100;

This is to certify that the
thesis entitled

FLUORESCENCE LINESHAPE STUDIES OF THERMAL
UNFOLDING AND SECONDARY STRUCTURE
TRANSITIONS IN PROTEINS

presented by

Sara E. McFadden

has been accepted towards fulfillment
of the requirements for the

MS. degree in Chemistry

 

 

 

Major Professor’s Signature

Date

 

MSU is an Affinnative Action/Equal Opportunity Institution

 

LIBRARY
Michigan State
University

 

 

 

PLACE IN RETURN BOX to remove this checkout from your record.
T0 AVOlD FINES return on or before date due.
MAY BE RECALLED with earlier due date if requested.

 

DATE DUE DATE DUE DATE DUE

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

2/05 p:/ClRC/DateDue.indd-p.1

 

‘1‘-

IIIIIIIOIIU.

flit-(nil

FLUORESCENCE LINESHAPE STUDIES OF THERMAL UNFOLDING AND
SECONDARY STRUCTURE TRANSITIONS IN PROTEINS

By

Sara E. McFadden

A THESIS
Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of
MASTER OF SCIENCE
Department of Chemistry

2006

ABSTRACT

FLUORESCENCE LINESHAPE STUDIES OF THERMAL UNFOLDING AND
SECONDARY STRUCTURAL TRANSITIONS IN PROTEINS

By

Sara E. McFadden

Using l-anilino-8-naphthalenesulfonic acid (ANSA) as an extrinsic hydrophobic
fluorescent probe, we have monitored protein and solvent motions in the thermal
unfolding and secondary structural transitions in proteins. ANSA is sensitive to the
polarity of its environment By first characterizing the lineshape of the AN SA
fluorescence spectrum with respect to changes in solvent polarity, we were able to relate
the changes of the lineshape to the relative change in polarity of the ANSA-binding site
in the proteins under study. Analysis of all the parameters associated with the lineshape
of the spectrum provided more structural information than monitoring the intensity of the
fluorescence alone.

In this thesis, we examine the thermal unfolding of the molten globule of bovine
a-lactalbumin and the a—helix to B—sheet secondary structure transition of poly-L-Iysine.
The results presented here show that the thermal unfolding of the molten globule of
a-lactalbumin is not a two-state transition as was previously suggested. Using the ANSA
fluorescence spectrum lineshape and Rayleigh light scattering from poly-L-lysine, we
show that the transition from the a-helix form to the B-sheet form is preceded by the

formation of a molten-globule—like intermediate, which then rapidly aggregates.

TO
Kate, Michele and my family

ACKNOWLEDGMENTS

I would like to thank my advisor, Dr. Warren Beck, and the Beck group for their help
and support during my time at MSU. I also want to thank my family and friends for all

their support.

iv

TABLE OF CONTENTS

Pages
LIST OF TABLES ................................................................................... vi
LIST OF FIGURES ................................................................................. vii
LIST OF ABBREVIATIONS ...................................................................... xi
CHAPTER] INTRODUCTION .................................................................. 1
References ........................................................................... 9
CHAPTER 2 LINESHAPE CHARACTERIZATION OF THE FLUORESCENCE
SPECTRUM OF l-ANILINO-S-NAPHTHALENE SULFONIC ACID
DURING THE THERMAL UNFOLDING OF THE a-LACTALBUMIN
MOLTEN GLOBULE
Introduction ........................................................................ 12
Sample Preparation ................................................................ 14
Experimental Setup ............................................................... 15
Results .............................................................................. 17
Discussion .......................................................................... 30
References .......................................................................... 36
CHAPTER 3 STUDY OF THE a—HELIX TO B-SHEET TRANSITION OF POLY-L-

LYSINE USING 1-ANILINO-8-NAPHTHALENESULFONIC ACID

AND LIGHT SCATTERING

Introduction ........................................................................ 38
Sample Preparation ................... . ............................................ 41
Experimental .......................... I ............................................. 41
Results .............................................................................. 43
Discussion .......................................................................... 53
References .......................................................................... 58

LIST OF TABLES

Table
Page
2.1 Parameters obtained by fitting the 5 uM ANSA fluorescence
spectrum from methanol, 5:10 uM ANSAzBLA at 4 °C and 90 °C,

and water to log-normal lineshapes ................................... 33

vi

Figure
l. l

1.2

1.3

2.1

2.2

2.3

2.4

LIST OF FIGURES

Page
Structure of 1,8-anilinonaphthalene sulfonic acid ..................... 5

Cartoon representation of the interaction of the dipole moment of
the chromophore and the surrounding solvent molecules. The
large circle with an arrow represents the chromophore and its
dipole moment and the small circles represent the solvent
molecules and their dipole moments. After Lakowicz.”. . . . . ......7

Energy level diagram corresponding to a solvatochromic probe in
solvents with different polarities. S0 is the ground state of the
chromophore, S1 is the first excited state of the chromophore, big,
is the energy of the absorption photon, th and th' is the energy
of the fluorescence photon in a less and more polar solvent,
respectively. After Lakowicz.29 ........................................... 8

Structure of bovine a-lactalbumin with the Ca2+ ion rendered as a
sphere. This picture created from protein data bank entry 1F6S
using PyMOL ............................................................. I3

Schematic diagram of the fluorescence spectrometer used in
experiments discussed in this chapter. Symbols: LED, 380 nm
light-emitting diode light source; 6 1, 1- inch focal length lens; 3,, 1-
inch focal length lens; 63, 4-inch focal length lens; Ml, collimating
mirror; M2, focusing mirror; Grating, 600 gr/mm grating; CCD,
2048 x 512 pixel CCD detector. The temperature-controlled
sample holder was 3 Quantum Northwest TLC 50 Fluorescence
Cuvette Holder ............................................................. 16

Fluorescence (Flu’) and absorption (A/ v) dipole-strength spectra
from a solution of 5 uM ANSA in pH 2.0 water. The fluorescence
spectrum is centered at 16940 cm'l and the absorption spectrum is
centered at 28620 cm'1 ................................................... 18

Fluorescence dipole-strength spectra from 20 uM AN SA solutions
with decreasing methanolzwater ratios. The percentage of methanol
in the methanol/water solvent mixture is shown beside each

spectrum ................................................................... 20

vii

2.5

2.6

2.7

2.8

2.9

2.10

2.11

Progression of the integrated relative fluorescence dipole strength
from 20 uM AN SA solutions with increasing methanolzwater
ratios. Each data point represents the total dipole strength of the
AN SA fluorescence spectrum obtained by fitting the spectrum to a
log-normal lineshape. The bottom panel shows a plot of In(re|ative
dipole strength) versus percent methanol is linear .............. , ..... 21

Progression of the center frequency of the ANSA fluorescence
spectrum from 20 uM ANSA solutions with increasing
methanolzwater ratios. Each data point represents the center
frequency of the AN SA fluorescence spectrum obtained by fitting
the spectrum to a log-normal lineshape ............................... 22

Progression of the half-width at half-maximum of the ANSA
fluorescence spectrum from 20 11M ANSA solutions with
increasing methanolzwater rato. Each data point represents the
half-width at half-maximum of the ANSA fluorescence spectrum
obtained by fitting the spectrum to a log—normal lineshape ........ 23

Progression of the AN SA fluorescence spectrum from a 5:10 uM
AN SAzBLA solution as the temperature of the sample was
increased from 4 °C to 90 °C. The temperature (°C) at which the
spectrum was collected is indicated. With the increase in
temperature, the AN SA fluorescence spectrum broadens and red
shifts from being centered at 20180 cm’1 at 4 °C to 18310 cm‘1 at
90 °C ....................................................................... 25

Evolution of the integrated dipole strength of the ANSA
fluorescence spectrum in a 5:10 uM ANSA:BLA solution as the
solution was heated from 4 °C to 90 °C. The average integrated
dipole strength from a set of three experiments is plotted. Error
bars mark +/- one standard deviation .................................. 27

Evolution of the center frequency of the ANSA fluorescence
spectrum from a 5:10 uM ANSA:BLA solution as a function of
temperature from 4 °C to 90 °C. The average center frequency
from a set of three experiments is shown. Error bars mark +/— one
standard deviation ........................................................ 28

Evolution of the half-width at half-maximum of the AN SA
fluorescence spectrum in a 5:10 uM ANSA:BLA solution upon
heating of the solution from 4 °C to 90 °C. The average half-width
at half-maximum from a set of three experiments is shown. Error
bars mark +/- one standard deviation .................................. 29

viii

2.12

3.1

3.2

3.3

3.4

3.5

3.6

Comparison of the fluorescence dipole-strength spectrum of 5 11M
ANSA in methanol, water and a 5:10 uM ANSA:BLA solution at
4 and 90 °C. The bottom panel shows an expanded view of the

AN SA fluorescence spectrum from the 5:10 uM AN SAzBLA
solution at 90 °C and 5 uM ANSA in pH 2.0 water ................. 32

Time evolution of the temperature of the sample solution as
measured by a thermocouple. The temperature controller of the
sample holder was changed from 23 °C to 55 °C at the plotted
zero of time ................................................................ 42

Fluorescence dipole—strength (F/V‘) spectra from a 7:14 11M

AN SAzPLL solution as a function of time following a change in
temperature, from 23 °C to 55 °C. The scattered-light peak,
centered at 26600 cm", is superimposed with a fit to a Lorentzian
lineshape (short dashes). The ANSA fluorescence spectrum,
centered at 19000 cm", is superimposed with a fit to a log-normal
lineshape (long dashes). The indicated time is that following the
sample temperature change ............................................. 44

Expanded view of the time evolution of the 380-nm scattered-light
peak detected in the 7:14 uM ANSA:PLL solution. The indicated

time is that following the sample temperature change from 23 °C
to 55 °C ................................................................... 45

Expanded view of the time evolution of the ANSA fluorescence
spectrum detected in the 7:14 11M ANSA:PLL solution. The

indicated time is that following the sample temperature change
from 23 °C to 55 °C .................................................... 46

Time evolution of the area (integrated dipole strength) of the
scattered-light peak observed in a 7:14 11M ANSA:PLL solution.
The average area from a set of six experiments is shown. Error
bars mark +/— one standard deviation. The area of the scattered
light peak at the plotted zero of time (the time at which the
temperature change from 23 °C to 55 °C occurred) was subtracted
from the plotted areas .................................................... 48

Time evolution of the integrated dipole strength of the ANSA
fluorescence spectrum in a 7:14 1.1M AN SAzPLL solution. The
average integrated dipole strength from a set of six experiments is
plotted. Error bars mark +/- one standard deviation. The sample
temperature was changed from 23 °C to 55 °C at the plotted zero
of time ..................................................................... 49

ix

3.7

3.8

3.9

Time evolution of the average center frequency of the ANSA
fluorescence spectrum observed in a 7:14 11M AN SAzPLL
solution. The average center frequency from a set of six
experiments is shown with error bars marking +/- one standard
deviation. The sample temperature was changed from 23 °C to

55 °C at the plotted zero of time ....................................... 51

Time evolution of the half-width at half-maximum of the ANSA
fluorescence spectrum in a 7:14 11M AN SAzPLL solution. The
average half-width at half-maximum from a set of six experiments
is shown. Error bars mark +/- one standard deviation. The sample
temperature was changed from 23 °C to 55 °C at the plotted zero
of time ..................................................................... 52

Comparison of the time evolution of the ANSA fluorescence
spectrum (area (0), center frequency (X), half-width at half-
maximum (A)) and the scattered-light peak (+) from the 7:14 uM
ANSA:PLL solution following a 23 °C to 55 °C temperature
change. The signals have been normalized to zero at toand to unity
at t = 7250 s. The ANSA half-width at half-maximum response
has been inverted ......................................................... 55

AN SA

BLA

CD

FTIR

HCl

LED

MG

NaOH

NMR

PLL

LIST OF ABBREVIATIONS
l-anilino -8-naphtha1enesulfonic acid
Bovine a-lactalbumin
Circular Dichroism
Fourier Transform Infrared Spectroscopy
Hydrochloric Acid
Light-Emitting Diode
Molten Globule
Sodium Hydroxide
Nuclear Magnetic Resonance

Poly-L-lysine

xi

CHAPTER 1

INTRODUCTION

There are four basic levels of protein structure. The sequence of amino acids in the
protein is known as the primary structure and the physical arrangement of the primary
structure into a-helices or b-strands is known as the secondary structure. These small
regions of secondary structure fold and interact with one another to form the native
tertiary structure. In large proteins, domains of tertiary structure interact with one
another to form the quaternary structure.1

One important process that has yet to be understood is that by which a sequence of
amino acids folds into a specific three—dimensional protein structure?3 An understanding
of this process would theoretically allow a researcher with a need for a specific
three-dimensional structure to determine the sequence of amino acids that would fold
into the target structure. A better knowledge of how a protein folds may also give insight
into what causes a protein to fold improperly or to lose its folded—state stability.

Protein misfolding is thought to be a direct cause for the transmissible spongiform
encephalopathy diseases that include Cruetzfeldt-Jakob disease in humans, scrapie in
sheep, and bovine spongiform encephalopathy (“mad cow disease”).“'5 Alzheimer’s
disease and Parkinson’s disease are amyloidoses which are also associated with protein
misfolding.“7 These diseases are thought to be associated with the accumulation of
amyloid fibrils, but the critical step seems to be the decay of the native folded form of the

protein in question.2

Native proteins have both hydrophilic topologies and hydrophobic cores. This
combination allows the protein to be soluble and to protect its hydrophobic core from the
solvent environment The protection of the hydrophobic core from the solvent plays a
critical role in the stability of the native fold.8 Amyloid fibrils, which are derived from
native proteins, are insoluble.9 Proteins that fail to assume the proper native fold form
small soluble aggregates that associate into protofibrils. The protofibrils go on to form
insoluble mature fibrils. The initial aggregation, however, is always associated with a
misfolded or partially unfolded state of the proteins"'0

The ability of polypeptides to form amyloid fibrils is apparently a general property
and not necessarily dependent on specific amino—acid sequences or native—fold
topologies.ll The molten globule state of a—lactalbumin at low pH, and other members of
the lysozyme family of proteins, have been shown to form amyloid fibrils at high salt
concentrations.” Regardless of the protein of origin, the fully formed fibrils exhibit
nearly identical structures.5"° The rates at which the misfolded proteins aggregate,
however, depend strongly on the physicochemical properties of the proteins, such as the
spatial organization of charges, a preference for certain secondary structural features, and
hydrophobicity.l3

The current trend in the debate over what role misfolded or aggregated protein
structures play in the progress of disease points to small aggregate structures from
misfolded proteins as the disease—causing species. It was found that structurally
disorganized pre-fibrilar aggregates are highly cytotoxic while structurally well-defined

mature fibrilar forms of the same protein are mostly harmless?”-I The pre—fibrilar forms

are thought to be toxic because they expose hydrophobic regions of the protein to the
surrounding solvent.ls

In the so—called new view of protein folding, there are thought to be many possible
folding trajectories between the unfolded and native states on the folding
potential—energy surface, which has been termed an energy landscape.um A
funnel—shaped topology of the potential-energy surface by which folding proteins are
steered toward their native states has been suggested.2022 This view requires consideration
of the dynamics of an ensemble of folding trajectories rather than the standard two—state
reaction-coordinate that is suggested by most biochemical and spectroscopic
experiments.3*23'“ It is possible that protein misfolding results from alternate Gibbs
free—energy minima on the landscape that act as kinetic sinks. The picture of a protein
folding funnel with local Gibbs free—energy minima suggest that the native structure is
formed under a combination of kinetic and thermodynamic control.” Because folded
proteins are only marginally stable, misfolded proteins may also form spontaneously
from the native state.20 Given that the topology of the funnel would suggest steeply
sloping sides close to the native state, it is likely that most unfolding trajectories lead to
refolding of the native structure.

Assuming the process by which a protein folds is similar to the process by which it
unfolds, the study of protein unfolding should give information about the
potential—energy surface of its folding funnel. For example, intermediates in the
unfolding of a protein would suggest that there are local energy minima on the walls of

the funnel. If the intermediates that collect in these kinetic traps could be isolated and

characterized, one can imagine that specific pharmaceuticals might be designed to
recognize intermediates and retard the formation of the disease—causing products.

The formation of the hydrophobic core is thought to provide most of the energetic
stabilization of the native fold.7 Redfield and coworkers have shown that in
a—lactalbumin, the native—like three-dimensional fold is formed through interactions of
regions that are at opposite ends in the sequence code.26 They attribute the stability of the
fold to the grouping of hydrophilic and hydrophobic residues.

In this thesis, we examine the nature of events in the unfolding of the hydrophobic
core in two protein systems. We use an extrinsic chromophore,
l—anilino—8—naphthalenesulfonic acid (AN SA), that is sensitive to changes in polarity to
sense protein and solvent motions in partially unfolded protein states and in changes of
secondary structure. First, we monitor the flow of solvent molecules into the
hydrophobic core of the a-lactalbumin molten globule as it unfolds. We then monitor the
formation of a molten-globule—like intermediate as poly-L-lysine undergoes a
conformational change from a—helix to B—sheet.

Because of the sensitivity of ANSA (see Figure 1.1) to polarity, it has been widely
used as an extrinsic probe in the study of proteins and membrane composition and
function?”29 The fluorescence quantum yield of AN SA in water is less than 10". As the
polarity of the solvent surrounding the ANSA molecule decreases, the quantum yield
increases.” A decrease in the quantum yield of the ANSA fluorescence with increasing
solvent polarity can be attributed to a larger probability of non-radiative transitions as the
larger dipole of the excited state interacts more readily with a more polar solvent?0 A low

fluorescence quantum yield of ANSA in water allows for any signal of unbound dye to be

Figure 1.1. Structure of 1,8—anilinonaphthalenesulfonic acid.

neglected. Therefore, if the probe molecule is bound to a protein, any change in the

AN SA spectrum can be attributed to a motion of the protein which affects the polarity of
the ANSA environment. A decrease in polarity of the solvent also causes the emission
maximum to shift to the blue.

The spectral changes with respect to polarity are due to solvatochromism. When the
ground—state (S0 state) probe molecule absorbs a photon, it is promoted to the first
excited singlet state (S,). In this excited state, the dipole moment of the molecule is
larger than it is in the ground state. Solvent molecules that surround the excited state
probe molecule rotate to align their dipole moments with the dipole moment of the
excited species and lower the energy of the excited state. The more polar the solvent, the
more the energy of the excited state is decreased. 3‘ (See Figures 1.2 and 1.3) This
results in a shift of the spectrum to the red as the polarity of the solvent increases.

In this thesis, we show that a fit of the ANSA fluorescence spectrum to a log-normal
lineshape32 can provide structural information that is not provided by an'analysis of
intensity alone. The fit to a log-normal lineshape gives three spectral parameters: area,
center frequency, and half-width at half-maximum. The area of the log-normal lineshape
is proportional to the fluorescence quantum yield of the probe, the center frequency is
sensitive to the polarity of the probe environment, and the half—width at half—maximum is
sensitive to the range of structures and binding sites in the ensemble. A comparison of the
rates at which these parameters change can provide considerably more insight into the

change in probe environment with motions of the protein.

Solvent
Relaxation (D d)

(D
@809 W 35,01:
0

Absorption 1 Fluorescence

L
'

 

(I)
<13 <I>

® V 11
a0 6 Rita; "
‘9 Q 9

Figure 1.2. Cartoon representation of the interaction of the dipole moment of the
chromophore and the surrounding solvent molecules. The large circle with an arrow
represents the chromophore and its dipole moment and the small circles represent the
solvent molecules and their dipole moments. After Lakowicz.31

 

 

 

 

 

 

 

 

 

E
‘ E
A

SI i ~~~$ Less Polar
Solvent
More Polar
Solvent
hVF th'
h VA
IV V
4/”’
S0

 

 

Figure 1.3. Energy level diagram corresponding to a solvatochromic probe in solvents
with different polarities. S0 is the ground state of the chromophore, S1 is the first excited
state of the chromophore, th is the energy of the absorption photon, hvF and th' is the
energy of the fluorescence photon in a less and more polar solvent, respectively. After
Lakowicz.31

References

(1)

(2)
(3)

(4)

(5)

(6)

(7)

(8)

(9)

(10)

(11)

(12)

(13)

(14)

Branden, C.; Tooze, J. Introduction to Protein Structure; 2nd ed.; Garland
Publishing, Inc.: New York, 1999.

Dobson, C. M. “Protein folding and misfolding.” Nature 2003, 426, 884—890.

Dill, K. A.; Chan, H. S. “From Levinthal to pathways to funnels.” Nature Struct.
Biol. 1997, 4, 10—19.

Dobson, C. M. “Protein misfolding, evolution and disease.” Trends BiochemSci.
1999, 24, 329—332.

Prusiner, S. B. “Prions.” Proc. Natl. Acad. Sci. USA 1998, 95, 13363—13383.

Caughey, B.; Lansbury, P. T. “Protofibrils, pores, fibrils, and neurodegeneration:
Separating the responsible protein aggregates from the innocent bystanders.”
Annu. Rev. Neurosci. 2003, 26, 267—298.

Selkoe, D. J. “Folding proteins in fatal ways.” Nature 2003, 426, 900—904.

Kauzmann, W. “Some factors in the interpretation of protein denaturation.” Adv.
Protein. Chem. 1959, 14, 1-63.

Dobson, C. M. "The structural basis of protein folding and its links with human
disease.” Philos. Trans. R. Soc. Lond. Ser. B-Biol. Sci. 2001, 356, 133-145.

Sunde, M.; Blake, C. “The structure of amyloid fibrils by electron microscopy and
X-ray diffraction.” In Adv. Prot. Chem, 1997; Vol. 50, p 123—159.

Chiti, F.; Webster, P.; Taddei, N.; Clark, A.; Stefani, M.; Ramponi, G.; Dobson,
C. M. “Designing conditions for in vitro formation of amyloid protofilaments and
fibrils.” Proc. Natl. Acad. Sci. USA 1999, 96, 3590—3594.

Goers, J .; Perrnyakov, E. A.; Perrnyakov, S. E.; Uversky, V. N.; Pink, A. L.
“Conformational Prerequisites for a—Lactalbumin Fibrillation.” Biochemistry
2002, 41, 12546.

Chiti, F.; Stefani, M.; Taddei, N.; Ramponi, G.; Dobson, C. M. “Rationalization
of the effects of mutations on peptide and protein aggregation rates.” Nature
2003, 424, 805—808.

Hartley, D. M.; Walsh, D. M.; Ye, C. P. P.; Diehl, T.; Vasquez, S.; Vassilev, P.
M.; Teplow, D. B.; Selkoe, D. J. “Protofibrillar intermediates of amyloid beta-
protein induce acute electrophysiological changes and progressive neurotoxicity
in cortical neurons.” J. Neurosci. 1999, I 9, 8876—8884.

(15)

(16)

(17)

(18)

(19)

(20)

(21)

(22)

(23)

(24)

(25)

(26)

(27)

Bucciantini, M.; Giannoni, E.; Chiti, F.; Baroni, F.; Forrnigli, L.; Zurdo, J. S.;
Taddei, N.; Ramponi, G.; Dobson, C. M.; Stefani, M. “Inherent toxicity of
aggregates implies a common mechanism for protein misfolding diseases.”
Nature 2002, 416, 507—51 1.

Bucciantini, M.; Calloni, G.; Chiti, F.; Formigli, L.; Nosi, D.; Dobson, C. M.;
Stefani, M. “Prefibrillar amyloid protein aggregates share common features of
cytotoxicity.” J. Biol. Chem. 2004, 279, 31374—31382.

Cleary, J. P.; Walsh, D. M.; Hofmeister, J. J.; Shankar, G. M.; Kuskowski, M. A.;
Selkoe, D. J.; Ashe, K. H. “Natural oligomers of the amyloid-protein specifically
disrupt cognitive function.” Nat. Neurosci. 2005, 8, 79—84.

Frauenfelder, H.; Sligar, S. G.; Wolynes, P. G. “The energy landscapes and
motions of proteins.” Science 1991, 254, 1598-1603.

Brooks III, C. L.; Onuchic, J. N.; Wales, D. J. “Statistical Thermodynamics:
Taking a Walk on a Landscape.” Science 2001, 293, 612-613.

Onuchic, J. N.; Luthey—Schulten, Z.; Wolynes, P. G. “Theory of Protein Folding:
The Energy Landscape Perspective.” Annu. Rev. Phys. Chem. 1997, 48, 545—600.

Bryngelson, J. D.; Wolynes, P. G. “Spin glasses and the statistical mechanics of
protein folding.” Proc. Natl. Acad. Sci. U. S. A. 1987, 84, 7524—7528.

Leopold, P. E.; Montal, M.; Onuchic, J. N. “Protein folding funnels: a kinetic
approach to the sequence-structure relationship.” Proc. Natl. Acad. Sci. U. S. A.
1992, 89, 8721-8725.

Chan, H. S.; Dill, K. A. “Protein folding in the landscape perspective: Chevron
plots and Non-Arrhenius kinetics.” Proteins: Struct. F unct. Genet. 1998, 30,
2—33.

Dill, K. A. “Polymer principles and protein folding.” Protein Sci. 1999, 8,
1166—1180.

Baker, D.; Agard, D. A. “Kinetics Versus Thermodynamics in Protein-Folding.”
Biochemistry 1994, 33, 7505—7509.

Redfield, C.; Schulman, B. A.; Milhollen, M. A.; Kim, P.; Dobson, C. M.
“a—Lactalbumin forms a compact molten globule in the absence of disulfide
bonds.” Nat. Struct. Biol. 1999, 6, 948—952.

Slavik, J. “Anilinonaphthalene Sulfonate as a Probe of Membrane Composition
and Function.” Biochimica Et Biophysica Acta 1982, 694, 1-25.

10

(28)

(29)

(30)

(31)

(32)

Turner, D. C.; Brand, L. “Quantitative Estimation of Protein Binding Site Polariy.
Fluorescence of N—Arylaminonaphthalenesulfonates.” Biochemistry 1968, 7,
3381-3390.

Weber, G.; Daniel, E. “Cooperative Effects in Binding by Bovine Serum
Albumin. II. The Binding of 1-Anilino—8—napthalenesulfonate. Polarization of
the Ligand Fluorescence and Quenching of the Protein Fluorescence.”
Biochemistry 1966, 5, 1900—1907.

Edelman, G. M.; McClure, W. O. “Fluorescent Probes and the Conformation of
Proteins.” Accounts of Chemical Research 1968, I , 65-70.

Lakowicz, J. R. Principles of Fluorescence Spectroscopy; Second ed.; Kluwer
Academic/Plenum Publishers: New York, 1999.

Siano, D. B.; Metzler, D. E. “Band shapes of the electronic spectra of complex
molecules.” Journal of Chemical Physics 1969, 51, 1856-1861.

11

CHAPTER 2
LINESHAPE CHARACTERIZATION OF THE FLUORESCENCE SPECTRUM OF
1-ANILINO-8-NAPHTHALENESULFONIC ACID DURING THE THERMAL
UNFOLDING OF THE a—LACI‘ALBUMIN MOLTEN GLOBULE
Introduction

In this chapter, we characterize the change in lineshape of the
1-anilino-8-naphthalenesulfonic acid (AN SA) fluorescence spectrum during the thermal
unfolding of the a-lactalbumin molten globule. As a first step, we characterize the
sensitivity of the fluorescence spectrum lineshape to a change in polarity by varying the
concentrations of methanol and water. Once the changes in spectral lineshape that occur
in response to the change in methanol concentration have been established, we will use it
as a gauge to detect changes in the hydrophobic core of the molten globule of
a-lactalburnin as it unfolds.

Bovine a-lactalbumin (BLA) is a relatively small protein (~14 kDa) that plays a vital
role in the production of milk in mammals.1 It has been widely studied, however,
because of its ability to readily form a stable molten globule (MG).2 The MG state is a
partially denatured state that is characterized by a native-like secondary structural
component but with few native tertiary interactions and a fluctuating hydrophobic core.3
BLA has two main domains; one domain that is primarily a -helical and a second domain
that contains predominantly a B-sheet structure (see Figure 2.1).4 Dobson and co-workers
have used NMR techniques coupled with hydrogen exchange studies to show that the
MG states of or-lactalbumins are structurally similar to the native states, but have
expanded hydrophobic cores and lack the tight side-chain packing of the native states.”8

Fluorescence studies of a-lactalbumin using tryptophan residues also indicate that the

12

 

Figure 2.1. Structure of bovine a-lactalbumin with the Ca2+ ion rendered as a sphere.
This picture created from protein data bank entry 1F6S using PyMOL.

l3

MG state has native-like tertiary interactions but a more globular form than the native
state.9 Native BLA has been known to partially unfold to the molten globular state under
the following conditions: acidic pH, high temperature, low concentration of chemical
denaturants, and removal of the Ca2+ ion.lo Recently, NMR data has been used to
determine a “coarse-grained” free energy landscape for the folding of the MG form of
human cr-lactalbumin.ll

AN SA has been shown to preferentially bind to the MG state of proteins rather than
to the native or fully unfolded states. Even in proteins with hydrophobic side chains on
the surface, AN SA will not readily bind to the native or fully unfolded states.”"13 The
expanded hydrophobic core of the MG allows the AN SA to bind within the core of the
protein.

Using AN SA as an extrinsic probe, we can monitor the flow of solvent molecules into
the hydrophobic core of MG a-lactalbumin as it unfolds. In this chapter we show that the
transition from the MG state to the fully unfolded state is not a two-state transition as was
previously believed from the results of Dobson and coworkers and Arai and

Kuwajima.“15

Sample Preparation

ANSA (8-anilino—l—naphthalenesulfonic acid ammonium salt) and BLA
(a-lactalbumin from bovine milk) were obtained from Sigma and used as received.
Samples were dissolved in methanol (Spectrum, spectrophotometric grade) or water as
stated to obtain an absorbance of less than 0.15 at 350 nm with a 1-cm path length. The

aqueous BLA samples were brought to pH 2.0 with addition of HCl. Prior to collection

14

of fluorescence spectra, the samples were passed through 0.22 um filters. A 1:2
chromophore:protein molar ratio was used to limit the possibilities of a protein having

more than one bound chromophore.”3

Experimental Setup

Figure 2.2 shows a schematic diagram of the fluorescence spectrometer used in
experiments discussed in this chapter. An 380-nm light-emitting diode (Ocean Optics,
LS-450) was used as the excitation source. Light from the LED was directed through an
l-m optical fiber. It was then focused into the sample cuvette by a 1-inch focal length
fused-silica lens.

A Quantum Northwest TLC 50 Fluorescence Cuvette Holder was used as the sample
holder. The use of this sample holder allowed for heating, cooling, and stirring of the
sample. The TLC 50 fluorescence cuvette holder employs a Peltier device for
temperature control and requires a flow of water to cool the Peltier chip’s heat sink. To
prevent condensation buildup during temperature changes, the sample compartment was
purged with nitrogen.

Within the sample cuvette, the focus of the 380—nm light from the excitation source
was directed slightly toward the collection optics to minimize self-absorption of the
fluorescence signal by the sample. A 1-inch focal length fused—silica lens was placed
directly after the optical window at 90° with respect to the excitation source to collect the
fluorescence emitted from the sample. A 4-inch focal length fused-silica lens focused the

fluorescence signal into a Spex 270M Spectrograph. The entrance slit of the

15

 

Computer

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Sample
. e Holder

Optical 1
Fiber

CD [2

C £3 CCD
S!“ C —

" Grating

I

I

I

I

I

I

I

I

I

I

/ \ '
I

I

I

I

l

I

l

l

l

I

I

I

I
I

 

 

M1 Spex 270M ,
Spectrograph M2

 

 

Figure 2.2. Schematic diagram of the fluorescence spectrometer used in experiments
discussed in this chapter. Symbols: LED, 380 nm light—emitting diode light source; (I, 1-

inch focal length lens; (2, l-inch focal length lens; (3, 4-inch focal length lens; Ml,
collimating mirror; M2, focusing mirror; Grating, 600 gr/mm grating; CCD, 2048 x 512

pixel CCD detector. The temperature-controlled sample holder was a Quantum
Northwest TLC 50 Fluorescence Cuvette Holder.

16

spectrograph was fixed at 0.5 mm giving the spectrograph an effective band pass of

3.09 nm. After passing through the entrance slit of the spectrograph, the fluorescence
signal was directed at a 600-groove/mm diffraction grating by a large collimating mirror.
Following the grating, a focusing mirror focused the diffracted spectrum onto a Spex
SpectrumOne 2048 x 512 pixel CCD detector. The CCD was positioned at the slit plane
of the spectrograph so that the emission was focused onto the CCD detector elements. A
shutter on the entrance slit of the spectrograph controlled the amount of time the CCD
was exposed during each experiment.

Spex SpectraMax software, as provided with the CCD detector, was used for data
collection. Automation of data collection was possible using QuicKeys software to
control the sequence of events in the SpectraMax software. The baseline signal from the
CCD detector was removed from the fluorescence spectra through subtraction of blank
solvent-only spectra that were recorded under the same measurement conditions with the
same cuvette. When plotted as a function of wavenumber, the fluorescence intensity was
multiplied by the square of the wavelength as required to compensate for the fixed
effective spectral band pass of the spectrometer.'7

Absorption spectra were obtained with a 2-nm spectral band pass at room temperature

with a Hitachi U-2000 spectrometer and LabVIEW (National Instruments) routines.
Results

Figure 2.3 shows the fluorescence (F/ 123) and absorption (A/v) relative dipole-strength

spectra of 5 11M ANSA in pH 2.0 water. The dipole strength measures the square of the

17

 

 

  

 

 

 

I
l
I
.: Ah I
‘g': I
m I \"’ - ~ ~ I
b I \ l
U) I \ I
2 ,’ \ .’
8. I \\ I
5 I \\ [I
a, F/v3 ,’ \ ,
..>. I _/
E I
G)
D:
16.0 20.0 24.0 28.0 32.0

Energy (103 cm")

Figure 2.3. Fluorescence (Flt/3) and absorption (A/ u) dipole-strength spectra from a
solution of 5 uM AN SA in pH 2.0 water. The fluorescence spectrum is centered at

16940 cm'1 and the absorption spectrum is centered at 28620 cm".

18

transition-dipole moment.l8 The fluorescence and absorption spectra have log-normal
lineshapes'9 that are mirrored to one another with respect to the direction of the skew, or
asymmetry, of the lineshape. The fluorescence spectrum is centered around 16940 cm".
The absorption spectrum is centered around 28620 cm". The optical density of the 20 11M
ANSA solution in neat methanol was 0.12. The AN SA absorption spectrum arises from a
series of overlapping bands, with the one at 28620 cm'1 associated with the lowest singlet
state from which the fluorescence arises.

The fluorescence dipole-strength spectra from solutions of 20 11M ANSA with
decreasing methanol to water ratios are shown in Figure 2.4. The fluorescence spectra
shown in Figure 2.4 were collected with 15-second integration times and 15 averaged
accumulations. As the methanol to water ratio decreases, the polarity of the solution
increases. With the increase in solvent polarity, the ANSA fluorescence spectrum
decreases in relative dipole strength and shifts to the red.

Figures 2.5 — 2.7 report the change in lineshape of the ANSA fluorescence spectrum
with the change in polarity of solvent. The fluorescence spectrum obtained at a given
methanol concentration was fit to a log-normal lineshape. The skew, or asymmetry, of
the fitted log-normal lineshape was held constant at 1.14; this parameter does not change
measurably as the methanol concentration is changed. Figure 2.5 shows that as the
percentage of methanol increases, the area increases exponentially. The plot of In(dipole
strength) versus percent methanol is linear. The area is equivalent to the integrated
relative dipole strength, which is proportional to the fluorescence quantum yield.18 The
area of the ANSA fluorescence spectrum in 100 percent methanol is more than 75 times

larger than that of the ANSA fluorescence spectrum in 99.5 percent water.

19

 

0.5

Relative Dipole Strength

 

30

 

50

 

90

100

 

 

I
20.0

Energy (103 cm")

25.0

Figure 2.4. Fluorescence dipole-strength spectra from 20 uM AN SA solutions with
decreasing methanolzwater ratios. The percentage of methanol in the methanol/water

solvent mixture is shown beside each spectrum.

20

 

2.0x 105

 

 

 

 

%, 1.0x105-

C

e

(75

2 0.0_

.3

D

.3 —1.0x1o5-

g

Q

E

E —2.0x105-
_ l I I I I l_ I l 1 l
3”“05 o 20 4o 60 so 100

Percent Methanol

Figure 2.5. Progression of the integrated relative fluorescence dipole strength from

20 11M ANSA solutions with increasing methanolzwater ratios. Each data point represents
the In(total dipole strength) of the AN SA fluorescence spectrum obtained by fitting the
spectrum to a log-normal lineshape.

21

 

 

 

' o
,3." O
|E 1.9'104 '-
9, o
5‘ o
C ..
é o
u. o
1: 1.8-104 A
9 O
C
o o
0 .
o
1'7.1O4 '- t J I l l n
0 20 40 60 80 100

Percent Methanol

Figure 2.6. Progression of the center frequency of the AN SA fluorescence spectrum
from 20 uM ANSA solutions with increasing methanolzwater ratios. Each data point

represents the center frequency of the ANSA fluorescence spectrum obtained by fitting
the spectrum to a log-normal lineshape.

22

 

 

 

 

- o
A 3600 - o
T o
5 o
.C ' 0
:15
E 3400 _ ’
a O
I o
' o
o
3200 —
I l I I I I
O 20 40 60 80 100

Percent Methanol

Figure 2.7. Progression of the half-width at half-maximum of the AN SA fluorescence
spectrum from 20 [1M ANSA solutions with increasing methanolzwater ratio. Each data
point represents the half-width at half-maximum of the AN SA fluorescence spectrum
obtained by fitting the spectrum to a log-normal lineshape.

23

 

Figure 2.6 shows the center frequency of the AN SA fluorescence spectrum as a
function of methanol concentration. The observed linear response arises from the change
in polarity of the solvent.” With a decrease in polarity of the solvent, the ANSA
fluorescence spectrum blue shifts from 17047 cm‘1 in 99.5 percent water to 19423 cm'1 in
100 percent methanol. As the polarity decreases, the spectrum shifts to the blue because
the S0 —) Sl transition increases the dipole moment of the chromophore.

The half-width at half-maximum of the AN SA fluorescence spectrum, as shown in
Figure 2.7, decreases linearly with an increase in methanol concentration. In the solution
with 100 percent methanol, the half-width at half-maximum of the AN SA fluorescence
spectrum is 3220 cm", whereas in the 0.5 percent methanol solution the half-width at
half-maximum of the AN SA fluorescence spectrum is 3700 cm". This relatively small
change in the width of the spectrum probably reports the change in hydrogen-bonding
structure that accompanies the solvent change.

Having recorded the change in the AN SA fluorescence spectrum as a function of
methanol concentration, we can now use the change in lineshape as a measure of solvent
polarity. This information can be used to monitor the flow of water molecules into the
hydrophobic core of an unfolding protein.

The progression of the ANSA fluorescence spectrum from a 5:10 uM ANSA:BLA
solution as a function of temperature from 4 °C to 90 °C is shown in Figure 2.8. All
spectra were collected with 30-second integrations and 40 averaged accumulations. The
sample was allowed to equilibrate for approximately ten minutes following each
temperature change before collection of the spectrum. The ANSA fluorescence spectrum

shifts to the red and decreases in area as the temperature increases.

24

 

 

 

 

 

 

' 4

5 2.0 -
U)
c
9
37) - 2
2
8.
'- 30
D _
g 1.0 60
.. 40
E 7090
g . 50

0'0 I 1 I l . I

16.0 20.0 24.0

Energy (1 03 cm”)

Figure 2.8. Progression of the ANSA fluorescence spectrum from a 5:10 11M
ANSA:BLA solution as the temperature of the sample was increased from 4 °C to 90 °C.
The temperature (°C) at which the spectrum was collected is indicated. With the increase
in temperature, the ANSA fluorescence spectrum broadens and red shifts from being
centered at 20180 cm" at 4 °C to 18310 cm'I at 90 °C.

25

Figures 2.9-2.11 report the change in lineshape of the ANSA fluorescence spectrum
as a function of temperature. The spectrum obtained at a given temperature was fit to a
log-normal lineshape. As in the methanol solution, the skew of the lineshape was held
constant at 1.14; this parameter does not change measurably as the temperature of the
solution changes.

Figure 2.9 shows the progression of the area of the ANSA fluorescence spectrum with
the change in solution temperature. Each data point represents an average from a set of
three experiments. Error bars mark +/- one standard deviation. The area of the ANSA
fluorescence spectrum is proportional to the relative total dipole strength, which is
proportional to the fluorescence quantum yield. As the temperature increases from 4 °C
to 90 °C, the area of the ANSA fluorescence spectrum decreases by a factor of 12.

The center frequency of the ANSA fluorescence spectrum as a function of
temperature is plotted in Figure 2.10. Each data point represents an average from a set of
three experiments. Error bars mark +/- one standard deviation. The center frequency of
the ANSA spectrum exhibits a shift to the red a total of 1800 cm", from 20170 cm‘1 at
4 °C to 18370 cm'1 at 90 °C. This 1800 cm'1 shift in center frequency corresponds to a
red shift of 48.6 nm.

The half-width at half-maximum of the ANSA fluorescence spectrum, as shown in
Figure 2.11, increases with the change in solution temperature. From 3207 cm" at 4 °C,
the half-width at half-maximum of the spectrum increases to 3383 cm‘1 at 90 °C. The
half-width of the spectrum in the BLA solution at 90 °C is comparable to that from the

70 percent methanol solution (see Figure 2.7).

26

 

6.0103 . . . , . , . . .

 

 

 

.: I
6 . .
g I
25 I
f5; 40103 - I -
E .
8 - ‘
'- I
,3 2.0-103 - .. -
E
(‘2 :I:
. 2 .
I I I
o l I I I l I r I 4
0 20 40 60 80

Temperature (°C)

Figure 2.9. Evolution of the integrated dipole strength of the AN SA fluorescence
spectrum in a 5:10 uM ANSA:BLA solution as the solution was heated from 4 °C to

90 °C. The average integrated dipole strength from a set of three experiments is plotted.
Error bars mark +/- one standard deviation.

27

 

 

 

 

' II:
TA 2.0-104- I I I
E», I
I
E‘ L
g a:
o-
9
"- I
3;} 1.0104-
c:
8 I
. I
I
I I I I I I I I I
O 20 4O 60 80

Temperature (°C)

Figure 2.10. Evolution of the center frequency of the AN SA fluorescence spectrum from
a 5:10 uM ANSA:BLA solution as a function of temperature from 4 °C to 90 °C. The

average center frequency from a set of three experiments is shown. Error bars mark +/-
one standard deviation.

28

 

3600

3500 1' .-

O
3.31 ,

TI --

Halt Width (cm“)

 

 

 

 

 

 

3200 — I I

I J I I I I

 

 

0 40 60 80

Temperature (°C)
Figure 2.11. Evolution of the half-width at half-maximum of the AN SA fluorescence
spectrum in a 5:10 IIM ANSA:BLA solution upon heating of the solution from 4 °C to
90 °C. The average half-width at half-maximum from a set of three experiments is
shown. Error bars mark +/- one standard deviation.

29

Discussion

The results in this chapter show that it is possible to monitor solvent motions in the
unfolding of a MG through use of a solvatochromic probe. Analysis of the ANSA
fluorescence spectrum from solutions with varying methanol concentrations provides a
characterization of the spectral lineshape with respect to the solvent polarity. The
polarity of the ANSA environment in the BLA solution can be interpreted based on this
initial lineshape characterization.

As the percentage of methanol increases from 0.5 to 100 percent, the area of the
ANSA fluorescence spectrum increases 78-fold, the center frequency blue shifts
2375 cm", and the half-width at half-maximum narrows 470 cm". These changes in
lineshape are consistent with a solvatochromic dye being located in an increasingly less
polar environment”20

A comparison to the results from the methanol experiments provides a rough estimate
of the polarity of the AN SA environment in the BLA solution. Changes in the polarity of
the probe’s environment indicate a flow of solvent, in this case water, into or out of the
hydrophobic core of the protein. As the temperature of the BLA solution increases from
4 °C to 90 °C, the area of the AN SA spectrum decreases by a factor of 12, the center
frequency red shifts 1800 cm", and the half-width at half-maximum broadens by
180 cm“. These parameters indicate that as the temperature increases, the average
polarity of the environment surrounding the probe increases. This data is consistent with

the ANSA binding to the hydrophobic core of the MG at 4 °C and then being exposed to

water as the MG unfolds with the increase in temperature.

30

The center frequency and half-width at half-maximum of the AN SA fluorescence
spectrum in the BLA solution at 4 °C both indicate that the ANSA is initially in a region
that is less polar than the polarity of the 100 percent methanol sample. The center
frequency of the ANSA spectrum in BLA at 50 °C is equivalent to the center frequency
of the spectrum from the 100 percent methanol sample. The half-width at half-maximum
of the ANSA spectrum in the 100 percent methanol sample is equal to the half-width at
half-maximum of the spectrum from BLA at 15 °C. In the BLA solution at 90 °C, the
center frequency of the ANSA spectrum is equal to that from a 60 percent methanol
sample and the half-width at half-maximum is equal to that expected from a solution of
70 percent methanol.

Figure 2.12 compares the 5 uM ANSA dipole-strength fluorescence spectrum from
four solutions. The ANSA fluorescence spectrum in methanol, 5:10 1.1M ANSA:BLA at
4 °C and 90 °C, and pH 2.0 water is shown in the upper panel. The bottom panel shows
an expanded view of the ANSA fluorescence spectrum from the BLA solution at 90 °C
and from the pH 2.0 water. The center frequency of the ANSA fluorescence spectrum is
blue shifted 750 cm‘1 in the ANSA:BLA solution at 4 °C as compared to the spectrum in
the methanol sample. This suggests that at 4 °C, the ANSA in the BLA solution is located
in a region that is less polar than methanol.

At 90 °C, the area of the AN SAzBLA fluorescence spectrum is four times larger than
the fluorescence spectrum in the water sample. The center frequency of the AN SA:BLA
spectrum at 90 °C also shifts approximately 1375 cm'I to the blue of the center frequency
of the ANSA spectrum in water. The parameters of these spectra obtained from fits to

log-normal lineshapes are listed in Table 2.1.

31

 

 

 

 

 

 

 

 

 

  

 

3.0
2.5 -
Methanol
.c 2.0 -
E”
S 9
gig 1.5 - BLA,
2 4 °c BLA.
mg 1.0 _ 90 °C
0 Water
0.5 b /\
0.0 -— ,
I I I I I I
16 0 20.0 24 0
0.2
.C
E» .
g a
3.325 0.1 -
68% .
.9- -
o 0.0 "' 4 l I I

 

 

 

 

I I
16.0 20.0 24.0
Energy (103 cm“)
Figure 2.12. Comparison of the fluorescence dipole-strength spectrum of 5 11M ANSA in
methanol, water and a 5:10 11M AN SAzBLA solution at 4 and 90 °C. The bottom panel

shows an expanded view of the AN SA fluorescence spectrum from the 5:10 uM
ANSA:BLA solution at 90 °C and 5 uM ANSA in pH 2.0 water.

32

Table 2.1. Parameters obtained by fitting the 5 uM AN SA fluorescence spectrum from
methanol, 5:10 11M ANSA:BLA at 4 °C and 90 °C, and water to a log-normal lineshape.
The relative area of the ANSA spectrum from methanol has been set to unity.

 

 

. _, Half-width at
Sample Relative Area Center (cm ) half-maximum (cm")
Methanol 1.0 19200 3310
5:10 uM
ANSA:BLA, 4 °C 0.84 20181 3220
5:10 uM
ANSA:BLA, 90 °C 0.067 18313 3490
Water 0.017 16936 3930

33

Dobson and coworkers report that the process by which apo-BLA, BLA in the
absence of the Ca2+ ion, in the fully unfolded state refolds into a MG can be fit to a

1.14 They studied the fluorescence anisotropy of apo-BLA as it refolds

two-state mode
from the guanidinium chloride-denatured fully unfolded state to the native state (pH 7.0)
and to the MG (pH 2.0) by mixing with a refolding buffer and attribute the burst-phase
anisotropy in the folding to the native state to the formation of a MG. They fit the
anisotropy data from the formation of the MG to a two-state model. Arai and Kuwajima
also attribute the burst-phase intermediate in their stopped-flow far UV CD and
fluorescent measurements to the formation of a kinetic MG during the refolding of
a-lactalbumin from the unfolded state to the native state.15 They report that the only
intermediate in the refolding process is the MG and model the UV CD data of unfolding
of the MG state to a two-state transition. Not only do they look at the kinetic unfolding
transition, they also report the equilibrium and theoretical unfolding transition as being a
two-state transition.

A two-state model suggests that as the MG-state of BLA unfolds, there is a smooth
transition between only two forms, the MG state and the unfolded state. For a two-state
transition, all measures of spectroscopic data should change at the same time in
accordance with the change in position of the two-state equilibrium.3 The mid-points of
the center frequency and half-width at half-maximum of the unfolding transition would
be expected to occur at the same temperature if the transition were really two-state in
character (see Figures 2.5 and 2.6).

The data in this chapter show the mid-point temperature of the center-frequency

transition is approximately 53 °C, whereas that of the half-width at half-maximum

34

transition is roughly 33 °C. This implies that the unfolding of the MG state to the
unfolded state is highly heterogeneous. The ensemble of proteins apparently moves
through a variety of forms, each with a distinct contribution to the ANSA spectrum and
with a distinct temperature, as the unfolding proceeds.

Semisotnov and Dobson have reported that as the acid-denatured MG of
a-lactalbumin is heated, it unfolds to a fully unfolded state.”21 Semisotnov also reports
that the fully unfolded state does not bind ANSA.2l The data presented here, however
(see Figure 2.12 and Table 2.1), suggest there is residual structure in the BLA that
provides a binding site for the AN SA even at 90 °C. Redfield also reports that AN SA in
solution with the chemically-denatured, fully unfolded state of human a-lactalbumin
exhibits a fluorescence spectrum that differs from the dye in free in water.22

If the BLA were fully unfolded to a random-coil state, the dye molecule would be
surrounded by water and have spectral properties very similar to the free dye in water.
Our results indicate that at the end of the thermal unfolding transition, the BLA molecule
retains a folded region that binds ANSA and partially protects it from the bulk aqueous
environment.

The results in this chapter show that the lineshape of the ANSA probe provides a
great deal of information over that provided by the intensity alone. The center frequency
of the spectrum is a rather sensitive probe to polarity. The width of the spectrum may be
useful as a measure of the breadth of the structural ensemble in some cases, but here its

sensitivity to the bonding structure is the main effect noted.

35

References

(1)

(2)

(3)

(4)

(5)

(6)

(7)

(8)

(9)

(10)

(11)

(12)

Voet, D.; Voet, J. G.; Pratt, C. W. Fundamentals of Biochemistry; John Wiley &
Sons, Inc.: New York, 1999.

Ptitsyn, O. “The Molten Globule State.” In Protein Folding; Creighton, T., Ed.;
W. H. Freeman and Company: New York, 1992, p 243-300.

Fersht, A. Structure and Mechanism in Protein Science; W. H. Freeman and
Company: New York, 1999.

Chrysina, E. D.; Brews, K.; Acharya, K. R. “Crystal Structures of Apo- and H010-
bovine a-Lactalbumin at 2.2-A Resolution Reveal an Effect of Calcium on Inter-

lobe Interactions.” J. Biol. Chem. 2000, 275, 37021—37029.

Baum, J .; Dobson, C. M.; Evans, P. A.; Hanley, C. “Characterization of a Partly
Folded Protein by NMR Methods: Studies on the Molten Globule State of the
Guinea Pig a—Lactalbumin.” Biochemistry 1989, 28, 7—13.

Schulman, B. A.; Redfield, C.; Peng, Z.; Dobson, C. M.; Kim, P. 8. “Different
Subdomains are Most Protected from Hydrogen Exchange in the Molten Globule
and Native States of Human a—Lactalbumin.” J. Mol. Biol. 1995, 253, 651-657.

Balbach, J.; Forge, V.; Lau, W. S.; Jones, J. A.; van Nuland, N. A. J.; Dobson, C.
M. “Detection of residue contacts in a protein folding intermediate.” Proc. Natl.
Acad. Sci. USA 1997, 94, 7182—7185.

Forge, V.; Wijesinha, R. T.; Balbach, J.; Brew, K.; Robinson, C. V.; Redfield, C.;
Dobson, C. M. “Rapid Collapse and Slow Structural Reorganisation During the
Refolding of Bovine a-Lactalbumin.” J. Mol. Biol. 1999, 288, 673-688.

Chakraborty, S.; Ittah, V.; Bai, P.; Luo, L.; Haas, E.; Peng, Z. “Structure and
Dynamics of the a—Lactalbumin Molten Globule: Fluorescence Studies Using
Proteins Containing a Single Tryptophan Residue.” Biochemistry 2001, 40,
7228—7238.

Kuwajima, K. “The molten globule state of a-lactalbumin.” FASEB J. 1996, 10,
102-109.

Vendruscolo, M.; Paci, E.; Karplus, M.; Dobson, C. M. “Structures and relative
free energies of partially folded states of proteins.” Proc. Natl. Acad. Sci. USA
2003, 100, 14817.

Semisotnov, G. V.; Kihara, H.; Kotova, N. V.; Kimura, K.; Amemiya, Y.;
Wakabayashi, K.; Serdyuk, I. N.; Timchenko, A. A.; Chiba, K.; Nikaido, K.;
Ikura, T.; Kuwajima, K. “Protein Globularization During Folding. A Study by
Synchrotron Small-angle X-ray Scattering.” J. Mol. Biol. 1996, 262, 559-574.

36

(13)

(14)

(15)

(16)

(17)

(18)

(19)

(20)

(21)

(22)

Rawitch, A. B.; Hwan, R.-Y. “Anilinonaphthalene sulfonate as a probe for the
native structure of bovine alpha lactalbumin: Absence of binding to the native,

monomeric protein.” Biochemical and Biophysical Research Communications
1979, 91, 1383—1389.

Canet, D.; Doering, K.; Dobson, C. M.; Dupont, Y. “Hi gh—Sensitivity
Fluorescence Anisotropy Detection of Protein—Folding Events: Application to
a—Lactalbumin.” Biophys. J. 2001, 80, 1996-2003.

Arai, M.; Kuwajima, K. “Rapid formation of a molten globule intermediate in
refolding of a—lactalbumin.” Folding and Design 1996, 1, 275-287.

Vanderheeren, G.; Hanssens, 1. “Thermal unfolding of bovine or-lactalbumin.”
J. Biol. Chem. 1994, 269, 7090—7094.

Lakowicz, J. R. Principles of Fluorescence Spectroscopy; Second ed.; Kluwer
Academic/Plenum Publishers: New York, 1999.

Cantor, C. R.; Schimmel, P. R. Biophysical Chemistry. Part II: Techniques for the
Study of Biological Structure and Function; W. H. Freeman and Company: San
Francisco, 1980.

Siano, D. B.; Metzler, D. E. “Band shapes of the electronic spectra of complex
molecules.” J. Chem. Phys. 1969, 51, 1856—1861.

Turner, D. C.; Brand, L. “Quantitative Estimation of Protein Binding Site Polariy.
Fluorescence of N—Arylaminonaphthalenesulfonates.” Biochemistry 1968, 7,
3381—3390.

Semisotnov, G. V.; Rodionova, N. A.; Razgulyaev, O. 1.; Uversky, V. N.; Gripas,
A. F.; Gilmanshin, R. I. “Study of the "Molten Globule" Interrneidate Sate in
Protein Folding by a Hydrophobic Fluorescenct Probe.” Biopolymers 1991, 31,
119—128.

Ramboarina, S.; Redfield, C. “Structural Characterisation of the Human
a—Lactalbumin Molten Globule at High Temperature.” J. Mol. Biol. 2003, 330,
1177—1188.

37

CHAPTER 3
STUDY OF THE a-HELIX TO B—SHEE'T TRANSITION OF POLY-L-LYSINE USING

1-ANILINO-8-NAPHTHALENESULFONIC ACID AS A FLUORESCENT PROBE
AND LIGHT SCATTERING

Introduction

In the 1960’s, Davidson and F asman showed that by heating poly-L-lysine (PLL) at
pH 11 to around 50 °C, the initially a-helical form converts quantitatively to the B—sheet
form in a few minutes. 1’2 These early CD measurements suggest that this conformational
transition involves an equilibrium between the a-helix form and a random coil at
temperatures below the temperature required for the B-sheet formation. 1’3 At
temperatures above that required for the B-sheet formation, however, the rate of
formation of the B-sheet is very rapid.4 Witz and Van Duuren then studied the three
forms of PLL, the a-helix form, random coil, and B-sheet form, with a hydrophobic
fluorescence probe, 2-p-toluidiny1naphthalene-6-sulfonate (TNS), which is very similar
in structure to ANSA.5 They conclude that TNS binds preferentially to the B-form of
PLL and does not bind to either the a—helix or random-coil species. Dzwolak and
coworkers have recently used F TIR spectroscopy to detect a transition-state species that
precedes the formation of the B-sheet form.6 They suggest that the structure of the
transition state contains neither a true a-helix nor a pure B-sheet secondary-structural
compliment and that it partially exposes the hydrophobic regions to the bulk solvent.7

We are interested in the transition of PLL from the a-helix to the B—sheet form
because of the relevance of such a transition in the formation of amyloid fibrils that are

associated with diseases. The tendency to form amyloid fibrils is apparently a general

38

property of polypeptides and not necessarily dependent on specific amino-acid sequences
or native—fold topologies.8 Even proteins with globular structures that have primarily
a—helical secondary structure have been shown to form amyloid fibrils.”10 In order for
these amyloid fibrils to form, some of the a helices have to unfold and then refold to
form B-sheet-rich structures.9

In this chapter, we monitor the conformational transition of PLL from the a—helical
form to the B-sheet form with an extrinsic probe, ANSA. The lineshape of the ANSA
fluorescence spectrum has been characterized with respect to polarity in Chapter 2 of this
thesis. We chose to characterize the ANSA fluorescence spectrum in the MG of
a-lactalbumin first because of its well-characterized structure.”"2

As PLL goes through the transition to the B-sheet form, the a-helical form must
somehow first unfold to a less structured form, such as 3 MG. If a MG forms, it will bind
ANSA in its hydrophobic region. By monitoring the AN SA fluorescence lineshape and
the Rayleigh light-scattering peak during the secondary-structural transition, we can
determine the relative rates of change of the spectral parameters and the rate of change of
the molecular weight, respectively. Given our previous knowledge of the spectral
response of AN SA in a well-characterized MG, we can determine if a MG-like state is
formed during the PLL structural transition and what, if any, role this state plays in the
rate of aggregation.

Within our spectral range for data collection, we can monitor both the AN SA
fluorescence spectrum and the Rayleigh scattering peak simultaneously. The molecular

weight of the solute in a polymer solution is related to the intensity of the Rayleigh

scattering according to Equation 3.1.13

39

- 2
l_9:(I+C(2)S mKCMw (3.1)
o r

N

In Equation 3.1, i, is the intensity of the Rayleigh scattering from the sample, 10 is the
intensity of the incident radiation, r is the distance from the sample to the detector, 0 is
the angle of the data collection from the incident beam, C is the concentration of the
solute, M.; is the average molecular weight of the solute, and K is a constant for a given
solution of macromolecules that depends on the refractive index and the monitoring
wavelength.

Since we do not change the concentration of the solute during the course of the
experiment, and since K is a constant for the solution at a specific wavelength, a change
in the intensity of the Rayleigh-scattering peak can be attributed to a change in the
molecular weight of the solute. This allows us to monitor the rate of aggregation of the
peptide simultaneously with the measurement of the fluorescence spectrum of ANSA,

which gives us information on the polarity of the environment that binds the probe.

Sample Preparation

ANSA (l-anilino—8-naphthalenesulfonic acid ammonium salt) and PLL
(poly-L-lysine; molecular weight 30,000 — 70,000) were obtained from Sigma and used
as received. Samples of 7:14 11M ANSA:PLL were dissolved in water to obtain an
absorbance less than 0.05 at 350 nm with a 1-cm path length. The aqueous PLL samples
were brought to pH 11.0 with NaOH. Prior to collection of fluorescence spectra, samples
were passed through 0.22 um filters (Millipore, Millex-GV). All fluorescence spectra
were collected on the spectrofluorimeter described in Chapter 2. The conversion of the

a-helical form to the B-sheet form was induced by increasing the temperature of the

40

solution from 23 °C to 55 °C using the temperature-controlled sample holder. These

sample conditions were chosen based on previous work by Davidson and Fasman.l

Experimental

For each experiment, spectra were collected at pre-set time intervals over a minimum
of two hours. Before the temperature change, the fluorescence spectra were collected by
exposing the CCD for 10 accumulations of a 20-s integration time every 240 3. After the
temperature change, thirty spectra were collected with 10 accumulations of a 10-s
integration at 120 3 intervals for the first 3600 3. Once the rate of change in the AN SA
fluorescence spectrum began to decrease, the number of accumulations averaged for each
spectrum was increased to 20 and the time interval between collections was increased to
900 3. Each spectrtun was normalized by the exposure time of the CCD to compensate

for the different exposure times over the course of the experiment.

Results

The temperature profile for a water-filled cuvette in the same experimental setup as
described above is shown in Figure 3.1. The temperature on the controller was changed
from 23 °C to 55 °C at the plotted zero of time. Within 330 s of changing the
temperature on the controller, the solution had reached 50 °C. The temperature of the

solution equilibrated to 55 °C after 540 s. Stirring was used to prevent the formation of

temperature gradients within the sample.

41

+++++++++++++

 

 

+ + + +
+
50— +
+
+
45fi +
Q +
2 40'-
: +
E
i
+9 35- +
30- +
25- +
+
+
I I I I I
0 200 400 600 800
Time (s)

Figure 3.1. Time evolution of the temperature of the sample solution as measured by a
thermocouple immersed in a blank sample of water of the same volume as used in the
fluorescence experiments. The temperature controller of the sample holder was changed
from 23 °C to 55 °C at the plotted zero of time. The half-time for the transition is 140 s.

42

After establishing the temperature profile associated with our experimental setup and
conditions, we collected a series of time-resolved fluorescence spectra from solutions of
7: 14 11M AN SAzPLL . Figure 3.2 shows the fluorescence spectrum collected from a
7:14 [J.M AN SAzPLL solution at 0 s, 580 s, and 11180 s, respectively, after the sample
temperature was changed from 23 °C to 55 °C. In each spectrum, the AN SA
fluorescence spectrum and scattered-light peak have been superimposed with a fit to a
log-normall4 and a Lorentzian lineshape, respectively. In the spectrum at 23 °C, prior to
the change in temperature, the scattered-light peak from the 380-nm excitation source is
relatively small and centered around 26600 cm". The ANSA fluorescence spectrum,
centered around 19000 cm", is broad and relatively weak. Following the change in
temperature, the relative dipole strength of the scattered-light peak increases and the
AN SA fluorescence spectrum increases in dipole strength and shifts to the blue.

Expanded views of the scattered-light peak and the AN SA fluorescence spectrum, are
shown in Figures 3.3 and 3.4, respectively. Figure 3.3 shows the relative dipole strength
of the scattered-light peak continues to increase with time following the change in
temperature without a change in lineshape. The AN SA fluorescence spectrum in
Figure 3.4 reports an increase in intensity and a blue shifi after the temperature change.
Unlike the scattered-light peak, the AN SA fluorescence spectrum does not continue to
change dramatically after 580 s. As will be shown later, the integrated dipole strength of
the AN SA fluorescence spectrum levels off after the fast, initial change with the increase
in temperature.

The time evolution of the average area of the scattered-light peak from a set of six

experiments is plotted in Figure 3.5. Error bars mark +/- one standard deviation. The

43

 

 

 

 

 

 

 

 

0.15 y I I I 1 I
l- .
0.10)“ -I
a I- a
dig 0.05- .
£037) 000. d
613%- . - -
5 -0.05- _
—0.10- i ..
n I I I I I l "
' I ' I ' l ' III
II
t=580$ II
1.00- 'I ..
I
I
.c . I .
E” I
Q .
22 _ I
Ei _ ’I’ .
‘5 7 . ’
0.00- -- -— _
I I I I I I
I ' I I I I I n‘
300- t=111803 I. -
. I .
I
5 2.00- I -
.‘9’ I
as ,1 .
:32 1.00_ II 3
8' .
'5 ’ ‘
0.00- —- - —- ——— -
I 4 I I I

 

I I
16.0 20.0 24.0

Energy (103 cm“)

Figure 3.2. Fluorescence dipole-strength (F/ V") spectra from a 7: 14 11M ANSA:PLL
solution as a function of time following a change in temperature, from 23 °C to 55 °C.
The scattered-light peak, centered at 26600 cm“, is superimposed with a fit to a
Lorentzian lineshape (short dashes). The ANSA fluorescence spectrum, centered at
19000 cm“, is superimposed with a fit to a log-normal lineshape (long dashes). The
indicated time is that following the sample temperature change. Note that the ordinate is
autoscaled for each panel.

 

   

 

 

 

 

I I j I I I fir
..I"\
3.00 - ./ s.-
/
.C
a - / ‘
E t = 11180 s//
"’ 2.00 - _
a)
.02 ./
.3 .
o
g 1.00 /'\-
E
(D 4
m t=0s
0.00 -
I I 4 l I L I
25.0 25.5 26.0 26.5

Energy (103 cm")

Figure 3.3. Expanded view of the time evolution of the 380 nm scattered-light peak
detected in the 7: 14 [1M ANSA:PLL solution. The indicated time is that following the
sample temperature change from 23 °C to 55 °C. The dashed lines represent the fit to a
Lorentzian lineshape.

45

 

    

0.20- , -
t=11180$ .

   

 

 

.C
4.0 of
U) o,
C ..7. ’. .
g P .6214”
(D
.2
.3 0.10—
0 . 5'
Q g.”
.2 It;
as - .
'5 1., ‘
m K: I
0.00 ..
I I I I I

 

 

I
16.0 18.0 20.0 22.0
Energy (103 cm‘1)

Figure 3.4. Expanded view of the time evolution of the AN SA fluorescence spectrum
detected in the 7: 14 uM ANSA:PLL solution. The indicated time is that following the
sample temperature change from 23 °C to 55 °C. The dashed lines represent the fit to a

log-normal lineshape.”

scattered-light peak in each spectrum was fit to a Lorentzian lineshape. From the fit to the
Lorentzian lineshape, the area of the scattered-light peak was determined. The area of the
scattered light peak at the zero of time was subtracted from the plotted areas. In Figures
3.5-3.8, the zero of time represents the time at which the temperature on the controller
was changed from 23 °C to 55 °C. Note that the area of the scattered—light peak does
not change immediately with the change in temperature. There is a 340 s delay after the
temperature change before the area of the scattered—light peak changes significantly. The
time constant associated with the initial exponential rise is approximately 570 s. The
intensity of the scattered-light peak increases by a factor of 27, indicating the average
molecular weight of the PLL increases from the initial 52 kDa to approximately

1400 kDa.

The fit of the AN SA fluorescence spectrum to a log—normal lineshape gives the total
dipole strength (area), center frequency, and half—width at half—maximum of the AN SA
fluorescence spectrum. The time evolution of these parameters are plotted in
Figures 3.6-3.8. Each point on the time evolution plot represents the average value from
a set of six experiments. Error bars mark +/— one standard deviation.

The time evolution of the total integrated dipole strength of the ANSA fluorescence
spectrum in a 7: 14 uM ANSA:PLL solution is shown in Figure 3.6. In response to the
increase in temperature to 55 °C, the area of the ANSA fluorescence spectrum increases.
The integrated area is proportional to the total dipole strength, which is proportional to
the relative quantum yield.'5 The increase in area of the ANSA fluorescence spectrum is
indicative of the ANSA probe being in an environment that is less polar. There appear to

be at least two components in the time evolution of the area of the ANSA fluorescence

47

 

 

 

 

 

 

 

 

4.00 -
.c 0
I? 3.00 - . O
9 .
(75 .L
{i 2.00 -
a
o I
.g
% 1.00 -
(I
P I
0.00 - ! I . ’
I I I I I L I L
0 2000 4000 6000 8000

Time (s)

Figure 3.5. Time evolution of the area (integrated dipole strength) of the scattered-light
peak observed in a 7: 14 [4M ANSA:PLL solution. The average area from a set of six
experiments is shown. Error bars mark +/- one standard deviation. The area of the
scattered light peak at the plotted zero of time (the time at which the temperature change
from 23 °C to 55 °C occurred) was subtracted from the plotted areas.

 

 

 

 

 

700- - W. I U T -.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

g, 4?
E 600 _ . 'W i
(I)
2 J. -*- — J
.3 500 ' ‘-
0 P L
35'!
8
é) 400 -
g
0)
CE
300 f
200 I I I I I I I I
0 2000 4000 6000 8000

Time (s)

Figure 3.6. Time evolution of the integrated dipole strength of the ANSA fluorescence
spectrum in a 7: l4 uM AN SAzPLL solution. The average integrated dipole strength from

a set of six experiments is plotted. Error bars mark +/- one standard deviation. The
sample temperature was changed from 23 °C to 55 °C at the plotted zero of time.

49

spectrum. The initial, rapid component shows an exponential rise that is completed after
approximately 1500 s. A second, slower component accompanies a decrease in the area
of the ANSA fluorescence spectrum.

The time evolution of the average center frequency and the half-width at
half-maximum of the AN SA fluorescence spectrum in the sample described above are
shown in Figures 3.7 and 3.8, respectively. The responses of the center frequency and
the half-width at half-maximum of the ANSA fluorescence spectrum to the temperature
change are consistent with the ANSA probe being in a less polar environment after the
increase in temperature. As shown in Figure 3.7, the center frequency of the AN SA
fluorescence spectrum to shifts to the blue by 1033 cm'1 (32 nm) within 500 s of the
temperature change. Following the initial rise, the average center frequency of the AN SA
fluorescence spectrum continues a slight shift to the blue over the length of the
experiment.

Figure 3.8 shows that the time evolution of the half-width at half-maximum of the
AN SA fluorescence spectrum shows a rapid narrowing that is completed approximately
1000 s after the temperature change. Following the initial component, the lineshape
continues a slow narrowing as time progresses. The presence of two phases in all of the
parameters associated with the AN SA fluorescence spectrum suggests that there is an
initial dramatic change in the structure of the PLL following the change in temperature
and then a slower, more subtle change in structure occurs as the molecular weight of the

PLL increases.

50

 

    
  

 
    
 

 

'.""..vvvvvv

'0' AAA.
'I""x IIIAIIAI““
.“.“ “

  

 

  

1.85-104

 

1.80104

Center Frequency (cm“)

 

Hi

I L I I I I I I

0 2000 4000 6000 8000
Time (s)

 

 

 

Figure 3.7. Time evolution of the average center frequency of the ANSA fluorescence
spectrum observed in a 7: 14 uM ANSA:PLL solution. The average center frequency

from a set of six experiments is shown with error bars marking +/— one standard
deviation. The sample temperature was changed from 23 °C to 55 °C at the plotted zero

of time.

51

 

 

 

 

 

 

5000 -
r
T ‘9
E 4500 -
.C
3*: .
3
:5 4000 — - _
3500 — f 1 I
I l I l l l I
0 2000 4000 6000 8000
Time (s)

Figure 3.8. Time evolution of the half-width at half-maximum of the ANSA
fluorescence spectrum in a 7: 14 uM ANSA:PLL solution. The average half-width at
half-maximum from a set of six experiments is shown. Error bars mark +/- one standard
deviation. The sample temperature was changed from 23 °C to 55 °C at the plotted zero

of time.

52

Discussion

With the change in temperature from 23 °C to 55 °C, the relative dipole strength of
the scattered-light peak from the AN SAzPLL solution shows a 340-3 delay in response, a
rapid increase that is completed within the first 2000 s, and then a slower, continual
increase with time. The concentration of the sample is constant throughout the
experiment, so the increase in the Rayleigh scattering can be attributed to the increase in
average molecular weight of the solute (see Equation 3.1).

The area, center frequency, and half-width at half-maximum of the AN SA
fluorescence spectrum show a delay of approximately 220 s that precedes the initial rapid
response to the increase in temperature. The responses of these parameters appear to be
associated with at least two time constants, a fast initial response and a second, much
slower response. The area of the fluorescence spectrum doubles within 2200 s after the
increase in temperature and then shows a slight decrease with time. Over the time of the
data collection, the center frequency of the spectrum shifts 1240 cm'1 to the blue and the
half-width at half-maximum decreases by 925 cm’1 after the initial 220 s delay in
response.

The experimental setup used in this thesis has an inherent time lag associated with the
change in sample temperature. The 220-s delay in the response of the AN SA
fluorescence spectrum most likely arises from the sample being at a temperature below
that required for the B«sheet formation due to this time lag. As shown in Figure 3.1,

220 s after the temperature change on the controller, the temperature of the solution is
approximately 45 °C, which should be a high enough temperature to allow formation of

B—sheet under our conditions.I

S3

Figure 3.9 shows the relative rates of the spectral responses from the AN SAzPLL
solution shown in Figures 3.5—3.8. The relative rates of change in order from fastest to
slowest are as follows: center frequency, half-width at half-maximum, area, and light
scattering. Snell and Fasman report a rate of B—sheet formation at 55 °C that has an
exponential time constant of approximately 50 s, but the differences in our experimental
setups do not allow direct comparison of these data with ours.4 It is likely that once our
system has reached 55 °C, the formation of the B-sheet occurs on the same time scale.

The center frequency of the ANSA spectrum is the parameter that is most sensitive to
the polarity of the environment. Its relatively fast rate of change suggests that a species
with hydrophobic binding sites for the probe is formed rapidly upon heating. This
species, which binds AN SA, can be considered MG-like.

Dzwolak and coworkers have suggested a transition-state species that is less
structurally defined than either a pure a-helix or a pure B-sheet and has hydrophobic
regions exposed . They suggest that the formation of a “loose” helix would allow
exposure of hydrophobic binding sites for AN SA.7 Our data is in agreement with the
formation of a species that binds AN SA prior to the formation of aggregates. It is
difficult, however, to say whether we can differentiate between this MG-like state with
“loose” secondary structure and the pure B-sheet form of PLL prior to aggregation.

Using the relative rates of change presented in Figure 3.9, we propose the following

model for the sequence of events in the a—fi transition:

a-—>MG—>(MG,,—>[3,

ggregate )

54

 

 

0'.
o
1.0a .
. x
>‘(xxxxx xxxxxxxxxxxxxxxxxxx x
0.8— x t +
++fiifffti ‘
x ‘A‘AA‘ A:A“:#++*
0.6— +
'6 ‘ +++
a +
'(T) : +
< 0.4— +
+
A
0.2-4 A
A 8+
_ is
x
x o
o
l l l I I
Time(s)

Figure 3.9. Comparison of the time evolution of the ANSA fluorescence spectrum (area
(0), center frequency (X), half-width at half-maximum (A)) and the scattered-light peak
(+) from the 7: l4 uM ANSA:PLL solution following a 23 °C to 55 °C temperature
change. The signals have been normalized to zero at toand to unity at t = 7250 s. The
ANSA half-width at half-maximum response has been inverted.

55

Parentheses are placed around the MGn to Baggngm transition because we are not able to
differentiate the time ordering of the formation of the aggregates from the formation of
the B-sheet from our results. In the following, we discuss how the proposed mechanism is
suggested by the experimental observations.

A comparison of the change in center frequency of the ANSA upon binding to the
PLL with the change in the center frequency of the MG state of BLA supports the idea of
a MG-like intermediate in PLL. The change in center frequency of AN SA during the
thermal unfolding of the MG of BLA is 1800 cm’1 (48 nm, see Chapter 2) and the change
in center frequency during the a—B transition in PLL is 1371 cm’1 (42 nm). The absolute
values of the center frequencies, however, suggest that the environment of the ANSA in
the PLL is more polar than the MG of BLA. In the PLL solution, the AN SA goes from
being essentially free in water to an environment that is partially sheltered from the
water. There is only an increase by a factor of two in the area of the AN SA fluorescence
spectrum in PLL, whereas there is more than a ten-fold increase in the area of the AN SA
fluorescence spectrum in the MG BLA. The relatively small increase in area in the PLL
solution is likely to involve the partially-exposed nature of the binding site for AN SA in
PLL. The decrease in the area over long time accompanies the formation of larger and
larger aggregates. As these aggregates form, the ANSA-binding site is gradually
increasingly exposed to solvent. This indicates an evolution of the initial aggregate to a
different structural form.

Figure 3.8 shows that as the temperature is increased, the width of the AN SA
spectrum decreases. A decrease in the width of the ANSA spectrum indicates that the

range of structures in the ensemble is decreasing. As time progresses, the ANSA

56

spectrum continues to narrow as the aggregation (see Figure 3.5) of PLL increases. This
suggests that a more ordered form accumulates as the aggregation continues.

The rate of change of the scattered-light peak is the slowest of the four parameters
analyzed. This suggests that before aggregation can occur, the B-rich state of PLL must
form. Experiments similar to the one reported here only with different concentrations of
PLL suggest that the rates of change do not change significantly with a change in
concentration of peptide (data not shown). This suggests that the initial formation of the
B—sheet-rich PLL is intramolecular.

All of these changes in spectral parameters are consistent with the mechanism
proposed above. In the transition from a-helix to B—sheet, the rate of formation of the
fl-sheet is dependent on the initial formation of the MG-like state. Note that formation of
aggregates following a secondary structural transition of a protein is reminiscent of the
prion and prion-related diseases. Prions are proteins that undergo conformational changes
that result in accumulation of aggregate forms that are cytotoxic.16 It may be possible that
the conformational transition of PLL could be furthered considered as a model system for

understanding the mechanism by which prions aggregate.

57

References

(1)

(2)

(3)

(4)

(5)

(6)

(7)

(8)

(9)

(10)

(11)

Davidson, B.; Fasman, G. D. “The Conformational Transitions of Unchargcd
Poly-L-lysine. 0t Helix-Random Coil-B Structure.” Biochemistry 1967, 6,

1616—1629.

Davidson, B.; Fasman, G. D. “The Single-Stranded Polyadenylic
Acid-Poly-L-lysine Complex. A Conformational Study and Characterization.”
Biochemistry 1969, 8, 4116—4126.

Wooley, S.-Y. C.; Holzwarth, G. “Intramolecular B-PIeated-Sheet Formation by
Poly-L-lysine in Solution.” Biochemistry 1970, 9, 3604—3608.

Snell, C. R.; Fasman, G. D. “Kinetics and Thermodynamics of the or Helix-[3
Transconformation of Poly(L-lysine) and L-Leucine Copolymers. A
Compensation Phenomenon.” Biochemistry 1973, 12, 1017—1025.

Witz, G.; Van Duuren, B. L. “Hydrophobic Fluorescence Probe Studes with
Poly-L-lysine.” J. Phys. Chem. 1973, 77, 648—651.

Dzwolak, W.; Muraki, T.; Kato, M.; Taniguchi, Y. “Chain-length dependence of
alpha-helix to beta-sheet transition in polylysine: Model of protein aggregation
studied by temperature-tuned Ff IR spectroscopy.” Biopolymers 2004, 73,
463-469.

Dzwolak, W.; Smirnovas, V. “A conformational a-helix to B-sheet transition
accompanies racemic self-assembly of polylysine: an FT-IR spectroscopic study.
Biophys. Chem. 2005, 115, 49-54.

’9

Chiti, F.; Webster, P.; Taddei, N .; Clark, A.; Stefani, M.; Ramponi, G.; Dobson,
C. M. “Designing conditions for in vitro formation of amyloid protofilaments and
fibrils.” Proc. Natl. Acad. Sci. USA 1999, 96, 3590—3594.

Pertinhez, T. A.; Bouchard, M. L.; Tomlinson, E. J.; Wain, R.; Ferguson, 8. J.;
Dobson, C. M.; Smith, L. J. “Amyloid fibril formation by a helical cytochrome.”
Febs Lett. 2001, 495, 184—186.

Fandrich, M.; Forge, V.; Buder, K.; Kittler, M.; Dobson, C. M.; Diekmann, S.
“Myoglobin forms amyloid fibrils by association of unfolded polypeptide
segments.” Proc. Natl. Acad. Sci. USA 2003, 100, 15463—15468.

Baum, J .; Dobson, C. M.; Evans, P. A.; Hanley, C. “Characterization of a Partly
Folded Protein by NMR Methods: Studies on the Molten Globule State of the
Guinea Pig a-Lactalbumin.” Biochemistry 1989, 28, 7—13.

58

 

(12)

(13)

(14)

(15)

(16)

Schulman, B. A.; Redfield, C.; Peng, Z.; Dobson, C. M.; Kim, P. S. “Different
Subdomains are Most Protected from Hydrogen Exchange in the Molten Globule
and Native States of Human a—Lactalbumin.” J. Mol. Biol. 1995, 253, 651—657.

van Holde, K. E.; Johnson, W. C.; Ho, P. S. Principles of Physical Biochemistry;
Second ed.; Pearson Education, Inc: Upper Saddle River, NJ, 2006.

Siano, D. B.; Metzler, D. E. “Band shapes of the electronic spectra of complex
molecules.” Journal of Chemical Physics 1969, 51, 1856—1861.

Cantor, C. R.; Schimmel, P. R. Biophysical Chemistry. Part II: Techniques for the
Study of Biological Structure and Function; W. H. Freeman and Company: San
Francisco, 1980.

Prusiner, S. B. “Prions.” Proc. Natl. Acad. Sci. USA 1998, 95, 13363—13383.

59