydrophobic lipid core is showed to be required for lipid induced rPrP conversion.

The interaction between lipid and the middle region of PrP is important for

generating the PrPSc-like C-terminal PK-resistant fragment. When lipid vesicles

are disrupted by detergent, PrP aggregation is necessary to maintain the anionic

114 lipid-induced C-terminal PK-resistant conformation. It is also shown here that

PrP-lipid interaction is influenced by lipid oxidation, metal ion, and RNA in the

same manner as they affect PrPSc.

Lastly, a role of PrP-lipid interaction in prion biology is strengthened by the

findings that GSS-associated P105L mutation and the 129 polymorphic residues alter PrP-lipid interaction. These findings support both the physiological relevance of previously reported lipid-induced PrP conformational change [148,

150, 151, 170, 171] and the proposition that lipid membranes may provide a

support for the PrPSc conformation [153, 164].

Relevance of studying rPrP-lipid interaction to the pathogenesis in prion

disease

The transgenic mouse experiment revealed that PrP without its GPI

anchor (PrPΔGPI), which is similar to rPrP, without GPI anchor and mostly

unglycosylated, still supports the conformation change and the propagation of

prion infectivity [176]. The generation of prion infectivity in these mice does not

exclude the involvement of lipid membrane. Secreted anchorless PrP may

interact with lipid membranes it normally would not encounter, which may be

responsible for the massive accumulation of PK-resistant PrP around vascular

endothelial cells [176]. Authors of that study postulated a possible facilitating

115 effect by lipid membrane-associated proteoglycans [176]. It is also possible that

the lipid membranes themselves play a role in this process.

The role of lipid in PrP conversion seems paradoxical to the observation

that PrP was efficiently converted in the presence of detergent in a modified

version of protein misfolding cyclic amplification (PMCA) [191] or in PMCA [137].

However, detergent extraction may not completely remove lipid molecules from

aggregated protein, which is supported by the presence of lipid molecules in

highly purified prion rod [92] and the co-purification of lipid with PrPC in the minimal components used for generation of native prion in vitro [5]. In addition, these in vitro PrP conversion assays are performed with brain homogenates and the presence of other polyanions like RNA and proteoglycan may substitute for the role of anionic lipid membranes [140, 141]. Furthermore, detergent itself is a bipolar molecule, which may replace the role of lipid to certain extent. It has been shown that anionic detergent sarkosyl induces PrP aggregation and fiber formation, but without proper PK resistance [192]. Low concentrations of SDS maintain a β-sheet-rich, transiently soluble state of PrP(27-30) [147]. Thus, the presence of detergent in the in vitro PrP conversion assays does not necessary rule out the involvement of lipid membrane in PrP conversion in a biological system. On the contrary, the electrostatic interaction-initiated PrP-lipid contact is consistent with the observations that charged molecules like polyamines or

116 sulfated molecules inhibit PrPSc production in cultured cells [154]. In addition, the

cell surface-localized anionic proteoglycans, the raft-localized gangliosides with

bulky carbohydrate structures and terminal sialic acids, and the N-linked

oligosaccharides on PrP would prevent PrP-lipid interaction, which would be

consistent with the rarity of spontaneous PrP conversion.

Different interactions between rPrP and lipids result in different PK-

resistant rPrP species

An intriguing observation in this study is the concurrent appearance of full-

length, N-terminal and C-terminal PK-resistant rPrP species. The amounts of these three PK-resistant species vary from each other under the same experimental conditions. And for the same species, the amount varies as well depending on the compositions of lipid vesicles used in different experimental setups. These observations strongly suggest that different PK-resistant rPrP species were resulted from different mechanisms. This conclusion is supported

by several observations in this study. Under the most complex experimental

condition in this study (Figure 4.15), when lipid vesicles composed of zwitterionic

PAPC, cationic DOTAP, and anionic RNA were used, more full-length PK-

resistant rPrP were generated than C-terminal PK-resistant rPrP, while no N-

terminal PK-resistant rPrP has been detected (Supplemental Figure 4.2). On

117 the contrary, when lipid vesicles composed of POPG alone or POPS plus POPC

were used, little full-length PK-resistant rPrP but abundant C-terminal and N-

terminal PK-resistant rPrP were detected (Figure 4.7). And, when Arachidonic

acid (AA) was used to induce rPrP conversion, no N-terminal PK-resistant

fragment was detected, even after rPrP was incubated with AA plus POPC

(Figure 4.19). rPrP variants also support that the three PK-resistant species are

derived from different mechanisms. In contrast to wild type recombinant mouse

PrP, mutant rPrPΔ111-131 produced more full-length PK-resistance while no C-

terminal PK-resistance was detected (Figure 4.14), indicating that the C-terminal

PK-resistance of wild-type rPrP resulted from specific hydrophobic interaction with involving the conserved middle region of PrP, and the full-length PK-

resistance is formed by other mechanism. When mutant ΔKKRPK was compared

to wild type recombinant human PrP, the C-terminal PK-resistance was similar

while no full-length PK-resistance was observed with ΔKKRPK (Figure 4.24).

Together, my results suggest that the three PK-resistant rPrP species are

derived from different mechanisms.

The C-terminal PK-resistant rPrP conformation results from lipid-induced

rPrP conformational change

These findings also strongly argue that the C-terminal PK-resistance

118 detected in this study is indeed a result from rPrP conformational change instead

of lipid vesicle protection. Had it been non-specific lipid vesicle protection, the

lipid vesicle bound rPrPΔ111-131 would have produced corresponding C-terminal

PK-resistant band. Furthermore, different levels of C-terminal PK-resistance,

associated with lipid-interacting ΔKKRPK and P105L (Figure 4.24), suggest that,

instead of non-specific lipid vesicle protection, the specific interaction between lipid and the middle region of rPrP is crucial for generating C-terminal PK-

resistance.

The appearance of both N- and C-terminal PK-resistant fragments after

lipid binding is suspiciously similar to that of the transmembrane from of PrP

[193]. However, it is unlikely that the transmembrane PrP was formed in this

study since neither signal sequences nor translocon was present. Generally, a

protein cannot by itself penetrated across a lipid membrane. Although some

protein toxins, such as diphtheria and anthrax toxin, are known to form

proteinaceous channels and translocate part of the toxin across membrane [194],

PrP is not known to have any of these characteristics. In addition, C-terminal PK-

resistant rPrP was formed after incubation with either POPG or AA, both of which

tend to form micelles without an interior aqueous core and would prohibit a

transmembrane topology. In addition, only when rPrP was incubated with POPG,

N-terminal PK-resistance was generated, and when rPrP was incubated with AA,

119 no N-terminal PK-resistance was formed. These analyses suggest that the presence of two PK-resistant form of rPrP is not due to the formation of transmembrane PrP. Instead, a lipid-induced PrP conformational change is a more plausible explanation.

The authentic PrPSc has a highly PK-resistant C-terminal fragment [1] and the specific C-terminal PK resistance differentiates between nonspecific PrP aggregation and the PrPSc conformation. While N-terminal and full-length PK- resistance varied with different factors, the 15 kDa PK-resistant fragments observed under various conditions in this study were demonstrated to be at the

C-terminus of rPrP. So it is plausible to investigate the lipid-PrP interaction and it’s role in the pathogenesis of prion disease by monitoring the C-terminal PK- resistant rPrP.

Relevance of lipid-PrP interaction to in vivo prion biology

In addition to PrP conformational change, the PrP-lipid interaction may also play a role in PrP folding and/or its physiological function. In vitro, folding of recombinant PrP is tethered to the nickel column [149], presumably due to the removal of interference derived from unstructured N-terminal part of PrP [119,

120]. In vivo, lipid interaction may play a similar role by binding to the positively charged N-terminal region of PrP, which would facilitate the folding of highly

120 structured C-terminal part of PrP. In this case, some of the pathogenic mutations located at the middle and N-terminal unstructured regions of PrP, such as P105L, may disrupt PrP folding by interfering PrP-lipid interaction, and increased misfolded PrP ultimately leads to neurodegeneration. Given the fact that GSS- associated P105L mutation is generally not associated with generating PK- resistant PrPSc [189, 195], disrupting PrP folding and/or normal physiological function appears to be a more plausible alternative explanation. In accordance with this notion, transgenic mice expressing middle region deletion PrP mutants developed spontaneous neurodegeneration on PrP null background [196-198], and this result is explained by a hypothesis that the interaction between PrP middle region and a putative ligand is important of neuronal survival [196, 198].

Notably, lipid molecules have been postulated as a potential ligand [196], and the above result, that the middle region is involved in the PrP-lipid interaction, would be consistent with this hypothesis.

In PrP middle region, polymorphism at codon 129 (MM, MV or VV) has a dramatic effect on both susceptibility to and phenotype of human prion diseases

[190]. Structure studies of PrP variants bearing M or V at codon 129 revealed that no PrP conformation and stability changes have been found between these two PrP forms when they are alone in solution [199], which suggests that their effects on pathogenesis of prion disease are not resulted from their tertiary

121 structures. In this study, I found that the M and V variants differentiated from each other by their hydrophobic interactions with lipid. The hydrophobic

interaction between rPrP129M and lipid was stronger than that between rPrP129V and lipid (Figure 4.25). It is interesting that the hydrophobicity of residue valine is actually higher than that of residue methionine, which seems to be in contrary to our observation in this study. As shown in this study, PrP undergoes conformational changes due to the hydrophobic interaction with lipid, which involves the middle hydrophobic region of PrP and the hydrophobic core of lipid.

Therefore, it is not only residue 129, but multiple residues in the hydrobic region of rPrP are involved in this interaction. Substitution of methionine with valine at

129 might change the way in which the surrounding residues interact with lipid.

And this change could reduce the total hydrophobic interaction between lipid and the middle region, despite of the hydrophobicity increase at 129. Nonetheless, this finding confirms the result that the middle region is involved in the PrP-lipid interaction and, more importantly, provides an alternative perspective for the study of the 129-polymorphism effect on human prion disease.

Aggregation, but not formation of amyloid fibers, stabilizes the C-terminal

PK-resistance

A novel finding in this study is that the rPrP PK-resistance conformation

122 does not correlate to amyloid fiber formation. Previous reports of full-length rPrP conversion are all associated with the generation of amyloid fibers, which are formed in an acidic environment in the presence denaturants [157], by denaturing at high temperature [200], or by incubation with lipid bicelles followed by a 65°C heating step [148]. It has been shown that full-length PrPSc isolated from

diseased brain is not in an amyloid fiber state [201, 202] and the most infectious

prion protein particles are PrP oligomers instead of amyloid fibers [203].

Moreover, it is showed here that lipid oxidation or copper reduces the formation

of C-terminal PK-resistant rPrP, while both ferrous and ferric ions enhance it,

which is similar to their effects on PrPSc [161, 187, 204, 205]. The requirement of

middle region of PrP has been found to be essential for PrPSc propagation [206-

210]. These comparisons revealed that the non-amyloid-associated C-terminal

PK-resistant conformation identified in this study at least recapitulates certain

biochemical features of PrPSc.

Conclusion and future directions

Collectively, the results presented in this thesis provide strong evidence

supporting the relevance of PrP-lipid interaction to the pathogenesis of prion

disease and/or in the normal biological function of PrP.

However, the PK-resistance does not equal to prion infectivity and the

123 unique property of a prion resides in its self-perpetuation ability. To determine

whether the lipid-induced PK-resistant rPrP is capable of self-perpetuating and/or

is infectious, it is imperative to allow the PK-resistant rPrP to form infection-

competent aggregates [203]. It is showed here that aggregation stabilizes the C-

terminal PK-resistant conformation, but how it aggregates might be crucial for

creating a real prion. Moreover, as suggested by previous studies in prion

infected cells [106, 167], specific compositions of lipid may also be playing a role

in generating infection-competent PrP form. These possibilities need to be explored in future studies.

124 REFERENCES

1. Prusiner, S.B., Prions. Proc Natl Acad Sci U S A, 1998. 95(23): p. 13363- 83.

2. Weissmann, C., Molecular biology of prion disease. Trends in Cell Biol., 1994. 4: p. 10.

3. Aguzzi, A. and F.L. Heppner, Pathogenesis of prion diseases: a progress report. Cell Death Differ, 2000. 7(10): p. 889-902.

4. Caughey, B. and B. Chesebro, Prion protein and the transmissible spongiform encephalopathies. Trends in Cell Biol., 1997. 7: p. 56-62.

5. Deleault, N.R., et al., From the Cover: Formation of native prions from minimal components in vitro. Proc Natl Acad Sci U S A, 2007. 104(23): p. 9741-6.

6. Gambetti, P., et al., Sporadic and familial CJD: classification and characterisation. Br Med Bull, 2003. 66: p. 213-39.

7. Prusiner, S.B., et al., Purification and structural studies of a major scrapie prion protein. Cell, 1984. 38(1): p. 127-34.

8. Oesch, B., et al., A cellular gene encodes scrapie PrP 27-30 protein. Cell, 1985. 40(4): p. 735-46.

9. Gambetti, P., P. Parchi, and R. Petersen, et al., Inherited prion diseases, in Prion biology and diseases, P. SB, Editor. 1999, Cold Spring Harbor Laboratory Press: Cold Spring Harbor (NY). p. 509-584.

125 10. Aguzzi, A., F. Baumann, and J. Bremer, The prion's elusive reason for being. Annu Rev Neurosci, 2008. 31: p. 439-77.

11. Gajdusek, D.C. and V. Zigas, Degenerative disease of the central nervous system in New Guinea, the endemic occurrence of kuru in the native population. N. Engl. J. Med., 1957. 257: p. 974-978.

12. Gajdusek, D.C., Unconventional viruses and the origin and disappearance of kuru. Science, 1977. 197(4307): p. 943-60.

13. Gajdusek, D.C., C.J. Gibbs, and M. Alpers, Experimental transmission of a Kuru-like syndrome to chimpanzees. Nature, 1966. 209(5025): p. 794-6.

14. Gajdusek, D.C., C.J. Gibbs, Jr., and M. Alpers, Transmission and passage of experimenal "kuru" to chimpanzees. Science, 1967. 155(759): p. 212-4.

15. Duffy, P., et al., Letter: Possible person-to-person transmission of Creutzfeldt-Jakob disease. N Engl J Med, 1974. 290(12): p. 692-3.

16. Brown, P., et al., Iatrogenic Creutzfeldt-Jakob disease at the millennium. Neurology, 2000. 55(8): p. 1075-81.

17. Will, R.G., et al., A new variant of Creutzfeldt-Jakob disease in the UK. Lancet, 1996. 347(9006): p. 921-5.

18. Collinge, J., et al., Molecular analysis of prion strain variation and the aetiology of 'new variant' CJD [see comments]. Nature, 1996. 383(6602): p. 685-90.

19. Bruce, M.E., et al., Transmissions to mice indicate that 'new variant' CJD is caused by the BSE agent. Nature, 1997. 389(6650): p. 498-501.

126 20. Hill, A.F., et al., The same prion strain causes vCJD and BSE. Nature, 1997. 389(6650): p. 448-50, 526.

21. Scott, M.R., et al., Compelling transgenetic evidence for transmission of bovine spongiform encephalopathy prions to humans. Proc Natl Acad Sci U S A, 1999. 96(26): p. 15137-42.

22. Aguzzi, A. and M. Glatzel, Prion infections, blood and transfusions. Nat Clin Pract Neurol, 2006. 2(6): p. 321-9.

23. Ladogana, A., et al., Mortality from Creutzfeldt-Jakob disease and related disorders in Europe, Australia, and Canada. Neurology, 2005. 64(9): p. 1586-91.

24. Aguzzi, A., C. Sigurdson, and M. Heikenwaelder, Molecular mechanisms of prion pathogenesis. Annu Rev Pathol, 2008. 3: p. 11-40.

25. Budka, H., et al., Neuropathological diagnostic criteria for Creutzfeldt- Jakob disease (CJD) and other human spongiform encephalopathies (prion diseases). Brain Pathol, 1995. 5(4): p. 459-66.

26. Prusiner, S.B., Shattuck lecture--neurodegenerative diseases and prions. N Engl J Med, 2001. 344(20): p. 1516-26.

27. Collinge, J., Prion diseases of humans and animals: their causes and molecular basis. Annu Rev Neurosci, 2001. 24: p. 519-50.

28. du Plessis, D.G., Prion protein disease and neuropathology of prion disease. Neuroimaging Clin N Am, 2008. 18(1): p. 163-82; ix.

29. Brown, P. and R. Bradley, 1755 and all that: a historical primer of transmissible spongiform encephalopathy. Bmj, 1998. 317(7174): p. 1688-

127 92.

30. Schreuder, B.E., General aspects of transmissible spongiform encephalopathies and hypotheses about the agents. Vet Q, 1993. 15(4): p. 167-74.

31. Novakofski, J., et al., Prion biology relevant to bovine spongiform encephalopathy. J Anim Sci, 2005. 83(6): p. 1455-76.

32. Anderson, R.M., et al., Transmission dynamics and epidemiology of BSE in British cattle. Nature, 1996. 382(6594): p. 779-88.

33. Glatzel, M. and A. Aguzzi, The shifting biology of prions. Brain Res Brain Res Rev, 2001. 36(2-3): p. 241-8.

34. Cockcroft, P.D., The similarity of the physical sign frequencies of bovine spongiform encephalopathy and selected differential diagnoses. Vet J, 2004. 167(2): p. 175-80.

35. Wilesmith, J.W., et al., Bovine spongiform encephalopathy: epidemiological studies. Vet Rec, 1988. 123(25): p. 638-44.

36. Wilesmith, J.W., J.B. Ryan, and M.J. Atkinson, Bovine spongiform encephalopathy: epidemiological studies on the origin. Vet Rec, 1991. 128(9): p. 199-203.

37. Wilesmith, J.W., et al., Bovine spongiform encephalopathy: epidemiological features 1985 to 1990. Vet Rec, 1992. 130(5): p. 90-4.

38. Williams, E.S. and S. Young, Chronic wasting disease of captive mule deer: a spongiform encephalopathy. J Wildl Dis, 1980. 16(1): p. 89-98.

128 39. Spraker, T.R., et al., Spongiform encephalopathy in free-ranging mule deer (Odocoileus hemionus), white-tailed deer (Odocoileus virginianus) and Rocky Mountain elk (Cervus elaphus nelsoni) in northcentral Colorado. J Wildl Dis, 1997. 33(1): p. 1-6.

40. Miller, M.W., M.A. Wild, and E.S. Williams, Epidemiology of chronic wasting disease in captive Rocky Mountain elk. J Wildl Dis, 1998. 34(3): p. 532-8.

41. Williams, E.S. and S. Young, Neuropathology of chronic wasting disease of mule deer (Odocoileus hemionus) and elk (Cervus elaphus nelsoni). Vet Pathol, 1993. 30(1): p. 36-45.

42. Miller, M.W. and E.S. Williams, Prion disease: horizontal prion transmission in mule deer. Nature, 2003. 425(6953): p. 35-6.

43. Miller, M.W. and E.S. Williams, Chronic wasting disease of cervids. Curr Top Microbiol Immunol, 2004. 284: p. 193-214.

44. Hamir, A.N., et al., Transmission of sheep scrapie to elk (Cervus elaphus nelsoni) by intracerebral inoculation: final outcome of the experiment. J Vet Diagn Invest, 2004. 16(4): p. 316-21.

45. Browning, S.R., et al., Transmission of prions from mule deer and elk with chronic wasting disease to transgenic mice expressing cervid PrP. J Virol, 2004. 78(23): p. 13345-50.

46. Kong, Q., et al., Chronic wasting disease of elk: transmissibility to humans examined by transgenic mouse models. J Neurosci, 2005. 25(35): p. 7944-9.

47. Raymond, G.J., et al., Evidence of a molecular barrier limiting susceptibility of humans, cattle and sheep to chronic wasting disease.

129 Embo J, 2000. 19(17): p. 4425-30.

48. Chandler, R.L., Encephalopathy in mice produced by inoculation with scrapie brain material. Lancet, 1961. 1(7191): p. 1378-9.

49. Hadlow, W.J., Scrapie and kuru. Lancet, 1959. ii: p. 289-290.

50. Klatzo, I., D.C. Gajdusek, and V. Zigas, Pathology of Kuru. Lab Invest, 1959. 8(4): p. 799-847.

51. Gibbs, C.J., Jr., et al., Creutzfeldt-Jakob disease (spongiform encephalopathy): transmission to the chimpanzee. Science, 1968. 161(839): p. 388-9.

52. Sigurdsson, B., Rida, a chronic encephalitis of sheep. British Veterinary Journal, 1954. 110: p. 341-354.

53. Alper, T., D.A. Haig, and M.C. Clarke, The exceptionally small size of the scrapie agent. Biochem Biophys Res Commun, 1966. 22(3): p. 278-84.

54. Alper, T., et al., Does the agent of scrapie replicate without nucleic acid? Nature, 1967. 214(5090): p. 764-6.

55. Latarjet, R., et al., Inactivation of the scrapie agent by near monochromatic ultraviolet light. Nature, 1970. 227(5265): p. 1341-3.

56. Gibbs, C.J., Jr., D.C. Gajdusek, and R. Latarjet, Unusual resistance to ionizing radiation of the viruses of kuru, Creutzfeldt-Jakob disease, and scrapie. Proc Natl Acad Sci U S A, 1978. 75(12): p. 6268-70.

57. Griffith, J.S., Self-replication and scrapie. Nature, 1967. 215(5105): p. 1043-4. 130 58. Pattison, I.H. and K.M. Jones, The possible nature of the transmissible agent of scrapie. Vet Rec, 1967. 80(1): p. 2-9.

59. Gibbons, R.A. and G.D. Hunter, Nature of the scrapie agent. Nature, 1967. 215(5105): p. 1041-3.

60. Hunter, G.D., R.H. Kimberlin, and R.A. Gibbons, Scrapie: a modified membrane hypothesis. J Theor Biol, 1968. 20(3): p. 355-7.

61. Field, E.J., et al., Susceptibility of scrapie agent to ionizing radiation. Nature, 1969. 222(5188): p. 90-1.

62. Hunter, G.D., Scrapie: a prototype slow infection. J Infect Dis, 1972. 125(4): p. 427-40.

63. Pattison, I.H., Resistance of the Scrapie Agent to Formalin. J Comp Pathol, 1965. 75: p. 159-64.

64. Hartley, E.G., Action of disinfectants on experimental mouse scrapie. Nature, 1967. 213(5081): p. 1135.

65. Lavelle, G.C., Large fraction of scrapie virus resistant to lipid solvents. Proc Soc Exp Biol Med, 1972. 141(2): p. 460-2.

66. Hunter, G.D. and G.C. Millson, Attempts to release the scrapie agent from tissue debris. J Comp Pathol, 1967. 77(3): p. 301-7.

67. Hunter, G.D., et al., Further studies of the infectivity and stability of extracts and homogenates derived from scrapie affected mouse brains. J Comp Pathol, 1969. 79(1): p. 101-8.

68. Millson, G.C., G.D. Hunter, and R.H. Kimberlin, The physico-chemical 131 nature of the scrapie agent. Front Biol, 1976. 44: p. 243-66.

69. Brown, P., et al., Effect of chemicals, heat, and histopathologic processing on high-infectivity hamster-adapted scrapie virus. J Infect Dis, 1982. 145(5): p. 683-7.

70. Prusiner, S.B., Novel proteinaceous infectious particles cause scrapie. Science, 1982. 216(4542): p. 136-44.

71. Prusiner, S.B., et al., Sedimentation properties of the scrapie agent. Proc Natl Acad Sci U S A, 1977. 74(10): p. 4656-60.

72. Prusiner, S.B., et al., Molecular properties, partial purification, and assay by incubation period measurements of the hamster scrapie agent. Biochemistry, 1980. 19(21): p. 4883-91.

73. Bolton, D.C., M.P. McKinley, and S.B. Prusiner, Identification of a protein that purifies with the scrapie prion. Science, 1982. 218(4579): p. 1309-11.

74. Prusiner, S.B., et al., Further purification and characterization of scrapie prions. Biochemistry, 1982. 21(26): p. 6942-50.

75. Chesebro, B., et al., Identification of scrapie prion protein-specific mRNA in scrapie-infected and uninfected brain. Nature, 1985. 315(6017): p. 331- 3.

76. Meyer, R.K., et al., Separation and properties of cellular and scrapie prion proteins. Proc Natl Acad Sci U S A, 1986. 83(8): p. 2310-4.

77. Locht, C., et al., Molecular cloning and complete sequence of prion protein cDNA from mouse brain infected with the scrapie agent. Proc Natl Acad Sci U S A, 1986. 83(17): p. 6372-6.

132 78. Bolton, D.C., et al., Isolation and structural studies of the intact scrapie agent protein. Arch Biochem Biophys, 1987. 258(2): p. 579-90.

79. Caughey, B.W., et al., Secondary structure analysis of the scrapie- associated protein PrP 27-30 in water by infrared spectroscopy. Biochemistry, 1991. 30(31): p. 7672-80.

80. Pan, K.M., et al., Conversion of alpha-helices into beta-sheets features in the formation of the scrapie prion proteins. Proc Natl Acad Sci U S A, 1993. 90(23): p. 10962-6.

81. Stahl, N., et al., Structural studies of the scrapie prion protein using mass spectrometry and amino acid sequencing. Biochemistry, 1993. 32(8): p. 1991-2002.

82. McKinley, M.P., D.C. Bolton, and S.B. Prusiner, A protease-resistant protein is a structural component of the scrapie prion. Cell, 1983. 35(1): p. 57-62.

83. Bolton, D.C., M.P. McKinley, and S.B. Prusiner, Molecular characteristics of the major scrapie prion protein. Biochemistry, 1984. 23(25): p. 5898- 906.

84. Hsiao, K.K., et al., Serial transmission in rodents of neurodegeneration from transgenic mice expressing mutant prion protein. Proceedings of the National Academy of Sciences of the United States of America, 1994. 91(19): p. 9126-30.

85. Piccardo, P., et al., Phenotypic variability of Gerstmann-Straussler- Scheinker disease is associated with prion protein heterogeneity. J Neuropathol Exp Neurol, 1998. 57(10): p. 979-88.

86. Oesch, B., et al., Search for a scrapie-specific nucleic acid: a progress

133 report. Ciba Found Symp, 1988. 135: p. 209-23.

87. Kellings, K., et al., Further analysis of nucleic acids in purified scrapie prion preparations by improved return refocusing gel electrophoresis. J Gen Virol, 1992. 73 ( Pt 4): p. 1025-9.

88. Kellings, K., et al., Analysis of nucleic acids in purified scrapie prion preparations. Arch Virol Suppl, 1993. 7: p. 215-25.

89. Riesner, D., et al., Prions and nucleic acids: search for "residual" nucleic acids and screening for mutations in the PrP-gene. Dev Biol Stand, 1993. 80: p. 173-81.

90. Kellings, K., S.B. Prusiner, and D. Riesner, Nucleic acids in prion preparations: unspecific background or essential component? Philos Trans R Soc Lond B Biol Sci, 1994. 343(1306): p. 425-30.

91. Diringer, H., et al., Highly infectious purified preparations of disease- specific amyloid of transmissible spongiform encephalopathies are not devoid of nucleic acids of viral size. Intervirology, 1997. 40(4): p. 238-46.

92. Klein, T.R., et al., Prion rods contain small amounts of two host sphingolipids as revealed by thin-layer chromatography and mass spectrometry. Biol Chem, 1998. 379(6): p. 655-66.

93. Alper, T., The scrapie enigma: insights from radiation experiments. Radiat Res, 1993. 135(3): p. 283-92.

94. Gabizon, R., M.P. McKinley, and S.B. Prusiner, Purified prion proteins and scrapie infectivity copartition into liposomes. Proc Natl Acad Sci U S A, 1987. 84(12): p. 4017-21.

134 95. Baron, G.S., et al., Mouse-adapted scrapie infection of SN56 cells: greater efficiency with microsome-associated versus purified PrP-res. J Virol, 2006. 80(5): p. 2106-17.

96. Pattison, I.H., Scrapie in the welsh mountain breed of sheep and its experimental transmission to goats. Vet Rec, 1965. 77(47): p. 1388-90.

97. Prusiner, S.B., et al., Transgenetic studies implicate interactions between homologous PrP isoforms in scrapie prion replication. Cell, 1990. 63(4): p. 673-86.

98. Bruce, M., et al., Transmission of bovine spongiform encephalopathy and scrapie to mice: strain variation and the species barrier. Philos Trans R Soc Lond B Biol Sci, 1994. 343(1306): p. 405-11.

99. Puckett, C., et al., Genomic structure of the human prion protein gene. Am J Hum Genet, 1991. 49(2): p. 320-9.

100. Schatzl, H.M., et al., Prion protein gene variation among primates. J Mol Biol, 1995. 245(4): p. 362-74.

101. Wopfner, F., et al., Analysis of 27 mammalian and 9 avian PrPs reveals high conservation of flexible regions of the prion protein. J Mol Biol, 1999. 289(5): p. 1163-78.

102. Kretzschmar, H.A., et al., Scrapie prion proteins are synthesized in neurons. Am J Pathol, 1986. 122(1): p. 1-5.

103. Mobley, W.C., et al., Nerve growth factor increases mRNA levels for the prion protein and the beta-amyloid protein precursor in developing hamster brain. Proc Natl Acad Sci U S A, 1988. 85(24): p. 9811-5.

135 104. Manson, J., et al., The prion protein gene: a role in mouse embryogenesis? Development, 1992. 115(1): p. 117-22.

105. Wickner, R.B., Cell biology of PrP, in Prion diseases of mammals and yeast : molecular mechanisms and genetic features. 1997, Chapman & Hall: New York. p. 51-57.

106. Taraboulos, A., et al., Cholesterol depletion and modification of COOH- terminal targeting sequence of the prion protein inhibit formation of the scrapie isoform. J Cell Biol, 1995. 129(1): p. 121-32.

107. Schroeder, R., E. London, and D. Brown, Interactions between saturated acyl chains confer detergent resistance on lipids and glycosylphosphatidylinositol (GPI)-anchored proteins: GPI-anchored proteins in liposomes and cells show similar behavior. Proc Natl Acad Sci U S A, 1994. 91(25): p. 12130-4.

108. Brown, D.A. and E. London, Structure and function of sphingolipid- and cholesterol-rich membrane rafts. J Biol Chem, 2000. 275(23): p. 17221-4.

109. Brown, D., Structure and function of membrane rafts. Int J Med Microbiol, 2002. 291(6-7): p. 433-7.

110. Caughey, B. and G.J. Raymond, The scrapie-associated form of PrP is made from a cell surface precursor that is both protease- and phospholipase-sensitive. J Biol Chem, 1991. 266(27): p. 18217-23.

111. Shyng, S.L., M.T. Huber, and D.A. Harris, A prion protein cycles between the cell surface and an endocytic compartment in cultured neuroblastoma cells. J Biol Chem, 1993. 268(21): p. 15922-8.

112. McKinley, M.P., et al., Ultrastructural localization of scrapie prion proteins in cytoplasmic vesicles of infected cultured cells. Lab Invest, 1991. 65(6):

136 p. 622-30.

113. Taraboulos, A., et al., Synthesis and trafficking of prion proteins in cultured cells. Mol Biol Cell, 1992. 3(8): p. 851-63.

114. 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.

115. Borchelt, D.R., A. Taraboulos, and S.B. Prusiner, Evidence for synthesis of scrapie prion proteins in the endocytic pathway. J Biol Chem, 1992. 267(23): p. 16188-99.

116. Harris, D.A., Cellular biology of prion diseases. Clin Microbiol Rev, 1999. 12(3): p. 429-44.

117. Hornemann, S. and R. Glockshuber, Autonomous and reversible folding of a soluble amino-terminally truncated segment of the mouse prion protein. J Mol Biol, 1996. 261(5): p. 614-9.

118. Hornemann, S., et al., Recombinant full-length murine prion protein, mPrP(23-231): purification and spectroscopic characterization. FEBS Lett, 1997. 413(2): p. 277-81.

119. Riek, R., et al., NMR characterization of the full-length recombinant murine prion protein, mPrP(23-231). FEBS Lett, 1997. 413(2): p. 282-8.

120. Donne, D.G., et al., Structure of the recombinant full-length hamster prion protein PrP(29- 231): the N terminus is highly flexible. Proc Natl Acad Sci U S A, 1997. 94(25): p. 13452-7.

137 121. 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.

122. Lopez Garcia, F., et al., NMR structure of the bovine prion protein. Proc Natl Acad Sci U S A, 2000. 97(15): p. 8334-9.

123. Viles, J.H., et al., Copper binding to the prion protein: structural implications of four identical cooperative binding sites. Proc Natl Acad Sci U S A, 1999. 96(5): p. 2042-7.

124. Cohen, F.E., et al., Structural clues to prion replication. Science, 1994. 264(5158): p. 530-1.

125. Huang, Z., S.B. Prusiner, and F.E. Cohen, Scrapie prions: a three- dimensional model of an infectious fragment. Fold Des, 1995. 1(1): p. 13-9.

126. Cohen, F.E., Protein misfolding and prion diseases. J Mol Biol, 1999. 293(2): p. 313-20.

127. Swietnicki, W., et al., Familial mutations and the thermodynamic stability of the recombinant human prion protein. J Biol Chem, 1998. 273(47): p. 31048-52.

128. Liemann, S. and R. Glockshuber, Influence of amino acid substitutions related to inherited human prion diseases on the thermodynamic stability of the cellular prion protein. Biochemistry, 1999. 38(11): p. 3258-67.

129. Zhang, Y., et al., Solution structure of the E200K variant of human prion protein. Implications for the mechanism of pathogenesis in familial prion diseases. J Biol Chem, 2000. 275(43): p. 33650-4.

130. Bueler, H., et al., Normal development and behaviour of mice lacking the

138 neuronal cell-surface PrP protein. Nature, 1992. 356(6370): p. 577-82.

131. Basler, K., et al., Scrapie and cellular PrP isoforms are encoded by the same chromosomal gene. Cell, 1986. 46(3): p. 417-28.

132. Hope, J., et al., The major polypeptide of scrapie-associated fibrils (SAF) has the same size, charge distribution and N-terminal protein sequence as predicted for the normal brain protein (PrP). Embo J, 1986. 5(10): p. 2591- 7.

133. Stahl, N., et al., Glycosylinositol phospholipid anchors of the scrapie and cellular prion proteins contain sialic acid. Biochemistry, 1992. 31(21): p. 5043-53.

134. Wille, H., et al., Structural studies of the scrapie prion protein by electron crystallography. Proc Natl Acad Sci U S A, 2002. 99(6): p. 3563-8.

135. Govaerts, C., et al., Evidence for assembly of prions with left-handed beta- helices into trimers. Proc Natl Acad Sci U S A, 2004. 101(22): p. 8342-7.

136. Kocisko, D.A., et al., Cell-free formation of protease-resistant prion protein. Nature, 1994. 370(6489): p. 471-4.

137. Saborio, G.P., B. Permanne, and C. Soto, Sensitive detection of pathological prion protein by cyclic amplification of protein misfolding. Nature, 2001. 411(6839): p. 810-3.

138. Atarashi, R., et al., Ultrasensitive detection of scrapie prion protein using seeded conversion of recombinant prion protein. Nat Methods, 2007. 4(8): p. 645-50.

139. Castilla, J., et al., In vitro generation of infectious scrapie prions. Cell,

139 2005. 121(2): p. 195-206.

140. Deleault, N.R., et al., Protease-resistant prion protein amplification reconstituted with partially purified substrates and synthetic polyanions. J Biol Chem, 2005. 280(29): p. 26873-9.

141. Deleault, N.R., R.W. Lucassen, and S. Supattapone, RNA molecules stimulate prion protein conversion. Nature, 2003. 425(6959): p. 717-20.

142. Wong, C., et al., Sulfated glycans and elevated temperature stimulate PrP(Sc)-dependent cell-free formation of protease-resistant prion protein. Embo J, 2001. 20(3): p. 377-86.

143. Baron, G.S. and B. Caughey, Effect of glycosylphosphatidylinositol anchor-dependent and -independent prion protein association with model raft membranes on conversion to the protease-resistant isoform. J Biol Chem, 2003. 278(17): p. 14883-92.

144. Jackson, G.S., et al., Reversible conversion of monomeric human prion protein between native and fibrilogenic conformations. Science, 1999. 283(5409): p. 1935-7.

145. Baskakov, I.V., et al., Pathway complexity of prion protein assembly into amyloid. J Biol Chem, 2002. 277(24): p. 21140-8.

146. Baskakov, I.V., et al., Folding of prion protein to its native alpha-helical conformation is under kinetic control. J Biol Chem, 2001. 276(23): p. 19687-90.

147. Leffers, K.W., et al., Assembly of natural and recombinant prion protein into fibrils. Biol Chem, 2005. 386(6): p. 569-80.

140 148. Luhrs, T., R. Zahn, and K. Wuthrich, Amyloid formation by recombinant full-length prion proteins in phospholipid bicelle solutions. J Mol Biol, 2006. 357(3): p. 833-41.

149. Zahn, R., C. von Schroetter, and K. Wuthrich, Human prion proteins expressed in Escherichia coli and purified by high-affinity column refolding. FEBS Lett, 1997. 417(3): p. 400-4.

150. Morillas, M., et al., Membrane environment alters the conformational structure of the recombinant human prion protein. J Biol Chem, 1999. 274(52): p. 36859-65.

151. Sanghera, N. and T.J. Pinheiro, Binding of prion protein to lipid membranes and implications for prion conversion. J Mol Biol, 2002. 315(5): p. 1241-56.

152. Huber, L.A., K. Pfaller, and I. Vietor, Organelle proteomics: implications for subcellular fractionation in proteomics. Circ Res, 2003. 92(9): p. 962-8.

153. Caughey, B., Prion protein conversions: insight into mechanisms, TSE transmission barriers and strains. Br Med Bull, 2003. 66: p. 109-20.

154. Weissmann, C. and A. Aguzzi, Approaches to therapy of prion diseases. Annu Rev Med, 2005. 56: p. 321-44.

155. Aguzzi, A., Prion diseases of humans and farm animals: epidemiology, genetics, and pathogenesis. J Neurochem, 2006. 97(6): p. 1726-39.

156. Swietnicki, W., et al., Aggregation and fibrillization of the recombinant human prion protein huPrP90-231. Biochemistry, 2000. 39(2): p. 424-31.

157. Bocharova, O.V., et al., In vitro conversion of full-length mammalian prion

141 protein produces amyloid form with physical properties of PrP(Sc). J Mol Biol, 2005. 346(2): p. 645-59.

158. Adler, V., et al., Small, highly structured RNAs participate in the conversion of human recombinant PrP(Sen) to PrP(Res) in vitro. J Mol Biol, 2003. 332(1): p. 47-57.

159. Qin, K., et al., Copper(II)-induced conformational changes and protease resistance in recombinant and cellular PrP. Effect of protein age and deamidation. J Biol Chem, 2000. 275(25): p. 19121-31.

160. Brown, D.R., et al., Consequences of manganese replacement of copper for prion protein function and proteinase resistance. Embo J, 2000. 19(6): p. 1180-6.

161. Basu, S., et al., Modulation of proteinase K-resistant prion protein in cells and infectious brain homogenate by redox iron: implications for prion replication and disease pathogenesis. Mol Biol Cell, 2007. 18(9): p. 3302- 12.

162. Yin, S., et al., Human prion proteins with pathogenic mutations share common conformational changes resulting in enhanced binding to glycosaminoglycans. Proc Natl Acad Sci U S A, 2007. 104(18): p. 7546-51.

163. Horonchik, L., et al., Heparan sulfate is a cellular receptor for purified infectious prions. J Biol Chem, 2005. 280(17): p. 17062-7.

164. Kazlauskaite, J. and T.J. Pinheiro, Aggregation and fibrillization of prions in lipid membranes. Biochem Soc Symp, 2005(72): p. 211-22.

165. White, S.H., Translocons, thermodynamics, and the folding of membrane proteins. FEBS Lett, 2003. 555(1): p. 116-21.

142 166. White, S.H., et al., How membranes shape protein structure. J Biol Chem, 2001. 276(35): p. 32395-8.

167. Naslavsky, N., et al., Sphingolipid depletion increases formation of the scrapie prion protein in neuroblastoma cells infected with prions. J Biol Chem, 1999. 274(30): p. 20763-71.

168. Caughey, B., et al., Normal and scrapie-associated forms of prion protein differ in their sensitivities to phospholipase and proteases in intact neuroblastoma cells. J Virol, 1990. 64(3): p. 1093-101.

169. Stahl, N., D.R. Borchelt, and S.B. Prusiner, Differential release of cellular and scrapie prion proteins from cellular membranes by phosphatidylinositol-specific phospholipase C. Biochemistry, 1990. 29(22): p. 5405-12.

170. Kazlauskaite, J., et al., Structural changes of the prion protein in lipid membranes leading to aggregation and fibrillization. Biochemistry, 2003. 42(11): p. 3295-304.

171. Critchley, P., et al., Binding of prion proteins to lipid membranes. Biochem Biophys Res Commun, 2004. 313(3): p. 559-67.

172. Caughey, B., G.J. Raymond, and R.A. Bessen, Strain-dependent differences in beta-sheet conformations of abnormal prion protein. J Biol Chem, 1998. 273(48): p. 32230-5.

173. Safar, J., et al., Thermal stability and conformational transitions of scrapie amyloid (prion) protein correlate with infectivity. Protein Sci, 1993. 2(12): p. 2206-16.

174. Elfrink, K., et al., Structural changes of membrane-anchored native PrP(C). Proc Natl Acad Sci U S A, 2008. 105(31): p. 10815-9.

143 175. Legname, G., et al., Synthetic mammalian prions. Science, 2004. 305(5684): p. 673-6.

176. Chesebro, B., et al., Anchorless prion protein results in infectious amyloid disease without clinical scrapie. Science, 2005. 308(5727): p. 1435-9.

177. Vanik, D.L., K.A. Surewicz, and W.K. Surewicz, Molecular basis of barriers for interspecies transmissibility of mammalian prions. Mol Cell, 2004. 14(1): p. 139-45.

178. Jones, E.M. and W.K. Surewicz, Fibril conformation as the basis of species- and strain-dependent seeding specificity of mammalian prion amyloids. Cell, 2005. 121(1): p. 63-72.

179. Re, F., et al., Prion protein structure is affected by pH-dependent interaction with membranes: a study in a model system. FEBS Lett, 2008. 582(2): p. 215-20.

180. Ma, J. and S. Lindquist, De novo generation of a PrPSc-like conformation in living cells. Nat Cell Biol, 1999. 1(6): p. 358-361.

181. Ma, J. and S. Lindquist, Conversion of PrP to a self-perpetuating PrPSc- like conformation in the cytosol. Science, 2002. 298(5599): p. 1785-8.

182. Wang, X., et al., The interaction between cytoplasmic prion protein and the hydrophobic lipid core of membrane correlates with neurotoxicity. J Biol Chem, 2006. 281(19): p. 13559-65.

183. Polymenidou, M., et al., Coexistence of multiple PrPSc types in individuals with Creutzfeldt-Jakob disease. Lancet Neurol, 2005. 4(12): p. 805-14.

184. Pan, T., et al., Epitope scanning reveals gain and loss of strain specific

144 antibody binding epitopes associated with the conversion of normal cellular prion to scrapie prion. J Neurochem, 2004. 90(5): p. 1205-17.

185. Pan, T., et al., Biochemical fingerprints of prion infection: accumulations of aberrant full-length and N-terminally truncated PrP species are common features in mouse prion disease. J Virol, 2005. 79(2): p. 934-43.

186. Nazor, K.E., et al., Immunodetection of disease-associated mutant PrP, which accelerates disease in GSS transgenic mice. Embo J, 2005. 24(13): p. 2472-80.

187. Orem, N.R., et al., Copper (II) ions potently inhibit purified PrPres amplification. J Neurochem, 2006. 96(5): p. 1409-15.

188. Brown, D.R., et al., The cellular prion protein binds copper in vivo. Nature, 1997. 390(6661): p. 684-7.

189. Collins, S., C.A. McLean, and C.L. Masters, Gerstmann-Straussler- Scheinker syndrome,fatal familial insomnia, and kuru: a review of these less common human transmissible spongiform encephalopathies. J Clin Neurosci, 2001. 8(5): p. 387-97.

190. Ironside, J.W., D.L. Ritchie, and M.W. Head, Phenotypic variability in human prion diseases. Neuropathol Appl Neurobiol, 2005. 31(6): p. 565- 79.

191. Nishina, K., et al., In vitro prion protein conversion in detergent-solubilized membranes. Biochemistry, 2004. 43(9): p. 2613-21.

192. Xiong, L.W., et al., Conformational change, aggregation and fibril formation induced by detergent treatments of cellular prion protein. J Neurochem, 2001. 79(3): p. 669-78.

145 193. Hegde, R.S., et al., A transmembrane form of the prion protein in neurodegenerative disease. Science, 1998. 279(5352): p. 827-34.

194. Falnes, P.O. and K. Sandvig, Penetration of protein toxins into cells. Curr Opin Cell Biol, 2000. 12(4): p. 407-13.

195. Yamada, M., et al., An inherited prion disease with a PrP P105L mutation: clinicopathologic and PrP heterogeneity. Neurology, 1999. 53(1): p. 181-8.

196. Baumann, F., et al., Lethal recessive myelin toxicity of prion protein lacking its central domain. Embo J, 2007. 26(2): p. 538-47.

197. Li, A., et al., Neonatal lethality in transgenic mice expressing prion protein with a deletion of residues 105-125. Embo J, 2007. 26(2): p. 548-58.

198. Shmerling, D., et al., Expression of amino-terminally truncated PrP in the mouse leading to ataxia and specific cerebellar lesions. Cell, 1998. 93(2): p. 203-14.

199. Hosszu, L.L., et al., The residue 129 polymorphism in human prion protein does not confer susceptibility to Creutzfeldt-Jakob disease by altering the structure or global stability of PrPC. J Biol Chem, 2004. 279(27): p. 28515- 21.

200. Bocharova, O.V., et al., Annealing prion protein amyloid fibrils at high temperature results in extension of a proteinase K-resistant core. J Biol Chem, 2006. 281(4): p. 2373-9.

201. Prusiner, S.B., et al., Scrapie prions aggregate to form amyloid-like birefringent rods. Cell, 1983. 35(2 Pt 1): p. 349-58.

202. McKinley, M.P., et al., Scrapie prion rod formation in vitro requires both

146 detergent extraction and limited proteolysis. J Virol, 1991. 65(3): p. 1340- 51.

203. Silveira, J.R., et al., The most infectious prion protein particles. Nature, 2005. 437(7056): p. 257-61.

204. Solassol, J., et al., A novel copper-hydrogen peroxide formulation for prion decontamination. J Infect Dis, 2006. 194(6): p. 865-9.

205. Suyama, K., et al., Assessment of prion inactivation by fenton reaction using protein misfolding cyclic amplification and bioassay. Biosci Biotechnol Biochem, 2007. 71(8): p. 2069-71.

206. Enari, M., E. Flechsig, and C. Weissmann, Scrapie prion protein accumulation by scrapie-infected neuroblastoma cells abrogated by exposure to a prion protein antibody. Proc Natl Acad Sci U S A, 2001. 98(16): p. 9295-9.

207. Pankiewicz, J., et al., Clearance and prevention of prion infection in cell culture by anti-PrP antibodies. Eur J Neurosci, 2006. 23(10): p. 2635-47.

208. Feraudet, C., et al., Screening of 145 anti-PrP monoclonal antibodies for their capacity to inhibit PrPSc replication in infected cells. J Biol Chem, 2005. 280(12): p. 11247-58.

209. Holscher, C., H. Delius, and A. Burkle, Overexpression of nonconvertible PrPc delta114-121 in scrapie-infected mouse neuroblastoma cells leads to trans-dominant inhibition of wild-type PrP(Sc) accumulation. J Virol, 1998. 72(2): p. 1153-9.

210. Norstrom, E.M. and J.A. Mastrianni, The AGAAAAGA palindrome in PrP is required to generate a productive PrPSc-PrPC complex that leads to prion propagation. J Biol Chem, 2005. 280(29): p. 27236-43.

147