University of Southern California Dissertations and Theses

Structural studies on functional amyloids and the mechanism of aggregation and disaggregation of Huntingtin Exon 1

(USC Thesis Other)

Structural studies on functional amyloids and the mechanism of aggregation and disaggregation of Huntingtin Exon 1

PDF

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content

Structural Studies on Func0onal Amyloids and the
Mechanism of Aggrega0on and Disaggrega0on
of Hun0ng0n Exon 1

by

Silvia A. Cervantes Cortes

A Disserta3on Presented to the
FACULTY OF THE USC GRADUATE SCHOOL
UNIVERSITY OF SOUTHERN CALIFORNIA
In Par3al Fulfillment of the
Requirements for the Degree
DOCTOR OF PHILOSOPHY
(MEDICAL BIOPHYSICS)

August 2023
ii
Dedication

I dedicate this thesis to the memory of my dad. For all that he did for me. For the sacrifices
he made to give me the opportunities he never had and for his unwavering faith in me. His love
and support are one of the reasons I am who I am today.

iii
Acknowledgments
I would like to thank my funding support:
• NIH National Institute of Neurological Disorders and Stroke: Research Supplement to
Promote Diversity in Health-Related Research, 2015-2018
• NIH National Institute of Neurological Disorder and Stroke: Ruth L. Kirschstein National
Research Service Award (NRSA) Individual Predoctoral Fellowship (Parent F31) – Diversity,
2018-2021
• University of Southern California – Programs in Biomedical and Biological Sciences (PIBBS)
Fellowship, 2015
I would also like to sincerely thank the members of my thesis committee; Dr. Ralf Langen,
Dr. Tobias Ulmer and Dr. Ansgar Siemer for their experimental advice, their fruitful discussion
and their overall guidance during my PhD. Thank you also for creating such a collaborative
environment, it has been such a pleasure to be part of the Protein Structure Center.
I would especially like to thank my mentor Dr. Ansgar Siemer for his advice, support, and
patience thorough out my time here. It has been such a great experience to have you as a mentor.
Thank you for your trust and faith in me, for helping me preserver and for guiding me to become
a better scientist.
I would also like to thank my NIH Fellowship - F31 letter writers: Dr. Kensaku Nakayama,
Dr. Robert Farley, and Dr. Zoltan Tokes. Thank you for supporting my application, for your
guidance and for your genuine interest in my well-being and achievements.
iv
Thank you also to Dr. Pragna Patel, for her advice and encouragement and for her efforts
to create a more inclusive community for first generation, minority students such as myself. We
all appreciate your work and support.
Thank you also to Dr. Janine Kirstein and her lab members for their collaboration on our
work with DnaJB1.
I would also like to thank my Siemer lab mates, both present and former members. You
guys are truly the best and I am so grateful that I got to meet you all. It was great to work with
you all and to collaborate with you. I truly value the relationships we have built, and I am so
happy to know that you are all doing great things because you are wonderful people. Thank you
all for your support!
I also want to thank the rest of the members of the Langen lab and Ulmer lab for their
help and support. Thank you especially to Dr. Jose M. Bravo Arredondo, Dr. Mario J. Isas, Dr.
Anoop Rawat, Dr. Nitin K. Pandey, Dr. Jobin Varkey, and Mr. Alan Situ for their overall support
and assistance and experimental guidance throughout the years.
Finally, I would like to thank my family and friends, my loved ones, without whom all of
this would not have been possible. Thank you to all of you for your constant support and love,
for believing in me and for motivating me to be better every day. Thank you especially to my
mom. There will never be enough words to express how grateful I am to have you in my life, to
thank you for all that you have done and continue to do for me, and for all your unceasing support
and love. You are my rock!

v
Table of Contents
Dedication ...................................................................................................................................... ii
Acknowledgments ......................................................................................................................... iii
List of Tables ................................................................................................................................. vii
List of Figures ............................................................................................................................... viii
Abstract .......................................................................................................................................... x
Chapter 1: Introduction to Amyloid Fibrils ..................................................................................... 1
1.1 Amyloid Fibrils ...................................................................................................................... 1
1.2 Amyloid Fibrils in Disease ..................................................................................................... 3
1.3 Functional Amyloids ............................................................................................................. 5
1.4 Techniques Used in This Study ............................................................................................. 5
Chapter 2: Structural Studies on Huntingtin Exon 1 Monomer and Huntingtin Exon 1 Fibrils with
the Co-Chaperone DnaJB1 .............................................................................................................. 8
2.1 Introduction .......................................................................................................................... 8
2.2 Characterizing the Structure of HTTex1 Monomers on Agarose Beads ............................. 14
2.2.1 Results ......................................................................................................................... 14
2.2.2 Discussion .................................................................................................................... 20
2.3 Characterizing the Binding Interface of HTTex1 Fibrils and DnaJB1 ................................... 21
2.3.1 Results ......................................................................................................................... 21
2.3.2 Discussion .................................................................................................................... 39
2.4 Materials and Methods ...................................................................................................... 46
Chapter 3: Studies on Functional Amyloids .................................................................................. 55
3.1 Introduction ........................................................................................................................ 55
3.2 Metal Binding Properties of the N-terminus of the Functional Amyloid Orb2 ................... 59
3.2.1 Results ......................................................................................................................... 59
3.2.2 Discussion .................................................................................................................... 66
3.3 Characterization of the F5Y Mutant of Orb2A88 ................................................................ 70
3.3.1 Results ......................................................................................................................... 70
3.3.2 Discussion .................................................................................................................... 73
3.4 Materials and Methods ...................................................................................................... 75
Appendix A: ssNMR Characterization of ⍺-synuclein Mutants and Glycosylated ⍺-Synuclein ..... 80
Background ............................................................................................................................... 80
Results and Discussion .............................................................................................................. 82
Materials and Methods .......................................................................................................... 101
vi
Appendix B: Characterization of the HetS, ⍺-Synuclein Chimer (HET-sCT) and its Interaction with
DnaJB1 ........................................................................................................................................ 104
Background ............................................................................................................................. 104
Results and Discussion ............................................................................................................ 105
Materials and Methods .......................................................................................................... 118
References .................................................................................................................................. 124

vii
List of Tables
Table 3.1. Thermodynamic Values for Orb2A87 Binding to Divalent Metal Ions as Derived
from Fitting the ITC Data Shown in Figure 3.3 to a Model Assuming a Single
Binding Site .................................................................................................................. 62
Table 3.2. Thermodynamic Values for Orb2A87 Mutants Binding to Ni
2+
Derived from Fitting the
ITC Data Shown in Figure 3.4 ....................................................................................... 63
Table B1. Parameters for Protein Expression Screen ................................................................. 106

viii
List of Figures
Figure 1.1. Structure of Amyloids ................................................................................................... 1
Figure 1.2. High Resolution Structures Highlight the Structural Diversity of Amyloid Proteins ..... 2
Figure 1.3. Aggregation Pathway of Amyloid Fibrils ....................................................................... 4
Figure 1.4. Thioflavin T Provides a Means to Measure Amyloid Formation ................................... 6
Figure 2.1. Domain Organization and Structure of HTTex1 Fibrils ................................................. 9
Figure 2.2. Monomer Trapped on Agarose Beads ........................................................................ 10
Figure 2.3. Domain Organization of DnaJB1 ................................................................................. 13
Figure 2.4. HTTex1 with 6X-His Tag at C-terminus ....................................................................... 14
Figure 2.5. HTTex1 Q16 Monomer is Characterized by Structural Heterogeneity ....................... 15
Figure 2.6. The Dynamic Domains of HTTex1 Q16 Monomer Resemble Those of Q46 Fibrils ..... 16
Figure 2.7. HTTex1 with 6X-His at the Middle and N-terminus of the Construct ......................... 17
Figure 2.8. HTTex1 Q25 Monomer is Characterized by Dynamic Heterogeneity ......................... 19
Figure 2.9. DnaJB1 (40 kDa) Preferentially Binds Fibrils Made at Room Temperature (RT) ......... 23
Figure 2.10. DnaJB1 Affects the Morphology of HTTex1 Fibrils ................................................... 24
Figure 2.11. DnaJB1 Interaction Results in Narrowing of Fibril Diameter for All Three Fibril
Polymorphs and Fragmentation about Every 35 nm ................................................ 25
Figure 2.12. DnaJB1 Decorates the Surface of HTTex1 Fibrils When Added at Higher Ratios ...... 26
Figure 2.13. Incubation with Different Ratios of DnaJB1 Leads to Different Outcomes
for Fibrils ................................................................................................................... 27
Figure 2.14. DnaJB1 Binding Does Not Affect the Cross-b Core of HTTex1 .................................. 29
Figure 2.15. DnaJB1 Binding Results in Chemical Shift Perturbations in the Dynamic PRD
of HTTex1 .................................................................................................................. 30
Figure 2.16. HSQC and HNCA Show Major Chemical Shift Perturbation at the C12 Region of
HTTex1 ...................................................................................................................... 32
Figure 2.17. Addition of DnaJB1 Leads to Immobilization of Proline Residues in T Fibrils ........... 33
Figure 2.18. Temperature Screen Confirms Immobilization of Proline Residues ......................... 35
Figure 2.19. Binding of DnaJB1 to RT Fibrils Generate Low Molecular Weight Bands ................. 36
Figure 2.20. 25 kDa Band is Absent in Samples of HTTex1 with DnaJB1 in TFA dH 2O .................. 36
Figure 2.21. DnaJB1 Antibody Detects Band at 12 kDa, but not the Band at 25 kDa ................... 38
Figure 2.22. DnaJB1 C-terminus Antibody Detects Band at 25 kDa ............................................. 39
Figure 2.23. Low Molecular Weight Bands Might be the Result of DnaJB1 Fragmentation ......... 46
Figure 3.1. The N-terminal 88 Amino Acids of Orb2A are Essential for Aggregate Formation in S2
Cells and can Form Amyloid Fibrils in-vitro ................................................................ 58
Figure 3.2. Orb2A87 Binds Several Transition Metal Ions in the Absence of a His-tag ................ 60
Figure 3.3. Orb2A87 Binds Ni
2+
, Zn
2+
, and Cu
2+
, but not Mg
2+
...................................................... 61
Figure 3.4. Orb2A87 Mutants H29A, H46A, and H60Y Still Bind to Ni
2+
, but Only with Half a Binding
Site per Monomer ...................................................................................................... 63
Figure 3.5. Mg
2+
and Ca
2+
Do Not Change the Structure of Orb2A87 ........................................... 64
Figure 3.6. Addition of Ni
2+
Induces Rapid Increase of Thioflavin T (ThT) Fluorescence .............. 65
Figure 3.7. Addition of Ni
2+
Induces Aggregation Distinct from Fibril Formation ......................... 66
Figure 3.8. F5Y can Form Amyloid Fibrils ...................................................................................... 71
ix
Figure 3.9. F5Y fibrils are Morphologically Similar to Orb2A88 Fibrils ......................................... 71
Figure 3.10. ThT Kinetic Assay Highlights Similarities in Orb2A88 and F5Y Aggregation .............. 73
Figure A1. Nine Residues of a-synuclein can be O-GlcNAcylated ................................................ 81
Figure A2. 1D
13
C Spectra and 2D
13
C-
13
C Correlation Spectra of gS87 and WT a-synuclein Show
That They Have Different Structures and Dynamic Properties ................................... 84
Figure A3. gS87 a-synuclein Fibrils are Structurally Similar to a-synuclein Ribbons Described
Previously .................................................................................................................... 87
Figure A4. Unseeded a-synuclein Fibrils are Similar in Structure to a-synuclein Fibrils Described
Previously .................................................................................................................... 88
Figure A5. gS87 a-synuclein Fibrils are Structurally Distinct From Other a-synuclein
Fibril Types .................................................................................................................. 90
Figure A6. Unmodified a-synuclein Fibrils are Different in Structure to Other Fibril Types ........ 91
Figure A7. a-synuclein S87N Fibrils are Structurally Different from WT a-synuclein Fibrils ........ 93
Figure A8. a-synuclein TN Fibrils are Structurally Different From WT a-synuclein Fibrils ........... 94
Figure A9. a-synuclein A53T S87N (TN) Fibrils are More Dynamic than S87N Fibrils .................. 96
Figure A10. a-synuclein A53T S87N (TN) Fibrils are Structurally Distinct From gS87 Fibrils ........ 97
Figure A11. S87N a-synuclein Fibrils are Different in Structure to Other Fibril Types ................. 99
Figure A12. TN a-synuclein Fibrils are Different in Structure to Other Fibril Types ................... 100
Figure A13. BMRB Entry 18207 for a-synuclein A53T is a Good Match to TN Fibrils ................. 101
Figure B1. Domain Organization and Primary Structure of the HETs and a-synuclein Chimer (HET-
sCT) ............................................................................................................................ 105
Figure B2. HET-sCT Expresses at Temperatures Used for HETs and a-synuclein ....................... 107
Figure B3. HET-sCT is Not in Inclusion Bodies ............................................................................. 109
Figure B4. HET-sCT Purifies Well Using Soluble Purification Conditions .................................... 110
Figure B5. HET-sCT Purifies Well Using Denaturing Conditions .................................................. 110
Figure B6. DnaJB1 Shows Preference for One HET-sCT Chimer Sample ..................................... 112
Figure B7. DnaJB1 Shows Strong Affinity for HET-sCT Chimer ................................................... 113
Figure B8. DnaJB1 Binding Improves After Longer Fibril Incubation .......................................... 114
Figure B9.
13
C
15
N HET-sCT is Bound by DnaJB1 .......................................................................... 116
Figure B10. DnaJB1 Contributes to HET-sCT Fibril Breakage on EM Grids ................................. 117
Figure B11. The Static Amyloid Core of HET-sCT is Unaffected by Addition of DnaJB1 .............. 118

x
Abstract
Amyloid fibrils are proteinaceous deposits that are insoluble, resistant to degradation and
are rich in b-sheet structure. Amyloid fibrils are composed of b-strands that run perpendicular to
the fibril axis, thereby forming a cross-b sheet of potentially indefinite length. Although this cross-
b motif is the fundamental structure of amyloid fibrils, amyloid fibril structures are elaborate and
diverse, and can give rise to an abundance of functions. In some cases, these functions are
deleterious in nature, like in the case of neurodegenerative disorders, which are characterized
by the accumulation of misfolded protein that aggregates into amyloid fibrils.
In this work, we investigate two such amyloid proteins, Huntingtin, which causes
Huntington’s Disease and a-synuclein, which is associated with Parkinson’s disease. For
Huntingtin, we investigate its mechanism of aggregation by characterizing the structure of its
monomeric state using solid state NMR. In addition, we provide a better understanding of its
mechanism of disaggregation and overall toxicity on the cell, by characterizing the interaction of
Huntingtin fibrils with the mammalian chaperone DnaJB1. Chaperones are known promoters of
protein homeostasis and decrease amyloid protein toxicity through their work in identifying,
folding and re-folding misfolded protein. Using various biophysical tools, including solid state
NMR we identify the binding site of DnaJB1 on Huntingtin exon 1 fibrils. For a-synuclein, there
are known post translational modifications and mutations that influence the proteins’
aggregation and toxicity. In this work, we characterize the effect of glycosylation on Serine 87
(gS87), and the effect of S87N and S87N A53T (TN) mutations on the structure of these fibrils.
While in the case of neurodegenerative disorders and other amyloid diseases, amyloid
formation results from a deleterious misfolding event that affects the protein and causes it to
xi
further misfold and aggregate and ultimately become toxic to cells. In contrast, in the case of
functional amyloids, their formation is a tightly regulated process. In this context, amyloid
formation serves a purpose, is beneficial to cells, and is ultimately necessary for host survival. In
this work we characterize the functional amyloid Orb2, a member of the Cytoplasmic
Polyadenylation Element Binding protein family which is associated with memory formation and
maintenance. Specifically, we investigate the first 88 amino acids of the isoform Orb2A (Orb2A88)
and characterize its metal binding properties and the effect these have on its aggregation, as well
as the effect of a point mutation (F5Y) on the fibrils formed by Orb2A88.

1
Chapter 1: Introduction to Amyloid Fibrils
1.1 Amyloid Fibrils

Amyloid fibrils are proteinaceous deposits that are insoluble and resistant to degradation.
Traditionally amyloid proteins have been shown to be reactive to Congo Red dye (CR) and show
an apple green birefringence when observed using light microscopy and polarized light [1]. As
observed using electron microscopy (EM), amyloids are fibrillar in morphology. These stable
structures are the result of peptide or protein copies assembling on to themselves, thereby
creating a repetitive ordered arrangement with unchanging spacing (figure 1.1). At a molecular
level, their structure is composed of b-strands that run perpendicular to the fibril axis, thereby
forming a cross-b sheet of potentially indefinite length [2]. b-sheets can be arranged in a parallel
or anti-parallel conformation. In addition, b-sheets are repetitively spaced and give rise to a
typical fiber diffraction pattern (figure 1.1). These b-sheet structures usually form the amyloid
core of fibrils, and amyloid fibrils themselves are usually bundled protofilaments.
2
Although this cross-b motif is the fundamental structure of amyloid fibrils, the structures
formed by these proteins are elaborate and diverse. Via the arrangement of their individual
sheets, kinks, and loops, amyloid proteins can assemble into different types of fibrils with distinct
morphologies, even within the same amyloid protein or polypeptide [7] (figure 1.2). Today, the
molecular structure of many of these amyloids have been resolved. These high-resolution
structures have not only provided great insight into the overall organization and diversity of some
of these amyloid proteins, but in addition, have further highlighted the intricacy of the b-sheet
conformation.
3
Amyloid proteins grow through the recruitment of corresponding soluble protein, and
hence, this self-propagating nature makes them analogous to prion proteins. This self-
perpetuating nature means that in certain conditions, amyloid fragments can function as seeds
for the growth of fibrils. This templating behavior is a quality which is commonly used in in-vitro
studies to accelerate the aggregation of amyloid proteins as well as to generate homologous fibril
samples for structural characterization.
Among protein folds, the cross-b structure is unique and can give rise to an abundance of
functions [2]. In some cases, these functions are deleterious in nature, like in the case of disease
associated fibrils. While in other circumstances amyloid formation serves a purpose and is
beneficial to cells and host survival, such is the case of functional amyloids. In the following, we
discuss the role of fibrils in the context of both.
1.2 Amyloid Fibrils in Disease

Amyloid fibrils play a role in a variety of diseases. In this context, amyloid formation is the
result of a misfolding event that occurs early in the protein, and which leads to further misfolding
and aggregation of an otherwise soluble protein. Because this process of misfolding and
aggregation is one that the protein would normally not undergo, it contributes to disease via two
mechanisms: 1. Amyloid formation leads to an inability of the protein to perform its native
function, this is referred to as “loss of function” and 2. Amyloid formation and amyloid deposit
affects, and interferes, with normal cellular metabolism, this is called “toxic gain of function” [3].
Since the process of misfolding and aggregation is central to the formation of amyloid fibrils,
studies have therefore, focused on providing a better understanding of the mechanism of
4
misfolding and on characterizing the various conformers that participate in the aggregation
pathway of these proteins (figure 1.3).

Although amyloid formation is the hallmark of many diseases collectively called
amyloidosis [4], the proteins that participate in each of the diseases, and the aggregates formed
by each are unique to each disorder. Perhaps the most prominent diseases associated with
amyloid fibrils are neurodegenerative disorders. These progressive brain disorders include
disease such as Alzheimer’s disease (AD), which is caused by the accumulation of aggregated
amyloid b (Ab) in the form of plaques and tau in the form of tau tangles [5]. Huntington’s disease
(HD) on the other hand is caused by the misfolding and aggregation of the protein Huntingtin
(HTT), while Parkinson’s disease (PD) is linked to the deposition of ⍺-synuclein in Lewy bodies.
The latter of these two diseases, HD and PD, are discussed in this thesis. In it we
characterize their structure and association with chaperones.
5
1.3 Functional Amyloids

Although amyloid fibrils are associated with the pathology of many diseases, amyloids
also function in a variety of biological processes. In contrast to disease amyloids, which are the
result of an aberrant event, the formation of functional amyloids is a tightly regulated process.
The importance of amyloids in a functional context is becoming more apparent, as more and
more studies have highlighted their role in chemical storage, structure, signaling and storage of
information, among others [6]-[7]. Peptide hormones, for example, function in hormone
signaling and have been shown to be stored as amyloid-like cross b-sheet rich structures [8]. The
amyloid protein Curli is a component of the extracellular matrix of e. coli and functions in cell
adhesion, invasion, and biofilm formation [9]. The family of neuronal Cytoplasmic
Polyadenylation Element Binding proteins (CPEB) has been shown to be essential in the
formation and maintenance of long-term memory across species [10].
In this thesis we investigate the latter of these functional amyloids CPEBs. Namely, we
characterize the aggregates formed by the CPEB protein Orb2A.
1.4 Techniques Used in This Study

We characterized the morphology of fibrils in our study using Transmission Electron
Microscopy (TEM). We measured the aggregation of proteins using the amyloid binding dye
Thioflavin T (ThT). ThT is a fluorescent marker that can bind to amyloids by intercalating with the
b sheet strands of these fibrils, and so as the amyloid protein continues to propagate, the ThT
signal increases [11].

6

Although at first glance, amyloid proteins appear to be simplistic in nature, having a
repetitive structure, it is this same characteristic that makes them difficult to characterize with
conventional protein techniques. Therefore, to study the structure and interaction of amyloid
fibrils in this study, we used solid state Nuclear Magnetic Resonance (ssNMR). NMR takes
advantage of the magnetic properties of the nuclei and using different pulse programs, the
magnetization generated can be transferred in varying ways between them. Utilizing this
method, we can extract the secondary structure of the protein from its chemical shift
information. For our studies, we used uniformly isotope labeled
13
C
15
N proteins. The Insensitive
Nuclei Enhancement by Polarization Transfer (INEPT), INEPT Hetcor, INEPT TOBSY (through bond
correlation spectra), and Heteronuclear Single Quantum Coherence (HSQC) are experiments
which transfer magnetization through one bond J couplings and are experiments which we use
here to characterize the dynamic domains of the proteins. The cross polarization (CP)
experiments and its 2D versions CP-PARIS (Phase Alternated Recoupling Irradiation Schemes),
DREAM, and DARR (dipolar assisted rotational resonance) work through magnetization transfer
7
via dipolar coupling, what is commonly referred to as “thorough space”. These experiments are
therefore used to characterize the static components of the protein.

8
Chapter 2: Structural Studies on Huntingtin Exon 1 Monomer and
Huntingtin Exon 1 Fibrils with the Co-Chaperone DnaJB1
2.1 Introduction

HD is a fatal neurodegenerative disorder caused by a mutation in exon 1 of the Huntingtin
(htt) gene. The mutation specifically affects a CAG trinucleotide repeat region that, when
translated, results in an over expanded polyglutamine (polyQ) tract in the corresponding protein
huntingtin (HTT). Aberrant splicing and proteolysis lead to the generation of HTT fragments of
which HTT exon 1 (HTTex1) is the most prominent [12]-[13]. HTTex1 is composed of three main
domains: the N17 which is composed of the N-terminal 17 amino acids, the polyglutamine region
(polyQ), which forms the core of the fibrils, and the C-terminal proline (Pro) rich domain (PRD)
(figure 2.1). In addition to containing the pathological polyQ expansion, the importance of the
HTTex1 fragment is further highlighted by the fact that this fragment is sufficient to form fibrils
in vitro and cause disease in animal models. In HD, a polyQ length beyond 36Q is associated with
protein misfolding, aggregation and ultimately toxicity to cells [14]. In addition, the length of the
polyQ expansion is inversely related to the age of onset for the disease [15].

9

It is widely acknowledged that many HTTex1 conformers play a significant role in the
aggregation pathway and contribute to the cellular dysregulation and overall toxicity caused by
HTTex1. Of the many conformers comprising the aggregation pathway of HTTex1, the focus of
this study is on HTTex1 monomers and fibrils.
Monomers represent the beginning of the fibril formation pathway in HD and in other
amyloid diseases (figure 1.3). There is mounting evidence highlighting the toxicity of these and
other prefibrillar conformers [16]-[17]. Additional work also suggests that prefibrillar conformers
may be more toxic than fibrils themselves, and that conformational changes occurring prior to
fibril formation are the source of toxicity in amyloid proteins [18]. Therefore, characterizing the
effects of the polyQ expansion on the structural features of monomers is critical to understanding
how this expansion results in disease. More specifically, information on the molecular
organization, dynamic properties, and structural changes these conformers undergo during the
fibril formation process is essential to our understanding of HD pathology. Although there have
10
been advances in the characterization of these prefibrillar conformers, structural information
about their organization, especially of those with pathologic polyQ lengths, is still largely absent
[19]. Limiting such investigation is the fact that the monomeric state is short lived, explaining
why, until recently, monomers had not been identified as sources of toxicity [20]. This ephemeral
nature makes isolating and characterizing monomers quite difficult. To study these conformers,
we used a technique in which we bound monomeric HTTex1 protein to Nickel NTA (Ni-NTA) and
Copper NTA (Cu-NTA) agarose beads, thereby preventing further aggregation and allowing for
prolonged spectroscopic measurements (figure 2.2).

By essentially “trapping” the protein on the beads we aimed to investigate the structural
features of monomers for both wild type and mutant polyQ lengths. Such experiments could
allow for a better understanding of what differentiates disease from non-disease associated
conformers and add to our comprehension of the process of misfolding, aggregation and,
ultimately, toxicity in HD.
11
We were particularly interested in characterizing the C-terminus of HTTex1. The C-
terminus directly follows the polyQ domain and is composed of poly-proline and proline rich
sequences. Research on HTTex1 fibrils has revealed that this region is much more structurally
and dynamically complex than previously thought [21]. Rather than having a single secondary
structure, Pro residues are organized into PPII helical (ProA) or random coil (ProB) conformations
depending on their proximity to the polyQ domain. This relationship between the polyQ domain
and the C-terminus is further supported by previous EPR studies, which show that residues
flanking the polyQ region experience different dynamics depending on their proximity to the
polyQ region [22].

This data suggests a close association between the polyQ domain and the C-
terminus, one that makes the C-terminus highly sensitive and susceptible to influence by
structural changes in the polyQ domain. Previously, the polyQ-polyproline relationship was
examined by investigating the ability of Pro rich sequences to destabilize and inhibit the
aggregation of extended polyQ tracts [23].

The results revealed that such sequences are capable
of inhibiting the aggregation of polyQ tracts of up to 55Q as well as reduce aggregate toxicity.
This suggests that perhaps the normal function of poly-proline residues in HTTex1 is somehow
altered by the polyQ expansion, rendering it unable to inhibit the aggregation of the polyQ
domain. Taken together these findings point to the importance of characterizing the C-terminus
of HTTex1 monomers and investigating the role this domain plays in the aggregation process of
HTT.
Although more studies are now focusing on investigating pre-fibrillar aggregates, amyloid
fibrils continue to be the hallmark of HD [24]. This is because HD is characterized by an
internuclear accumulation of HTTex1 fibrils, the end product of the misfolding and aggregation
12
pathway (figure 1.3). In addition, like smaller HTT conformers, fibrils have been shown to induce
cellular toxicity and have been the focus of a variety of studies [25]-[26].

Characterizing the
structural features of amyloid fibrils and the regions in them that facilitate their interaction with
other cellular components is, therefore, critical to understanding the mechanism underlying HD.
Due to the efforts of many, we now have a better understanding of the overall
organization of the various HTTex1 domains in mature fibril. Such studies have highlighted the
fact that, when aggregated, the PRD of HTTex1 protrudes out and away from the fibril core and
is the most dynamic domain of the three [22]-[21], [27]-[28] (figure 2.1). In addition, recent
studies have further validated the importance of the PRD by showing that this domain can
become more or less bundled depending on the temperature used during fibril formation and
that the extent of bundling is directly related to the toxicity [29]. And so, although the polyQ
region comprises the core of the amyloid fibrils, it is the PRD that holds the key to understanding
how HTTex1 fibrils associate with other cellular components and ultimately how they contribute
to cytotoxicity.
Molecular chaperones are cellular components which function to promote efficient
protein folding and prevent protein aggregation, including HTTex1 misfolding [30]-[33]. Due to
their potency in influencing HTT aggregate formation, chaperones provide a possible target for
therapeutic development towards HD and have therefore been the focus of extensive studies.
Such work has revealed that, in addition to modulating HTTex1 aggregation, chaperones are able
to ameliorate the toxic effects of HTTex1 on cells [30]-[32]. These findings highlight the
importance of studying the chaperone-fibril interaction, because such investigations will allow
13
for a better understanding of how HTTex1 fibrils associate with cellular components, and in
addition, generate new hypotheses about what fibril regions mediate toxicity in the cell.
Although various chaperones have the ability to influence the aggregation of HTTex1 and
its effect on the cell, thus far, only the Hsc70, DnaJB1, and Apg2 chaperone complex has been
shown to be effective at both completely suppressing HTTex1 aggregation and disaggregating
formed fibrils in the presence of ATP [34]-[35]. While all members of the trimeric complex play a
vital role in maintaining protein homeostasis, DnaJB1 is the subunit which is essential for
initiating this process. DnaJB1, which receives its name from its conserved J domain, is a member
of the Hsp40 co-chaperone family [34] and is, therefore, the subunit responsible for substrate
identification and binding. In addition to its J domain, DnaJB1 also contains a C-terminal domain
(CTD), which it uses to interact with substrate, and which contains a conserved nine amino acid
sequence, located in the linker region between CT I and CT II, which is used to bind HTTex1 (figure
2.3). Although studies have shown that DnaJB1 binds the PRD of soluble HTTex1 [35], the DnaJB1-
HTTex1 fibril interface has not been identified yet. For this reason and because DnaJB1 has a
much higher affinity to fibrils compared to the soluble form of the respective protein [35]-[36]
we wanted to directly study the DnaJB1-HTTex1 fibril complex.
Specifically, we investigated the polymorph-specificity and binding interface of HTTex1
fibrils and DnaJB1 using sedimentation assays, EM, and solid-state NMR.
14

2.2 Characterizing the Structure of HTTex1 Monomers on Agarose Beads

2.2.1 Results

A"achment of HTTex1 via its C-terminal Polyhis:dine-Tag Leads to Reduced Mobility of the
PRD

We conducted our initial monomer studies on HTTex1 with a glutamine length of 16 and
one with a pathogenic length of 46. Both constructs contained a His tag at their C-terminus
following the PRD domain (figure 2.4). The agarose beads were prepared by washing them with
deionized water and then with buffer. After purification and cleavage of the Thioredoxin (TRX)
fusion tag, the protein was incubated with the agarose beads. The sample was then packed into
one of our 4mm rotors and measured. We prepared and measured a sample of agarose beads
alone for the purpose of comparison to the sample.

We collected 1D
13
C data on our samples using CP, INEPT and direct excitation (DE)
experiments. INEPT allows for the detection of dynamic regions, whereas CP, which relies on the
dipolar interaction between nuclei, exclusively detects static ones.

DE detects both static and
dynamic regions. Overall, these samples showed little signal to noise. For the Q16 sample,
specifically, in the CP spectrum we observed some signal for peaks coming from our Gln region,
15
however, the majority of the peaks detected corresponded to the agarose beads (figure 2.5). In
the INEPT spectrum we detected signal coming from the Gln residues in our protein. On closer
inspection, we observed that the dynamics and structural organization of this Q16 closely
resembled that of Q46 fibril samples we’ve measured previously (figure 2.6). Unfortunately, the
Q46 sample yielded very little signal and we were not able to obtain substantial information from
that sample.

16

HTTex1 Q25 Monomer is Characterized by Structural Heterogeneity

We wondered whether we could improve the signal to noise and overall dynamics of the
samples by attaching it to the agarose beads via a different domain. In these early constructs, the
protein was anchored via its C-terminus, which limited the mobility of this region overall. In
addition, anchoring via this domain left the N-terminus free which limited the amount of protein
we could attach to the beads. In HTTex1 aggregation, it is the N-terminus that comes together
and facilitates the aggregation of the protein, adding more protein to the beads could lead to
aggregation on them while conducting measurements. To circumvent these problems, we
designed and used two different constructs. One of the constructs contained the His tag in
between HTTex1 and the thioredoxin tag, so that after EkMax digestion, the construct would
retain the His tag and attach via the N-terminus to the beads (figure 2.7 A). We also utilized a

17

Q46 construct with a His tag at the N-terminus preceding the TRX tag (figure 2.7 B). We purified
and used this construct undigested, as digestion would mean loss of the 6X His tag.

We were successful at expressing the recombinant, uniformly labeled
13
C
15
N HTTex1 Q25
protein (figure 2.7 A). After Nickel column (His 60) and ion exchange chromatography
purification, we attached the protein to Ni-NTA resin via its 6X-His tag. One-dimensional DE,
INEPT, and CP experiments were undertaken to get an overview of the dynamics in this construct.
All three experiments show good signal to noise, and we were able to obtain 2D measurements
of the sample. In figure 2.8 A, we show the results for the 1D experiments conducted on this
sample. Similar to the earlier sample, we observed Pro peaks corresponding to the Pro residues
in both the CP and INEPT spectra, but this time we also detected the polyQ region of the protein.
The
13
C
1
H INEPT-HETCOR, like the INEPT 1D, is sensitive to dynamic parts of the protein. This
spectrum was primarily dominated by Pro residues (figure 2.8B). In contrast, the 2D DARR,
sensitive to static regions, had very few cross peaks (figure 2.8C), which indicates that although
18
the residues detected in this spectrum are more static compared to those detected in the INEPT-
HETCOR, they are still relatively dynamic. Interestingly, the fact that we see Gln peaks in both
dynamic and static selective spectra shows that this sample is relatively heterogenous. In these
initial attempts we were unable to obtain data with sufficient signal to noise for the Q46
construct (figure 2.7B).
19

20
2.2.2 Discussion

With these studies, we showed that it is possible to characterize soluble, monomeric
HTTex1 protein using solid state NMR. Using this technique, we were able to show that HTTex1
monomers are heterogenous in structure and mobility. We detected the C-terminus of the
protein in both the static selective and the dynamic selective spectra. Similarly, glutamine regions
were detected in both the static selective and dynamic selective data we collected. The fact that
we can detect these protein signals in both types of experiments tells us that there are Gln and
Pro residues that are quite stable, while others remain very mobile. This finding can be explained
via two possibilities, the first is that the structuring of Pro and Gln residues in the monomer is
partly a result of its transient interaction with the agarose beads. The second of these possibilities
is that there is some structuring of the polyQ domain even in non-pathogenic HTTex1 constructs.
The second of these possibilities is further supported by other structural studies, which
have shown that there is an overall similarity in the general structural features of non-pathogenic
and pathogenic Q length monomers. Specifically, these studies showed that there is a gradual
compaction of the monomer as the polyQ increases, the N17 and polyQ regions adopt a tadpole
like architecture defined by a globular head, while the C-terminus composes the tail [37],[92].
The NMR studies on HTTex1 Q25, specifically, also showed that, although there is a lack of
persistent secondary structure, the N17 and polyQ regions exhibited unusual rigidity [92].
Residues closer to the N17 experience greater organization that those further away. The
observed rigidity in the polyQ region for this construct is in line with our own observations, which
showed that there is immobilization of some of the Gln residues. The structural heterogeneity in
the HTTex1 monomer was also highlighted by studies using Electron Paramagnetic Resonance
21
and solution NMR [93]-[94]. These studies detected a mobility gradient as well, with residues
closer to the N17 showing greater rigidity as compared to those further away. Overall, the
heterogeneity in the monomeric structure indicates there are similarities in the behavior of the
protein, at this early stage, regardless of the polyQ length. It is, therefore, not surprising that we
would see these similarities translated in our structural studies as well. However, to further
disregard the influence of the agarose beads on the structuring of our constructs, samples with
longer linkers could be utilized, thereby furthering the HTTex1 monomer from the influence of
the beads.
Although we did not proceed with measurements, these first tests are a proof of concept
in that it shows that the technique is feasible. Further investigations in the area would require
improvement in signal to noise. For this the protein could be deuterated and attached to copper
beads. Unlike Ni, Cu is paramagnetic and induces faster
1
H R1 relaxation times allowing for
shorter recycle delay and increasing the number of scans obtained in a given period of time.
1
H
decoupling in deuterated proteins requires much less radio frequency power, shortening the
duty cycle of the probe. Overall, these modifications will result in improved spectral quality,
sharper lines, and overall increase in signal to noise.

2.3 Characterizing the Binding Interface of HTTex1 Fibrils and DnaJB1

2.3.1 Results

DnaJB1 Preferen:ally Binds to RT Fibrils

To study the interaction of DnaJB1 with HTTex1 fibrils, we produced HTTex1 fibrils with a
polyQ length of 46. Because our previous work highlighted the effect of aggregation temperature
22
on the dynamics and toxicity of the resulting fibrils as well as their ability to bind to fibril-specific
antibodies [29], we fibrilized HTTex1 at three different temperatures, 4 °C, 22 °C, and 37 °C.
Fibrils produced at 4°C (in the following called T fibrils) are predominantly unbundled. Fibrils
created at 22 °C (RT fibrils) and 37 °C (N fibrils) tend to be longer and more bundled (Figure 2.9
A). T fibrils are more toxic, especially as compared to N fibrils [29], [38]. We tested the interaction
of the three fibril types with DnaJB1 using a sedimentation assay. The assay consisted of mixing
DnaB1 with preformed HTTex1 fibrils and then incubating the protein at room temperature. After
the incubation period, the protein mixture was centrifuged to separate the soluble, unbound
chaperone from the insoluble portion containing the HTTex1 fibrils with bound chaperone (Figure
2.9 B). The soluble portion was then analyzed using SDS PAGE. To ensure that the assay gave a
clear-cut answer regarding whether there was an interaction between the HTTex1 fibrils in
question and DnaJB1, all sedimentation experiments were run using a 4:1 molar HTTex1:DnaB1
ratio. Thereby guaranteeing that the fibrils would pull down all the chaperone from solution.
Although DnaJB1 interacts with all three fibril types, we observed a preference for RT-fibrils, as
indicated by the intensity difference in bands observed in the SDS PAGE (Figure 2.9 C). In the case
of T and N fibrils, a substantial amount of DnaJB1 remains in the supernatant post incubation
with fibrils, as indicated by the bands remaining at 40 kDa in the gel. In the case of RT fibrils, no
DnaJB1 remains in the supernatant fraction, indicating that all available chaperone is bound to
the HTTex1 fibrils.
23

DnaJB1 Interac:on Makes HTTex1 Fibrils Bri"le

The sedimentation assay demonstrated that the DnaJB1 interaction with HTTex1Q46
fibrils is polymorph-specific. We next asked if DnaJB1, in turn, affects the morphology of HTTex1
fibrils. To answer this question, we analyzed all three HTTex1 polymorphs before and after
incubation with DnaJB1 at a 4:1 ratio (HTTex1:DnaJB1) using negative stain EM. The EM images
in figure 2.10 show that DnaJB1 interaction led to an unbundling of the fibrils in particular for the
N and RT fibrils that showed considerable lateral association of fibrils in the absence of DnaJB1,
but numerous unbundled fibrils after the addition of DnaJB1. To further probe for changes in
morphology, we measured the diameter of all polymorphs. Before the addition of DnaJB1 T, RT,
and N-fibrils had diameters of 13.5±2.8 nm, 13.1±2.2 nm, and 12.4±2.3 nm, respectively. This
slight decrease in diameter is significant (T to N p<0.001). After the addition of DnaJB1 all fibril
24
polymorphs got narrower with diameters for T, RT, and N-fibrils of 10.6±2.4 nm, 9.7±2.2 nm, and
9.2±2.1 nm, respectively (figure 2.11).

Interestingly, the addition of DnaJB1 not only led to the narrowing of HTTex1 fibrils but
also to fragmentation on the EM grid as can be seen by the zoomed in images in figure 2.10. This
fragmentation was observed for all three polymorphs but was most prominent in the RT-fibrils
25
that showed the strongest binding in our sedimentation assay. The fibril fragments observed line
up in a row suggesting that the fibril might have fragmented on the EM grid itself. In addition,
the fragments have a relatively consistent length of 34.5±11.5 nm (figure 2.11).

Interestingly, at an equal HTTex1:DnaJB1 ratio these fragments are not visible anymore
but the otherwise relatively smooth HTTex1 fibrils become more rugged in appearance indicative
of increased DnaJB1 interaction. The EM images in figure 2.12 illustrate this on the very samples
used for our NMR measurements described below. Where the fibrils in the absence of DnaJB1
26
are featureless i.e. have no twist and relatively even contrast, the fibrils to which DnaJB1 was
added at a 1:1 ratio show additional densities bulging out perpendicular to the fibril axis. To test
if this change in fibril appearance was an effect caused by the slightly different protocol used to
make isotope labeled HTTex1 fibrils, we repeated the experiment. This time we incubated
unlabeled HTTex1 fibrils with DnaB1 at different molar ratios and saw the same effect: HTTex1
fibrils exhibited fragmentation on the EM grid at a HTTex1:DnaJB1 ratio of 4:1 but not at a 1:1
ratio where the fibrils rather appeared covered with DnaJB1 resulting in additional densities at
the outside of the fibrils (figure 2.13).

27

DnaJB1 Binds the C-terminal PRD of HTTex1 Fibrils

Why is the binding of DnaJB1 to HTTex1 fibrils so polymorph specific? To get insight into
this question, we determined the DnaJB1 binding site of HTTex1 using solid-state NMR
spectroscopy. For these experiments, we made uniformly
13
C-
15
N labeled HTTex1 RT fibrils and
28
first tested the DnaJB1 binding efficiency using the sedimentation assay shown in Figure 2.9. We
then incubated these fibrils with unlabeled (natural abundance) DnaJB1 at a 1:1 ratio. HTTex1
fibrils alone in buffer served as a control. EM images of our NMR samples are shown in figure
2.10.
Because NMR chemical shifts and intensities are sensitive to changes induced by
molecular interactions, the comparison of NMR spectra of the bound and unbound fibrils should
indicate the binding site. The static fibril core HTTex1 can be spectroscopically separated from
the more dynamic C-terminus using NMR experiments that rely on non-motionally averaged
dipolar couplings [21], [39]. figure 2.14 shows overlays of such 1D
13
C CP spectra and 2D
13
C-
13
C
PARIS spectra where the unbound fibrils are shown in black and the bound fibrils in purple. These
spectra are dominated by signals coming from the polyQ domain featuring two types of Gln (Gln
A, and Gln B) but also show signals coming from the less dynamic regions of the PRD (Pro A and
few weak Pro B peaks) [21], [40]. Overall, the almost perfect overlap, both in terms of signal
intensity as seen from the 1D spectra, and chemical shift as seen from the 2D PARIS spectra,
indicates that DnaJB1 binding leaves the fibril core of HTTex1 fibrils unaffected.

29

Using a different set of NMR experiments, we are able to selectively detect the dynamic
domains that often surround the cross-β core. These experiments are based on J-couplings,
which under the moderate MAS frequencies used in this study only result in signals from highly
mobile regions of the protein [39], [28]. Figure 2.15 shows the result of two of these experiments
namely a 2D
13
C-
13
C INEPT-TOBSY and a 2D
1
H-
13
C INEPT-HETCOR. Although the majority of the
signals in these spectra still overlap quite well, there are some specific differences between the
spectrum recorded in the absence (black) and presence (purple) of DnaJB1. Direct chemical shift
30
perturbation i.e. a shift in cross-peak location could be observed for Ala and Glu. In addition,
signals of the only Gly (G102) and of Pro in the cis conformation increased in intensity, whereas
signals for His, Arg, and Val decreased with the addition of DnaJB1. Other residue types such as
Leu, Met, Gln, and the majority of the Pro signal, were unaffected by DnaJB1 binding.

All these changes are compatible with DnaJB1 binding the dynamic PRD of HTTex1.
Specifically, the last 12 residues of this domain, termed C12, which include the only Gly (G102)
and Arg (R112) of HTTex1. To further confirm this result, we recorded a
1
H-
15
N HSQC experiment
31
that predominantly contains signals from this region [27]. Figure 2.16 A illustrates the relatively
significant changes between the HSQC spectra of unbound fibrils and of fibrils in the presence of
DnaJB1. There is again a noticeable increase in Gly signal intensity and decrease in the (original)
R112 signal. In addition, we see the signal of L110 disappear and several new signals appear in
the spectrum of fibrils with DnaJB1. To assign these new, additional signals we recorded a 3D
HNCA spectrum. However, most of the signals were too weak to be detected in this spectrum
with a few exceptions notably the new Ala resonance which could uniquely be assigned to A104
due to the fact that it is the only Ala in the sequence that is preceded by a Pro residue (see Figure
2.16 B). Figure 2.16 C illustrates the results from our NMR measurements with no visible changes
seen for residues unique to the N17 and polyQ regions and the majority of changes observed for
resonances compatible with a location in the C-terminal C12 region.
32

DnaJB1 Affects the Dynamics of the PRD of HTTex1 T Fibrils

Our sedimentation assay in Figure 2.9, showed that DnaJB1 has a preference for RT fibrils,
as evident by the fact that it is completely pulled down by these fibrils. However, the assay also
shows that DnaJB1 binds T fibrils and N fibrils, we can see this by the fact that there is a decrease
in the intensity of the bands for these samples as well. Since T fibrils are known to be more toxic
than RT or N fibrils, we wondered what effect DnaJB1 would have on the structure of these
33
fibrils? Using uniformly labeled
1
H
13
C HTTex1, we made T fibrils and incubated them with DnaJB1.
Like with RT fibrils we made samples of both fibrils alone and fibrils with DnaJB1. We obtained
1D CP, DE and INEPT data on the samples to gage the difference in dynamics. We saw a sharp
increase in the intensity of peaks belonging to Pro residues in our CP spectra upon addition of
DnaJB1 (figure 2.17 A). Because the cross-polarization experiment is sensitive to static regions of
the protein, this tells us that Pro residues are becoming immobilized upon addition of the
chaperone (figure 2.17 B). Additionally, we see that T fibrils precipitate in the sample tube upon
addition of DnaJB1 (figure 2.17 C). This points to a potential bundling of the fibrils by DnaJB1.
These changes upon addition on DnaJB1 to the fibrils were reproduced with other samples.

34
To probe whether the observation that DnaJB1 leads to immobilization of Pro residues is
the result of a change in dynamics rather than just an artifact, we ran a temperature screen on
the samples. We obtained 1D
13
C CP, DE, and INEPT data for both samples at 0 ºC, -10 ºC, -20 ºC,
and -30 ºC. At -20 ºC, all dynamics are frozen out as confirmed by the lack of signal in our INEPT
experiments. In figure 2.18, we see that at this temperature, the spectrum of HTTex1 with DnaJB1
and HTTex1 fibrils alone perfectly overlap. This confirms our finding and shows that the
chaperone is interacting with Pro residues. Because Pro residues are only found in the C-terminus
of HTTex1, this tells us that DnaJB1 is interacting with the C-terminus of T fibrils. The fact that
we do not see a change in the Gln residues, tells us that the fibril core is unaffected by DnaJB1.
Overall, we showed that similar to RT fibrils, DnaJB1 interacts with T fibrils via their C-
terminus. In contrast to RT fibrils, we did not observe chemical shift perturbations.
35

DnaJB1 Binding to RT Fibrils, but not to T Fibrils, Leads to the Appearance of Addi:onal Low
Molecular Weight Bands

For our studies we used the sedimentation assay shown in figure 2.9 B to verify the
interaction of DnaJB1 with our fibril samples. Interestingly we observed that binding of DnaJB1
to RT samples produced low molecular bands, which we did not see in the T fibril samples (figure
2.19). The bands observed were approximately at 25 and 12 kDa. For the majority of our
experiments, we used chaperone in hepes solution, but for a few of our assays, we used
chaperone in TFA dH 2O solution and chaperone in phosphate buffer. The activity and binding
capability of our chaperone, in these alternative conditions, was tested by our collaborator. The
chaperone was found to be functional and showed the same binding activity as chaperone in
36
hepes. Using these additional conditions, we consistently found that although the band at 25 kDa
was present in the sedimentation assays carried out using DnaJB1 in hepes and in phosphate, it
was not present in samples in the TFA dH 2O condition (figure 2.20). The band at 12 kDa was also
less pronounced in these samples.

37

We investigated the identity of the low molecular bands via antibody detection and the
use of Western blots. For our initial Western blots, we included samples of chaperone alone,
HTTex1 alone, Ulp1 (the protease used to cleave the SUMO tag off DnaJB1), and samples from
our sedimentation assays. We tested the membranes using a DnaJB1 specific antibody and a 6X
His antibody, because HTTex1 contains a His tag and could possibly be one of the fragments. We
show the results for one of these trials in figure 2.21. Using the DnaJB1 antibody, we primarily
detected bands at 40 kDa and 80 kDa in the chaperone samples, which coincide with a DnaJB1
monomer and dimer. To our surprise, we did not detect the 25 kDa band, but we did detect the
12 kDa one in the HTTex1-DnaJB1 sedimentation assay sample. We did not detect either of the
bands in question using the 6X His antibody. To further clarify their identity, we isolated each of
the bands and sent them for mass spectrometry analysis. Unfortunately, the results we obtained
were inconclusive and could not specifically link either of the bans to HTTex1 nor DnaJB1.
38

We followed our investigation into the identity of the low molecular bands with an
antibody specific to the C-terminus of DnaJB1. For this analysis we included two DnaJB1 samples
(DnaJB1 in hepes and in phosphate buffer), as well as one HTTex1-DnaJB1 sedimentation assay
sample, and one HET-sCT-DnaJB1 sedimentation assay sample (discussed in Appendix B). The set
of samples were blotted with both the DnaJB1 polyclonal antibody used in figure 2.21 and the C-
terminus specific DnaJB1 antibody. The DnaJB1 polyclonal antibody detects a band between 10
and 15 kDa, as well as a chaperone band at 40 kDa and 80 kDa, which is consistent with our
previous Western blots using this antibody. In contrast, the antibody specific to the C-terminus
of DnaJB1 detects a 40 kDa band in both DnaJB1 control samples and the 25 kDa band in both
sedimentation assay samples. This antibody does not detect the 12 kDa band (figure 2.22).
39

2.3.2 Discussion

Our sedimentation binding assay showed that although DnaJB1 had some affinity for all
three fibril polymorphs used in our study, it had the highest affinity for fibrils made at room
temperature. This finding indicates that DnaJB1 binding to HTTex1 fibrils is polymorph specific.
However, the reason why RT-fibrils are preferred is not obvious. Our previous study on HTTex1
polymorphs showed that less bundled T fibrils and completely unbundled P fibrils were more
readily bound by the fibril-specific antibody MW8 than the more bundled N-fibrils [29]. We
40
interpreted this data such that the epitope of the MW8 antibody, located at the PRD of HTTex1,
would be more easily accessible in unbundled fibrils with more dynamic PRD. In contrast, DnaJB1
prefers an intermediate state prepared at room temperature. One potential explanation could
be that fibrils made at this temperature have a higher likelihood of presenting the correct, pre-
formed epitope for DnaJB1 binding. Another potential reason could be that RT fibrils have a
higher likelihood at presenting two DnaJB1 epitopes at the correct distance so that DnaJB1, being
a homodimer, binds HTTex1 fibrils with both monomers at the same time.
Besides a slight unbundling of HTTex1 fibrils, we observed a significant narrowing by
about 3 nm of all three fibril polymorphs upon the addition of DnaJB1. The fibril diameter of the
unbound HTTex1 fibrils of about 13 nm is in line with what has been observed previously by us
and others [21], [28 ]-[29],[95]. We also saw a small but significant increase in fibril diameter
from N to T fibrils. Lin et al. also observed a difference in diameter between HTTex1 fibrils made
at 37 ºC and 22 ºC although larger than 3 nm [41]. Interestingly, the addition of DnaJB1 reduced
the diameter of our HTTex1 fibrils but maintained the trend with N fibrils being significantly
narrower than T fibrils. What could be the origin of this change in diameter with the addition of
DnaJB1? One potential explanation could be an increase in dynamics of the C-terminal domain
making it less visible in negatively stained EM. Jain et al. recently reported similar narrowing of
the HTTex1 fibril diameter when fibrils were formed in the presence of curcumin indicating that
fibril narrowing in the presence of a HTTex1 binder could be a more general phenomenon [95].
In addition to fibril narrowing, we also observed that our HTTex1 fibrils became more
brittle when adding DnaJB1 especially RT fibrils and when HTTex1 was in excess. The resulting
fragmentation on the EM grid came as somewhat of a surprise because disaggregation of HTTex1
41
fibrils involving DnaJB1 had previously only been observed in combination with Hsc70, the
nucleotide exchange factor Apg2, and in the presence of ATP [34]-[35]. The fact that this
fragmentation occurred at regular intervals (every ~35 nm) and that fragments were often found
lined up in a row on the EM grid, indicates that the fragmentation occurred either when the fibrils
attached on the grid or during staining. One possible explanation for this fragmentation could be
that DnaJB1 binds preferentially at specific, periodic regions of the HTTex1 fibrils thereby
destabilizing the fibril integrity at regular intervals. In other words, DnaJB1 interaction would
make HTTex1 fibrils more brittle rather than disaggregating them itself. On the EM grid the thus
destabilized fibrils could break at regular intervals either when binding or during staining. That
we didn’t observe this fragmentation at equal molar ratios of HTTex1 and DnaJB1 might be
because DnaJB1 now binds HTTex1 at all available binding sites, not only the preferred regions.
Similar fragmentation of cross-β fibrils in the presence of chaperones has been observed by
others: Gao et al. observed fragmentation of α-synuclein fibrils in the presence of DnaJB1, Hsc70,
and Apg2 with the addition of ATP. Also, in this case fibrils fragments were found to line up in a
row on the EM grid suggesting that the fragments were still connected when the fibril attached
to the grid [36]. Similarly, Stephanenko et al. observed the fragmentation of β 2-microglobulin and
lysozyme fibrils at low pH by the heat shock protein alpha-β-crystallin. However, in both cases
the resulting fibril fragments were relatively irregular in length in contrast to what we observed
in the present study.
Our solid-state NMR data indicate that DnaJB1 is not binding to the polyQ domain which
is the site of mutation in HD and which forms the static fibril core of HTTex1 fibrils. Even the more
42
static Pro signals that likely originate from Pro close to the polyQ domain are not affected neither
in their chemical shift nor their overall intensity.
Using NMR techniques selective for the dynamic framing sequences of HTTex1 fibrils, we
did observe changes in intensity and chemical shifts of a few residues in the presence of DnaJB1.
These residues were compatible with a location at the very end of the PRD termed C12. Mariscal
et al. recently reported the binding of DnaJB1 to soluble HTTex1 using cross linking experiments.
They found the binding site to be located at the P10 region i.e. the second of the two polyP
stretches of the PRD just N-terminal of the C12 region [35]. We did observe a slight increase in
the intensity of Pro in a cis conformation upon interaction with DnaJB1 which could be
compatible with this observation. However, the majority of Pro signals remain unchanged
between the bound and unbound form of HTTex1. Considering that the key residues showing
chemical shift changes in the C12 region remain dynamic, it is possible that the major changes
observed are adjacent to the binding site rather than the binding site itself which should be more
immobilized and not detectable in our J-based NMR spectra.
In summary, we find the interaction of DnaJB1 with HTTex1 fibrils to be surprisingly
polymorph-specific with fibrils made at room temperature bound preferentially. DnaJB1
interaction leads to fibril narrowing and fragmentation at high HTTex1:DnaJB1 ratios. At equal
molar ratios, fibrils lose their smooth appearance indicative of DnaJB1 binding. Our solid-state
NMR data confirm the interaction of DnaJB1 with the C-terminal PRD. The polymorph specificity
of DnaJB1 binding might have important consequences for the efficiency of the Hsc70
disaggregation of existing HTTex1 fibrils, while likely leaving the prevention of HTTex1 fibril
formation unaffected.
43
Interestingly, previous work on DnaJB1, Hsc70, Apg2 disaggregation of HTTex1 fibrils used
fibrils made at room temperature [34]. In contrast to this, cryo-EM tomography of polyQ
aggregates in cellular inclusions showed rather unbundled fibrils reminiscent of our T fibril
preparations [42]. Determining how this polymorph specificity affects the binding and efficiency
of the entire chaperone network will be an important next step.
The fact that DnaJB1 binding to RT fibrils leads to the appearance of low molecular bands,
as observed in our sedimentation assay, is quite interesting, and may be related to the intimate
relationship these RT fibrils have with the chaperone. We know that DnaJB1 remains in the pellet
with the fibrils because of the sedimentation assay data, but also from unbinding assays in which
we have disaggregated the pellet and detected DnaJB1 in it as well as from dot blot studies which
also detect DnaJB1 in the pellet.
Based on sedimentation assay experiments, we know that DnaJB1 does not pull down
with protofibrils nor formic acid treated fibrils, even at high centrifugation speeds. And while we
see some pulldown of the chaperone with T fibrils, it is not to the same degree as that observed
with RT fibrils. Using ssNMR, we see that in the case of T fibrils, chaperone addition leads to
immobilization of the dynamic domains, but with no further effect on the chemical environment
of the fibrils. What these fibril types have in common is that they are less bundled than RT fibrils.
As stated above, the preference for these fibrils may arise from the fact that due to their overall
organization, they present the correct epitope and provide the right distance for maximum
binding of the chaperone.
In addition to this, we suggest that the preference for these fibrils, may also come from
the fact that this fibril type is able to hold on to the chaperone better than other fibril types. In
44
other words, once DnaJB1 binds RT fibrils, it becomes intertwined with their PRD domains and
remains attached. The finding that chaperones and other cellular components can be
sequestered by amyloid fibrils is not uncommon, as several studies have found chaperones co-
clustered with aggregates [43]. That DnaJB1 may be sequestered with our HTTex1 RT fibrils helps
explain the appearance of the additional low molecular weight bands in the following way.
Sequestration of the chaperone by the PRD domains leads to pressure on the chaperone
structure and may push some of the chaperone towards fragmentation at its binding site. This is
similar to our observations indicating that DnaJB1 facilitates the breakage of HTTex1 on our EM
grids by making the fibrils more brittle. In addition, this possibility is further supported by the fact
that the low molecular weight bands appear only for certain samples incubated with HTTex1, but
never for DnaJB1 alone. In all our assays we include a DnaJB1 alone control, in these samples
DnaJB1 remains in the supernatant, as a 40 kDa band with no decrease in its intensity, regardless
of the incubation time. This tells us that the bands are the result of the interaction between the
two proteins, rather than just degradation of DnaJB1.
DnaJB1 binds HTTex1 via its linker domain located between CTI and CTII, if once bound
the chaperone were to undergo added pressure and break at this site the fragments that arise
would be of 26 kDa and 12 kDa (figure 2.23). From our western blot experiments (figure 2.21 and
2.22), we know that the 12 kDa and 25 kDa bands are recognized by antibodies specific for
DnaJB1. The question of why the 25 kDa band was not detected by the polyclonal antibody might
have to do with inaccessibility of this antibody’s epitope in this fragment. Furthermore, our
hypothesis that the low molecular bands correspond to DnaJB1 comes from the fact that we see
these same bands appear in sedimentation assays conducted on the HET-sCT chimer discussed
45
in Appendix B. The Western blot results confirm that the low molecular bands observed in this
case are also DnaJB1. This is an interesting finding, given that we know that the chaperone uses
CTDI and CTDII to bind substrate, these results could indicate that, similar to HTTex1, DnaJB1 uses
the same linker region to bind the HET-sCT fibrils as well.
The finding that the 25 kDa band is not present in the TFA samples, adds to our hypothesis
in the following way. When chaperone in hepes or phosphate solution is added to fibrils, the
addition of buffer contributes to greater bundling of the PRD domains. Therefore, once bound
the additional bundling, results in the co-clustering of DnaJB1 with the PRD. In TFA solution this
does not happen, because TFA works to unbundle fibrils rather than bundle them, therefore, in
this situation, DnaJB1 can bind the PRD domain, but does not become trapped by it and
ultimately does not fragment.
Thus far, this is the only hypothesis that can possibly reconcile our findings, but further
testing is necessary to fully validate this. Future efforts to characterize the HTTex1-DnaJB1
interface will most likely focus on using the TFA dH 2O condition.
46

2.4 Materials and Methods

Plasmid Cloning

HTTex1 monomer
HTTex1 Q25 with a 6X His tag at its N-terminus was obtained from GenScript Inc. The
construct was cloned into a pET32 vector using the restriction sites NdeI and BamHI and was
codon optimized. The construct was designed with a glycine linker between the thioredoxin tag
and the HTTex1 construct. The final translated sequence is as follows:

MSDKIIHLTDDSFDTDVLKADGAILVDFWAEWCGPCKMIAPILDEIADEYQGKLTVAKLNIDQNPGTAPKYGI
RGIPTLLLFKNGEVAATKVGALSKGQLKEFLDANLAGSGSGERQHMDSPDLGTDDDDKGGGHHHHHHMA
TLEKLMKAFESLKSFQQQQQQQQQQQQQQQQQQQQQQQQQPPPPPPPPPPPQLPQPPPQAQPLLPQP
QPPPPPPPPPPGPAVAEEPLHRP

47
Protein Expression and Purifica:on

Huntingtin exon 1 (HTTex1)
HTTex1 plasmids were transformed into chemically competent BL21(DE3) cells. Starter
cultures were grown at 37 ºC for 3-4 hours and then expanded into LB medium containing
kanamycin for pET28 constructs and ampicillin for pET32 constructs. Expanded cultures were
then grown at 37 ºC until an OD 600 of 0.6-0.8, at which point expression was induced with
Isopropyl b-D-1-thiogalactopyranoside (IPTG) at a final concentration of 1 mM. The temperature
was then decreased to 18 ºC and cultures were incubated for 48 hours. Uniformly labeled
13
C,
15
N HTTex1 was expressed in the same way but using M9 minimal media and following the
protocol by Marley et al. [96]. Cells were centrifuged at 4,000 rpm for 20 minutes using an F9-
6x1000 LEX rotor and pellets were stored at -80 ºC until further use. For purification, cells were
lysed in a buffer containing 20 mM Tris-HCl pH 7.4, 300 mM NaCl, 10 mM imidazole, and 1 mM
DTT and sonicated using a QSonic Ultrasonic Sonicator. The cell lysate was then centrifuged at
18,000 rpm for 20 minutes using an F21-8x50y rotor. The supernatant was incubated on His60
Superflow resin at 4 ºC for 45-60 min. The column was then washed using a buffer composed of
20 mM Tris-HCl pH 7.4, 300 mM NaCl, 50 mM imidazole, and 1 mM DTT. The protein was eluted
using a buffer containing 20 mM Tris-HCl pH 7.4, 300 mM NaCl, 300 mM imidazole, and 1 mM
DTT. The protein was then diluted using a 10 mM Tris-HCl pH 7.4 solution and loaded on to a
HiTrap-Q XL anion column (GE). The column was washed using buffer A (20 mM Tris-HCl pH 7.4,
20 mM NaCl) and the protein was eluted using a salt gradient composed of buffer A and buffer B
(20 mM Tris-HCl pH 7.4, 1 M NaCl). For fibril formation, after elution, protein was diluted to a
final concentration of 25 μM using a 20 mM Tris-HCl pH 7.4 solution and was digested and
prepared according to steps described in the fibril formation section below.
48
For monomer studies, the protein was prepared in two ways; One type of preparation
involved diluting the protein to 25 μM and then attaching it to agarose beads without EKMax
digestion. For the second preparation, fusion protein was diluted to 5 µM and then digested for
50 min with EKMax (1 unit EKMax per mL of reaction volume). The reaction was quenched with
the addition of 4M urea. The protein mixture was then purified using a reverse phase C4 column
(Phenomenex) and using buffer A (99.9% H 2O, 0.1% TFA) and buffer B (90% acetonitrile, 9.9%
H 2O, 0.1% TFA). Purified HTTex1 samples were then lyophilized and stored.

Ulp1 Protease
Ulp1 protease catalytic domain (402-621), with His tag at its C-terminus, was transformed
into chemically competent BL21(DE3) cells. Starter cultures were grown at 30 ºC for 14-16 hours
and then expanded into 500 mL LB medium cultures in the presence of kanamycin. Cultures were
then grown at 37 ºC until an OD 600 of 0.6-0.8. Isopropyl 1-thiol-β D-galactopyranoside (IPTG), at
a final concentration of 1 mM, was then added to induce protein expression. Cultures were
further incubated at 37 ºC for an additional 4 hours, after which cells were centrifuged at 4,000
rpm for 20 minutes using a F9-6x1000 LEX rotor (Thermo Scientific). Pellets were then stored at
-80 ºC until further use. For purification, cells were resuspended in 20 mM Tris-HCl pH 8.00, 300
mM NaCl, and 10 mM imidazole containing 20% sucrose (w/v), 0.2% Triton X-100 (v/v) and β-
mercaptoethanol at a final concentration of 1 mM. Cells were then sonicated using a QSonic
Ultrasonic Sonicator and the cell lysate was centrifuged at 18,000 rpm for 20 minutes using a F21-
8x50y rotor (Thermo Scientific). The supernatant was then incubated on His60 Ni Superflow resin
(Takara Bio USA, Inc.) and shaken at 4 ºC for 20-30 minutes. The column was then washed with
20 mM Tris-HCl pH 8.00, 300 mM NaCl, and 50 mM imidazole containing 1 mM β-
49
mercaptoethanol. Protein was eluted in 20 mM Tris-HCl pH 8.00, 300 mM NaCl, and 300 mM
imidazole with β-mercaptoethanol and dialyzed against 25 mM Tris-HCl pH 8.00, 250mM NaCl,
1% Triton X-100 (v/v), and 50% glycerol (v/v) containing 0.5 mM dithiothreitol. After dialysis,
protein aliquots were stored at -80 ºC.

DnaJB1 and Hsc70
DnaJB1 and Hsc70 with N-terminal His-SUMO tag were transformed into chemically
competent BL21(DE3) cells. Starter cultures were grown overnight for 14-16 hours at 30 ºC and
then expanded into LB medium containing appropriate antibiotics. Cultures were then grown at
37 ºC to an OD 600 of 0.6-0.8 and protein expression was induced using IPTG at a final
concentration of 1 mM. The temperature was lowered, and cells were incubated at 20 °C for 20
hours. Cells were harvested via centrifugation at 4,000 rpm for 20 minutes using a F9-6x1000 LEX
rotor and stored at -80 ºC for further use. For purification, cells were resuspended in buffer
composed of 30 mM Hepes pH 7.4, 500 mM KCl, 5 mM MgCl 2, 30 mM imidazole, and 10% glycerol
containing phenylmethylsulfonyl fluoride (PMSF), Pierce Protease Inhibitor EDTA-free, DNase I,
and β-mercaptoethanol. Cells were then sonicated using a QSonic Ultrasonic Sonicator and cell
lysate was centrifuged for 30 minutes at 20,000 rpm using an F21-8x50y rotor. The supernatant
was added to His60 Ni Superflow resin and incubated on a shaker at 4 ºC for 1 hour. The column
was then washed with high salt buffer (30 mM Hepes pH 7.4, 1 M KAc, 5 mM MgCl 2, 25 mM
imidazole, and 10% glycerol) followed by low salt buffer (30 mM Hepes pH 7.4, 50 mM KAc, 5
mM MgCl 2, 25 mM imidazole, and 10% glycerol) both containing β-mercaptoethanol. The protein
was then eluted with buffer containing, 30 mM Hepes pH 7.4, 100 mM KAc, 5 mM MgCl 2, and
300 mM imidazole and β-mercaptoethanol. The eluted protein was combined with Ulp1 protease
50
at a ratio of 400 μg of protease per 1 mL of substrate and dialyzed against 30 mM Hepes pH 7.4,
100 mM KAc, 5 mM MgCl 2, and 10% glycerol with β-mercaptoethanol. For certain experiments,
the protein was also dialyzed into buffer composed of 20 mM phosphate, 20 mM KCl pH 6.00 or
a Trifluoroacetic Acid (TFA)-dH 2O mixture (1:4000) pH 2.5. After dialysis the sample was
centrifuged for 20 min at 4,000 rpm using an Eppendorf A-4-44 rotor and the supernatant was
then added to equilibrated His60 Ni Superflow resin. The sample was incubated at 4 ºC for 20
minutes, after which the column was placed back on a stand and the flowthrough, containing the
cleaved chaperone, was collected. Cleavage of SUMO tag was verified by running samples on an
SDS-PAGE gel.
HTTex1 fibril forma:on

HTTex1 fibril seed preparation was initiated through the removal of the thioredoxin fusion
tag from the HTTex1 fusion protein. 1 unit EkMax (Invitrogen) was used per 280 μg of soluble
fusion protein. Upon addition of EkMax, the solution was quiescently incubated, and fibrils
isolated as described previously [89]. Fibrils for this study were formed in a similar manner, after
diluting purified fusion protein to 25 μM, using 20 mM Tris-HCl pH 7.4, 5% molar ratio of seeds
was added, followed by 1 unit EkMax per 280 μg of protein. The fibril mixture was then incubated,
without agitation, for two weeks [89]. For generation of 4 ºC fibrils, the fibril mixture was
incubated at 4 ºC, room temperature fibrils were incubated at 20-22 °C, and 37 ºC fibrils at 37 ºC.
Fibrils were then isolated by centrifugation for 1 h at 13,500 rpm using an Eppendorf 5804R
centrifuge (F45-30-11) and analyzed using EM microscopy. Fibril samples were stored in a
Trifluoroacetic acid-water mixture (1:4000, water in excess) and NaN 3 at a final concentration of
0.02% (w/v) was added to all preparations.
51
Binding of Protein to Agarose Beads

50 µL of Nickle NTA (Ni-NTA) (His 60 - Takara BioLabs) or Copper NTA (Cu-NTA) (Copper
Chelating Agent-Gbioscience) agarose beads were centrifuged at 4,500 rpm using a F9-6x1000
LEX rotor for 10 min. The supernatant was then removed, and the beads were washed with 50
µL of dH 2O. The beads were centrifuged again to remove the dH 2O and were then resuspended
in buffer composed of 20 mM phosphate at pH 8.00. The agarose beads were centrifuged again
to remove the buffer. HTTex1 sample was then added to the beads. The mixture was then
incubated on a shaker at room temperature for 45 min to 1hr. After the incubation period, the
sample mixture was centrifuged again, and after removal of the supernatant it was packed into
a 4mm ssNMR rotor.
Sedimenta:on Assay

HTTex1 Q46 fibrils were incubated with DnaJB1 at a molar ratio of 4:1 (i.e. fibrils in
excess). DnaJB1 and HTTex1 alone with the equivalent amount of buffer served as controls.
Samples were incubated at room temperature on an IKA rocker 2D shaker and then centrifuged
for 1 h, 13,500 rpm and 4 ºC (Eppendorf 5804R) to separate the insoluble portion (fibril pellet),
containing bound DnaJB1 from the soluble portion (supernatant), composed of unbound DnaJB1.
The supernatant was then analyzed using SDS PAGE. Fibril pellets containing bound DnaJB1 were
then resuspended in buffer or deionized water alone and used to prepare EM grids as described
below.
Dot Blots and Western Blots

For dot blot analysis, HTTex1 fibrils and DnaJB1 were mixed and incubated as detailed
above in sedimentation assay. Samples were then centrifuged, and pellets were isolated. Pellets
52
were then resuspended and blotted on to a nitrocellulose blotting membrane (GE Healthcare Life
Sciences Amersham). Membranes were blocked with a 10% milk solution and blotted with either
a DnaJB1 antibody (Rabbit Polyclonal - Proteintech 13174-1-AP) prepared in a TBST and BSA
solution or a His antibody (Rabbit Polyclonal - Rockland 600-401-382) prepared in the same
manner. An anti-rabbit antibody conjugated to peroxidase (Goat - Rockland 611-1302), and
prepared in the same TBST and BSA solution, was used as a secondary antibody. Membranes
were developed using an Amersham ECL western blotting analysis system and following the
instructions provided. Membranes were imaged with a Syngene G:Box imager.
For Western Blots, after the sedimentation assay was performed, the supernatant was
isolated and run on an SDS page gel. Proteins were then transferred to a nitrocellulose botting
membrane using a Bio-Rad Mini Protean Tetra Cell system run at 100 kV. Membranes were
blocked with a 10% milk solution and blotted with the same primary antibodies used for dot blot
experiments in addition to a C-term DnaJB1 antibody (Rabbit polyclonal-Antibodies-online.com
ABIN1107003). The same anti-rabbit antibody conjugated to peroxidase was used as a secondary
in these experiments. Western blot membranes were developed and imaged in the same manner
as described for dot blots.
ssNMR Sample Prepara:on and Measurements

For HTTex1 monomer experiments, samples were packed into 4.0 mm magic angle
spinning (MAS) rotors using a home-built centrifuge tool. NMR measurements were done on an
Agilent DD2 600 MHz solid-state NMR spectrometer using a triple-resonance, 4.0 mm probe set
at 9 kHz (MAS). Temperatures of 0 ºC, -10 ºC, and -20 ºC were used for experiments. Hard pulses
of 100 kHz (
1
H) and 50kHz (
13
C) were used. For the 1D
13
C CP experiments, the Hartman-Hahn
53
match at 60 kHz (
13
C) and 85 kHz (
1
H) and 60 kHz XiX [97] during detection was used. 2D
13
C-
1
H
INEPT- HETCOR (heteronuclear chemical-shift correlation) and DARR (
13
C-
13
C dipolar assisted
rotational resonance) spectra for this sample started with the same CP followed by a 50 ms long
mixing time using a
1
H recoupling field of 10 kHz and the N = 0.5 condition [98]. The spectral
widths were 50 kHz in both dimensions and 500 indirect, complex increments were detected
using 32 acquisitions each.
For measurements of HTTex1 fibrils alone and fibrils with the chaperone DnaJB1, fibrils
were incubated with equal or excess chaperone (1:1 or 2:1 molar ratio) or in buffer alone and as
described above. The pellets of fibrils alone or fibrils with bound chaperone were then isolated
and packed into 1.6 mm magic angle spinning (MAS) rotors using a home-built centrifuge tool.
NMR measurements were done on the same Agilent DD2 600 MHz solid-state NMR spectrometer
used for monomer measurements but using a triple-resonance, 1.6 mm probe set at 25 kHz
(MAS). A temperature of 0°C was used for all experiments. Hard pulses of 200 kHz (
1
H), 100 kHz
(
13
C), and 50 kHz (
15
N) were used. For the 1D
13
C CP experiments, the Hartman-Hahn match at 60
kHz (
13
C) and 85 kHz (
1
H) and 140 kHz XiX decoupling during detection was used. 2D PARIS
experiments started with the same CP followed by a 50 ms long mixing time using a
1
H recoupling
field of 10 kHz and the N = 0.5 condition [98]. The spectral widths were 50 kHz in both dimensions
and 500 indirect, complex increments were detected using 32 acquisitions each. The 2D
13
C-
13
C
INEPT TOBSY, 2D
1
H-
13
C INEPT HETCOR, 2D
1
H-
15
N HSQC, and 3D HNCA spectra were recorded as
described previously [27]-[28].
54

Electron Microscopy

For EM, 10 μl of sample was pipetted onto parafilm. EM grids (150 mesh copper) were
then placed on top of the samples and allowed to incubate for 5 minutes. Negative staining was
achieved by incubating the grids on 10 μl of 1% uranyl acetate solution for 2 minutes. Excess
sample and uranyl acetate stain were removed by blotting grids onto filter paper. For washing,
grids were tapped over uranyl acetate followed by deionized water. Grids were then allowed to
air dry. All EM grid images were captured using a Gatan digital camera and a JEOL-JEM 1400
transmission electron microscope at 100 kV.
Fibril Measurements

All fibril measurements were done using the Fiji - ImageJ straight line tool. The straight-
line selection was then measured and automatically tabulated using the measure command on
Image J, thereby generating a .csv file. Fibril width measurements were taken perpendicular to
the fibril axis and spanned from side to side of the negatively stained fibrils. Only fibrils with clear
and defined edges were used for measurement. Fibril fragment lengths were measured in the
same manner except that measurements were taken along the axis of the fibril and measured
from end to end of the fragment. As above, only fibril fragments with clear and defined ends
were measured.

55
Chapter 3: Studies on Functional Amyloids
3.1 Introduction

Memory can be grouped into two main categories: short term memory, which only
persists for a few minutes to hours, and which only requires short storage and recall of
information, and long-term memory which can last several days or longer and, therefore,
requires the ability to store information for extended periods of time [44]. Short term memory
formation depends on covalent modification of already existing proteins, which can then
modulate properties of the nerve cell and their synaptic connections [45]. In contrast, long term
memory formation requires the activation of effector genes, protein modification, protein
synthesis and ultimately a new synaptic connection [45]. Long term memory formation is,
therefore, a highly orchestrated process and one in which amyloid fibrils play a critical role.
Work with Xenopus, Aplysia, Hermissenda, and Drosophila has contributed significantly
to our understanding of the mechanism underlying memory formation. Such studies have
identified CPEB proteins as key regulators of protein synthesis at the synapse. Cytoplasmic
polyadenylation element binding proteins (CPEBs) are a family of RNA binding proteins, which
function to polyadenylate mRNA, thereby activating them for translation at the synapse [46].
Studies by Si and co-workers found that in Aplysia, CPEB (apCPEB) is required for long term
maintenance of facilitation at the synapse and that its depletion leads to inhibition of this process
[10]. This same study also found that the N-terminus of apCPEB has features which are similar to
those of the prion-like domain in yeast. Later studies by the same group further confirmed
apCPEBs prion like properties and additionally found that the aggregated form of the protein is
56
its most active state. Si and colleagues found that soluble apCPEB did not bind RNA, instead the
RNA was bound by small amyloid like aggregates of apCPEB [47].
CPEB proteins are highly conserved among species, all CPEB proteins have two RNA
bindings motifs and a zinc finger domain at its C-terminus but are different in their N terminal
composition. In vertebrates, there are four CPEB proteins (CPEB 1 – 4) which have functions
similar to that of apCPEB and which have also been shown to aggregate. Although their
importance in the regulation of RNA is well established, this group of CPEB proteins has not been
studied as extensively as apCPEB. Additional work is being done towards this aim, and studies,
thus far, have shown that in mice, knock outs of CPEB 1, 2 and 3 affect the formation and
maintenance of spatial and fear memory [48]. For CPEB 4, there are several studies which
demonstrated its function in both mitotic and meiotic cell cycle progression, as well as its
importance in supporting neuron viability [99]. However, its role in memory formation, is not yet
as clear.
In Drosophila, CPEB members are Orb and Orb2A. While Orb functions in oogenesis, Orb2
is critical to the formation and maintenance of long-term memory in the fruit fly. Like apCPEB,
Orb2 was found to be capable of existing in a soluble or aggregated form an to be enriched in the
synaptic region of neurons, where it contributes to the formation of long-term memory [49]-[50].
Further studies also found that Orb2 also functions as a repressor in its soluble form and as an
activator in its aggregate status [51]. Orb2 is composed of two isoforms, Orb2A and Orb2B.
Although the majority of the Orb2 aggregates are composed of Orb2B, Orb2A is more efficient at
aggregating and has been shown to be essential for the nucleation and regulation of fibrils made
by Orb2 [50]. Although Orb2A and Orb2B are similar in the composition of their C-terminus, and
57
both have a glutamine rich domain (Q-rich), their N-terminus is quite different. Orb2B has 178
residues preceding its Q-rich region, Orb2A only has 21 residues, with residues 1 to 9 being
unique to this isoform.
Work by Majumdar and co-workers further established the importance of the N-terminus
of Orb2A by showing that the first 88 amino acids (Orb2A88) of this isoform are capable of
forming amyloid like aggregates in-vivo. In our own studies, we showed that the Orb2A88
construct is capable of forming amyloid fibrils in-vitro as well [52]. Additionally, we found that
the first 21 residues of Orb2A (Orb2A1-22) can form fibrils in-vitro and contain the static fibril
core of the Orb2A88 fragment [52] (figure 3.1). These 21 N terminal residues contain the 8
residues unique to Orb2A, and which studies have shown to be important for Orb2A fibril
formation. Although these 8 amino acids are not sufficient to form amyloid fibrils on their own,
in-vivo studies have shown that their deletion greatly affects Orb2A aggregation [50]. These same
studies identified the conserved phenylalanine residue at position 5 (F5) as critical to reducing
aggregate formation. A point mutation of this residue, phenylalanine to tyrosine (F5Y), not only
reduced amyloid like oligomerization in S2 cells, but in addition, led to impairment of long-term
memory formation in fruit flies. An analysis of puncta formation in S2 cells revealed that this
mutant exhibited diffused fluorescence. Additional studies showed that F5Y puncta are more
dynamic, and that F5Y aggregates are nonhomogeneous in nature. Overall, this points to a
variation in the structural packing of F5Y as compared to Orb2A88 [50].

58

Having established that the N-terminal 21 residues preceding the Q/H-rich domain of Orb2A
are sufficient for fibril formation and that these same N-terminal residues can bind to negatively
charged lipid vesicles and subsequently form a helical structure that prevents the formation of
amyloid fibrils [53]. We wondered: what is the role of the Q/H-rich domain of Orb2 if it is not
involved in the formation of the amyloid core as originally thought? Interestingly, the Q/H-rich
region of Orb2 strikingly resembles the glutamine and histidine rich bacterial Hpn-like protein
(Hpnl) found in Helicobacter pylori, which colonizes the human gut. Hpnl has specific Ni
2+
, Cu
2+
,
Co
2+
, and Zn
2+
binding activity that plays a role in nickel storage, which is important for the
colonization of the host [54]. Several histidine residues of Hpnl were found to be essential for
metal binding, while the glutamine residues were responsible for stabilizing the Hpnl-metal
complex [54]-[55].
59
The high histidine content in Orb2’s N-terminus and its similarity to Hpnl suggests that Orb2
might be able to bind transition metals similar to Hpnl. Interestingly, several other amyloid
forming proteins, such as amyloid b (Aβ), α-synuclein, and the prion protein (PrP), have metal
binding activity [56]. In the case of Aβ, Cu
2+
and Zn
2+
, binding not only modulates aggregation
kinetics but was also hypothesized to cause toxicity [57].

Aims of the study

In our aims to further characterize Orb2A88, we set out to describe the affinity of Orb2A87
to several transition metal ions and characterize the effect of metal binding on aggregation
kinetics. Our results, discussed below, lead to some new hypotheses suggesting that metal
interaction might also play a role in modulating the aggregation kinetics of Orb2 in vivo and could
play a role in the regulation of long-term memory in Drosophila.
Similarly, we wondered about the importance of the F5Y mutation. Given that studies by
Majumdar and co-workers highlighted the importance of this point mutation, we set out to
investigate how this substitution translates to studies in-vitro and specifically, how it would affect
the aggregation and structure of Orb2A88.
3.2 Metal Binding Properties of the N-terminus of the Functional Amyloid Orb2
* This work was originally published in Volume 7, issue 3, page 57 of Biomolecules 2017 (doi:
10.3390/biom7030057) as “Metal Binding Properties of the N-Terminus of the Functional Amyloid Orb2”.
The following is an adaptation of that publication.

3.2.1 Results

The N-terminus of Orb2A Binds Transi:on Metals in the Absence of a Polyhis:dine-Tag

To test whether Orb2 binds metals as hypothesized, we cloned the first 88 residues of
Orb2A (Orb2A88-His) with a C-terminal polyhistidine-tag and later removed the polyhistidine-tag
60
by mutating in an earlier stop-codon (Orb2A87). Both Orb2A88-His and Orb2A87 bind to nickel
charged, nitrilotriacetic acid (Ni-NTA) resin although the binding affinity of Orb2A87 is reduced.
All experiments described in the following were done on Orb2A87, i.e., without additional
polyhistidine-tag. To test whether Orb2A87 binds to metals other than nickel, we tested Orb2A87
binding to NTA resin that was stripped from metal ions and re-charged with Ni
2+
, Cu
2+
, Co
2+
, or
Zn
2+
. Soluble recombinant Orb2A87 was bound to these resins at pH 8.0 and eluted from the
resin at pH 3.75, at which point the histidine residues are expected to be protonated and lose
their metal-binding affinity. As can be seen from figure 3.2, Orb2A87 is able to bind to all of these
metal ions, but not to the NTA resin without chelated bivalent metal ions. Although the same
amount of protein was loaded to the same amount of resin, the intensities of the elution from
the Cu-NTA resin is lower, probably because of the lower specificity and higher affinity of this
resin [58].

61
Isothermal Titra:on Calorimetry to Characterize Metal Binding Affinity of Orb2A87

To determine the dissociation constant and number of binding sites of the metal–
Orb2A87 complex, we measured isothermal titration calorimetry (ITC) using the following metals
ions: Ni
2+
, Cu
2+
, Zn
2+
, Ca
2+
, and Mg
2+
. To slow amyloid formation during these measurements, we
measured ITC in the presence of 1 M urea (see Materials and Methods section). Orb2A87 did not
have any measurable binding affinity to Ca
2+
and Mg
2+
. However, we measured μM dissociation
constants for Ni
2+
, Cu
2+
, and Zn
2+
(figure 3.3). The ITC data of both Ni
2+
and Zn
2+
fit well to a model
of approximately one binding site per monomer. The results of the fits are summarized in table
3.1. The binding parameters for Cu
2+
could not be determined with confidence because it could
only be measured in the presence of a weak ligand (glycine), which complicated its analysis [56].
However, the titration curve in figure 3.3 C indicates that Cu
2+
has the highest affinity of all
metals.

62

Several His:dines are Important for the Metal Binding Affinity of Orb2A87

To determine the location of the metal binding site of Orb2A87, we introduced several
point mutations in the Q/H-rich region, H29A, H46A, H60A, and H61Y. We mutated histidine to
tyrosine in the case of H61 because the H61Y mutation was previously shown by Majumdar and
co-workers to limit the ability of Orb2A to aggregate in S2 cells [50]. Otherwise, we preferred the
more neutral mutation from histidine to alanine. Orb2A87 H29A, H46A, and H60A could still be
purified using Ni-NTA affinity chromatography indicating that the metal binding properties of
these mutants remained intact. However, H61Y could not be purified this way, indicating that
this mutation entirely disrupted the metal affinity of the protein (figure 3.4 D). The ITC data
shown in figure 3.4 A-C confirm that H29A, H46A, and H60A still bind nickel. As can be seen from
table 3.2, the nickel dissociation constants of H29A, H46A, and H60A are still in the μM range,
while the number of binding sites (n) is decreased to less than 1 per monomer of Orb2A87.

63

Orb2A-Metal Interac:on Affects the Aggrega:on of Orb2A87

To test the effect of metal binding on the structure of Orb2A87, we recorded circular
dichroism (CD) spectra of Orb2A87 in the presence and absence of several metal ions. As can be
seen from figure 3.5, Orb2A87 has a predominantly random coil conformation in the absence of
metal ions as indicated by the minimum at about 202 nm. The addition of the non-binding metals
Mg
2+
and Ca
2+
did not have any effect on the CD spectrum of Orb2A87. Cu
2+
, Ni
2+
, and Zn
2+
, which
64
bind to Orb2A87, led to a decrease in the CD signal. The overall shape of these less intense spectra
stayed the same giving no indication of structural changes. However, the reduction of the CD
signal and the absence of an isosbestic point also indicate that part of the sample aggregated and
was thus not detected in the presence of Cu
2+
, Ni
2+
, and Zn
2+
.

To study the effect of metal binding on Orb2A87 aggregation and amyloid formation, we
measured thioflavin T (ThT) fluorescence kinetics in the absence of metal and in the presence of
Ni
2+
. We used Mg
2+
, which did not bind to Orb2A87, as a negative control. As can be seen
from figure 3.6, the addition of Ni
2+
results in an immediate increase in ThT fluorescence,
whereas Orb2A87 alone or in the presence of Mg
2+
shows an increase in fluorescence after a lag
phase of about 3 h. After 10 h, the ThT fluorescence in the presence of Ni
2+
starts to decrease
again whereas it keeps increasing in the absence of Ni
2+
. To test whether the immediate increase
65
in ThT fluorescence with the addition of Ni
2+
was the result of metal induced fibril formation, we
took EM images from samples that were negatively stained 6 h after the beginning of the
aggregation study (i.e., when the ThT fluorescence of the sample containing Ni
2+
was close to its
maximum). As can be seen in figure 3.7, both Orb2A87 in the absence of metal or in the presence
Mg
2+
shows the formation of bundled fibrils. However, in the presence of Ni
2+
, no clear fibrillar
structures could be observed and the aggregates looked rather non-specific.

66

3.2.2 Discussion

The N-terminus of Orb2A Has an Affinity for Bivalent Metal Ions

In their study of Hpnl, Zeng et al. found that Hpnl binds bivalent metal ions using ITC. The
presence of a similar histidine-rich domain in Orb2 led us to hypothesize that it too possesses a
similar binding affinity for bivalent metal ions. This study confirms this hypothesis using metal
affinity chromatography. As shown in figure 3.2, Orb2A87 binds to Ni
2+
, Zn
2+
, Cu
2+
, and
Co
2+
chelated to NTA resin. The fact that Orb2A87 did not also elute from the Cu-NTA column,
might be a result of its higher binding affinity to this metal [59]-[60]. Our ITC data in figure 3.3
specify that Orb2A87 has one binding site for Ni
2+
, Zn
2+
, Cu
2
with micromolar affinity. However,
Orb2A87 does not bind Mg
2+
or Ca
2+
ions, indicating that the metal binding affinity of Orb2 is
specific to transition metal ions. Because the Cu
2+
binding experiment had to be performed in the
presence of glycine, a weak ligand, quantitative analysis was not performed, although this
interaction seems to have the lowest dissociation constant. Point mutations of several histidine
67
residues in the Q/H-rich region showed that Orb2A87 H61Y abolishes Ni
2+
binding. In contrast
Orb2A87 H29A, H46A, and H60A, still bound Ni
2+
with micromolar affinity although we observed
an apparent decrease in the number of binding sites per Orb2A87 monomer. This reduced
number of binding sites suggests that the absence of histidine residues at position 29, 46, and 60
may either reduce the number of monomers capable of binding metal or require two protein
molecules to maintain the metal binding properties of the protein. These observations confirm
that the metal binding site of Orb2A87 is indeed located in the Q/H-rich region, specifically in the
latter half of the domain, with the H61 residue being necessary for binding. Our CD spectra show
that soluble Orb2A87 is mostly disordered in the absence of Ni
2+
, Zn
2+
, Cu
2+
, which agrees with
our previous CD and Electron Paramagnetic Resonance (EPR) data on the soluble form of
Orb2A88 [53]. Finally, our CD, ThT, and EM data show that Ni
2+
binding induces the formation of
ThT active Orb2A87 aggregates that are different from the amyloid fibrils observed in the absence
of metal. Although an increase in ThT fluorescence has been shown to be relatively specific to
amyloid fibrils [61], there are examples of similar increases in fluorescence when ThT binds to
monomeric proteins [62] and amorphous aggregates [63]. Although aggregates formed in the
presence of Ni
2+
do not resemble amyloid fibrils, the increase in ThT fluorescence indicates that
a ThT binding pocket and possibly a cross β-sheet rich structure is formed upon metal binding.
In summary, our data confirm the original hypothesis that the Q/H rich domain of Orb2
can bind to transition metals via its His residues. Hence, the Q/H rich domain behaves like other
H-rich domains, most of which were shown to be metal binding domains. Examples of such
domains not only include the Q/H-rich domains of HpnI, but also H-rich domains that are not Q-
68
rich such as the histidine-rich glycoproteins (HRG), Cu and Zn-superoxide dismutase found
in Haemophilus, and histidine-proline-rich glycoprotein (HPRG) [64]-[66].
What is the role of the glutamines in Orb2 in the context of metal binding? One possibility
is that they help stabilize the metal binding complex similar to Hpn1 [55]. Another possibility is
that metal binding induces a structural change, or stabilizes a specific conformation of the
glutamines, or both. Finally, metal binding could play a role in stabilizing potential protein-protein
interaction of the Q/H-rich domain. Polyglutamine (Poly-Q) domains have been suggested to be
involved in coiled-coil protein–protein interfaces [67]. The additional histidines could potentially
initiate and strengthen this interaction upon metal binding. For example, ligand interaction of
HRGs is enhanced by its interaction with Zn
2+
[64]. When the Q/H-rich domain of Orb2 is plotted
on a helical wheel, the first six histidines fall on the same side, highlighting the potential of a
metal stabilized zipper motif for Orb2. Our data clearly show that metal interaction does affect
the structure of Orb2A87. However, it will be important to find out in which structural contexts
metal binding occurs to determine its exact structural role.
What is the potential function of this metal interaction in long-term memory
of Drosophila? Majumdar and co-workers showed that the H61Y mutant affected puncta
formation of full length Orb2A in S2 cells. They further showed that a different mutation with a
similar effect on puncta formation also inhibited long-term memory in flies [50]. Interestingly,
H61Y was the only histidine mutation that completely abolished metal binding of Orb2A87. These
data suggest that the metal interaction of Orb2 might play a role in its ability to aggregate in cells.
Our data show that metal interaction has a pronounced effect on the structure of Orb2A87. We
think that the same transition metals will also bind to the Q/H rich domain of full length Orb2A
69
and Orb2B and have a similar effect on the structure of these proteins. Furthermore, it is possible
that in a cellular environment this interaction is crucial for Orb2 aggregation to occur. Since the
Q/H rich domain is part of both the rare isoform Orb2A and the predominant isoform Orb2B, the
effect of metals might be the same for both isoforms and not necessarily be involved in
aggregation initiation which requires the unique N-terminus of Orb2A.
If transition metals are an important factor for the function of Orb2, which metal is most
likely binding Orb2 in vivo? Several metals have been shown to be important for memory
formation. For example, iron and zinc deficiencies affect neuronal development and can lead to
memory loss [68]-[70]. In particular, Zn
2+
has been associated with memory [71]. It has been
shown that Zn
2+
enters the presynaptic and postsynaptic neurons when released from synaptic
vesicles together with glutamate and that this translocation is important for LTP at mossy fibers
in rat [72]. Zinc could similarly act as a second messenger on Orb2 during long-term memory
formation in Drosophila by introducing a structural change or by regulating protein–protein
interactions. The basal concentration of free Zn
2+
in the cytosol is <10
−9
M, well below the
dissociation constant that we determined via ITC. However, during excitation, the intracellular
Zn
2+
concentration can significantly increase [73] potentially inducing structural changes in the
Q/H-rich domain of Orb2. The role of these structural changes could be either to initiate the
formation of functional Orb2 aggregates, or to induce a non-amyloidogenic, reversible off
pathway aggregation that would postpone amyloid formation until the Zn
2+
concentration goes
back to equilibrium.
Further research is necessary to reveal the role of metal binding on the glutamine/histidine
rich domain of Orb2 and the effect of this interaction on long-term memory in Drosophila.
70

3.3 Characterization of the F5Y Mutant of Orb2A88

3.3.1 Results

F5Y can be Purified Using Orb2A88 Condi:ons and can Form Amyloid Fibrils in-vitro

We generated the F5Y mutant using site directed mutagenesis and were successful at
purifying it using the same purification protocol used for Orb2A88. We then set out to investigate
F5Ys aggregation propensity, and whether aggregates formed by this mutant resembled or
differed from Orb2A88 fibrils. We tested F5Y in various buffers to optimize fibrilization. We
generally found that Renaturing buffer (50 mM NaH 2PO 4, 200 mM NaCl, 10% glycerol pH 8.0)
worked the best. In this buffer, we observed ThT fluorescence for both Orb2A88 and F5Y at week
1, 95 a.u. and 53 a.u., respectively. At week 2 we saw an increase in ThT fluorescence for both
Orb2A88 and F5Y, both constructs more than doubled in absorbance units to 253 a.u. (Orb2A88)
and 122 a.u. (F5Y). At week 3 we saw a decrease in the fluorescence of both proteins, but even
then, the ThT emission value for Orb2A88 was double that of F5Y, 202 a.u. and 106 a.u.
respectively (figure 3.8).
We followed the ThT study with electron micrographs of the samples. At week 1 we
observed structures that resembled amyloid fibrils in the F5Y sample. The fibrils appeared more
defined and assembled into larger aggregates by week 2 (figure 3.9). Morphologically, the F5Y
fibrils resembled those of Orb2A88.

71

72
Aggrega:on Kine:cs of Orb2A88 and F5Y

Our initial ThT studies showed that the F5Y mutant is capable of aggregating and that
morphologically the aggregates resemble those of Orb2A88 (figures 3.8 and 3.9). We next asked
how the fibrilization kinetics of F5Y compared to those of Orb2A88. To answer this question, we
ran ThT kinetic assays for Orb2A88 and F5Y in Renaturing buffer at pH 7.4. We observed a small
lag phase for both samples prior to their increase in fluorescence, both started at about the same
ThT emission value (approximately 20 a.u.) and increased gradually within the first 200 minutes
of measurement (figure 3.10). Although, initially, the increase in ThT emission is sharper for
Orb2A88, F5Y emission continues to increase as well. F5Y ThT emission values are statistically
similar to those of Orb2A88. In the end, both constructs end up reaching the same ThT emission
range.
We complemented the ThT kinetic studies with EM and made grids for the samples at 1
day, 2 days, and 1 week. Fibrils imaged resembled those observed in figure 3.9. Similar to
previous observations, we found that F5Y fibrils were similar in morphology to Orb2A88 fibrils.

73

3.3.2 Discussion

Our study on Orb2A88 and F5Y revealed that the F5Y mutant can form amyloid fibrils in-
vitro and that these fibrils are similar in morphology to those of Orb2A88. In both cases we saw
early fibrous aggregates which eventually matured into larger aggregates. Additionally, we found
that under certain conditions, both constructs aggregate at similar rates. For Orb2A88 fibril
conditions play a big role on the rate and morphology of our final fibril product. Similarly, with
F5Y, we found that using different conditions allowed us to influence the time required for
aggregate formation. In initial studies, we tested F5Y in phosphate buffer with NaF at pH 6.8 and
found that under these conditions, it aggregated quite slowly. We did not measure Orb2A88 in
the same conditions, therefore, we cannot compare their aggregation rate under these
circumstances, but the fact that F5Y aggregated slower in this buffer speaks to the importance of
fibrilization conditions. These findings are in line with In vivo studies, which showed that although
74
the F5Y mutation affected puncta formation, it did not impede it. And so, the effect of the F5Y
mutation on fibrillization can be mitigated in vitro via the use of different aggregation conditions.
In addition, later studies by Hervas and colleagues found that the amyloid core for full length
Orb2 encompasses the Q rich domain [74]. This implies that although the N-terminus is important
for fibril formation, perhaps by providing the initial interaction sites and forming a transitory core,
it does not form part of the final core domain of the fibrils and ultimately other domains of Orb2
can contribute to fibril formation.
In vivo studies on F5Y showed that the mutation affected the ability of the protein to form
puncta, and that the puncta that was formed had different characteristics to that of Orb2A88.
Similarly, in our studies we found that although F5Y is able to aggregate and that morphologically
these fibrils are similar to Orb2A88, F5Y behaves differently. As highlighted by the ThT study in
figure 3.8, under certain conditions, Orb2A88 is a better fibril former than F5Y. This observation
is also highlighted by our attempts to characterize the F5Y mutant using ssNMR. These studies
were impeded by the fact that we were not able to generate sufficient sample for a direct
comparison to data of Orb2A88. Altogether, this speaks to the variation in structural packing that
was suggested by the in vivo studies, which would ultimately affect the amount and quality of
fibrils generated by F5Y. In the end, further characterization of the mutant is needed to more
precisely identify the differences in its structure as compared to Orb2A88, ideally this future work
would be carried out using ssNMR.

75
3.4 Materials and Methods

Cloning

To create the Orb2A88 F5Y mutant, the Orb2A88 WT plasmid in a pET28b expression
vector was cloned using site-directed mutagenesis and a Q5 Master Mix kit from New England
Bio Labs.
Protein Expression

E. coli Rosetta 2 (DE3) cells (Novagen-EMD Millipore, Billerica, MA, USA) were transformed
with corresponding pET28b vectors. Transformed cells were grown in Luria Bertani (LB) Miller
medium with 35 μg/L chloramphenicol and 50 μg/L kanamycin at 30 °C for 15-18 h. Cultures were
diluted into LB Miller medium with 35 μg/L chloramphenicol and 50 μg/L kanamycin and grown
at 37 ºC until the optical cell density at 600 nm reached 0.6. An amount of 1 mM isopropyl 1-
thiol-D-galactopyranoside was then added to induce protein expression. Induction for Orb2A87
proceeded for 16 h while cells remained shaking at 37 ºC. For F5Y, after induction cells were kept
at 25 ºC for 16 to 18 hours. Cells were then pelleted by centrifugation at 4000 rpm for 20 min at
4 ºC using a Sorvall SLC-6000 rotor (Thermo Fisher Scientific Inc., Waltham, MA, USA). Cell pellets
were immediately stored at -80 ºC.
Purifica:on

Orb2A87
For ITC samples, cells were defrosted on ice and resuspended in denaturing buffer (8 M
urea, 10 mM citrate, 100 mM sodium phosphate, 10% glycerol, 0.05% 2-mercaptoethanol) at pH
8.0. The suspended cells were further lysed by sonication for 6 minutes using a cell disruptor
sonicator (Heat Systems Model W-220F Qsonica, Newtown, CT, USA). The lysate was later
76
centrifuged at 20,000 rpm for 20 min using a Sorvall ss-34 rotor (Thermo Fisher Scientific Inc.).
The resulting supernatant was loaded onto pre-equilibrated Sigma HIS-select Ni-NTA resin
(Sigma-Aldrich, St. Louis, MO, USA) and incubated on a shaker at room temperature for at least
1 h. Following incubation, the flow through was collected and the column was washed with
denaturing buffer at pH 8.0 to which 0.5% Triton X-100 and 500 mM NaCl were added. The
column was subsequently washed with denaturing buffer pH 6.75. Protein was eluted with a pH
step gradient in denaturing buffer. The protein predominantly eluted at pH 4.25. The resulting
pure protein was either used immediately or frozen in 2 mL aliquots using liquid N 2 and stored at
-80 ºC.
For samples used in ThT fluorescence and CD assays, protein was purified in the same
manner as above. Except for the following modifications: (1) denaturing buffer contained no
glycerol and 10 mM Tris instead of citrate; (2) in addition to the Triton X-100, 500 mM NaCl, and
pH 6.75 washes, the column was further washed with renaturing buffer (50 mM sodium
phosphate, 200 mM NaCl, 10% glycerol, 0.05% 2-mercaptoethanol at pH 8.00) and renaturing
buffer containing 20 mM imidazole; (3) the protein was eluted in renaturing buffer containing
250 mM imidazole.
Orb2A88 WT and Orb2A88 F5Y
Cells were resuspended and lysed in Denaturing Buffer (10 mM Tris, pH 8.0, 8 M urea,
100 mM NaH 2PO 4 and 0.05% v/v β-mercaptoethanol) and lysis was carried out using Cell
Disruptor Sonicator (Heat Systems Model W-220F). Soluble parts were then isolated through
centrifugation at 20,000 rpm for 20 minutes using a Sorvall SS-34 rotor. The resulting supernatant
was poured into a pre-equilibrated Ni-NTA column and incubated on a shaker at room
temperature for a minimum of 1 hour. The flowthrough was subsequently collected, and the
77
column was washed with Denaturing Buffer containing 0.5% Triton X-100 followed by Denaturing
Buffer containing 500 mM NaCl. The column was also washed with Denaturing Buffer pH 6.75
and Renaturing Buffer (200 mM NaCl, pH 8.0, 50 mM NaH 2PO 4, 10% glycerol, 0.05% v/v β-
mercaptoethanol) with 20 mM imidazole. Protein was eluted with an imidazole step gradient in
Renaturing Buffer. The majority of protein eluted in the 100 mM and 150 mM imidazole fractions.
Resulting protein was either used immediately or frozen as 2 mL aliquots using liquid N 2 and
stored at -80 °C.
Fibril Forma:on

To exchange Orb2A88 and F5Y proteins into fibril forming conditions, PD10 desalting
columns (GE Healthcare, Buckinghamshire, UK) were used. The proteins were exchanged into
Renaturing Buffer containing sodium azide (0.02% final concentration) and kept on a shaker at
room temperature. Orb2A87 samples were firbrillized as described in the Orb2A87 section for
ThT assay below.
Thioflavin T (ThT) Fluorescence Assay

Orb2A87
Orb2A87 samples were prepared using GE Healthcare PD-10 desalting columns, and a 20
mM hepes, 100 mM NaCl pH 7.4 buffer. The eluted samples were pooled and then aliquoted into
three fractions of equal volume. Each fraction was aggregated in the presence or absence of the
appropriate ligand (Ni
2+
or Mg
2+
) in a 2.5:1 ligand to protein molar ratio. Samples were measured
in triplicate (200 μL fractions plus 2.5 μL of 5 mM thioflavin T stock), using an Eppendorf Plate
Reader AF2200 fluorimeter (Eppendorf, Hauppauge, NY, USA). Samples were excited at 440 nm
with bandwidth of 20 nm and emission was recorded at 484 nm with a bandwidth of 25 nm.
Fluorescence measurements were taken every 15 min, with 2 seconds shaking prior to
78
measurement, and over the course of 24 h. Collected data points were averaged and the standard
deviation was calculated.
Orb2A88 WT and Orb2A88 F5Y
Orb2A88 WT and F5Y mutant aliquots were thawed, and buffer exchanged into
Renaturing buffer pH 8.00 using PD-10 desalting columns. Concentration was measured using the
Fluorescamine assay described below. Samples of similar concentration were then combined
with thioflavin T (200 µL fractions, 2.5 µL of 5mM thioflavin T stock) and measured using the
Eppendorf Plate reader described above, using the same parameters.
Fluorescamine Protein Assay

To determine protein concentration, a fluorescamine protein assay from FluoProbes and
described in Soria et al. was used [53]. Lysozyme was used to generate standards of known
concentration and from this a concentration standard curve. Samples and lysozyme standards
were treated with 0.1 M borate and 1% sodium dodecyl sulfate (pH 9.0). They were then boiled
and allowed to cool to room temperature. Fluorescamine in acetone was added dropwise to each
with vortexing. Fluorescence was measured using a PlateReader AF2200 (Eppendorf -Hamburg,
Germany) and using a filter with an excitation wavelength of 360 nm and an emission wavelength
of 465 nm. A slit width of 35 nm was used for both excitation and emission.
Circular Dichroism

For Orb2A87
Purified Orb2A87 aliquots were buffer exchanged into 75 mM phosphate and 100 mM NaF
buffer pH 7.6 using a PD-10 desalting column (GE Healthcare, Chicago, IL, USA). Protein
concentration was determined using UV absorbance at 280 nm and calculated to be 19 μM. NiCl 2,
MgCl 2, ZnCl 2, CaCl 2, and CuCl 2 were dissolved in deionized H 2O to make 10 mM stock solutions.
79
Metal stocks were added to protein in a 2.5 molar excess of metal over protein, 0.98 μL of each
metal stock. For protein with no metals, 0.98 μL deionized H 2O was added. Circular dichroism
(CD) was measured from 260 nm to 195 nm on a Jasco J-810 spectropolarimeter (Jasco Inc.,
Easton, MD, USA) at a scan speed of 50 nm/min with data points at every 0.5 nm. Background
measurements were taken with buffer and metal or buffer and water to correlate with each
sample. Sample and background measurements were averaged over 16 scans. The background
spectra were subtracted from each respective sample spectrum and the results were plotted.
Electron Microscopy

Fibrils were adsorbed onto copper mesh electron microscopy grids (Electron Microscopy
Sciences, Hatfield, PA) for 5 min. These grids were negatively stained with 1% uranyl acetate for
2 min, rinsed with deionized water and dried. Subsequently, the grids were examined with a JEOL
JEM-1400 electron microscope (JEOL, Peabody, MA) at 100 kV and photographed using a Gatan
digital camera.

80
Appendix A: ssNMR Characterization of ⍺-synuclein Mutants and
Glycosylated ⍺-Synuclein
* The following is the result of a collaboration with Dr. Matthew Pratt’s lab (USC). All proteins were
expressed by S.A. Cervantes. Purification was done by A. Balana from the Pratt lab. All ssNMR sample
preparation, data collection and analysis was done by S.A. Cervantes

Background

Parkinson’s disease (PD), Parkinson’s disease with dementia (PDD), and dementia with
Lewy Bodies (DLB) are the second most common types of neurodegenerative dementias [75].
These disorders are characterized by the presence of intra-neuronal Lewy Bodies (LB) which are
primarily composed of aggregated ⍺-synuclein [76]. ⍺-synuclein is a protein expressed in the
central nervous system and is composed of 140 amino acids. Its structure can be divided into
three domains: N-terminus, central region, and C-terminus [75] (figure A1). In its biological role,
⍺-synuclein interacts with synaptic vesicles at the neuronal synapse, and it is believed to
participate in neurotransmitter release [77]. Although the mechanism by which ⍺-synuclein
aggregates is not fully understood, the role of post translational modifications (PTMs) and certain
gene mutations on the aggregation of the protein are well established.
81

PTMs like phosphorylation, nitration, ubiquitination, truncation, Sumoylation, and O-
GlcNAcylation have been shown to influence the stability and toxicity of the protein, in addition
to its aggregation. O-GlcNAcylation modifications, in particular, have been shown to decrease
the aggregation propensity and influence the toxicity of various amyloid proteins, including ⍺-
synuclein [78]. O-GlcNAcylation (O-GlcNAc) of ⍺-synuclein can happen at nine different positions,
including serine 87 (S87) (figure A1). Studies by Levine and co-workers have shown that
glycosylation of S87 affects the nucleation of ⍺-synuclein, thereby increasing the lag phase for
fibril formation and leading to an overall decrease in the speed of aggregation for the protein.
Glycosylation at this site has, in addition, been linked to better disease prognosis. Electron
micrograph analysis of the fibrils formed by O-GlcNAcylated S87 (gS87) showed a morphological
difference to unmodified fibrils. In addition, Proteinase K digestion revealed a possible difference
in the structural organization of gS87 fibrils as compared to that of unmodified ones, thereby
82
making glycosylation at S87 one of the few O-GlcNAc modifications to affect the conformation of
⍺-synuclein [78].
Similar studies on the effect of certain mutations on the pathogenicity of ⍺-synuclein have
revealed that while both the A53T and S87N mutations contribute to faster fibrilization of ⍺-
synuclein, the double mutant A53T S87N (TN) is less toxic to neurons and mice [79]. These same
studies also highlighted a possible structural similarity between gS87 and the TN double mutant.
Although most cases of ⍺-synuclein are sporadic, characterizing such mutations and the role they
play in disease pathogenicity can help to provide a better understanding of the mechanism of ⍺-
synuclein aggregation [80].
Having established the importance of gS87, S87N and the A53T S87N mutants, the
question that arose was, what is the structure of these glycosylated fibrils and what is the
structural difference between the S87N and TN mutants and WT ⍺-synuclein? Although protease
digestion showed a possible structural difference between these fibrils and unmodified fibrils,
the method was not sufficient to offer molecular information regarding the specific structural
changes occurring as a result of these modifications. We contributed to this need by helping to
express uniformly labeled
13
C
15
N ⍺-synuclein and then obtaining ssNMR data for unmodified
fibrils, gS87 fibrils, A53T S87N, and S87N fibrils.
Results and Discussion

The Sta:c Fibril Core of ⍺-Synuclein gS87 is Different From That of Unmodified Fibrils

What is the structure of gS87of ⍺-synuclein fibrils and how is it different from unmodified
fibrils? To answer these questions, we measured uniformly
13
C-
15
N labeled gS87 and unmodified
⍺-synuclein fibrils using solid-state NMR. Both fibril types were packed into 1.6 mm rotors and
83
measured in 20 mM phosphate buffer, pH 7.4. We recorded 1D
13
C MAS spectra as well as 2D
13
C-
13
C DREAM and PARIS spectra of gS87 fibrils and compared them to spectra for unmodified
fibrils. Figure A2-A shows 1D
13
C CP MAS spectra, which detect the static domains of the fibrils,
and which are often equivalent with their cross-β core [81]-[82], [39]. Because both cores share
roughly the same amino acid composition 1D CP spectra roughly overlap. However, since the Cα
and Cβ shifts of the aliphatic region shown here are particularly sensitive to secondary structure
[90], the imperfect overlap of a large number of peaks suggests that the core of both fibril types
is different in structure. The refocused INEPT spectrum shown in figure A2-A is, under the
conditions used, selective to highly flexible regions outside the cross-β core [39]. In these spectra
the peak positions do not change between the two fibril types, which is expected because
intrinsically disordered (aka random coil) regions often have very similar
13
C chemical shifts.
However, the intensity of the two INEPT spectra is very different indicating that the non-seeded
control fibrils have more dynamic framing sequences.

84

85
We next acquired 2D
13
C -
13
C homonuclear correlation spectra (DREAM and PARIS) for the
samples. Like our 1D
13
C CP MAS spectra, both DREAM and PARIS experiments are sensitive to
static domains in the protein, but in contrast to our 1D experiment, the added dimension allows
for the further separation of overlapping signals and therefore greater information about the
chemical shifts and coupling constants of residues. We detected the same resonances and
acquired the same information from both experiments. We show the results of the DREAM
spectra in figure A2-B. As we saw in the 1D experiments, the core of the gS87 fibrils is different
from that of WT fibrils. We can see this from the imperfect overlap of the spectra. From the 2D
comparison we can also see that the gS87 sample has better signal to noise than the unmodified
fibrils. This is evident from the fact that we are able to detect more and better-defined
resonances for gS87 sample as compared to the unmodified fibrils. This finding further adds to
our observations in the 1D experiments, in which we saw that the unmodified sample has greater
dynamics as compared to gS87.
We wanted to define the changes in the core more precisely and so we obtained ⍺-
synuclein resonance assignments from the Biological Magnetic Resonance Bank (BMRB) and used
them to assign the residues detected in our spectra. Doing so allowed us to observe that, the
differences between the fibrils are particularly pronounced in alanine and threonine residues
(highlighted in figure A2-B). We also observed the difference in the glycosylated S87 residue in
these experiments as well. The resonance for this residue is weak in the spectra of WT fibrils, but
in the spectra corresponding to gS87, the S87 residue is strong and well defined. This tells us that
the S87 region is more static in the glycosylated fibrils. This finding is also supported by the fact
86
that we detect, A85 and A90 in the spectra for gS87 and not in the spectra of WT fibrils. This again
points to the structuring and lack of mobility in this region for gS87 fibrils.
⍺-Synuclein gS87 Fibrils are Structurally Similar to Previously Reported ⍺-Synuclein Fibril
Structures

How do the structures of α-synuclein gS87 and unmodified fibrils compare to previously
reported ⍺-synuclein polymorphs? To answer this question, we again used BMRB depository to
access solid state NMR entries of wild-type, full-length ⍺-synuclein. We found entries primarily
coming from the labs of Beat Meier (ETH Zurich) and Chad Rienstra (UW-Madison). Using the
Computer Aided Resonance Assignment (CARA) software, we overlaid the deposited assignments
(BMRB entries; 17498, 18860, 25535, 16939, and 26890) on to our gS87 and unmodified fibril
13
C-
13
C DREAM and PARIS spectra.
Comparison of gS87 to BMRB entries revealed that the structure of these glycosylated
fibrils is similar to those of “ribbon” fibrils characterized by Gath and co-workers [83]. We
overlayed the BMRB entry for these fibrils (BMRB entry 17498) (shown in blue) on our gS87 fibrils
and found that they were almost a perfect match (figure A3). The good fit is further highlighted
by the zoomed in images in the figure. Here we can see that the assignments from entry 17498
overlap alanine, serine, and threonine regions in the spectrum quite well. The fact that the
overlay is such a good fit to the gS87 fibrils tells us that these two fibril types are similar in
structure.

87

Comparison of our unmodified WT ⍺-synuclein with the “fibrils” described in the same
publication by Gath, BMRB entry 18860, also showed similarities in structure. The overlay for this
BMRB assignment on our unmodified fibrils is shown in figure A4. We see from the overlay that;
the fibril similarities are particularly pronounced in the alanine and threonine regions of this
sample. The signal to noise ratio of the spectra obtained from the unmodified fibrils was worse
as compared to the gS87 fibrils, and this observation is further evident by the fact that there are
several assignment points (red crosses), for which there is no resonance detected in the spectra.
However, the good match between the residues that are detected in the spectra and the
overlayed assignments, indicates that the fibril structures compared are very similar.

88

The fact that gS87 fibrils are similar in structure to the “ribbons” described by Gath and
co-workers and that our unmodified fibrils are similar to the “fibrils” described in this same
publication was a surprise to us. However, upon analysis of the conditions used to prepare the
“ribbons” and “fibrils” described by Gath, we found that the buffer conditions are similar to those
used to generate gS87 and unmodified fibrils. Gath and co-workers used: 20 mM Tris, pH 7.5, 150
mM KCl to make “fibrils” and 5 mM tris, pH 7.5 to generate “ribbons”, while gS87 and unmodified
fibrils used in our studies were made using 50 mM tris, pH 7.5 and 150 mM NaCl. Because ⍺-
synuclein is susceptible to changes depending on the conditions used for fibrilization, it makes
sense that these similar conditions would yield similar fibril structures. One additional similarity
between the fibril samples made by Gath and the fibrils we examined is that for their study Gath
found that “ribbon” fibrils were different in morphology and toxicity to the “fibrils” they
89
generated. Overall, “ribbons” were found to be less toxic than “fibrils”. Similarly, the gS87 fibrils
and unmodified fibrils were tested in a later study, which found that gS87 fibrils are less toxic and
morphologically different than unmodified fibrils [78].
Comparison of gS87 and WT a-synuclein fibrils to the other BMRB entries further
confirmed entries 17498 and 18860 as the best fit for the glycosylated and WT fibrils respectively.
Figure A5 compares the 2D gS87
13
C PARIS correlation spectrum, shown in black, with the spectra
of the BMRB entries overlaid in color. Spectra overlays for ⍺-synuclein fibrils made in Tris-HCl
buffer conditions are shown in green and those for fibrils made in phosphate buffer conditions
are shown in orange. Although there is some resonance overlap in the overlays, the comparison
shows a clear structural difference between the modified gS87 fibrils, and the fibril types overlaid
for comparison. These structural differences are particularly evident in the serine and threonine
region of the spectra. Similarly, in the case of the unmodified ⍺-synuclein fibrils, the overlaid
assignments show that there is a structural difference between our WT fibrils and the fibrils used
for comparison (figure A6).

90

91

⍺-Synuclein S87N and A53T S87N (TN) are Similar in Structure

How is the structure of ⍺-synuclein affected by mutations? To answer this question, we
examined fibrils made from an S87N mutant and a A53T S87N (TN) double mutant. The A53T and
S87N mutations are known to accelerate fibril formation independently. In the Cryo EM structure
described by Guerrero-Ferreira and co-workers, the A53T mutation faces the interface between
92
the two protofibrils and thereby participates in providing fibril stability [84]. The S87 position is
found in the NAC domain of ⍺-synuclein. While the same study by Guerrero-Ferreira showed that
this residue faces towards the outside of the fibril, previous studies have found that this residue
is positioned towards the inside of the protofilament core. Overall, these findings highlight the
importance of these residues in the fibril core of ⍺-synuclein. Because like gS87, these
modifications can influence the fibril structure.
We started our study by looking at the structural differences between these mutants and
⍺-synuclein WT. Like with our studies on gS87, we recorded 1D
13
C MAS spectra of the static and
dynamic domains for the S87N and TN mutants. Compared to ⍺-synuclein WT, the S87N mutant
is more static, as observed by the reduced signal in the INEPT
13
C spectra (figure A7). We also
noted a difference in the core structure of this mutant as compared to WT. We can see that there
is slight peak shifting in the 1D
13
C CP spectra, which implies that the structural composition of
the fibril core is different for the two. We next compared the structure of the TN double mutant
to that of ⍺-synuclein WT. Similar to the S87N mutant, we noted a difference in the static
components of the fibrils. This difference is marked by the imperfect overlap of the 1D
13
C CP
spectra for the two samples. The peak shifting and overall changes observed for this sample, as
compared to the WT fibrils, was more pronounced (figure A8). We also noted that this sample
appeared more dynamic than the S87N mutant. We can see from the INEPT spectra, that
although not equal to the WT spectra, we do detect greater signal for this double mutant.

93

94

To compare the differences between the two mutants with greater attention, we
overlayed their spectra. We see in figure A9-A that there are some differences in the static
selective spectra (
13
C CP), thereby indicating that the fibril core structure for these two is not
identical. However, the differences between the two are less substantial than the differences
observed between mutants and WT fibrils. This is a reasonable finding, given that the two
mutants share one site in common, they both contain the S87N substitution. Therefore, the
differences we see in their spectra are most likely due to the presence of the A53T mutation. The
similarities in the static domains of the protein are more obvious when we compare the 2D
13
C-
95
13
C DREAM spectra for the samples (figure A9-B). Here we see that resonances for both samples
overlap well with minimum differences.
In the plot comparing the INEPT spectra, we see that although there is good overlap
between the samples, the A53T S87N sample is more dynamic, we detect higher signal to noise
for this sample. Previous studies have highlighted the importance of residue S87 in influencing
the behavior of ⍺-synuclein. It is well established that this site is commonly phosphorylated and
glycosylated. From our own studies, we have found that glycosylating this site leads to
pronounced changes in the fibril structure. In addition, based on the current structures of ⍺-
synuclein, others have hypothesized that modifying S87 leads to a smaller fibril core since a
modification at this site, inhibits the participation of the last 10 to 15 amino acids [78]. This
hypothesis would, therefore, partially explain why we see a change in dynamic for these mutants,
however, it does not fully explain why we only see an increase in dynamics for the double mutant
and not for the S87N mutant alone.
96

97

We next decided to compare the TN double mutant to gS87. Studies by Luk and co-
workers highlighted a possible structural similarity between TN and gS87 using protease
digestion and found that TN fibrils, similar to gS87 fibrils, are less toxic [79]. Surprisingly, we found
that the structure of these two proteins is quite different. In figure A10, we can see the imperfect
overlap in the 1D
13
C cross polarization, telling us that the core of these two constructs is
drastically different. In addition, we see that although both proteins have the same amount of
mobility as observed by the similar intensity in their INEPT spectra, their IDD domains are also
structurally distinct. We see that the overlay of the INEPT spectra is imperfect and observe peak
shifting at several locations.

98
⍺-Synuclein S87N and A53T S87N (TN) Mutants are Structurally Similar to Previously Reported
⍺-Synuclein Fibril Structures

How does the structure of these mutants compare to other structures of ⍺-synuclein? To
answer this question, we used the CARA software, and overlayed the following BMRB entries;
17498, 18860, 25535, 16939, and 26890 on our mutant
13
C -
13
C PARIS spectra. In addition, on the
TN double mutant, we overlaid BMRB entries: 17649 and 18207, which are entries for the A53T
mutant. In the overlays for the S87N mutant, found in figure A11, we can see that most
assignments are a good fit, as observed by the fact that most of the assignments fall on or are
close to resonances detected on the spectra. We also see that BMRB entry 18860 is the best fit
of the group for this mutant. Not only do most assignments sit directly on resonances, but in
addition we see that in the Threonine region (highlighted by the green arrow) these assignments
are an almost perfect match to the resonances for these residues. This entry, 18860 is the one
that corresponds to the fibrils described by Gath and co-workers and which we found was the
best match to the unmodified fibrils. The fact that this assignment shows a good fit is in
agreement with our 1D
13
C CP data which compares S87N fibrils to WT ⍺-synuclein and shows
that although different in structure, they are not drastically distinct (figure A7).
We obtained similar results for our comparison of A53T S87N (TN) fibrils with the
overlays. Again, although there was a good match between the assignments and resonances,
none was a perfect match. In addition, we also found entry 18860 to be the best match for these
mutant fibrils (figure A12). Since this mutant contains the A53T mutation we wanted to see how
an overlay of assignments for this mutation would compare. Although entry 17649 has few
assignments, they all match the TN spectra well. Similarly, entry 18027 matches this mutant well.
99
BMRB entry 18207 has more assignments, we see that the majority of them fall on resonance
peaks. We show the overlay for these entries on the TN spectrum in figure A13.

100

101

Materials and Methods

Expression

LB agarose plates containing BL21(DE3) cells transformed with ⍺-Synuclein WT and A53T
S87N (TN) and S87N mutants in a pRK172 vector were obtained from Dr. Matthew Pratt’s lab. A
single colony was then used to inoculate 20 mL LB medium flasks containing ampicillin (100
μg/mL). The starter cultures were grown at 30 ºC overnight for 14 hours and then expanded into
X8 500 mL flasks containing LB medium with Ampicillin. The cultures were grown to an optical
density OD 600 of 0.7-0.8 after which they were centrifuged at 4,000 rpm for 20 minutes using a
Sorvall SLC-6000 rotor (Thermo Fisher Scientific Inc.). The pellets were then resuspended in 1L
total of 1X wash buffer, components of which are detailed in the previous publication (Marley
2001) and centrifuged again for 20 minutes at 4,000 rpm. The resulting pellets were then
102
resuspended in 1L total of M9 minimal media (list of components in Marley 2001) and then split
into x2 500 mL fractions. The cultures were incubated for one hour at 25 ºC after which
expression was induced by adding Isopropyl 1-thio-β-d-galactopyranoside at a final
concentration of 0.5 mM. The cultures were further incubated at 25 ºC overnight for 20 hours
and then centrifuged at 4,000 rpm for 20 minutes. Pellets were stored at -80 ºC until further use.
Purification of the proteins, as well as seeding with gS87 ⍺-Synuclein for generation of the gS87
variant, was carried out by the Pratt Lab.
Solid-State NMR Sample Prepara:on

Fibril solutions were obtained from the Pratt lab and kept at -80 ºC until use, at which
point they were thawed and centrifuged at 13,500 rpm for 1 hour using an Eppendorf F45-30-11
rotor (5840R centrifuge, Eppendorf AG Hamburg, Germany). The pellets were then resuspended
in deionized water and centrifuged again using the same parameters. Phosphate buffer (20 mM
phosphate, 20 mM KCl at pH 7.4) was then used to pack fibril samples into 1.6 mm magic angle
spinning (MAS) rotors. All NMR fibril samples (gS87, unmodified WT α-synuclein, and α-synuclein
mutants) were prepared in the same manner.
Solid-State NMR Spectroscopy

All spectra were recorded on an Agilent 600 MHz solid-state NMR spectrometer using a
T3 1.6 mm probe operating at 25 kHz and 0°C. Hard pulses of 200 kHz and 100kHz were applied
on the
1
H and
13
C channels, respectively and the recycle delay was 3 s. Cross polarization (CP)
transfers were done with Hartman-Hahn match conditions at 60 kHz (
13
C) and 85 kHz (
1
H) and a
10% amplitude ramp. XiX
1
H decoupling of 140 kHz was applied during the direct and indirect
detection. 2D
13
C-
13
C PARIS spectra [98] were acquired using 10 kHz recoupling field on
1
H with
103
phase inversion every half rotor period. The mixing time was 50 ms. The spectral width was 50
kHz in both dimensions and 40 and 32 acquisitions were co-added for each indirect increment on
the gS87 and control sample, respectively.
Electron Microscopy

EM grids (150 mesh copper) were incubated on top of the samples for 5 minutes. Negative
staining was achieved by incubating the grids on 10 μl of 1% uranyl acetate solution for 2 minutes.
Excess sample and uranyl acetate stain was removed by blotting grids onto filter paper. Grids
were further washed with uranyl acetate followed by deionized water and then allowed to air
dry. All EM grid images were captured using a Gatan digital camera and a JEOL-JEM 1400
transmission electron microscope at 100 kV.

104
Appendix B: Characterization of the HetS, ⍺-Synuclein Chimer (HET-
sCT) and its Interaction with DnaJB1
Background

Neurodegenerative diseases are characterized by the accumulation of amyloid proteins.
Much of the research surrounding these disorders has, therefore, focused on characterizing the
mechanism by which these proteins misfold and aggregate as well as the structure of their
amyloid core. This is because the amyloid core usually contains the mutation that gives rise to
the pathogenic process of aggregation. In the case of HTTex1, for example, the fibril core is
composed of the polyglutamine expansion which causes Huntington’s disease. Although research
of amyloid proteins continues to focus on the structure of the amyloid core, the importance of
investigating other amyloid protein regions (N and C-terminus) has also gained importance.
Research has shown that these regions play an important role in the formation of fibrils and that,
in addition, once the fibrils are formed, they are the regions that surround the fibril core [29] and
therefore the parts of the protein that can interact with other fibrils and cellular components. In
HTTex1, the N-terminus functions to promote fibril formation, while the C-terminus deters the
aggregation of the protein, and once fibrils are formed, it is the C-terminus that protrudes out of
the fibril core and remains mobile [21]. Because these regions retain mobility and do not
assemble into any specific fold or form, they are commonly referred to as Intrinsically Disordered
Domains (IDDs).
Much like HTTex1, ⍺-synuclein is a protein composed of different domains, all of which
contribute to the end fibril structure. In ⍺-synuclein the central region, containing the NAC
domain, forms the core of the fibrils. Its N and C-terminus do not form the amyloid core but are
105
regions that have been shown to interact with many cellular components and antibodies [85],
[86]. ⍺-synuclein, unlike HTTex1, aggregates into various polymorphs depending on the
conditions used to fibrilize it. Although these polymorphs are composed of the same primary
structure, their overall organization, in terms of the fibril core, is different. Because of the
intimate relationship between the fibril core and the IDDs we wondered what effect the different
polymorphs have on the features of the IDDs. Specifically, we wondered how the different cores
influence the interaction of ⍺-synuclein IDDs with the chaperone DnaJB1.
One way we addressed this question was by attaching the C-terminus of ⍺-synuclein on
the fibril core of the HET-s prion protein (a.a. 218-289) [87]. We termed this chimer, HET-sCT, the
following is the characterization of this chimer.

Results and Discussion

The HET-sCT Chimer Expresses Well Using HET-s and ⍺-synuclein Condi:ons

We began our work with the chimer by attempting to express it. We summarize the
expression parameters tested below (table B1).

106

Starting from the same overnight cultures, the cells were expanded into fresh LB broth
and grown at 37 ºC. When it came time to add IPTG, the cultures were separated into the three
temperature groups. We decided to use both 1mM and 0.5 mM IPTG (final concentrations) to
account for the fact that, although HET-s is induced with 1 mM IPTG, ⍺-synuclein expression only
requires 0.5 mM IPTG. The expression gels, shown in figure B2, clearly show that the chimer was
expressed in all three temperature conditions (band at approximately 17 kDa). However, when
we compare the expression gels more carefully, we can see that expression at 25 ºC and 37 ºC is
more efficient than at 18 ºC. In addition to temperature, we also tested expression using different
end concentrations of IPTG, 1 mM vs 0.5 mM. We show these in figure B2 as values boxed in
yellow and values boxed in green respectively. In comparing the expression of the chimer band,
we can see that the final concentration of IPTG did not play a role in influencing expression levels.
Based on these results we decided to proceed with our HET-sCT studies using 25 ºC and
37 ºC expression and using 1 mM IPTG (final concentration).

107

HET-sCT Purifies Well Using Soluble Purifica:on Condi:ons

For purification of the HET-sCT, we tried a range of purifications, which included: inclusion
body purification - Purification I (figure B3), soluble purification - Purification II and Purification
III (figure B4), as well as a purification using denaturing conditions - Purification IV (figure B5) for
both pellets grown at 25 ºC and 37 ºC. Purification II and III differ only in the starting conditions
used to lyse the cells. In Purification II, the cells were exposed to 8 M urea during lyses, in
Purification III they were not. In all purification cases we obtained a single, clean band
corresponding to our protein of interest. The protein yield for the purifications, except for
108
Purification I – inclusion body, was also good and concentrations ranged from 63 µM to 75 µM.
Proteins were dialyzed into buffer composed of Tris and NaCl at 100 mM Tris, 150 mM NaCl, pH
8.0.
Although all purifications led to pure products and a good yield, we noticed that for the
37 ºC pellets, there was a loss of product, which we did not observe in the 25 ºC pellets. In figure
B4, we can see that in addition to the chimer (17 kDa), we also expressed a low molecular weight
protein that runs at about 12 kDa. For protein coming from the 25 ºC pellets, this 12 kDa band is
removed in the flowthrough. While in the case of the 37 ºC pellets, although most of the 12 kDa
band is also lost in the flowthrough, so is HET-sCT. This could potentially point to a problem with
the expression at 37 ºC, the fact that we lose HET-sCT in the flowthrough implies that there could
be a problem with the 6XHis tag on the protein. In addition, the fact that we see it in Purification
II, in which the protein lysis contained 8 M urea, tells us that the issue is not that the His tag is
inaccessible, because under denaturing conditions all protein regions should accessible. In all
cases we see that the remaining 12 kDa impurity is removed by the wash we incorporated.
Although, here too we see that we lose the chimer more so in the 37 ºC condition than in the 25
ºC one.
Overall, these purification efforts showed that the chimer could be obtained, and that it
purifies pure and in good yields. For fibril formation, all purified samples were set to incubate at
room temperature on a shaker after dialysis.

109

110

111
DnaJB1 Binds the HET-sCT Chimer

We next set out to investigate whether DnaJB1 would interact with the HET-sCT chimer.
We know that DnaJB1 can interact with ⍺-synuclein WT fibrils at a 1:10 molar ratio, with fibrils in
excess. Additionally, in our hands, we also saw DnaJB1 interaction with WT fibrils made in
phosphate buffer, and described by Tuttle et al. 2016, at a 1:5 molar ratio [88]. We allowed the
chimer samples to fibrilize at room temperature for approximately two weeks and then tested
them using the sedimentation assay described in the HTTex1-DnaJB1 section and using a 1:10
molar ratio, with fibrils in excess. Since the inclusion body purification did not yield any protein,
we only tested five samples: two from purification II (25 ºC, 37 ºC), two from purification III (25
ºC, 37 ºC), and one from the last purification using denaturing conditions (37 ºC). At the two week
time point in which we tested, DnaJB1 only bound to fibrils made from purification III and coming
from the 25 ºC growth. As can be seen from the absence of the chaperone band in the
supernatant samples run on the SDS Page gel (figure B6). It should be noted that we also observed
some binding of the chaperone to the 37 ºC sample purified using the same soluble conditions.
Although samples from Purification II were purified in a similar way to those of Purification III, we
did not observe chaperone binding in the earlier case. It should be noted that the difference
might be due to the temperature in which the samples were dialyzed. Samples for Purification II
were dialyzed at room temperature, and we observed aggregate formation for these samples
after only a few hours of dialysis. Samples for Purification III, on the other hand, were dialyzed at
4 ºC and we did not observe the same aggregate formation while on dialysis. This could indicate
that samples stemming from Purification II were more bundled than those from Purification III
and therefore sites on the fibrils were inaccessible for binding.
112

We further tested the 25 ºC sample from purification III at higher chaperone to fibril
ratios. We tested the interaction of chaperone and fibrils at: 1:5, 1:4, 1:3, 1:2, and 1:1 (chaperone
to fibrils) and found that DnaJB1 binds fibrils at all these ratios as observed by the missing
chaperone band in the supernatant (figure B7). This, overall, indicates that the binding of DnaJB1
to HET-sCT is stronger than that of DnaJB1 to ⍺-synuclein WT. This is an interesting finding since
it supports our initial hypothesis that the fibril core has an influence on the IDD domains. By
taking the C-terminus of ⍺-synuclein and attaching it to a better fibril former like HET-s we were
able to greatly improve the binding of DnaJB1. For WT ⍺-synuclein, even at 1:10 we did not see
full disappearance of the chaperone band from the supernatant. However, with the chimer, we
observed biding even at 1:1.
113

Since we saw partial binding of DnaJB1 to the 37 ºC, Purification III sample, we decided to
test for binding at a later time point. Again, we tested all the five samples listed above and, again,
found that DnaJB1 binds the 25 ºC sample from Purification III (figure B8). This time, however,
we also observed full binding of the 37 ºC sample from Purification III. When this sample was
previously tested, we only observed partial binding (figure B6). In addition to observing full
binding for both of the Purification III samples, this time we also saw partial binding of the
Purification II, 37 ºC sample (figure B8). We can only attribute this improvement of binding to the
fact that the samples were allowed to fibrilize for longer, thereby making them more mature and
more bundled. From our studies with DnaJB1-HTTex1, we have found that the chaperone has a
preference for slightly bundled fibrils (RT fibrils).

114

HET-sCT-DnaJB1 Samples Show the Same Addi:onal Bands Observed in the HTTex1-DnaJB1
Samples

In contrast to the sedimentation assay for DnaJB1 with ⍺-synuclein WT, we see that
DnaJB1 binding to HET-sCT leads to the appearance of the same low molecular bands we see
when testing binding of the chaperone to HTTex1. The low molecular bands run at the same
molecular weights for those observed in the HTTex1-DnaJB1 sedimentation assay, namely 25
kDa, and 12 kDa approximately. In addition, we see that, like in the case of DnaJB1-HTTex1, the
appearance of the 25 kDa and 12 kDa bands are affected by the amount of chaperone present.
For samples in which we have more chaperone present (1:1) we see a more pronounced 25 kDa
band, whereas in the samples where fibrils are in excess, we see a stronger 12 kDa band (figure
B7). As mentioned in Chapter 2 (HTTex1-DnaJB1) our current hypothesis is that these bands are
the result of fragmented chaperone, which is the result of binding the fibrils and then being
115
sequestered by its C-terminus. We confirmed the identity of the bands using Western blot assays
(figure 2.21 and 2.22). That we don’t see the same low molecular weight bands appear in the WT
⍺-synuclein sedimentation assays, although we see a decrease in the intensity of the chaperone
band in the supernatant, suggests that this difference is the result of the different core. The core
for WT ⍺-synuclein is easily depolymerized, in our own studies we have found that low
temperature and acidic conditions leads to a breakdown of the fibrils and increases the dynamics
of the protein. In contrast the HET-sCT chimer has the HET-s core, which is much more robust
than that of WT ⍺-synuclein. These fibrils can therefore have a more stable structure than WT ⍺-
synuclein. In addition to this, in our own studies with WT ⍺-synuclein and the HET-sCT chimer,
we have seen that while WT ⍺-synuclein produces clear pellets, indicating unbundled, smaller
structures, the HET-sCT chimer produces large white aggregates, which is usually indicative of a
much more bundled species.
Fibril Fragmenta:on is Observed in HET-sCT Samples Containing DnaJB1

Having screened expression, purification, and binding conditions, we set out to make a
ssNMR sample of HET-sCT fibrils alone and HET-sCT fibrils with DnaJB1. For our labeled sample,
we decided to express the protein at 25 ºC and to purify it using the soluble purification protocol
previously discussed above. Although all conditions led to good protein expression and pure
protein, we decided to go with these conditions because the fibrils formed using them were
bound the best by the chaperone. We tested the fibrils for binding at 1:10 and 1:5 molar ratios
(figure B9).
We examined the effect of the chaperone on the morphology of the HET-sCT fibrils. To
do this we looked at HET-sCT fibrils alone and in the presence of DnaJB1. The HET-sCT fibrils alone
116
sample was composed primarily of thin, expanded networks of fibrils. In some sections of the
grids, we also observed long fibrils that resembled strands (figure B10-A). The gird of HET-sCT
plus DnaJB1, on the other hand, mostly contained fibrils that were similar to long strands (figure
B10-B). In various sections of the grid, we observed breakage of the long fibril strands. This
observation is similar to what we see with HTTex1 in the presence of DnaJB1 (figure B10-B). We
know from our studies of HTTex1-DnaJB1 that this breakage happens on the grid, and although
DnaJB1 is not capable of cleaving amyloid fibrils, it is capable of contributing to weaknesses in
the fibril structure thereby making them more prone to breakage during the process of EM grid
preparation.

117

Characteriza:on of the Effect of DnaJB1 on the Structure of HET-sCT

We next looked at the effect of DnaJB1 on the structure of HET-sCT fibrils using ssNMR.
Unfortunately, our efforts to characterize the effect of the HET-s core on the ⍺-synuclein
disordered domains was impeded by the low signal to noise. In the sample for fibrils alone, we
obtained very little signal from the dynamic domains. In the sample with DnaJB1, we obtained
even less signal. However, we were able to detect the static domains of the chimer using the
cross-polarization experiment, which is sensitive to these domains. From a comparison of the
spectra, we can see that there is almost perfect overlap between the sample of fibrils alone and
118
fibrils with DnaJB1 (figure B11). This tells us that addition of the chaperone does not affect the
structure of the fibril core.
Overall, were able to show that we can express and purify the HET-sCT chimer. In addition,
we showed that it interacts with DnaJB1, and that this interaction does not affect the fibril core
of the fibrils. Future efforts will focus on investigating the identity of the additional bands we see
in the sedimentation assay as well as on optimizing the ssNMR sample for measurements.

Materials and Methods

Expression Screen

The HET-sCT chimer (pET28) was ordered from GenScript Biotech. The lyophilized DNA
was dissolved in deionized water and according to the instructions provided by the manufacturer.
119
It was then transformed into e. coli XL10 commercial cells. Purified DNA was obtained using a
Zymo Research plasmid DNA miniprep kit and sequenced. After sequence verification, the chimer
DNA was transformed into BL21(DE3) cells and plated onto kanamycin LB agar plates. Two starter
cultures (25 mL LB) were inoculated with transformed colonies and grown overnight at 30 ºC for
approximately 14 hours. One overnight culture was chosen and then expanded into X6 100 mL
cultures of LB containing kanamycin. All six cultures were then grown at 37 ºC, 225 rpm and OD
was monitored. The cultures were grown to an OD of 0.70-0.80, and then induced with IPTG
(either 1 mM or 0.5 mM final concentration). After induction, the six cultures were separated
into three temperature groups for expression. One set of cultures was placed at 37 ºC and 200
rpm, the second group was placed at 25 ºC and 200 rpm, and the third group was placed at 18 ºC
and 160 rpm. Aliquots were taken at the times specified in the table below to monitor protein
expression. The Aliquots were frozen at -20 ºC and then analyzed together using SDS page gel
electrophoresis.
Protein Expression

For a full natural abundance growth, HET-sCT pET28 was transformed into BL21(DE3)
commercial cells. For overnight cultures, X2 50 mL LB broth cultures were inoculated with one
transformed colony and kanamycin. The starter cultures were incubated overnight at 30 ºC, 200
rpm for a total of 14-16 hrs. After the overnight incubation, 1 mL of the overnight culture was
used to expand into 500 mL of LB broth containing antibiotic. The cultures were then grown at
37 ºC and 200 rpm until they reached an OD of 0.8 at which point, they were induced with IPTG
at a final concentration of 1 mM. The cultures were then separated into two groups, one for
expression at 37 ºC and 200rpm and one at 25 ºC and 200 rpm. Expression was allowed to
120
proceed for a 24-hour period. Uniformly labeled
13
C,
15
N protein was expressed in the same way
but using the M9 minimal media and protocol by Marley, et al. 2001 and cells were only
expressed at 25 ºC and 200rpm. Cells were harvested using a F9-6x1000 LEX rotor (Thermo
Scientific) run for 20 min at 4,000 rpm and 4 ºC. Cells were then stored at -80 ºC until used.
Purifica:on

Inclusion body purification
Pellets (25 ºC growth and 37 ºC growth) were thawed and resuspended in 35 mL
phosphate Denaturing buffer (100 mM NaH 2PO 4, 10 mM Tris, 8 M urea, pH 8.00). Protease
inhibitor (Pierce protease inhibitor MINI tablets, EDTA free – Thermo fisher), lysozyme (.04 g),
and b-mercaptoethanol (0.05% final) were added to the resuspended pellets. The pellets were
then incubated on a shaker at room temperature for 10 min and then sonicated on ice for a total
of 8, 30 second rounds at an amplitude of 80% (QSonic Ultrasonic Sonicator). The pellet solution
was allowed to rest for 30 seconds between sonication. The sonicated solutions were then
centrifuged at 20,000 rpm and 4 ºC for 20 minutes using an F21-8x50y rotor. The supernatant
was kept for analysis and stored at -80 ºC. The pellets were resuspended in 20 mL of Denaturing
buffer with vortexing. The resuspended pellets were then sonicated again using the same initial
parameters. The sonicated solution was then centrifuged for 15 min at 10,000 rpm and 4 ºC. The
supernatant was removed, and the pellets were resuspended as before. The process of
resuspending the pellets and sonicating them was repeated three more times, for a total of 4
times. All supernatant and pellet aliquots taken were analyzed using SDS gel electrophoresis.
Purification of soluble portion
The supernatant portions from the inclusion body purification, which were stored at -80
ºC, were thawed. Two gravity columns containing approximately 8 mL slurry solution of His60 Ni-
121
NTA resin (Takara) were equilibrated with 30 mL of buffer containing 20 mM tris, 300 mM NaCl,
and 10 mM imidazole pH 8.00. Imidazole (20 mM final) was also added to the supernatant
samples. The supernatant was then poured into the columns and incubated on a shaker at room
temperature for approximately 1 hour. The flowthrough was then collected. Each column was
washed with 50 mL of buffer composed of 20 mM tris, 50 mM imidazole, 300 mM NaCl pH 8.00.
The protein was eluted with 30 mL of a similar buffer, except that it contained 300mM imidazole.
The eluted protein was then dialyzed against 3L total of 100 mM tris, 150 mM NaCl, pH 8.00 at
room temperature.
This purification was repeated but this time the protein was not exposed to denaturing
conditions (i.e. 8 M urea in lysis buffer for the first trial). In addition, the following changes were
added; 1. The pellet after centrifugation was stored at -80 ºC, and 2. The eluted proteins were
dialyzed at 4 ºC instead of room temperature.
Denaturing conditions purification
The 37 ºC pellet from the soluble purification was thawed and resuspended in 150 mM
NaCl, 100 mM tris pH 8.00 for a pellet wash. The resuspended pellet was then centrifuged at
20,000 rpm for 20 minutes and at 4 ºC (F21-8x50y rotor). The supernatant was removed, and the
pellet was resuspended in the same buffer as before, but with 6 M GuHCl. The pellet did not
dissolve well, so it was left for incubation in the buffer overnight at 4 ºC. The pellet solution was
filtered using a 0.45 µm filter. His60 Ni-NTA resin was equilibrated with 30 mL buffer composed
of 150 mM NaCl, 100 mM tris, pH 8.00 and the filtered protein solution was poured into the
column. The column was then incubated for 1 hour at room temperature on a shaker. The column
was washed with 50 mL of 150 mM NaCl, 100 mM tris, 8 M urea, and pH 8.00 buffer. The protein
was eluted with 30 mL of a similar buffer, except that it contained 200mM imidazole. The eluted
122
proteins were dialyzed against 3L total of 100 mM tris, 150 mM NaCl, pH 8.00 at room
temperature.
Fibril forma:on

For fibrilization, all proteins were place on a shaker and were allowed to aggregate in the
dialysis buffer (100 mM tris, 150 mM NaCl, pH 8.00) and were analyzed using EM and tested for
binding using the sedimentation assay described below.
Sedimenta:on Assay

HTTex1 Q46 fibrils were incubated with DnaJB1 at a molar ratio of 4:1 (i.e. fibrils in
excess). DnaJB1 and HTTex1 alone with the equivalent amount of buffer served as controls.
Samples were incubated at room temperature on an IKA rocker 2D shaker and then centrifuged
for 1 h, 13,500 rpm and 4°C (Eppendorf 5804R) to separate the insoluble portion (fibril pellet),
containing bound DnaJB1 from the soluble portion (supernatant), composed of unbound DnaJB1.
The supernatant was then analyzed using SDS PAGE. Fibril pellets containing bound DnaJB1 were
then resuspended in buffer or deionized water alone and used to prepare EM grids as described
below.
ssNMR Spectroscopy

For measurements of Het-sCT fibrils alone and fibrils with DnaJB1, fibrils were incubated
with equal amounts of chaperone (1:1) or in buffer alone and as described above. The pellets of
fibrils alone or fibrils with bound chaperone were then isolated and packed into 1.6 mm magic
angle spinning (MAS) rotors using a home-built centrifuge tool. NMR measurements were done
on an Agilent DD2 600 MHz solid-state NMR spectrometer using a triple-resonance, 1.6 mm
probe set at 25 kHz (MAS). A temperature of 0°C was used for all experiments. Hard pulses of
123
200 kHz (
1
H), 100 kHz (
13
C), and 50 kHz (
15
N) were used. For the 1D
13
C CP experiments, the
Hartman-Hahn match at 60 kHz (
13
C) and 85 kHz (
1
H) and 140 kHz XiX decoupling during detection
was used. 2D PARIS experiments started with the same CP followed by a 50 ms long mixing time
using a
1
H recoupling field of 10 kHz and the N = 0.5 condition [98]. The spectral widths were 50
kHz in both dimensions and 500 indirect, complex increments were detected using 32
acquisitions each. The 2D
13
C-
13
C INEPT TOBSY, 2D
1
H-
13
C INEPT HETCOR, 2D
1
H-
15
N HSQC, and
3D HNCA spectra were recorded as described previously [27]-[28].
Electron Microscopy

EM grids (150 mesh copper) were incubated on top of the samples for 5 minutes. Negative
staining was achieved by incubating the grids on 10 μl of 1% uranyl acetate solution for 2 minutes.
Excess sample and uranyl acetate stain was removed by blotting grids onto filter paper. Grids
were further washed with uranyl acetate followed by deionized water and then allowed to air
dry. All EM grid images were captured using a Gatan digital camera and a JEOL-JEM 1400
transmission electron microscope at 100 kV.

124
References
[1] R. N. Rambaran, L. C. Serpell, Prion 2008, 2 (3), 112, doi.org/10.4161/pri.2.3.7488.

[2] J. Greenwald, R. Riek, Structure (London) 2010, 18 (10), 1244, doi.org/10.1016/j.str.2010.08.009.

[3] K. F. Winklhofer, J. Tatzelt, C. Haass, The EMBO journal 2008, 27 (2), 336,
doi.org/10.1038/sj.emboj.7601930.

[4] Z. L. Almeida, R. M. M. Brito, Molecules (Basel, Switzerland) 2020, 25 (5), 1195,
doi.org/10.3390/molecules25051195.

[5] G. S. Bloom, JAMA neurology 2014, 71 (4), 505, doi.org/10.1001/jamaneurol.2013.5847.

[6] D. Otzen, R. Riek, Cold Spring Harbor perspectives in biology 2019, 11 (12), a033860,
doi.org/10.1101/cshperspect.a033860.

[7] A. B. Siemer, Critical Reviews in Biochemistry and Molecular Biology 2022, 57 (4), 399,
doi.org/10.1080/10409238.2022.2113030.

[8] S. K. Maji, M. H. Perrin, M. R. Sawaya, S. Jessberger, K. Vadodaria, R. A. Rissman, P. S. Singru, K. P.
R. Nilsson, R. Simon, D. Schubert, D. Eisenberg, J. Rivier, P. Sawchenko, W. Vale, R. Riek, Science
(American Association for the Advancement of Science) 2009, 325 (5938), 328,
doi.org/10.1126/science.1173155.

[9] M. R. Chapman, L. S. Robinson, J. S. Pinkner, R. Roth, J. Heuser, M. Hammar, S. Normark, S. J.
Hultgren, Science (American Association for the Advancement of Science) 2002, 295 (5556), 851,
doi.org/10.1126/science.1067484.

[10] K. Si, M. Giustetto, A. Etkin, R. Hsu, A. M. Janisiewicz, M. C. Miniaci, J.-H. Kim, H. Zhu, E. R. Kandel,
Cell 2003, 115 (7), 893, doi.org/10.1016/S0092-8674(03)01021-3.

[11] M. Biancalana, S. Koide, BIOCHIMICA ET BIOPHYSICA ACTA-PROTEINS AND PROTEOMICS 2010,
1804 (7), 1405, doi.org/10.1016/j.bbapap.2010.04.001.

[12] M. DiFiglia, E. Sapp, K. O. Chase, S. W. Davies, G. P. Bates, J. P. Vonsattel, N. Aronin, Science
(American Association for the Advancement of Science) 1997, 277 (5334), 1990,
doi.org/10.1126/science.277.5334.1990.

[13] K. Sathasivam, A. Neueder, T. A. Gipson, C. Landles, A. C. Benjamin, M. K. Bondulich, D. L. Smith,
R. L. M. Faull, R. A. C. Roos, D. Howland, P. J. Detloff, D. E. Housman, G. P. Bates, Proceedings of
the National Academy of Sciences - PNAS 2013, 110 (6), 2366, doi.org/10.1073/pnas.1221891110.

[14] F. O. P. Walker, The Lancet (British edition) 2007, 369 (9557), 218, doi.org/10.1016/S0140-
6736(07)60111-1.

125
[15] J. M. Lee, E. M. Ramos, J. H. Lee, T. Gillis, J. S. Mysore, M. R. Hayden, S. C. Warby, P. Morrison, M.
Nance, C. A. Ross, R. L. Margolis, F. Squitieri, S. Orobello, S. Di Donato, E. Gomez-Tortosa, C. Ayuso,
O. Suchowersky, R. J. A. Trent, E. McCusker, A. Novelletto, M. Frontali, R. Jones, T. Ashizawa, S.
Frank, M. H. Saint-Hilaire, S. M. Hersch, H. D. Rosas, D. Lucente, M. B. Harrison, A. Zanko, R. K.
Abramson, K. Marder, J. Sequeiros, J. S. Paulsen, G. B. Landwehrmeyer, R. H. Myers, M. E.
MacDonald, J. F. Gusella, Neurology 2012, 78 (10), 690,
doi.org/10.1212/WNL.0b013e318249f683.

[16] P. Lajoie, E. L. Snapp, PloS one 2010, 5 (12), e15245, doi.org/10.1371/journal.pone.0015245.

[17] L. G. Nucifora, K. A. Burke, X. Feng, N. Arbez, S. Zhu, J. Miller, G. Yang, T. Ratovitski, M. Delannoy,
P. J. Muchowski, S. Finkbeiner, J. Legleiter, C. A. Ross, M. A. Poirier, The Journal of biological
chemistry 2012, 287 (19), 16017, doi.org/10.1074/jbc.M111.252577.

[18] Y. Nagai, T. Toda, T. Inui, H. A. Popiel, N. Fujikake, K. Hasegawa, Y. Urade, Y. Goto, H. Naiki, Nature
structural & molecular biology 2007, 14 (4), 332, doi.org/10.1038/nsmb1215.

[19] M. Baias, P. E. S. Smith, K. Shen, L. A. Joachimiak, S. Żerko, W. Koźmiński, J. Frydman, L. Frydman,
Journal of the American Chemical Society 2017, 139 (3), 1168, doi.org/10.1021/jacs.6b10893.

[20] G. Hoffner, P. Djian, Brain sciences 2014, 4 (1), 91, doi.org/10.3390/brainsci4010091.

[21] J. M. Isas, R. Langen, A. B. Siemer, Biochemistry (Easton) 2015, 54 (25), 3942,
doi.org/10.1021/acs.biochem.5b00281.

[22] C. W. Bugg, J. M. Isas, T. Fischer, P. H. Patterson, R. Langen, The Journal of biological chemistry
2012, 287 (38), 31739, doi.org/10.1074/jbc.M112.353839.

[23] A. Bhattacharyya, A. K. Thakur, V. M. Chellgren, G. Thiagarajan, A. D. Williams, B. W. Chellgren, T.
P. Creamer, R. Wetzel, Journal of molecular biology 2006, 355 (3), 524,
doi.org/10.1016/j.jmb.2005.10.053; M. F. Perutz, T. Johnson, M. Suzuki, J. T. Finch, Proceedings
of the National Academy of Sciences - PNAS 1994, 91 (12), 5355,
doi.org/10.1073/pnas.91.12.5355.

[24] L. Pieri, K. Madiona, L. Bousset, R. Melki, Biophysical journal 2012, 102 (12), 2894,
doi.org/10.1016/j.bpj.2012.04.050.

[25] W. C. Duim, Y. Jiang, K. Shen, J. Frydman, W. E. Moerner, ACS chemical biology 2014, 9 (12), 2767,
doi.org/10.1021/cb500335w.

[26] K.-Y. Liu, Y.-C. Shyu, B. A. Barbaro, Y.-T. Lin, Y. Chern, L. M. Thompson, C.-K. J. Shen, J. L. Marsh,
Human molecular genetics 2015, 24 (6), 1602, doi.org/10.1093/hmg/ddu574.

[27] B. G. Caulkins, S. A. Cervantes, J. M. Isas, A. B. Siemer, The journal of physical chemistry. B 2018,
122 (41), 9507, doi.org/10.1021/acs.jpcb.8b09213.

[28] A. S. Falk, A. B. Siemer, Journal of biomolecular NMR 2016, 66 (3), 159, doi.org/10.1007/s10858-
016-0069-2.
126
[29] J. Mario Isas, N. K. Pandey, H. Xu, K. Teranishi, A. K. Okada, E. K. Fultz, A. Rawat, A. Applebaum, F.
Meier, J. Chen, R. Langen, A. B. Siemer, Nature communications 2021, 12 (1), 4272,
doi.org/10.1038/s41467-021-24411-2.

[30] A. H. Erbse, J. N. Wagner, K. N. Truscott, S. K. Spall, J. Kirstein, K. Zeth, K. Turgay, A. Mogk, B.
Bukau, D. A. Dougan, The FEBS journal 2008, 275 (7), 1400, doi.org/10.1111/j.1742-
4658.2008.06304.x.

[31] A. Scior, K. Juenemann, J. Kirstein, Essays in biochemistry 2016, 60 (2), 153,
doi.org/10.1042/EBC20160002.

[32] E. H. Koo, P. T. Lansbury, J. W. Kelly, Proceedings of the National Academy of Sciences - PNAS 1999,
96 (18), 9989, doi.org/10.1073/pnas.96.18.9989.

[33] A. Ahmad, A. Bhattacharya, R. A. McDonald, M. Cordes, B. Ellington, E. B. Bertelsen, E. R. P.
Zuiderweg, Proceedings of the National Academy of Sciences - PNAS 2011, 108 (47), 18966,
doi.org/10.1073/pnas.1111220108.

[34] A. Scior, A. Buntru, K. Arnsburg, A. Ast, M. Iburg, K. Juenemann, M. L. Pigazzini, B. Mlody, D.
Puchkov, J. Priller, E. E. Wanker, A. Prigione, J. Kirstein, The EMBO journal 2018, 37 (2), 282,
doi.org/10.15252/embj.201797212.

[35] S. M. Ayala Mariscal, M. L. Pigazzini, Y. Richter, M. Özel, I. L. Grothaus, J. Protze, K. Ziege, M. Kulke,
M. ElBediwi, J. V. Vermaas, L. Colombi Ciacchi, S. Köppen, F. Liu, J. Kirstein, Nature
communications 2022, 13 (1), 4692, doi.org/10.1038/s41467-022-32370-5.

[36] X. Gao, M. Carroni, C. Nussbaum-Krammer, A. Mogk, Nadinath B. Nillegoda, A. Szlachcic, D. L.
Guilbride, Helen R. Saibil, Matthias P. Mayer, B. Bukau, Molecular cell 2015, 59 (5), 781,
doi.org/10.1016/j.molcel.2015.07.012.

[37] J. B. Warner, K. M. Ruff, P. S. Tan, E. A. Lemke, R. V. Pappu, H. A. Lashuel, Journal of the American
Chemical Society 2017, 139 (41), 14456, doi.org/10.1021/jacs.7b06659.

[38] Y. Nekooki-Machida, M. Kurosawa, N. Nukina, K. Ito, T. Oda, M. Tanaka, Proceedings of the
National Academy of Sciences - PNAS 2009, 106 (24), 9679, doi.org/10.1073/pnas.0812083106.

[39] A. B. Siemer, Solid state nuclear magnetic resonance 2020, 106, 101643,
doi.org/10.1016/j.ssnmr.2020.101643.

[40] C. L. Hoop, H.-K. Lin, K. Kar, G. Magyarfalvi, J. M. Lamley, J. C. Boatz, A. Mandal, J. R. Lewandowski,
R. Wetzel, P. C. A. van der Wel, Proceedings of the National Academy of Sciences - PNAS 2016, 113
(6), 1546, doi.org/10.1073/pnas.1521933113.

[41] H.-K. Lin, J. C. Boatz, I. E. Krabbendam, R. Kodali, Z. Hou, R. Wetzel, A. M. Dolga, M. A. Poirier, P.
C. A. van der Wel, Nature communications 2017, 8 (1), 15462, doi.org/10.1038/ncomms15462.

127
[42] F. J. B. Bäuerlein, I. Saha, A. Mishra, M. Kalemanov, A. Martínez-Sánchez, R. Klein, I. Dudanova, M.
S. Hipp, F. U. Hartl, W. Baumeister, R. Fernández-Busnadiego, Cell 2017, 171 (1), 179,
doi.org/10.1016/j.cell.2017.08.009.

[43] H. Yang, H. Y. Hu, The FEBS journal 2016, 283 (20), 3705, doi.org/10.1111/febs.13722.

[44] A. Casadio, K. C. Martin, M. Giustetto, H. Zhu, M. Chen, D. Bartsch, C. H. Bailey, E. R. Kandel, Cell
1999, 99 (2), 221, doi.org/10.1016/S0092-8674(00)81653-0.

[45] P. Goelet, V. F. Castellucci, S. Schacher, E. R. Kandel, Nature (London) 1986, 322 (6078), 419,
doi.org/10.1038/322419a0.

[46] L. L. McGrew, E. Dworkin-Rastl, M. B. Dworkin, J. D. Richter, Genes & development 1989, 3 (6),
803, doi.org/10.1101/gad.3.6.803.

[47] K. Si, S. Lindquist, E. R. Kandel, Cell 2003, 115 (7), 879, doi.org/10.1016/S0092-8674(03)01020-1.

[48] W.-H. Lu, N.-H. Yeh, Y.-S. Huang, Cell reports (Cambridge) 2017, 21 (7), 1783,
doi.org/10.1016/j.celrep.2017.10.073.

[49] K. Keleman, B. J. Dickson, S. Krüttner, M. Alenius, Nature neuroscience 2007, 10 (12), 1587,
doi.org/10.1038/nn1996.

[50] A. Majumdar, Wanda C. Cesario, E. White-Grindley, H. Jiang, F. Ren, M. R. Khan, L. Li, Edward M.-
L. Choi, K. Kannan, F. Guo, J. Unruh, B. Slaughter, K. Si, Cell 2012, 148 (3), 515,
doi.org/10.1016/j.cell.2012.01.004.

[51] Mohammed R. Khan, L. Li, C. Pérez-Sánchez, A. Saraf, L. Florens, Brian D. Slaughter, Jay R. Unruh,
K. Si, Cell 2015, 163 (6), 1468, doi.org/10.1016/j.cell.2015.11.020.

[52] S. A. Cervantes, T. H. Bajakian, M. A. Soria, A. S. Falk, R. J. Service, R. Langen, A. B. Siemer, Scientific
reports 2016, 6 (1), 38265, doi.org/10.1038/srep38265.

[53] M. A. Soria, S. A. Cervantes, T. H. Bajakian, A. B. Siemer, Biophysical journal 2017, 113 (1), 37,
doi.org/10.1016/j.bpj.2017.05.039.

[54] Y.-B. Zeng, N. Yang, H. Sun, Chemistry : a European journal 2011, 17 (21), 5852,
doi.org/10.1002/chem.201100279.

[55] N. M. Chiera, M. Rowinska-Zyrek, R. Wieczorek, R. Guerrini, D. Witkowska, M. Remelli, H.
Kozlowski, Metallomics 2013, 5 (3), 214, doi.org/10.1039/c3mt20166j.

[56] L. Hong, J. D. Simon, Metallomics 2011, 3 (3), 262, doi.org/10.1039/c0mt00052c.

[57] V. Tõugu, A. Tiiman, P. Palumaa, Metallomics 2011, 3 (3), 250, doi.org/10.1039/c0mt00073f.

128
[58] J. Volz, F. Uwe Bosch, M. Wunderlin, M. Schuhmacher, K. Melchers, K. Bensch, W. Steinhilber, K.
P. Schäfer, G. Tóth, B. Penke, M. Przybylski, JOURNAL OF CHROMATOGRAPHY A 1998, 800 (1), 29,
doi.org/10.1016/S0021-9673(97)00877-7.

[59] F. H. Arnold, Bio/technology (New York, N.Y. 1983) 1991, 9 (2), 151, doi.org/10.1038/nbt0291-
151.

[60] E. K. M. Ueda, P. W. Gout, L. Morganti, JOURNAL OF CHROMATOGRAPHY A 2003, 988 (1), 1,
doi.org/10.1016/S0021-9673(02)02057-5.

[61] H. LeVine, Amyloid 1995, 2 (1), 1, doi.org/10.3109/13506129509031881.

[62] M. Groenning, L. Olsen, M. van de Weert, J. M. Flink, S. Frokjaer, F. S. Jørgensen, Journal of
structural biology 2007, 158 (3), 358, doi.org/10.1016/j.jsb.2006.12.010.

[63] S. Ghosh, N. K. Pandey, A. Singha Roy, D. R. Tripathy, A. K. Dinda, S. Dasgupta, PloS one 2013, 8
(9), e74336, doi.org/10.1371/journal.pone.0074336.

[64] A. L. Jones, M. D. Hulett, C. R. Parish, Immunology and cell biology 2005, 83 (2), 106,
doi.org/10.1111/j.1440-1711.2005.01320.x.

[65] A. Battistoni, F. Pacello, A. P. Mazzetti, C. Capo, J. S. Kroll, P. R. Langford, A. Sansone, G.
Donnarumma, P. Valenti, G. Rotilio, The Journal of biological chemistry 2001, 276 (32), 30315,
doi.org/10.1074/jbc.M010527200.

[66] M. Ranieri-Raggi, A. J. G. Moir, A. Raggi, Biomolecules 2014, 4 (2), 474,
doi.org/10.3390/biom4020474.

[67] M. H. Schaefer, E. E. Wanker, M. A. Andrade-Navarro, Nucleic acids research 2012, 40 (10), 4273,
doi.org/10.1093/nar/gks011.

[68] S. J. B. Fretham, E. S. Carlson, M. K. Georgieff, Advances in nutrition (Bethesda, Md.) 2011, 2 (2),
112, doi.org/10.3945/an.110.000190.

[69] M. Han, J. Kim, PloS one 2015, 10 (3), e0120609, doi.org/10.1371/journal.pone.0120609.

[70] A. Takeda, H. Tamano, Biometals 2016, 29 (2), 177, doi.org/10.1007/s10534-015-9907-z.

[71] E. P. Huang, Proceedings of the National Academy of Sciences - PNAS 1997, 94 (25), 13386,
doi.org/10.1073/pnas.94.25.13386.

[72] Y. Li, C. J. Hough, C. J. Frederickson, J. M. Sarvey, The Journal of neuroscience 2001, 21 (20), 8015,
doi.org/10.1523/JNEUROSCI.21-20-08015.2001.

[73] A. Takeda, International Journal of Alzheimer＇s Disease 2011, 2011, 491597,
doi.org/10.4061/2011/491597.

129
[74] R. Hervas, M. J. Rau, Y. Park, W. Zhang, A. G. Murzin, J. A. J. Fitzpatrick, S. H. W. Scheres, K. Si,
Science (American Association for the Advancement of Science) 2020, 367 (6483), 1230,
doi.org/10.1126/science.aba3526.

[75] N. d. O. Manzanza, L. Sedlackova, R. N. Kalaria, Frontiers in aging neuroscience 2021, 13, 690293,
doi.org/10.3389/fnagi.2021.690293.

[76] M. G. Spillantini, M. L. Schmidt, V. M. Y. Lee, J. Q. Trojanowski, R. Jakes, M. Goedert, Nature
(London) 1997, 388 (6645), 839, doi.org/10.1038/42166.

[77] L. Maroteaux, J. T. Campanelli, R. H. Scheller, The Journal of neuroscience 1988, 8 (8), 2804,
doi.org/10.1523/jneurosci.08-08-02804.1988.

[78] P. M. Levine, A. Galesic, A. T. Balana, A.-L. Mahul-Mellier, M. X. Navarro, C. A. De Leon, H. A.
Lashuel, M. R. Pratt, Proceedings of the National Academy of Sciences - PNAS 2019, 116 (5), 1511,
doi.org/10.1073/pnas.1808845116.

[79] K. C. Luk, D. J. Covell, V. M. Kehm, B. Zhang, I. Y. Song, M. D. Byrne, R. M. Pitkin, S. C. Decker, J. Q.
Trojanowski, V. M. Y. Lee, Cell reports (Cambridge) 2016, 16 (12), 3373,
doi.org/10.1016/j.celrep.2016.08.053.

[80] P. T. Lansbury, K. A. Conway, J. D. Harper, Nature medicine 1998, 4 (11), 1318,
doi.org/10.1038/3311.

[81] H. Heise, K. Seidel, M. Etzkorn, S. Becker, M. Baldus, Journal of magnetic resonance (1997) 2005,
173 (1), 64, doi.org/10.1016/j.jmr.2004.11.020.

[82] A. B. Siemer, C. Ritter, M. O. Steinmetz, M. Ernst, R. Riek, B. H. Meier, Journal of biomolecular
NMR 2006, 34 (2), 75, doi.org/10.1007/s10858-005-5582-7.

[83] J. Gath, L. Bousset, B. Habenstein, R. Melki, A. Böckmann, B. H. Meier, PloS one 2014, 9 (3),
e90659, doi.org/10.1371/journal.pone.0090659.

[84] R. Guerrero-Ferreira, N. M. I. Taylor, A.-A. Arteni, P. Kumari, D. Mona, P. Ringler, M. Britschgi, M.
E. Lauer, A. Makky, J. Verasdock, R. Riek, R. Melki, B. H. Meier, A. Böckmann, L. Bousset, H.
Stahlberg, eLife 2019, 8, e48907, doi.org/10.7554/eLife.48907.

[85] E. Monsellier, V. Redeker, G. Ruiz-Arlandis, L. Bousset, R. Melki, The Journal of biological chemistry
2015, 290 (5), 2560, doi.org/10.1074/jbc.M114.603332.

[86] S. H. Shahmoradian, J. G. Galaz-Montoya, M. F. Schmid, Y. Cong, B. Ma, C. Spiess, J. Frydman, S. J.
Ludtke, W. Chiu, eLife 2013, 2013 (2), e00710, doi.org/10.7554/eLife.00710.

[87] C. Wasmer, A. Lange, H. Van Melckebeke, A. B. Siemer, R. Riek, B. H. Meier, Science (American
Association for the Advancement of Science) 2008, 319 (5869), 1523,
doi.org/10.1126/science.1151839.

130
[88] M. D. Tuttle, G. Comellas, A. J. Nieuwkoop, D. J. Covell, D. A. Berthold, K. D. Kloepper, J. M.
Courtney, J. K. Kim, A. M. Barclay, A. Kendall, W. Wan, G. Stubbs, C. D. Schwieters, V. M. Y. Lee, J.
M. George, C. M. Rienstra, Nature structural & molecular biology 2016, 23 (5), 409,
doi.org/10.1038/nsmb.3194.

[89] J. M. Isas, A. Langen, M.C. Isas, N. K. Pandey, A. B. Siemer, Biochemistry 2017, 56 (28), 3579-3586,
doi.org/10.1021/acs.biochem.7b00138.

[90] S. Spera and A. Bax, Journal of the American Chemical Society 1991, 113 (14), 5490-5492,
doi.org/10.1021/ja00014a071.

[91] R. Diaz-Espinoza, Frontiers in Aging Neuorscience 2021, 19 (13), 782617,
doi.org/10.3389/fnagi.2021.782617.

[92] S.A. Newcombe, K.M. Ruff, A. Sethi, A.R. Ormsby, Y.M. Ramdzan, A. Fox, A.W. Purcell, P.R. Gooley,
R.V. Pappu, D.M. Hatters, Journal of Molecular Biology 2018, 430 (10), 1442-1458,
doi.org/10.1016/j.jmb.2018.03.031.

[93] J.M. Bravo-Arredondo, N.C. Kegulian, T. Schmidt, N.K. Pandey, A.J. Situ, T.S. Ulmer, R. Langen,
Journal of Bilogical Chemistry 2018, 293 (51), 19613-19623, doi.org/10.1074/jbc.RA118.004808.

[94] M. Baias, P.E. Shen, K. JOachimiak, L.A. Zerko, S. Kozaminski, J. Frydman, L. Frydman, Journal of
the American Chemical Society 2017, 139 (3), 1168-1176, doi.org/10.1021/jacs.6b10893.

[95] G. Jain, M. Trombetta-Lima, I. Matlahov, H.T. Ribas, A.M. Dolga, P. Van der Wel, bioRxiv 2023,
p.2023.04.24.537565, doi.org/10.1101/2023.04.24.537565.

[96] J. Marley, M. Lu, C.A. Bracken, Journal of Biomolecular NMR 2001, 20 (1),
doi.org/10.1023/a:1011254402785.

[97] A. Detken, E.H Hardy, M. Ernst, B.H. Meier, Chemical Physical Letters 2002, 356 (3-4), 298-304,
doi.org/10.1016/S0009-2614(02)00335-4.

[98] M. Weingarth, D.E. Demco, G. Bodenhausen, P. Tekely, Chemical Physical Letters 2009, 469 (4-6),
342-348, doi.org/10.1016/j.cplett.2008.12.084.

[99] L.Y Tsai, Y.W. Chang, P.Y. Lin, H.J. Chou, T.J. Liu, P.T. Lee, W.H. Huang, Y.L. Tsou, Y.S. Huang, PLoS
One 2013, 30 (12), doi.org/10.1371/journal.pone.0084978.

Abstract (if available)

Abstract Amyloid fibrils are proteinaceous deposits that are insoluble, resistant to degradation and are rich in beta-sheet structure. Amyloid fibrils are composed of beta-strands that run perpendicular to the fibril axis, thereby forming a cross-beta sheet of potentially indefinite length. Although this cross-beta motif is the fundamental structure of amyloid fibrils, amyloid fibril structures are elaborate and diverse, and can give rise to an abundance of functions. In some cases, these functions are deleterious in nature, like in the case of neurodegenerative disorders, which are characterized by the accumulation of misfolded protein that aggregates into amyloid fibrils.
In this work, we investigate two such amyloid proteins, Huntingtin, which causes Huntington’s Disease and alpha-synuclein, which is associated with Parkinson’s disease. For Huntingtin, we investigate its mechanism of aggregation by characterizing the structure of its monomeric state using solid state NMR. In addition, we provide a better understanding of its mechanism of disaggregation and overall toxicity on the cell, by characterizing the interaction of Huntingtin fibrils with the mammalian chaperone DnaJB1. Chaperones are known promoters of protein homeostasis and decrease amyloid protein toxicity through their work in identifying, folding and re-folding misfolded protein. Using various biophysical tools, including solid state NMR we identify the binding site of DnaJB1 on Huntingtin exon 1 fibrils. For alpha-synuclein, there are known post translational modifications and mutations that influence the proteins’ aggregation and toxicity. In this work, we characterize the effect of glycosylation on Serine 87 (gS87), and the effect of S87N and S87N A53T (TN) mutations on the structure of these fibrils.
While in the case of neurodegenerative disorders and other amyloid diseases, amyloid formation results from a deleterious misfolding event that affects the protein and causes it to further misfold and aggregate and ultimately become toxic to cells. In contrast, in the case of functional amyloids, their formation is a tightly regulated process. In this context, amyloid formation serves a purpose, is beneficial to cells, and is ultimately necessary for host survival. In this work we characterize the functional amyloid Orb2, a member of the Cytoplasmic Polyadenylation Element Binding protein family which is associated with memory formation and maintenance. Specifically, we investigate the first 88 amino acids of the isoform Orb2A (Orb2A88) and characterize its metal binding properties and the effect these have on its aggregation, as well as the effect of a point mutation (F5Y) on the fibrils formed by Orb2A88.

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

Structure and kinetics of the Orb2 functional amyloid

PDF

Structural characterization of the functional amyloid Orb2A using EPR spectroscopy

PDF

Regulating functional amyloid formation: the promiscuous behavior of the Orb2A N-terminal amphipathic region

PDF

Before they were amyloid: understanding the toxicity of disease-associated monomers and oligomers prior to their aggregation

PDF

Interaction of monoclonal antibodies MW1 and PHP1 with huntingtin exon1 protein

PDF

Overexpression and interaction of Orb2 in S2 insect cells

PDF

Oligomer formation of functional amyloid protein - Orb2A

PDF

Glycine to alanine mutations affect the structure and dynamics of micelle bound alpha-synuclein

PDF

Facilitating unambiguous NMR assignment by solid–state NMR using segmental isotope labeling through split-inteins

PDF

Molecular basis of CD33 receptor function, and effects of a destabilizing transmembrane motif on αXβ2 integrin

PDF

The structure and function of membrane curving proteins on different membrane shapes and their regulation by post-translational modifications

PDF

Role of CD33 structure and function in Alzheimer’s disease

PDF

Role of mitochondrially derived peptides in the inhibition of IAPP misfolding

PDF

Huntington’s disease vaccine development using fungal prion HET-s

PDF

Enhancing and inhibiting diabetic amyloid misfolding

PDF

Uncovering the influence of N-terminal phosphorylation on conformational dynamics of huntingtin exon 1 monomer

PDF

Study of integrin αIIbβ3 transmembrane and cytosolic domain interaction and the expression of extracellular domain by Pichia Pastoris

PDF

Dual effects of transmembrane proline residues on integrin function

PDF

Structural features and modifiers of islet amyloid polypeptide: implications for type II diabetes mellitus

PDF

Structural studies of the IAPP membrane-mediated aggregation pathway

Asset Metadata

Creator Cervantes Cortes, Silvia Angelica (author)

Core Title Structural studies on functional amyloids and the mechanism of aggregation and disaggregation of Huntingtin Exon 1

School Keck School of Medicine

Degree Doctor of Philosophy

Degree Program Medical Biophysics

Degree Conferral Date 2023-08

Publication Date 08/09/2023

Defense Date 07/14/2023

Publisher University of Southern California. Libraries (digital)

Tag alpha-synuclein,amyloid fibrils,amyloid proteins,amyloids,huntingtin,OAI-PMH Harvest,Orb2A,solid state NMR

Language English

Contributor Electronically uploaded by the author (provenance)

Advisor Langen, Ralf (committee chair), Siemer, Ansgar (committee member), Tobias, Ulmer (committee member)

Creator Email angie.crvts@gmail.com,silviace@usc.edu

Permanent Link (DOI) https://doi.org/10.25549/usctheses-oUC113297696

Unique identifier UC113297696

Identifier etd-CervantesC-12228.pdf (filename)

Legacy Identifier etd-CervantesC-12228

Document Type Dissertation

Rights Cervantes Cortes, Silvia Angelica

Internet Media Type application/pdf

Type texts

Source 20230809-usctheses-batch-1082 (batch), University of Southern California (contributing entity), University of Southern California Dissertations and Theses (collection)

Access Conditions The author retains rights to his/her dissertation, thesis or other graduate work according to U.S. copyright law. Electronic access is being provided by the USC Libraries in agreement with the author, as the original true and official version of the work, but does not grant the reader permission to use the work if the desired use is covered by copyright. It is the author, as rights holder, who must provide use permission if such use is covered by copyright. The original signature page accompanying the original submission of the work to the USC Libraries is retained by the USC Libraries and a copy of it may be obtained by authorized requesters contacting the repository e-mail address given.

Repository Name University of Southern California Digital Library

Repository Location USC Digital Library, University of Southern California, University Park Campus, Los Angeles, California 90089, USA

Repository Email cisadmin@lib.usc.edu