Conformational Transitions of 18.5-Kilodaltons Myelin Basic

Conformational Transitions of 18.5-Kilodaltons Myelin Basic Protein
studied by Fluorescence Spectroscopy and Förster Resonance Energy
Transfer
by
Andrew Daniel Jenkins
A Thesis
presented to
The University of Guelph
In partial fulfilment of requirements
for the degree of
Master of Science
in
Molecular and Cellular Biology
Guelph, Ontario, Canada
© Andrew D. Jenkins, April, 2015
ABSTRACT
Conformational Transitions of 18.5-Kilodaltons Myelin Basic Protein
studied by Fluorescence Spectroscopy and Förster Resonance Energy
Transfer
Andrew Daniel Jenkins
University of Guelph, 2015
Advisor:
Professor G. Harauz
The intrinsically-disordered, 18.5-kiloDalton (kDa) isoform of myelin basic protein
(MBP) is a peripheral membrane protein that is essential to proper myelin formation in the
central nervous system. MBP multifunctionality arises from its high conformational plasticity
and its ability to undergo reversible disorder-to-order transitions. One such transition is the
disorder-to-α-helical conformational change which is induced upon MBP-membrane binding.
We have investigated the disorder-to-α-helical transition of MBP-derived α-peptides as well as
of the full-length 18.5-kDa protein. The data suggest that the disorder-to-α-helical transition of
MBP follows a three-state model: disordered↔intermediate↔α-helical, with each of the
identified equilibrium states likely representing a conformational ensemble. The disordered state
is characterized by slight compaction, whereas the intermediate globally more compact. This
study suggests that multifunctionality in MBP could arise from differences in the population of
energetically distinct ensemble clusters in different conditions and also provides an example of
an IDP that undergoes cooperative global conformation change.
ACKNOWLEDGEMENTS
This master's degree was really the collective doing of an extended group of individuals
who supported me throughout the years.
First and foremost I would like to thank my graduate supervisor, Dr. George Harauz, who
afforded me the opportunity to work, learn and grow as scientist in his lab. I am truly grateful for
your patience and guidance.
I would also like to thank and my advisory committee members, Dr. Rod Merrill and Dr.
Matthew Kimber, for their guidance and constructive input throughout the process that was my
graduate education.
To the two post-doctoral fellows in our lab, Dr. Vladimir Bamm and Dr. Kenrick Vassall,
who were responsible for bending my mind and wrapping it around many of the concepts that I
touch on in this thesis, I would not have been near as successful or as well trained without your
efforts. To past and present labbies (Aggie, Danielle, Miguel, Sergio), thank you for your advice
and endless hilarity that dragged me through the days.
But most of all, I’d like to thank my family, especially my mom – thank you for all the
help, love, and support. You have made this all possible.
iii
TABLE OF CONTENTS
ABSTRACT .................................................................................................................................... ii
ACKNOWLEDGEMENTS ........................................................................................................... iii
TABLE OF CONTENTS ............................................................................................................... iv
LIST OF TABLES ........................................................................................................................ vii
LIST OF FIGURES ..................................................................................................................... viii
LIST OF ABBREVIATIONS ......................................................................................................... x
Chapter 1.0: Introduction ................................................................................................................ 1
1.1 Myelin architecture and Myelin Basic Protein (MBP) ..................................................................... 1
1.2 Multiple Sclerosis (MS) and Myelin Basic Protein........................................................................... 3
1.3 Myelin basic protein in health and disease..................................................................................... 4
1.4 Post-translational modifications on MBP in MS pathology ............................................................ 5
1.5 The Intrinsically-Disordered Protein (IDPs) Family ......................................................................... 6
1.6 MBP is an IDP and has alpha-helical propensity ............................................................................. 7
1.7 The recombinant MBP 18.5-kDa isoform........................................................................................ 8
1.8 TFE-Induced secondary structural changes .................................................................................. 14
1.9 Fluorescence and Fluorescence Anisotropy ................................................................................. 15
1.10 Fluorescence Resonance Energy Transfer (FRET) and its use in IDPs ......................................... 17
1.11 Measuring the ‘disorder-to-order’ transition(s) of MBP in TFE in order to probe the ‘major
thermodynamic ensemble populations’ ............................................................................................. 18
Chapter 2.0: Experimental Plan .................................................................................................... 20
2.1 Hypotheses.................................................................................................................................... 20
2.2 Objectives...................................................................................................................................... 20
2.2.1 Develop a labelling protocol ...................................................................................................... 20
2.2.2 Compare using CD and anisotropy............................................................................................. 21
2.2.3 Map intramolecular distances ................................................................................................... 21
2.2.4 Combined analysis ..................................................................................................................... 21
2.2.5 Mapping intramolecular distances in order to identify the ensemble populations .................. 21
iv
2.2.6 Structural and thermodynamic comparison .............................................................................. 22
2.3 Significance ................................................................................................................................... 22
2.4 Statement of Publication .............................................................................................................. 22
Chapter 3.0: Experimental Methods ............................................................................................. 23
3.1 Chemicals and Reagents ............................................................................................................... 23
3.2 Recombinant murine full-length untagged 18.5-kDa MBP-UTC1 and MBP-UTC8........................ 23
3.3 Labelling of Cysteine-substituted MBP-UTC1 and MBP-UTC8 variants with IAEDANS................. 26
3.4 Circular Dichroism Measurements................................................................................................ 27
3.5 Fluorescence measurements ........................................................................................................ 28
3.6 Steady-state Fluorescence Anisotropy Measurements ................................................................ 28
3.7 ANS fluorescence .......................................................................................................................... 28
3.8 Fluorescence resonance energy transfer measurements............................................................. 29
3.9 TFE-titration curves monitored by tryptophan fluorescence quenching ..................................... 30
3.10 Quantitative analysis of TFE-titration curves .............................................................................. 30
Chapter 4.0: Results ...................................................................................................................... 33
4.1 MBP Site-directed mutagenesis .................................................................................................... 33
4.2 MBP purification ........................................................................................................................... 33
4.3 MBP labelling with IAEDANS ......................................................................................................... 36
4.4 Determining MBP labelling efficiency ........................................................................................... 37
4.5 Circular Dichroism Measurements of MBP ................................................................................... 37
4.6 ANS fluorescence in MBP .............................................................................................................. 42
4.7 Steady-state Fluorescence Anisotropy Measurements of MBP ................................................... 47
4.8 Fluorescence resonance energy transfer (FRET) measurements of MBP ..................................... 47
4.9 Measuring the ‘disorder-to-order’ conformational transition(s) of MBP in TFE .......................... 52
4.10 The thermodynamic analysis of the ‘disorder-to-order’ transition of MBP in TFE in order to
probe the ‘ensemble populations with different free energies’ ........................................................ 52
4.11 Deconvolution of the major thermodynamic ensembles along the ‘disorder-to-order’ transition
of MBP................................................................................................................................................. 57
Chapter 5.0: Discussion ................................................................................................................ 63
5.1 MBP-UTC1 in solution ................................................................................................................... 63
v
5.2 Membrane mimetic solvents and their uses in IDPs..................................................................... 65
5.3 Thermodynamic analysis of the disorder-to-α-helical transition in full-length 18.5-kDa MBPUTC1 using 2,2,2-trifluoroethanol (TFE) ............................................................................................. 67
5.4 Defining a physiological role for MBP-UTC8: A structural and thermodynamic comparison of the
MS-MBP isoform ................................................................................................................................. 68
Chapter 6.0: Conclusions, Significance and Future Directions .................................................... 70
References .......................................................................................................................................... 72
vi
LIST OF TABLES
Table #1A and #1B. Steady-state fluorescence anisotropy (r) measurements of MBP-UTC1 (A)
and MBP-UTC8 (B). ..................................................................................................................... 45
Tables #2A and 2B. Fluorescence resonance energy transfer (FRET) of MBP labelled with
IAEDANS ..................................................................................................................................... 49
Table #3. FRET measurements in 6 M guanidinium chloride LIST OF TABLES (guanidine
hydrochloride, GdmCl). ................................................................................................................ 51
Table #4. Thermodynamic values were determined at 25 °C from global fitting of TFE titration
curves monitored by changes in CD signal and Trp fluorescence quenching by IAEDANS....... 59
vii
LIST OF FIGURES
Figure
Title
Page
#1
The structure of myelin in the CNS and the primary function of 18.5-kDa
12
MBP
#2
Notable secondary structural elements of 18.5-kDa MBP
#3
Simple Jablonski diagram of three-state model of fluorescence
#4
#5
#6
#7
#8
#9
#10
Amino acid sequence and secondary structural map of murine 18.5-kDa
MBP-UTC1 and MBP-UTC8 (168 residues)
Purification of MBP-UTC1 and MBP-UTC8
Changes in circular dichroic (CD) spectra of 18.5-kDa MBP with
increasing trifluoroethanol (TFE) concentration
Comparison of TFE-titration curves of IEADANS-labelled MBP-UTC1S17C and S159C 18.5-kDa MBP with the unmodified protein
Circular dichroic (CD) spectra of unmodified 18.5-kDa MBP-UTC18
and MBP-UTC IAEDANS-labelled cysteine variants.
Change in ANS fluorescence enhancement by 18.5-kDa MBP upon TFE
titration
Change in ANS fluorescence enhancement by 18.5-kDa MBP-UTC8
upon TFE titration
19-20
24
42
43
47
48
49
51
52
#11
TFE-titration curves of full-length 18.5-kDa MBP
59
#12
Comparison of TFE-titration curves of IEADANS-labelled S17C and
60
viii
S159C 18.5-kDa MBP-UTC1 with the unmodified protein
#13
#14
#15
Preliminary TFE-titration curves of full-length 18.5-kDa MBP-UTC8
Fractional population of equilibrium conformers of MBP-UTC1 and
MBP-UTC8.
Thermodynamic mechanism of disorder-to-α-helical transition of MBP
and fractional population of equilibrium conformers
61
65
66-67
ix
LIST OF ABBREVIATIONS
Abbreviation
Term
α1, α2, α3
α-helix-containing peptide fragments of murine 18.5-kDa MBP:
segments (A22-K56), (S72-S107), (S133-S159), respectively.
ANS
anilinonaphthalene-1-sulfonic acid
AP
action potential
AU
arbitrary units
BCA
bicinchoninic acid
CD
circular dichroism
CNS
central nervous system
DNA
deoxyribonucleic acid
DPC
dodecylphosphocholine
E. coli
Escherichia coli
EAE
experimental autoimmune encephalomyelitis
EPR
electron paramagnetic resonance
FPLC
fast protein liquid chromatography
FRET
fluorescence (or Förster) resonance energy transfer
FTIR
Fourier transform infrared
golli
gene in the oligodendrocyte lineage
GdmCl
guanidinium chloride
HPLC
high-performance liquid chromatography
IAEDANS
5-((((2-iodoacetyl) amino) ethyl) amino) naphthalene-1-sulfonic acid
x
IDP
intrinsically-disordered protein
IPL
intraperiod line
IPTG
isopropyl β-D-thiogalactopyranoside
kDa
kilodaltons
LB
Luria-Bertani
MAP
mitogen-activated protein
MBP
myelin basic protein (usually the 18.5-kDa size isoform)
MDL
major dense line
MoRF
molecular recognition fragment
MS
Multiple Sclerosis
NaCl
salt, sodium chloride
NMR
nuclear magnetic resonance
PAD-2
peptidylarginine deiminase-2
PCR
polymerase chain reaction
PLP
proteolipid protein
PTM
post-translational modifications
rmMBP
recombinant murine MBP (generic designation, usually with C-terminal
LEH6 tag yielding 176 residues)
SDS-PAGE
sodium dodecyl sulphate-polyacrylamide gel electrophoresis
TCEP
Tris-2-carboxyethyl-phosphine
TEM
transmission electron microscopy
TFA
trifluoroacetic acid
xi
TFE
trifluoroethanol
UTC1
unmodified C1 charge component of MBP (untagged, 168 residues)
UTC8
pseudo-deiminated C8 charge component of MBP (untagged, 168
residues)
UTC1, UTC8
“untagged” recombinant variants of 18.5-kDa murine MBP, unmodified
and pseudo-deiminated, respectively
UV
ultraviolet
xii
Chapter 1.0: Introduction
1.1 Myelin architecture and Myelin Basic Protein (MBP)
The myelin sheath is a distinctive membrane complex formed when multiple spiralwrapped lamellae surround neuronal axons. Glial cells known as oligodendrocytes produce
myelin in the central nervous system (CNS), which includes the spinal cord and brain. The
oligodendrocyte membrane extends towards and makes physical contact with the axon and
initiates myelin sheath formation, concentrically winding itself onto an axon (1, 2).
The myelin sheath is important for proper neuronal function, as its integrity determines
the speed at which action potential(s) (APs) can propagate (3, 4). Myelin itself has historically
been viewed as an axonal insulator, working to limit the dissipation of ion current away from ion
channels, and limiting depolarization events to the nodes of Ranvier (5). These nodes remain
uninsulated and, due to the abundance of voltage-gated ion channels located within them, are
capable of generating vast amounts of electrical activity. These nodes are spaced approximately
1 µm in distance away from each other, and thus AP conduction and propagation along the
neuron “hops” or “jumps” from node to node: this mechanism of AP conduction and propagation
along a myelinated nerve is known as saltatory conduction (6, 7). The process of myelination is
thought to have facilitated fast and efficient communication in the brain, giving rise to animals
with complex motor, sensory, and cognitive functions (8). Any disruption of the myelin sheath
integrity compromises the ability of nerves to propagate APs effectively, and leads to impaired
neurological function, where symptoms are highly correlated with the areas of the brain that are
affected (5, 7, 9).
Figure #1 shows the structure of myelin sheath itself, composed of repeating lipid
bilayers separated by 3 to 4 nm aqueous layers (10) and forms distinct regions: the intraperiod
line (IPL) formed by the apposition of extracellular membrane layers, and the major dense line
(MDL) formed by the apposition of the cytoplasmic membrane surfaces (11). The MDL contains
a cytosolic, peripheral membrane protein, known as Myelin Basic Protein (MBP) which
composes up to 30% of the dry weight of the total protein in myelin, making it the second most
abundant myelin protein, following proteolipid protein (PLP) (3, 12, 13).
1
Figure #1. The structure of myelin in the CNS and the primary function of 18.5-kDa
MBP. The myelin sheath is a distinctive membrane complex formed when multiple
spiral-wrapped lamellae surround neuronal axons of nerve fibers. Glial cells known as
oligodendrocytes produce myelin in the central nervous system (CNS), which includes
the spinal cord and brain. The myelin sheath is composed of repeating lipid bilayers
separated by 3 to 4 nm aqueous layers and forming distinct regions: the intraperiod line
formed by the apposition of extracellular membrane layers and the major dense line
formed by the apposition of the cytoplasmic membrane surfaces. A cytosolic, peripheral
membrane protein, known as myelin basic protein (MBP), comprises up to 30% of the
dry weight of protein and is present in the major dense line. The predominant 18.5-kDa
isoform of MBP maintains myelin structure by conjoining the cytoplasmic leaflets of the
oligodendrocyte membrane. MBP is the 'executive molecule' and the only structural
protein essential for formation and maintenance of normal myelin. Adapted with
permission from Ishiyama and colleagues (2001).
2
When discussing MBP, it must be clarified that it is not one single protein, but a large
family of protein isoforms derived from alternative splicing events of the Golli (gene in the
oligodendrocyte lineage) gene complex, with the common human MBP splice isoforms being
17.2, 18.5, 20.2, and 21.5 kDa in nominal molecular mass (3). These protein isoforms are
developmentally-expressed, translocatable, and highly post-translationally modified, and show a
diversity of binding partners (3). Historically, the 18.5-kDa MBP isoform’s role has been thought
to maintain the structural integrity of the myelin sheath via adhesion to cytoplasmic leaflets of
the oligodendrocyte membrane. But today, it is known that MBP is highly multi-functional,
serving as a 'hub' protein within myelin. Thus, in addition to its main role of maintaining the
structural integrity of myelin via interactions with bilayers and/or lipids (14), common roles for
the 18.5-kDa MBP isoform include chemical signalling (15); cytoskeletal assembly (16);
interactions with Cu2+ and Zn2+ (17, 18), with calcium-activated calmodulin (19), and with SH3domain containing proteins (20-22). As the focus of this thesis and the experimental work has
been performed on this classical 18.5-kDa MBP isoform, it will henceforth and throughout be
referred to as MBP.
1.2 Multiple Sclerosis (MS) and Myelin Basic Protein
Multiple sclerosis (MS) is a debilitating disease directed against myelin and the myelin
components in the CNS (3, 23). MS is characterized by inflammation, demyelination, lesion
formation, and progressive neurodegeneration (24). Progressive deterioration of myelin causes
impaired saltatory conduction resulting in increasing loss of neurological function with
symptoms including cognitive dysfunction, depression, spasticity, muscle weakness, and
paralysis (24).
Schools of thought on the origin of MS typically involve the degeneration (or failure to
regenerate) of the insulating layer around neuronal axons, and generation of auto-antigens
against components of the CNS myelin, specifically MBP which is a well-known candidate autoantigen in MS (25-27). Although MS is synonymous with auto-immune disease, the exact
causative mechanism remains elusive (3, 27-29). Increasing experimental evidence suggests that
a genetic contribution manifested by the up-regulation of enzymes that govern specific posttranslational modifications (PTMs) of MBP may be involved (30-33). The mechanisms through
3
which post-translational modifications and interacting partners alter the structure of MBP require
further investigation (3).
1.3 Myelin basic protein in health and disease
MBP maintains myelin structure by conjoining the cytoplasmic leaflets of the
oligodendrocyte membrane and linking the membrane to the underlying cytoskeleton, including
the cytoskeletal proteins actin and tubulin (34), whose assembly MBP strongly enhances (16).
Deletion or mutation of the mbp gene causes the loss of compact CNS myelin as was shown in
the shiverer mice (35), and the shaker rat (36). As the 'executive molecule' and only structural
protein essential for formation and maintenance of normal myelin, the loss of MBP, or change in
MBP’s interaction with the bilayer, can have dramatic physiological and/or neurological effects
(3, 37).
The association between MBP and oligodendrocyte membranes is thought to be dictated
by hydrophobic and electrostatic forces working in cooperation to maintain the interactions
between MBP residues and anionic lipids of the myelin bilayer (38). Alterations made to these
hydrophobic or electrostatic interactions between MBP and myelin layers results in reduced
adhesion and sheath instability (39).
Some post-translational modifications of MBP, specifically deimination (conversion of
arginine to citrulline, resulting in a decrease in the net positive charge) is highly correlated with
MS pathology (32, 40, 41) with the degree of arginine lost by deimination being around 45% in
chronic MS, and 20% in normal healthy adult brain (42). But more relevantly, highly deiminated
MBP is thought to be ‘less structurally ordered’ and also shows less affinity to maintain binding
membrane lipids (40, 43). Yet, there is still a poor understanding of the role that deimination
plays in MS pathogenesis, and the mechanisms through which PTMs and interacting partners
alter the structure of MBP require further investigation (3).
4
1.4 Post-translational modifications on MBP in MS pathology
Recently, bioinformatics analysis has suggested that proteins with post-translational
target sites that are great in number and variation, such as MBP, are more prone to be involved in
disease (44). PTMs have also long been known to alter the nature and extent of secondary
structure in proteins, permitting them to adopt various conformations required for binding and
signalling events, as is seen in proteins such as tau and α-synuclein which are very similar to
MBP in such properties (45, 46). The PTMs of MBP include methylation, deimination,
deamidation, and phosphorylation (3), and give rise to a multitude of charge components and
potentially a vast number of conformers (31, 47-49).
More specifically, the enzyme peptidylarginine deiminase type 2 (PAD2) is known to
cause a highly relevant PTM of MBP that is associated with MS (31, 32, 50). Citrullination
causes the irreversible conversion of the amino acid arginine into citrulline through replacement
of the terminal ketamine group (=NH) on arginine with a ketone group (=O) by PAD.
Deimination has been shown to alter the electrostatic and hydrophobic interactions between
citrullinated residues and negatively-charged lipid head groups (3, 52) work to decrease MBP’s
ability to interact with and organize lipid bilayers into compacted myelin layers (53).
Deimination of MBP is thought to represent the primary defect that precedes demyelination and
progressive neurodegeneration that is seen in MS pathology (3, 12). In MS research in our
laboratory, we distinguish between two major charge components: MBP-UTC1 which is the least
modified and most cationic variant of MBP with a net charge of +19 at neutral pH; and MBPUTC8 which represents a more highly modified form and differs by deimination of 6 to 11
arginyl residues, resulting in a decrease of net-positive charge to +13 at neutral pH (32, 51). The
MBP-UTC8 variant is associated with MS pathology as well as with myelin development (12).
Electron paramagnetic resonance (EPR) and nuclear magnetic resonance (NMR) studies
suggest that deimination results in a loss of structural integrity of an intrinsic α-helix formed by
MBP (discussed below). This PTM causes the α-helix to become shorter in length, and more
surface-exposed when bound to the cytoplasmic leaflet of the myelin membrane, and is thus
more susceptible to proteolysis as compared to MBP-UTC1 (52, 54). Other EPR studies on
MBP-UTC1 and MBP-UTC8 determined that MBP-UTC8 formed a shorter, more surface5
exposed α-helix in MBP-UTC8 as compared to MBP-UTC1 (52), which nicely complemented
studies using myelin-associated proteases such as cathepsin D, which showed an increase in the
ability to digest MBP-UTC8 (3, 43). All in all, the specific effect of deimination of MBP-UTC1
to yield MBP-UTC8 results in a conformational change that leads to the cytoplasmic exposure of
a Phe86-Phe87 pair making the α-helix more vulnerable to proteolytic attack and digestion (52).
Therefore, the effect(s) of deimination on the structure of MBP, and the resultant conformational
changes that lead to the exposure of this 'molecular switch' may define a new pathological role
for MBP in MS (55-58). Furthermore, this model both complements and challenges the historical
dogma of MS research, suggesting that MS may be in fact be a post-translational disease and that
PTMs may represent the primary defect that leads to MS pathology (33, 59). But as discussed
below, characterizing MBP’s structure, or the effects of PTMs and the resultant conformational
changes, have historically been somewhat challenging.
1.5 The Intrinsically-Disordered Protein (IDPs) Family
MBP’s structure continues to elude researchers largely because MBP is not any one
single protein. A central tenet of biochemistry, the so-called structure-function paradigm, states
that having a well-defined 3D structure is a prerequisite for the proper function of proteins (60).
Traditionally, proteins are characterized by a rigid, well-folded, three-dimensional structure,
defined by a single set of backbone side-chain atom positions and dihedral angles, represented by
an energetically-minimal assembly where limited fluctuations of conformation occur (61, 62).
This interpretation of protein structure-function relationships revolves around two main
principles. The first suggests that the network of enzymatic, protein-protein, and/or proteinnucleic acid interactions inside the cell would not occur unless proteins act in a highly "specific"
manner (63-66). Secondly, the "lock-and-key" concept, used to understand specificity of ligand
action, states that without a well-defined geometry and the implied rigidity, particularly in
regions of molecular binding, specificity of interactions could not be assured (63-66).
Today, it is understood that proteins can exist in a trinity of structures (namely the
ordered, molten globule, and random coil states (60)), that native protein structure and function
can correspond to any of the three states, and finally, that intrinsically-disordered proteins (IDPs)
are distinguished by their incapacity to adopt a well-defined structure under near-physiological
6
conditions (65). Most IDPs typically show low hydrophobicity, high net charge, and exist as
dynamic, conformational ensembles where dihedral angles and atom arrangements vary
significantly over-time, therefore attaining no equilibrium (66). Thus, IDPs lack the specific
three-dimensional structure required for crystallography (67, 68). It was not until the late 1990's
that this central tenet of biochemistry was challenged as more advanced methodologies for
analysis allowed the discovery of functional proteins which were either completely or partially
disordered.
In 2013, the database of protein disorder (Disprot) listed over 694 proteins, with 1539
disordered regions (69). Amino acid sequence analysis has suggested disorder is an encoded
property and software packages predict that 45-50% of eukaryotic proteins contain partially, or
completely, disordered regions in solution (70). Disordered proteins are now known to be
functionally important, partaking in cellular processes such as signalling, transcription,
replication, cell-cycle control, chaperoning, and more importantly, have been associated with
various human pathologies, including multiple neurological disorders (71, 72).
Intrinsic disorder is advantageous as it can allow for rapid association with one or
multiple binding partners, in response to alterations in environmental or chemical signals (73).
Consequently, IDPs can form a large number of interactions, acting as network hubs, which
allow protein-protein, protein-ligand, and protein-membrane interactions that facilitate the
transmission of information across subcellular systems (49, 66, 73). These associations remain
fairly specific as the energy of secondary and/or tertiary folding can act as a barrier to
interaction(s) (60).
1.6 MBP is an IDP and has alpha-helical propensity
All variants of MBP are IDPs (3). Previous Fourier-transform infrared (FTIR) (74) and
NMR (75) studies of recombinant 18.5-kDa MBP have shown that MBP does not adopt a
significant proportion of regular secondary structure elements in aqueous solution (76). But
whereas the behaviour of MBP is consistent with a random-coil conformation, the central region
is known to have some degree of ordered secondary structure (77). MBP has demonstrated the
potential to adopt long-range folding interactions, thought to be critical for compact assembly of
7
myelin (3). Hydrophobic moment analysis of the amino-acid sequence of MBP (78), and
solution/solid state nuclear magnetic resonance (NMR) spectroscopy, demonstrate the existence
of three amphipathic α-helices located in the N- and C-termini and in the central regions of the
protein (78, 79). Regions with high α-helical propensity in MBP include: Thr33-Asp46, Val83Thr92, and Try142-Leu154, designated as α1, α2, and α3, respectively (79-81) (Figure #2A and
#2B). Depth measurements of unmodified MBP (MBP-UTC1) in simple lipid systems, via spin
labelling and electron paramagnetic resonance (EPR) spectroscopy, have confirmed that these
regions are embedded in the lipid membrane and have α-helical properties (82).
The α-helical regions, and more specifically the central region, which contains α2
followed by a poly-proline sequence, is present in almost all vertebrates, with residues Asp81Gln118 being highly conserved in mammals (78). This region also contains mitogen-activated
protein (MAP)-kinase sites located at T92 and T95 (murine 18.5-kDa sequence numbering), and
includes the immunodominant epitope (murine N81-PVVHFFKNIVTPRTPPP-S99) used to
induce EAE (experimental autoimmune encephalomyelitis) in model animals, and so may also
act as a possible "molecular switch" (83). Characterizations of α2 and the molecular switch
regions have been performed in aqueous buffer, in membrane-mimetic environments such as
2,2,2-trifluoroethanol (TFE) (83), and in dodecylphosphocholine (DPC) micelles (84-86).
Generally, PTMs (with net positive charge reduction) have the potential to modify MBP
electrostatic interactions with its binding partners, which result in conformational changes in
both the molecular switch and/or α-helical regions. These changes are, in turn, thought to cause
the destabilization of myelin sheath, as seen in EAE and MS (85).
1.7 The recombinant MBP 18.5-kDa isoform
The recombinant murine MBP (rmMBP) 18.5-kDa protein has been demonstrated as a
successful substitute to purifying the protein from natural sources (87). Recombinant proteins
from a bacterial over-expression system can sometimes be misrepresentative of a protein from a
natural source for two main reasons: (1) if the recombinant protein is not biochemically
representative of its natural counter-part, and (2) the normal eukaryotic post-translational
modifications are absent (87, 88). With respect to the first point, there is no known functional
assay to determine the structural integrity of MBP (3, 89), and therefore the structure of
8
recombinant murine MBP variants has typically been characterized by spectroscopic techniques
such as nuclear magnetic resonance (NMR) (78-80), FTIR (74), circular dichroism (CD) (90),
and electron paramagnetic resonance (EPR) (19, 33, 91).
9
Figure #2A and #2B. Notable secondary structural elements of 18.5-kDa MBP. Three
regions previously shown to have α-helical propensity (Thr33-Asp46, Val83-Thr92, and Tyr142Leu154) are displayed as red helices. Hydrophobic moment plot of three regions previously
shown to have α-helical propensity (Thr33-Asp46, Val83-Thr92, and Tyr142-Leu154; shown in
part A as segments of red helix) reveals that these helical regions are strongly amphipathic. This
figure was reproduced with permission from Harauz et al. (2009). (B) Probability of regular
secondary structure types in His6-tagged murine recombinant 18.5-kDa MBP in 30% TFE (v/v).
The plot was obtained by analyzing previously-derived Hα, HN, Cα, Cβ, and C′ solution NMR
chemical shift assignments (BMRB 6100) using the δ2D software and the associated web server
created by the Vendruscolo group (Camilloni et al., 2012). The regions with highest propensity
for α-helical structure are at the center (encompassed here by the α2-peptide) and at the Cterminus (encompassed here by the α3-peptide). Significant propensity for α-helical structure is
also seen near the N-terminus (~I35-F42, murine species numbering) (encompassed here by the
α1-peptide). The regions with the highest probabilities for α-helical content in 30% TFE (v/v)
correspond to regions predicted to form membrane-associated α-helices based on a hydrophobic
moment plot (Harauz et al., 2009). The locations of the α-helical regions also agree with CD
experiments of different MBP α-peptides, and with solution NMR experiments of the central
MBP segments in a dodecylphosphocholine micelle environment. These regions were also
shown in site-directed spin labelling electron paramagnetic resonance measurements to associate
with the lipid bilayer in reconstituted myelin-mimetic systems.
10
Figure2A
11
Figure2B
12
Studies on 18.5-kDa rmMBP have determined that the recombinant protein is similar to
the natural bovine 18.5-kDa isoform in terms of (1) immunoreactivity to antibodies, (2) the
ability to cause experimental autoimmune encephalomyelitis (EAE) in animal model systems
(87, 92), (3) the degree of secondary structure in aqueous solution via CD (90), and (4) adhesion
to lipids (93). The lack of normal eukaryotic PTMs in bacterial systems is evident when looking
at differences between natural bovine MBP (bMBP) and rmMBP. For instance, bMBP arginine
residue 106 is either non-, mono-, or di-methylated, and the amino-terminal end of bMBP is
acetylated (94, 95). Similar modifications are present in natural human MBP 18.5-kDa (3, 90,
94) but the fact that PTMs typical of eukaryotic species are absent in bacterial systems can
actually be beneficial to researchers, acting as a 'tabula rasa'.
Recombinant DNA technology allows for the production of homogenous preparations of
recombinant proteins (88) and pseudo-modification of amino acids by site-directed mutagenesis,
which would otherwise be impossible (50, 54, 87, 88). The incorporation of probes to converted
residues can then be used to study the effects of specific PTMs individually, or in a
combinatorial manner, on structure or interaction with proteins, small ligands, and other
interacting partners. In our studies, MBP-UTC8 refers to recombinant murine MBP modified at
these 6 particular sites (the murine residues R25, R33, K119, R127, R157, and R168 correspond
to primary arginine deimination sites in the C8 component of the human protein. The MBP posttranslational sites were recombinantly modified to best mimic the size, as well as the charge
reduction, associated with the particular PTM. The murine protein’s sites of pseudo-citrullination
(5 Arg + 1 Lys, each with a net charge +1) were modified to glutamine (net charge 0). This
strategy has been employed previously in our lab to study MBP PTMs, and has been cited to be
effective in mimicking structural changes and modulating ligand associations seen with in vivo
PTMs in other proteins (50). As these modifications are determinate for normal MBP structure
and/or function, they are therefore critical for the understanding of MS pathology (3).
13
1.8 TFE-Induced secondary structural changes
Conformational analysis of peptides and small proteins is commonly carried out in
structure-stabilizing solvents such as TFE. Experimentally, TFE induces and stabilizes α-helicity
in peptides and proteins (96). TFE is a hydrophilic and hydrogen-bonding solvent (97) and can
also strongly influence secondary structure and global folding of peptides and proteins. It has
been demonstrated that TFE induces pre-existing helix-coil formation in proteins limited to when
the effect is saturated at low concentration of TFE (<40% vol) (98), and solvation of the helix
structure by TFE stabilizes the intramolecular hydrogen bonds (99, 100).
Trifluoroethanol has been shown to induce α-helical structure mainly in regions which
already have α-helix-forming propensity, and it has been shown to be an effective membranemimetic solvent for many proteins, including MBP (75, 99). In corroboration, analysis of
previously published NMR data of MBP in 30% TFE (v/v) has demonstrated that the pattern of
α-helical formation in TFE is consistent with what has been found in lipid (101). At low
concentrations, it has been suggested that TFE can effectively simulate membrane surface
conditions by lowering the dielectric constant of the solution, whereas at higher concentrations,
there is the added effect of the preferential interaction of TFE with the protein, which can
effectively simulate the effects of protein-membrane binding (99, 102). Thus, TFE is useful in
analyzing the disorder-to-α-helical transition of MBP that occurs as it approaches the
oligodendrocyte membrane surface and eventually interacts.
TFE is also compatible with a number of techniques including fluorescence and circular
dichroism (CD). CD/TFE titration curves are particularly useful as they can be used to determine
the average α-helicity of peptides and small proteins (83). This method can also allow for the
calculation of the solvation free energy and free energy change associated with the coil-helix
transition, and defining the thermodynamics for α-helical formation of IDPs (98). Previous CDmonitored TFE-titration curves of four α2-peptides (unmodified, phosphorylated T92 and/or
T95) have been performed (83); however, the effects of other typical PTMs such as deimination
in this respect have yet to be determined.
14
1.9 Fluorescence and Fluorescence Anisotropy
Over the last two decades, fluorescence has become a standard tool for deciphering
complex biological systems. Fluorescence is the release or emission of radiation produced by
some molecules (fluorophores) due to the absorption of incident radiation of a shorter
wavelength (103). A three-stage process responsible for the fluorescence of fluorophores is
illustrated by the simple electronic-state diagram shown in Figure #3. Excitation is the process
of photon absorption creating an excited electronic singlet state (S1). The S1 state is then
dissipated, yielding a relaxed state from which fluorescence emission originates (104), and
finally by fluorescence emission, where a photon of energy is emitted, returning the fluorophore
to its ground state S0 (103). Measurable parameters of a fluorophore, such as absorption
wavelength (λ), emission λ, quantum yield, fluorescence quenching, fluorescence intensity and
anisotropy, can provide a wealth of information pertinent to the structure of IDPs, and
specifically MBP (103, 105-107).
For example, excitation of some molecules by polarized light in one dimension (x, y, or
z) subsequently causes the emission of polarized light within the same dimension (103). A
molecule’s anisotropy is defined as the difference between the intensity of polarized light in the
vertical direction and the intensity of polarized light in the horizontal direction, normalized to the
total intensity of light (103). Fluorophores with non-zero anisotropies display changes in the
dimension of polarized light emission, and the extent of this emission is defined by its relative
anisotropy (r), a unit-less measure (103). Proteins in aqueous solution, particularly IDPs which
are highly mobile, tumble during the excited state. Consequently, the polarized light is emitted in
various directions as compared to the incident light (103, 108). The displacement of polarized
light is dependent upon the speed and freedom of rotational diffusion during the lifetime of a
fluorophore’s excited state (108). Rotational diffusion of a protein depends on the solvent, and
the protein’s size and intrinsic shape (103, 109). Macromolecules and protein-binding
interactions with various binding partners can also be measured by anisotropy as a result of
changes in rotational diffusion (108, 109). Experimentally, a sample is excited by polarized light
which is oriented in the vertical or the z-axis (103). The intensity of the emission is measured
through a polarizer, and if emission is oriented parallel (║) to the direction of the excitation light,
15
Figure #3. Simple Jablonski diagram of three-state model of fluorescence. (1)
Excitation is the process whereby a photon is supplied by an external source and
absorbed by the fluorophore, creating an excited electronic singlet state (S1'). (2) The
excited state lifetime is finite, lasting 1–10 nanoseconds (in most cases), and during
this time, a fluorophore is subjected to a multitude of possible interactions with its
molecular environment. The energy of S1' is partially dissipated, yielding a relaxed
singlet excited state (S1) from which fluorescence emission originates. (3) A photon is
emitted, returning the fluorophore to its ground state S0. Due to energy dissipation
during the excited-state lifetime, the energy of this photon is lower, and therefore of
longer wavelength, than the excitation photon (Figure created by Author).
16
the observed intensity is called I (║). If the polarizer is oriented perpendicular (┴) to the excitation
light, the intensity is called I (┴) (109). These values can be used to calculate anisotropy (r), in
order to infer the relative environment of a fluorescent probe as well as to determine if the probe
has free-rotation or is restricted. Data will be collected for both MBP-UTC1 and MBP-UTC8 in
aqueous buffer and in TFE.
1.10 Fluorescence Resonance Energy Transfer (FRET) and its use in IDPs
The realization that MBP is an IDP shifted focus to understanding the ‘binding-induced
folding’ or ‘disorder-to-order transitions’ influenced by interactions with its cellular partners
(zinc, salt, etc.) as reviewed in (94), or/and in the presence of membrane-mimetic environments
such as TFE or synthetic lipid(s) (39, 110). As mentioned previously, multifunctionality in IDPs
such as MBP is related to their conformational plasticity and their ability to undergo disorder-toorder transitions. It is important to understand the mechanism of these structural conversions
(47). As the organization of the protein conformation may already start forming “in solution”
before it interacts fully with the two leaflets, MBP therefore has the potential to interact with
various cellular partners before being inserted into the membrane (111). It is, therefore,
important to obtain more information on the changes in ensemble conformation of MBP; as such,
FRET will be performed in a series of environments.
Fluorescence (or Förster) resonance energy transfer (FRET) can be, and has been,
employed for the determination of long-range interactions between amino acid residues in IDPs
(46, 106, 112), often exploiting the intrinsic tryptophan residue(s) of proteins as the donor, and
the selection of a suitable fluorescent probe as an acceptor (46, 113). The FRET approach is
uniquely suited to the study of IDP structure, and by extension MBP’s structure, as the technique
has high binding specificity, accuracy at angstrom (Å) level distances and/or molecular
dimensions, and the ability to resolve distributions of intramolecular distances of fast fluctuating
disordered protein(s) regions (106). Minimally, FRET requires that: (1) the absorption spectrum
of the acceptor must overlap the fluorescence emission spectrum of the donor, (2) the donor and
acceptor molecules must be in proximity (typically 10–100 Å), and (3) donor and acceptor
transition dipole orientations must be roughly parallel (103, 113, 114). The FRET involves the
distance-dependent irradiative transfer of energy from a donor molecule to an acceptor molecule
17
via a dipole-dipole coupling mechanism (113). The donor molecule is the chromophore that
initially absorbs energy when excited by photons and which subsequently transfers energy to an
acceptor chromophore (46, 113) Energy transfer efficiency is determined by the ratio between
donor fluorescence intensity in the presence and absence of acceptor, which can then be used to
determine the physical distance between the donor and acceptor (114). Previous studies have
demonstrated that fluorescence techniques, including FRET, are well-suited to study the
conformational ensemble of MBP caused by PTMs in aqueous buffers, in membrane-mimetic
solvents such as TFE, in protein denaturants, and, finally, in the presence of small-interacting
ligands such as zinc and NaCl (46, 103, 113, 115). For the reasons outlined above, it is clear that
FRET is a natural complement to previous studies that have been conducted by the Harauz group
on MBP using NMR, EPR, and CD spectroscopy.
1.11 Measuring the ‘disorder-to-order’ transition(s) of MBP in TFE in order
to probe the ‘major thermodynamic ensemble populations’
As part of an integration of multiple sources of data, it is also possible to define the
mechanism of the disordered-to-α-helical transition of MBP. This can be accomplished by
thermodynamic analysis of the disordered-to-α-helical transition that is induced upon titration
with TFE. It is important to monitor conformational changes using multiple probes specific for
secondary and/or tertiary (or global) structural changes. In this study, this analysis will be
performed by CD spectroscopy, which probes for changes in secondary structure, and will also
be monitored by tryptophan fluorescence quenching by IAEDANS, which can be used to probe
for concomitant changes in global organization. By utilizing combinations of techniques,
including circular dichroism (CD) spectroscopy, tryptophan fluorescence quenching, and 1anilinonaphthalene-8-sulfonic acid (ANS)-binding, it is possible to define the ‘major
thermodynamic clusters of ensemble populations’ along the transition pathway, as previously
done with other prominent IDPs, such as tau and α-synuclein (46, 115, 116),
MBP likely undergoes global conformational changes resulting in the population of
various energetically distinct sets of conformational ensembles which are regulated by PTMs, as
has been seen for a variety of IDPs (46, 117), each of which potentially having varying levels of
compaction and secondary structure. A full thermodynamic description of the mechanism of
18
disorder-to-order transition is useful both for defining the protein’s conformational space which
may be important in binding-induced folding mechanisms, and also for identifying populations
of energetically distinct conformations along the transition pathway which may be instrumental
in conformational selection.
19
Chapter 2.0: Experimental Plan
2.1 Hypotheses
First, we posit that the MBP-UTC1 variant adopts a hairpin or paperclip shape in the
presence of membrane-mimetic solvents, as observed in previous studies of other IDPs similar to
MBP, like tau (46, 115). This conformation is important for close apposition of oligodendrocyte
membrane bilayers, yet MBP is conformationally adaptive and is subject to change upon
interaction with its various binding partners, and in the presence of different solvent systems,
which is a key property for its multifunctionality.
Second, we hypothesise that deimination of MBP associated with charge reduction will
alter the global structure of the MBP-UTC8 variant. In particular, we believe that the extensive
deimination of MBP-UTC1 (+19) to MBP-UTC8 (+13) will result in different long-range
intramolecular folding interactions. Subsequently, this conformational change in MBP may lead
to the destabilization of the compact myelin sheath and to the degeneration seen in MS.
Third, we suggest that the multifunctionality seen in MBP could arise from differences in
the population of energetically distinct ensemble clusters in different conditions which can be
measured as energetically distinct states.
Finally, we hold the expectation that if pseudo-deimination in MBP-UTC8 causes global
structural changes as compared to MBP-UTC1, there will be also a measurable thermodynamic
difference in the population of energetically distinct ensemble clusters between MBP-UTC1 and
MBP-UTC8 (47).
2.2 Objectives
2.2.1 Develop a labelling protocol
Using single cysteine mutants of the MBP-UTC1 and MBP-UTC8 variants, we will
develop a labelling protocol with the fluorescent probe 5-((((2-iodoacetyl) amino) ethyl) amino)
naphthalene-1-sulfonic acid (1, 5-IAEDANS, henceforth just IAEDANS, for simplicity) to create
FRET pairs with the intrinsic single tryptophan at position 113.
20
2.2.2 Compare using CD and anisotropy
We will compare IAEDANS-labelled MBP-UTC1 and MBP-UTC8 using CD
spectroscopy and fluorescence anisotropy measurements to ensure that site-directed
mutagenesis and label incorporation did not alter MBP-UTC1 and MBP-UTC8 variant
secondary structural propensities.
2.2.3 Map intramolecular distances
We will map intramolecular distances using fluorescence resonance energy transfer
(FRET) in full-length MBP-UTC1 and MBP-UTC8 variants, with the goal of determining the
location, flexibility, and distribution of various MBP domains in aqueous buffer, TFE, NaCl,
and 6 M guanidinium chloride, and in the presence of zinc. These distances will complement
our previous EPR, NMR, FTIR, and CD studies of MBP-UTC1 and MBP-UTC8 variants, and
will facilitate the understanding of MBP topology in the myelin sheath.
2.2.4 Combined analysis
Using a combined analysis of CD spectroscopy, ANS fluorescence, and FRET we aim to
determine the minimal applicable mechanism of disorder-to-α-helical-transition of full-length
18.5-kDa MBP. This will be performed on both MBP-UTC1 and MBP-UTC8. We also hope to
determine the free energy change associated with the coil-helix transition of MBP. This
transition will be induced through titration of the membrane-mimetic solvent trifluoroethanol
(TFE) into protein solutions. A full thermodynamic description of the mechanism of disorder-toorder transition is useful both for defining the protein’s conformational landscape, which may
be important in binding-induced folding mechanisms.
2.2.5 Mapping intramolecular distances in order to identify the ensemble
populations
We hope to identify populations of distinct conformations along the transition pathway
which may be instrumental for understanding the conformational landscape of MBP.
21
2.2.6 Structural and thermodynamic comparison
We hope to draw a structural and thermodynamic comparison of MBP-UTC1 and MBPUTC8 in order to better define a physiological role for these isoforms.
2.3 Significance
This work will allow us to define more precisely the role of MBP in the underlying
molecular architecture of CNS myelin, including MBP polymorphism and modifications linked
to multifunctionality. In addition, understanding the molecular basis of induction of
demyelination, and failure to remyelinate in MS, will aid development of therapeutic strategies
for MS.
2.4 Statement of Publication
Parts of this thesis work have been published and several figures are reproduced with
permission as noted (K.A. Vassall, A.D. Jenkins, V.V. Bamm, G. Harauz. “Thermodynamic
analysis of the disorder-to-α-helical transition of 18.5-kDa myelin basic protein reveals
equilibrium intermediate representing the most compact conformation”. Journal of Molecular
Biology, volume 427, pp 1977-1992, 2015, doi: 10.1016/j.jmb.2015.03.011).
22
Chapter 3.0: Experimental Methods
3.1 Chemicals and Reagents
The
reagents
5-((2-[(iodoacetyl)amino]ethyl)amino)naphthalene-1-sulfonic
acid
(IAEDANS) and 1-anilinonaphthalene-8-sulfonic acid (ANS) were obtained from Invitrogen
Life Technology (Burlington, ON, Canada). All other chemicals were reagent grade and acquired
from either Fisher Scientific (Unionville, ON, Canada), or Sigma-Aldrich (Oakville, ON,
Canada).
3.2 Recombinant murine full-length untagged 18.5-kDa MBP-UTC1 and
MBP-UTC8
Seven MBP variants containing single cysteine substitutions (S17C, S44C, S67C, H85C,
S99C, S129C, and S159C) were produced for each MBP charge variant. Targets for mutagenesis
were chosen as in previous EPR studies (118), and were constructed by the polymerase chain
reaction (PCR) with site-directed mutagenesis using the QuickChange® protocol (RocheStratgene, La Jolla, CA). Previously described expression vector(s) pET22b (+) MBP-UTC1 and
MBP-UTC8 were used as the template DNA (87). Primers were supplied by the Guelph
Molecular Supercenter (University of Guelph, ON, Canada) and were supplied at a minimum
concentration ~0.400 nmole/µL. Mutant strand synthesis was performed using a BioRad (ON,
Canada) thermal cycler PCR system. The PCR mutant plasmid products were subjected to
digestion by the restriction enzyme FastDigest DpnI (Fermentas, Ottawa, ON) to remove
parental DNA. A volume of 1 µL of DpnI (10 units) was added to 20 µL of each of the PCR
products, and was placed in a 37°C water bath for 2 hours. Following digestion, plasmid DNA
was stored at -20°C.
Mutant plasmids were transformed via heat shock into competent E. coli-DH5α cells. An
amount of 5 µL of the digested PCR product was added to competent E. coli-DH5 cells and
incubated on ice for 5 minutes. The cells were then transferred to a 42°C water bath for 45
seconds and then placed back on ice for another 2 minutes. Then, 300 μL of LB media was
added and the mixture was incubated at 37°C with shaking for 1 hour. The entire volume of each
mutant culture was plated on agar with 50 µg/mL ampicillin. Plates were incubated overnight at
23
37°C. Both positive and negative control plates were used. Selected E. coli-DH5α colonies from
the ampicillin agar plates for each mutant were inoculated in 10 mL of LB media (+50 µg/ mL
ampicillin) and were incubated with shaking for ~17 hours at 37°C. A miniprep plasmid
extraction kit (Roche Diagnostics, Laval, Quebec) was used according to the provided protocol
for the small-scale extraction of plasmid DNA. The concentration of plasmid DNA was
determined using a POLARstar Omega® (BMG Labtech, NC, U.S.A.) micro-plate reader. The
amount of DNA was quantified by absorbance at 260 nm (A260) using the following equation:
[DNA] = A260 / (ε x l), where (ε) is the extinction coefficient (mL mg-1 cm-1) and (l) is the path
length (cm). The ε coefficient used was 50 mL mg-1 cm-1 and the path length was 1 cm. A
volume of 10 µL of each mutant plasmid was collected and sent for sequencing (Guelph
Molecular Supercenter, University of Guelph, ON).
Mutant plasmids were transformed into E. coli-BL21-Codon Plus (DE3)-RP cells (see
Cloning Techniques). Then 300 µL of LB media was added and cells were incubated at 37°C
with shaking for 1 hour. The cell cultures were added to a 10 mL LB media with 50 µg/mL
ampicillin and 34 µg/mL chloramphenicol, and incubated at 37°C with shaking for ~14 hours.
Each culture was then added to a flask containing 150 mL of LB broth with 50 µg/mL ampicillin
and 34 µg/mL chloramphenicol. The cells incubated at 37°C with shaking until the optical
density at 600 nm (A600) was between 0.6 - 0.8 (arbitrary units, a.u.) as measured by an
Ultraspec-2000® (Pharmacia-Biotech, QC, Canada) with LB broth as a reference. Cultures were
transferred into centrifuge bottles and spun at 2000 rpm (Beckman-Coulter Rotor JA-10) for 15
minutes at 20°C. Supernatant was disposed. Cell pellets were re-suspended into 1.5 L of M9
media with additions of 3 mL of 1 M MgSO4, 150 µL of 1 M CaCl2, 1.5 mL of 50 µg/mL
ampicillin, 1.5 mL of 34 µg/mL chloramphenicol, and 30 mL of 20% glucose solution.
The M9 cultures were incubated at 37°C with shaking until an A600 = 0.7 - 0.9 (A.U.) was
achieved. A 1-mL negative-induction sample was taken for each variant. Protein induction was
accomplished with the addition of 1.5 mL of 0.238 g/mL isopropyl β-D-1-thiogalactopyranoside
(IPTG), and cultures were incubated for ~5 hours. Then 1-mL positive-induction samples were
taken for each variant. The entire cell culture of 1.5 L for each variant was pelleted by
centrifugation at 7000 rpm (Beckman-Coulter Rotor JA-10) for 15 minutes at 4°C. The pellet
24
was frozen at -20°C for storage. Negative and positive induction samples were analyzed by
sodium dodecyl sulphate-polyacrylamide gel electrophoresis (SDS-PAGE).
Cell lysis was performed by re-suspension of each variant cell pellet into 60 mL of lysis
buffer (8 M urea, 100 mM NaH2PO4, 500 mM NaCl, 10 mM Tris-base, 10 mM βmercaptoethanol, 1% v/v Tween-20, pH 8.0) and homogenization. The lysate was frozen at 80°C, thawed, and further stirred at 20°C for 1 hour. Lysate was then spun at 15,000 rpm using a
JA-25.5 Beckman-Coulter rotor for 20 minutes at 4°C. Supernatant was collected and dialyzed
for more than 5 hours against 2L IEX buffer (6 M urea, 80 mM glycine, pH 10.0) using 60008000 Da molecular weight cut-off (MWCO) dialysis tubing.
Purification of MBP variants was achieved by fast protein liquid chromatography (FPLC)
using a BioRad-DuoLogic® apparatus and a 1-mL Pall AcroSep® cation exchange column. The
IEX buffer filtered through a 0.45-µm filter composed the mobile phase. Detection was
performed at 280 nm, the flow rate was 1 mL/minute, and the column temperature was 4°C. A
linear gradient of salt in IEX buffer (0-500 mM) was developed over the total volume of 50 mL
with a flow rate of 2 mL/minute. The 1.5-mL fractions were collected automatically. Pure
fractions were observed and confirmed by SDS-PAGE using a standard protein ladder. Isolated
fractions then were transferred into 6000-8000 Da dialysis tubing and dialyzed against 2 x 2L of
Buffer A (50 mM Tris-base, 250 mM NaCl, pH 7.4), 2 x 2L of Buffer B (100 mM NaCl), and 4 x
2L distilled water. Buffer changes were made every ~5 hours. Following dialysis, the variant
proteins were collected and lyophilized.
After ion exchange chromatography, the MBP single cysteine-substituted variants were
treated with a 10-fold molar excess of β-mercaptoethanol for 20 min at 25°C, in order to ensure
that all cysteine residues were reduced. Finally, unmodified MBP as well as the cysteine variants
were further purified by reversed-phase high-performance liquid chromatography (HPLC), using
a Waters (Mississauga, ON) apparatus with a Symmetry 300, C18, 5 mm, 4.6 x 250 mm column.
Acetonitrile constituted the mobile phase, and 0.1% trifluoroacetic acid was used as the ionpairing agent. Detection was performed at 214 nm and 280 nm, the flow rate was 1 mL/minute,
and the column temperature was 40°C. The elution gradient initiated at 75% Solvent A (sterile
water and 0.1% TFA) and 25% Solvent B (acetonitrile and 0.1% TFA). The gradient was run at
25
1% Solvent B per minute for 15 minutes, and then increased to 100% Solvent B at 2 mL/minute
for over a period of 2 minutes. Fractions representing the main peak were manually collected.
The collected protein fractions were freeze-dried using a Labconco FreeZone freeze-dryer
(Fisher Scientific, Markham, ON, Canada). Upon solubilising, protein purity was assessed by
SDS-PAGE and HPLC. Reduced and non-reduced protein samples were analyzed by SDSPAGE under non-reducing conditions. The final sample purity was analyzed by HPLC using the
same chromatographic conditions as above, and was determined to be greater than 95% for all
MBP variants.
3.3 Labelling of Cysteine-substituted MBP-UTC1 and MBP-UTC8 variants
with IAEDANS
Freeze-dried MBP cysteine variants were re-solubilized in a solution containing 20 mM
Hepes-NaOH, pH 7.4, to a final protein concentration of ~1 mg/mL (~54 μM). Next, IAEDANS
was added to this protein solution to a final concentration that was 20-fold in excess of the
protein (by moles). The labelling reaction occurred for 3 hours at 25° C, and was terminated
using 12.5 mM β-mercaptoethanol. Unreacted IAEDANS and β-mercaptoethanol were removed
by HPLC (utilizing the same method that was used for MBP purification), and the collected
protein fraction was then freeze-dried. Protein concentrations were determined by measuring the
absorbance of re-solubilized protein in water, using an extinction coefficient at 276 nm of 8421.4
M-1 cm-1 for MBP-UTC1 and 8432.9 M-1 cm-1 for MBP-UTC8, which was experimentally
determined by resuspending a known mass of protein in a known volume of water, and
measuring absorbance using the following equation:
=
(Eq. 1 )
To determine the concentration of protein and IAEDANS in the labelled protein samples,
multifactorial analysis of UV absorption spectra in water was used and produced the following
system of equations:
[] =
 −([]×


(Eq. 2 )
26
[] =
 −([]×


(Eq. 3 )
where εMBP-UTC1276 (extinction coefficient of MBP at 276 nm) is 8,421.4 M-1 cm-1, εMBP-UTC1260
(extinction coefficient of MBP at 260 nm) is 5,544.9 M-1 cm-1, εMBP-UTC8276 (extinction coefficient
of MBP at 276 nm) is 8432.9 M-1 cm-1, εMBP-UTC8260 (extinction coefficient of MBP at 260 nm) is
5685.9 M-1 cm-1, εI276 (extinction coefficient of IAEDANS at 276 nm) is 1,633.5 M-1 cm-1, and
εI260 (extinction coefficient of IAEDANS at 260 nm) is 13,423.5 M-1 cm-1. The solution to these
systems of equations is:
[MBP-UTC1] = 129.1A276 − 15.7A260
(Eq. 4 )
[IAEDANS-MBP-UTC1] = 81.0A260 − 53.3A276
(Eq. 5 )
[MBP-UTC8] = 129.4A276 − 14.1A260
(Eq. 6 )
[IAEDANS-MBP-UTC8] = 79.0A260 − 51.2A276
(Eq. 7 )
In addition to Equation 2, the concentration of MBP was also determined by the
bicinchoninic acid (BCA) assay (119). In the BCA assay, a protein standard was prepared by
dissolving a known mass of freeze-dried MBP into a known volume of water. The calculated
concentrations of MBP and IAEDANS were subsequently used in the determination of labelling
efficiency (fA) using the following equation:
 =
[]
[]
(Eq. 8 )
3.4 Circular Dichroism Measurements
All CD data were acquired using a Jasco J-815 spectropolarimeter (Japan Spectroscopic,
Tokyo, Japan) using a quartz cuvette with a 1-mm path length (Hellma, Concord, ON, Canada),
with thermostatting at 25°C using a Jasco PTC-424S/15 Peltier temperature controller (Japan
Spectroscopic, Tokyo, Japan). The CD spectra were collected with solutions containing 10 µM
protein, 5 mM HEPES-NaOH, pH 7.4, and various concentrations of TFE (0-30%, v/v). The
reagent tris(2-carboxyethyl)phosphine (TCEP) was added to a final concentration of 0.5 mM
when the samples contained cysteine-substituted variants, in order to maintain a reducing
environment. Each spectrum was collected at a scan rate of 50 nm min-1, and represents an
27
average of 4 scans. Corresponding buffer scans were subtracted from sample scans before
analysis and presentation of the data. Experiments were typically repeated 2 or more times with
independently-prepared samples.
3.5 Fluorescence measurements
All fluorescence data were obtained with a Fluoromax Spectrophotometer (Photon
Technology International, Edison, NJ, USA). Temperature was maintained at 25° C with a
circulating water bath in all experiments.
3.6 Steady-state Fluorescence Anisotropy Measurements
Steady-state fluorescence anisotropy (r) measurements were performed using 2-nm
excitation, and 8-nm emission, slit widths. Time-based polarization was measured for both
tryptophan and IAEDANS using excitation at 295 nm and emission at 350 nm, and excitation at
336 nm and emission at 500 nm, respectively. The “T-format” was used for the simultaneous
detection of intensities for vertically- (IVV) and horizontally- (IVH) polarized emitted light
following sample excitation by vertically-polarized light. The “G” factor was measured as
IVH/IHH, where IVH and IHH are the intensities of the vertically- and horizontally-polarized emitted
light, respectively, from horizontally-polarized excitation light. A signal integration time of 30 s
with the rate of 10 points per second was used, and each anisotropy value is the average of three
replicates. The final anisotropy (r) values for all labelled cysteine variants for both MBP-UTC1
and MBP-UTC8 were calculated as shown in the subsequent equation:

− 
 =  + 


(Eq. 9 )
3.7 ANS fluorescence
The ANS fluorescence in the absence and presence of full-length unmodified MBP was
measured at different concentrations of TFE. In these experiments, stock solutions of ANS were
prepared by dissolving ANS in water to a final concentration of ~1.5 mM. This solution was
centrifuged at 18,000 x g for 10 minutes to remove undissolved ANS, and the final stock
concentration of ANS was determined by measuring the absorbance at 350 nm using an
extinction coefficient of 5,000 M-1 cm-1 (120-122). From these stock solutions, sets of samples
28
were prepared containing 50 µM ANS, 20 mM HEPES-NaOH, pH 7.4, concentrations of TFE
ranging from 0-55% (v/v), and either 0 or 7.5 µM MBP. Measurements were made using a
quartz cuvette (Hellma, Concord, ON, Canada) with a 3-mm path-length with the excitation
wavelength set to 350 nm (2-nm slit width), and emission was measured from 400-600 nm (6-nm
slit width). In addition to collection of full spectra, emission was also measured at the single
wavelength of 480 nm and presented as an average of 30 readings taken at a rate of 1 s-1. All
experiments were repeated a minimum of 3 times on independently-prepared samples.
3.8 Fluorescence resonance energy transfer measurements
All fluorescence data were obtained with a Fluoromax Spectrophotometer (Photon
Technology International, Edison, NJ, USA). Temperature was maintained at 25° C with a
circulating water bath in all experiments. In FRET experiments involving MBP, the lone intrinsic
tryptophan (W113) was used as the donor, and an IAEDANS-labelled cysteine served as the
acceptor. Prepared samples contained 2-4 μM unlabelled or IAEDANS-labelled MBP, 20 mM
HEPES-NaOH, 0.5 mM TCEP, at pH 7.4, and were measured in various environmental
conditions. Samples were prepared under various environmental conditions such as; in presence
of 6 M GdmCl, 100 mM NaCl, Zn2+ (10x excess Zn to protein in moles), and finally at varying
concentrations of TFE (0%, 10%, 15%, or 30% TFE (v/v)). Measurements were performed using
a 5-mm path-length quartz cuvette (Hellma, Concord, ON, Canada). Tryptophan excitation was
at 295 nm, and emission was measured from 305 nm to 550 nm; slit widths were each set to 4
nm. Each spectrum was recorded as an average of 3 scans and manually blank-corrected by
subtracting buffer scans. The FRET efficiency (EFRET) was calculated based on the extent of
tryptophan fluorescence quenching by IAEDANS using the following equation:
 = ( −



) ( )

(Eq. 10 )
where DA and D are the tryptophan fluorescence intensities at 355 nm of labelled and unlabelled
MBP, respectively, and ƒA is the labelling efficiency (fraction of labelled protein). All FRET
efficiency measurements were repeated using a minimum of 3 independently-made samples,
using protein derived from at least 2 independent IAEDANS labelling reactions.
29
After obtaining FRET efficiencies, the distance (r) between the donor and acceptor was
then calculated using the following equation:
 =  (
− /

)
(Eq. 11 )
where Ro is the Förster distance, which was taken as 22.0 Å for the tryptophan-IAEDANS pair
(46). Measurements in the various environments were then compared to the Flory random-coil
model for denatured proteins in 6 M guanidinium chloride (123).
3.9 TFE-titration curves monitored by tryptophan fluorescence quenching
The MBP variants S17C-MBP and S159C-MBP were further analyzed by obtaining
equilibrium TFE-titration curves monitored by the change in tryptophan fluorescence quenching
by IAEDANS. In these quenching experiments, separate samples of unlabelled and IAEDANSlabelled MBP were excited at 295 nm, and emission was measured at 355 nm. A 3-mm quartz
cuvette was used (Hellma, Concord, ON, Canada), and slit widths were 2 nm for excitation and 8
nm for emission. A total of 30 readings were acquired over 30 s, which were subsequently
manually averaged. The labelled fluorescence signal was then subtracted from the unlabelled
signal to determine quenched fluorescence. Samples were prepared identically to those used for
FRET measurements (see above) with the exception that the range of TFE concentrations used
was broader, ranging from 0-35% (v/v), and many more TFE concentrations were considered.
Samples were incubated at 25 °C for a minimum of 2 h before measuring. Similarly to the CDmonitored curves, reversibility was confirmed by collecting reverse-titration curves in which
samples were incubated at high concentration of TFE (~40%) for ~1 h, and then diluted to lower
concentrations before measuring.
3.10 Quantitative analysis of TFE-titration curves
All quantitative analysis was performed using OriginPro 8 (OriginLab Corporation,
Northampton, MA, USA). In the analysis of TFE-titration curves the curves were fit to a 3-state
model (disordered↔intermediate↔α-helical). In this analysis, we assume a linear dependence of
ΔG (change in free energy) with TFE concentration for each protein variant (124-126), in which
the equilibrium constant (KTFE) at any given TFE concentration can be defined as:
30
 = 
−

=
[]−

=
[]−[]

(Eq. 12 )
where, ΔGTFE is the free energy change at a given concentration of TFE, ΔGH2O is the free energy
change in the absence of TFE (i.e., in 20 mM Hepes-NaOH, pH 7.4), m is the dependence of
ΔGTFE on [TFE], [TFE]mid is the concentration of TFE where the transition is half completed, R is
the universal gas constant, and T is the temperature in degrees Kelvin. For both a 2-state and 3state equilibrium model, the conformational transitions can be mathematically related to the
measured optical signal (127). The 2-state transition was defined by:
 =
( + [])+( + [])
(Eq. 13 )
+
where Yobs is the measured optical signal, YD and YH are the intercepts of the disordered
conformation and helical conformation baselines, respectively, and SD and SH are the slopes of
the disordered conformation and α-helical conformation baselines, respectively. The values of
YD, YH, SD, and SH were determined by linear regression analysis and subsequently fixed when
fitting to Equation 9, while m and [TFE]mid were floating parameters. The 3-state transition was
defined by:
 =
 
( + [])+( + [])
 +( + []) 
 
+
 + 
(Eq. 14 )
where YI and SI are the slopes and intercept of the intermediate, respectively, and K1TFE and K2TFE
are
the
equilibrium
constants
of
the
first
(disordered↔intermediate)
and
second
(intermediate↔α-helical) transitions, respectively. In 3-state fitting, all the CD and fluorescence
quenching curves for both S17C-MBP and S159C-MBP were globally fit with ΔG1H2O and
ΔG2H2O (free energy change for the 1st and 2nd transitions, respectively), and m1 and m2 (the
dependence of ΔG on TFE for the 1st and 2nd transitions, respectively) as shared floating
parameters. For the CD data, the parameters defining the disordered and α-helical conformation
baselines were determined by linear regression, and subsequently fixed during global fitting. It
was assumed that the CD signal of the intermediate conformation is indistinguishable from the
disordered conformation, and hence the baseline values of the intermediate conformation were
set to be equal to those of the disordered conformation during fitting. For the fluorescence
31
quenching data, it was assumed that all slopes were zero; however, the intercepts of all three
baselines were left as floating parameters.
32
Chapter 4.0: Results
*Statement of Work Performed
The work described in this chapter was performed entirely by the author, except for the
TFE titration curves monitored by CD spectroscopy, ANS-binding, and tryptophan fluorescence
quenching, along with subsequent data fitting to equilibrium thermodynamic models.
4.1 MBP Site-directed mutagenesis
Using site-directed mutagenesis via PCR, seven MBP-UTC1 and seven MBP-UTC8
protein constructs were produced, each containing single-cysteine substitution at positions:
S17C, S44C, S67C, H85C, S99C, S129C, and S159C (Figure #4). Primers for mutagenesis were
designed according to the QuikChange® protocol, ordered, and supplied by the Guelph Molecular
Supercenter. The PCR conditions for mutation were performed as per the protocol described in
the previous chapter. Transformation by heat-shock into E. coli-DH5α cells was confirmed using
agar plates containing ampicillin as indicated by positive and negative controls. Plasmid DNA
was purified using a mini-prep kit for all fourteen protein constructs, and their concentrations
were determined via A260 measurements. Sequences of the isolated plasmids were confirmed by
the Guelph Molecular Supercenter, Guelph, ON.
4.2 MBP purification
Cysteine-containing MBP variants were successfully transformed and expressed in E.
coli BL-21(DE3)-Codon Plus-RP cells. Induction by IPTG produced significant quantities of
variant protein(s), as confirmed by the comparison of the SDS-PAGE profile of induced and
uninduced controls (Figure #5A). Extraction and purification of MBP from the bacterial cells
were performed under denaturing conditions as described by (87) with minor changes to the
protocol for purification. According to the chromatogram, MBP was eluted from the column at
an approximate salt concentration of 160 mM. As the buffer used had a pH of 10.0, we would
expect that all bacterial proteins should have been removed by FPLC, and therefore, it is likely
33
Figure #4. Amino acid sequence and secondary structural map of murine 18.5-kDa MBP-UTC1 and MBP-UTC8
(168 residues). MBP-UTC1 and MBP-UTC8 are intrinsically-disordered in aqueous solution but have three distinct
segments which have strong α-helical propensity α1, α2, α3 (top to bottom). These regions consist of residues which can
form a amphipathic α-helices when associated with lipids or in the presence of membrane-mimetic solvents. For purposes
of this study, recombinant murine MBP-UTC1 and MBP-UTC8 constructs containing a series of point mutations at
positions S17C, S44C, S67C, H85C, S99C, S129C, and S159C were generated. Cysteinyl-containing proteins were
utilized for labelling with the thiol-reactive fluorescent probe 1,5-IAEDANS, in order to create a FRET donor-acceptor
pair with an intrinsic tryptophan at position 113. Adapted and modified with permission from Vassall et al. (2013).
34
A
1
B
2
3
1 2
12
C
3
4
5
6
7
8
9 10 11
1
2
3
Figure #5. Purification of MBP-UTC1 and MBP-UTC8. (A) 14% SDS-PAGE gel for MBPC1 cysteine-containing variants in pET22b(+) expression vector in E. coli BL21-CodonPlus
(DE3)-RP cells induced with 1 mM IPTG. Samples were prepared by dissolving variant cell
pellets in 5x reducing loading dye and boiled at >100°C for 2 minutes. Lane A-1 = Post-IPTG
induction, Lane A-2 = Pre-IPTG induction, Lane A3 = Pink Plus Prestained Protein Ladder from
BioRadTM. The gel was stained with Coomassie brilliant blue for 15 minutes, and destained with
2% methanol and 1% acetic acid. Results show overexpressed 18.5-kDa MBP as indicated in
Lane #3 by the large band occurring at ~21 kDa. Data shown are for the MBP-C1-S17C variant.
(B) 14% SDS-PAGE gel for FPLC-pure MBP-UTC1 cysteine-containing variant. Samples were
prepared as above. Lanes B2-B12 are shown to contain overexpressed MBP-UTC1 protein
variant (~18.50 kDa). Data shown are for the MBP-UTC1-S17C variant. (C) 14% SDS-PAGE
gel for MBP-UTC1 cysteine containing variant reduction by 125 mM β-mercaptoethanol.
Fractions were analyzed by SDS-PAGE. Samples of each fraction were prepared by addition of
5x reducing loading dye to make 20 µL samples, and were then boiled at >100°C for 2 minutes.
Lane 1 = Reducing loading buffer. Lane 2 = Non-reducing loading buffer. Lane 3 = Postreduction with non-reducing loading buffer. Lane 4 =Pink Plus Prestained Protein Ladder from
BioRad. The reduction of dimers is shown be successful as indicated by the absence of bands
occurring between 42 and 29 kDa in Lane 3. Data shown are for the MBP-C1-S17C variant.
35
that the observed low molecular mass impurities are truncations of full-length 18.5-kDa MBP.
High molecular mass impurities were observed on the SDS-PAGE gel following FPLC
purification (Figure #5B). Regarding these species, one possible explanation is the result of
cross-linking of free cysteine residues, resulting in dimerization or oligomerization of MBP
monomers. Analytical samples for each variant confirmed the presence of minor contaminants
from dialyzed fractions of FPLC-pure variant protein. Purification of samples was successful as
indicated by the HPLC chromatogram (data not shown). According to the chromatogram, MBP
was eluted at approximately 37% acetonitrile, and exhibited a double-peak pattern. This pattern
is indicative of MBP monomers and dimers. The nature of MBP dimers are likely the result of
disulfide bonds between the introduced cysteine residues. In order to achieve high efficiency
labelling, these dimers must be reduced. Reduction of variant protein dimers was achieved by
incubation with 125 mM β-mercaptoethanol and subsequent analysis by SDS-PAGE (Figure
#5C) revealed only monomeric MBP. In order to remove excess β-mercaptoethanol the samples
were purified by HPLC under the same chromatographic conditons as mentioned previously. The
final purity was assessed by HPLC and was determined to be greater than 95% for all MBP
variants (data not shown).
4.3 MBP labelling with IAEDANS
Protein labelling was achieved by incubation of variant protein with 15-20x molar excess
IAEDANS fluorescent label in 20 mM Hepes Buffer at pH 7.4. The reaction was terminated
using 0.125 mM β-mercaptoethanol after a 4-hour period. Excess label, and the βmercaptoethanol used to terminate the reaction, were successfully removed by HPLC under the
same chromatographic conditons as stated above, though the absorbance channels that were
monitored were altered to 214 nm and 336 nm, with 336 nm representing the IAEDANS label
(data not shown). The fractions representing the main, labelled protein peak were manually
collected and lyophilized.
36
4.4 Determining MBP labelling efficiency
Absorbance wavelength spectra were collected from 220 nm to 700 nm for both
unlabelled and labelled samples (data not shown). A characteristic 3-peak shape was observed
for all labelled MBP-UTC1 and MBP-UTC8, suggesting the successful linking of label to
protein. In order to resolve the peaks and determine the amount of bound label, a system of
equations was developed to subtract the IAEDANS contribution at A276 to determine protein
concentration, as well as to subtract the protein contribution at A260 to determine the
concentration of the amount of bound IAEDANS label (Section 3.3). A series of extinction
coefficients were determined for both MBP-UTC1 and MBP-UTC8, as well as IAEDANS label,
which were determined by solubilizing known weights in water. The final values calculated are
listed: εMBP-UTC1276 (extinction coefficient of MBP-UTC1 at 276 nm) is 8,421.4 M-1 cm-1, εMBPUTC1
260
(extinction coefficient of MBP-UTC1 at 260 nm) is 5,544.9 M-1 cm-1, εMBP-UTC8
276
(extinction coefficient of MBP-UTC8 at 276 nm) is 8432.2.1 M-1 cm-1, εMBP-UTC8260 (extinction
coefficient of MBP-UTC8 at 260 nm) is 5685.9 M-1 cm-1. The values for IAEDANS are εI276
(extinction coefficient of IAEDANS at 276 nm) is 1,633.5 M-1 cm-1, and εI260 (extinction
coefficient of IAEDANS at 260 nm) is 13,423.5 M-1 cm-1. Both unlabelled stock variant protein
and labelled variant protein concentrations were successfully determined via this multifactorial
analysis, which was confirmed by performing the bicinchoninic acid protein assay (BCA) for
protein concentration. The BCA Protein Assay Kit (Thermo-Scienftic, ON, Canada) was
performed as per the provided protocol, with the standard being modified to be either pure MBPUTC1 or pure MBP-UTC8 protein (128). The standard concentration range used was from 200
µg/mL to 0.5 µg/mL, and was developed by serial dilution of pure MPB-UTC1 or MBP-UTC8
with known concentration, which were determined by solubilizing known weights in water. The
labelling event was successful providing a labelling efficiency of >60%.
4.5 Circular Dichroism Measurements of MBP
In order to reaffirm that neither the incorporation of the cysteine residue nor the labelling
with IAEDANS fluorescent probe resulted in alteration of MBP-UTC1 and MBP-UTC8 protein
structure, we performed CD spectroscopy. Previously, NMR (16, 84, 129) and CD (130)
spectroscopic studies have demonstrated that MBP is a disordered, monomeric protein in
37
solution (59, 77). However, in the presence of membrane-mimetic solvents such as TFE, MBP
has been shown to adopt significant α-helical secondary structure (131). Previous studies using
site-directed spin labelling and EPR spectroscopy (57, 82, 132), solid-state NMR spectroscopy in
a lipid milieu (80, 84), as well as solution NMR spectroscopy in 30% TFE (v/v) (133, 134), have
all demonstrated that α-helical formation is localized predominantly in three regions
(approximately T33-F43, P82-V91, and Y141-K153). The disorder-to-α-helical transition of
MBP can be directly monitored by measuring the change in far-UV CD spectra with increasing
concentrations of TFE. This was done originally for natural protein purified from brain
representing a heterogeneous population of modified forms, then again previously by us for a
hexa-histidine tagged construct of MBP (135), and has been repeated here for the MBP-UTC1
(Figure #6) and MBP-UTC8 (data not shown) cysteine variants. Far-UV CD spectra of untagged
MBP-UTC1 demonstrate here that there is pronounced α-helical content at TFE concentrations
>25% (v/v) and conversely, there is relatively little ɑ-helical content between 0-10%. This
disorder-to-α-helical transition of MBP in TFE is protein concentration independent, suggesting
that α-helical formation is not accompanied by oligomerization of the protein (Figure #7).
Significantly, there is an isodichroic point at 203 nm, which is usually suggestive of a 2-state
transition (136, 137). However, we were cautious in analyzing the data via a 2-state model before
analyzing the disorder-to-α-helical transition using additional probes (see below).
The CD spectra collected at 0% TFE of the labelled UTC1 and UTC8 MBP variants
displayed the typical minimum of ∼200 nm, indicating a random coil structure. Conversely, CD
spectra of MBP-UTC1 and MBP-UTC8 protein samples measured in 30% TFE displayed a αhelical rich secondary structure profile. It should be noted that far-UV CD spectra, of the hexahistidine tagged construct of MBP, revealed that a second transition also occurs at ~90% TFE
(101). We have identified a similar, well-separated, second transition in the untagged protein
(data not shown), but have not considered it in our analysis in light of prior experimental
evidence that α-helical formation above 40% TFE may not be physiologically relevant (122).
Overall, control CD spectra collected for all the variants at 0% and 30% TFE suggest that the
presence of the cysteine-substitutions and label do not significantly perturb the protein’s native
secondary structure composition (Figure #8).
38
Figure #6. Changes in circular dichroic (CD) spectra of 18.5-kDa MBP-UTC1 with
increasing trifluoroethanol (TFE) concentration. Samples contained 10 µM MBP, 5 mM
Hepes-NaOH, pH 7.4, and measurements were made at 25 °C. Blank-subtracted representative
data are shown and data were normalized to mean residue ellipticity. The α-helical content
increases with increasing TFE concentrations (0% to 35%, v/v), and there is an isodichroic point
at ~203 nm.
39
Figure #7. Comparison of TFE-titration curves of IEADANS-labelled MBP-UTC1-S17C
and S159C 18.5-kDa MBP with the unmodified protein. The representative titration curves
were monitored by change in circular dichroic signal at 222 nm and normalized to facilitate
comparison. Unmodified MBP samples contained 4 μM, 7.7 μM, and 25 μM protein
concentration, as indicated, whereas S17C and S159C MBP samples contained 4 μM protein. All
solutions were buffered with 20 mM Hepes-NaOH, pH 7.4, and measurements were performed
at 25 °C.
40
UT-MBP
H85C
20
S17C
S99C
S44C
S129C
A
B
C
D
S67C
S159C
10
-10
-20
20
-3
2
-1
[] *10 (deg cm dmol )
0
10
0
-10
-20
200
225
250 200
225
250
Wavelength (nm)
Figure #8. Circular dichroic (CD) spectra of unmodified 18.5-kDa MBP-UTC18 and MBPUTC IAEDANS-labelled cysteine variants. The labeling efficiency of the MBP cysteinesubstituted variants was >60% in all cases. All samples contained 10 µM MBP, 5 mM HepesNaOH, pH 7.4, and measurements were made at 25 °C. Blank-subtracted data are shown, and
data were normalized to mean residue ellipticity. Panels (A) and (B) show far-UV CD spectra
collected in 0% and 30% TFE for MBP-UTC1 variants (v/v), respectively. Panels (C) and (D)
show far-UV CD spectra collected in 0% and 30% TFE for MBP-UTC8 variants (v/v),
respectively. All variants displayed consistent levels of secondary structure in both 0% and 30%
TFE.
41
4.6 ANS fluorescence in MBP
Although it is clear that MBP forms amphipathic α-helices in at least three different
segments in the presence of TFE, there is little known about the evolution of global structure that
may accompany α-helix formation. Here, we obtained preliminary information on the evolution
of tertiary structure for both MBP-UTC1 and MBP-UTC8 by utilizing the well-known
hydrophobic fluorescent dye ANS. ANS is known to bind to solvent-exposed hydrophobic
pockets in proteins, resulting in a blue shift in fluorescence maxima and an enhancement in
intensity. Hence, this dye is commonly used to assess conformational changes within various
proteins (120, 121), including IDPs (138, 139).
We found that MBP-UTC1 in the absence of TFE showed modest enhancement of ANS
fluorescence. There is a sigmoidal increase in the wavelength of maximal ANS fluorescence
enhancement from 0-45% TFE (Figure #9A), with maximal enhancement occurring close to 477
nm. Marked changes in the pattern of enhancement were seen when TFE was titrated into the
system (Figure #9B). Additionally, there is a substantial increase in the intensity of enhancement
between 5 and 10% TFE, compared to 0% TFE, and intensity is at a maximum at 10-17.5% TFE
(Figure #9C). From 17.5-30% TFE, there is a steep decline in ANS fluorescence enhancement
intensity, and from 30-45% TFE there is virtually no enhancement. Results for MBP-UTC8 were
largely consistent with MBP-UTC1 (Figure #10) though maximal ANS enhancement was shown
to occur at slightly lower percentages of TFE.
Overall, the pattern of ANS fluorescence enhancement between 0 and 45% TFE is bellshaped, suggesting the presence of one or more intermediates along the disorder-to-α-helical
pathway. This finding is in contrast to the CD results in which an isodichroic point at 203 nm is
present when far-UV scans of MBP are conducted at different TFE concentrations, suggesting a
simple 2-state transition. These results can be reconciled if we consider that the intermediate
conformer may have a similar local secondary structure composition to the disordered
conformer, but a different global structure. Thus, it may be difficult to differentiate between the
disordered and intermediate conformations in the far-UV CD scans, as they are only sensitive to
changes in secondary structure (124).
42
Figure #9. Change in ANS fluorescence enhancement by 18.5-kDa MBP upon TFE
titration. Samples contained 20 mM Hepes-NaOH, pH 7.4, 50 µM ANS, and 7.5 µM
MBP, and measurements were conducted at 25 °C with excitation at 350 nm. All data are
presented after subtraction of corresponding data from samples that did not contain MBP.
Panel (A) shows the effect of increasing TFE concentration on the enhancement of ANS
fluorescence intensity by MBP (data are normalized) whereas panel (B) shows the
corresponding wavelengths of maximal enhancement. Panel (C) shows the enhancement at
480 nm for each concentration of TFE relative to the enhancement in 0% TFE. Data points
are an average of a minimum of 3-independently-made samples and error bars are standard
deviations.
43
Relative Fluorescence Enhancement.
[TFE] (% v/v)
Figure #10. Change in ANS fluorescence enhancement by 18.5-kDa MBPUTC8-S17C upon TFE titration. Samples contained 20 mM Hepes-NaOH, pH
7.4, 50 µM ANS, and 7.5 µM MBP, and measurements were conducted at 25 °C
with excitation at 350 nm. All data are presented after subtraction of corresponding
data from samples that did not contain MBP. The enhancement at 480 nm for each
concentration of TFE relative to the enhancement in 0% TFE. Data points are an
average of a minimum of 3 independently-made samples and error bars are
standard deviations.
44
Tables #1A and #1B. Steady-state fluorescence anisotropy (r) measurements of MBP-UTC1 (A) and MBP-UTC8 (B).
Time-based polarization was measured for both tryptophan and IAEDANS using “T-format” was used for the simultaneous detection
of intensities for vertically (IVV) and horizontally (IVH) polarized emitted light. “G” factor was measured as IVH/IHH, where IVH
and IHH are the intensities of the vertically and horizontally polarized emitted light, respectively, from horizontally polarized
excitation light. A signal integration time of 30 s with the rate of 10 points per second was used, and each anisotropy value is the
average of three replicates. Final values for (r) were calculated via equation #9.
Table 1A
TFE concentration
UTC1-S17C -Trp
UTC1-S44C-Trp
UTC1-S67C-Trp
UTC1-H85C-Trp
UTC1-S99C
Trp
UTC1-S129C-Trp
UTC1-S159C-Trp
0% TFE
0.038 +/- 0.052
0.032 +/- 0.057
0.032 +/- 0.054
0.039 +/- 0.052
0.048 +/- 0.048
0.043 +/- 0.051
0.030 +/- 0.052
10% TFE
0.053 +/- 0.049
0.062 +/- 0.057
0.064 +/- 0.050
0.054 +/- 0.049
0.064 +/- 0.046
0.063 +/- 0.043
0.057 +/- 0.049
15% TFE
0.059 +/- 0.043
0.063 +/- 0.058
0.063 +/- 0.048
0.072 +/- 0.053
0.071 +/- 0.053
0.065 +/- 0.046
0.067 +/- 0.047
30% TFE
0.054 +/- 0.035
0.050 +/- 0.05
0.052 +/- 0.047
0.042 +/- 0.048
0.042 +/- 0.048
0.056 +/- 0.042
0.056 +/- 0.046
TFE concentration
UTC1-S17CIAEDANS
UTC1-S44C-IAEDANS
UTC1-S67C-IAEDANS
UTC1-H85C-IAEDANS
UTC1-S99C-IAEDANS
UTC1-S129C-IAEDANS
UTC1-S159CIAEDANS
0% TFE
0.017 +/- 0.025
0.022 +/- 0.023
0.023 +/- 0.024
0.026 +/- 0.022
0.024 +/- 0.018
0.022 +/- 0.018
0.018 +/- 0.023
10% TFE
0.029 +/- 0.021
0.038 +/- 0.022
0.037 +/- 0.022
0.036 +/- 0.020
0.032 +/- 0.016
0.034 +/- 0.016
0.032 +/- 0.020
15% TFE
0.036 +/- 0.019
0.040 +/- 0.022
0.040 +/- 0.021
0.049 +/- 0.019
0.037 +/- 0.015
0.036 +/- 0.017
0.037 +/- 0.018
30% TFE
0.034 +/- 0.015
0.031 +/- 0.017
0.031 +/- 0.021
0.034 +/- 0.017
0.034 +/- 0.014
0.034 +/- 0.005
0.030 +/- 0.017
45
Table 1B
TFE
concentration
UTC8-S17C -Trp
UTC8-S44CTrp
UTC8-S67CTrp
UTC8-H85CTrp
UTC8-S99CTrp
UTC8-S129C-Trp
UTC8-S159C-Trp
0% TFE
0.031 +/- 0.082
0.035 +/- 0.034
0.029 +/- 0.085
0.039 +/- 0.069
0.059 +/- 0.073
0.116 +/- 0.074
0.035 +/- 0.073
10% TFE
0.035 +/- 0.074
0.062 +/- 0.05
0.023 +/- 0.071
0.048 +/- 0.074
0.030 +/- 0.073
0.037 +/- 0.077
0.084 +/- 0.070
15% TFE
0.042 +/- 0.074
0.046 +/- 0.048
0.027 +/- 0.072
0.065 +/- 0.073
0.078 +/- 0.067
0.026 +/- 0.071
0.143 +/- 0.074
30% TFE
0.031 +/- 0.075
0.033 +/- 0.045
0.051 +/- 0.077
0.028 +/- 0.073
0.029 +/- 0.074
0.107 +/- 0.074
0.046 +/- 0.074
TFE
concentration
UTC8-S17C-IAEDANS
UTC8-S44C-IAEDANS
UTC8-S67C-IAEDANS
UTC8-H85C-IAEDANS
UTC8-S99C-IAEDANS
UTC8-S129C-IAEDANS
UTC8-S159C-IAEDANS
0% TFE
0.057 +/- 0.059
0.037 +/- 0.059
0.031 +/- 0.063
0.047 +/- 0.052
0.055 +/- 0.053
0.051 +/- 0.049
0.42 +/- 0.47
10% TFE
0.041 +/- 0.047
0.046 +/- 0.033
0.014 +/- 0.050
0.047 +/- 0.048
0.029 +/- 0.052
0.049 +/- 0.048
0.081 +/- 0.061
15% TFE
0.046 +/- 0.048
0.051 +/- 0.045
0.026 +/- 0.045
0.065 +/- 0.056
0.065 +/- 0.048
0.039+/- 0.053
0.110 +/- 0.054
30% TFE
0.043 +/- 0.045
0.039 +/- 0.042
0.048 +/- 0.043
0.038 +/- 0.048
0.050 +/- 0.060
0.040 +/- 0.048
0.053 +/- 0.047
46
4.7 Steady-state Fluorescence Anisotropy Measurements of MBP
Tables #1A and #1B show the apparent fluorescence anisotropy values for MBP-UTC1
and MBP-UTC8 IAEDANS-labelled protein variants. Both tryptophan and cysteine-bound
IAEDANS anisotropy values were measured for labelled proteins in various concentrations of
TFE. All values that were determined showed relatively low fluorescence anisotropy values
(<0.075), consistent with solvent-exposed sites and a high degree of probe mobility (109).
4.8 Fluorescence resonance energy transfer (FRET) measurements of MBP
The fluorescence spectra of unlabelled variants, with excitation at 295 nm, are consistent
with the presence of a solvent-exposed tryptophan (residue Trp113) in both MBP-UTC1 and
MBP-UTC8 (46), and show a fluorescence maximum at ~350 nm. The corresponding
fluorescence spectra of the IAEDANS-labelled protein also have a fluorescence maximum at
~350 nm, but with diminished intensity, while there is another large peak in the spectra at ~500
nm, indicating resonance energy transfer between the donor tryptophan and the acceptor
IAEDANS, even as the protein concentrations remained identical at 2.5 µM. The FRET
efficiency between the lone intrinsic tryptophan at position 113 and the IAEDANS label at each
of the seven positions was determined from the extent of tryptophan fluorescence quenching, and
the distances between the label and Trp113 were also determined for both MBP-UTC1 and
MBP-UTC8 (Table #2A and #2B). The FRET measurements were also successfully performed
in 10x excess zinc (by moles) and again with 150 mM NaCl, and their results are compared to
aqueous buffer (data not shown)
Comparisons were also made between the distances obtained in 0% TFE and theoretical
random coil values (Table #3). All distances were found to be significantly less than would be
expected for a random coil conformation indicating that MBP in aqueous buffer (0% TFE)
retains a conformation which is somewhat globally compact. Additionally, denaturation was
observed to occur upon addition of 6 M GdmCl (Table #3). This denaturation is demonstrated by
the marked decrease in the transfer efficiencies, and the resulting increase in distances, as
compared to native conditions. These results are consistent with recent small-angle X-ray
47
scattering (SAXS) experiments on MBP which also concluded that the protein is more compact
than an ideal flexible polymer but, however, less compact than a globular protein (140).
48
Tables #2A and 2B. Fluorescence resonance energy transfer (FRET) of MBP labelled with IAEDANS. Samples contained 2.5
µM protein, 20 mM Hepes-KOH, pH 7.4, and 0.5 mM TCEP. Measurements were conducted at 25 °C and the excitation wavelength
was 295 nm. Panel (A) shows the FRET efficiencies and distances calculated from donor (W113) quenching with IAEDANS at
different positions within the primary sequence of 18.5-kDa MBP-UTC1 as indicated. Panel (B) shows the FRET efficiencies and
distances calculated from donor (W113) quenching with IAEDANS at different positions within the primary sequence of 18.5-kDa
MBP-UTC8 as indicated. Data are averages of 3-6 independent measurements, and the error bars represent the standard error. The
general trend is for increased efficiency at 10% and/or 15% TFE, and lower efficiencies at 0% TFE and 30% TFE. This trend is most
prominent at position S17, H85, and S159 indicating a global transition in MBP and the possible presence of an intermediate
conformation.
Table 2A
0% TFE (v/v)
10% TFE (v/v)
15% TFE (v/v)
30% TFE (v/v)
Efficiency (%)
Distance (Å)
Efficiency (%)
Distance (Å)
Efficiency (%)
Distance (Å)
Efficiency (%)
Distance (Å)
UTC1-S17C
9.8±3.9
32.2±2.1
23.2±3.6
26.9±0.9
26.4±0.6
26.1±0.1
12.8±3.4
30.5±1.9
UTC1-S44C
21.5±2.8
27.4±0.8
24.9±3.0
26.5±0.7
30.2±4.8
25.3±1.0
14.8±4.0
30.0±1.8
UTC1-S67C
24.5±4.9
26.6±1.2
32.4±4.8
25.0±0.9
29.3±3.9
25.5±0.8
17.8±3.0
28.7±1.3
UTC1-H85C
35.1±4.4
24.4±0.8
51.1±7.9
21.9±1.2
46.0±7.8
22.6±1.2
29.5±5.0
25.4±0.9
UTC1-S99C
51.2±8.7
22.3±0.6
62.9±3.9
20.2±0.6
61.1±3.7
20.4±0.5
48.3±3.2
22.1±0.3
UTC1-S129C
44.0±8.4
23.2±0.9
56.9±11.7
21.1±1.7
51.6±8.0
21.8±1.2
40.4±3.1
23.5±0.5
UTC1-S159C
23.8±8.0
27.1±1.3
40.9±6.1
23.5±0.9
31.7±13.6
24.9±1.9
19.0±6.7
27.5±1.3
49
Table 2B
0% TFE (v/v)
10% TFE (v/v)
15% TFE (v/v)
30% TFE (v/v)
Efficiency (%)
Distance (Å)
Efficiency (%)
Distance (Å)
Efficiency (%)
Distance (Å)
Efficiency (%)
Distance (Å)
UTC8-S17C
12.1±4.0
30.9±2.1
33.6±11.8
26.5±3.7
29.8±6.5
25.5± 1.4
8.5±3.2
33.0±2.2
UTC8-S44C
34.1±.4.0
24.1±0.9
42.8±6.4
23.1±1.0
41.9±6.7
23.3±1.1
21.7±4.0
27.3±1.0
UTC8-S67C
35.9±9.7
24.4±2.0
41.9±4.3
23.2±0.7
47.3±6.3
22.4±0.9
34.9±9.7
24.5±1.8
UTC8-H85C
39.9±12.2
23.7±2.1
54.4±6.6
21.4±0.9
47.2±3.6
22.4±0.5
39.5±9.5
23.7±1.7
UTC8-S99C
57.3±6.6
21.0±1.0
59.9±8.5
20.6±1.2
52.6±9.2
21.6 ±2.2
48.7±6.1
22.2±0.9
UTC8-S129C
52.1±12.4
21.8±1.9
68.6±11.4
19.2±1.8
49.6±6.7
19.2±1.0
43.3±10.2
23.1±1.7
UTC8-S159C
49.9±7.7
22.0±1.1
54.5±3.1
21.3±0.4
51.0±4.1
21.3 ±0.6
36.4±11.2
24.3±2.1
0% TFE (v/v)
10% TFE (v/v)
15% TFE (v/v)
30% TFE (v/v)
50
Table #3. FRET measurements in 6 M guanidinium chloride (guanidine hydrochloride,
GdmCl). Comparison of FRET-derived distances in IAEDANS-labelled 18.5-kDa MBP samples
with theoretical random coil predictions. MBP was labelled with IAEDANS via engineered free
cysteines at the positions indicated. The FRET was measured between IAEDANS and Trp113.
Samples contained 20 mM Hepes-NaOH, pH 7.4.Theoretical distances are the mean end-to-end
distances for a Gaussian coil and were calculated as 8.37√N, where N is the number of amino
acids separating IAEDANS and Trp113.
Table 3
Under Native
conditions (Å)
Theoretical Random
Coil (Å)
UTC1 S17C
32.2±2.1
UTC1-S44C
MBP-UTC1
in 6 M GdmCl (Å)
MBP-UTC8
in 6 M GdmCl (Å)
80.1
>34
>34
27.4±0.8
68.0
n.d
n.d
UTC1-S67C
26.6±1.2
55.7
>34
>34
UTC1-H85C
24.4±0.8
43.5
n.d
n.d
UTC1-S99C
22.3±0.6
30.9
25.1±0.5
25.2±0.1
UTC1-S129C
23.2±0.9
33.0
n.d
n.d
UTC1-S159C
27.1±1.3
55.7
>34
>34
51
4.9 Measuring the ‘disorder-to-order’ conformational transition(s) of MBP in
TFE
Measurements were performed in various concentrations of TFE ranging from 0-30%
TFE v/v for both MBP-UTC1 and MBP-UTC8. For both variants in the absence of TFE, the data
are consistent with a disordered, flexible conformation, as the determined FRET efficiencies are
qualitatively inversely proportional to the distance in primary sequence between the site of
labelling and Trp113, and this general trend is also true in the presence of TFE. However, there
is a clear trend for FRET efficiency to be higher at 10% and 15% TFE, and lower at 0% and 30%
TFE (Table #2A and #2B), suggesting a tendency to populate more compact conformations at
intermediate TFE concentrations. This general trend holds across all labelling positions, and is
most prevalent with IAEDANS at positions S17C, H85C, and S159C. There are relatively larger
changes in the FRET efficiencies between 0-30% for positions S17C and S159C compared to
positions closer to Trp113. This observation suggests that both UTC1 and UTC8 MBP have
relatively more flexible termini and a more rigid core, which is consist with what has been seen
in previous studies (140). Overall, the FRET data suggest that both 18.5-kDa MBP-UTC1 and
MBP-UTC8 structural ensembles are relatively extended at low and high TFE concentrations,
but that more compact intermediate species are populated at TFE concentrations of 10-15%
(v/v), in general agreement with the ANS-binding data for both variants (Figure #9A-C and
Figure #10).
4.10 The thermodynamic analysis of the ‘disorder-to-order’ transition of MBP
in TFE in order to probe the ‘ensemble populations with different free
energies’
Previously, we had determined the free energy change (ΔG) associated with the disorderto-α-helical transition of the MBP α2-peptide (S72-S107) by fitting CD-monitored TFE titration
curves to a 2-state mechanism (83, 136), and this analysis and mechanism was recently extended
to two other MBP α-peptides, A22-K56 and S133-S159 (Vassall et al., Journal of Molecular
Biology, 427 (10), pp 1977-1992, 2015, doi: 10.1016/j.jmb.2015.03.011).
A combined analysis of the CD, ANS, and FRET data suggest that the minimal
applicable mechanism of disorder-to-α-transition in full-length 18.5-kDa MBP is a 3-state
52
transition (disordered↔intermediate↔α-helical). The structural probes indicate that the
intermediate conformation, when compared to the disordered conformation, has a different
global structure but largely the same local α-helical content. We utilized the fact that the
intermediate is more compact than both the disordered and α-helical conformations, and
systematically monitored the TFE-titration curves of full-length 18.5-kDa MBP-UTC1 by
measuring the change in tryptophan fluorescence quenching by IAEDANS bound at position
S17C or S159C (Figure #11). In these TFE-titration curves, the intensity of tryptophan
fluorescence was measured as TFE was titrated into different solutions containing either
unlabelled or IAEDANS-labelled MBP. The extent of tryptophan fluorescence quenching was
then determined by subtracting the fluorescence intensity of the labelled protein from the
corresponding intensity of the unlabelled protein. Separate experiments using variants S17C and
S159C for MBP-UTC1 were performed in order to assess whether conformational changes in
18.5-kDa MBP are global and if they propagate across the entire primary sequence. We also
collected the corresponding CD-monitored titration curves for IAEDANS-labelled S17C and
S159C for both variants, and found that these curves were highly congruent with the CD curves
collected for unmodified MBP, suggesting again that the presence of the cysteine-substitution
and label do not perturb the conformational transitions of 18.5-kDa MBP (Figure #12A and
#12B). Preliminary experiments for MBP-UTC8 S17C have also been performed (Figure #13).
Reverse-titration curves were also collected which showed that, similar to all three
peptides (83), the conformational transition of full-length 18.5-kDa MBP-UTC1 is reversible and
can be analyzed thermodynamically (Figure #12C and #12D). In selecting the applicable
thermodynamic model with which to analyze the data, we compared the titration curves
monitored by far-UV CD spectroscopy, which follows α-helical formation, with curves
monitored by fluorescence quenching, which follows the formation of new long-range contacts.
In a two-state model, we would expect that new long-range contacts would occur concurrently
with the formation of α-helical structure and hence, titration curves monitored by these two
probes would be superimposable. However, when comparing the CD- and fluorescence
quenching-monitored curves here, it is immediately apparent that the two cannot be
superimposed, as the former is sigmoidal whereas the latter is bell shaped and bi-phasic. This
53
Figure #11. TFE-titration curves of full-length 18.5-kDa MBP-UTC1. Samples contained 4
µM MBP (labelled with IAEDANS), 20 mM Hepes-NaOH, pH 7.4, 0.5 mM TCEP. Experiments
were conducted at 25 °C. Panels (A) and (B) are CD-monitored curves of S17C-MBP and
S159C-MBP, respectively. Panels (C), (D), and (E) are curves of S17C-MBP, S159C-MBP, and
a second independently-labelled S159C-MBP respectively, monitored by Trp113 fluorescence
quenching by IAEDANS. Each data point is the mean of at least 2 independently-prepared
samples, and error bars are the average deviation from the mean. Data were globally fit to a 3state model: disordered↔intermediate↔α-helical (red lines) and fitted values are in Table #4.
All data were normalized before plotting.
54
A
B
C
D
Figure #12. Comparison of TFE-titration curves of IEADANS-labelled S17C and
S159C 18.5-kDa MBP-UTC1 with the unmodified protein. Forward (closed
symbols) and reverse (open symbols) TFE-titration curves. Of IAEDANS-labelled
MBP-UTC1-S17C and S159C variants. Panels (A and B) are representative titration
curves monitored by the change in circular dichroic signal at 222 nm, whereas panels
(C) and (D) show titration curves monitored by tryptophan fluorescence-quenching
by IAEDANS. Samples contained 4 µM protein, 20 mM Hepes-NaOH, pH 7.4, and
measurements were made at 25 °C after a minimum of 2 hours incubation.
55
A
Normalized CD (A.U.)
1.0
0.8
0.6
0.4
0.2
0.0
0
5
15
20
25
30
35
25
30
35
[TFE] (% v/v)
B
Normalized Fluorescence
(A.U.)
10
1.0
0.8
0.6
0.4
0.2
0.0
0
5
10
15
20
[TFE] (% v/v)
Figure #13. Preliminary TFE-titration curves of full-length 18.5-kDa MBPUTC8. Samples contained 4 µM MBP (labelled with IAEDANS), 20 mM HepesNaOH, pH 7.4, 0.5 mM TCEP. Experiments were conducted at 25 °C. Panel (A)
shows the CD-monitored curves of MBP-UTC8-S17C. Panel (B) shows MBPUTC8-S17C monitored by Trp113 fluorescence quenching by IAEDANS. Each
data point is the mean of at least 2 independently-prepared samples, and error bars
are the average deviation from the mean. Data were globally fit to a 3-state
model: disordered↔intermediate↔α-helical (red lines) and fitted values are in
Table 4. All data were normalized before plotting
56
result provides further confirmation that the disordered-to-α-helical transition in MBP does not
occur in a simple two-state process, but involves the presence of at least one intermediate (124).
The data collected for both MBP-UTC1 S17C and S159C, as well as preliminary results for
MBP UTC8 S17C across both probes (CD and fluorescence quenching), were globally fit to a 3state model, comprising disordered ↔ intermediate ↔ α-helical), in which ΔG and m values
were set as shared parameters, and there is good agreement between the data and the fits (Figure
#12, Table #4). From the fits, the ΔG1 (free energy change of disordered↔intermediate) and
ΔG2 (free energy change of intermediate↔α-helical) for MBP-UTC1 values are 1.3±0.5 and
2.5±0.6 kcal mol-1, respectively. The ΔGtotal (total free energy change of disordered↔α-helical
transition), which is calculated as the sum of ΔG1 and ΔG2, is 3.8±1.1 kcal mol-1 (Table #4).
Preliminary data analysis for MBP-UTC8 S17C indicate a ΔG1 and ΔG2 of 1.0±1.2 and 4.8±0.9
kcal mol-1, respectively, whereas the ΔGtotal is 5.8±2.1 kcal mol-1 (Figure #13)
4.11 Deconvolution of the major thermodynamic ensembles along the
‘disorder-to-order’ transition of MBP
The fitted thermodynamic values were also used to calculate the fractional population of
each equilibrium species at different TFE concentrations (Figure #14) for both MBP-UTC1 and
MBP-UTC8. The data concerning MBP-UTC1 show that the disordered conformation is present
until ~17.5% TFE, whereas the population of intermediate conformation reaches a maximal
fractional population of ~0.6 at 14% TFE. The α-helical conformation begins to become
significantly populated at ~15% TFE, and by >30% TFE, this is virtually the only 18.5-kDa
MBP conformation present at equilibrium. The variant MBP-UTC8 shows a large change in the
behaviour of both the disordered and the intermediate states. The disordered conformation is
only present until ~7% TFE, and the intermediate conformation begins to populate the ensemble
at much lower concentrations of TFE, reaching a maximal fractional population of ~1.0 at 10%
TFE. Similarly to MBP-UTC1, the α-helical conformation of MBP-UTC8 begins to become
significantly populated at ~16% TFE, and by >30% TFE completely dominates the population.
Using the FRET efficiencies of both MBP-UTC1 and MBP UTC8 measured in different
concentrations of TFE (Tables #2A and #2B), along with the fractional populations from
thermodynamic analysis, we were able to deconvolve the individual contributions of each of the
57
equilibrium ensemble sets to the measured FRET efficiencies, and also the corresponding
tryptophan-to-IAEDANS distances (Figures #15A, #15B, and #15C). We find that the distances
between IAEDANS and Trp113 in the intermediate conformation are lower at all labelling
positions, compared to the disordered and α-helical conformations, with the largest relative
changes occurring at the S17C and S159C positions. This result is consistent with recent smallangle X-ray scattering data modelled by a coarse-grained structural ensemble, which suggested
that the disordered 18.5-kDa MBP conformation has a relatively rigid core with flexible ends
(140). When combining all the data, we propose that the mechanism of disorder-to-α-helical
formation in 18.5-kDa MBP involves global compaction of the disordered conformation to an
intermediate conformation, with little to no α-helical content, followed by re-expansion to an
extended α-helical-rich conformation.
58
Table #4. Thermodynamic values were determined at 25 °C from global fitting of TFE
titration curves monitored by changes in CD signal and Trp fluorescence quenching by
IAEDANS. Errors (±) were estimated by the fitting program from the global fit. Thermodynamic
values were determined at 25 °C from fitting CD-monitored TFE titration curves. All errors (±)
are standard deviations from at least three independent experiments. Subscripts 1 and 2 refer to
thermodynamic parameters corresponding to the first (D↔I) and second (I↔H) transition,
respectively, for the three-state model. Also, m is the dependence of ∆G on [TFE] and Cmid is the
[TFE] at the mid-point of the transition (i.e., [TFE] when half the peptide molecules contain a αhelical conformation). Both values were determined directly from fitting. e∆GH2O is the change in
free energy of the folding transition in the absence of TFE and was calculated as ∆GH2O = Cmid
•m.fmtotal = m1 +m2; ΔGtotalH2O= ΔG1H2O+ ΔG2H2O.
Table 4
3-state model (D↔I↔H)
m1
ΔG1H2O
m2
ΔG2H2O
mtotal
ΔGtotalH2O
(kcal mol-1
M-1)
(kcal
mol-1)
(kcal mol-1
M-1)
(kcal
mol-1)
(kcal mol-1
M-1)f
(kcal mol-1)f
MBP-UTC1
1.1±0.5
1.3±0.5
1.0±0.2
2.5±0.6
2.1±0.7
3.8±1.1
MBP-UTC8
3.0.0±2.0
1.0±1.2
1.7±0.3
4.8±0.9
4.7±2.3
5.8±2.1
59
1.2
Fractional Population
1
Fi-C1
0.8
Fh-C1
0.6
Fd-C1
Fi-C8
0.4
Fh-C8
0.2
Fd-C8
0
0
10
20
30
40
[TFE] (%vol/vol)
Figure #14. Fractional population of equilibrium conformers of MBP-UTC1 and MBPUTC8. The populations were determined from thermodynamic values (Table 1) obtained from
global fitting of MBP TFE-titration curves to a 3-state model (disordered↔intermediate↔αhelical).
60
Figure 15A
∆G1-UTC1= 1.3
∆G2-UTC1= 2.5
∆G1-UTC8= 1.0
∆G2-UTC8= 5.8
Distance (Å) from tryptophan
residue #113
Figure 15B
40
C
35
30
25
20
15
S17C
S44C
S67C
H85C
S99C S129C S159C
MBP-UTC1 cysteine variant
S17C
S44C
S67C
H85C
S99C S129C S159C
MBP-UTC8 cysteine variant
61
Figure #15A and #15B Thermodynamic mechanism of disorder-to-α-helical transition of
MBP and fractional population of equilibrium conformers. Panel (A) shows the
thermodynamic disorder-to-α-helical transition of MBP with the associated ΔGs for MBP-UTC1
and MBP-UTC8 which populations were obtained from global fitting of MBP TFE-titration
curves to a 3-state model (disordered↔intermediate↔α-helical) (Table #4). The fractional
populations of each equilibrium conformer of MBP used to deconvolute the contribution of each
conformer to the measured FRET efficiencies (Table #2A and #2B), allowed the distances
between the indicated positions and Trp113 within MBP to be calculated (Equation #11). The
distances determined for MBP-UTC1 are shown in Panel (B), and Panel (C) shows the distances
determined for MBP-UTC8.
62
Chapter 5.0: Discussion
5.1 MBP-UTC1 in solution
The earliest conformational studies showed that MBP is predominantly a “random coil”
in aqueous solution (142) and that MBP’s lack of defined tertiary structure in solution prevents
analysis by conventional techniques such as protein X-ray crystallography (67). Subsequent
studies hinted at conformational transitions and some kind of “folded conformation” even in
solution (e.g., (77, 143-145)). Although MBP is largely disordered in aqueous solution, many
studies have revealed the presence of α-helical molecular recognition fragments (α-MoRFs)
which are important for modulating the interaction of the protein with membranes and with
protein partners (reviewed in (78, 131, 146)). As previously mentioned, the inherent intrinsic
disorder of MBP is advantageous as it can allow for rapid association with one or multiple
binding partners, in response to alterations in environmental or chemical signals (73), and it is
well known that PTMs of MBP determine and regulate its conformation (55, 83, 147).
Fluorescence techniques such as FRET, when used in conjunction with other structural
techniques, have been employed in IDPs similar to MBP such as tau and α-synuclein (46, 115).
These techniques are well-suited to study the conformational ensembles of MBP, and structural
changes caused by PTMs in aqueous buffers, in membrane-mimetic solvents such as TFE, in
denaturants, and in the presence of small-interacting ligands such as zinc (115, 148, 149). In
experiments presented here, seven MBP-UTC1 and seven MBP-UTC8 protein variants were
used, each containing single-cysteine substitutions which were designed to span the entire
primary sequence of the 168-residue protein (89, 150) in order to develop a ‘molecular ruler’.
This strategy should give a sense of the ‘global’ conformational changes that occur in MBP
during the disorder-to-order transitions of MBP isoforms. Additionally, by comparing the
distances measured by FRET in solution with those predicted theoretically, we can gain insight
into the nature of MBP conformational ensembles in solution. The FRET was measured between
the center of the IAEDANS fluorescent probe and the center of Trp113.
As mentioned previously, MBP has been described as lacking ordered secondary
structural elements in solution, yet it may be possible that MBP maintains some residual
63
compacted conformation in solution (77, 148). Recent studies have indicated that the Flory
random-coil model can accurately predict dimensions of completely denatured (random coil)
proteins successfully (148). This model may also be applicable to IDPs that have little to no
long-range interactions. A comparison of FRET distances for MBP to the Flory model
demonstrated that MBP is compacted relative to a random coil (Table #3). Using in accordance
with the Flory random coil model, theoretical distances are the mean end-to-end distances for a
Gaussian coil and were calculated as 8.37√N, where N is the number of amino acids separating
IAEDANS and Trp113. Our experimental FRET distances of ~20-30 Å are much smaller than
those predicted by the Flory model, suggesting that the assumption of a pure random coil is not
applicable, and thus that MBP maintains some level of residual structure in both MBP-UTC1 and
MBP-UTC8 whilst in solution.
The FRET measurements were also performed in 150 mM NaCl, as MBP can be viewed
as a ‘polyampholyte’ which refers to a polymer composed of negatively- and positively-charged
monomers interspersed along a protein chain. These chains are either neutral or have a net charge
of one sign (151). If the net charge is high as is the case with MBP, the expectation is for the
chain to behave like a conventional polyelectrolyte with all charges of the same sign, having a
net repulsion and forming an ‘extended rod-like’ molecule, whereas the behaviour of neutral
polyampholyte chains is entirely different as the net electrostatic forces are attractive, so that the
chains have a tendency to collapse into a globule (151). When an ionic solution such as salt
(NaCl) is added, a charge-screening effect alters the interactions and reduces the electrostatic
attraction(s). The globule begins to increase in size when the concentration of salt becomes
larger than the concentration of charge on the polymer itself. This screening effect occurs in
ionic solutions and works by creating an "atmosphere" of charges of the opposite sign in order to
screen the long-range interactions, and gives rise to a net gain in electrostatic energy (151). We
measured the FRET for both MBP-UTC1 and MBP-UTC8 in the presence 150 mM NaCl in
order to monitor the extent of change screening of the transfer efficiencies (data not shown).
Preliminary results show no statistical change in FRET efficiency for either MBP-UTC1 or
MBP-UTC8 in the presence of NaCl as compared to aqueous solution. More replicates must be
performed in order to confirm this result.
64
MBP in CNS myelin is also known to interact with the small ligand zinc. Zinc is found in
high concentration (~50 µM) in the myelin cytoplasm, and increases up to 3-fold in the
erythrocytes during MS pathology (152). MBP is known to be a prominent zinc-binding protein
via intrinsic histidyl-residues (74). Divalent cations, such as Zn2+, can induce a random coil to
order transition in several IDPs, including MBP (153). Zinc is believed to stabilize the protein's
association with the lipids in order to form compact myelin in the CNS (154), and thus is
proposed to fulfill a similar role to that of zinc in the peripheral nervous system myelin (155).
Previous studies by the Harauz group characterized the effect of zinc on MBP by FTIR, which
spectroscopy revealed that there was a rearrangement of secondary structure components upon
addition of zinc (74). Results for FRET distances and efficiencies in the presence of 10x excess
Zn (by moles) suggest no significant change to either MBP-UTC1 or MBP-UTC8 as compared
to aqueous solution. Previous studies, in true membrane- environments, showed changes in
secondary structure upon increasing the amount of zinc, suggesting a synergistic effect of the
myelin membrane and divalent cations on MBP structural rearrangement (156). Thus, it could be
due to the lack of a lipid component that we see no significant change in conformation of MBPUTC1 or MBP-UTC8. Furthermore, ANS binding (discussed below) performed in the presence
of 10x molar excess Zn for both MBP-UTC1 and MBP-UTC8 showed no shift in the maximal
enhancement (data not shown) as compared to aqueous solution.
5.2 Membrane mimetic solvents and their uses in IDPs
In this study, we sought to characterize the mechanism of 18.5-kDa MBP disorder-to-αhelical transition through a conformational and thermodynamic analysis of MBP-UTC1 and
MBP-UTC8. From previous structural studies, we have produced a two-dimensional map of
MBP (Figure #1, #2A, and #2B) revealing three regions with α-helical propensity
(approximately T33-F43, P82-V91, Y141-K153) that we have shown are important for tethering
MBP to the oligodendrocyte membrane. Other studies have focused on the three MBP α1-, α2-,
and α3-peptides (A22-K56, S72-S107, and S133-S159, respectively, murine 18.5-kDa species
numbering), and found that both the α1 and α3 regions are important interaction sites for Ca2+activated calmodulin, and that the α1-containing region may also interact with actin (158). The α2
region is part of a molecular switch, and may be conformationally linked to an adjacent prolinerich structure that is an SH3-ligand (20, 83, 141, 158). These three regions in 18.5-kDa MBP
65
may adopt the α-helical conformation upon interaction with its partners in a binding-induced
folding mechanism, or alternatively, interaction may occur through conformational selection
whereby a population of the α-helical conformer is present and available for interaction (66, 159,
160). However, the applications of such conventional ensemble spectroscopic techniques such as
NMR, EPR, and CD have not determined all the necessary conformational information about
MBP (distance restraints) in order to generate a structural model. Not all structured and
unstructured domains of MBP, and/or the effect of PTMs on structure, have yet been resolved
(131). There is also a lack of information regarding the mechanism of disorder-to-order transition
which is useful both for defining the protein’s conformational space and may be important in
binding-induced folding mechanisms. Furthermore, identifying the populations of energeticallydistinct sets of ensembles along the transition pathway may be instrumental for determining the
potential for conformational selection.
The propensity to form α-helical structure in the full-length MBP-UTC1 and MBP-UTC8
18.5-kDa protein variants was probed using TFE-titration curves which have long been used to
determine the free energy of α-helix formation of peptides as well as of proteins, through fitting
to thermodynamic folding models (124-126).
Trifluoroethanol has been shown to induce α-helical structure mainly in regions which
already have α-helix-forming propensity, and it has been shown to be an effective membranemimetic solvent for many proteins, including MBP, when used judiciously (97, 158, 161-163).
Consistent with this idea, analysis of previously published NMR data of MBP in 30% TFE has
demonstrated that the pattern of α-helical formation in TFE corresponds to what has been found
in reconstituted lipid bilayer environments (see Figure #2B and accompanying figure legend).
At low concentrations, it has been suggested that TFE can effectively simulate membrane surface
conditions by lowering the dielectric constant of the solution, whereas at higher concentrations,
there is the added effect of the preferential interaction of TFE with the protein, which can
effectively simulate the effects of protein-membrane binding (153). Thus, TFE is useful in
analyzing the disorder-to-α-helical transition of MBP that occurs as it approaches the
oligodendrocyte membrane surface and eventually interacts (164, 165). Of direct relevance to
this work is an independent study by the group of Israelachvili et al., who demonstrated using a
66
surface forces apparatus and quartz crystal microbalance that MBP in solution first adsorbs to a
solid surface such as mica or SiO2, then undergoes a conformational change (165). It can be
predicted that the same phenomenon occurs with membrane surfaces (111).
5.3 Thermodynamic analysis of the disorder-to-α-helical transition in fulllength 18.5-kDa MBP-UTC1 using 2,2,2-trifluoroethanol (TFE)
In this study, we performed a thermodynamic analysis of TFE-titration curves monitored
by CD spectroscopy, which probed changes in secondary structure, and tryptophan fluorescence
quenching by IAEDANS, which was used to probe for concomitant changes in global
organization (Figure #11A-E, #13A and #13B and Table #4). The data are well described by a
3-state transition model (disordered↔intermediate↔α-helical), indicating that intermediate
formation in MBP is a first-order process involving global cooperative interactions. The
observed global cooperativity of the disordered↔α-helical transition of 18.5-kDa MBP suggests
that the conformational space of each equilibrium species is somewhat limited. Nevertheless, it is
important to note that each of the three identified equilibrium species will still have significant
conformational freedom, and thus that their structural properties likely represent an average of
ensemble conformations, with each ensemble arranged in distinct clusters that are separated by
energy barriers (Figure #14).
The ΔGtotal of the disordered-to-α-helical transition in MBP-UTC1 is a fairly modest
3.8±1.1 kcal mol-1, and we found that the ΔG of the first transition (disordered↔intermediate)
was more than 50% lower than the second transition (intermediate↔α-helical). Due to the
relatively low ΔG1 value, there is a significant population of intermediate (fractional population
of ~0.1) even in the absence of TFE (Figure #15). Surprisingly, upon transitioning from the
intermediate to the α-helical conformation, we find that 18.5-kDa MBP globally re-expands. The
α-helical conformation in MBP is highly populated above 25% TFE (v/v), and could be
stabilized by preferential interaction with this membrane-mimetic solvent.
The observed collapse and subsequent re-expansion mechanism of the disorder-to-order
transition in 18.5-kDa MBP helps to explain the observed pattern of ANS-enhancement at
different TFE concentrations, which largely mirrors the fractional population curve of the
intermediate (Figures #9, #10 and #15). These data thus suggest that it is predominantly the
67
intermediate that binds to ANS, whereas there is little ANS interaction with the more extended
disordered or α-helical conformations. The ability to bind ANS is a commonly observed
characteristic of collapsed protein conformations which have solvent-exposed hydrophobic
pockets, and is a property associated with molten-globule and pre-molten globule conformations.
Significantly, both the molten-globule and pre-molten globule conformations possess some
ordered secondary structure which is notably absent from the MBP intermediate identified here.
This phenomenon suggests that collapse in the intermediate state is not driven by cooperative
interactions between the α-helical domains, as α-helical formation occurs concurrently with
global re-expansion in a second transition. The presence of hydrophobic pockets in the
intermediate suggests that compaction could be, at least in part, driven by the classical
hydrophobic collapse model. However, since 18.5-kDa MBP, like most IDPs, has a low overall
hydrophobicity, it is unlikely that this is the dominant mechanism for global collapse. Rather
than being the driving force of collapse, the presence of hydrophobic patches may have
functional significance; for example, they may be important for binding of the intermediate MBP
conformer to a partner or surface (150, 164). It has been suggested that molecular recognition
fragments in disordered proteins or regions may have evolved to have a high proportion of
hydrophobic patches, which can serve as non-specific binding sites that mediate induced folding
(166). The fact that the population of the intermediate is highest at ~14% TFE suggests that
collapse is mediated by hydrogen bonding, consistent with the new models of protein collapse
which argue that backbone interactions are the dominant force (167).
5.4 Defining a physiological role for MBP-UTC8: A structural and
thermodynamic comparison of the MS-MBP isoform
Preliminary FRET results for MBP-UTC8 in solution (0% TFE) are largely similar to
those seen for MBP-UTC1 showing the presence of a flexible but non-random coil state that
completely lacks any secondary structure as monitored by CD spectroscopy. One discrepancy is
that MBP-UTC8-S159C appears to remain much closer to W113 as compared to MBP-UTC1S159C when in solution, and remains so over the entire range of TFE concentrations studied.
This observation suggests a more compacted C-terminus, and possibly a more compact global
structure.
68
From thermodynamic analysis and the resulting fractional populations (Figure #14 and
#15 we see that the fractional population of the disordered ensemble is about ~0.8 at 0% TFE v/v
as compared to ~0.9 for MBP-UTC1. Furthermore, in MBP-UTC8 the fractional population of
the disordered species declines to only ~0.01 at 8% TFE. This result is in stark contrast to what is
seen for MBP-UTC1, in which the disordered species remains populated over a much larger
range of TFE concentrations, and decreases to ~0.01 at 20% TFE v/v. More importantly, the
intermediate ensemble for MBP-UTC8 is maximally populated at much lower concentrations of
TFE, totalling >0.95 of the ensemble population at as little as 8% TFE v/v, and reaching 1.0 at
~10% TFE v/v. The relatively higher population of the intermediate species when comparing
MBP-UTC1 and MBP-UTC8 mainly results from ΔG2 being significantly higher in the latter,
suggesting that the intermediate is more stable compared to the α-helical conformation. In
relating these preliminary thermodynamic data of MBP-UTC8 to a possible physiological role in
MS, it is possible that a larger energy barrier between intermediate and α-helical ensembles may
impede myelin stability, as formation of the α-helices is required for membrane association.
While this is an interesting proposition, more data will have to be collected to verify this finding.
69
Chapter 6.0: Conclusions, Significance and Future Directions
Intrinsically-disordered proteins are highly flexible and dynamic, allowing them to
sample a wide variety of conformations. They are usually envisioned as extended structures but
are known to also sample collapsed conformations containing various degrees of ordered
secondary structure. The conformational energy landscape of IDPs, in the absence of a binding
partner, is often represented as either a shallow funnel, or as a totally random landscape in which
all conformational states are readily accessible with little to no thermodynamic barriers (167168). Here, in our analysis of 18.5-kDa MBP, which is a stereotypical IDP, we find that the
disorder-to-α-helical transition follows a biphasic mechanism at equilibrium in which three
distinct states (or more likely, three ensemble clusters) are separated by measurable energy
barriers (Figure #14A and #14B). The FRET results of MBP in solution are consistent with the
presence of a relatively disordered conformation, yet more compacted than a random coil state,
that completely lacks any ordered secondary structure as monitored by CD (Figure #8) and also
by NMR spectroscopy (169). These kinds of disordered collapsed conformations have been
observed in many IDPs including Sic1 (170) and osteopontin (171), and also in amyloidogenic
IDPs such α-synuclein (172), where they may represent important aggregation precursors. It is,
therefore, of significant interest to characterize these collapsed conformations within IDP
ensembles both structurally and mechanistically.
Interestingly, when considering the thermodynamic data as well as the results from FRET
measurements, we find that it is the intermediate conformational ensemble of 18.5-kDa MBP
that is the most compact, and not the α-helical ensemble. A α-helical rich state that is relatively
globally extended may, in fact, be functionally important, as it could potentially allow MBP to
better bridge the two cytoplasmic leaflets in the major dense line of myelin, which is roughly 3
nm in size. Additionally, a more extended α-helical state may facilitate faster membrane binding
by providing more surface area for interaction and by giving the α-helical segments greater
flexibility to attain the correct orientation for insertion into the membrane. Furthermore, we
suggest that differences observed here in both the structural conformation of MBP-UTC1 and
MBP-UTC8 as well the differences that observed in the thermodynamic values for the
70
disordered↔intermediate↔α-helical transition provide evidence that further supports the
suggestion that deimination of MBP may be an early event that precipitates MS pathogenesis.
Interesting avenues of future research would be to study MBP in a natural lipidic
environment which is both physiologically and structurally pertinent. Even when membraneassociated, MBP is still largely mobile and able to interact with various cellular components
(146, 173). Previously, DPC micelles have been used to induce α-helicity of MBP-UTC1 and its
associated peptides by NMR and CD spectroscopy, but remains to be completed for the MBPUTC8 variant. Furthermore, performing FRET studies in the presence of DPC would be wellsuited for a structural comparison similar to the work completed here. The effects of other PTMs
such as phosphorylation on the disorder-to-α-helical transition of MBP should also be measured.
This modification, which reduces the overall net change of MBP and also alters the charge
distribution on the protein (22, 83), could modulate the ensemble conformations sampled by the
protein. Additionally, these PTMs could also directly affect α-helical formation. For example,
using the MBP α2-peptide, the Harauz lab have shown previously that phosphorylation at
mitogen-activated protein kinase sites (T92 and T95, murine 18.5-kDa numbering) alters the
disorder-to-α-helical transition of the second α-helical segment (P82-I90) by destabilizing the Cterminal end of the α-helix (141). The methodology that we have illustrated here would allow us
to quantitatively assess the effects of PTMs on the global disorder-to-α-helical transition of
MBP, which could provide insights into the regulation of myelination, multiple sclerosis
pathogenesis and a better theoretical framework of the factors that control the conformational
energy landscape of IDPs.
71
References
1. Simons, M. and Lyons, D.A. Axonal selection and myelin sheath generation in the central
nervous system. Curr. Opin. Cell. Biol. 2013;25(4):512-519.
2. Aggarwal, S., Yurlova, L., and Simons, M. Central nervous system myelin: Structure,
synthesis and assembly. Trends Cell. Biol. 2011;21(10):585-593.
3. Harauz, G. and Boggs, J.M. Myelin management by the 18.5 and 21.5-kDa classic myelin
basic protein isofroms. J. Neurochem. 2013;125(3):334-361.
4. Paez, P.M., Fulton, D., and Colwell, C.S., Campagnoni, A.T. Voltage-operated ca(2+) and
na(+) channels in the oligodendrocyte lineage. J. Neurosci.Res. 2009;87(15):3259-3266.
5. Spiegel, I. and Peles, E. Cellular junctions of myelinated nerves (review). Mol. Membr. Biol.
2002 04;19(2):95-101.
6. Lublin, F.D. Multiple sclerosis classification and overview. In: Lazzarini, R.A., Griffin, J.W.,
Lassman, H., Nave, K., Miller, R.H., Trapp, B.D., editors. San Diego: Elsevier Academic Press;
2004. p. 691-699.
7. Hartline, D. What is myelin? Neuron glia biology. 2008;4(2):153-163.
8. Zhang, K. A universal scaling law between gray matter and white matter of cerebral cortex.
Proc. Natl. Acad. Sci. U.S.A. 2000;97(10):5621-5626.
9. Prineas, J.W. Multiple sclerosis. pathology of recurrent lesions. Brain. 1993;116 (3):681-693.
10. Kirschner, D.A., Inouye, H., Ganser, A.L., and Mann, V. Myelin membrane structure and
composition correlated: A phylogenetic study. J. Neurochem. 1115 11;53(5):1599-1609.
11. Ishiyama, N. Electron crystallography of myelin basic protein. 2000. MSc. Thesis, Molecular
Biology and Genetics, University of Guelph, Canada.
12. Boggs, J.M. Myelin basic protein: A multifunctional protein. Cell. Mol. Life. Sci. 2006;
63:1945-1961.
13. Kalwy, S., Marty, M.C., Bausero, P., and Pessac, B. Myelin basic protein-related proteins in
mouse brain and immune tissues. J. Neurochem. 1998;70 (1):435-438.
72
14. Ahmed, M.A.M., Bamm, V.V., Shi, L., Steiner-Mosonyi, M., Dawson, J.F., Brown, L., et al.
Induced secondary structure and polymorphism in an intrinsically disordered structural linker of
the CNS: Solid-state NMR and FTIR spectroscopy of myelin basic protein bound to actin.
Biophys. J. 2009; 96(1):180-191.
15. Boggs, J.M., Rangaraj, G., Gao, W., and Heng, Y.M. Effect of phosphorylation of myelin
basic protein by MAPK on its interactions with actin and actin binding to a lipid membrane in
vitro. Biochemistry. 2006; 45(2):391-401.
16. Dyer, C.A., Philibotte, T.M., Billings-Gagliardi, S., and Wolf, M.K. Cytoskeleton in myelinbasic-protein-deficient shiverer oligodendrocytes. Dev. Neurosci. 1995;17(1):53-62
17. Baran, C., Smith, G.S., Bamm, V.V., Harauz, G., and Lee, J.S. Divalent cations induce a
compaction of intrinsically disordered myelin basic protein. Biochem. Biophys. Res. Commun.
2010; 391(1):224-229.
18. Bund, T. Copper uptake induces self-assembly of 18.5 kDa myelin basic protein (MBP).
Biophys. J. 2010;99(9):3020-3028.
19. Homchaudhuri, L., De Avila, M., Nilsson, S.B., Bessonov, K., Smith, G.S., Bamm,
V.V., Musse, A.A., Harauz, G., and Boggs, J.M. Secondary structure and solvent accessibility of
a calmodulin-binding C-terminal segment of membrane-associated myelin basic protein.
Biochemistry. 2010; 49 (41):8955-8966.
20. De Avila, M., Vassall, K.A., Smith, G.S.T., Bamm,V.V., and Harauz, G. The proline-rich
region of 18.5-kDa myelin basic protein binds to the SH3-domain of fyn tyrosine kinase with the
aid of an upstream segment to form a dynamic complex in vitro. Biosci. Rep. Biochem. J.
2014;34(6):775-788.
21. Smith, G.S.T., Homchaudhuri, L., Boggs, J.M., and Harauz, G. Classic 18.5- and 21.5-kDa
myelin basic protein isoforms associate with cytoskeletal and SH3-domain proteins in the
immortalized N19-oligodendroglial cell line stimulated by phorbol ester and IGF-1. Neurochem.
Res. 2012;37(6):1277-1295.
22. Polverini, E., Rangaraj, G., Libich, D.S., Boggs, J.M., and Harauz, G. Binding of the prolinerich segment of myelin basic protein to SH3-domains - spectroscopic, microarray, and modelling
studies of ligand conformation and effects of post-translational modifications. Biochemistry.
2008;47(1):267-282.
23. Steinman, M.D. Multiple sclerosis: A coordinated immunological attack against myelin in
the central nervous system. Cell. 1996;85(3):299-302.
73
24. Compston, A., and Coles, A. Multiple sclerosis. Lancet. 2002;359 (9313):1221-1231.
25. Miller, R.H., and Fyffe-Maricich, S.L. Restoring the balance between disease and repair in
multiple sclerosis: Insights from mouse models. Dis. Model Mech. 2010; (3):535-539.
26. Franklin, R.J. Why does remyelination fail in multiple sclerosis? Nat. Rev. Neurosci. 2002;9
(3):705-714.
27. Givogri, M.I. New insights on the biology of myelin basic protein gene: The neural-immune
connection. J. Neurosci. Res. 2000;59(2):153-159.
28. Stys, P.K. Pathoetiology of multiple sclerosis: Are we barking up the wrong tree? F1000
Prime. Rep.2013;(5):20.
29. Stys, P.K. Multiple sclerosis: Autoimmune disease or autoimmune reaction? Can. J. Neurol.
Sci. 2010; (37):16-23.
30. Witalison, E.E., Thompson, P.R., Hofseth, L.J. Protein arginine deiminases and associated
citrullination: Physiological functions and diseases associated with dysregulation. Curr. Drug
Targets. 2015. [E-pub]
31. Mastronardi, F.G., and Moscarello, M.A. Deimination of myelin basic protein by PAD
enzymes, and their role in multiple sclerosis. In: Boggs JM, editor. New York: Nova Science
Publishers; 2008. p. 31-49.
32. Harauz, G., and Musse, A.A. A tale of two citrullines - structural and functional aspects of
myelin basic protein deimination in health and disease. Neurochem. Res. 2007;32(2):137-58.
33. Moscarello, M.A., Mastronardi, F.G. and Wood, D.D. The role of
citrullinated proteins suggests a novel mechanism in the pathogenesis of multiple sclerosis.
Neurochem. Res. 2007;32:251-256
34. Boggs, J.M., Rangaraj, G., Hill, C.M., Bates, I.R., Heng, Y.M., Harauz, G. Effect of arginine
loss in myelin basic protein, as occurs in its deiminated charge isoform, on mediation of actin
polymerization and actin binding to a lipid membrane in vitro. Biochemistry. 2005;44(9):35243534.
35. Readhead, C. and Hood, L. The dysmyelinating mouse mutations shiverer (shi) and myelin
deficient (shimld). Behav. Genet. 1990;20(2):213-34.
74
36. Carr‚ J.L., Goetz, B.D., O'Connor, L.T., Bremer, Q., and Duncan, I.D. Mutations in the rat
myelin basic protein gene are associated with specific alterations in other myelin gene
expression. Neurosci Lett. 2002; 13;330(1):17-20.
37. Wood, D.D and Moscarello, M.A. Molecular biology of the glia: Components of myelin myelin basic protein - the implication of post-translational changes for demyelinating disease. In:
Russell W.C., editor. Chichester: John Wiley and Sons; 1997. p. 37-54.
38. Milner-White, E.J., and Bell, L.H., Maccallum, P.H. Pyrrolidine ring puckering in cis and
trans-proline residues in proteins and polypeptides. different puckers are favoured in certain
situations. J. Mol. Biol. 1992;228(3):725-734.
39. Rispoli, P., Carzino, R., Svaldo-Lanero, T., Relini, A., Cavalleri, O., Fasano, A., Liuzzi,
G.M., Carlone, G., Riccio, P., Gliozzi, A., and Rolandi, R. P. A thermodynamic and structural
study of myelin basic protein in lipid membrane models. Biophys. J. 2007;93(6):1999-2010.
40. Cao, L., Sun, D., and Whitaker, J.N. Citrullinated myelin basic protein induces experimental
autoimmune encephalomyelitis in lewis rats through a diverse T cell repertoire. J.
Neuroimmunol. 1998;88(1-2):21-29.
41. Zand, R., Jin, X., Kim, J., Wall, D.B., Gould, R., and Lubman, D.M. Studies of
posttranslational modifications in spiny dogfish myelin basic protein. Neurochem. Res.
2001;26(5):539-547.
42. Wood, D.D., Bilbao, J.M., O'Connors, P., and Moscarello, M.A. Acute multiple sclerosis
(marburg type) is associated with developmentally immature myelin basic protein. Ann. Neurol.
1996;40(1):18-24.
43. Pritzker, L.B., Joshi, S., Gowan, J.J., Harauz, G., and Moscarello, M.A. Deimination of
myelin basic protein. 1. effect of deimination of arginyl residues of myelin basic protein on its
structure and susceptibility to digestion by cathepsin D. Biochemistry. 2000;39(18):5374-5381.
44. Al-Ghobashy, M.A., Cucheval, A., Williams, M.A., Laible, G., and Harding, D.R.. Probing
the interaction between recombinant human myelin basic protein and caseins using surface
plasmon resonance and diffusing wave spectroscopy. J. Mol. Recognit. 2009;23(1):84-92.
45. Trexler, A and Rhoades, E. Function and dysfunction of ɑ-synclein: probing conformational
changes and aggregation by single molecule fluorescence. Mol. Neurobiol. 2013;47(2):622-631.
46. Jeganathan, S. Global hairpin folding of tau in solution. Biochemistry. 2006;45(7):22832293.
75
47. Mao, A.H., Crick, S.L., Vitalis, A., Chicoine, C.L., and Pappu, R.V. Net charge per residue
modulates conformational ensembles of intrinsically disordered proteins. Proc. Natl. Acad. Sci.
U. S. A. 2010 ;107(18):8183-8.
48. Dunker, A.K., Cortese, M.S., Romero, P., Iakoucheva, L.M., and Uversky, V.N. Flexible
nets. the roles of intrinsic disorder in protein interaction networks. FEBS J. 2005;272(1742-464;
20):5129-5148.
49. Dunker, A.K., Silman, I., Uversky, V.N., and Sussman, J.L. Function and structure of
inherently disordered proteins. Curr. Opin. Struct. Biol. 2008;18(1879-033; 6):756-764.
50. Bates, I.R., Libich, D.S., Wood, D.D., Moscarello, M.A., and Harauz, G. An arg/lys-->Gln
mutant of recombinant murine myelin basic protein as a mimic of the deiminated form
implicated in multiple sclerosis. Protein Expr. Purif. 2002;25(2):330-41.
51. Zand, R., Li, M.X., Jin, X., and Lubman, D. Determination of the sites of posttranslational
modifications in the charge isomers of bovine myelin basic protein by capillary electrophoresismass spectroscopy. Biochemistry.1998;37(8):2441-2449.
52. Musse, A. and Harauz, G. Deimination of membrane-bound myelin basic protein in multiple
sclerosis exposes an immunodominant epitope. Proc. Natl. Acad. Sci. U.S.A.
2006;103(12):4422-4427.
53. Wood, D.D., and Moscarello, M.A. Is the myelin membrane abnormal in multiple sclerosis?
J. Membr. Biol. 1984;79(3):195-201.
54. Bates, I.R., and Harauz, G. Molecular dynamics exposes ɑ-helices in myelin basic protein.
J.Mol.Mod. 2003;9(5):290-297.
55. Harauz, G., and Musse, A.A. A tale of two citrullines - structural and functional aspects of
myelin basic protein deimination in health and disease. Neurochem. Res. 2007;32(2):137-158.
56. Lamensa, J.W.E., and Moscarello, M.A. Deimination of human myelin basic protein by a
peptidylarginine deiminase from bovine brain. J. Neurochem. 1993;61(3): 987-996
57. Bates, I.R., Feix, J.B., Boggs, J.M., and Harauz, G. An immunodominant epitope of myelin
basic protein is an amphipathic alpha-helix. J Biol Chem. 2004;279(7):5757-5764.
58. Harauz, G., Ladizhansky, V., and Boggs, J.M. Structural polymorphism and
multifunctionality of myelin basic protein. Biochemistry. 2009;48(34):8094-8104.
76
59. Harauz, G. and D.S. Libich The classic protein of myelin - conserved structural motifs and
the dynamic molecular barcode involved in membrane adhesion, protein-protein interactions, and
pathogenesis in multiple sclerosis. 2013;1:1-53.
60. Wright, P.E., and Dyson, H.J. Intrinsically unstructured proteins: re-assessing the protein
structure-function paradigm. J. Mol. Biol. 1999;293(2):321-331.
61. Dunker, A.K., Silman, I., Uversky, V.N., and Sussman, J.L. Function and structure of
inherently disordered proteins. Curr. Opin. Struct. Biol. 2008;18(6):756-764.
62. Uversky, V.N. Intrinsically disordered proteins from A to Z. Int. J. Biochem. Cell. Biol.
2011;43(8):1090-1103.
63. Tompa, P. Intrinsically unstructured proteins. Trends. Biochem. Sci. 2007;27(10):527-533.
64. Uversky, V.N. What does it mean to be natively unfolded? Eur. J. Biochem. 2001;269(1):212.
65. Uversky, V.N. Introduction to intrinsically disordered proteins (IDPs). Chem. Rev.
2014;114(13):6557-6560.
66. Rezaei-Ghaleh N, Blackledge MF, Zweckstetter M. Intrinsically disordered proteins: From
sequence and conformational properties toward drug discovery. ChemBioChem : Eur. J.
Biochem. 2012; 13(7):930-950.
67. Sedzik, J., and Kirschner, D.A. Is myelin basic protein crystallizable? Neurochem. Res.
1992;17(2):157-166.
68. Uversky, V.N. Unusual biophysics of intrinsically disordered proteins. Biochim. Biophys.
Acta. 2013;1834(5):932-51.
69. Sickmeier, M. DisProt: The database of disordered proteins. Nucleic Acids Res.
2007;35:786-793.
70. Xue, B., Dunker, A.K., and Uversky, V.N. Orderly order in protein intrinsic disorder
distribution: Disorder in 3500 proteomes from viruses and the three domains of life. J. Biomol.
Struct. Dyn. 2012;30(2):137-149.
71. Cumberworth, A., Lamour, G., Babu, M.M., and Gsponer, J. Promiscuity as a functional
trait: Intrinsically disordered regions as central players of interactomes. Biochem. J.
2013;454(3):361-369.
77
72. Uversky, V.N. Intrinsic disorder in proteins associated with neurodegenerative diseases.
Front Biosci. 2009;14(1):5188-5238.
73. Uversky, V.N., Oldfield, C.J., and Dunker, A.K.. Showing your ID: Intrinsic disorder as an
ID for recognition, regulation and cell signaling. J. Mol. Recognit. 2005;18(5):343-84.
74. Smith, G.S.T., Chen, L., Bamm, V.V., Dutcher, J.R., and Harauz, G. The interaction of zinc
with membrane-associated 18.5 kDa myelin basic protein: An attenuated total reflectance-fourier
transform infrared spectroscopic study. Amino Acids. 2010; 39(3):739-50.
75. Harauz G, and Ladizhansky V. Structure and dynamics of the myelin basic protein family by
solution and solid-state NMR. In: Boggs JM, editor. New York: Nova Science Publishers; 2008.
p. 197-232.
76. Krigbaum, W.R. Molecular conformation of bovine A1 basic protein, a coiling
macromolecule in aqueous solution. Biochemistry. 1975;14(11):2542-2546.
77. Nowak, M.W., and Berman, H.A. Fluorescence studies on the interactions of myelin basic
protein in electrolyte solutions. Biochemistry.1991;30(30):7642-7651.
78. Harauz, G., and Libich, D.S. The classic basic protein of myelin - conserved structural motifs
and the dynamic molecular barcode involved in membrane adhesion and protein-protein
interactions. Current Protein and Peptide Science. 2009;10(3):196-215.
79. Ahmed, M.A.M., Bamm, V.V., Harauz, G., and Ladizhansky, V. Solid-state NMR
spectroscopy of membrane-associated myelin basic Protein "Conformation and dynamics of an
immunodominant epitope". Biophys. J. 2010;99:1-9.
80. Zhong, L., Bamm, V.V., Ahmed, M.A., Harauz, and G, Ladizhansky V. Solid-state NMR
spectroscopy of 18.5 kDa myelin basic protein reconstituted with lipid vesicles: Spectroscopic
characterisation and spectral assignments of solvent-exposed protein fragments.
Biochim.Biophys.Acta (Biomembranes). 2007;1768(12):3193-3205.
81. Reinl, H.M. and Bayerl, T.M. Interaction of myelin basic protein with single bilayers on a
solid support: an NMR, DSC and polarized infrared ATR study. Biochim.Biophys.Acta
(Biomembranes). 1993; 1151(2);127-136.
82. Boggs, J.M., Bates, I.R., Musse, A.A., and Harauz, G. Interactions of the 18.5 kDa myelin
basic protein with lipid bilayers: Studies by electron paramagnetic resonance spectroscopy and
implications for generation of autoimmunity in multiple sclerosis. In: Boggs JM, editor. New
York: Nova Science Publishers; 2008. p. 105-125.
78
83. Vassall, K.A., Bessonov, K., De Avila, M., Polverini, E., and Harauz G. The effects of
threonine phosphorylation on the stability and dynamics of the central molecular switch region
of 18.5-kDa myelin basic protein. PLoS One. 2013;8(7): 19.
84. Ahmed, M.A.M., Bamm, V.V., Harauz, G., and Ladizhansky, V. Solid-state NMR
spectroscopy of membrane-associated myelin basic protein - conformation and dynamics of an
immunodominant epitope. Biophys J. 2010;99(4):1247-1255.
85. Fares, C., and Davis, J.H. The search for high-resolution NMR methods for membrane
peptide structure. In: Burnell EE, deLange CA, editors. Kluwer Academic; 2003. p. 191-213.
86. Mendz, G.L., Moore, W.J., Kaplin, I.J., Cornell, B.A., Separovic, F., Miller, D.J., et al.
Characterization of dodecylphosphocholine/myelin basic protein complexes. Biochemistry.
1988;27(1):379-386.
87. Bates, I.R., Matharu, P., Ishiyama, N., Rochon, D., Wood, D.D., Polverini, E., et al.
Characterization of a recombinant murine 18.5-kDa myelin basic protein. Protein Expr. Purif.
2000;20(2):285-299.
88. Jayaraj, R. So you need a protein - A guide to the production of recombinant proteins. The
open veterinary science journal. 2009;3(1):28-34.
89. Deibler, G. E., Martenson R. E., and Kie M.W. Large Scale Preparation of Myelin Basic
Protein from Central Nervous Tissue of Several Mammalian Species. Prep. Biochem. 1972;
2(2):139-165
90. Libich, D.S., Hill, C.M.D., Bates, I.R., and Hallett, F.R., Armstrong S, Siemiarczuk A, et al.
Interaction of the 18.5-kD isoform of myelin basic protein with Ca2+-calmodulin: Effects of
deimination assessed by intrinsic trp fluorescence spectroscopy, dynamic light scattering, and
circular dichroism. Protein science : a publication of the Protein Society. 2003;12(7):1507-1521.
91. Polverini, E., Boggs, J.M., Bates, I.R., Harauz, G., and Cavatorta, P. Electron paramagnetic
resonance spectroscopy and molecular modelling of the interaction of myelin basic protein
(MBP) with calmodulin (CaM)-diversity and conformational adaptability of MBP CaM-targets.
J. Struct. Biol. 2004;148(3):353-369.
92. Oettinger, H.F., al Sabbagh, A., Jingwu, Z., LaSalle, J.M., Weiner, H.L., and Hafler, D.A.
Biological activity of recombinant human myelin basic protein. J. Neuroimmunol.
1993;44(2):157-162.
79
93. Bamm, V.V., Ahmed, M.A.M., Harauz, G. Interaction of myelin basic protein with actin in
the presence of dodecylphosphocholine micelles. Biochemistry. 2010; 49(32):6903-6915.
94. Harauz, G., Ishiyama, N., Hill, C.M.D., Bates, I.R., Libich, D.S, and Fares C. Myelin basic
protein - diverse conformational states of an intrinsically unstructured protein and its roles in
myelin assembly and multiple sclerosis. Micron. 2004;35(7):503-542.
95. Costentino, M., Pritzker, L., Boulias, C., and Moscarello, M.A. Acylation of myelin basic
protein peptide 1-21 with alkyl carboxylates 2-10 carbons long affects secondary structure and
posttranslational modification. Biochemistry. 1994;33(14):4155-4162.
96. Buck, M. Trifluoroethanol and colleagues: Cosolvents come of age. recent studies with
peptides and proteins. Q. Rev. Biophys. 1998 08;31(3):297-355.
97. Buck, M. Radford, S.E., and Dobson, C.M. A partially folded state of hen egg white
lysozyme in trifluoroethanol: structural characterization and implication for protein folding.
Biochemistry.1993;32(2):669-678.
98. Imai, T., Kovalenko, A., Hirata, F., and Kidera, A. Molecular thermodynamics of
trifluoroethanol-induced helix formation: Analysis of the solvation structure and free energy by
the 3D-RISM theory. Interdiscip. Sci. 2009;1(2):156-160.
99. Povey, J.F., Smales, C.M., Hassard, S.J., and Howard, M.J.. Comparison of the effects of
2,2,2-trifluoroethanol on peptide and protein structure and function. J. Struct. Biol. 2007.
;157(2):329-238.
100. Sonnichsen, F.D., Eyk, J.E.V., Hodges, R.S., Sykes, B.D. Effect of trifluoroethanol on
protein secondary structure: An NMR and CD study using a synthetic actin peptide.
Biochemistry. 1992;31(37):8790-8798.
101. Libich, D.S., and Harauz, G. Solution NMR and CD spectroscopy of an intrinsically
disordered, peripheral membrane protein: Evaluation of aqueous and membrane-mimetic solvent
conditions for studying the conformational adaptability of the 18.5 kDa isoform of myelin basic
protein (MBP). Eur. Biophys. J. 2008;37(6):1015-1029.
102. Main, E.R., and Jackson, S.E. Does trifluoroethanol affect folding pathways and can it be
used as a probe of structure in transition states? Nat. Struct. Biol. 1999;6(9):831-835.
103. Lakowicz, J.R. Principles of Fluorescence. 3rd Ed. Kluwer Academic - Plenum Publishers
2006:- 443-475.
80
104. Valeur, B. A brief history of fluorescence and phosphorescence before the emergence of
quantum theory. J. Chem. Educ. 2011(6):731-738.
105. Grohmann, D., Werner, F., and Tinnefeld, P. Curr. Opin. Chem. Biol. 2013;17(4):691-698.
106. Haas, E. Ensemble FRET methods in studies of intrinsically disordered proteins. Methods
Mol Biol. 2012;(895):467-498.
107. Basak, S. Studies of protein folding and dynamics using single molecule fluorescence
spectroscopy. Phys. Chem. Chem. Phys.. 2014;16(23):11139-11149.
108. LiCata, V.J., and Wowor, A.J. Applications of fluorescence anisotropy to the study of
protein-DNA interactions. Method. Cell Bio. JID - 0373334. 1206(0091-679; 0091-679).
109. Musse, A.A., Wang, J., Deleon, G.P., Prentice, G.A., London, E., Merrill, A.R.. Scanning
the membrane-bound conformation of helix 1 in the colicin E1 channel domain by site-directed
fluorescence labelling. J. Biol. Chem. 2006;281(2):885-895.
110. Ishiyama, N. The effects of deimination of myelin basic protein on structures formed by its
interaction with phosphoinositide-containing lipid monolayers. J. Struct. Biol. 2001;136(1):3045.
111. Lee, D. Lipid domains control myelin basic protein adsorption and membrane interactions
between model myelin lipid bilayers. Proc. Natl. Acad. Sci. U. S. A. 2014;111(8):768-775.
112. Nath, A., Sammalkorpi, M., DeWitt, D.C., Trexler, A.J., Elbaum-Garfinkle, S., O'Hern,
C.S., et al. The conformational ensembles of alpha-synuclein and tau: Combining singlemolecule FRET and simulations. Biophys. J. 2012;103(9):1940-1949.
113. Deniz, A. A. Site-specific protein labelling for single-molecule fluorescence studies. In:
Roberts, G.K., editor. Springer Berlin Heidelberg; 2013. p. 2364-2368.
114. Taraska, J.W.. Mapping membrane protein structure with fluorescence. Curr. Opin. Struct.
Biol. 2012 (1879-033; 0959-440; 4):507-13.
115. Trexler, A.J., Rhoades, E. Function and dysfunction of α-synuclein: Probing conformational
changes and aggregation by single molecule fluorescence. Mol. Neurobio. 2012(2):622.
116. Mao, A. Net charge per residue modulates conformational ensembles of intrinsically
disordered proteins. Proc. Natl. Acad. Sci. U. S. A. 2010;107(18):8183-8138.
81
117. Dunker, A.K., and Obradovic, Z. The protein trinity - linking function and disorder. Nat
Biotechnol. 2001;19(9):805-806.
118. Polverini, E., Boggs, J.M., Bates, I.R., Harauz, G., and Cavatorta, P. Electron paramagnetic
resonance spectroscopy and molecular modelling of the interaction of myelin basic protein
(MBP) with calmodulin (CaM) diversity and conformational adaptability of MBP CaM-targets. J
Struct Biol. 2004;148(3):353-369.
119. Smith, P.K., Krohn, R.I., Hermanson, G.T., Mallia, A.K., Gartner, F.H., Provenzano, M.D.,
et al. Measurement of protein using bicinchoninic acid. Anal. Biochem. 1985;150(1):76-85.
120. Semisotnov, G.V., Rodionova, N.A., Razgulyaev, O.I., Uversky, V.N., Gripas', A.F., and
Gilmanshin, R.I. Study of the "molten globule" intermediate state in protein folding by a
hydrophobic fluorescent probe. Biopolymers. 1991;31(1):119-128.
121. Engelhard, M., Evans, P.A. Kinetics of interaction of partially folded proteins with a
hydrophobic dye: Evidence that molten globule character is maximal in early folding
intermediates. Protein Sci. 1995 04(8):1553-1562.
122. Myers, J.K., Pace, C.N., and Scholtz, J.M. A direct comparison of helix propensity in
proteins and peptides. Proc. Natl. Acad. Sci. U. S. A. 1997;94(7):2833-2837.
123. Fitzkee, N.C., and Rose, G.D. Reassessing random-coil statistics in unfolded proteins. Proc.
Natl. Acad. Sci. U. S. A. 2004;101(34):12497-12502.
124. Sancho, J. The stability of 2-state, 3-state and more-state proteins from simple spectroscopic
techniques - plus the structure of the equilibrium intermediates at the same time. Arch. Biochem.
Biophys. 2013;531(1-2):4-13.
125. Jasanoff, A., and Fersht, A.R.. Quantitative determination of helical propensities from
trifluoroethanol titration curves. Biochemistry. 1994;33(8):2129-35.
126. Soler-Gonzalez, A.S. and Fersht, A.R. Helix stability in barstar peptides. Eur. J. Biochem.
1997;249(3):724-32.
127. Santoro, M.M., and Bolen, D.W.. Unfolding free energy changes determined by the linear
extrapolation method. 1. unfolding of phenylmethanesulfonyl alpha-chymotrypsin using different
denaturants. Biochemistry. 1988;27(21):8063-8068.
128. Smith, P.K., Krohn, R.I., Hermanson, G.T., Mallia, A.K., Gartner, F.H., Provenzano, M.D.,
et al. Measurement of protein using bicinchoninic acid. Anal. Biochem. 1985; 150(1):76-85.
82
129. Ahmed, M.A., De Avila, M., Polverini, E., Bessonov, K., Bamm, V.V., and Harauz, G.
Solution nuclear magnetic resonance structure and molecular dynamics simulations of a murine
18.5 kDa myelin basic protein segment (S72-S107) in association with dodecylphosphocholine
micelles. Biochemistry. 2012;51(38):7475-7487.
130. Bates, I.R., Feix, J.B., Boggs, J.M., Harauz, G. An immunodominant epitope of myelin
basic protein is an amphipathic alpha-helix. J Biol Chem. 2004;279(7):5757-64.
131. Pritzker, L.B., Joshi, S., and Harauz, G. Deimination of myelin basic protein. 2. Effect of
methylation of MBP on its deimination by peptidylarginine deiminase. Biochemistry. 2000;
39(18): 6310-6316
132. Homchaudhuri, L., Polverini, E., Gao, W., Harauz, G., and Boggs, J.M. Influence of
membrane surface charge and post-translational modifications to myelin basic protein on its
ability to tether the fyn-SH3 domain to a membrane in vitro. Biochemistry. 2009;48(11):23852393.
133. Fares, C., Libich, D.S., and Harauz, G. Solution NMR structure of an immunodominant
epitope of myelin basic protein. conformational dependence on environment of an intrinsically
unstructured protein. FEBS J. 2006;273(1742-464; 3):601-614.
134. Libich, D.S.. Conformational choreography of myelin basic protein by solution NMR
spectroscopy [dissertation]. University of Guelph; 2008.
135. Harauz, G., and Ladizhansky, V. Structure and dynamics of the myelin basic protein family
by solution and solid-state NMR. In: Boggs, J.M., editor. New York: Nova Science Publishers;
2008. p. 197-232.
136. Holtzer, M.E., and Holtzer, A. Alpha-helix to random coil transitions: Determination of
peptide concentration from the CD at the isodichroic point. Biopolymers. 1992;32(12):16751677.
137. Nelson, J.W., and Kallenbach, N.R. Stabilization of the ribonuclease S-peptide alpha-helix
by trifluoroethanol. Proteins. 1986;1(3):211-217.
138. Feinstein, M.B., and Felsenfeld, H. Reactions of fluorescent probes with normal and
chemically modified myelin. Biochemistry.1975;14(14):3041-3048.
139. Bailey, R.W., Dunker, A.K., Brown, C.J., Garner, E.C., and Griswold, M.D.. Clusterin, a
binding protein with a molten globule-like region. Biochemistry. 2001;40(39):11828-40.
83
140. Stadler, A.M., Stingaciu, L., Radulescu, A., Holderer, O., Monkenbusch, M., Biehl, R., et
al. Internal nanosecond dynamics in the intrinsically disordered myelin basic protein. J. Am.
Chem. Soc. 2014;136(19):6987-94.
141. Polverini, E., Coll, E.P., Tieleman, D.P., and Harauz, G. Conformational choreography of
a molecular switch region in myelin basic protein--molecular dynamics shows induced folding
and secondary structure type conversion upon threonyl phosphorylation in both aqueous and
membrane-associated environments. Biochim. Biophys. Acta. 2011;1808(3):674-683.
142. Gow., A and Smith, R. The thermodynamically stable state of myelin basic protein in
aqueous solution is a flexible coil. J. Biochem.1989;257(2):535-540.
143. Randall, C.S., and Zand, R. Spectroscopic assessment of secondary and tertiary structure in
myelin basic protein. Biochemistry. 1985;24(8):1998-2004.
144. Epand, R.M., Moscarello, M.A., Zierenberg, B., and Vail, W.J. The folded conformation of
the encephalitogenic protein of the human brain. Biochemistry.1974;13(6):1264-7.
145. Moscarello, M.A., Gagnon, J., Wood, D.D., Anthony, J., and Epand, R. Conformational
flexibility of a myelin protein. Biochemistry. 1973;12(18):3402-6.
146. Libich, D.S., Ahmed, M.A.M., Zhong, L., Bamm, V.V., Ladizhansky, V., and Harauz, G.
Fuzzy complexes of myelin basic protein - NMR spectroscopic investigations of a polymorphic
organizational linker of the central nervous system. Biochem. Cell Biol. 2010; 88(2); 143-155.
147. Uversky, V.N. Introduction to intrinsically disordered proteins (IDPs). Chem Rev.
2014;114(13):6557-6560.
148. Fitzkee, N. Reassessing random-coil statistics in unfolded proteins. Proc. Natl. Acad. Sci.
U. S. A. 2004;101(34):12497-12502.
149. Uversky, V.N., Permyakov, S.E., Zagranichny, V.E., Rodionov, I.L., Fink, A.L., Cherskaya
A.M., et al. Effect of zinc and temperature on the conformation of the gamma subunit of retinal
phosphodiesterase: A natively unfolded protein. J. Proteome. Res. 2002;1(2):149-59.
150. Bates, I.R., Boggs, J.M., Feix, J.B., and Harauz, G. Membrane-anchoring and charge effects
in the interaction of myelin basic protein with lipid bilayers studied by site-directed spin
labelling. J. Biol. Chem. 2003;278(31):29041-29047.
151. Higgs, P.G., and Joanny, J.F. Theory of polyampholyte solutions. Chem.Phys.
1991;94:1543-1554.
84
152. Koh, J.Y. Zinc and disease of the brain. Mol. Neurobiol. 2001; 24(1-3):99-106.
153. Uversky, V.N. Intrinsically disordered proteins and their environment: Effects of strong
denaturants, temperature, pH, counter ions, membranes, binding partners, osmolytes, and
macromolecular crowding. Protein J. 2009;28:305-25.
154. Boggs, J. Myelin basic protein: A multifunctional protein. Cellular and molecular life
sciences. 2006;63(17):1945-1961.
155. Inouye, H., and Kirschner, D.A. Effects of ZnCl2 on membrane interactions in myelin of
normal and shiverer mice. Biochim Biophys Acta. 1984;776(2):197-208.
156. Smith, G.S.T., Chen, L., Bamm, V.V., Dutcher, J., and Harauz, G. The interaction of zinc
with membrane-associated 18.5 kDa myelin basic protein: An attenuated total reflectance-fourier
transform infrared spectroscopic study. Amino Acids. 2010: 39:739-750.
157. Bamm, V.V., Avila, M.D., Smith, G.S., Ahmed, M.A., and Harauz, G. Structured functional
domains of myelin basic protein: Cross talk between actin polymerization and ca(2+)-dependent
calmodulin interaction. Biophys J. 201;101(5):1248-1256.
158. De Avila, M., Ahmed, M.A.M., Smith, G.S.T., Boggs, J.M., and Harauz G. Modes of SH3domain interactions of 18.5 kDa myelin basic protein in vitro and in oligodendrocytes. Biophys.
J. 2011.
159. Vacic, V., Oldfield, C.J., Mohan, A., Radivojac, P., Cortese, M.S., Uversky, V.N., et al.
Characterization of molecular recognition features, MoRFs, and their binding partners.
J.Proteome Res. 2007;6(6):2351-66.
160. Uversky, V.N., and Dunker, A.K.. The case for intrinsically disordered proteins playing
contributory roles in molecular recognition without a stable 3D structure. F1000 Biol Rep.
2013;5(1757-594; 1757-594):1.
161. Main, E.R., and Jackson, S.E. Does trifluoroethanol affect folding pathways and can it be
used as a probe of structure in transition states? Nat. Struct. Biol. 1999;6(9):831-5.
162. Damaschun, G., Damaschun. H., Gast, K., and Zirwer, D.. Proteins can adopt totally
different folded conformations. J. Mol. Biol. 1999;291(3):715-25.
163. Libich, D.S., and Harauz, G. Backbone dynamics of the 18.5-kDa isoform of myelin basic
protein reveals transient alpha-helices and a calmodulin-binding site. Biophys. J.
2008;94(12):4847-66.
85
164. Aggarwal, S., Snaidero, N., Pahler, G., Frey, S., Sanchez, P., Zweckstetter, M., et al. Myelin
membrane assembly is driven by a phase transition of myelin basic proteins into a cohesive
protein meshwork. PLoS Biol. 2013;11(6):e1001577.
165. Lee, D.W., Banquy, X.F., Kristiansen, K.F., Min, Y.F., Ramachandran, A. et al. Adsorption
mechanism of myelin basic protein on model substrates and its bridging interaction between the
two surfaces. Langmuir : the ACS journal of surfaces and colloids. 2015
166. Vacic, V., Oldfield, C.J., Mohan, A., Radivojac, P., Cortese, M.S., Uversky, V.N., et al.
Characterization of molecular recognition features, MoRFs, and their binding partners. J.
Proteome Res. 2007 06;6(6):2351-2366.
167. Teufel, D.P., Johnson, C.M., Lum, J.K., and Neuweiler, H. Backbone-driven collapse in
unfolded protein chains. J. Mol. Biol. 201;409(2):250-62.
168. Haran, G. How, when and why proteins collapse: The relation to folding. Curr. Opin. Struct.
Biol. 2012;22(1):14–20.
169. Libich, D.S., Monette, M.M., Robertson, V.J., and Harauz, G. NMR assignment of an
intrinsically disordered protein under physiological conditions: The 18.5 kDa isoform of myelin
basic protein. Biomol. NMR Assign. 2007;1(1):61-3.
170. Brocca, S., Samalikova, M., Uversky, V.N., Lotti, M., Vanoni, M., Alberghina, L., et al.
Order propensity of an intrinsically disordered protein, the cyclin-dependent-kinase inhibitor
Sic1. Proteins. 2009 ;76(3):731-746.
171. Kurzbach, D., Platzer, G., Schwarz, T.C., Henen, M.A., Konrat, R., Hinderberger, D.
Cooperative unfolding of compact conformations of the intrinsically disordered protein
osteopontin. Biochemistry. 2013;52(31):5167-5175.
172. Lee, J.C., Langen, R., Hummel, P.A., Gray, H.B., Winkler, J.R. Alpha-synuclein structures
from fluorescence energy-transfer kinetics: Implications for the role of the protein in parkinson's
disease. Proc. Natl. Acad. Sci. U. S. A. 2004;101(47):16466-16471.
173. Bates, I.R., Boggs, J.M., Feix, J.B., and Harauz, G. Membrane-anchoring and charge effects
in the interaction of myelin basic protein with lipid bilayers studied by site-directed spin
labelling. J Biol. Chem. 2003;278(31):29041-29047.
86