Citation
Binding studies of c-reactive protein to lipid-coated gold nanoparticles and isoform differentiation by quenching experiments

Material Information

Title:
Binding studies of c-reactive protein to lipid-coated gold nanoparticles and isoform differentiation by quenching experiments
Creator:
Messersmith, Reid Elliot
Publication Date:
Language:
English
Physical Description:
1 electronic file : ;

Subjects

Subjects / Keywords:
Cooperative binding (Biochemistry) ( lcsh )
C-reactive protein ( lcsh )
C-reactive protein ( fast )
Cooperative binding (Biochemistry) ( fast )
Genre:
bibliography ( marcgt )
theses ( marcgt )
non-fiction ( marcgt )

Notes

Bibliography:
Includes bibliographical references.
Thesis:
Department of Chemistry
Statement of Responsibility:
by Reid Elliott Messersmith.

Record Information

Source Institution:
|University of Colorado Denver
Holding Location:
|Auraria Library
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
858281362 ( OCLC )
ocn858281362

Auraria Membership

Aggregations:
Auraria Library
University of Colorado Denver

Downloads

This item has the following downloads:


Full Text
BINDING STUDIES OF C-REACTIVE PROTEIN TO LIPID-COATED GOLD
NANOPARTICLES AND ISOFORM DIFFERENTIATION BY QUENCHING
EXPERIMENTS
by
Reid Elliott Messersmith
B.S. Chemistry, Pepperdine University, 2010
A thesis submitted to the
University of Colorado Denver
in partial fulfillment of the
requirements for the degree of
Master of Science
Chemistry
2012


This thesis for the Master of Science
degree by
Reid Elliott Messersmith
has been approved for
by
Scott Reed
Mark Anderson
Jefferson Knight
Date: April 5th, 2012
11


Messersmith, Reid, Elliott (M.S., Chemistry)
Binding Studies Of C-Reactive Protein To Lipid-Coated Gold Nanoparticles And Isoform
Differentiation By Quenching Experiments
Thesis directed by Professor Scott Reed.
ABSTRACT
C-reactive protein (CRP) is an acute phase protein that has been implicated in
cardiovascular disease. CRP exists in serum in a pentameric form, but upon binding to
phosphatidylcholine on damaged cell membranes it dissociates into a biologically active
form: modified CRP. Lipid-coated gold nanoparticles were synthesized as membrane
mimics so the effects of membrane structure on CRP binding could be observed. Stable
mimics were created with propanethiol anchors and a mixed bilayer containing
phosphatidylcholine and cholesterol surrounding the gold nanoparticle. (3-
mercaptoethanol was utilized to separate the hybrid lipid bilayer with the bound CRP
from the gold nanoparticles allowing for analysis of the CRP tryptophan
fluorescence. Conformational changes in CRP on binding to PC-AuNP were studied by
quenching the intrinsic tryptophan fluorescence of the protein with succinimide and gel
assays.
This abstract accurately represents the contents of the candidates thesis.
I recommend its publication.
Approved: Scott Reed


ACKNOWLEDGMENTS
I would like to thank all the members of the Reed lab for their help and support.
In particular, I appreciate Scott Reed for allowing me to work in his lab and his vision for
this project. Min Wang for her laboratory expertise and help as a knowledgeable
resource. Audrea Piper for being an accommodating lab mate and continued assistance
throughout the project. Additionally, I would like to acknowledge my friends and family
for their support during this journey.
IV


TABLE OF CONTENTS
LIST OF TABLES..........................................................viii
LIST OF FIGURES...........................................................ix
LIST OF SCHEMES..........................................................xii
LIST OF ABBREVIATIONS...................................................xiii
CHAPTER
1. INTRODUCTION............................................................1
Introduction to C-reactive protein.........................................1
CRP structure.......................................................2
Ligands of CRP......................................................3
Function of C-reactive protein......................................4
Cardiovascular disease..............................................6
Introduction to lipid coated gold nanoparticles............................8
AuNP and Thiols.....................................................9
Biological Membranes and Supported Surfaces........................11
Lipid-coated Nanoparticles.........................................13
Lipid coated AuNP..................................................14
Introduction to Fluorescence..............................................16
Fluorescence Quenching.............................................17
Tryptophan Fluorescence and Quenching..............................19
Binding studies of CRP to PT-PC-AuNP......................................21
2. EXPERIMENTAL...........................................................23
v


Materials....................................................................23
Gold nanoparticle preparation................................................23
Lipid-coated, thiol-anchored AuNP synthesis..................................23
CRP binding studies with PT-PC-AuNP..........................................24
Monomeric CRP preparation....................................................25
Gel electrophoresis..........................................................25
Fluorescence analysis........................................................26
Fluorescence quenching studies...............................................26
Cyanide stability studies....................................................27
3. RESULTS..........................................................................28
Formation of PT-PC-AuNP..............................................................28
AuNP characterization........................................................28
Maximizing ion impermeability of PT-PC-AuNP..................................28
Binding studies of CRP with PT-PC-AuNP.......................................31
Lipid disruption utilizing BME or DTT........................................32
Quenching Experiments................................................................34
Quenching in buffer by succinimide and acrylamide............................34
Succinimide quenching in PT-PC-AuNP background...............................36
Binding Studies......................................................................37
Utilizing BME to disrupt PT-PC-AuNP..........................................37
Sequestering of BME by NMM...................................................39
4. Discussion.......................................................................41
PT-PC-AuNP preparation...............................................................41
AuNP size and shape determination............................................41
vi


Oleate concentration
42
Cholesterol concentration..............................................43
Thiol concentration....................................................44
Binding studies of pCRP with PT-PC-AuNP.......................................45
Quenching issues and BME...............................................45
Quenching with succinimide and acrylamide in buffer....................46
Succinimide quenching with PT-PC-AuNP background.......................47
BME and NMM............................................................48
5. Conclusions................................................................50
REFERENCES....................................................................52
vii


LIST OF TABLES
Table 1. Summarizes succinimide fluorescence quenching data. Buffer background
refers to quenching fluorescence in 10 mM phosphate buffer, 140 mM NaCl and 2.5 mM
CaCh. PT-PC-AuNP supernatant background refers to making PT-PC-AuNP,
centrifuging at 15,000 rpm for 20 minutes, and adding mCRP, pCRP and tryptophan to
the supernatant..................................................................34
Table 2. Succinimide quenching data of binding studies samples containing CRP
incubation with PT-PC-AuNP and BME. The 1:1 BME:NMM also contains 1 mM
NMM and all samples were centrifuged before quenching analysis...................37
vm


LIST OF FIGURES
Figure 1. (3-face of c-reactive protein viewed down the five-fold symmetry axis. The PC
binding site is located near the calcium ions (green) and the pocket for the choline
group.1,2..........................................................................1
Figure 2. a-face of c-reactive protein viewed down the five-fold symmetry axis. Clq
binding site located in cleft near a-helix (green).1,2.............................2
Figure 3. Fluorescent tryptophan residues in c-reactive protein shown in green. There
are 6 tryptophan residues per monomer, 30 tryptophan residues total.1,2...........16
Figure 4. The absorbance retained after 24 hours of incubation with cyanide compared
with the concentration of oleate used to prevent aggregation. The error bars are the
standard deviation of at least 3 trials..........................................28
Figure 5. The absorbance retained after 24 hour of incubation with cyanide plotted
against the percentage of lipid layer that was cholesterol. The remainder of the lipid layer
was always PC and the error bars are the standard deviation of at least 3 trials.29
Figure 6. The amount of nanomoles of propanethiol injected into a 1 mL solution of PC-
AuNP compared with absorbance retained after 24 hours of incubation with cyanide. The
error bars are the standard deviation of 10 trials...............................30
Figure 7. Gel electrophoresis and Western blotting were preformed on the supernatant
(S) and suspended pellet (P) of samples that did and did not contain EDTA. Controls of
pCRP and mCRP were run for comparison............................................31
Figure 8. All samples contained PT-PC-AuNP and cyanide. One sample contained 1
mM BME (), another 1 mM DTT () and the last did not contain any other reactants
(). The error bars are the standard deviation of 10 trials......................32
Figure 9. All samples contained HT-PC-AuNP and cyanide. One sample contained 1
mM BME (), another 1 mM DTT () and the last did not contain any other reactants
(). The error bars are the standard deviation of 7 trials.......................32
Figure 10. Gel electrophoresis and Western blotting of samples containing BME and
pCRP. DTT and pCRP. pCRP and mCRP................................................33
Figure 11. Stern-Volmer plot of pCRP (), mCRP () and tryptophan () quenched by
succinimide in 10 mM phosphate buffer, 140 mM NaCl and 2.5 mM CaCU The
equation of the best-fit line for tryp: y = 0.0135x + 1.010; mCRP: y = 0.0088x + 1.044;
pCRP: y = 0.0031 + 0.975......................................................... 34
Figure 12. Stern-Volmer plot of pCRP (), mCRP () and tryptophan () quenched by
acrylamide in 10 mM phosphate buffer, 140 mM NaCl and 2.5 mM CaCh. The equation
IX


of the best-fit line for tryp: y = 0.0264x + 0.0660; mCRP: y = O.OOlOx + 1.044; pCRP: y
= 0.0052 + 0.926.................................................................... 35
Figure 13. Stern-Volmer plot of pCRP (), mCRP () and tryptophan () quenched by
succinimide in the supernatant of PT-PC-AuNP. The equation of the best-fit line for
tryp: y = 0.0078x + 1.056; mCRP: y = 0.0029x + 1.057; pCRP: y = 0.0011 + 1.011......36
Figure 14. Stern-Volmer plots of the succinimide quenching of the supernatants of the
binding studies. One sample contained calcium () and isoform conversion occurred (y
= 0.0012x + 1.025). The other sample contained EDTA () and binding did not occur (y
0.005\ 1.0103).....................................................................37
Figure 15. Western blot of supernatants of samples where CRP was incubated with PT-
PC-AuNP and then BME before centrifugation...........................................38
Figure 16. Western blot of binding studies of CRP with PT-PC-AuNP, after incubation
with 1 mM BME, 23 mM NMM and centrifugation. The supernatant (S) and pellet (P) of
samples containing EDTA and not containing EDTA are shown...........................39
Figure 17. Adapted from Basu and co-workers. >wx is plotted against the diameter,
determined by TEM, of AuNP also synthesized by Frens method (x). The average >wx
for the AuNP created by this analyst (O) is plotted to show the expected corresponding
size.................................................................................41
x


LIST OF EQUATIONS
Equation 1. Oxidation of gold(O) to gold(l) by cyanide...............................10
Equation 2. Stern-Volmer equation where F0 is the initial fluorescence, F is the
fluorescence with a certain concentration of quencher [Q] and KSv is the quenching
constant..............................................................................17
Equation 3. The effect of efficiency (y) on the value of the KSv.....................18
Equation 4. Correlation between extinction coefficient (e) and AuNP diameter (d) with
two constants k and a.................................................................41
xi


LIST OF SCHEMES
Scheme 1. Displacement of thiol coated gold nanoparticles by a shorter thiol........10
Scheme 2. One possible mechanism for lipid deposition onto planar surfaces..........11
Scheme 3. Oleate coated gold nanoparticles are incubated with PC liposomes. The thiol
anchor is added to stabilize the hybrid lipid bilayer against cyanide penetration....14
Scheme 4. Propanethiol stabilized PC-AuNP become completely cyanide instable after 2
hours when mixed with BME............................................................22
Scheme 5. Isoform conversion as a result of calcium dependent binding of c-reactive
protein to PT-PC-AuNP. Removal of the hybrid lipid bilayer by BME for fluorescence
quenching analysis...................................................................22
Xll


LIST OF ABBREVIATIONS
CRP C-reactive protein
pCRP Pentameric c-reactive protein
mCRP Monomeric c-reactive protein
AuNP Gold nanoparticles
PC Phosphatidylcholine
PC-AuNP Lipid coated gold nanoparticles
PT-PC-AuNP Propanethiol stabilized lipid coated gold nanoparticles
HT-PC-AuNP Hexanethiol stabilized lipid coated gold nanoparticles
BME (3 mercaptoethanol
DTT Dithiothreitol
LDL Low density lipoproteins
EDTA Ethylenediaminetetraacetic acid
SDS Sodium dodecyl sulfate
LSPR Localized surface plasmon resonance
DLS Dynamic light scattering
Ksv Stern-Volmer quenching constant
Xlll


CHAPTER 1. INTRODUCTION
Introduction to C-reactive protein
C-reactive protein (CRP) was discovered in 1930 when it precipitated with
Streptococcus pneumoniae,3 4 The unknown substance was labeled Fraction C because
of its reactivity to c-polysaccharide, which is what led to the naming of the protein.5 CRP
is a prototypic acute phase protein that increases in concentration due to a number of
stimuli and is involved with the innate immune system.6 CRP has a low basal level (1
mg/L) but can increase 500, 1,000, or even 10,000 times in 48 hours in response to
infection, inflammation or tissue damage.57 High baseline levels of CRP indicate
systemic inflammation and when linked
with cholesterol concentrations can predict
future cardiovascular events.8 Ridker and
co-workers showed that the most potent
predictor of future cardiovascular events is
CRP levels independent of cholesterol
. Q- If)
concentration.
CRP is produced mainly by
Figure 1. (3-face of c-reactive protein
viewed down the five-fold symmetry
axis. The PC binding site is located near
the calcium ions (green) and the pocket
for the choline group.12
hepatocytes in the liver, more specifically in the endoplasmic reticulum and is
transcriptionally controlled by cytokine interleukin-6. CRP has no known deficiencies or
polymorphism in humans and has high phylogenetic conservation.
1


CRP structure
CRP is a member of the pentraxin family and is composed of five identical, non-
covalently linked subunits totaling 115,135 Da.1 Each protomer or monomer has a
molecular weight of 23,027 Da and consists of 206 amino acids composed in two
antiparallel P sheets with flattened jellyroll topology.1 The pentamer has an overall
diameter -102 A with a central pore diameter -30 A and each monomer has a diameter
-36 A.4 Each monomer also contains two calcium-binding sites (Figure 1) on one face
and a Fc receptor/Clq binding site on the opposing side (Figure 2). The P-face binds in a
calcium dependent manner to phosphatidylcholine (PC) while the a-face binds to Fc
receptors and is the Clq binding site
conditionally.11 It is unclear exactly what happens
to serum or pentameric C-reactive protein (pCRP)
when binding occurs to phosphocholine but it is
clear that new epitopes are exposed that can
activate the classical complement pathway or
Figure 2. a-face of c-reactive
increase phagocytosis.12 The lipid bound protein is Prtein viewed down the five-fold
symmetry axis. Clq binding site
referred to as modified C-reactive protein located in cleft near a-helix
(green).1;2
(modCRP) and is biologically active.13 Forced
conversion utilizing a variety of different denaturants also leads to a biologically active
form known as monomeric C-reactive protein (mCRP).14,15
The PC-binding p-face contains two calcium ions 4 A from each other and both
are involved with ligand binding. The phosphate group on PC co-ordinates with the
bound calcium ions and the choline group rests in a hydrophobic pocket.11,16 The
2


hydrophobic pocket has been shown to be imperative for PC binding using site-directed
mutagenesis. Replacement of the less polar amino acids in the pocket greatly decreases
mutant CRPs attraction to PC.17
Clq and Fc receptor both bind on the a-face of CRP but it is uncertain if they
utilize the same or distinct binding sites. The a-face contains an unusual cleft, near the
a-helix, that is the binding site for Clq and possibly the binding site Fc receptors.4 The
cleft is near the c-terminus of the protein and mutational data has confirmed specific
residues related to Clq activation. Specifically, Glu 88 seems to assist in the
conformational change of Clq that activates the complement system.18 The Fc receptor
binding site has not been as well characterized as the Clq site, but it is suspected to
interact with CRP in the same cleft.
Ligands of CRP
Due to steric hindrance it is unlikely that one pCRP can have multipoint
attachment to a single lipid source and activate Clq. Previously, it was believed that
activation was caused by multiple pCRP substituents binding to one substrate in order to
become active.19 For example, six pCRP molecules could bind near each other and
together activate one Clq molecule. An alternative theory is that upon PC-binding the
monomers of pCRP break apart and rotate.20 This would allow for multipoint attachment
and enhanced signaling from a single molecule.13 It would also explain why bound CRP
allows for Clq-binding and increased phagocytosis, which serum, pentameric CRP does
not induce.5 The latter theory also explains why forcibly monomerized CRP, in vitro, can
activate the complement pathway or phagocytosis.21
3


Pentameric dissociation into mCRP is accomplished in vitro by three major
pathways. 1) Monomerization also occurs when pCRP is heated in the presence of urea
and ethylenediaminetetraacetic acid (EDTA). 2) pCRP has also been shown to convert to
mCRP utilizing sodium dodecyl sulfate (SDS) and heat. 3) pCRP can be change to
mCRP when treated with acid and then neutralized. Addiontally, pCRP, in the absence
of calcium, will decompose in prolonged storage into mCRP. The urea/EDTA method
has been the most widely used and established method for creating mCRP; however,
complete characterization of mCRP has not yet been published.1415 pCRP bound to
membranes that has become biologically active (modCRP) and forced isoform
conversion (mCRP) may be identical, but this has not yet been unambiguously confirmed
because neither has been fully characterized.
Function of C-reactive protein
CRPs exact biological function is unknown; however, CRP is able to recognize
damaged cells, pathogens and remove them by activation of complement or uptake by
phagocytic cells.5,22 PC is present on the outer leaflet of most cell membranes, so it may
seem surprising at first that CRP does not bind to all mammalian cells. CRP is only
found on apoptotic cell walls and it has been demonstrated that CRP only binds to lipids
containing some amount of lysophospholipids. It has been suggested that cell damage
causes membrane flip-flop or exposure of hydrophobic tail groups of PC allows for CRP
binding.23
Once PC-bound, CRP recruits Clq and C3 convertase is effectively initiated
through the classical complement pathway.24 For successful activation of Clq, either
multiple CRP molecules must be bound in a localized area or the CRP monomers of a
4


single molecule must rotate and separate.25 This is another piece of evidence to support
the alternative rotation and dissociation model, of CRP binding to PC. According to this
model, only one Clq binding is available for pCRP due to steric hinderance. However, if
rotation and dissociation occurs on lipid membranes then fewer CRP molecules could
activate classical complement pathway.5
CRP binds PC in a calcium dependent manner.26 Only PC-bound CRP activates
complement or can bind to the FcyRI and FcyRII receptors on phagocytic cells.5 It was
found that mice with FcyRI and FcyRII deficiencies were not able to uptake CRP-bound
apoptotic cells in macrophages.27 Along with a parallel study involving monocytic cell
lines expressing FcyRI and FcyRII confirmed that FcyRI and FcyRII are receptors for
CRp 27 jhgj-g have been multiple conflicting studies regarding CRP uptake by
macrophages in the past and a possible explanation for this is the expression of different
FcR receptors leading to inhibitory or activation signals. CRP acts as an opsonin for
particles and leads to uptake by white blood cells through the FcRyl receptors.22 CRP
mediated opsonization and uptake has been shown to occur with bacteria as well as
apoptotic cells.28 Free, serum pCRP does not interact with either phagocytic cells or
activate complement. This indicates that serum CRP is a proprotein that requires binding
to PC to become biologically active.21
Cholesterol is transported around the human body in lipoprotein particles to
provide for cells.29 High-density lipoproteins carry cholesterol from peripheral cells and
carry it back to the liver and typically have diameters from 5-15 nm.30 Low-density
lipoproteins carry cholesterol from the liver to cells and have diameters around 15-30
nm. CRP does not bind to native LDL, but does bind with oxidized or enzymatically
5


modified LDL.31 Modified LDL bound by CRP is then consumed by a macrophage to
create foam cells.32,33 Foam cells deposit in arterial walls and contribute to
atherosclerosis, which can lead to myocardial infarction.34,35 Ridker and co-workers have
employed the concentration of CRP in serum as a predictor of risk factor of future
cardiovascular events.32,36,37
Cardiovascular disease
CRP may have a causal role in inflammation leading to arteriosclerotic
lesions.10,38 Previously, LDL cholesterol concentrations were the main diagnostic tool for
establishing risk of future coronary events; however, as high sensitivity assays for CRP
were developed CRP levels were found to have more prognostic value for determining
future risk.36 For example, half of all acute ischemic events occur in patients without
hyperlipidemia indicating that LDL concentrations are not necessarily the best method
for predicting future events.39 Fortunately, statins, commonly used to lower LDL
concentrations, also lower CRP concentrations even in patients were LDL concentrations
were already low.37 The risk of future events in patients with low LDL and high CRP
levels was higher than that of patients who had high LDL and low CRP levels confirming
that CRP is a better predictor of cardiovascular events.37
Currently, it is still unclear if CRPs role in cardiovascular disease is simply as a
marker or if high CRP levels are a cause of cardiovascular events.39 There have been
many contradictory results in this area of study because the two isoforms of CRP
(monomeric and pentameric) behave differently in respect to their atherogenicity. This
role for a member of the innate immune system was not initially agreed upon because it
opposes the findings of other host-defense molecules. However, differentiating between
6


isoforms in vivo is still not well established and is the likely cause of the inconsistent
results.40
pCRP has been reported to have pro- and anti-inflammatory characteristics.
pCRP upregulated the expression of adhesion molecules to endothelial cells and stirred
monocyte chemotactic protein release both of which are precursors to inflammation.41
pCRP also provoked cytokine-release including: interleukin 1, 6, 8, 18 and tumor
necrosis factor-a.42 Conversely, pCRP has also been shown to have vasculoprotective
potential. pCRP induces expression of the anti-inflammatory interleukin-10 as well as
interleukin-1 receptor antagonist.43 The above findings have been questioned because of
the presence of lipopolysaccharide and sodium azide as part of the experimental
procedures.5 These contaminants could play a part in the inconsistent results observed.
Additionally, stored pCRP spontaneously converts to mCRP over extended periods of
time especially in solutions lacking calcium.14 When contaminants and monomerization
were controlled pCRP has since been shown to have anti-inflammatory properties.5
pCRP inhibited platelet aggregation and did not prompt cytokine release or adhesion
molecule expression. These new studies indicate that pCRP does not have pro-
inflammatory effects and may even have anti-inflammatory potential.40
Utilizing mice as model for studies involving CRP has also lead to contradictory
evidence regarding the involvement of CRP in atherogenesis. Mice do have a small
quantity of serum CRP but it does not behave in the same manner as human CRP does in
humans.5 Human CRP expressed in mice was shown to protect against certain infections
and it was assumed to proceed by the classical complement pathway.44 However, when
testing was expanded to include atherogenic studies the results were inconclusive. pCRP
7


was demonstrated to increase, decrease or not effect the development of atherosclerotic
plaque.8 Contradicting previous results, it was demonstrated that human CRP did not
activate mouse compliment (in studies not using sodium azide); therefore, mouse models
are likely not ideal for extrapolation to humans when mice are injected with human CRP.
Rat models have provided much more reliable data when examining stroke and
myocardial infarction because human CRP activates rat complement. When human CRP
was injected into rats there was a clear and consistent development of atherosclerotic
plaque.38 This evidence supports the idea that CRP plays a contributory role in
atherosclerotic lesions.5
Effectively differentiating mCRP and pCRP has been of the utmost importance, as
more evidence emerges indicating that protein isoform determines functionality. mCRP
epitopes are revealed in vivo when pCRP interacts with lysophosphatidylcholine, which is
readily found on activated platelets and apoptotic cells.43 Interaction with this lipid
facilitates isoform conversion from serum pCRP to pro-inflammatory mCRP. mCRP has
not been detected in circulation but has been detected deposited in atherosclerotic
lesions.34 These pieces of evidence support the theory that mCRP is formed when pCRP
binds to specific substrates.
Introduction to lipid coated gold nanoparticles
Biological systems are immensely complex and isolating and identifying specific
interactions is quite challenging. Cell mimics were employed to reduce the number of
possible complications.45 A gold nanoparticle core was chosen because of its easily
tunable size allowing it to be made to replicate LDL. Additionally, cyanide stability tests
can be used to determine if an ion impermeable lipid layer surrounds all of the gold
8


nanoparticles (AuNP).46 This prevents the gold core from directly interacting with CRP
and eliciting an unwanted response.
AuNP and Thiols
Functionalized nanoparticles have been a focus of much research because of their
tunable properties and promise in catalysis, biosensing, electronics, and
nanomedicine.4748 Modifications to AuNP have been widely studied, especially thiol
amended because of the strong affinity between sulfur and gold. There are a variety of
ligands that can be attached to gold because of this interaction; therefore, an immense
amount of applications are possible with functionalized AuNP.49 Nanoparticle size is
also an important factor in determining the possible functions of the complex. In 1973,
Fren described a method for reducing gold(III) with varying concentrations of citrate in
order to produce uniform AuNP of different sizes.50 These gold cores have been
augmented in an array of fashions by various methods but frequently utilizing thiols.
AuNP appear red or pink in solution with a size and shape dependent light
absorption. The absorption is described as localized surface plasmon resonance (LSPR),
which involves restricted electron movement on AuNP.51 The incident light causes
resonating oscillations of restricted electrons on the gold surface, which effectively leads
to light absorption. The observed light absorption is the addition of the cross-sections of
absorbance and scattering.52 As gold nanoparticle size increases there is a distinct red
shift in agreement with Mie theory and theoretical predictions of spectra are possible.
Mie theory is mainly useful in dilute solutions were particles do not readily interact.51
This can also be a useful tool for confirming the size of the nanoparticles created since
the wavelength is dependent on the gold nanoparticle size. Unfortunately, the AuNP
9


absorb all of the emission from proteins when excited at 285 nm; this phenomenon is
called inner fdter effect and is a common problem for fluorescent studies.53
AuNP are typically aggregates of gold(O) atoms and are not soluble in aqueous
environments. However, if they are functionalized with a thiol that contains a
hydrophilic group opposing the sulfur end then they can become suspendable in water.54
This technique has been utilized in a series of different fashions to obtain distinct uses for
the water-soluble AuNP. Cyanide was introduced into the solution to establish the
protective capability of the thiol-ligands. Cyanide oxidizes the gold(O) nanoparticles and
forms a dicyanogold(I) complex that has no localized surface plasmon resonance
4Au + 8CN' + 02 + 2H20 4[Au(CN)2]' + 40H' (1)
(Equation l).55 Many different studies have characterized the ion permeability of thiol-
ligands shells by this process, which is easily observed using a UV-vis spectrometer. The
stability of these partially protected, water soluble, thiol coated AuNP was attributed to
the organization of ligands on the surface. As steric hindrance of the thiol ligands was
increased the packing efficiency decreased and the nanoparticles became more
susceptible to cyanide.56
Long-chain alkanethiols were demonstrated to be the most effective ligand for
providing cyanide stability. Bulky ligands, like dendrimers, or short-chain alkane thiols
were found to be less protective against cyanide stability.56 Thiol displacement reactions
Scheme 1. Displacement of thiol coated gold nanoparticles by a shorter thiol.


showed that thiols like dithiothreitol (DTT), dihydrolipoic acid and glutathione could
replace existing long-chain alkane thiols on the surface of AuNP and expose the gold(O)
core to cyanide.57 This displacement proceeded quickly in aqueous environments were
the opposing polar end groups helped stabilize the AuNP. Smaller thiols were more
effective at the displacement of other thiols likely because their molecular size allows for
easier penetration of the existing ligand (Scheme 1). Nuclear magnetic resonance
spectroscopy has also been utilized to observe thiol exchange reactions on gold cores.56
It has been suggested that the displacement occurs by an associative mechanism where
the incoming thiol forces the bound thiol to detach from the surface.57
Biological Membranes and Supported Surfaces
Membranes are vital features of biological
systems, serving as interfaces between the interior
and exterior of cells as well as vesicles.58 These
membranes are organized as large liposomes with
hydrophilic outer ends with a hydrophobic interior.
Typical animal membranes consist of equal parts
lipids and proteins by weight and control the
transfer of signals into and out of the cell.59
Membranes are fluid structures that are difficult to
characterize because typical characterization techniques like x-ray crystallography do not
give accurate representation of hydrated lipids.60 Deposition of membranes onto solid
supports is a popular method of creating biological mimics.61 With the wide range of
surfaces that can be synthesized and the increase in the number of surface-sensitive
[_Z__If______________'_____________i
Scheme 2. One possible
mechanism for lipid deposition
onto planar surfaces.
11


characterization techniques, supported lipid bilayers have become a successful way to
model complex cellular structures.62,63
The precise mechanism by which liposomes form bilayers depends on multiple
factors but clearly involves the rupture of vesicles and adhesion to supports.64 Most
models predict an accumulation of liposomes near the solid surface and when a specific
surface density is obtained adhesion occurs.58 This process can vary in length from
seconds to hours and depends on the ionic strength of the solution, constituents of the
liposome and the surface roughness. Asymmetry of lipid bilayers is common in nature
and can also exist on supported lipid bilayers especially on highly curved membranes.
Identifying defects in the lipid layer is important because analyte membrane
interactions can be significantly different from analyte support interactions.65 Planar
supported surface techniques have previously yielded successful results when analyzing
interactions of lipids with CRP.66
12


Lipid-coated Nanoparticles
Lipid-coated nanoparticles have a large range of biomedical applicability acting
as simple models to identify integral portions of binding interactions. Control over the
size and shape of the nanoparticles as well as membrane composition allows for clear
correlation between protein interactions and specific substrates. Supported lipid bilayers
have been shown to maintain fluidity, incorporate proteins and are impermeable to ionic
species like natural liposomes (Scheme 2).65 Planar substrates were originally the most
commonly investigated substrate for membrane deposition; however, in the past several
years 100 nm and 50 nm silica nanoparticles and various sizes of AuNP have been coated
with lipid bilayers.646768
There are many natural occurrences of high membrane curvature including
membrane blebbing during apoptosis, vesicle trafficking and lipoproteins. Creating
mimics for these structures can be a valuable tool for assessing the properties of different
proteins that interact with these biological structures. Free-floating vesicles typically
exist with very large diameters and forcing the lipids onto smaller nanoparticles leads to
strain. Lipid bilayers are fluid, so reorganization and packing can relieve some of this
strain but lipid composition can also impact the membranes ability to fit to specific
curvature. Pores, holes in the lipid bilayer, can occur when a membrane is unable to
cover a specific area due to various surface deformations. Atomic force microscopy has
been one the most commonly utilized methods for evaluating lipid uniformity and
detecting defections on supported membranes.64
Membranes can also be easily and quantitatively modified to assess numerous
aspects of lipid composition. In particular, cholesterol reduces membrane fluidity by
13


increasing membrane packing, which can lead to a more stable membrane.69 The flat and
rigid cholesterol molecule with a small polar headgroup imposes local ordering in the
hydrophobic portion of lipid membranes. Cholesterol is mainly located in the
membranes of cells and the alteration of membrane function by cholesterol content is an
important area of study.70 Supported bilayer membranes containing cholesterol have
even been demonstrated to be air stable during analysis. PC vesicles, containing up to 50
mole percent cholesterol, have demonstrated decreased ion permeability.71
Lipid coated AuNP
AuNP of various sizes have been coated in PC in order to create cell mimics.
Under specific conditions they can be stable in aqueous or organic solvents and switching
Scheme 3. Oleate coated gold nanoparticles are incubated with PC liposomes. The thiol
anchor is added to stabilize the hybrid lipid bilayer against cyanide penetration.
between two solvent conditions is possible.67 PC coated AuNP (PC-AuNP) can be
synthesized by at least two methods: either mixing gold(III) with phospholipids before
reduction or by reducing gold(III) to gold(0) before the addition of lipids 46,72 To
synthesize ion impermeable membranes surrounding the gold core a thiol was used to
increase packing of the lipid membrane (Scheme 3). The ion impermeability of thiol
14


stabilized PC membranes surrounding gold(O) nanoparticles cores can be evaluated by
cyanide etching.46 If the hybrid lipid bilayer is complete then cyanide will be unable to
oxidize the gold core, which can be observed by the lack of a visible absorption
decrease.73 It has been suggested that the thiol replaces some portion of the inner leaflet
of the bilayer and reduces the strain of the curved membrane.46,74 The exact location of
the thiols within the bilayer has still not been exclusively determined.75
AuNP have also been functionalized to mimic to high-density lipoproteins.76 This
involves first coating the nanoparticle with apolipoprotein A-l before coating with PC.
This process can be observed by dynamic light scattering (DLS), which evaluates particle
size utilizing Brownian motion. The nanoparticle size systematically increases with each
additional additive in accordance with the size of the reactant. Determination of
monolayer (~5 nm increase) or bilayer (~10 nm increase) deposition is possible using
dynamic light scattering.76
15


Introduction to Fluorescence
Fluorescence is the subcategory of luminescence, where a molecule emits light
from an electronically excited state. Excited states are typically achieved by light
absorption from a source that outputs high enough energy to ensure electron excitation
from a ground state to an excited state. Internal conversion can occur, which leads to
non-radiative decay by vibrational relaxation. Intersystem crossing, involving a spin flip,
to a triplet state is also possible but that process is completely suppressed in aqueous
solutions. After internal conversion, a
comparably fast process, the excited electron
will be at the lowest vibrational level of the
excited state. The drop from the lowest
vibrational level of the first electronic state to
the ground electronic state will result in the
emission of light.53
Tryptophan, tyrosine and phenylalanine
Figure 3. Fluorescent tryptophan
are three fluorescent amino acids that can residues in c-reactive protein shown in
green. There are 6 tryptophan residues
sustain excited states in their conjugated % per monomer, 30 tryptophan residues
total.1,2
systems. Tryptophan fluorescence has been
commonly used to identify solvent exposure to various parts of proteins. Excitation at
285-295 nm yields the highest emission intensity at roughly 345-350 nm for most
proteins.53 Tryptophan fluoresces more readily in hydrophobic environments where non-
radiative decay is less prevalent. Protein folding can create hydrophobic pockets where
tryptophan residues can fluoresce77 Protein folding can also decrease fluorescent signal
M

-j: \ / /* ^


;Si ,
OV ^ . \s r a'


16


because of absorption by other amino acids. Additionally, inner filter effects caused by
AuNP prevent fluorescent analysis because of their optical density. When unfolding
occurs, tryptophan residues are exposed to aqueous solvent and non-radiative decay by
energy transfer to the solvent becomes the most prominent form of energy release. C-
reactive protein has six tryptophan residues per monomer, totaling 30 tryptophan residues
per pentamer (Figure 3). Individual fluorescence will depend on the electronic
environment of the tryptophan and conversion from pCRP to mCRP has been shown to
lead to decreased intensity of fluorescence due to an increase in the amount polar solvent
exposure to tryptophan residues.78
Fluorescence Quenching
Fluorescence quenching has been a commonly utilized technique for identifying
locations of fluorescent amino acids in proteins. The most frequently utilized type of
quenching is dynamic or collisional quenching, where the excited fluorophore contacts
the quencher that absorbs the energy in a non-radiative manner. The mechanism by
which the energy is transferred is not well characterized; however, electron transfer is the
leading theory because of certain studies and models.79 Static quenching, where the
quencher attaches to the fluorophore, is a common complicating feature of quenching
analysis. Quenching is considered a non-destructive technique because typically there is
no photochemical reaction between the quencher and the fluorophore.53
The Stern-Volmer equation relates the decrease in the fluorescence intensity to
y=l + Ksv[Q]
(2)
17


the quencher concentration as shown in Equation 2. This equation relates the ratio of the
decrease in fluorescence (F0 initially, F in the presence of quencher) to the concentration
of the quencher, [Q], and the Stem-Volmer constant, KSv- Quenching data is plotting
with Fo / F on the y-axis and [Q] on the x-axis.53 A linear relationship between quencher
concentration and F0 / F typically indicates that fluorophores have a similar electronic
environment and are equally accessible; however, a complex system with multiple
fluorophores in different electronic environments can lead to multiple complications.80
The type of quencher utilized can also affect the slope of the plot because of interactions
with other portions of the molecule. For example, large ionic quenchers do not typically
effectively quench tryptophan residues that are located inside hydrophobic regions of the
protein. This leads to a curved quenching plot because the interior fluorophores are
continuing to fluoresce despite the ionic quencher. Conversely, acrylamide, a common
quencher, has been established to invade proteins, to a certain extent, and to quench
internal tryptophan residues.81
Additional thought also needs to be given to the efficiency of the quencher (y).
For efficient quenchers, every collision leads deactivation of the excited fluorophore, y
approaches unity and the determining factor in KSv is diffusion rate limited (Equation
yKSV KsVobserved (3)
3).82 Efficient and established quenchers include: oxygen, acrylamide and iodide.
Inefficient quenchers include: succinimide, bromate and imidazole. Efficiency can play a
role in the linearity of the Stern-Volmer plot as well as fluorophore availability.
Inefficiency may cause a false curve to appear or fortuitously cause a straight line to
appear when a combination of inefficiency in both static and dynamic quenching
18


occurs.81 Efficiency is important to consider when attempting to characterize
fluorophores of proteins.
An ionic quencher that is unable to quench interior fluorophores in a protein may
demonstrate a negative curve on the Stern-Volmer plot, as certain fluorophores may
remain unaffected by the ionic quencher. Divergently, oxygen is one of the best-known
quenchers; however, it is not as useful for establishing exterior or interior tryptophan
residues in a protein because it is easily able to permeate most proteins and therefore
quench multiple tryptophan populations.53 All quenchers have different properties that
suit them for specific experiments and choosing the optimal quencher is typically
determined experimentally.
Tryptophan Fluorescence and Quenching
Tryptophan has a large amount of sensitivity to its local electronic environment
and has additional complications because it has two excited states, !La and ^b.
Depending on multiple factors, including solvent exposure, polarity and nearby amino
acids either of the two excited states can dominate emission by being lower in energy.53
The !La is the lower energy excited state for most proteins in aqueous solution and
typically is the major excited state when excited between 285-290 nm. When excited in
aqueous solvents, tryptophan, not an efficient fluorophore, has also been found to have a
quantum yield of ~0.13, which is one of the major factors against intrinsic fluorescence
investigations.83 The emission spectrum of tryptophan undergoes a red shift when
dissolved in more polar solvents. Beginning at an emission maximum around 300 nm in
cyclohexane and moving to 340 nm in ethanol and 350 nm in water when excited at 285
nm. It is important to note that when dissolved in cyclohexane the excited state is the
19


lower energy excited state unlike when tryptophan is dissolved in water or ethanol where
^La is the lowest excited state.53
Previously studied proteins have showed two distinct trends from their native to
denatured isoforms. The emission maximum undergoes a red shift, to varying
magnitudes, as well as the denatured protein having increased sensitivity to quenching.
The latter aspect is revealed to lead to a steeper slope in the Stem-Volmer plot or a larger
KSv- Szabo and co-workers showed that the protein azurin demonstrates a significant red
shift when denatured by guanidine hydrochloride.84 This is logical when considering that
when tryptophan is dissolved in increasingly polar solvents the emission shifts to longer
wavelengths; therefore, this study is analogous because the tryptophan residue is entering
a more polar environment and the emission wavelength shifts accordingly. Avigliano
and co-workers demonstrated that the protein stellacyanin contained tryptophan residues
with two different amount of quencher accessibility by quenching the native and
denatured protein with iodide.85 The larger KSv value of the denatured form of the
protein indicated that the tryptophan residues were more readily quenched by iodide,
which is most likely due to increased solvent exposure. This confirms other work that
iodide, an ionic quencher, is unable to penetrate certain proteins and quench inner
fluorophores, but iodide is capable of quenching solvent exposed tryptophan residues.83
A denatured protein typically loses its tertiary or quaternary structure and is more likely
to have solvent exposed residues like tryptophan.86 This work is still applicable even
though mCRP is not technically a denatured form of pCRP because the idea of increased
solvent accessibility to tryptophan residues is still consistent.78
20


Succinimide is a bulky, polar, inefficient quencher that can be utilized to detect
exterior tryptophan residues.81 It has been suggested that because of the size and polarity
of succinimide it is unable to quench the interior tryptophan residues in certain proteins.
Also, because succinimide is not an ionic quencher there are fewer considerations when
determining salt effects during quenching studies. This makes succinimide a unique
candidate for confirming solvent exposure tryptophan residues in multi-tryptophan
proteins.82 Even some larger, polar quenchers, like acrylamide, have demonstrated the
ability to invade proteins and quench interior tryptophan residues. Eftink and co-workers
utilized succinimide as a model to show that certain quenchers were incapable of
quenching interior proteins; therefore, only affecting the solvent exposed tryptophan
residues. Succinimide has been demonstrated to only be partially effective (y 70 %) at
quenching tryptophan residues dissolved in water; indicating that even certain solvent
exposed tryptophan residues may not be quenched by succinimide.81 Succinimide has
many unique characteristics that make it suitable for specific tasks in certain situations.
Binding studies of CRP to PT-PC-AuNP
PT-PC-AuNP is a cell mimic that effectively acts as an apoptotic bleb or LDL.
CRP has been shown to bind to these in a calcium dependent manner; therefore, EDTA
can be used as a control to prevent binding and confirm that observed isoform conversion
is due to calcium dependent binding to the cell mimics. The addition of BME disrupts
the lipid layer surrounding the AuNP, which can be demonstrated by cyanide instability
(Scheme 4). When BME is used with centrifugation the membrane bound mCRP can be
21


separated from the AuNP (Scheme 5). This is a key step because fluorescence studies
that cannot be conducted in the presence of AuNP due to their inner filter effects.
Succinimide was the fluorescence quencher that was utilized to differentiate between the
two isoforms of CRP and binding studies between CRP and PT-PC-AuNP were
confirmed by gel electrophoresis.
2-Mercaptoethanol
Scheme 4. Propanethiol stabilized PC-AuNP become completely
cyanide instable after 2 hours when mixed with BME.
&3
Scheme 5. Isoform conversion as a result of calcium dependent binding
of c-reactive protein to PT-PC-AuNP. Removal of the hybrid lipid bilayer
by BME for fluorescence quenching analysis.
22


CHAPTER 2. EXPERIMENTAL
Materials
All water was deionized to 18.2 MQ or better (Milli-q Water Purification
Systems, Millipore, Bellerica, MA), and reagents were reagent grade. Frens method was
utilized for the formation of AuNP and lipid coating and thiol anchoring were adapted
from previous reports with modifications.5087
Gold nanoparticle preparation
AuNP were created by dissolving hydrogen tetrachloroaurate(III) hydrate (18 mg,
Strem Chemicals, Newbury Port, MA) into 100 mL of water (530pM Au) and adding
sodium citrate (1.7 mM final concentration, JT Baker, Phillipsburg, NJ) to a refluxing
solution for 15 minutes. The AuNP were allowed to sit for 24 hours without stirring.
The sizes were determined by dynamic light scattering (DLS, Malvern Nanosizer) and
confirmed by UV-vis spectroscopy (PerkinElmer UV-vis spectrometer Lamba 35).
Lipid-coated, thiol-anchored AuNP synthesis
L-a-phosphatidylcholine (302 mg PC, Avanti Polar Lipids, Alabaster, AL) was
dissolved in 5 mL of chloroform (80 mM PC) and kept at 0 C when not in use.
Cholesterol (32 mg, Eastman Organic Chemicals, Kingsport, TN) was dissolved in
chloroform (17 mM) and kept at 0 C when not in use. Aliquots of 25 uL of the PC
solution and 12 pL of the cholesterol solution were added to a glass vial where they were
mixed, evaporated by a stream of nitrogen and the remaining chloroform was evaporated
overnight in a vacuum desiccator (91% PC and 9% cholesterol). PC films were
23


rehydrated with 10 mM 4-(2-hydroxyethyl)-l-piperazineethanesulfonic acid (HEPES,
200 pL, Sigma-Aldrich, Saint Louis, MO) and placed in a bath sonicator (Branson 1510,
Emerson Industrial Automation, Danbury, CT) for one hour. AuNP (1 mL) were
incubated with 5 pL of oleic acid sodium salt (100 iiM final concentration, TCI America,
Portland, OR) for thirty minutes. PC/cholesterol rehydrated in HEPES was added to the
AuNP solution (20 pL, 200 nmol PC, 20 nmol cholesterol final concentration) and
allowed to stir for one hour. Propanethiol (3 pL, Acros Chemicals, Belgium) was
dissolved in 10 mL of water (3.3 mM); all thiols were made fresh each day. Propanethiol
(10 pL, 32 pM final concentration) was added to each vial and allowed to stir for 30
minutes. Occasionally, hexanethiol was used in place of propanethiol for comparison
purposes. Hexanethiol (1 pL) was dissolved in 1 mL of ethanol (7.1 mM) and 3 pL were
added to each vial (21 nM final concentration).
CRP binding studies with PT-PC-AuNP
Purified CRP (16 nM final concentration, Academy Biochemical Company) was
added to 900 pL of PC coated, thiol anchored, AuNP (PT-PC-AuNP) for thirty minutes.
Ethylenediaminetetraacetic acid tetrasodium salt (2.08 g EDTA, Matheson Coleman and
Bell, Norwood, OH) was dissolved in 5 mL of water and 5 pL aliquots were used to
prevent calcium dependent binding of CRP. (i-mercaptoethanol (3 pL, BME) was added
to 500 pL of water (85 mM) and 12.5 pL of the BME solution were added to each vial
(1.16 mM BME final concentration) for two hours. Alternatively, DL-dithiothreitol
(DTT) was used for comparison displacement with BME. N-methyl maleimide (25 mg,
NMM) was dissolved in 2 mL of water (113 mM NMM) requiring small amounts of
24


sonication and 10 uL of the NMM solution (1.22 mM final concentration) was added to
the reaction vial and allowed to sit for thirty minutes. Other studies included 100 uL of
the 113 mM NMM stock solution in order to completely prevent in gel degradation of
CRP. The solutions were centrifuged (Eppendorf, Hamburg, Germany) at 15,000 rpm for
20 minutes and 800 uL of supernatant was collected. The pellet was suspended with 800
pL of water for analysis.
Monomeric CRP preparation
Monomeric C-reactive protein (mCRP) was formed from stock pentameric C-
reactive protein (pCRP) utilizing 8 M urea and 10 mM EDTA at 37 C for 2 hours. The
solution was dialyzed three times against PBS at 4 C and concentration was determined
using a Bicinchonic acid protein assay kit (Peirce).
Gel electrophoresis
Agarose gels (0.8%, 0.4 g, Bio-Rad, Hercules, CA) dissolved in 50 mL of 0.5 x
Tris/Borate/EDTA buffer (TBE) and 0.005% sodium dodecyl sulfate (SDS, USB
corporation, Cleveland, OH) was added after the agarose was dissolved in TBE. The 5 x
TBE buffer was created by dissolving 54 g of Tris (445 mM, JT Baker), 13.8 g of Boric
acid (445 mM, Mallinckrodt) and 2.9 g EDTA (10 mM) were dissolved in 1 L of water
and dilutions were made from this stock. The gel was run in 0.5 x TBE / 0.005% SDS
buffer at 40 V for thirty minutes. The gel was transblotted onto nitrocellulose membrane
(NTC, Schleicher and Schuell, Keene, NH) using a semi-dry transblotter at 25 V for 90
minutes. The NTC was blocked with 3% bovine serum albumin (BSA, OmniPur,
Gibbstown, NJ), 0.05% Tween-20 (VWR International, Westchester, PA) and 0.002%


NaN3 in PBS overnight. The NTC was washed three times with Tween-20/PBS solution
before equilibrating with biotinylated polyclonal anti-CRP antibody (1:5000, Academy
Biomedical Company) for one hour. The primary antibody could be used multiple times
and was stored at 4 C along with the secondary antibody. The NTC was washed three
times with Tween-20/PBS, incubated with streptavidin-IR800 for a half hour (1:5000, Li-
COR Biotechnology) and imaged on an Odyssey imager (Li-COR Biotechnology).
Fluorescence analysis
A spectrofluorometer (Photon Technology International, Birmingham, NJ) using
FelixG software was used to analyze the supernatant after centrifugation. Measurements
were carried out in a 1.5 mL quartz cuvette (Starna Cells, Atascadero, CA) at 37 C with
2 nm slit widths for the excitation monochromator and 8 nm slit widths for the emission
monochromator. The samples were excited at 285 nm and data was recorded from 300 to
400 nm in 1 nm increments and 0.5 second integration time. The spetra were corrected
for background absorption and the Raman peak of water.78,88
Fluorescence quenching studies
Quenching experiments were conducted on binding samples as well as pCRP,
mCRP and tryptophan controls. Supernatant collected from centrifugation (800 pL) was
evaluated for isoform determination after adding 23 pL of 5.2 M sodium chloride (144
mM NaCl, Mallinscrodt) and 3.5 pL of 600 mM calcium chloride (2.5 mM CaCh,
Sigma-Aldrich). Succinimide (0.2 g, Alfa Aesar, Ward Hill, MA) was dissolved in 1 mL
of water (< 3 minutes of sonication, 2.02 M) and adjusted to maintain a constant ionic
strength throughout the standard additions. Occasionally, acrylamide (0.15 g, Sigma-
26


Adlrich) was used for comparison and was dissolved in 1 mL of water (2.11 M) and
adjusted for ionic strength.
Cyanide stability studies
Potassium cyanide (0.2 g, Mallinckrodt) was dissolved in 5 mL of water (600
mM). Thiol-anchored, lipid-coated gold nanoparticles were tested for ion impermeability
by the addition of 10 pL of the cyanide stock solution (6 mM final concentration). The
UV-vis spectrum was taken of each sample, from 400 600 nm, before the addition of
cyanide, 1, 2, 3, 4, 5 and 24 hours after addition. The data was compared to the original
absorption at 525 nm to evaluate the loss of signal due to cyanide oxidation.
27


CHAPTER 3. RESULTS
Formation of PT-PC-AuNP
AuNP characterization
AuNP were created as a nanoparticle scaffold for binding studies between CRP
and LDL mimics. The average size of all AuNP created was 17.3 nm ( 1.2) by DLS and
the average Lmax for those AuNP was 525 nm. The optical density of AuNP at 525 nm
was 0.9, which was used to determine the concentration of AuNP and the expected
amount of thiols required to coat the AuNP.87 The average size of PC-AuNP was 26.1
nm ( 4.3) by DLS, an addition of 8.8 nm, the average size of two bilayers is 8 10 nm.59
Maximizing ion impermeability of PT-PC-AuNP
Optimizing the concentration
of oleate for PT-PC-AuNP was important for prevention of AuNP 110 g 100 Ol c 5 90 a a i 1 &
aggregation during lipid and thiol a S 80 .Q O
addition. Varying amounts of oleate < 70
were stirred, in separate vials, with 1 C ) 20 40 60 80 Oleate (uM) 100 120
mL of AuNP in each to determine the Figure 4. The absorbance retained at 525 nm
after 24 hours of incubation with cyanide
minimum concentration of oleate compared with the concentration of oleate used
to prevent aggregation. The error bars are the
required to maintain stability against standard deviation of at least j trials,
cyanide for 24 hours. For 1 mL of 1.5 nM AuNP: 33 nmol of propanethiol, 20 nmol of
cholesterol (9% of total lipid), 200 nmol of PC and 6 mM cyanide were all held constant.
28


Initial absorbance was recorded before the addition of cyanide and compared to
absorption after samples were allowed to incubate for 24 hours with cyanide.
Maintaining a constant amount of absorbance over a 24 hour period signifies that an ion
impermeable membrane surrounds the gold core. As shown in Figure 4, the PT-PC-
AuNP did not retain 100% cyanide stability until the concentration of oleate was 100 uM,
The lipid layer surrounding
the gold nanoparticle was modified
with cholesterol to increase ion
impermeability.69 Different amounts
of cholesterol were mixed with PC
while both were dissolved in
chloroform to ensure homogeneity
before evaporation by nitrogen.
These films were rehydrated in 10
mM HEPES buffer and equivalent
amounts of lipid were added to separate vials. Each vial also contained 33 nmol of
propanethiol, 100 gM oleate and 1 mL of 1.5 nM AuNP (17.3 nm diameter). The result
was an increase in absorbance retention as the cholesterol content was at least 9% by
mole of total lipid (Figure 5). Cholesterol concentrations as high as 20% also created ion
impermeable membranes.
Figure 5. The absorbance retained after 24
hour of incubation with cyanide plotted against
the percentage of lipid layer that was
cholesterol. The remainder of the lipid layer
was always PC and the error bars are the
standard deviation of at least 3 trials.
29


The concentration of propanethiol was optimized for minimum ion
impermeability, which indicates that
cyanide is unable to reach the gold
core. This implies that CRP will also
be unable to interact with the gold
core, which would cause undesirable
interactions. All other reagents were
held constant, including 9%
cholesterol/PC lipid layer and lOOpM
oleate for 1 mL of 1.5 nM AuNP
(17.3 nm diameter). A minimum
amount of 33 nmol propanethiol per 1 mL of PC-AuNP was essential to maintain 100%
retained absorbance and up to ~70 nmol propanethiol could be used before AuNP
aggregation occurred.
20 30 40 50
Propanethiol (nmols)
Figure 6. The amount of nanomoles of
propanethiol injected into a 1 mL solution of
PC-AuNP compared with absorbance retained
after 24 hours of incubation with cyanide. The
error bars are the standard deviation of 10 trials.
30


Binding studies of CRP with PT-PC-AuNP
CRP bindsn in a calcium dependent mannern to PC; therefore, when EDTA is
employed no binding occurs and the pCRP
remains unaltered in solution.1723 This
acts as a useful control to determine if
binding occurred as a results of calcium
dependent interactions with the PT-PC-
AuNP. pCRP was incubated with PT-PC-
AuNP for 30 minutes with and without
EDTA. Samples were then centrifuged,
run on a 0.8% agarose gel, transblotted
onto a nitrocellulose membrane and
visualized by Western blotting. The pCRP and mCRP controls were incubated in buffer
prior to loading in the well and the location of those bands after running and blotting the
gel onto nitrocellulose membrane denote the location of that isoform in the Figure 7. In
samples with EDTA, the pCRP remains in the supernatant; however, in samples without
EDTA, calcium dependent binding occurs and the mCRP is found in the pellet with the
PT-PC-AuNP after centrifugation (Figure 7). The LSPR of AuNP was a useful tool for
during the cyanide stability tests; however, the LSPR absorbs all of the electromagnetic
radiation during fluorescence analysis. In order to evaluate the isoform of CRP by
fluorescence analysis, CRP must be separated from the AuNP.
No EDTA EDTA
Figure 7. Gel electrophoresis and
Western blotting were preformed on the
supernatant (S) and suspended pellet (P)
of samples that did and did not contain
EDTA. Controls of pCRP and mCRP
were run for comparison.
31


Lipid disruption utilizing BME or DTT
The lipid bilayer surrounding PT-PC-AuNP became permeable when BME or
DTT was added. This helps separate the lipid bilayer from the AuNP so that fluorescence
analysis can be preformed on the
lipid-bound mCRP. All PT-PC-
AuNP were synthesized with 100
pM oleate, 9% cholesterol/PC lipid
layer and 33 nmol of propanethiol
per 1 mL of AuNP. As shown in
Figure 8. All samples contained PT-PC-AuNP
Figure 8, complete cyanide and cyanide. One sample contained 1 mM
BME (), another 1 mM DTT () and the last
instability was demonstrated when did not contain any other reactants (). The error
bars are the standard deviation of 10 trials.
PT-PC-AuNP were incubated with
BME after 2-3 hours and when incubated with DTT for 24 hours. With a final
concentration 1 mM the BME or DTT is able coat at least a portion of the AuNP and
expose the gold to the cyanide.
Previously, PC-AuNP were
stabilized with 9 nmol of hexanethiol
per 1 mL of PC-AuNP with an optical
density of 0.8 and 10 nm diameter
AuNP.87 No loss of absorbance was
Figure 9. All samples contained HT-PC-AuNP
and cyanide. One sample contained 1 mM observed when 1 mM BME or DTT
BME (), another 1 mM DTT () and the last
did not contain any other reactants (). The error was incubated with hexanethiol
bars are the standard deviation of 7 trials.
stabilized PC-AuNP (Figure 9).
32


Concentrations of 20 nmol of hexanethiol per 1 mL of PC-AuNP were required to
replicate the cyanide stability in this study. No loss of absorbance indicates that the
hexanethiol anchored PC-AuNP had an ion impermeable membrane, the gold core
surface could not be modified and lipid bilayer could not be disrupted by BME or DTT.
The apparent rise in absorption is peculiar but possibly due to evaporation of water,
which would increase the scattering and absorbance of the solution. The t-test indicates
that all three samples had a statistically significant increase in absorption from the 1st
hour to the 24th hour; however, there was not a statistically significant increase in
absorption from the 1st hour to the 2nd hour and the 5th hour to the 24th hour (99.9%
confidence interval).89 An increase in absorption does not indicate cyanide oxidation of
the gold core.
BME and DTT have been
BME&pCRP DTT&pCRP pCRP mCRP
demonstrated to cleave disulfide bonds in
proteins and disrupt quaternary structure;
therefore, gel electrophoresis was used in
Figure 10. Gel electrophoresis and
order to ensure that the structure of pCRP Western blotting of samples containing
BME and pCRP, DTT and pCRP, pCRP
was not influenced by the presence of BME an<^ mCRP.
or DTT.59 pCRP was incubated with 1 mM BME or DTT for 1 hour before being run
through an agarose gel and blotted in the Western method. The Western blot, Figure 10,
shows that pCRP does not undergo isoform conversion to mCRP under these conditions.
33


Quenching Experiments
Quenching in buffer by
succinimide and acrylamide
Herein, we display the ability
to discern at least two of the isoforms
of CRP employing the fluorescence
quencher succinimide. The data in
Figure 11 shows the result of
tryptophan, Urea/EDTA mCRP and
pCRP succinimide quenching and
Stern-Volmer analysis in buffer
containing 10 mM phosphate, 140 mM NaCl and 2.5 mM CaCh. Initial fluorescence was
recorded and the reduced fluorescence was plotted with the corresponding concentration
of quencher.53 There are 30 tryptophan residues per CRP molecule, so the concentration
of tryptophan was 30 times the concentration of CRP (240 mM) in order to match total
Figure 11. Stern-Volmer plot of pCRP (), mCRP
() and tryptophan () quenched by succinimide in
10 mM phosphate buffer, 140 mM NaCl and 2.5
mM CaCh. The equation of the best-fit line for
tryp: y = 0.0135x + 1.010; mCRP: y = 0.0088x +
1.044; pCRP: y = 0.0031 + 0.975.
Table 1. Summarizes succinimide fluorescence quenching data. Buffer
background refers to quenching fluorescence in 10 mM phosphate buffer, 140
mM NaCl and 2.5 mM CaCl2. PT-PC-AuNP supernatant background refers to
making PT-PC-AuNP, centrifuging at 15,000 rpm for 20 minutes, and adding
_______________mCRP, pCRP and tryptophan to the supernatant.________________
Succinimide Fluorophore MM'1) KsvmCRP/MCRP
Buffer background Tryp 13.5 0.8 2.8 0.4
mCRP 8.8 1
pCRP 3.1 + 0.2
PT-PC-AuNP supernatant background Tryp 7.8 0.1 2.6 0.2
mCRP 2.9 0.1
pCRP 1.1 0.05
34


fluorophore concentration.11 The slope, KSv, of tryptophan was the largest, 13.5 M'1,
because the tryptophan residues are completely exposed to the quencher (Figure 11). The
Ksv of urea/EDTA mCRP (8 nM) was 2.9 M'1 and the Ksv of pCRP (8 nM) was 1.1 M'1
(Table 1).
Acrylamide is a more efficient
quencher of fluorescence than
succinimide, but has been shown to
diffuse into proteins and quench
interior tryptophan residues.53 The
efficiency was confirmed by the data
shown in Figure 12, which shows the
KSv for tryptophan to be twice as
large for acrylamide (KSv = 26.4 M'1)
as succinimide (KSv =13.5 M'1) in
buffer (Figure 11). Additionally,
pCRP had a larger value (KSv = 5.2 M'1) than mCRP (KSv =1.0 M'1) indicating that
acrylamide was not only quenching exterior tryptophan residues but also interior
tryptophan residues.88 The lack of clarity of these acrylamide quenching results gives
support to using succinimide as the better quencher for discerning the two isoforms of
CRP in this particular examination. The succinimide data is consistent with the current
understanding of solvent accessibility to tryptophan and allows for a rational difference in
KSv values for the two isoforms of CRP.
Figure 12. Stem-Volmer plot of pCRP (),
mCRP () and tryptophan () quenched by
acrylamide in 10 mM phosphate buffer, 140 mM
NaCl and 2.5 mM CaCh. The equation of the
best-fit line for tryp: y = 0.0264x + 0.0660;
mCRP: y = O.OOlOx + 1.044; pCRP: y = 0.0052 +
0.926.
35


Succinimide quenching in PT-PC-AuNP background
The KSv of fluorophore-
quencher pairs can be influenced by
background conditions, so the
background was modified to be a
more accurate comparison for CRP
and PT-PC-AuNP binding studies.
Succinimide quenching of
tryptophan, pCRP and mCRP was
Figure 13. Stem-Volmer plot of pCRP (),
mCRP () and tryptophan () quenched by
succinimide in the supernatant of PT-PC-AuNP.
conducted in the supernatant of PT- The equation of the best-fit line for tryp: y =
0.0078x + 1.056; mCRP: y = 0.0029x + 1.057;
pCRP: y = 0.0011 + 1.011.
PC-AuNP in order to observe the
effects of background complications on KSv values. PT-PC-AuNP were created and
centrifuged to replicate the conditions the background experiences during the binding
study. Urea/EDTA mCRP, pCRP and tryptophan were added to the supernatant of PT-
PC-AuNP and quenched by succinimide (Figure 13). mCRP had a KSv of 2.9 M'1 and
pCRP had a KSv of 1.1 M'1, which are both lower in magnitude than the values observed
when mCRP and pCRP were quenched in buffer (Table 1). There is still a similar ratio
(KSvmCRP / KSvpCRP) between the two isoforms of CRP, so quenching by succinimide is
still a viable method for isoform determination.
36


Binding Studies
Utilizing BME to disrupt PT-PC-AuNP
Binding studies of pCRP with PT-PC-AuNP were conducted to evaluate isoform
conversion as a function of calcium dependent PC binding. Binding studies were
monitored by gel electrophoresis and fluorescence quenching experiments to determine
CRP isoform. pCRP was bound to cell mimics in a calcium dependent method; EDTA
was incubated with certain samples
to act as a control to make sure other
factors were not leading to isoform
conversion. In order to analyze
mCRP by succinimide quenching,
BME was utilized to disrupt the
hybrid lipid bilayer and
centrifugation separated the AuNP
Figure 14. Stem-Volmer plots of the
succinimide quenching of the supernatants of
the binding studies. One sample contained
calcium () and isoform conversion occurred (y
from the conceivably still membrane = 0 0012x + 1 025)' The other samPle contained
EDTA () and binding did not occur (y =
0.005x+ 1.0103).
bound mCRP (Scheme 5). To
Table 2. Succinimide quenching data of binding studies samples containing CRP
incubation with PT-PC-AuNP and BME. The 1:1 BME:NMM also contains 1 mM
_______NMM and all samples were centrifuged before quenching analysis._____
Succinimide Fluorophore MM'1) KsvmCRP/MCRP
No NMM Calcium (mCRP) 1.1610.06 2.510.3
EDTA (pCRP) 0.46 0.04
1:1 BME:NMM Calcium (mCRP) 1.19 + 0.05 2.210.2
EDTA (pCRP) 0.55 0.05
37


maintain uniformity, BME was also added to samples containing EDTA despite having
less of an active role since it was already demonstrated in Figure 7 that pCRP remains in
the supernatant.
Succinimide quenching and Stem-Volmer analysis were preformed on the
supernatants of the samples described above. The KSv values of mCRP and pCRP were
1.16 M'1 and 0.46 M'1 respectively (Figure 14). These absolute magnitudes were lower
than what was observed in buffer or in the PT-PC-AuNP supernatant. The KsvmCRP /
KSvpCRP was 2.5 (Table 2), comparable to the 2.8 observed in buffer and 2.6 observed in
PT-PC-AuNP supernatant (Table 1). Conversely, neither of the comparison values
contained BME, which may have effected the KSvmCRP / KSvpCRP BME is thought to
disrupt at least some portion of the lipid membrane surrounding the PT-PC-AuNP, which
would lead to a larger amount of lipids or BME in the supernatant. This variability in the
background has significant effects on the KSv values.
A complication arises when gel electrophoresis and Western blotting is preformed
on the binding samples. BME was used in the binding
studies to disrupt the lipid membrane; however, during gel
electrophoresis the presence of BME led to smearing (Figure
15). The lane that does not contain EDTA (where calcium
dependent binding can occur) shows mCRP as expected;
No EDTA EDTA
however, the smear in the lane that contains EDTA should
only contain one pCRP band.
Figure 15. Western blot
of supernatants of
samples where CRP was
incubated with PT-PC-
AuNP and then BME
before centrifugation.
38


Sequestering of BME by NMM
No EDTA EDTA
Figure 16. Western blot of
binding studies of CRP with
PT-PC-AuNP, after
incubation with 1 mM BME,
23 mM NMM and
centrifugation. The
supernatant (S) and pellet (P)
of samples containing EDTA
To sequester excess BME, N-methyl maleimide
(NMM) was added to the CRP PT-PC-AuNP binding
study before centrifugation. As shown in Figure 16,
NMM prevented the degradation of pCRP in the EDTA
containing samples and did not have effects in the
samples containing calcium. Centrifugation is important
because it allows for fluorescence analysis by removing
the inner filter effects of the AuNP. NMM was not
previously reported to interact with or affect CRP.90
NMM is a quencher of fluorescence and at the 1:23 mol
ratio (1 mM BME: 23 mM NMM) there is no
observable fluorescence signal in the supernatant of the CRP PT-PC-AuNP binding
studies. In a separate study, NMM was added to the CRP PT-PC-AuNP binding studies
samples in a 1:1 BME:NMM ratio in an attempt to prevent CRP degradation during gel
electrophoresis without quenching the entire fluorescent signal. This resulted in KSvmCRP
/ KSvpCRP = 2.2 from succinimide quenching of the supernatant, which is similar to the
KSvmCRP / KSvpCRP = 2.5 without any NMM (Table 2); however, there was still some
smearing in the gel. This serves as an internal confirmation that the degradation of the
pCRP (Figure 16) occurred during gel electrophoresis, and not during incubation with
PT-PC-AuNP and EDTA, because the KSvmCRP / KSvpCRP is similar to the ratio observed
in buffer and in PT-PC-AuNP supernatant (Table 1). The amount of signal from
tryptophan fluorescence was less in the 1:1 BME:NMM samples than samples that did
39


not contain any NMM, so there are additional factors to consider when comparing
Xr mCRP j tr pCRP
Ksv / ls-svF ratios.
40


Chapter 4. Discussion
PT-PC-AuNP preparation
AuNP size and shape determination
In order to ensure monodispersity in size and shape, AuNP were created by a
Figure 17. Adapted from Basu and co-workers.
Xmax is plotted against the diameter, determined by
TEM, of AuNP also synthesized by Frens method
(x). The average Xmax for the AuNP created by this
analyst (o) is plotted to show the expected
corresponding size.
Basu and co-workers demonstrated a correlation between AuNP diameter and X^x
(Figure 17).91 The AuNP from our analysis fits within this trend line by being an
appropriate size for the observed Xmax. This can be inferred as an additional confirmation
of the diameter and shape of the AuNP.
Liu and co-workers developed a method to calculate the concentration of AuNP
based on diameter and absorbance of the AuNP.92 The extinction coefficient (e) was
540
515
510
AuNP Size (nm)
common method and thoroughly
characterized. AuNP were
previously characterized by
transmission electron microscopy
(TEM), dynamic light scattering
(DLS) and UV-vis analysis.50,91
Our lab utilized the latter two
methods to determine the size and
concentration of AuNP
synthesized by the same method.
ln(e) = k ln(d) + a
(4)
41


determined by AuNP diameter (d) and Beers law was employed to determine final
concentration. The relationship between extinction coefficient and diameter is shown in
Equation 4, with k = 3.32111 and a = 10.80505.92 From the data published by
Mackiewicz and co-workers, the total surface area in their analysis was 14 cm2 and the
total surface area in this analysis was 8.6 cm2. The total surface area comparison
indicates that there is around half as much space available for thiols to interact with in
this analysis when compared to previous work. Accounting for the number of thiols,
there were ~4 thiols/nm2 in the Mackiewicz publication while there were ~23 thiols/nm2
in this analysis. It is important to note that hexanethiol was utilized to stabilize the PC-
AuNP in the Mackiewicz analysis and propanethiol was used in this analysis.87 Thiol to
surface area comparison between different studies may be more accurate based on the
current understanding of thiol and gold interactions.
Previous work demonstrated that hexanethiol and decanethiol stabilized PC-
AuNP were impermeable to cyanide etching. The completeness of the lipid layer
was evaluated using cyanide, which readily oxidizes gold(0) to gold(l), and can be
observed by the loss of visible light absorption.55 Novel, PT-PC-AuNP were shown here
to also by impermeable to cyanide penetration. However, concentrations of the other
reactants, including oleate, lipid composition and thiol concentration, had to be modified
because of the use of propanethiol as the anchor.
Oleate concentration
The sodium salt of oleic acid is a surfactant utilized to stabilize the gold AuNP
during the addition of the PC. The AuNP are free floating in solution because the citrate
used to reduce the gold(III) to gold(0) nanoparticles surrounds the AuNP. The role of
42


oleate is not completely understood, but when lipids are added to AuNP solutions without
oleate the AuNP aggregate and crash out of solution. It is likely that the surfactant helps
suspend the hydrophobic AuNP in the presence of lipids, while the citrate can only keep
the AuNP suspended in water. The oleate concentration, 100 iiM, is larger than
previously published oleate concentrations (20 iiM), which used hexanethiol or
decanethiol to stabilize the PC-AuNP.87 Aggregation is a visible color change and
typically occurs during the addition of thiols, so possibly the large amount of
propanethiol required more oleate.
Cholesterol concentration
Cholesterol, a common constituent of membranes, was added to the PC to
decrease membrane fluidity.69 Cholesterol was not previously required to create ion
impermeable hexanethiol stabilized PC-AuNP, but was necessary with the PT-PC-
AuNP.87 Natural vesicles exist with varying diameters typically > 26 nm, so forcing the
lipid layers onto solid supports that are smaller leads to an increased amount of strain.
Cholesterol has a small polar headgroup and large cone angle that if located on the inner
leaflet of the bilayer could help to reduce the strain associated with forming around the
AuNP.71 This would reduce the possible interstitial water layer volume and allow the
shorter propanethiol to assist in stabilizing the lipid bilayer.87 We suggest that in addition
to forming a less mobile bilayer the cholesterol also allowed for tighter packing of the
lipid around the AuNP.
43


Thiol concentration
The thiol anchor helps stabilize the ion impermeable lipid layers around the gold
core and determining the optimal concentration is important. When decanethiol was used
~6 thiols/surface gold atom were needed to prevent cyanide from oxidizing the gold core
of PC-AuNP.46 Yang and co-workers demonstrated the ability to create cyanide stable
lipid-coated gold nanoparticles with a octadecanethiol anchor using even lower amounts,
~2 thiol s/surface gold atoms.75 These reports lead to the suggestion that chain length
correlates to the amount of thiol required for ion impermeability; however, comparison to
previous work is complicated by the varying amount of surface area. More rigorous
analysis of surface area is required before direct comparisons of concentrations can be
made. As shown above, the amount of available surface area in this analysis was around
half that of previous work with smaller AuNP. This correlates poorly with the fact that
twice as much hexanethiol was needed to replicate those results. There are a variety of
factors to consider in this comparison including the modifications to the oleate
concentration and cholesterol content in the lipid layer that were not present in the
previous analysis. However, even higher amounts ~23 thiols/nm2 were required when
propanethiol was used as the anchor. We propose that higher concentrations are needed
for propanethiol because the shorter alkane chain has less ability to interact with the lipid
bilayer.
44


Binding studies of pCRP with PT-PC-AuNP
Quenching issues and BME
One of the first issues when assessing the binding of pCRP to PT-PC-AuNP by
fluorescence was the lack of observable signal because of the inner filter effects of the
AuNP. Separating the CRP from the AuNP without altering the isoform of the protein
became crucial for applying fluorescence quenching techniques to determine the isoform
of CRP. The main advantage of the PT-PC-AuNP over previous thiol stabilized PC-
AuNP is that the lipid layer can be disrupted with the addition of BME or DTT (Figure
8). The sulfur end of the BME has a high affinity for gold and the hydroxyl group on the
opposing end disrupts the tight lipid packing and allows for penetration of cyanide.
Because BME disrupted the hybrid lipid bilayer faster than DTT it was selected for the
lipid displacement in the binding studies. The extent to which the lipid bilayer is
detached is unknown, but it is possible that a significant portion of the PC membrane is
detached from the AuNP. The gel data confirmed that BME did not appear to affect
pCRP (Figure 10), and adding BME to all samples helped prevent multiple variable
analysis comparison issues.
Wang and co-workers also demonstrated that DTT does not affect the tertiary
structure of pCRP; however, they did show that the disulfide bonds in mCRP are reduced
when incubated with DTT, which reveals cholesterol-binding epitopes and implicated
reduced mCRP in proinflammatory activities. The reduced mCRP showed similar
tertiary structure and fluorescent properties to non-reduced mCRP.90 Agarose gels are
not sensitive enough to differentiate between reduced and non-reduced mCRP, so a
difference is not observed in the binding studies presented here.
45


Quenching with succinimide and acrylamide in buffer
Differentiating pCRP from mCRP and observing the isoform conversion are
important aspects of developing a more complete understanding of the reactivity of CRP.
Tryptophan quenching experiments and Stern-Volmer analysis has not been previously
utilized to differentiate the two isoforms. mCRP quenched by succinimide in buffer had
a KSv of 8.8 M'1 (Figure 11) and pCRP had a KSv of 2.9 M'1, which correlates with the
theory that pCRP has more interior tryptophan residues that are harder for a large, polar
quencher like succinimide to access. This leads to a lower KSv value because
succinimide would not be able interact with tryptophan residues that are located in
nonpolar, interior portions of proteins. This displays the capability of discriminating
between the two isoforms of CRP by a new method.
There are many possible explanations for the observed KSv values but they are
difficult to consider without mCRP being fully characterized. Additionally, modCRP and
forced conversion mCRP may not necessarily be identical despite having similar
reactivity.5,13 What is established is that the tryptophan residues of mCRP fluoresce
significantly less than pCRP likely due to solvent exposure and previous reports
determined the mCRP has an increased amount of a-helix portion relative to pCRP.90
Proteins frequently quench their own fluorescence and it is possible that during isoform
conversion mCRP became able to quench more of its own intrinsic fluorescence.78
Succinimide was selected as the optimal quencher of fluorescence over many
other quenchers including acrylamide. Acrylamide is one of the most commonly used
quenchers because of its high efficiency and using it avoids electrostatic problems
because it is an uncharged quencher.53,86 Acrylamide data was presented to represent the
46


issues that other quenchers had when attempting to differentiate the two isoforms of CRP.
A possible explanation for the inconsistency in this case may be due to the ability of
acrylamide to penetrate proteins, which has been highly susceptible to a variety of factors
dependent on the protein.8893 Differentiating the two isoforms of CRP in a logical
method is important when choosing a quencher.
Succinimide quenching with PT-PC-AuNP background
The goal of the fluorescence quenching studies was to establish a new method to
differentiate the two isoforms of CRP. Background changes can drastically affect the
quenching constants of protein/quencher pairs.53 Succinimide gave different absolute
values for the KSv of mCRP and pCRP depending on the background used for the
analysis. However, the ratio of KSvmCRP / KSvpCRP remained very similar (KSvmCRP /
KsvpCRP = 2.8 in buffer and KsvmCRP / KsvpCRP = 2.6 in PT-PC-AuNP supernatant
background), which indicates that the value of the information may not be lost due to
various background conditions (Table 1). CRP is a complex protein for quenching
analysis due to the large number of fluorophores and a thorough characterization of the
tryptophan residues was never the objective of this research. Effectively differentiating
mCRP from pCRP is the main aim and despite the KSv magnitude change the usefulness
of the resulting data does not decline. One possible explanation that exists is that the
increased complexity of the samples led to a more complex background and different
slopes.
47


BME and NMM
The addition of BME led to an increased amount of mCRP in the supernatant, but
degraded pCRP during gel electrophoresis. Importantly, the majority of the protein was
in the supernatants of the gel lanes and no longer in the suspended pellet lanes (Figure
15). This demonstrates the usefulness of BME especially in the samples containing
calcium, where binding occurred, because the mCRP is no longer in the pellet but has
been separated from the PT-PC-AuNP. The smearing in the lane containing EDTA,
where no isoform conversion previously occurred, was unfortunate but not
unprecedented. N-ethyl maleimide was previously utilized to prevent CRP degradation
during gel electrophoresis in experiments utilizing DTT.90 In this analysis, NMM was
utilized for the same purpose and as shown in Figure 16, was successful at preventing in-
gel degradation of pCRP.
It is reasonable to assume that the degradation of pCRP, in binding study samples
that contained EDTA, occurred during gel electrophoresis because the KSvmCRP / KSvpCRP
ratio without NMM or with a 1:1 BME:NMM was similar to the KSvmCRP / KSvpCRP in
buffer or with PT-PC-AuNP background (Tables 1 & 2). Isoform differentiation is
possible by KSvmCRP / KSvpCRP and is confirmed by gel electrophoresis in the presence of
1:1 BME:NMM. Intriguingly, the samples that do not contain EDTA (where calcium
dependent binding and isoform conversion occur) do not require NMM but do require
BME to separate the PT-PC-AuNP from the mCRP for fluorescence analysis.
Conversely, the samples that do contain EDTA (where pCRP remains in solution) do not
require BME, because separation of the lipid layer from the AuNP is not required, but do
need NMM to prevent smearing during gel electrophoresis. These problems could be
48


solved alternatively if BME was not added to samples containing EDTA and NMM was
not added to samples that do not contain EDTA.
49


Chapter 5. Conclusions
These studies indicate that intrinsic fluorescence quenching utilizing succinimide
and Stem-Volmer analysis can differentiate pCRP and mCRP. In order to separate the
mCRP from the AuNP, so fluorescence quenching can be employed, PT-PC-AuNP
needed to be fabricated. The lipid bilayer on the HT-PC-AuNP could not be disrupted by
BME or DTT, so a less stable, shorter alkanethiol anchors was used. The less stable
propanethiol anchor required higher concentrations of oleate, PC and thiol than HT-PC-
AuNP. PT-PC-AuNP also required cholesterol in the lipid layer in order to maintain tight
lipid packing demonstrated by stability against cyanide oxidation. Binding studies using
the PT-PC-AuNP allowed for the disruption of the hybrid lipid bilayer with the addition
of BME. This allowed for separation of mCRP from the AuNP and fluorescence
quenching differentiating the two isoforms of CRP without interfering inner filter effects.
This work demonstrates novel PT-PC-AuNP as well a new method for
differentiating mCRP and pCRP. PT-PC-AuNP is a cell mimic that has a displaceable
hybrid lipid bilayer when BME or DTT is added. This additional control gives future
investigators more options when modifications to the cell mimic are needed without
aggregation of the PC-AuNP. BME was determined to be more effective at disrupting
the bilayer than DTT because of the shorter time required to reach complete oxidation of
the AuNP. Displacing the lipid layer was essential for separating mCRP from the PC-
AuNP without allowing the interacting of CRP and the AuNP because bare AuNP can
cause isoform conversion. BME effectively detached the bound mCRP and lipid bilayer
and allowed for fluorescent quenching analysis. Succinimide was found to be the most
effective quencher for determining isoform despite some of its perceived shortcomings.
50


Established gel electrophoretic techniques confirmed the isoform conversion determined
by the quenching studies. The gel also indicated that a large portion of the mCRP was
removed from the AuNP because there was very little if any mCRP in the suspended
pellet lane. Future Clq studies would also help to confirm that NMM only prevents in
gel degradation of CRP and there are not other competing factors. Hopefully, a more
complete understanding of CRP could lead to a practical application in the prevention of
cardiovascular disease.
Cell mimics will continue to be useful tools for establishing models and isolating
specific interactions with proteins. Membranes are the interface of the exterior and
interior of cells and are very complex. Increasing the amount of possible modifications
and flexibility in membrane models will allow for a wider range of possible experiments.
Many other proteins bind to cell walls and examining these phenomenons in solution may
provide added insight to their mechanisms. Changing membrane composition would also
allow for further studies to examine and isolate necessary interactions. It is also possible
to change the composition of the nanoparticle in order to modify the size and shape of the
mimic desired.
51


REFERENCES
1. Shrive, A. K., Gheetham, G. M. T., Holden, D., Myles, D. A. A., Turned,
W. G., Volanakis, J. E., Pepys, M. B., Bloomer, A. C. & Greenhough, T.
J. (1996). Three Dimensional Structure of Human C-reactive Protein.
Nature Structural Biology 3, 346-354.
2. Moreland, J. L., Gramada, A., Buzko, O. V., Zhang, Q. & Bourne, P. E.
(2005). The Molecular Biology Toolkit (MBT): a modular platform for
developing molecular visualization applications. BMC Bioinformatics 6,
21.
3. Abernethy, T. J. & Avery, O. T. (1941). The occurrence during acute
infections of a protein not normally present in the blood: I. Distribution of
the reactive protein in patients' sera and the effect of calcium on the
flocculation reaction with c polysaccharide of pneumococcus. The Journal
of Experimental Medicine 73, 173-82.
4. Volanakis, J. E. (2001). Human C-reactive protein: expression, structure,
and function. Molecular Immunology 38, 189-97.
5. Eisenhardt, S. U., Thiele, J. R., Bannasch, H., Stark, G. B. & Peter, K.
(2009). C-reactive protein: how conformational changes influence
inflammatory properties. Cell Cycle 8, 3885-92.
6. Pepys, M. B. (1981). C-reactive protein fifty years on. Lancet 1, 653-7.
7. Pepys, M. B. & Hirschfield, G. M. (2003). C-reactive protein: a critical
update. The Journal of Clinical Investigation 111, 1805-1812.
8. Singh, S. K., Suresh, M. V., Voleti, B. & Agrawal, A. (2008). The
connection between C-reactive protein and atherosclerosis. Annals of
Medicine 40, 110-20.
9. Ridker, P. M., Hennekens, C. H., Buring, J. E. & Rifai, N. N. (2000). C-
Reactive Protein and Other Markers of Inflammation in the Prediction of
Cardiovascular Disease in Women. The New England Journal of Medicine
342, 836-843.
52


10. Ridker, P. M., Rifai, N., Rose, L., Buring, J. E. & Cook, N. R. (2002).
Comparison of C-reactive protein and low-density lipoprotein cholesterol
levels in the prediction of first cardiovascular events. The New England
Journal of Medicine 347, 1557-65.
11. Thompson, D., Pepys, M. B. & Wood, S. P. (1999). The physiological
structure of human C-reactive protein and its complex with
phosphocholine. Structure 7, 169-177.
12. Oliveira, E. B., Gotschlich, E. C. & Liu, T. Y. (1977). Primary structure of
human C-reactive protein. Proceedings of the National Academy of
Science of the United States of America 74, 3148-51.
13. Ji, S. R., Wu, Y., Zhu, L., Potempa, L. A., Sheng, F. L., Lu, W. & Zhao, J.
(2007). Cell membranes and liposomes dissociate C-reactive protein
(CRP) to form a new, biologically active structural intermediate:
mCRP(m). The Journal of the Federation of American Societies for
Experimental Biology 21, 284-94.
14. Potempa, L. A., Maldonado, B. A., Laurent, P., Zemel, E. S. & Gewurz,
H. (1983). Antigenic, electrophoretic and binding alterations of human C-
reactive protein modified selectively in the absence of calcium. Molecular
Immunology 20, 1165-75.
15. Potempa, L. A., Siegel, J. N., Fedel, B. A., Potempa, R. T. & Gewurz, H.
(1987). Expression, detection and assay of a neoantigen (Neo-CRP)
associated with a free, human C-reactive protein subunit. Molecular
Immunology 24, 531-541.
16. Kinoshita, C. M., Ying, S. C., Hugh, T. E., Siegel, J. N., Potempa, L. A.,
Jiang, H., Houghten, R. A. & Gewurz, H. (1989). Elucidation of a
protease-sensitive site involved in the binding of calcium to C-reactive
protein. Biochemistry 28, 9840-8.
17. Agrawal, A., Lee, S., Carson, M., Narayana, S. V., Greenhough, T. J. &
Volanakis, J. E. (1997). Site-directed mutagenesis of the phosphocholine-
binding site of human C- reactive protein: role of Thr76 and Trp67.
Annals of the New York Academy of Sciences 389, 235-250.
53


18. Volanakis, J. E. (1982). Complement activation by C-reactive protein
complexes. Annals New York Academy of Sciences 389, 235-50.
19. Richards, R. L., Gewurz, EL, Osmand, A. P. & Alving, C. R. (1977).
Interactions of C-reactive protein and complement with liposomes.
Proceedings of the National Academy of Sciences of the United States of
America 74, 5672-6.
20. Volanakis, J. E. & Narkates, A. J. (1981). Interaction of C-reactive protein
with artificial phosphatidylcholine bilayers and complement. The Journal
of Immunology 126, 1820-5.
21. Ji, S. R., Ma, L., Bai, C. J., Shi, J. M., Li, H. Y., Potempa, L. A., Filep, J.
G., Zhao, J. & Wu, Y. (2009). Monomeric C-reactive protein activates
endothelial cells via interaction with lipid raft microdomains. The Journal
of the Federation of American Societies for Experimental Biology 23,
1806-1816.
22. Gershov, D., Kim, S., Brot, N. & Elkon, K. B. (2000). C-Reactive protein
binds to apoptotic cells, protects the cells from assembly of the terminal
complement components, and sustains an antiinflammatory innate immune
response: implications for systemic autoimmunity. The Journal of
Experimental Medicine 192, 1353-64.
23. Volanakis, J. E. & Wirtz, K. W. (1979). Interaction of C-reactive protein
with artificial phosphatidylcholine bilayers. Nature 281, 155-7.
24. Kishore, U. & Reid, K. B. (2000). Clq: structure, function, and receptors.
Immunopharmacology 49, 159-70.
25. Richards, R. L., Gewurz, H., Siegel, J. & Alving, C. R. (1979).
Interactions of C-reactive protein and complement with liposomes. II.
Influence of membrane composition. The Journal of Immunology 122,
1185-9.
26. Roux, K. H., Kilpatrick, J. M., Volanakis, J. E. & Kearny, J. F. (1983).
Localization of the phosphocholine-binding sites on C-reactive protein by
immunoelectron microscopy. The Journal of Immunology 131, 2411-2415.
54


27. Mold, C., Gresham, H. D. & Du Clos, T. W. (2001). Serum amyloid P
component and C-reactive protein mediate phagocytosis through murine
Fc gamma Rs. The Journal of Immunology 166, 1200-5.
28. Chang, M. K. (2002). C-reactive protein binds to both oxidized LDL and
apoptotic cells through recognition of a common ligand:
Phosphorylcholine of oxidized phospholipids. Proceedings of the National
Academy of Sciences of the United States of America 99, 13043-13048.
29. Pentikainen, M. O., Oorni, K., Ala-Korpela, M. & Kovanen, P. T. (2000).
Modified LDL trigger of atherosclerosis and inflammation in the arterial
intima. Journal of Internal Medicine 247, 359-70.
30. Lima, E. S. & Maranhao, R. C. (2004). Rapid, simple laser-light-scattering
method for HDL particle sizing in whole plasma. Clinical Chemistry 50,
1086-8.
31. Singh, S. K., Suresh, M. V., Hammond, D. J., Rusinol, A. E., Potempa, L.
A. & Agrawal, A. (2009). Binding of the monomeric form of C-reactive
protein to enzymatically-modified low-density lipoprotein: effects of
phosphoethanolamine. Clinica Chimica Acta 406, 151-5.
32. Fu, T. & Borensztajn, J. (2002). Macrophage uptake of low-density
lipoprotein bound to aggregated C-reactive protein: possible mechanism of
foam-cell formation in atherosclerotic lesions. Biochemical Journal 366,
195-201.
33. Torzewski, J., Torzewski, M., Bowyer, D. E., Frohlich, M., Koenig, W.,
Waltenberger, J., Fitzsimmons, C. & Hombach, V. (1998). C-reactive
protein frequently colocalizes with the terminal complement complex in
the intima of early atherosclerotic lesions of human coronary arteries.
Arteriosclerosis, Thrombosis, and Vascular Biology 18, 1386-92.
34. Eisenhardt, S. U., Habersberger, J., Murphy, A., Chen, Y. C., Woollard,
K. J., Bassler, N., Qian, H., von Zur Muhlen, C., Hagemeyer, C. E.,
Ahrens, I., Chin-Dusting, J., Bobik, A. & Peter, K. (2009). Dissociation of
pentameric to monomeric C-reactive protein on activated platelets
localizes inflammation to atherosclerotic plaques. Circulation Research
105, 128-37.
55


35. Zhang, Y. X., Cliff, W. J., Schoefl, G. I. & Higgins, G. (1999). Coronary
C-reactive protein distribution: its relation to development of
atherosclerosis. Atherosclerosis 145, 375-9.
36. Ridker, P. M. (2008). High-sensitivity C-reactive protein as a predictor of
all-cause mortality: implications for research and patient care. Clinical
chemistry 54, 234-7.
37. Ridker, P. M. (2009). C-reactive protein: eighty years from discovery to
emergence as a major risk marker for cardiovascular disease. Clinical
chemistry 55, 209-15.
38. Pepys, M. B., Hirschfield, G. M., Tennent, G. A., Gallimore, J. R., Kahan,
M. C., Bellotti, V., Hawkins, P. N., Myers, R. M., Smith, M. D., Polara,
A., Cobb, A. J., Ley, S. V., Aquilina, J. A., Robinson, C. V., Sharif, I.,
Gray, G. A., Sabin, C. A., Jenvey, M. C., Kolstoe, S. E., Thompson, D. &
Wood, S. P. (2006). Targeting C-reactive protein for the treatment of
cardiovascular disease. Nature 440, 1217-21.
39. Ridker, P. M. (2007). C-reactive protein and the prediction of
cardiovascular events among those at intermediate risk: moving an
inflammatory hypothesis toward consensus. Journal of the American
College of Cardiology 49, 2129-38.
40. Khreiss, T., Jozsef, L., Potempa, L. A. & Filep, J. G. (2004).
Conformational rearrangement in C-reactive protein is required for
proinflammatory actions on human endothelial cells. Circulation 109,
2016-22.
41. Pasceri, V., Willerson, J. T. & Yeh, E. T. (2000). Direct proinflammatory
effect of C-reactive protein on human endothelial cells. Circulation 102,
2165-8.
42. Blake, G. J. & Ridker, P. M. (2001). Novel Clinical Markers of Vascular
Wall Inflammation. Circulation Research 89, 763-771.
43. Khreiss, T., Jozsef, L., Hossain, S., Chan, J. S., Potempa, L. A. & Filep, J.
G. (2002). Loss of pentameric symmetry of C-reactive protein is
associated with delayed apoptosis of human neutrophils. The Journal of
Biological Chemistry 277, 40775-81.
56


44. Agrawal, A., Simpson, M. J., Black, S., Carey, M. P. & Samols, D.
(2002). A C-reactive protein mutant that does not bind to phosphocholine
and pneumococcal C-polysaccharide. The Journal of Immunology 169,
3217-22.
45. Fendler, J. H. (1987). Atomic and molecular clusters in membrane
mimetic chemistry. Chemical Reviews 87, 877-899.
46. Sitaula, S., Mackiewicz, M. R. & Reed, S. M. (2008). Gold nanoparticles
become stable to cyanide etch when coated with hybrid lipid bilayers.
Chemical Communications, 3013.
47. Huang, C.-C., Yang, Z., Lee, K.-H. & Chang, H.-T. (2007). Synthesis of
Highly Fluorescent Gold Nanoparticles for Sensing Mercury(II).
Angewandte Chemie 119, 6948-6952.
48. Panigrahi, S., Basu, S., Praharaj, S., Pande, S., Jana, S., Pal, A., Ghosh, S.
K. & Pal, T. (2007). Synthesis and Size-Selective Catalysis by Supported
Gold Nanoparticles:,Aa Study on Heterogeneous and Homogeneous
Catalytic Process. The Journal of Physical Chemistry C 111, 4596-4605.
49. Templeton, A. C., Mostetler, M. J., Kraft, C. T. & Murray, R. W. (1998).
Reactivity of Monolayer-Protected Gold Cluster Molecules: Steric
Effects. Journal of the American Chemical Society 120, 1906-1911.
50. Frens, G. (1973). Controlled Nucleation for the Regulation of the Particle
Size in Monodisperse Gold Suspensions. Nature Physical Science 241, 20-
22.
51. Hutter, E. & Fendler, J. A. (2004). Exploitation of Localized Surface
Plasmon Resonance. Advanced Materials 16, 1685-1706.
52. Sepulveda, B., Angelome, P. C., Lechuga, L. M. & Liz-Marzan, L. M.
(2009). LSPR-based nanobiosensors. Nano Today 4, 244-251.
53. Lakowicz, J. R. (1999). Principles of Fluroescence Spectroscopy. 2 edit,
Kluwer Academic / Plenum Publishers, New York.
57


54. Agasti, S. S., You, C.-C., Arumugam, P. & Rotello, V., M. (2008).
Structural control of the monolayer stability of water-soluble gold
nanoparticles. Journal of Materials Chemistry 18, 70-73.
55. Kumar, A., Biebuyck, H. A., Abbott, N. L. & Whitesides, G. M. (1992).
The use of self-assembled monolayers and a selective etch to generate
patterned gold features. Journal of the American Chemical Society 114,
9188-9189.
56. Gopidas, K. R., Whitesell, J. K. & Fox, M. A. (2003). Nanoparticle-cored
dendrimers: synthesis and characterization. Journal of the American
Chemical Society 125, 6491-502.
57. Hong, R., Fernandez, J. M., Nakade, H., Arvizo, R., Emrick, T. & Rotello,
V. M. (2006). In situ observation of place exchange reactions of gold
nanoparticles. Correlation of monolayer structure and stability. Chemical
Communications, 2347-9.
58. Richter, R. P., Berat, R. & Brisson, A. R. (2006). Formation of Solid-
Supported Lipid Bilayers: An Integrated View. Langmuir 22, 3497-3505.
59. Voet, D., Voet, J. G. & Pratt, C. W. (2008). Fundamentals of
Biochemistry: Life at the Molecular Level. 3rd edit, Wiley.
60. Nagle, J. F. & Tristram-Nagle, S. (2000). Structure of lipid bilayers.
Biochim Biophys Acta 1469, 159-95.
61. Tanaka, M. & Sackmann, E. (2005). Polymer-supported membranes as
models of the cell surface. Nature 437, 656-663.
62. Moyano, D. F. & Rotello, V. M. (2011). Nano meets biology: structure
and function at the nanoparticle interface. Langmuir 27, 10376-85.
63. Groves, J. T. & Boxer, S. G. (2002). Micropattem Formation in Supported
Lipid Membranes. Accounts of Chemical Research 35, 149-157.
64. Sundh, M., Svedhem, S. & Sutherland, D. S. (2011). Formation of
Supported Lipid Bilayers at Surfaces with Controlled Curvatures:
58


Influence of Lipid Charge. The Journal of Physical Chemistry B 115,
7838-7848.
65. Roiter, Y., Omatska, M., Rammohan, A. R., Balakrishnan, J., Heine, D. R.
& Minko, S. (2009). Interaction of Lipid Membrane with Nanostructured
Surfaces. Langmuir 25, 6287-6299.
66. Mi, L. Z., Wang, H. W. & Sui, S. F. (1997). Interaction of rabbit C-
reactive protein with phospholipid monolayers studied by
microfluorescence film balance with an externally applied electric field.
Biophysical Journal 73, 446-51.
67. Mackiewicz, M. R., Ayres, B. R. & Reed, S. M. (2008). Reversible,
reagentless solubility changes in phosphatidylcholine-stabilized gold
nanoparticles. Nanotechnology 19, 115607.
68. Baksh, M. M., Jaros, M. & Groves, J. T. (2004). Detection of molecular
interactions at membrane surfaces through colloid phase transitions.
Nature 427, 139-41.
69. Yeagle, P. L. (1991). Modulation of membrane function by cholesterol.
Biochimie 73, 1303-10.
70. Spooner, P. J. & Small, D. M. (1987). Effect of free cholesterol on
incorporation of triolein in phospholipid bilayers. Biochemistry 26, 5820-
5.
71. Deng, Y., Wang, Y., Holtz, B., Li, J., Traaseth, N., Veglia, G., Stottrup, B.
J., Elde, R., Pei, D., Guo, A. & Zhu, X. Y. (2008). Fluidic and air-stable
supported lipid bilayer and cell-mimicking microarrays. Journal of the
American Chemical Society 130, 6267-71.
72. He, P. & Urban, M. W. (2005). Phospholipid-stabilized Au-nanoparticles.
Biomacromolecules 6, 1224-5.
73. Plant, A. L. (1999). Supported Hybrid Bilayer Membranes as Rugged Cell
Membrane Mimics. Langmuir 15, 5128-5135.
59


74. Kundu, J., Levin, C. S. & Halas, N. J. (2009). Real-time monitoring of
lipid transfer between vesicles and hybrid bilayers on Au nanoshells using
surface enhanced Raman scattering (SERS). Nanoscale 1, 114-7.
75. Yang, J. A. & Murphy, C. J. (2012). Evidence for Patchy Lipid Layers on
Gold Nanoparticle Surfaces. Langmuir.
76. Thaxton, C. S., Daniel, W. L., Giljohann, D. A., Thomas, A. D. & Mirkin,
C. A. (2009). Templated spherical high density lipoprotein nanoparticles.
Journal of the American Chemical Society Communications 131, 1384-5.
77. Chen, Y. & Barkley, M. D. (1998). Toward Understanding Tryptophan
Fluorescence in Proteins. Biochemistry 37, 9976-9982.
78. Chou, C., Hsu, H. Y., Wu, H. T., Tseng, K. Y., Chiou, A., Yu, C. J., Lee,
Z. Y. & Chan, T. S. (2007). Fiber optic biosensor for the detection of C-
reactive protein and the study of protein binding kinetics. Journal of
Biomedical Optics 12, 024025.
79. Guy, J., Caron, K., Dufresne, S. p., Michnick, S. W., Skene & Keillor, J.
W. (2007). Convergent Preparation and Photophysical Characterization of
Dimaleimide Dansyl Fluorogens: Elucidation of the Maleimide
Fluorescence Quenching Mechanism. Journal of the American Chemical
Society 129, 11969-11977.
80. Popov, A. V., Gladkikh, V. S. & Burshtein, A. I. (2003). Stern-Volmer
Law in Competing Theories and Approximations. The Journal of Physical
Chemistry A 107, 8177-8183.
81. Eftink, M. R. & Ghiron, C. A. (1981). Fluorescence quenching studies
with proteins. Analytical Biochemistry 114, 199-227.
82. Eftink, M. R. & Ghiron, C. A. (1984). Indole fluorescence quenching
studies on proteins and model systems: use of the inefficient quencher
succinimide. Biochemistry 23, 3891-3899.
83. Badley, R. A. (1975). The location of protein in serum lipoproteins: a
fluorescence quenching study. Biochimica et Biophysica Acta 379, 517-
28.
60


84. Szabo, A. G., Stepanik, T. M., Wayner, D. M. & Young, N. M. (1983).
Conformational heterogeneity of the copper binding site in azurin. A time-
resolved fluorescence study. Biophysical Journal 41, 233-44.
85. Avigliano, L., Finazzi-Agro, A. & Mondovi, B. (1974). Perturbation
studies on some blue proteins. The Journal of the Federation of American
Societies for Experimental Biology Letters 38, 205-208.
86. Komiyama, T. & Miwa, M. (1980). Fluorescence quenching as an
indicator for the exposure of tryptophyl residues in Streptomyces subtilisin
inhibitor. The Journal of Biochemistry 87, 1029-36.
87. Mackiewicz, M. R., Hodges, H. L. & Reed, S. M. (2010). C-reactive
protein induced rearrangement of phosphatidylcholine on nanoparticle
mimics of lipoprotein particles. Journal of Physical Chemistry B 114,
5556-62.
88. Calhoun, D. B., Vanderkooi, J. M., Holtom, G. R. & Englander, S. W.
(1986). Protein fluorescence quenching by small molecules: protein
penetration versus solvent exposure. Proteins 1, 109-15.
89. Skoog, D. A., West, D. M., Holler, F. J. & Crouch, S. R. (2004).
Fundamentals of Analytical Chemistry. 8 edit, Thomson / Brooks / Cole,
Belmont.
90. Wang, M. Y., Ji, S. R., Bai, C. J., El Kebir, D., Li, H. Y., Shi, J. M., Zhu,
W., Costantino, S., Zhou, H. H., Potempa, L. A., Zhao, J., Filep, J. G. &
Wu, Y. (2011). A redox switch in C-reactive protein modulates activation
of endothelial cells. The Journal of the Federation of American Societies
for Experimental Biology 25, 3186-96.
91. Basu, S., Pande, S., Jana, S., Bolisetty, S. & Pal, T. (2008). Controlled
Interparticle Spacing for Surface-Modified Gold Nanoparticle Aggregates.
Langmuir 24, 5562-5568.
92. Liu, X., Atwater, M., Wang, J. & Huo, Q. (2007). Extinction coefficient of
gold nanoparticles with different sizes and different capping ligands.
Colloids Surfaces B: Biointerfaces 58, 3-7.
61


93. Friedman, M. L., Schlueter, K. T., Kirley, T. L. & Halsall, H. B. (1985).
Fluorescence quenching of human orosomucoid. Accessibility to drugs
and small quenching agents. Biochemical Journal 232, 863-7.
62


Full Text

PAGE 1

BINDING STUDIES OF C REACTIVE PROTEIN TO LIPID COATED GOLD N ANOPARTICLES AND ISOFORM DIFFERENTIATION BY QUENCHING EXPERIMENTS by Reid Elliott Messersmith B.S. Chemistry, Pepperdine University, 2010 A thesis submitted to the University of Colorado Denver in partial fulfillment of the requirements for the degree of Master of Science Chemistry 2012

PAGE 2

ii This thesis for the Master of Science degree by Reid Elliott Messersmith has been approved for by Scott Reed Mark Anderson Jefferson Knight Date: April 5 th 2012

PAGE 3

iii Messersmith, Reid, Elliott (M.S., Chemistry) Binding Studies Of C Reactive Protein To Lipid Coated Gold N anoparticles And Isoform Differentiation By Quenching Experiments Thesis directed by Professor Scott Reed. ABSTRACT C reactive protein (CRP) is an acute phase protein that has been implicated in cardiovascular disease. CRP exists in serum in a pentameric form, but upon binding to phosphatidylcholine on damaged cell membranes it dissociates into a biologically active form: modified CRP. Lipid coated gold nanoparticles were synthesized as membrane mimics so the effects of mem brane structure on CRP binding could be observed. Stable mimics were created with propanethiol anchors and a mixed bilayer containing phosphatidylcholine and cholesterol surro unding the gold nanoparticle. mercaptoethanol was utilized to separate the hy brid lipid bilayer with the bound CRP from the gold nanoparticles allowing for analysis of the CRP tryptophan fluorescence. Conformational changes in CRP on binding to PC AuNP were studied by quenching the intrinsic tryptophan fluorescence of the protein with succinimide and gel assays. This abstract accurately represents the contents of the candidate's thesis. I recommend its publication. Approved : Scott Reed

PAGE 4

iv ACKNOWLEDGMENTS I would like to thank all the m embers of the Reed lab for their help and support. In particular, I appreciate Scott Reed for allowing me to work in his lab and his vision for this project. Min Wang for her laboratory expertise and help as a knowledgeable resource. Audrea Pi per for being a n accommodating lab mate and continued assistance throughout the project. Additionally, I would like to acknowledge my friends and family for their support during this journey.

PAGE 5

v TABLE OF CONTENTS LIST OF TABLES ................................ ................................ ................................ ........... viii LIST OF FIGURES ................................ ................................ ................................ ........... ix LIST OF SCHEMES ................................ ................................ ................................ ........ xii LIST OF ABBREVIATIONS ................................ ................................ .......................... xiii CHAPTER 1. INTRODUCTION ................................ ................................ ................................ .......... 1 Introduction to C reactive protein ................................ ................................ ....................... 1 CRP structure ................................ ................................ ................................ ...................... 2 Ligands of CRP ................................ ................................ ................................ ................... 3 Function of C reactive protein ................................ ................................ ............................ 4 Cardiovascular disease ................................ ................................ ................................ ........ 6 Introduction to lipid coated gold nanoparticles ................................ ................................ .. 8 AuNP and Thiols ................................ ................................ ................................ ................. 9 Biological Membranes and Supported Surfaces ................................ ............................... 11 Lipid coated Nanoparticles ................................ ................................ ............................... 13 Lipid coated AuNP ................................ ................................ ................................ ........... 14 Introduction to Fluorescence ................................ ................................ ............................. 16 Fluorescence Quenching ................................ ................................ ................................ ... 17 Tryptophan Fluorescence and Quenching ................................ ................................ ........ 19 Binding studies of CRP to PT PC AuNP ................................ ................................ ......... 21 2. EXPERIMENTAL ................................ ................................ ................................ ........ 23

PAGE 6

vi Materials ................................ ................................ ................................ ........................... 23 Gold nanoparticle preparation ................................ ................................ ........................... 23 Lipid coated, thiol anchored AuNP synthesis ................................ ................................ .. 23 CRP binding studies with PT PC AuNP ................................ ................................ .......... 24 Monomeric CRP preparation ................................ ................................ ............................ 25 Gel electrophoresis ................................ ................................ ................................ ............ 25 Fluorescence analysis ................................ ................................ ................................ ........ 26 Fluorescence quenching studies ................................ ................................ ........................ 26 Cyanide stability studies ................................ ................................ ................................ ... 27 3. RESULTS ................................ ................................ ................................ ..................... 28 Formation of PT PC AuNP ................................ ................................ .............................. 28 AuNP characterization ................................ ................................ ................................ ...... 28 Maximizing ion impermeability of PT PC AuNP ................................ ............................ 28 Binding studies of CRP with PT PC AuNP ................................ ................................ ..... 31 Lipid disruption utilizing BME or DTT ................................ ................................ ............ 32 Quenching Experiments ................................ ................................ ................................ .... 34 Quenching in buffer by succinimide and acrylamide ................................ ....................... 34 Succinimide quenching in PT PC AuNP background ................................ ...................... 36 Binding Studies ................................ ................................ ................................ ................. 37 Utilizing BME to disrupt PT PC AuNP ................................ ................................ ........... 37 Sequestering of BME by NMM ................................ ................................ ........................ 39 4. Discussion ................................ ................................ ................................ ..................... 41 PT PC AuNP preparation ................................ ................................ ................................ 41 AuNP size and shape determination ................................ ................................ ................. 41

PAGE 7

vii Oleate concentration ................................ ................................ ................................ ......... 42 Cholesterol concentration ................................ ................................ ................................ 43 Thiol concentration ................................ ................................ ................................ ........... 44 Binding studies of pCRP with PT PC AuNP ................................ ................................ ... 45 Quenching issues and BME ................................ ................................ .............................. 45 Quenching with succinimide and acrylamide in buffer ................................ .................... 46 Succinimide quenching with PT PC AuNP backg round ................................ .................. 47 BME and NMM ................................ ................................ ................................ ................ 48 5. Conclusions ................................ ................................ ................................ ................... 50 REFERENCES ................................ ................................ ................................ ................. 52

PAGE 8

viii LIST OF TABLES Table 1. Summarizes succinimide fluorescence quenching data. Buffer background refers to quenching fluorescence in 10 mM phosphate buffer, 140 mM NaCl and 2.5 mM CaCl 2 PT PC AuNP supernatant background refers to making PT PC AuNP, centrifuging at 15,000 rpm for 20 minutes, and adding mCRP, pCRP and tryptophan to the supernatant. ................................ ................................ ................................ ................. 34 Table 2. Succinimide quenching data of binding studies samples containing CRP incubation with PT PC AuNP and BME The 1:1 BME:NMM also contains 1 mM NMM and all samples were centrifuged before quenching analysis. ............................... 37

PAGE 9

ix LIST OF FIGURES Figure 1. face of c reactive protein viewed down the five fold symmetry axis. The PC binding site is located near the calcium ions (green) and the pocket for the choline group. 1; 2 ................................ ................................ ................................ ............................... 1 Figure 2. face of c reactive protein viewed down the five fold symmetry axis. C1q binding site located in cleft near helix (green). 1; 2 ................................ ........................... 2 Figure 3. Fluorescent tryptophan residues in c reactive protein shown in green. There are 6 tryptophan residues per monomer, 30 tryptophan residues total. 1; 2 ......................... 16 Figure 4. The absorbance retained after 24 hours of incubation with cyanide compared with the concentration of oleate used to prevent aggregation. The error bars are the standard deviation of at least 3 trials. ................................ ................................ ................ 28 Figure 5. The absorbance retained after 24 hour of incubation with cyanide plotted against the percentage of lipid layer that was cholesterol. The remainder of the lipid layer was always PC and the error bars are the standard deviation of at least 3 trials. ............. 29 Figure 6. The amount of nanomoles of propanethiol injected into a 1 mL solution of PC AuNP compared with absor bance retained after 24 hours of incubation with cyanide. The error bars are the standard deviation of 10 trials. ................................ .............................. 30 Figure 7. Gel electrophoresis and Western bl otting were preformed on the supernatant (S) and suspended pellet (P) of samples that did and did not contain EDTA. Controls of pCRP and mCRP were run for comparison. ................................ ................................ ..... 31 Figure 8. All samples contained PT PC AuNP and cyanide. One sample contained 1 mM BME ( ), another 1 mM DTT ( ) and the last did not contain any other reactants ( # ). The error bars are the standard deviation of 10 trials. ................................ ................ 32 Figure 9. All samples contained HT PC AuNP and cyanide. One sample contained 1 mM BME ( ), another 1 mM DTT ( ) and the last did not contain any other reactants ( # ). The error bars are the standard deviation of 7 trials. ................................ .................. 32 Figure 10. Gel electrophoresis and Western blotting of samples containing BME and pCRP, DTT and pCRP, pCRP and mCRP. ................................ ................................ ....... 33 Figure 11. Stern Volmer plot of pCRP ( # ), mCRP ( ) and tryptophan ( ) quenched by succinimide in 10 mM phosphate buffer, 140 mM NaCl and 2.5 mM CaCl 2 The equation of the best fit line for tryp: y = 0.0135x + 1.010; mCRP: y = 0.0088x + 1.044; pCRP: y = 0.0031 + 0.975. ................................ ................................ ............................... 34 Figure 12. Stern Volmer plot of pCRP ( # ), mC RP ( ) and tryptophan ( ) quenched by acrylamide in 10 mM phosphate buffer, 140 mM NaCl and 2.5 mM CaCl 2 The equation

PAGE 10

x of the best fit line for tryp: y = 0.0264x + 0.0660; mCRP: y = 0.0010x + 1.044; pCRP: y = 0.0052 + 0.926. ................................ ................................ ................................ .............. 35 Figure 13. Stern Volmer plot of pCRP ( # ), mCRP ( ) and tryptophan ( ) quenched by succinimide in the supernatant of PT PC AuNP. The equation of the best fit line for tryp: y = 0.0078x + 1 .056; mCRP: y = 0.0029x + 1.057; pCRP: y = 0.0011 + 1.011. ..... 36 Figure 14. Stern Volmer plots of the succinimide quenching of the supernatants of the binding studies. One sample contained calcium ( ) and isoform conversion occurred (y = 0.0012x + 1.025). The other sample contained EDTA ( # ) and binding did not occur (y = 0.005x + 1.0103). ................................ ................................ ................................ ........... 37 Figure 15. Western blot of supernatants of samples where CRP was incubated with PT PC AuNP and then BME before centrifugation. ................................ .............................. 38 Figure 16 Western blot of binding studies of CRP with PT PC AuNP, after incubation with 1 mM BME, 23 mM NMM and centrifugation. The supernatant (S) and pellet (P) of samples containing EDTA and not containing EDTA are shown. ................................ ... 39 Figure 17. Adapted from Basu and co workers. # max is plotted against the diameter, determined by TEM, of AuNP also synthesized by Fren's method (! ). The average # max for the AuNP created by this analyst ( $ ) is plotted to show the expected corresponding size. ................................ ................................ ................................ ................................ ... 41

PAGE 11

xi LIST OF EQUATIONS Equation 1. Oxidation of gold(0) to go ld(1) by cyanide. 10 Equation 2. Stern Volmer equation where F 0 is the initial fluorescence, F is the fluorescence with a certain concentration of quencher [Q] and K SV is the quenching constant ..17 Equation 3 The effect of efficiency ( $ ) on the value of the K SV ..18 Equation 4. Correlation between extinction coefficient ( % ) and AuNP diameter (d) with two constants k and a .41

PAGE 12

xii LIST OF SCHEMES Scheme 1. Displacement of thiol coated gold nanoparticles by a shorter thiol. .............. 10 Scheme 2. One possible mechanism for lipid deposition onto planar surfaces. .............. 11 Scheme 3. Oleate coated gold nanoparticles are incubated with PC liposomes. The thiol anchor is added to stabilize the hybrid lipid bilayer against cyanide penetration. ........... 14 Scheme 4. Propanethiol stabilized PC AuNP become completely cyanide instable after 2 hours when mixed with BME. ................................ ................................ .......................... 22 Scheme 5. Isoform conversion as a result of calcium dependent binding of c reactive protein to PT PC AuNP. Removal of the hybrid lipid bilayer by BME for fluorescence quenching analysis. ................................ ................................ ................................ ........... 22

PAGE 13

xiii LIST OF ABBREVIATIONS CRP C reactive protein pCRP Pentameric c reactive protein mCRP Monomeric c reactive protein AuNP Gold nanoparticles PC Phosphatidylcholine PC AuNP Lipid coated gold nanoparticles PT PC AuNP Propanethiol stabilized lipid coated gold nanoparticles HT PC AuNP Hexanethiol stabilized lipid coated gold nanoparticles BME mercaptoethanol DTT Dithiothreitol LDL Low density lipoproteins EDTA Ethylenediaminetetraacetic acid SDS Sodium dodecyl sulfate LSPR Localized surface plasmon resonance DLS Dynamic light scattering K SV Stern Volmer quenching constant

PAGE 14

1 CHAPTER 1 INTRODUCTION Introduction to C reactive protein C reactive protein (CRP) was discovered in 1930 when it precipitated with Streptococcus pneumoniae 3 ; 4 The unknown substance was labeled "Fraction C" because of its reactivity to c polysaccharide, which is what led to the naming of the protein. 5 CRP is a prototypic acute phase protein that increases in concentration due to a number of stimuli and is involved with the innate immune system. 6 CRP has a low basal level (1 mg/L) but can increase 500, 1,000, or even 10,000 times in 48 hours in response to infection, inflammation o r tissue damage. 5 ; 7 High baseline levels of CRP indicate systemic inflammation and when linked with cholesterol concentrations can predict future cardiovascular events. 8 Ridker and co workers showed that t he most potent predictor of future cardiovascular events is CRP levels independent of cholesterol concentration 9 ; 10 CRP is produced mainly by hepatocytes in the liver, more specifically in the endoplasmic reticulum and is transcriptionally controlled by cytokine interl e ukin 6. CRP has no known deficiencies or polymorphism in humans and has high phylogenetic conservation. 8 Figure 1 face of c reactive protein viewed down the five fold symmetry axis. The PC binding site is located near the calcium ions (green) and the pocket for the choline group. 1 ; 2

PAGE 15

2 CRP structure CRP is a member of the pentraxin family and is composed of five identical, non covalently linked subunits totaling 115,135 Da. 1 Each protomer or monomer has a molecular weight of 23,027 Da and consists of 206 amino acids composed in two antiparallel sheets with flattened jellyroll topology. 1 The pentamer has an overall diameter ~102  with a central pore diameter ~30  and each monomer has a diameter ~36 . 4 Each monomer also contains two calcium binding site s (Figure 1) on one face and a Fc receptor /C1q binding site on the opposing side (Figure 2) The face binds in a calcium dependent manner to phosphatidylcholine (PC) while the face binds to Fc receptors and is the C1q binding site conditionally. 11 It is unclear exactly what happens to serum or pentameric C reactive protein (pCRP) when binding occurs to phosphocholine but it is clear that new epitopes are exposed that can activate the classical complement pathway or increase p hagocytosis. 12 The lipid bound protein is referred to as modified C reactive protein (m od CRP) and is biologically active. 13 Forced conver sion utilizing a variety of different denaturants also leads to a biologically active form known as monomeric C reactive protein (mCRP) 14 ; 15 The PC binding face contains two calcium ions 4  from each other and both are involved with ligand binding. The phosphate group on PC co ordinates with the bound calcium ions and the choline group rests in a hydrophobic pocket. 11 ; 16 The Figure 2 face of c reactive protein viewed down the five fold symmetry axis. C1q binding site located in cleft near helix (green). 1 ; 2

PAGE 16

3 hydrophobic pocket has bee n shown to be imperative for PC binding using site directed mutagenesis. Replacement of the less polar amino acids in the pocket greatly decreases mutant CRP's attraction to PC. 17 C1q and Fc receptor both bind on the face of CRP but it is uncertain if they utilize the same or distinct bin ding site s The face contains an unusual cleft, near the helix, that is the binding site for C1q and possibly the binding site Fc receptors. 4 The cleft is near the c terminus of the protein and mutational data has confirmed specific residues related to C1q activation. Specifically, Glu 88 seems to assist in the conformational change of C1q that activates the complement system. 18 The Fc receptor binding site has not been as well charac terized as the C1q site, but it is suspected to interact with CRP in the same cleft. Ligands of CRP Due to steric hindrance it is unlikely that one pCRP can have multipoint attachment to a single lipid source and activate C1q Previously, it was believed that activation was caused by multiple pCRP substituents binding to one substrate in order to become active 19 For example, six pCRP molecules could bind near each other and together activate one C1q molecule. An alternative theory is that upon PC binding the monomers of pCRP break apart and rotate. 20 This would allow for multipoint attachment and enhanced signaling from a singl e molecule. 13 It would also explain why bound CRP allows for C1q binding and inc reased phagocytosis, which seru m pentameric CRP does not induce. 5 The latter theory also explains wh y forcibly monomerized CRP, in vitro can activate the complement pathway or phagocytosis. 21

PAGE 17

4 Pentameric dissociation into mCRP is accomplished in vitro by three major pathways. 1) Monomerization also occurs when pCRP is heated in the presence of urea and ethylenediaminetetraacetic acid ( EDTA ). 2) pCRP has also been shown to convert to mCRP utilizing sodium dodecyl sulfate (SDS) and heat. 3) pCRP can be change to mCRP when treated with acid and then neutralized. Addiontally, pCRP i n the absence of calcium will decompose in prolonged storage into mCR P. The urea/EDTA method has been the most widely used and established method for creating mCRP; however, complete characterization of mCRP has not yet been published. 14 ; 15 pCRP bound to membranes that has become biologically active (modCRP) and forced isoform conversion (mCRP) may be identical, but this has not yet been unambiguously confirmed because neither has been fully characterized. Function of C reactive protei n CRP's exact biological function is unknown; however, CRP is able to recognize damaged cells, pathogens and remove them by activation of complement or uptake by phagocytic cells. 5 ; 22 PC is present on the outer leaflet of most cell membranes so it may seem surprising at first that CRP does not bind to all mammalian cells. CRP is only found on apoptotic cell walls and it has been demonstrated that CRP only binds to lipids containing some amount of lysophospholipids. It has been suggested that cell damage causes membrane flip flop or exposure of hydrophobic tail groups of PC allows for CRP binding. 23 Once PC bound CRP recruits C1q and C3 convertase is effectively initiated through the classical complement pathway. 24 For successful activation of C1q, either multiple CRP molecules must be bound in a localized area or the CRP monomers of a

PAGE 18

5 single molecule must rotate and separate. 25 This is another piece of evidence to support the alternative rotation and dissociation model, of CRP binding to PC. According to this model, only one C1q binding is available for pCRP due to steric hinderance However, if ro tation and dissociation occu rs on lipid membranes then fewer CRP molecules could activate classical complement pathway 5 CRP binds PC in a calcium dependent manner 26 Only PC bound CRP activates complement or can bind to the Fc #RI and Fc #RII receptors on phagocytic cells. 5 It was found that mice with Fc # R I and Fc #RII deficiencies were not able to uptake CRP bound apoptotic cells in macrophages. 27 Along with a parallel study involving monocytic cell lines expressing Fc #RI and Fc #RII confirmed that Fc #RI and Fc #RII are receptors for CRP. 27 There have been multiple conflicting studies regarding CRP uptake by macrophages in the past and a possible explanation for this is the expression of different FcR receptors leading to inhibitory or activation signals. CRP acts as an opsonin for particles and leads to uptake by white blood cells through the FcR $ I receptors. 22 CRP mediated opsonization and uptake has been shown to occur with bacteria as well as apoptotic cells. 28 Free, serum p CRP does not interact with either phagocytic cells or activate complement. This indicates that serum CRP is a proprotein that requires binding to PC to become biologically active. 21 Cholesterol is transported around the human body in lipoprotein particles to provide for cells 29 High density lipoproteins carry cholesterol from peripheral cells and carry it back to the liver and typically have diamet ers from 5 15 nm. 30 Low density lipoproteins carry cholesterol from the liver to cells and have diameters around 15 30 nm. CRP does not bind to native LDL but does bind with oxidized or enzymatically

PAGE 19

6 modified LDL. 31 Modified LDL bound by CRP is then consumed by a macrophage to create foam cells. 32 ; 33 Foam cells deposit in arterial walls and contribute to atherosclerosis, which can lead to myocardial infarction. 34 ; 35 Ridker and co workers have employed t he concentration of CRP in serum as a predictor of risk factor of future cardiovascular events 32 ; 36 ; 37 Cardiovascular disease CRP may have a causal role in inflammation leading to arteriosclerotic lesions. 10 ; 38 Previously, LDL cholesterol concentrations were the main diagnostic tool for establishing risk of future coronary event s ; however, as high sensitivity assays for CRP were develop ed CRP levels were found to have more prognostic value for determining future risk. 36 For example, half of all acute ischemic events occur in patients without hyperlipidemia indicating that LD L concentrations are not necessarily the best method for predicting future events. 39 Fortunately, statins, commonly used to lower LDL concentrations, also lower CRP concentrations even in patients were LDL concentrations were already low. 37 The risk of future events in patients with low LDL and high CRP leve ls was higher than that of patients who had high LDL and low CRP levels confirming that CRP is a better predictor of cardiovascular events. 37 Currently, it is still unclear if CRP's role in ca rdiovascular disease is simply as a marker or if high CRP levels are a cause of cardiovascular events. 39 There have been many contradictory results in this area of study because the two isoforms of CRP (monomeric and pentameric) behave differently in respect to their atherogenicity. This role for a member of the innate immune system was not initially agreed upon because it opposes the findings of other host defense molecules. However, differentiating between

PAGE 20

7 isoforms in vivo is still not well established and is the likely cause of the inconsistent results. 40 pCRP has been reported to have pro and anti inflammatory characteri stics pCRP upregulated the expression of adhesion molecules to endothelial cells and stirred monocyte chemotactic protein release both of which are precursors to inflammation. 41 pCRP also provoked cytokine release including : interleukin 1, 6, 8, 18 and tumor necrosis factor 42 Conversely, pCRP has also been shown to have vasculoprotective potential. pCRP induces expression of the anti inflammatory interleukin 10 as well as interleukin 1 receptor antagonist. 43 The above findings have been questioned because of the presence of lipopolysaccharide and sodium azide as part of the experimental procedures. 5 These contaminants could play a part in the inconsistent results observed. Additionally, stored pCRP spontaneously converts to mCRP over extended periods of time especially in solutions lacking calcium. 14 When contaminants and monomerization were controlled pCRP has since been shown to have anti inflammatory properties. 5 pCRP inhibited platelet aggregation and did n ot prompt cytokine release or adhesion molecule expression. These new studies indicate that pCRP does not have pro inflammatory effects and may even have anti inflammatory potential 40 Utilizing mice as model for studies involving CRP has also lead to contradictory evidence regarding the involvement of CRP in atherogenesis. Mice do have a small quantity of serum CRP but it does not behave in the same manner as human CRP does in humans 5 Human CRP expressed in mice was shown to protect against certain infections and it was assumed to proceed by the classical complement pathway. 44 However, when testing was expanded to include atherogenic studies the results were inconclusive. pCRP

PAGE 21

8 was de monstrated to increase, decrease or not effect the development of atherosclerotic plaque. 8 Contradicting previous results, it was demonstrated that human CRP did not activate mouse compliment (in studies not using sodium azide); therefore, mouse models are likely not ideal for extrapolation to humans when mice are injected with human CRP. Rat models have provided much more reliable data when examining stroke and myocardial infarction because human CRP activates rat complement When human CRP was injected into rats there was a clear and consistent development of atherosclerotic plaque. 38 This evidence supports the idea that CRP plays a contributory role in atherosclerotic lesions. 5 Effectively differentiating mCRP and pCRP has been of the utmost importance, as more evidence emerges indicating that protein isoform deter mines functionality. mCRP epitopes are revealed in vivo when pCRP interacts with lysophosphatidylcholine, which is readily found on activated platelets and apoptotic cells. 43 Interaction with this lipid facilitates isoform conversion from serum pCRP to pro inflammatory mCRP. mCRP has not been detected in circulation but has been detected deposited in atherosclerotic lesions. 34 These pieces of evidence support the theory that mCRP is formed when pCRP binds to specific substrates. Introduction to lipid coated gold nanoparticles Biological systems are immensel y complex and isolating and identifying specific interactions is quite challenging. Cell mimics were employed t o reduce the number of possible complications. 45 A gold nanoparticle core was chosen because of its easily tunable size allowing it to be made to replicate LDL. Additionally, cyanide stability tests can be used to determine if an ion impermeable lipid layer surrounds all of the g old

PAGE 22

9 nanoparticles (AuNP). 46 This prevents the gold core from directly interacting with CRP and eliciting an unwanted response. AuNP and T hiols Functionalized nanoparticles have been a focus of much research beca use of their tunable properties and promise in catalysis, biosensing, electronics and nanomedicine. 47 ; 48 Modifications to AuNP have been widely studied, especially thiol amended because of the strong affinity between sulfur and gold. There are a variety of ligands that can be attached to gold because of this interaction; therefore, an immense amount of applications are possible with functionalized AuNP 49 Nanoparticle size is also an important factor in determining the possible functions of the complex. In 1973, Fren described a method for reducing gold( III) with varying concentrations of citrate in order to produce uniform AuNP of different sizes. 50 These gold cores have been augmented in an array of fashions by various methods but frequently utilizing thiols. AuNP appear red or pink in solution with a size and shape dependent light absorption. The absorption is described as localized surface plasmon resonan ce (LSPR), which involves restricted electron movement on AuNP 51 The incident light causes resonating oscillations of restricted electrons on the gold surface, which effectively leads to light absorption. The observed light absorption is the addition of the cross sections of absorbance and scattering. 52 As gold nanoparticle size increases there is a distinct red shift in agreement with Mie theory and theo retical predictions of spectra are possible. Mie theory is mainly useful in dilute solutions were particles do not readily interact. 51 This can also be a useful tool for confirming the size of the nanoparticles created since the wavelength is dependent on the gold nanoparticle size. Unfortunately, the AuNP

PAGE 23

10 absorb all of the emission from proteins when e xcited at 285 nm; this phenomenon is called inner filter effect and is a common problem for fluorescent studies. 53 AuNP are typically aggregates of gold( 0) atoms and are not soluble in aqueous environments. However, if they are functionalized with a thiol that contains a hydrophilic group opposing the sulfur end then they can become suspendable in water. 54 This technique has been utilized in a series of different fashions to obtain distinct uses for the water soluble AuNP Cyanide was introduced into the solution to establish the protective capability of the thiol ligands. Cyanide oxidizes the gold(0) nanoparticle s and forms a dicyanogold(I) complex that has no localized surface plasmon resonance (Equation 1) 55 Many different studi es have characterized the ion permeability of thiol ligands shells by this process, which is easily observed using a UV vis spectrometer. The stability of these partially protected, water soluble, thiol coated AuNP was attributed to the organization of ligands on the surface. As steric hindrance of the th iol ligands was increased the packing efficiency decreased and the nanoparticles became more susceptible to cyanide. 56 Long chain alkane thiols were demonstrated to be the most effective ligand for providing cyanide stability. Bulky ligands, like dendrimers, or short chain alkane thiols were found to be less protective against cyanide stability. 56 Thiol displacement reactions Scheme 1 Displacement of thiol coated gold nanoparticles by a shorter thiol. 4Au + 8CN + O 2 + 2H 2 O & 4[Au(CN) 2 ] + 4OH (1)

PAGE 24

11 showed that thiols like dithiothreitol (DTT) dihydrolipoic acid and glutathione could replace existing long chain alkane thiols on the surface of AuNP and expose the gold(0) core to cyanide. 57 This displacement proceeded quickly in aqueous environments were the opposing polar end groups helped stabilize the AuNP. Smaller thiols were more effective at the displacement of other thiols likely becaus e the ir molecular size allows for easier penetration of the existing ligand (Scheme 1) Nuclear magnetic resonance spectroscopy has also been utilized to observe thiol exchange reactions on gold cores. 56 It has been suggested that the displacement occurs by an associative mechanism where the incoming thiol forces th e bound thiol to detach from the surface. 57 Biological Membranes and Supported Surfaces Membranes are vital features of biological systems, serving as interfaces between the in terior and exterior of cells as well as vesicles 58 These membranes are organized as large liposomes with hydrophilic outer ends with a hydrophobic interior. Typical animal membranes consist of equal parts lipids and proteins by weight and control the transfer of signals into and out o f the cell. 59 Membranes are fluid structures that are difficult to characterize because typical characterization techniques like x ray crystallography do not give accurate representation of hydrated lipids. 60 Deposition of membranes onto solid supports is a popular method of creating biological mimics. 61 With the wide range of surfaces that can be synthesized and the increase in the number of surface sensitive Scheme 2 One possible mechanism for lipid deposition onto planar surfaces.

PAGE 25

12 characterization tech niques, supported lipid bilayers have become a successful way to model complex cellular structures. 62 ; 63 The precise mechanism by which liposomes form bilayers depends on multiple factors but clearly involves the rupture of vesicles and adhesion to supports. 64 Most models predict an accumulation of liposomes near the solid surface and when a specific surface density is obtained adhesion occurs. 58 This process can vary in length from seconds to hours and depends on the ionic strength of the solution, constituents of the liposome and the surface roughness. Asymmetry of lipid bilayers is common in nature and can also exist on supported lipid bilayers especially on highly curved membranes Identifying defects in the lipid layer is important because analyte membrane interactions can be significantly different from analyte support interactions. 65 Planar s upported surface techniques have previously yielded successful results when analyzing interactions of lipid s with CRP. 66

PAGE 26

13 Lipid coated Nanoparticles Li pid coated nanoparticles have a large range of biomedical applicability acting as simple models to identify integral portions of binding interactions. Control over the size and shape of the nanoparticles as well as membrane composition allows for clear co rrelation between protein interactions and specific substrates. Supported lipid bilayers have been shown to maintain fluidity, incorporate proteins and are impermeable to ionic species like natural liposomes (Scheme 2) 65 Planar substrates were originally the most commonly investigated substrate for membrane deposition; however, in the past several years 100 nm and 50 nm silica nanoparticles and various sizes of AuNP have been coated with lipid bilayers. 64 ; 67 ; 68 There are many natural occurrences of high membrane curvature including membrane blebbing during apoptosis, ves icle trafficking and lipoproteins. Creating mimics for these structures can be a valuable tool for assessing the properties of different proteins that interact with th ese biological structures. Free floating vesicles typically exist with ve ry large diameter s and forcing the lipids onto smaller nanoparticles lead s to strain. Lipid bilayers are fluid, so reorganization and packing can relieve some of this strain but lip id composition can also impact the membrane's ability to fit to specific curvature. Pores, holes in the lipid bilayer, can occur when a membrane is unable to cover a specific area due to various surface deformations. Atomic force microscopy has been one the most commonly utilized methods for evaluating lipid uniformity and detecting defections on supported membranes 64 Membranes can also be easily and quantitatively modified to assess numerous aspects of lipid composition. In particular, cholesterol reduces membrane fluidity by

PAGE 27

14 increa sing membrane packing, which can lead to a more stable membrane. 69 The flat and rigid cholesterol molecule with a small polar headgroup imposes local ordering in the hydrophobic portion of lipid membranes. Chole sterol is mainly located in the membranes of cells and the alteration of membrane function by cholesterol content is an important area of study. 70 Supported bilayer membranes containing cholesterol have even been demonstrated to be air stable during anal ysis. PC vesicles, containing up to 50 mole percent cholesterol, have demonstrated decreased ion permeability. 71 Lipid coated AuNP AuNP of various size s have been coated in PC in order to create cell mimics. Under specific conditions they can be stable in aqueous or organic solvents and switching between two solvent conditions is possible. 67 PC coated AuNP (PC AuNP) can be synthesized by at least two methods: either mixing gold(III) with phospholipids before reduction or by reducing gold(III) to gold(0) before the addition of lipids. 46 ; 72 To synthesize ion impermeable membranes surrounding th e gold core a thiol was used to increase packing of the lipid membrane (Scheme 3 ) The ion impermeability of thiol Scheme 3 Oleate coated gold nanoparticles are incubated with PC liposomes. The thiol anchor is added to stabilize the hybrid lipid bilayer against cyanide penetration.

PAGE 28

15 stabilized PC membranes surroundin g gold(0) nanoparticles cores can be evaluated by cyanide etching. 46 If the hybrid lipid bilayer is complete then cyanide will be unable to oxidize the gold core, which can be observed by the lack of a visible absorptio n decrease. 73 It has been suggested that the thiol replaces some portion of the inner leaflet of the bilayer and reduces the strain of the curved membrane. 46 ; 74 The exact location of the thiols within the bilayer has still not been exclusively determined 75 AuNP have also been functionalized to mimic to high density lipoproteins. 76 This involves first coating the nanoparticle with apolipoprotein A 1 before coating with PC This process can be observed by dynamic light scattering (DLS), which evaluates particle size utilizing Brownian motion. The nanoparticle size systematically increases with each additional additive in accordance with the size of the reactant. Determination of monolayer (~5 nm increase) or bilayer (~10 nm increase) deposition is possible usi ng dynamic light scattering. 76

PAGE 29

16 Introduc tion to Fluorescence Fluorescence is the subcategory of luminescence, where a molecule emits light from an electronically excited state. Excited states are typically achieved by light absorption from a source that outputs high enough energy to ensure electron excitation from a ground state to an excited state. Internal conversion can occur, which leads to non radiative decay by vibrational relaxation. Intersystem crossing, involving a spin flip, to a triplet state is also possible but that process is completely suppressed in aqueous solutions. After internal conversion, a comparably fast process, the excited electron will be at the lowest vibrational level of the excited state. The drop from the lowest vibrational level of the first electronic state to the ground electronic state will result in the emission of light. 53 Tryptophan, tyrosine and phenylalanine are three fluorescent amino acids that can sustain excited states in their conjugated $ systems Tryptophan fluorescence has been commonly used to identify solvent exposure to various parts of proteins. Excitation at 285 295 nm yields the highest emission intensity at roughly 345 350 nm for most proteins 53 Tryptophan fluoresces more readily in hydrophobic environments w h ere non radiative decay is less prevalent. Protein folding can create hydrophobic pockets where tryptophan residues can fluoresce. 77 Protein folding can also decrease fluorescent signal Figure 3 Fluorescent tryptophan residues in c reactive protein shown in green. There are 6 tryptophan residues per monomer, 30 tryptophan residues total. 1 ; 2

PAGE 30

17 because of absorption by other amino acids. Additionally, i nner filter effects caused by AuNP prevent fluorescent analysis because of their optical density. When unfolding occurs, tryptophan residues are exposed to aqueous solvent and non radiativ e decay by energy transfer to the solvent becomes the most prominent form of energy release. C reactive protein has six tryptophan residues per monomer, totaling 30 tryptophan residues per pentamer (Figure 3) Individual fluorescence will depend on the e lectronic environment of the tryptophan and conversion from pCRP to mCRP has been shown to lead to decreased intensity of fluorescence due to an increase in the amount polar solvent exposure to tryptophan residues 78 Fluorescence Quenching Fluorescence quenching has been a commonly utilized technique for identifying locations of fluorescent amino acids in proteins. The most frequently utilized t ype of quenching is dynamic or collisional quenching, where the excited fluorophore contacts the quencher that absorbs the energy in a non radiative manner. The mechanism by which the energy is transferred is not well characterized; however, electron tran sfer is the leading theory because of certain studies and models. 79 Static quenching, where the quencher attaches to the fluorophore, is a common complicating feature of quenching analysis. Quenching is considered a non destructive technique because typically there is no photochemical reaction between the quencher and the fluorophore. 53 The Stern Volmer equation relates the decrease in the fluorescence intensity to (2)

PAGE 31

18 the quen cher concentration as shown in E quation 2 This equation relates the ratio of the decrease in fluorescence (F 0 initially, F in the presence of quencher) to the concentration of the que ncher, [Q ], and the Stern Volmer constant, K SV Quenching data is plotting with F 0 / F on the y axis and [Q] on the x axis. 53 A linear relationship between quencher concentration and F 0 / F typically indicates that fluorophores have a similar electronic environment and are equally accessible; how ever, a complex system with multiple fluorophores in different electronic environments can lead to multiple complications. 80 The type of quencher utilized can also affect the slope of the plot because of interactions with other portions of the molecule. For example, large ionic quenchers do not typically effectively quench tryptophan r esidues that are located inside hydrophobic regions of the protein. This leads to a curved quenching plot because the interior fluorophores are continuing to fluoresce despite the ionic quencher Conversely, acrylamide, a common quencher, has been established to invade proteins, to a certain extent, and to quench internal tryptophan residues. 81 Additional thought also needs to be given to the efficiency of the quencher ( #). For efficien t quenchers, every collision leads deactiva tion of the excited fluorophore, # approaches unity and the determining factor in K SV is diffusion rate limited (Equation 3) 82 Efficient and established quenchers include: oxygen, acrylamide and iodide. Inefficient quenchers include: succinimide, bromate and imidazole. Efficiency can play a role in the linearity of the Stern Volmer plot as well as fluorophore availability. Ine fficiency may cause a false curve to appear or fortuitously cause a straight line to appear when a combination of inefficiency in both static and dynamic quenching # K SV = K SVobserved (3)

PAGE 32

19 occurs. 81 Efficiency is important to consider when attempting to character ize fluorophores of proteins. An ionic quencher that is unable to quench interior fluorophores in a protein may demonstrate a negative curve on the Stern Volmer plot, as certain fluorophores may remain unaffected by the ionic quencher. Divergently, oxyge n is one of the best known quenchers; however, it is not as useful for establishing exterior or interior tryptophan residues in a protein because it is easily able to permeate most proteins and therefore quench multiple tryptophan populations. 53 All quenchers have different properties tha t sui t them for spe cific experiments and choosing the optimal quencher is typically determined experimentally. Tryptophan Fluorescence and Quenching Tryptophan has a large amount of sensitivity to its local electronic environment and has additional com plications because it has two excited states, 1 L a and 1 L b Depending on multiple factors, including solvent exposure, polarity and nearby amino acids either of the two excited states can dominate emission by being lower in energy. 53 The 1 L a is the lower energy excited state for most proteins in aqueous solution and typically is the major excited state when excited between 285 290 nm. When excited in aqueous solvents, tryptophan not an efficient fluorophore has also been found to have a qu antum yield of ~0.13, which is one of the major factors against intrinsic fluorescence investigations. 83 The emission spectrum of tryptophan undergoes a red shift when dissolved in more polar solvents. Beginning at an emission maximum around 300 nm in cyclohexane and moving to 340 nm in ethanol and 350 nm in water when excited at 285 nm It is important to note that when dissolved in cyclohexane the 1 L b excited state is the

PAGE 33

20 lower energy excited state unlike when tryptophan is dissolved in water or ethanol where 1 L a is the lowest excited state. 53 Previously studied proteins have showed two distinct trends from their native to denatured isoforms. The emission maximum undergoes a red shift, to varying magnitudes, as well as the denatured protein having increased sensitivity to quenching. The latter aspect is revealed to lead to a steeper slope in the Stern Volmer plot or a larger K SV Szabo and co workers showed that the protein azurin demonstrates a significant red shift when denatured by guanidine hydrochloride. 84 This is logical when considering that when tryptophan is dissolved in increasingly polar solvents the emission shifts to longer wavelengths; therefore, this study is analogous because the tryptophan residue is entering a more polar environment and the em ission wavelength shifts accordingly. Avigliano and co workers demonstrated that the protein stellacyanin contained tryptophan residues with two different amount of quencher accessibility by quenching the native and denatured protein with iodide. 85 The larger K SV value of the denatured form of the protein indicated that the t ryptophan residues were more readily quenched by iodide, which is most likely due to increased solvent exposure. This confirms other work that iodide, an ionic quencher, is unable to penetrate certain proteins and quench inner fluorophores, but iodide is capable of quenching solvent exposed tryptophan residues. 83 A denatured protein typical ly loses its tertiary or quaternary structure and is more likely to have solvent exposed residues like tryptophan. 86 This work is still applicable even though mCRP is not technically a denatured form of pCRP because the idea of increased solvent accessibility to tryptophan residues is still consistent. 78

PAGE 34

21 Succinimide is a bulky polar, inefficient quencher that can be utilized to detect exterior tryptophan residues. 81 It has been suggested that because of the size and polari ty of succinimide it is unable to quench the interior tryptophan residues in certain proteins. Also, because succinimide is not an ionic quencher there are fewer considerations when determining salt effects during quenching studies. This makes succinimide a unique candidate for confirming solvent exposure tryptophan residues in multi tryptophan proteins. 82 Even some larger, polar quenchers, like acrylamide, have demons trated the ability to invade proteins and quench interior tryptophan residues. Eftink and co workers utilized succinimide as a model to show that certain quenchers were incapable of quenching interior proteins; therefore, only affecting the solvent expose d tryptophan residues. Succinimide has been demonstrated to only be partially effective ( # 70 %) at quenching tryptophan residues dissolved in water; indicating that even certain solvent exposed tryptophan residues may not be quenched by succinimide. 81 Succinimide has many unique characteristics that make it suitable for specific tasks in certain situations. Binding studies of CRP to PT PC AuNP PT PC AuNP is a cell mimic that effectively acts as an apoptotic bleb or LDL CRP has been shown to bind to these in a calcium dependent manner; therefore, EDTA can be used as a control to prevent binding and confirm that observed isoform conversion is due to calcium dependent binding to the cell mimics. The addition of BME disrupts the lipid layer surrounding the AuNP which can be demonstrated by cyanide instability (Scheme 4 ) W hen BME is used with centrifugation the membrane bound mCRP can be

PAGE 35

22 separated from the AuNP (Scheme 5 ) This is a key step because fluorescence studies that cannot be conducted in the presence of AuNP due to their inner filter effects. Succinimide was the fluorescence quencher that was utilized to differentiate between the two isoforms of CRP and binding studies between CRP and PT P C AuNP were confirmed by gel electrophoresis. Scheme 4 Propanethiol stabilized PC AuNP become completely cyanide instable after 2 hours when mixed with BME. Scheme 5 Isoform conversion as a result of calcium dependent binding of c reactive protein to PT PC AuNP. Removal of the hybrid lipid bilayer by BME for fluorescence quenching analysis.

PAGE 36

23 CHAPTER 2. EXPERIMENTAL Materials All water was deionized to 18.2 M % or bet ter (Milli q Water Purification Systems, Millipore, Bellerica, MA), and reage nts were reagent grade Fren's method was utilized for the formation of AuNP and lipid coating and thiol anchoring were adapted from previous reports with modifications. 50 ; 87 Gold na noparticle preparation AuNP were created by dissolving hydrogen tetrachloroaurate(III) hydrate (18 mg, Strem Chemicals, Newbury Port, MA) into 100 mL of water (530M Au) and adding sodium citrate (1.7 mM final concentration, JT Baker, Phillipsburg, NJ) to a refluxing solution for 15 minutes. The AuNP were allowed to sit for 24 hours without stirring. The sizes were determined by dynamic light scattering (DLS, Malvern Nanosizer) and confirmed by UV vis spectroscopy (PerkinElmer UV vis spectrometer Lamba 35 ). Lipid coated, thiol anchored AuNP synthesis L phosphatidylcholine ( 302 mg PC, Avanti Polar Lipids, Alabaster, AL) was dissolv ed in 5 mL of chloroform ( 80 mM PC ) and kept at 0 C when not in use. Cholesterol ( 32 mg Eastman Organic Chemicals, Kingsp ort, TN) was dissolved in chloroform ( 17 mM) and kept at 0 C when not in use. Aliquots of 25 L of the PC solution and 12 L of the cholesterol solution were added to a glass vial where they were mixed, evaporated by a stream of nitrogen and the remaining chloroform was evaporated overnight in a vacuu m desiccator (91% PC and 9% cholesterol ). PC films were

PAGE 37

24 rehydrated with 1 0 mM 4 (2 hydroxyethyl) 1 piperazineethanesulfonic acid (HEPES, 200 L, Sigma Aldrich, Saint Louis, MO) and placed in a bath sonicator (Branson 1510, Emerson Industrial Automation, Danbury, CT) for one hour. AuNP (1 mL) were incubated with 5 L of oleic acid sodium salt (100 M final concentration, TCI America, Portland, OR) for thirty minutes. PC /cholesterol rehydrated in HEPES was added to the AuNP solution (20 L, 200 nmol PC, 20 nmol cholesterol final concentration) and allowed to stir for one hour. Propanethiol (3 L Acr os Chemicals, Belgium) was dissolved i n 10 mL of water (3.3 mM ); all thiols were made fresh each day. Propanethiol (10 L, 32 M final concentration) was added to each vial and allowed to stir for 30 minut es. Occasionally, hexanethiol was used in place o f propanethiol for comparison purposes Hexanethiol (1 L) was dissolved in 1 mL of ethanol (7.1 mM) and 3 L were added to each vial (2 1 nM final concentration). CRP binding studies with PT PC AuNP Purified CRP (16 nM final concentration, Academy Biochemical Company) was added to 900 L of PC coated, thiol anchored AuNP ( PT PC AuNP) for thirty minutes. Ethylenediaminetetraacetic acid tetrasodium salt (2.08 g EDTA Matheson Coleman and Bell, Norwood, OH) was dissol ved in 5 mL of water and 5 L aliquots were used to prevent calcium dependent binding of CRP. mercaptoethanol (3 L, BME) was added to 500 L of water (85 mM) and 12.5 L of the BME solution were added to each vial (1.16 mM BME final concentration) for two hours. Alternatively, DL dithiothreitol ( DTT) was used for comparison displacement with BME. N methyl maleimide (25 mg, NMM) was dissolved in 2 mL of water (113 mM NMM) requiring small amounts of

PAGE 38

25 sonication and 10 L of the NMM solution (1.22 mM fina l concentration) was added to the reaction vial and allowed to sit for thirty minutes. Other studies included 100 L of the 113 mM NMM stock solution in order to completely prevent in gel degradation of CRP. The solutions were centrifuged (Eppendorf, Ham burg, Germany) at 15,000 rpm for 20 minutes and 800 L of supernatant was collected. The pellet was suspended with 800 L of water for analysis. Monomeric CRP preparation Monomeric C reactive protein (mCRP) was formed from stock pentameric C reactive protein (pCRP) utilizing 8 M urea and 10 mM EDTA at 37 C for 2 hours. The solution was dialyzed three times against PBS at 4 C and conce ntration was determined using a Bicinchonic acid protein assay kit (Peirce) Gel electrophoresis Agarose gels (0.8%, 0.4 g, Bio Rad, Hercules, CA) dissolved in 50 mL of 0.5 x Tris/Borate/EDTA buffer (TBE) and 0.005% sodium dodecyl sulfate (SDS, USB corporation, Cleveland, OH) was added after the agarose was dissolved in TBE The 5 x TBE buffer was created by dissolving 54 g of Tris (445 mM, JT Baker), 13.8 g of Boric acid (445 mM, Mallinckrodt) and 2.9 g EDTA (10 mM) were dissolved in 1 L of water and dilutions were made from this stock. The gel was run in 0.5 x TBE / 0.005% SDS buffer at 40 V for thirty minutes. The g el was transblotted onto nitrocellulose membrane (NTC, Schleicher and Schuell, Keene, NH) using a semi dry transblotter at 25 V for 90 minutes. The NTC was blocked with 3% bovine serum albumin (BSA, OmniPur, Gibbstown, NJ) 0.05% Tween 20 (VWR Internation al, Westchester, PA) and 0.002%

PAGE 39

26 NaN 3 in PBS overnight. The NTC was washed three times with Tween 20/PBS solution before equilibrating with biotinylated polyclonal anti CRP antibody (1:5000, Academy Biomedical Company) for one hour. The primary antibody c ould be used multiple times and was stored at 4 ¡ C along with the secondary antibody. The NTC was washed three times with Tween 20/PBS, incubated with streptavidin IR800 for a half hour (1:5000, Li COR Biotechnology) and imaged on an Odyssey imager (Li COR Biotechnology). Fluorescence analysis A spect rofluorometer (Photon Technology International, Birmingham, NJ) using FelixG software was used to analyze the supernatant after centrifugation. Measurements were carried out in a 1.5 mL quartz cuvette (Starna Cells, Atascadero, CA) at 37 C with 2 nm slit widths for the excitation monochromator and 8 nm slit widths for the emission monochromator. The samples were excited at 285 nm and data was recorded from 300 to 400 nm in 1 nm increments and 0.5 second integration time. The spetra were corrected for ba ckground absorption and the Raman peak of water. 78 ; 88 Fluorescence quenching studies Quenching experiments were conducted on binding samples as well as pCRP, mCRP and tryptophan controls. Supernatant collected from centrifugation (800 L) was evaluated for isoform determination after adding 23 L of 5.2 M sodium chloride ( 144 mM NaCl, Mal linscrodt) and 3.5 L of 600 mM calcium chloride (2.5 mM CaCl 2 Sigma Aldrich). Succinimide (0.2 g, Alfa Aesar, Ward Hill, MA) was dissolved in 1 mL of water (< 3 minutes of sonication, 2.02 M) and adjusted to maintain a constant ionic strength throughout the standard additions Occasionally, a crylamide (0.15 g, Sigma

PAGE 40

27 Adlrich) was used for comparison and was dissolved in 1 mL of water (2.11 M) and adjusted for ionic strength Cyanide stability studies Potassium c yanide (0.2 g Mallinckrodt) was dissolved in 5 mL of water (600 mM ). Thiol anchored, lipid coated gold nanoparticles were tested for ion impermeability by the addition of 10 L of the cyanide stock solution (6 mM final concentration ) The UV vis spectrum was taken of each sample, from 400 600 nm, before the a ddition of cyanide, 1, 2 3, 4, 5 and 24 hours after addition. The data was compared to the original absorption at 525 nm to evaluate the loss of signal due to cyanide oxidation

PAGE 41

28 CHAPTER 3. RE SU LTS Formation of PT PC AuNP AuNP characterization AuNP were created as a nanoparticle scaffold for binding studies between CRP and LDL mimics. The average size of all AuNP created was 17.3 nm ( 1.2) by DLS and the average # max for those AuNP was 525 nm. The optical density of AuNP at 525 nm was 0.9, which was used to determine the concentration of AuNP and the expected amount of thiols required to coat the AuNP. 87 The average size of PC AuNP was 26.1 nm ( 4.3) by DLS, an addition of 8.8 nm, the av era ge size of two bilayers is 8 10 nm 59 Maximizing ion impermeability of PT PC AuNP Optimizing the concentration of oleate for PT PC AuNP was important for prevention of AuNP ag gregation during lipid and thiol addition Varying amounts o f oleate were stirred in separate vials, with 1 mL of AuNP in each to determine the minimum concentration of oleate required to maintain stability against cyanide for 24 hours. For 1 mL of 1.5 nM AuNP: 33 nmol of propanethiol, 20 nmol of cholesterol (9% of total lipid ) 200 nmol of PC and 6 mM cyanide were all held constant. Figure 4 The absorbance retained at 525 nm after 24 hours of incubation with cyanide compared with the concentration of o leate used to prevent aggregation. The error bars are the standard deviation of at least 3 trials.

PAGE 42

29 Initial absorbance was recorded before the addition of cyanide and compared to absorption after samples were allowed to incubate for 24 hours with cyanide Maintaining a constant a mount of absorbance over a 24 hour period signifies that an ion impermeable membrane surrounds the gold core. As shown in Figure 4 the PT PC AuNP did not retain 100% cyanide stability until the concentration of oleate was 100 M. The lipid layer surrou nding the gold nanoparticle was modified with cholesterol to increase ion impermeability. 69 Different amounts of c holesterol were mixed with PC while both were dissolved in chlo roform to ensure homogeneity before evaporation by nitrogen. These films were rehydrated in 10 mM HEPES buffer and equivalent amounts of lipid were added to separate vial s Each vial also contained 33 nmol of propanethiol, 100 M oleate and 1 mL of 1.5 n M AuNP (17.3 nm diameter) The result was an increase in absorbance retention as the cholesterol content was at least 9% by mol e of total lipid (Figure 5) Cholesterol concentrations as high as 20% also created ion impermeable membranes. Figure 5 The absorbance retained after 24 hour of incubation with cyanide plotted against the percentage of lipid layer that was cholesterol. The remainder of the lipid layer was always PC and the error bars are the standard deviation of at least 3 trials.

PAGE 43

30 The concentrat ion of propanethiol was optimized for minimum ion impermeability which indicates that cyanide is unable to reach the gold core This implies that CRP will also be unable to interact with the gold core, which would cause undesirable interactions. All other reagents were held constant, includin g 9% cholesterol/PC lipid layer and 100 M oleate for 1 mL of 1.5 nM AuNP (17.3 nm diameter) A minimum amount of 33 nmol propanethiol per 1 mL of PC AuNP was essential to maintain 100% retained absorbance and up to ~70 nmol propanethiol could be used before AuNP aggregation occurred. Figure 6 The amount of nanomoles of propanethiol injected into a 1 mL solution of PC AuNP compared with absorbance retained after 24 hours of incubation with cyanide. The error bars are the standard deviation of 10 trials.

PAGE 44

31 B inding studies of CRP with PT PC AuNP CRP binds n in a calcium dependent manner n to PC; therefore, when EDTA is employed no binding occurs and the pCRP remains unaltered in solution. 17 ; 23 This acts as a useful control to determine if binding occurred as a results of calcium depe nd ent interactions with the PT PC AuNP. pCRP was incubated with PT PC AuNP for 30 minutes with and without EDTA. Samples were then centrifuged, run on a 0.8% ag a rose gel, transblotted onto a nitrocellulose membrane and visualized by Western blotting The pCRP and mCRP controls were incubated in buffer prior to loading in the well and the loc ation of those bands after running and blotting the gel onto nitrocellulose membrane denote the location of that isoform in the Figure 7. In samples with EDTA, t he pCRP re mains in the supernatant ; however, i n samples without EDTA, calcium dependent binding occurs and the mCRP is found in the pellet with the PT PC AuN P after centrifugation ( Figure 7 ). The LSPR of AuNP was a useful tool for during the cyanide stability tests; however, the LSPR absorbs all of t he electromagnetic radiation during fluorescence analysis. In order to evaluate the isoform of CRP by fluorescence analysis, CRP must be separated from the AuNP. Figure 7 Gel electrophoresis and Western blotting were preformed on the supernatant (S) and suspended pellet (P) of samples that did and did not contain EDTA. Controls of pCRP and mCRP were run for comparison.

PAGE 45

32 Lipid disruption utilizing BME or DTT The lipid bilayer surrounding PT PC AuNP became p ermeable when BME or DTT was added. This helps separate the lipid bilayer from the AuNP so that fluorescence analysis can be preformed on the lipid bound m CRP. All PT PC AuNP were synthesized with 100 M oleate, 9% cholesterol/PC lipid layer and 33 nmol of propanethiol per 1 mL of AuNP. As shown in Figure 8 complete cyanide instability was demonstrated when PT PC AuNP were incubated with BME after 2 3 hours and when incubated with DTT for 24 hours With a final concentration 1 mM the BME or DTT is abl e coat at least a portion of the AuNP and expose the gold to the cyanide. Previously, PC AuNP were stabilized with 9 nmol of hexanethiol per 1 mL of PC AuNP with an optical density of 0.8 and 10 nm diameter AuNP 87 No loss of absorbance was observed when 1 mM BME or DTT was incubated with hexaneth iol stabilized PC AuNP (Figure 9 ). Figure 8 All samples c ontained PT PC AuNP and cyanide. One sample contained 1 mM BME ( ) another 1 mM DTT ( ) and the last did not contain any other reactants ( # ). The error bars are the standard deviation of 10 trials. Figure 9 All samples contained H T PC AuNP and cyanide. One sample contained 1 mM BME ( ) another 1 mM DTT ( ) and the last did not contain any other reactants ( # ). The error bars are the standard deviation of 7 trials.

PAGE 46

33 Concentrations of 20 nmol of hexanethiol per 1 mL of PC AuNP were required to replicate the cyanide stability in this study No loss of absorbance indicates that the hexa nethiol anchored PC AuNP had an ion impermeable membrane the gold core surface could not be modified and lipid bilayer could not be disrupted by BME or DTT The apparent rise in abso rption is peculiar but possibly due to evaporation of water, which would increase the scattering and absorbance of the solution. The t test indicates that all three samples had a statistically significant increase in absorption from the 1 st hour to the 24 th hour; however, there was not a statistically significant increase in absorption from the 1 st hour to the 2 nd hour and the 5 th hour to the 24 th hour (99.9% confidence interval). 89 An increase in absorption do es not i ndicate cyan ide oxidation of the gold core. BME and DTT have been demonstrated to cleave disulfide bonds in proteins and disrupt quaternary structure ; therefore, g el electrophoresis was used in order to ensure that the structure of p CRP was not influenced by the presence of BME or DTT. 59 p CRP was incubated with 1 mM BME or DTT for 1 hour before being run through an agarose gel and blotted in the Western method. The Western blot, Figure 10 shows that pCRP does not unde rgo isoform conversion to mCRP under these conditions Figure 10 Gel electrophoresis and Western blotting of samples containing BME and pCRP, DTT and pCRP, pCRP and mCRP.

PAGE 47

34 Quenching Experiments Quenching in buffer by succinimide and acrylamide Herein, we display the ability to discern at least two of the isoforms of CRP employing the fluorescence quencher succinimide. The data in Figure 11 shows the result of tryptophan, Urea/EDTA mCRP and pCRP succinimide quenching and Stern Volmer analysis in buffer containing 10 mM phosphate, 140 mM NaCl and 2 .5 mM CaC l 2 Initial fluorescence was recorded and the reduced fluorescence was plotted with the corresponding concentration of quencher 53 There are 30 tryptophan residues per CRP molecule, so the concentration of tryptophan was 30 times the concentration of CRP (240 mM) in order to match total Figure 11 Stern Volmer p lot of pCRP ( # ), mCRP ( ) and tryptophan ( ) quenched by succinimide in 10 mM phosphate buffer, 140 mM NaCl and 2.5 mM CaCl 2 The equation of the best fit line for tryp: y = 0.0135x + 1.010; mCRP: y = 0.0088x + 1. 044; pCRP: y = 0.0031 + 0.975. Table 1 Summarizes succinimide fluorescence quenching data. Buffer background refers to quenching fluorescence in 10 mM phosphate buffer, 140 mM NaCl and 2.5 mM CaCl 2 PT PC AuNP supernatant background refers to making PT PC AuNP, centrifuging at 15,000 rpm for 20 minutes, and adding mCRP, pCRP and tryptophan to the supernatant

PAGE 48

35 fluorophore concentration. 11 The slop e, K SV of tryptophan was the largest, 13 5 M 1 because the tryptophan residues are completely exposed to the quencher (Figure 11 ). The K SV of urea/EDTA mCRP (8 nM) was 2.9 M 1 and the K SV of pCRP (8 nM) was 1.1 M 1 (Table 1) Acrylamide is a more efficient quencher of fluorescence than succinimide, but has been shown to diffuse into proteins and quench interior tryptophan residues. 53 The efficiency was confirm e d by the data shown in Figure 12 which shows the K SV for tryptopha n to be twice as large for acrylamide (K SV = 26.4 M 1 ) as succinimide (K SV = 13.5 M 1 ) in buffer (Figure 11) Additionally, pCRP had a larger value (K SV = 5.2 M 1 ) than mCRP (K SV = 1.0 M 1 ) indicating that acrylamide was not only quenching exterior tryptophan residues but also interior tryptophan residues. 88 The lack of clarity of these acrylamide quenching results gives support to using succinimide as the better quencher for discerning the two isoforms of CRP in this particular examinatio n. The succinimide data is consistent with the current understanding of solvent accessibility to tryptophan and allows for a rational difference in K SV values for the two isoforms of CRP. Figure 12 Stern Volmer plot o f pCRP ( # ), mCRP ( ) and tryptophan ( ) quenched by acrylamide in 10 mM phosphate buffer, 140 mM NaCl and 2.5 mM CaCl 2 The equation of the best fit line f or tryp: y = 0.0264x + 0.0660; mCRP: y = 0.0010x + 1.044; p CRP: y = 0.0052 + 0.926.

PAGE 49

36 Succinimide q uenching in PT PC AuNP background The K SV of fluorophore quencher pa i r s can be influenced by background conditions, so the background was modified to be a more accurate comparison for CRP and PT PC AuNP binding studies. Succinimide quenching of tryptophan, pCRP and mCRP was conducted in the supernatant of PT PC AuNP in order to observe the effects of background complications on K SV val ues PT PC AuNP were created and centrifuged to replicate the conditions the background experiences during the binding study. Urea/EDTA mCRP, pCRP and tryptophan were added to the supernatant of PT PC AuNP and quenched by succinimide (Figure 13 ) mCRP had a K SV of 2 9 M 1 and pCRP had a K SV of 1 1 M 1 which are both lower in magnitude than the values observed when mCRP and pCRP were quenched in buffer (Table 1) There is still a similar ratio (K SV mCRP / K SV pCRP ) between the two isoforms of C RP so qu enching by succinimide is still a viable method for isoform determination. Figure 13 Stern Volmer plot of pCRP ( # ), mCRP ( ) and tryptophan ( ) quenched by succinimide in the supernatant of PT PC AuNP. The equation of the best fit line for tryp: y = 0.0078x + 1.056; mCRP: y = 0.0029x + 1.057; pCRP: y = 0.0011 + 1.011.

PAGE 50

37 Binding Studies Utilizing BME to disrupt PT PC AuNP Binding studies of p CRP with PT PC AuNP were conducted to evaluate isoform conversion a s a function of calcium dependent PC binding. Binding s tudies were monitored by gel electrophoresis and fluorescence quenching experiments to determine CRP isoform. pCRP was bound to cell mimics in a calcium dependent method; EDTA was incubated with cert ain samples to act as a control to make sure other factors were not leading to isoform conversion. In order to analyze mCRP by succinimide quenching, BME was utilized to disrupt the hybrid lipid bilayer and centrifugation separated the AuNP from the conce ivably still membrane bound mCRP (Scheme 5) To Table 2 Succinimide quenching data of binding studies samples containing CRP incubation with PT PC AuNP and BME. The 1:1 BME:NMM also contains 1 mM NMM and all samples were centrifuged before quenching analysis. Figure 14 Stern Volmer plots of the succinimide quenching of the supernatants of the binding studies. One sample contained calcium ( ) and isoform conversion occurred ( y = 0.0012x + 1.025 ). Th e other sample contained EDTA ( # ) and binding did not occur ( y = 0.005x + 1.0103 ).

PAGE 51

38 maintain uniformity, BME was also added to samples containing EDTA despite having less of an active role since it was already demonstrated in Figure 7 that pCRP remains in the supernatant Succinimide quenching and Stern Volmer analysis were preformed on the supernatants of the samples described above The K SV valu es of mCRP and pCRP were 1 16 M 1 an d 0. 46 M 1 respectively (Figure 14 ). These absolute magnitudes were lower than what was observed in buf fer or in the PT PC AuNP supernatant. The K SV mCRP / K SV pCRP was 2. 5 (Table 2) comparable to the 2.8 observed in buffer and 2.6 observed in PT PC AuNP supernatant (Table 1) Conversely neither of the comparison values contain ed BME, which may have ef fected the K SV mCRP / K SV pCRP BME is thought to disrupt at least some portion of the lipid membrane surrounding the PT PC AuNP, which would lead to a larger amount of lipids or BME in the supernatant. This variability in the background has significant ef fects on the K SV values. A complication arises when gel electrophores is and W estern blotting is preformed on the binding samples. BME was used in the binding studies to disrupt the lipid membrane; however, during gel electrophoresis the presence of BME led to smearing (Figure 15 ). The lane that does not contain EDTA (where calciu m dependent binding can occur) show s mCRP as expected; however, the smear in the lane that contains EDTA should only contain one pCRP band. Figure 15 Western blot of s upernatants of samples where CRP was incubated with PT PC AuNP and then BME before centrifugation.

PAGE 52

39 Sequestering of BME by NMM To sequester excess BME, N methyl maleimide (NMM) was added to the CRP PT PC AuNP binding study before centrifugation. As shown in Figure 16 NMM prevented the degradation of pCRP in the EDTA containing samples and did not have effects in the samples containing calcium. Centrifugation is important because it allows for fluorescence ana lysis by removing the inner filter effects of the AuNP. NMM was not previously reported to interact with or affect CRP. 90 NMM is a quencher of fluorescence and at the 1:23 mol ratio ( 1 mM BME: 23 mM NMM) there is no observable fluorescence signal in the supernatant of the CRP PT PC AuNP binding stu dies. In a separate study NMM was added to the CRP PT PC AuNP binding studies samples in a 1:1 BME:NMM ratio in an attempt to prevent CRP degradation during gel electrophoresis without quenching the entire fluores cent signal. This resulted in K SV mCRP / K SV pCRP = 2.2 from succinimide quenching of the supernatant which is similar to the K SV mCRP / K SV pCRP = 2. 5 without any NMM (Table 2); however, there was still some smearing in the gel. This serves as a n internal confirmation that the degradation of the pCRP (Figure 16) occurred during gel electrophoresis, and not during incubation with PT PC AuNP and EDTA, because the K SV mCRP / K SV pCRP is similar to the ratio observed in buffer and in PT PC AuNP supernatant (Table 1 ). The amount of signal from tryp tophan fluorescence was less in the 1:1 BME:NMM samples than samples that did Figure 16 Western blot of b inding studies of CRP with PT PC AuNP, after incubation with 1 mM BME, 23 mM NMM and centrifugation. The supernatant (S) and pellet (P) of samples containing EDTA and not containing EDTA are shown.

PAGE 53

40 not contain any NMM, so there are additional factors to consider when comparing K SV mCRP / K SV pCRP ratios.

PAGE 54

41 Chapter 4. Discussion PT PC AuNP preparation AuNP size and shape determination In order to e nsure monodispersity in size and shape AuNP were created by a common method and thoroughly characterized. AuNP were previously characterized by transmission electron microscopy (TEM), dynamic l ight scattering (DLS) and UV vis analysis. 50 ; 91 Our lab utilized the latter two methods to determine the size and concentration of AuNP synthesized by the same method. Basu and co workers demonstrated a correlation between AuNP diameter and # max (Figure 17 ). 91 The AuNP from our analysis fits within this trend line by being an appropriate size for the observed # max This can be inferred as an additional confi rmation of the diameter and shape of the AuNP. Liu and co workers developed a method to calculate the concentration of AuNP based on diameter and absorbance of the AuNP. 92 The extinction coefficient ( % ) was Figure 17 Adapted from Basu and co workers. # max is plotted against the diameter, determined by TEM, of AuNP also synthesized by Fren's method ( ) The average # max for th e AuNP created by this analyst ( $ ) is pl otted to show the expected corresponding size. ln( % ) = k ln(d) + a (4)

PAGE 55

42 determined by AuNP diameter (d) and Beer's law was employed to determine final concentratio n. The relationship between extinction coefficient and diameter is shown in Equation 4 with k = 3.32111 and a = 10.80505 92 Fr om the data published by Mackiew icz and co workers, the t otal surface area in their analysis was 14 cm 2 and the total surface area in this analysis was 8.6 cm 2 The total surface area comparison indicates that there is around half as much space available for thiols to interact with in this analysis when compared to previous work. Accounting for the number of thiols, there were ~4 thiols/nm 2 in the Mackiewicz publication while there were ~23 thiols/nm 2 in t his analysis It is important to note that hexanethiol was utilized to stabil ize the PC AuNP in the Mackiewicz analysis and propanethiol was used in this analysis 87 Thiol to surface area comparison between different studies may be more accurate based on the current understanding of thiol and gold interactions. Previous work demonstrated that hexanethiol and decanethiol stabilized PC AuNP were impermeable to cyanide etching 46 ; 75 ; 87 The completenes s of the lipid layer was evaluated using cyanide, which readily oxidizes gold(0) to gold(1), and can be observed by the loss of visible light absorption. 55 Novel, PT PC AuNP were shown here to also by impermeable to cyanide penetration. However, concentrations of the other reactants, including oleate, lipid composition and thi ol concentration, had to be modified because of the use of propanethiol as the anchor. Oleate concentration The sodium salt of oleic acid is a surfactant utilized to stabilize the gold AuNP during the addition of the PC. The AuNP are free floating in solution because the citrate used to reduce the gold(III) to gold(0) nanoparticles surrounds the AuNP. The role of

PAGE 56

43 oleate is not completely understood, but when lipids are added to AuNP solutions without oleate the AuNP aggregate and crash out of solution. It is likely that the surfactant helps suspend the hydrophobic AuNP in the presence of lipids, while the citrate can only kee p the AuNP suspended in water. The oleate concentration 100 M, is larger than previously published olea te concentrations (20 M) which used hexanethiol or decanethiol to stabilize the PC AuNP. 87 Aggregation is a visible color change and typically occurs during the addition of thio ls, so possibly the large amount of propanethiol required more oleate. Cholesterol concentration Cholesterol, a common constituent of membranes, was added to the PC to decrease membrane fluidity. 69 Chol esterol was not previously required to create ion impermeable hexane thiol stabilized PC AuNP, but was necessary with the PT PC AuNP. 87 Natural vesi cles exist with varying diameters typically > 26 nm, so forcing the lipid layers onto solid supports that are smaller leads to an increased amount of strain. Cholesterol has a small polar headgroup and large cone angle that if located on the inner leaflet of the bilayer could help to reduce the strain associated with forming around the AuNP. 71 This would reduce the possible interstitial water layer volume and allow the shorter propanethiol to assist in stabilizing the lipid bi layer. 87 We suggest that in addition to forming a less mobile bilayer the cholesterol also allowed for tighte r packing of the lipid around the AuNP.

PAGE 57

44 Thiol concentration The thiol anchor helps stabilize the ion impermeable lipid layers around the gold core and determining the optimal concentration is important. When decanethio l was used ~6 thiols/surface gold atom were needed to prevent cyanide from oxidizing the gold core of PC AuNP. 46 Yang and co workers demonstrated the ability to create cyanide stable lipid coated gold nanoparticles with a octadecanethiol anchor using ev en lower amounts ~2 thiol s/surface gold atoms 75 These reports lead to the suggestion that chain length correlates to the amount of thiol required for ion impermeability; however, c omparison to pre vious work is complicated by the varying amount of surface area. More rigorous analysis of surface area is required before direct comparisons of concentrations can be made. As shown above, the amount of available surface area in this analysis was around half that of previous work with smaller AuNP. This correla tes poorly with the fact that twice as much hexanethiol was needed to replicate those results There are a variety of factors to consider in this comparison including the modifications to the oleate concentration and cholesterol content in the lipid layer that were not present in the previous analysis. However, e ven higher amounts ~23 thiols/nm 2 we re required when propanethiol was used as the anchor. We propose that higher concentrations are needed for propanethiol because the shorter alkane chain has less ability to interact with the lipid bilayer.

PAGE 58

45 Binding studies of pCRP with PT PC AuNP Quench ing issues and BME One of the first issues when assessing the binding of p CRP to PT PC AuNP by fluorescence was the lack of observable signal because of the inner filter effects of the AuNP. Separating the CRP from the AuNP without altering the isoform of the protein became crucial for applying fluorescence quenching techniques to determine the isoform of CRP. The main advantage of the PT PC AuNP over previous thiol stabilized PC AuNP is that the lipid layer can be disrupted with the addition of BME or DT T (Figure 8 ) The sulfur end of the BME has a high affinity for gold and the hydroxyl group on the opposing end disrupts the tight lipid packing and allows for penetration of cyanide. Because BME disrupted the hybrid lipid bilayer faster than DTT it was selected for the lipid displacement in the binding studies. The extent to which the lipid bilayer is detached is unknown, but it is possible that a significant portion of the PC membrane is detached from the AuNP. The gel data confirmed that BME did not appear to affect pCRP (Figure 10 ), and adding BME to all samples helped prevent multiple variable analysis comparison issues. Wang and co workers also demonstrated that DTT does not affect the tertiary structure of pCRP; however, they did show that the dis ulfide bonds in mCRP are reduced when incubated with DTT, which reveals cholesterol binding epitopes and implicated reduced mCRP in proinflammatory activities. The reduced mCRP showed similar tertiary structure and fluorescent properties to non reduced mC RP. 90 Agarose gels are not sensitive enough to differentiate betwe en reduced and non reduced mCRP, so a difference is not observed in the binding studies presented here.

PAGE 59

46 Quenching with succinimide and acrylamide in buffer Differentiating pCRP from mCRP and observing the isoform conversion are impo rtant aspects of developing a more complete understanding of the reactivity of CRP. Tryptophan quenching experiments and Stern Volmer analysis has not been previously utilized to differentiate the two isoforms. mCRP quenched by succinimide in buffer had a K SV of 8.8 M 1 (Figure 11 ) and pCRP had a K SV of 2.9 M 1 which correlates with the theory that pCRP has more interior tryptophan residues that are harder for a large, polar quencher like succinimide to access. This leads to a lower K SV value because su ccinimide would not be able interact with tryptophan residues that are located in nonpolar, interior portions of proteins. This displays the capability of discriminating between the two isoforms of CRP by a new method. There are many possible explanations for the observed K SV values but they are difficult to consider without mCRP being fully characterized. Additionally modCRP and forced conversion mCRP may not necessarily be identical despite having similar reactivity 5 ; 13 What is established is that the tryptophan residues of mCRP fluoresce significantly less than pCRP likely due to solvent exposure and previous reports determined the mCRP has an increased amount of heli x portion relative to pCRP. 90 Proteins frequ ently quench their own fluorescence and it is possible that during isoform conversion mCRP became able to quench more of its own intrinsic fluorescence. 78 Succinimide was selected as the optimal quencher of fluorescence over many other quenchers including acrylamide. Acrylamide is one of the most commonly used quenchers because of its high efficiency and using it avoid s electrostatic problems because it is an uncharged quencher 53 ; 86 Acrylamide data was presente d to represent the

PAGE 60

47 issues that other quenchers had when attempting to differentiate the two isoforms of CRP. A possible explanation for the inconsistency in this case may be due to the ability of acrylamide to penetrate proteins which has been highly sus ceptible to a variety of factors dependent on the protein. 88 ; 93 Differentiating the two isoforms of CRP in a logical method is important when choosing a quencher. Succinimide quenching with PT PC AuNP background The goal of the fluorescence quenching studies was to establish a new method to differentiate the two isoforms of CRP. Background changes can drastically affect the quenching constants of protein/quencher pairs. 53 Succinimide gave different absolute values for the K SV of mCRP and pCRP de pending on the background used for the analysis. However, the ratio of K SV mCRP / K SV pCRP remained very similar (K SV mCRP / K SV pCRP = 2.8 in buffer and K SV mCRP / K SV pCRP = 2.6 in PT PC AuNP supernatant background), which indicates that the value of the in formation may not be lost due to various background conditions (Table 1) CRP is a complex protein for quenching analysis due to the large number of fluorophores and a thorough characterization of the tryptophan residues was never the objective of this re search. Effectively differentiating mCRP from pCRP is the main aim and despite the K SV magnitude change the usefulness of the resulting data does not decline. One possible explanation that exists is that the increased complexity of the samples led to a more complex b ackground and different slopes.

PAGE 61

48 BME and NMM The addition of BME led to an increased amount of mCRP in the supernatant, but degraded pCRP during gel electrophoresis. Importantly, the majority of the protein was in the supernatants of the gel lanes and no longer in the suspended pellet lanes (Figure 15 ) This demonstrates the usefulness of BME especially in the samples containing calcium, where binding occurred, because the mCRP is no longer in the pellet but has been separated from the PT PC AuNP. The smeari ng in the lane containing EDTA, where no isoform conversion previously occurred, was unfortunate but not unprecedented N ethyl maleimide was previously utilized to prevent CRP degradation during gel electrophoresis in experiments utilizing DTT. 90 In this analysis, NMM was utilized for the same purpose and as shown in Figure 16 was successful at preventing in gel degradation of pCRP. It is reasonable to assume that the degradation of p CRP, in binding study samples that contained EDTA, occurred during gel electrophoresis because the K SV mCRP / K SV pCRP ratio without NMM or with a 1:1 BME:NMM was similar to the K SV mCRP / K SV pCRP in buffer or with PT PC AuNP background (Tables 1 & 2). Isoform differentiation is possible by K SV mCRP / K SV pCRP and is confirmed by gel electrophoresis in the presence of 1:1 BME: NMM. Intriguingly, the samples that do not contain EDTA (where calcium dependent binding and isoform conversion occur) do not require NMM but do require BME to separate the PT PC Au NP from the mCRP for fluorescence analysis. Conversely the samples that do contain EDTA (where pCRP remains in solution) do not require BME, because separation of the lipid layer from the AuNP is not required, but do need NMM to prevent smearing during g el electrophoresis. These problems could be

PAGE 62

49 solved alternatively if BME was not added to samples containing EDTA and NMM was not added to samples that do not contain EDTA.

PAGE 63

50 Chapter 5. Conclusions These studies indicate that intrinsic fluorescence quenching utilizing succinimide and Stern Volmer analysis can differentiate pCRP and mCRP In order to separate the mCRP from the AuNP, so fluorescence quenching can be employed, PT PC AuNP needed to be fabricated. The lipid bilayer on the HT PC AuNP cou ld not be disrupted by BME or DTT so a less stable, shorter alkanethiol anchors was used. The less stable propanethiol anchor required higher concentrations of oleate, PC and thiol than HT PC AuNP. PT PC AuNP also required cholesterol in the lipid layer in order to maintain tight lipid packing demonstrated by stability against cyanide oxidation B inding studies using the PT PC AuNP allowed for the disruption of the hybrid lipid bilayer with the addition of BME. This allowed for separation of mCRP from the AuNP and fluorescence quenching differentiating the two isoforms of CRP without interfering inner filter effects This work demonstrates novel PT PC AuNP as well a new method for dif ferentiating mCRP and pCRP. PT PC AuNP is a cell mimic that has a dis placeable hybrid lipid bilayer when BME or DTT is added. This additional control gives future investigators more options when modifications to the cell mimic are needed without aggregation of the PC AuNP. BME was determined to be more effective at disrup ting the bilayer than DTT because of the shorter time required to reach complete oxidation of the AuNP Displacing the lipid layer was essential for separating mCRP from the PC AuNP without allowing the interacting of CRP and the AuNP because bare AuNP ca n cause isoform conversion BME effectively detached the bound m CRP and lipid bilayer and allowed for fluorescent quenching analysis. Succinimide was found to be the most effective quencher for determining isoform despite some of its perceived shortcoming s.

PAGE 64

51 Established gel electrophoretic techniques confirmed the isoform conversion determined by the quenching studies. The gel also indicated that a large portion of the mCRP was removed from the AuNP because there was very little if any mCRP in the suspend ed pellet lane. Future C1q studies would also help to confirm that NMM only prevents in gel degradation of CRP and there are not other competing factors Hopefully, a more complete understanding of CRP could lead to a practical application in the prevention of cardiovascular disease. Cell mimics will continue to be useful tools for establishing models and isolating specific interactions with proteins. Membranes are the interface of the exterior and interior of cells and are very complex. Increasing the amount of possible modifications and flexibility in membrane models will allow for a wider range of possible experiments. Many other proteins bind to c ell walls and examining these phenomenons in solution may provide added insight to their mechanisms. Changing membrane composition would also allow for further studies to examine and isolate necessary interactions. It is also possible to change the compo sition of the nanoparticle in order to modify the size and shape of the mimic desired.

PAGE 65

52 REFERENCES 1. Shrive, A. K., Gheetham, G. M. T., Holden, D., Myles, D. A. A., Turnell, W. G., Volanakis, J. E., Pepys, M. B., Bloomer, A. C. & Greenhough, T. J. (1996). Three Dimensional Structure of Human C reactive Protein. Nature Structural Biology 3 346 354. 2. Mor eland, J. L., Gramada, A., Buzko, O. V., Zhang, Q. & Bourne, P. E. (2005). The Molecular Biology Toolkit (MBT): a modular platform for developing molecular visualization applications. BMC Bioinformatics 6 21. 3. Abernethy, T. J. & Avery, O. T. (1941). The occurrence during acute infections of a protein not normally present in the blood: I. Distribution of the reactive protein in patients' sera and the effect of calcium on the flocculation reaction with c polysaccharide of pneumococcus. The Journal of Exper imental Medicine 73 173 82. 4. Volanakis, J. E. (2001). Human C reactive protein: expression, structure, and function. Molecular Immunology 38 189 97. 5. Eisenhardt, S. U., Thiele, J. R., Bannasch, H., Stark, G. B. & Peter, K. (2009). C reactive protein: how conformational changes influence inflammatory properties. Cell Cycle 8 3885 92. 6. Pepys, M. B. (1981). C reactive protein fifty years on. Lancet 1 653 7. 7. Pepys, M. B. & Hirschfield, G. M. (2003). C reactive protein: a critical update. The Journal of Clinical Investigation 111 1805 1812. 8. Singh, S. K., Suresh, M. V., Voleti, B. & Agrawal, A. (2008). The connection between C reactive protein and atherosclerosis. Annals of Medicine 40 110 20. 9. Ridker, P. M., Hennekens, C. H., Buring, J. E. & Rifai, N. N. (2000). C Reactive Protein and Other Markers of Inflammation in the Prediction of Cardiovascular Disease in Women. The New England Journal of Medicine 342 836 843.

PAGE 66

53 10. Ridker, P. M., Rifai, N., Rose, L., Buring, J. E. & Cook, N. R. (2002 ). Comparison of C reactive protein and low density lipoprotein cholesterol levels in the prediction of first cardiovascular events. The New England Journal of Medicine 347 1557 65. 11. Thompson, D., Pepys, M. B. & Wood, S. P. (1999). The physiological st ructure of human C reactive protein and its complex with phosphocholine. Structure 7 169 177. 12. Oliveira, E. B., Gotschlich, E. C. & Liu, T. Y. (1977). Primary structure of human C reactive protein. Proceedings of the National Academy of Science of the United States of America 74 3148 51. 13. Ji, S. R., Wu, Y., Zhu, L., Potempa, L. A., Sheng, F. L., Lu, W. & Zhao, J. (2007). Cell membranes and liposomes dissociate C reactive protein (CRP) to form a new, biologically active structural intermediate: mCRP( m). The Journal of the Federation of American Societies for Experimental Biology 21 284 94. 14. Potempa, L. A., Maldonado, B. A., Laurent, P., Zemel, E. S. & Gewurz, H. (1983). Antigenic, electrophoretic and binding alterations of human C reactive protein modified selectively in the absence of calcium. Molecular Immunology 20 1165 75. 15. Potempa, L. A., Siegel, J. N., Fedel, B. A., Potempa, R. T. & Gewurz, H. (1987). Expression, detection and assay of a neoantigen (Neo CRP) associated with a free, human C reactive protein subunit. Molecular Immunology 24 531 541. 16. Kinoshita, C. M., Ying, S. C., Hugli, T. E., Siegel, J. N., Potempa, L. A., Jiang, H., Houghten, R. A. & Gewurz, H. (1989). Elucidation of a protease sensitive site involved in the binding o f calcium to C reactive protein. Biochemistry 28 9840 8. 17. Agrawal, A., Lee, S., Carson, M., Narayana, S. V., Greenhough, T. J. & Volanakis, J. E. (1997). Site directed mutagenesis of the phosphocholine binding site of human C reactive protein: role of Thr76 and Trp67. Annals of the New York Academy of Sciences 389 235 250.

PAGE 67

54 18. Volanakis, J. E. (1982). Complement activation by C reactive protein complexes. Annals New York Academy of Sciences 389 235 50. 19. Richards, R. L., Gewurz, H., Osmand, A. P. & Alving, C. R. (1977). Interactions of C reactive protein and complement with liposomes. Proceedings of the National Academy of Sciences of the United States of America 74 5672 6. 20. Volanakis, J. E. & Narkates, A. J. (1981). Interaction of C reactive pr otein with artificial phosphatidylcholine bilayers and complement. The Journal of Immunology 126 1820 5. 21. Ji, S. R., Ma, L., Bai, C. J., Shi, J. M., Li, H. Y., Potempa, L. A., Filep, J. G., Zhao, J. & Wu, Y. (2009). Monomeric C reactive protein activat es endothelial cells via interaction with lipid raft microdomains. The Journal of the Federation of American Societies for Experimental Biology 23 1806 1816. 22. Gershov, D., Kim, S., Brot, N. & Elkon, K. B. (2000). C Reactive protein binds to apoptotic cells, protects the cells from assembly of the terminal complement components, and sustains an antiinflammatory innate immune response: implications for systemic autoimmunity. The Journal of Exper imental Medicine 192 1353 64. 23. Volanakis, J. E. & Wirtz, K. W. (1979). Interaction of C reactive protein with artificial phosphatidylcholine bilayers. Nature 281 155 7. 24. Kishore, U. & Reid, K. B. (2000). C1q: structure, function, and receptors. Imm unopharmacology 49 159 70. 25. Richards, R. L., Gewurz, H., Siegel, J. & Alving, C. R. (1979). Interactions of C reactive protein and complement with liposomes. II. Influence of membrane composition. The Journal of Immunology 122 1185 9. 26. Roux, K. H., Kilpatrick, J. M., Volanakis, J. E. & Kearny, J. F. (1983). Localization of the phosphocholine binding sites on C reactive protein by immunoelectron microscopy. The Journal of Immunology 131 2411 2415.

PAGE 68

55 27. Mold, C., Gresham, H. D. & Du Clos, T. W. (2001) Serum amyloid P component and C reactive protein mediate phagocytosis through murine Fc gamma Rs. The Journal of Immunology 166 1200 5. 28. Chang, M. K. (2002). C reactive protein binds to both oxidized LDL and apoptotic cells through recognition of a c ommon ligand: Phosphorylcholine of oxidized phospholipids. Proceedings of the National Academy of Sciences of the United States of America 99 13043 13048. 29. Pentikainen, M. O., Oorni, K., Ala Korpela, M. & Kovanen, P. T. (2000). Modified LDL trigger of atherosclerosis and inflammation in the arterial intima. Journal of Internal Medicine 247 359 70. 30. Lima, E. S. & Maranhao, R. C. (2004). Rapid, simple laser light scattering method for HDL particle sizi ng in whole plasma. Clinical Chemistry 50 1086 8. 31. Singh, S. K., Suresh, M. V., Hammond, D. J., Rusi–ol, A. E., Potempa, L. A. & Agrawal, A. (2009). Binding of the monomeric form of C reactive protein to enzymatically modified low density lipoprotein: effects of phosphoethanolamine. Clinica Chimica Acta 406 151 5. 32. Fu, T. & Borensztajn, J. (2002). Macrophage uptake of low density lipoprotein bound to aggregated C reactive protein: possible mechanism of foam cell formation in atherosclerotic lesions. Biochemical Journal 366 195 201. 33. Torzewski, J., Torzewski, M., Bowyer, D. E., Frohlich, M., Koenig, W., Waltenberger, J., Fitzsimmons, C. & Hombach, V. (1998). C reactive protein frequently colocalizes with the terminal complement complex in the inti ma of early atherosclerotic lesions of human coronary arteries. Arteriosclerosis, Thrombosis, and Vascular Biology 18 1386 92. 34. Eisenhardt, S. U., Habersberger, J., Murphy, A., Chen, Y. C., Woollard, K. J., Bassler, N., Qian, H., von Zur Muhlen, C., Ha gemeyer, C. E., Ahrens, I., Chin Dusting, J., Bobik, A. & Peter, K. (2009). Dissociation of pentameric to monomeric C reactive protein on activated platelets localizes inflammation to atherosclerotic plaques. Circulation Research 105 128 37.

PAGE 69

56 35. Zhang, Y. X., Cliff, W. J., Schoefl, G. I. & Higgins, G. (1999). Coronary C reactive protein distribution: its relation to development of atherosclerosis. Atherosclerosis 145 375 9. 36. Ridker, P. M. (2008). High sensitivity C reactive protein as a predictor of all cause mortality: implications for research and patient care. Clinical chemistry 54 234 7. 37. Ridker, P. M. (2009). C reactive protein: eighty years from discovery to emergenc e as a major risk marker for cardiovascular disease. Clinical chemistry 55 209 15. 38. Pepys, M. B., Hirschfield, G. M., Tennent, G. A., Gallimore, J. R., Kahan, M. C., Bellotti, V., Hawkins, P. N., Myers, R. M., Smith, M. D., Polara, A., Cobb, A. J., Ley S. V., Aquilina, J. A., Robinson, C. V., Sharif, I., Gray, G. A., Sabin, C. A., Jenvey, M. C., Kolstoe, S. E., Thompson, D. & Wood, S. P. (2006). Targeting C reactive protein for the treatment of cardiovascular disease. Nature 440 1217 21. 39. Ridker, P M. (2007). C reactive protein and the prediction of cardiovascular events among those at intermediate risk: moving an inflammatory hypothesis toward consensus. Journal of the American College of Cardiology 49 2129 38. 40. Khreiss, T., Jozsef, L., Potemp a, L. A. & Filep, J. G. (2004). Conformational rearrangement in C reactive protein is required for proinflammatory actions on human endothelial cells. Circulation 109 2016 22. 41. Pasceri, V., Willerson, J. T. & Yeh, E. T. (2000). Direct proinflammatory e ffect of C reactive protein on human endothelial cells. Circulation 102 2165 8. 42. Blake, G. J. & Ridker, P. M. (2001). Novel Clinical Markers of Vascular Wall Inflammation. Circulation Research 89 763 771. 43. Khreiss, T., Jozsef, L., Hossain, S., Chan J. S., Potempa, L. A. & Filep, J. G. (2002). Loss of pentameric symmetry of C reactive protein is associated with delayed apoptosis of human neutrophils. The Journal of Biological Chemistry 277 40775 81.

PAGE 70

57 44. Agrawal, A., Simpson, M. J., Black, S., Carey M. P. & Samols, D. (2002). A C reactive protein mutant that does not bind to phosphocholine and pneumococcal C polysaccharide. The Journal of Immunology 169 3217 22. 45. Fendler, J. H. (1987). Atomic and molecular clusters in membrane mimetic chemistry. Chemical Reviews 87 877 899. 46. Sitaula, S., Mackiewicz, M. R. & Reed, S. M. (2008). Gold nanoparticles become stable to cyanide etch when coated with hybrid lipid bilayers. Chemical Communications 3013. 47. Huang, C. C., Yang, Z., Lee, K. H. & Chang, H. T. (2007). Synthesis of Highly Fluorescent Gold Nanoparticles for Sensing Mercury(II). Angewandte Chemie 119 6948 6952. 48. Panigrahi, S., Basu, S., Praharaj, S., Pande, S., Jana, S., Pal, A., Ghosh, S. K. & Pal, T. (2007). Synthesis and Size Selective Catalysis by Supported Gold Nanoparticles:€‰ Study on Heterogeneous and Homogeneous Catalytic Process. The Journal of Physical Chemistry C 111 4596 4605. 49. Templeton, A. C., Mostetler, M. J., Kraft, C. T. & Murray, R. W. (1998). Reactivity of Monolaye r Protected Gold Cluster Molecules:& Steric Effects. Journal of the American Chemical Society 120 1906 1911. 50. Frens, G. (1973). Controlled Nucleation for the Regulation of the Particle Size in Monodisperse Gold Suspensions. Nature Physical Science 241 20 22. 51. Hutter, E. & Fendler, J. €. (2004). Exploitation of Localized Surface Plasmon Resonance. Advanced Materials 16 1685 1706. 52. Sepulveda, B., Angelome, P. C., Lechuga, L. M. & Liz Marzan, L. M. (2009). LSPR based nanobiosensors. Nano Today 4 2 44 251. 53. Lakowicz, J. R. (1999). Principles of Fluroescence Spectroscopy 2 edit, Kluwer Academic / Plenum Publishers, New York.

PAGE 71

58 54. Agasti, S. S., You, C. C., Arumugam, P. & Rotello, V., M. (2008). Structural control of the monolayer stability of water soluble gold nanoparticles. Journal of Materials Chemistry 18 70 73. 55. Kumar, A., Biebuyck, H. A., Abbott, N. L. & Whitesides, G. M. (1992). The use of self assembled monolayers and a selective etch to generate patterned gold features. Journal of the A merican Chemical Society 114 9188 9189. 56. Gopidas, K. R., Whitesell, J. K. & Fox, M. A. (2003). Nanoparticle cored dendrimers: synthesis and characterization. Journal of the American Chemical Society 125 6491 502. 57. Hong, R., Fernandez, J. M., Nakade, H., Arvizo, R., Emrick, T. & Rotello, V. M. (2006). In situ observation of place exchange reactions of gold nanoparticles. Correlation of monolayer structure and stability. Chemical Communications 2347 9. 58. Richter, R P., Berat, R. & Brisson, A. R. (2006). Formation of Solid Supported Lipid Bilayers: An Integrated View. Langmuir 22 3497 3505. 59. Voet, D., Voet, J. G. & Pratt, C. W. (2008). Fundamentals of Biochemistry: Life at the Molecular Level 3rd edit, Wiley. 6 0. Nagle, J. F. & Tristram Nagle, S. (2000). Structure of lipid bilayers. Biochim Biophys Acta 1469 159 95. 61. Tanaka, M. & Sackmann, E. (2005). Polymer supported membranes as models of the cell surface. Nature 437 656 663. 62. Moyano, D. F. & Rotello, V. M. (2011). Nano meets biology: structure and function at the nanoparticle interface. Langmuir 27 10376 85. 63. Groves, J. T. & Boxer, S. G. (2002). Micropattern Formation in Supported Lipid Membranes. Accounts of Chemical Research 35 149 157. 64. Sund h, M., Svedhem, S. & Sutherland, D. S. (2011). Formation of Supported Lipid Bilayers at Surfaces with Controlled Curvatures:

PAGE 72

59 Influence of Lipid Charge. The Journal of Physical Chemistry B 115 7838 7848. 65. Roiter, Y., Ornatska, M., Rammohan, A. R., Balak rishnan, J., Heine, D. R. & Minko, S. (2009). Interaction of Lipid Membrane with Nanostructured Surfaces. Langmuir 25 6287 6299. 66. Mi, L. Z., Wang, H. W. & Sui, S. F. (1997). Interaction of rabbit C reactive protein with phospholipid monolayers studied by microfluorescence film balance with an externally applied electric field. Biophysical Journal 73 446 51. 67. Mackiewicz, M. R., Ayres, B. R. & Reed, S. M. (2008). Reversible, reagentless solubility changes in phosphatidylcholine stabilized gold nanoparticles. Nanotechnology 19 115607. 68. Baksh, M. M., Jaros, M. & Groves, J. T. (2004). Detection of molecular interactions at membrane surfaces th rough colloid phase transitions. Nature 427 139 41. 69. Yeagle, P. L. (1991). Modulation of membrane function by cholesterol. Biochimie 73 1303 10. 70. Spooner, P. J. & Small, D. M. (1987). Effect of free cholesterol on incorporation of triolein in phosp holipid bilayers. Biochemistry 26 5820 5. 71. Deng, Y., Wang, Y., Holtz, B., Li, J., Traaseth, N., Veglia, G., Stottrup, B. J., Elde, R., Pei, D., Guo, A. & Zhu, X. Y. (2008). Fluidic and air stable supported lipid bilayer and cell mimicking microarrays. Journal of the American Chemical Society 130 6267 71. 72. He, P. & Urban, M. W. (2005). Phospholipid stabilized Au nanoparticles. Biomacromolecules 6 1224 5. 73. Plant, A. L. (1999). Supported Hybrid Bilayer Membranes as Rugged Cell Membrane Mimics. Lang muir 15 5128 5135.

PAGE 73

60 74. Kundu, J., Levin, C. S. & Halas, N. J. (2009). Real time monitoring of lipid transfer between vesicles and hybrid bilayers on Au nanoshells using surface enhanced Raman scattering (SERS). Nanoscale 1 114 7. 75. Yang, J. A. & Murphy C. J. (2012). Evidence for Patchy Lipid Layers on Gold Nanoparticle Surfaces. Langmuir 76. Thaxton, C. S., Daniel, W. L., Giljohann, D. A., Thomas, A. D. & Mirkin, C. A. (2009). Templated spherical high density lipoprotein nanoparticles. Journal of the American Chemical Society Communications 131 1384 5. 77. Chen, Y. & Barkley, M. D. (1998). Toward Understanding Tryptophan Fluorescence in Proteins. Biochemistry 37 9976 9982. 78. Chou, C., Hsu, H. Y., Wu, H. T., Tseng, K. Y., Chiou, A., Yu, C. J., Lee, Z. Y. & Chan, T. S. (2007). Fiber optic biosensor for the detection of C reactive protein and the study of protein binding kinetics. Journal of Biomedical Optics 12 024025. 79. Guy, J., Caron, K., Dufresne, S. p., Michnick, S. W., Skene & Keillor, J. W. ( 2007). Convergent Preparation and Photophysical Characterization of Dimaleimide Dansyl Fluorogens: Elucidation of the Maleimide Fluorescence Quenching Mechanism. Journal of the American Chemical Society 129 11969 11977. 80. Popov, A. V., Gladkikh, V. S. & Burshtein, A. I. (2003). Stern Volmer Law in Competing Theories and Approximations. The Journal of Physical Chemistry A 107 8177 8183. 81. Eftink, M. R. & Ghiron, C. A. (1981). Fluorescence quenching studies with proteins. Analytical Biochemistry 114 199 227. 82. Eftink, M. R. & Ghiron, C. A. (1984). Indole fluorescence quenching studies on proteins and model systems: use of the inefficient quencher succinimide. Biochemistry 23 3891 3899. 83. Badley, R. A. (1975). The location of protein in serum lipo proteins: a fluorescence quenching study. Biochimica et Biophysica Acta 379 517 28.

PAGE 74

61 84. Szabo, A. G., Stepanik, T. M., Wayner, D. M. & Young, N. M. (1983). Conformational heterogeneity of the copper binding site in azurin. A time resolved fluorescence stu dy. Biophysical Journal 41 233 44. 85. Avigliano, L., Finazzi Agro, A. & Mondovi, B. (1974). Perturbation studies on some blue proteins. The Journal of the Federation of American Societies for Experimental Biology Letters 38 205 208. 86. Komiyama, T. & Miwa, M. (1980). Fluorescence quenching as an indicator for the exposure of tryptophyl residues in Streptomyces subtilisin inhibitor. The Journal of Biochemistry 87 1029 36. 87. Mackiewicz, M. R., Hodges, H. L. & Reed, S. M. (2010). C reactive protein in duced rearrangement of phosphatidylcholine on nanoparticle mimics of lipoprotein particles. Journal of Physical Chemistry B 114 5556 62. 88. Calhoun, D. B., Vanderkooi, J. M., Holtom, G. R. & Englander, S. W. (1986). Protein fluorescence quenching by smal l molecules: protein penetration versus solvent exposure. Proteins 1 109 15. 89. Skoog, D. A., West, D. M., Holler, F. J. & Crouch, S. R. (2004). Fundamentals of Analytical Chemistry 8 edit, Thomson / Brooks / Cole, Belmont. 90. Wang, M. Y., Ji, S. R., B ai, C. J., El Kebir, D., Li, H. Y., Shi, J. M., Zhu, W., Costantino, S., Zhou, H. H., Potempa, L. A., Zhao, J., Filep, J. G. & Wu, Y. (2011). A redox switch in C reactive protein modulates activation of endothelial cells. The Journal of the Federation of A merican Societies for Experimental Biology 25 3186 96. 91. Basu, S., Pande, S., Jana, S., Bolisetty, S. & Pal, T. (2008). Controlled Interparticle Spacing for Surface Modified Gold Nanoparticle Aggregates. Langmuir 24 5562 5568. 92. Liu, X., Atwater, M., Wang, J. & Huo, Q. (2007). Extinction coefficient of gold nanoparticles with different sizes and different capping ligands. Colloids Surfaces B: Biointerfaces 58 3 7.

PAGE 75

62 93. Friedman, M. L., Schlueter, K. T., Kirley, T. L. & Halsall, H. B. (1985). Fluoresce nce quenching of human orosomucoid. Accessibility to drugs and small quenching agents. Biochemical Journal 232 863 7.