University of Southern California Dissertations and Theses

Vitronectin misfolding and aggregation: implications for the pathophysiology of age-related diseases

(USC Thesis Other)

Vitronectin misfolding and aggregation: implications for the pathophysiology of age-related diseases

PDF

Download

Open document

Flip pages

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content

VITRONECTIN MISFOLDING AND AGGREGATION: IMPLICATIONS FOR
THE PATHOPHYSIOLOGY OF AGE-RELATED DISEASES

by

Thuzar Myo Shin

____________________________________________________________________

A Dissertation Presented to the
FACULTY OF THE GRADUATE SCHOOL
UNIVERSITY OF SOUTHERN CALIFORNIA
In Partial Fulfillment of the
Requirements for the Degree
DOCTOR OF PHILOSOPHY
(PATHOBIOLOGY)

August 2007

Copyright 2007 Thuzar Myo Shin

ii
Table of Contents
Dedication
Acknowledgements
List of Tables
List of Figures
Abstract
Chapter I Introduction
Chapter II Purification and characterization of recombinant
human vitronectin

Chapter III Formation of soluble oligomers and amyloid fibrils
by the multifunctional protein vitronectin
Chapter IV Analysis of vitronectin aggregation in human
atherosclerotic plaques
Concluding Remarks
Bibliography

1
19
38
68
iii
vi
ix
iv
v
84
86

iii
Dedication

To my parents with love and thanks

iv
Acknowledgements
This work would not have been possible without the guidance of my mentors,
Drs. Jeannie Chen and Ralf Langen. In addition to their advice and direction
throughout my academic program, they supported me through several difficult
family events. I would also like to thank past and present members of the Chen and
Langen labs. Each one of them has helped me along the way, and I am lucky to have
worked with such kind and considerate folks. To committee members Dr. David
Hinton, Dr. Michael Stallcup, and Dr. Timothy Triche, thank you for your time,
thoughtful discussions, suggestions, and words of encouragement. Lastly, I would
like to thank my parents, sisters, family, and friends for loving me, supporting me,
and always believing in me.

v
List of Tables
Table 3.1 Optimization of vitronectin oligomer and fibril formation.
Table 4.1 Estimated amounts of vitronectin as a fraction of total protein in
atherosclerotic plaques.
79
43

vi
List of Figures
Figure 1.1 Genomic organization and amino acid sequence of
vitronectin.
Figure 1.2 Vitronectin is present in disease-associated plaques.
Figure 1.3 Fundoscopic photographs of the human retina and
schematic depiction of drusen formation.
Figure 1.4 Age-related amyloid diseases and their associated
proteins.
Figure 1.5 Human ocular drusen contain toxic nonfibrillar amyloid
oligomers.
Figure 1.6 Typical oligomers and fibrils formed by amyloid β.
Figure 1.7 Proposed relationship between soluble oligomers and
fibrils.
Figure 2.1 Map of pSE420 vector used for prokaryotic expression
of full-length human vitronectin.
Figure 2.2 Representative standards for circular dichroism
spectroscopy and secondary structure of plasma-purified and
recombinant vitronectin.
Figure 2.3 Agarose gel electrophoresis of pSE420 vitronectin
inserts digested with NcoI and HindIII.
Figure 2.4 Representative chromatogram from the purification of
recombinant vitronectin expressed in Escherichia coli DH5 α cells.
Figure 2.5 Affinity-purified recombinant vitronectin extracted
from inclusion bodies.
Figure 2.6. ELISA of recombinant vitronectin is comparable to
commercially-available human vitronectin purified from plasma.
Figure 2.7. Recombinant vitronectin promotes RPE cell
attachment.
Figure 2.8. Representative chromatogram from heparin affinity
purification of vitronectin from human serum.

2
22
26
28
29
30
6
11
8
10
15
17
33
34

vii
Figure 2.9. Fractions collected during heparin affinity
chromatography of vitronectin purification from human serum.
Figure 3.1 A cell-free, fluorescence-based assay to monitor
membrane leakage.
Figure 3.2 Human vitronectin forms amyloid fibrils and oligomers.
Figure 3.3 Vitronectin oligomers and fibrils enhance thioflavin T
fluorescence.
Figure 3.4 Vitronectin forms soluble nonfibrillar amyloid
oligomers.
Figure 3.5 Soluble nonfibrillar vitronectin oligomers compromise
cell viability.
Figure 3.6 Vitronectin oligomers induce membrane leakage.
Figure 3.7 Vitronectin fibrils contain a protease-resistant core.
Figure 3.8 Domain structure of vitronectin.
Figure 3.9 Gel filtration of recombinant vitronectin oligomers
using the Superdex 75 column.
Figure 3.10 Gel filtration of recombinant vitronectin oligomers
using the Superdex 200 column.
Figure 3.11 Colorimetric TUNEL assay of cultured RPE cells
treated with recombinant vitronectin oligomers.
Figure 4.1 Protocol for the extraction of extracellular, cytoplasmic,
membrane-associated, and insoluble proteins from human
atherosclerotic plaque material.
Figure 4.2 Soluble and insoluble protein fractions extracted from
human basilar artery plaque.
Figure 4.3 Vitronectin in present in the insoluble fraction of human
basilar artery plaque.
Figure 4.4 Atherosclerotic plaque proteins fractionated according
to cellular compartment

50
51
55
59
67
62
65
58
66
53
37
72
76
77
81
47

viii
Figure 4.5 Vitronectin immunoblot of human atherosclerotic
plaque proteins separated by cellular compartment.
82

ix
Abstract
Classic amyloidopathies are characterized by the accumulation of misfolded
proteins in the form of plaques. Whilst fibril deposition is pathognomonic for
amyloidopathies, recent data suggest that prefibrillar species may mediate the
development and progression of disease. Soluble amyloid oligomers exhibit similar
morphologic and cytotoxic properties, suggesting a shared pathogenic mechanism
among amyloid proteins. The oligomer-specific A11 antibody is a useful tool to
study amyloid diseases which lack abundant fibril deposition. As part of a
collaborative study on the pathophysiology of age-related macular degeneration, we
studied protein misfolding in human ocular drusen and demonstrated the presence
of prefibrillar oligomers using the A11 antibody. However, the oligomer-forming
protein in drusen has not yet been identified. Historically, the amyloidogenic
protein is oftentimes the most abundant protein near or within the disease plaque.
Vitronectin is one of the most abundant drusen proteins, is contained in all drusen,
and is also present within the insoluble deposits associated with Alzheimer disease,
atherosclerosis, systemic amyloidoses, and glomerulonephritis. These deposits
stain positive with thioflavin, indicating an underlying protein misfolding process.
The extent to which vitronectin contributes to amyloid formation within these
plaques and the role of vitronectin in the pathophysiology of the aforementioned
diseases is currently unknown. The investigation of vitronectin misfolding and
aggregation is significant since the formation of oligomers and fibrils is a common
ability of amyloid proteins, although they share neither sequence nor native

x
structural homology. In this study, we tested the hypothesis that vitronectin is
amyloidogenic. Our results demonstrate that human vitronectin readily forms
spherical oligomers and typical amyloid fibrils. Vitronectin oligomers are toxic to
cultured cells, possibly via a membrane-dependent mechanism, as they cause
leakage of synthetic vesicles. Vitronectin fibrils contain a C-terminal protease-
resistant domain which likely contains the essential residues for amyloid formation.
Consistent with our data that illustrate the inherent propensity of vitronectin to self-
associate in vitro, vitronectin fragments were detected in the insoluble fraction of
human atherosclerotic plaques, suggesting that vitronectin aggregation likely occurs
in vivo. Our results put forth the possibilities that accumulation of misfolded
vitronectin may contribute to amyloid plaque formation and the pathophysiology of
age-related amyloid diseases.

1
Chapter I. Introduction
1.1 Vitronectin structure and function
1.1.1. Genomic organization and sites of synthesis
The human vitronectin gene is located on chromosome 17. It is comprised of
eight exons and seven introns, and encodes a mature 75 kDa polypeptide of 459
amino acids (Jenne and Stanley 1987). The genomic organization, amino acid
sequence and functional domains of vitronectin are depicted in Figure 1.1.
Vitronectin is present in blood, urine, and amniotic fluid at a concentration of 0.2-
0.45 mg/ml, constituting 0.1-0.5% of plasma protein (Preissner, Wassmuth et al.
1985; Kitagaki-Ogawa, Yatohgo et al. 1990). The liver is the major source of
vitronectin, although sites of extrahepatic synthesis, including retina, brain, heart,
and skeletal muscle, have also been reported (Seiffert, Crain et al. 1994; Anderson,
Hageman et al. 1999; Hageman, Mullins et al. 1999; Ozaki, Johnson et al. 1999).
Two published versions of the vitronectin cDNA sequence reveal a genetic
polymorphism in the nucleotides encoding amino acid 381 (Suzuki, Oldberg et al.
1985; Jenne and Stanley 1987). The presence of a threonine versus methionine at
position 381 increases the susceptibility to proteolytic cleavage of vitronectin at Arg
379 (Tollefsen, Weigel et al. 1990). This results in a mixture of the 75 kDa single-
chain form and the 65+10 kDa two-chain form (linked by a disulfide bond),
producing a characteristic two-band pattern upon reducing sodium dodecyl sulfate
polyacrylamide gel electrophoresis (SDS-PAGE), which separates proteins by
molecular weight. Three vitronectin phenotypes were described by Sun and Mosher

2

Figure 1.1 Genomic organization and amino acid sequence of vitronectin [figure
from (Preissner 1991)]. Top, Genomic organization of the vitronectin gene, located
on chromosome 17. The eight exons comprising the gene (black) and the
corresponding protein domains are depicted. The signal sequence (-19 to 1) is not
present in the mature polypeptide of 459 amino acids. The somatomedin B is
contained within the hatched box. Hexagons represent targets of N-glycosylation,
and SO
4
and PO
4
indicate sulfation and phosphorylation sites, respectively. The
arrow denotes the site of proteolytic cleavage to generate 65 + 10 kDa vitronectin.
Bottom, Vitronectin amino acid sequence. Several functional domains are
highlighted: somatomedin B domain/ PAI-1 binding D1-T44 (aqua), RGD peptide
(yellow), plasminogen binding site A332-K348 (blue), polycationic heparin-binding
domain, K348-R379, involved in the inhibition of cell lysis by MAC and perforin
pore formation (green), and putative amyloid core A380-R427 (red, see Chapter 3).

3
(Sun and Mosher 1989): type 1-1 (61-90% single-chain), type 1-2 (38-60% single-
chain) and type 2-2 (10-26% single-chain). Type 1-1 individuals are homozygous
for the Met 381 allele, and type 2-2 individuals are homozygous for the Thr 381
allele (Tollefsen, Weigel et al. 1990). A recent study suggests that the two-chain
form of vitronectin may be produced by furin cleavage in the liver (Seger and
Shaltiel 2000), although it is still unclear whether other proteases may be involved.
No functional differences between the two isoforms were identified in vitro (Gibson
and Peterson 2001). Other reported post-translational modifications include N-
glycosylation, sulfation, phosphorylation, and transglutaminase crosslinking.
1.1.2. Normal functions in homeostasis
Vitronectin is a multifunctional protein involved in a variety of physiologic
processes. The N-terminal somatomedin B domain (residues 1-42) binds
plasminogen activator inhibitor 1 (PAI-1) and regulates fibrinolysis (Seiffert,
Ciambrone et al. 1994). Phosphorylation of single-chain vitronectin at Ser 378 by
platelet-released protein kinase A (McGuire, Peacock et al. 1988; Mehringer, Weigel
et al. 1991) modulates the interaction between vitronectin and PAI-1 (Schvartz,
Kreizman et al. 2002). The RGD sequence (Arg-Gly-Asp, residues 45-47) binds a
variety of integrin receptors (including αvβ
3
, αvβ
5
,

αvβ
1
,

αvβ
6
, αv β
8
, αIIb β
3
) and
mediates cell attachment, spreading, and migration (Cherny, Honan et al. 1993;
Schvartz, Seger et al. 1999). Vitronectin contains a crosslinking site involved in
multimer assembly at Gln 93 (Skorstengaard, Halkier et al. 1990; Stockmann, Hess
et al. 1993), a collagen-binding domain (Izumi, Shimo-Oka et al. 1988; Ishikawa-

4
Sakurai and Hayashi 1993), and hemopexin-like repeats (Stanley 1986). The C-
terminal heparin-binding domain consists of a 40-residue polycationic cluster within
a cryptic site that is exposed upon conformational changes induced by heparin,
interaction with protein complexes such as thrombin-antithrombin III, adsorption to
surfaces, or denaturation (Suzuki, Pierschbacher et al. 1984; Preissner and Muller-
Berghaus 1987; Preissner 1991; Bittorf, Williams et al. 1993). Vitronectin prevents
both complement-induced cell lysis and perforin pore formation via its highly basic
heparin-binding domain (Dahlback and Podack 1985; Tschopp, Masson et al. 1988;
Su 1996). The biological structure of vitronectin has been studied using a number of
biophysical methods, including x-ray crystallography of truncated fragments, and
small-angle scattering. Infrared spectroscopy and circular dichroism show that the
secondary structure of native vitronectin contains both random coil and β-sheet
elements, with α-helical content (Pitt W.G. 1989). However, no high-resolution
structure of full-length vitronectin exists.
1.1.3. The role of vitronectin in disease
The significance of vitronectin as a multifunctional protein involved in many
physiologic processes has been established. However, the role of vitronectin in
disease in not well understood, though it has been implicated in a number of disease
processes. Vitronectin expression is upregulated in animal models of acute and
chronic inflammation, (Seiffert, Geisterfer et al. 1995) and in fibrotic tissues (Reilly
and Nash 1988; Kobayashi, Yamada et al. 1994). In addition, serum levels of
vitronectin are elevated in patients with atherosclerosis (Ekmekci, Sonmez et al.

5
2002), type 2 diabetes (Zhang, Barker et al. 2004), and Alzheimer disease (AD)
(Zhang, Barker et al. 2004). Vitronectin protein has been identified in deposits
associated with AD, age-related macular degeneration (AMD), atherosclerosis,
systemic amyloidoses, and glomerulonephropathy (Dahlback, Lofberg et al. 1987;
Dahlback, Lofberg et al. 1988; Guettier, Hinglais et al. 1989; Niculescu, Rus et al.
1989; Akiyama, Kawamata et al. 1991; Okada, Yoshioka et al. 1993; Eikelenboom,
Zhan et al. 1994; Ogawa, Yorioka et al. 1994; van Aken, Seiffert et al. 1997;
Mullins, Russell et al. 2000; Johnson, Leitner et al. 2001; Robert, Jacobin-Valat et al.
2006). Examples of vitronectin immunostaining in disease aggregates from patients
with AD, AMD, and atherosclerosis are shown in Figure 1.2. These insoluble
plaques exhibit thioflavin staining (Vallet, Guntern et al. 1992; Anderson, Talaga et
al. 2004; Rocken, Tautenhahn et al. 2006), indicating an underlying process of
protein misfolding and amyloid formation. The role of vitronectin in amyloid
formation within these aggregates and its contribution to the pathophysiology of the
aforementioned diseases is currently unknown. As part of a collaborative USC study
on the pathophysiology of age-related macular degeneration, we are studying protein
misfolding in human ocular drusen.
1.2 Drusen and age-related macular degeneration
1.2.1. Clinical aspects of age-related macular degeneration
Early age-related maculopathy (ARM) is defined as the presence of soft drusen
≥63 µm in diameter and retinal pigment epithelium (RPE) irregularities in the
macular area of a patient ≥ 50 years of age (Bird, Bressler et al. 1995), according to

6

Figure 1.2 Vitronectin is present in disease-associated plaques. Upper panels
Immunohistochemistry of vitronectin in human brain tissue (Akiyama, Kawamata et
al. 1991). Vitronectin staining is observed around blood vessels in control grey
matter cortex (left). In the AD brain, vitronectin staining is observed in both tangled
neurons (arrowheads) and plaques (arrows). Lower panels Immunostaining of a
human ocular druse (left, asterisk) (Hageman, Mullins et al. 1999) and
atherosclerotic plaque from human carotid artery (right, arrowheads) (van Aken,
Seiffert et al. 1997) demonstrate the presence of vitronectin in these insoluble
deposits.

7
an international classification scheme. Late ARM is a term reserved for advanced
age-related macular degeneration (AMD), characterized by geographic atrophy
(nonexudative or dry AMD) and/or choroidal neovascularization (exudative or wet
AMD). Figure 1.3 shows representative fundoscopic photographs of normal and
AMD-affected retinas. AMD is the leading cause of vision loss in older Americans.
A recent study reports that the estimated prevalence of ARM or AMD in individuals
≥ 55 years of age is more than 8 million (Bressler, Bressler et al. 2003). An
estimated 1.75 million individuals, comprising 1.47% of the population age 40 or
more, have advanced AMD (Friedman, O'Colmain et al. 2004). With the aging of
the Baby Boomer generation, this number is projected to increase to almost 3 million
by the year 2020 (Friedman, O'Colmain et al. 2004). Two consistently demonstrable
risk factors for AMD are increasing age and smoking. Caucasians and those with
family history of the disease also appear to be at increased risk for AMD. Although
there is no treatment for geographic atrophy, patients with neovascular disease may
undergo laser photocoagulation or photodynamic therapy. However, the disease and
accompanying vision loss may progress despite treatment. While there is no cure for
AMD, a recent nationwide, double-blind clinical trial showed that high-dose
antioxidant and zinc supplements can help slow the progression from early to late
AMD (AREDS 2001). To date, this is the only proven intervention to delay the
onset of advanced disease, although its ability to prevent the development of AMD
altogether has not yet been investigated.
Drusen are lipoproteinaceous deposits that accumulate between the basal lamina

8

Figure 1.3 Fundoscopic photographs of the human retina and schematic depiction of
drusen formation. Upper panel A, Photograph of a normal human retina
(http://webvision.med.utah.edu/sretina.html). Arrows point to the optic nerve and
the fovea, which is located at the center of the macula. B, Retina of a patient with
AMD (http://dro.hs.columbia.edu/amd/htm). Note the presence of numerous
yellowish drusen in the macular area, the first clinically detectable hallmarks of
AMD. Lower panel The upper diagram demonstrates an intact RPE/Bruch
membrane complex in a normal retina. The lower diagram represents what occurs in
the AMD eye. Drusen (yellow) accumulate between the basal lamina of the RPE
(Bl) and Bruch membrane (Bm). The overlying RPE is displaced, resulting in
degeneration and cell death.

9
of the RPE and Bruch membrane. A schematic diagram illustrating the location of
drusen formation is shown in Figure 1.3. Although drusen are the first clinically
detectable hallmarks of AMD, the pathophysiology of drusen formation is not well
understood. Hard drusen are < 63 µm in diameter with discrete edges, while soft
drusen are ≥ 63 µm with indistinct margins (Bird, Bressler et al. 1995). Whilst hard
drusen are thought to be associated with normal aging and have been removed from
the diagnostic criteria for ARM, soft drusen or numerous hard drusen are considered
major risk factors for developing AMD (Bird, Bressler et al. 1995). One proposed
mechanism for soft drusen formation is the fusion of one or more small, hard drusen
(Sarks, Sarks et al. 1994), implying that hard drusen may also progress to ARM and
AMD. A limiting factor in establishing the significance of hard drusen is that there
are no population-based studies that establish the number of drusen that is normal as
a function of age.
1.2.2. Drusen contain toxic nonfibrillar oligomers
Several important similarities exist between the drusen characteristic of AMD and
protein aggregates associated with age-related degenerative disorders, such as AD,
Type 2 diabetes, and Parkinson disease. Figure 1.4 demonstrates several diseases
wherein the age-related accumulation of these insoluble deposits appears to be either
associated with or a risk factor for disease. Although they share no sequence
homology, the proteins implicated in these diseases possess the ability to form
spherical and protofibrillar oligomers and amyloid fibrils. Studies in our laboratory
have demonstrated that human ocular drusen stain positively with a conformational

10

Figure 1.4 Age-related amyloid diseases and their associated proteins. Classic
amyloidopathies are characterized by accumulation and aggregation of misfolded
proteins in the form of plaques, which contain a variety of proteins and lipids. One
pathological hallmark is fibril deposition. Amyloid fibrils are ordered entities with
characteristic dye binding properties. They are approximately 5-10 nm in width and
possess a cross-beta ultrastructure. These disease-associated proteins have no
sequence homology, but all amyloid proteins have a shared ability to form
morphologically similar spherical and protofibrillar oligomers and fibrils.

11

Figure 1.5 Human ocular drusen contain toxic nonfibrillar amyloid oligomers.
Bright field (A, C) and confocal images (B, D) of human ocular drusen (Luibl, Isas et
al. 2006), which exhibit A11 immunoreactivity (B, D, green), demonstrating the
presence of soluble amyloid oligomers. Note the degeneration of the overlying RPE
(red). No staining was observed in age-matched controls (data not shown),
suggesting that nonfibrillar oligomers are specific to drusen. Scale bar = 10 μm.

12
antibody that is specific for toxic oligomers (Figure 1.5). No staining was observed
in age-matched controls. These data reveal that drusen contain nonfibrillar
oligomers, suggesting that protein misfolding may contribute to drusen formation.
The existence of amyloid formation suggests that the development and progression
of AMD and the aforementioned age-related degenerative diseases may occur via a
similar mechanism.
1.2.3. Vitronectin is a candidate for protein misfolding in drusen
The presence of toxic soluble oligomers in drusen illustrates amyloid formation
within these aggregates, which may in turn compromise RPE cell viability.
Historically, the most abundant protein within or near the disease plaque has been
associated with a specific amyloidopathy (Figure 1.4). A number of studies have
demonstrated that vitronectin is a major protein component of and present in all
human ocular drusen (Hageman, Mullins et al. 1999; Hageman, Luthert et al. 2001;
Johnson, Leitner et al. 2001; Crabb, Miyagi et al. 2002). An example of vitronectin
immunostaining in a human ocular druse is shown in Figure 1.2 (lower left panel).
Ambati and colleagues recently demonstrated the presence of vitronectin in drusen-
like deposits in an animal model of age-related macular degeneration (Ambati,
Anand et al. 2003). However, further investigation is needed to elucidate whether
vitronectin deposition is primary or secondary to inflammatory stimuli in that
particular disease model. Since RPE cells are closely associated with drusen and
rich in vitronectin mRNA (Anderson, Hageman et al. 1999; Hageman, Mullins et al.
1999; Ozaki, Johnson et al. 1999), they may be a primary source of vitronectin

13
deposition. Due to its widespread distribution, local synthesis by RPE, and
deposition in drusen, vitronectin is an ideal candidate protein to study drusen
pathogenesis.
1.3 Background on amyloid diseases
1.3.1. Characteristics of amyloidopathies and associated proteins
The term “amyloid” was coined by the pathologist Rudolph Virchow, which he
used to describe a specific reaction of iodine with abnormal macroscopic tissue
samples. In 1959, Cohen and Calkins observed the existence of fibrils of 80-100 Ǻ
width and indeterminate length in ultrathin sections of amyloid-laden tissue using
electron microscopy (Cohen and Calkins 1959). Subsequently, the criteria for the
definition of amyloid were established: Congo red staining with apple-green
birefringence under polarized light, and fibrillar morphology by negative-stain
electron microscopy. Solubilization of isolated fibrils in guanidine hydrochloride
and subsequent gel filtration enabled the identification of the unique protein
associated with each amyloidopathy [reviewed in (Sipe and Cohen 2000)]. Fibrils
extracted from tissue are identical to those formed by synthetic peptides (Glenner,
Eanes et al. 1974), justifying the study of amyloid proteins in vitro.
The investigation of vitronectin misfolding and aggregation is significant since
the formation of spherical and protofibrillar oligomers and fibrils is a common
ability of amyloid proteins, although they share neither sequence nor native
structural homology. Representative fibrils and oligomers formed by amyloid β
(A β), as seen under transmission electron microscopy (TEM), are shown in Figure

14
1.6. Amyloid fibrils are typically 8-10 nm in diameter, bind dyes such as Congo red
and thioflavin, and adopt a characteristic cross- β structure (Sunde, Serpell et al.
1997; Sipe and Cohen 2000). Soluble oligomers formed in vitro from a variety of
proteins are morphologically similar and appear spherical or protofibrillar. Toxicity
to cultured cells is a shared property of amyloid oligomers (Hartley, Walsh et al.
1999; Janson, Ashley et al. 1999; Chiti, Bucciantini et al. 2001; Baskakov, Legname
et al. 2002; Bucciantini, Giannoni et al. 2002; Chromy, Nowak et al. 2003; Hoshi,
Sato et al. 2003; Kayed, Head et al. 2003; Bucciantini, Calloni et al. 2004; Zhu, Han
et al. 2004).
1.3.2. Evidence supporting a common mechanism of pathophysiology among
amyloid diseases
While historically viewed as disease-causing entities, there are conflicting reports
in the literature regarding the toxicity of fibrils (Janson, Ashley et al. 1999;
Novitskaya, Bocharova et al. 2006). A growing body of evidence indicates that
soluble prefibrillar oligomers may be the primary pathogenic species in
amyloidopathies (Lue, Kuo et al. 1999; McLean, Cherny et al. 1999; Walsh, Klyubin
et al. 2002; Barghorn, Nimmrich et al. 2005; Cleary, Walsh et al. 2005). For
instance, the concentration of soluble A β in human AD brains has a higher
correlation with clinically-established disease severity than plaque burden (Lue, Kuo
et al. 1999; McLean, Cherny et al. 1999). A β oligomers impair long-term
potentiation (Walsh, Klyubin et al. 2002; Barghorn, Nimmrich et al. 2005) and
disrupt cognitive function in vivo (Cleary, Walsh et al. 2005). The structural

15

Figure 1.6 Typical oligomers and fibrils formed by amyloid β, as seen under
transmission electron microscopy. A β, a protein implicated in Alzheimer disease,
forms soluble spherical oligomers (left) and typical amyloid fibrils (right). The
ability to form oligomers and fibrils is characteristic of amyloid proteins, though they
share neither sequence nor structural homology. Whilst fibril deposition is
pathognomonic for amyloidopathies, recent data suggest that oligomers may be
pathogenic species that mediate the development and progression of amyloid
diseases. Scale bar = 100 nm. Used with permission from J. Mario Isas.

16
similarity and cytotoxicity of soluble nonfibrillar oligomers suggest a common
pathogenic mechanism among amyloid proteins and their associated diseases
(Hartley, Walsh et al. 1999; Janson, Ashley et al. 1999; Chiti, Bucciantini et al.
2001; Baskakov, Legname et al. 2002; Bucciantini, Giannoni et al. 2002; Chromy,
Nowak et al. 2003; Hoshi, Sato et al. 2003; Kayed, Head et al. 2003; Bucciantini,
Calloni et al. 2004; Zhu, Han et al. 2004). A proposed relationship between soluble
oligomers and amyloid fibrils is shown in Figure 1.7 (Glabe 2004).
The development of a conformation-specific, sequence-independent antibody that
recognizes soluble prefibrillar oligomers made from a number of proteins, but not
monomers or fibrils, has aided the analysis of these toxic entities (Kayed, Head et al.
2003). The A11 antibody facilitates the characterization of soluble oligomers as
markers of amyloid diseases and enables the identification of novel amyloid diseases
wherein there is oligomer accumulation without abundant fibril deposition. One
such disease is desmin-related cardiomyopathy (Sanbe, Osinska et al. 2004; Sanbe,
Osinska et al. 2005). Our recent report that human ocular drusen contain nonfibrillar
oligomers implies that age-related macular degeneration may also represent this type
of amyloidosis (Luibl, Isas et al. 2006). Vitronectin is an appealing candidate for
amyloid formation due to its widespread distribution in the body, abundance in
drusen, and its association with insoluble, disease-associated plaques. In this study,
we aim to demonstrate that vitronectin behaves as an amyloid protein in vitro.
Chapter II outlines the purification and characterization of recombinant human
vitronectin. Chapter III establishes vitronectin as an amyloid protein, as

17

Figure 1.7 Proposed relationship between soluble oligomers and fibrils (Glabe
2004). Oligomer and fibril formation are common properties of amyloid proteins.
This diagram presents one hypothesis regarding the role of oligomers in fibril
assembly. In this example, a misfolded or natively unfolded monomeric protein self-
associates to form soluble oligomers which serve as building blocks for the assembly
of protofibrils and amyloid fibrils. The development of a conformation-specific,
sequence-independent antibody has facilitated the detection of soluble oligomers in
vitro and in situ.

18
demonstrated by the formation of soluble oligomers and amyloid fibrils. Chapter IV
presents the initial investigation of vitronectin aggregation in vivo using
atherosclerotic plaques from the human basilar artery. The inherent amyloidogenic
propensity of vitronectin and toxicity of nonfibrillar vitronectin oligomers suggests
that vitronectin misfolding and aggregation may contribute to the pathophysiology of
age-related diseases.

19
Chapter II. Purification and characterization of recombinant human
vitronectin
2.1. Introduction/Rationale
Previous studies of full-length human vitronectin used protein purified from
human plasma, conditioned medium from transfected insect or mammalian cells, or
GST-tagged recombinant protein (Yatohgo, Izumi et al. 1988; Izumi, Yamada et al.
1989; Gibson and Peterson 2001; Wojciechowski, Chang et al. 2004). While no
functional differences between the vitronectin isoforms have been established in
vitro (Gibson and Peterson 2001), the inherent heterogeneity of plasma-purified
vitronectin renders human plasma as a source of protein less than ideal for the study
of vitronectin amyloid formation. Because the amyloidogenic propensity of
vitronectin is unknown, large quantities of protein are required to empirically
determine the appropriate solution conditions for oligomer and fibril formation.
Commercially-available vitronectin purified from human plasma is expensive, and
purchasing the large amounts necessary to conduct this study is neither practical nor
cost-effective. Moreover, vitronectin from one source consistently displays
immunoreactivity when incubated in secondary antibody alone, indicating the
presence of a contaminant. It has been observed that the heparin-binding fraction of
vitronectin purified from human plasma contains proteolytic activity and can cleave
intact 75 kDa vitronectin to the 65 + 10 kDa form (Izumi, Yamada et al. 1989),
further discouraging the use of plasma-purified protein.
In order to obtain a homogeneous, cost-effective, and renewable supply of

20
protein, we generated full-length, untagged human vitronectin for use in this research
project. A plasmid vector containing human vitronectin cDNA was constructed for
use in a prokaryotic protein expression system and a purification protocol based on a
previously published method was developed (Wojciechowski, Chang et al. 2004).
We chose the Met 381 allele in order to generate the full-length 75 kDa polypeptide
that is less susceptible to proteolytic cleavage.

2.2. Materials and Methods
2.2.1. Cloning, expression, and purification of full-length human vitronectin
A plasmid containing human vitronectin cDNA was purchased from American
Tissue Culture Company (ATCC; Rockville, MD) and a 1,385 bp fragment of
mature, full-length human vitronectin cDNA containing engineered N-terminal NcoI
and C-terminal HindIII restriction sites was synthesized by polymerase chain
reaction (PCR) using the following primer pair: 5’ CAT GCC ATG GAC CAA GAG
TCA TGC AAG GGC 3’ (sense) and 5’ CCC AAG CTT CTA CAG ATG GCC
AGG AGC TGG 3’ (antisense). Following agarose gel electrophoresis of the PCR
product, the 1.4 kb fragment was extracted using the QIAquick Gel Extraction Kit
(Qiagen; Valencia, CA) and digested overnight at 37
o
C with NcoI and HindIII
restriction enzymes (New England Biolabs; Ipswich, MA). The resulting two
fragments were separated by agarose gel electrophoresis, gel extracted, and
subcloned into an NcoI and HindIII-digested, dephosphorylated, pSE420 vector
(Figure 2.1; Invitrogen, Carlsbad, CA) containing an ampicillin resistance gene by

21
three-piece ligation. Following successful transformation and selection of
Escherichia coli DH5 α cells with the ligated construct, plasmid DNA was extracted
from ampicillin-resistant colonies using the QIAprep Spin Miniprep Kit (Qiagen;
Valencia, CA). Correct orientation of the insert was verified by DNA sequencing at
the USC/Norris Microchemical Core Facility.
DH5 α cells were transformed with the 4.27 kb construct and grown with shaking
at 37
o
C to an optical density of 0.6-0.8. Protein expression was induced by the
addition of 1 mM isopropyl-1-thio- β-D-galactopyranoside (IPTG) for 4 hours at
37
o
C. Cells were harvested by centrifugation at 5,000 x g for 15 minutes and pellets
were stored at -80
o
C. Vitronectin protein was purified using a previously described
protocol (Wojciechowski, Chang et al. 2004) with modifications. The pellet was
resuspended in lysis buffer [5 mM EDTA, 10 mM dithiothreitol (DTT), 1 mg/mL
lysozyme, and protease inhibitors, all in 10 mM sodium phosphate buffer, pH 7.4],
incubated on ice for 1 hour, then nutated at 4
o
C for 10 minutes with the addition of
24 U/mL DNase I, 60 mM MgCl
2
, and 1% Triton X-100. The suspension was
sonicated 3 x 1 minute on ice, and centrifuged at 20,000 x g for 30 minutes. The
pellet was resuspended in 2 M urea buffer (2 M urea, 5 mM EDTA, 10 mM DTT,
and protease inhibitors, all in 10 mM sodium phosphate buffer, pH 7.4), sonicated on
ice, centrifuged, and the supernatant decanted. The resulting pellet of inclusion
bodies was solubilized in cold 8 M urea buffer (8 M urea, 5 mM EDTA, 10 mM
DTT, and protease inhibitors, all in 10 mM sodium phosphate buffer, pH 7.4),
sonicated on ice, and centrifuged. The supernatant was applied to a HiTrap heparin

22

Figure 2.1 Map of pSE420 vector used for prokaryotic expression of full-length
human vitronectin. The superlinker was removed by digesting the vector with NcoI
and HindIII restriction enzymes and vitronectin cDNA with complementary sticky
ends was introduced. Ampicillin selection was used to identify plasmid-containing
colonies.

23
affinity column (Amersham Biosciences; Piscataway, NJ) equilibrated with 8 M urea
buffer and proteins were eluted using a linear NaCl gradient. Vitronectin-containing
fractions were identified by SDS-PAGE followed by Coomassie staining, pooled,
and dialyzed against 1% acetic acid, 10 mM phosphate buffer (pH 7.4), or
phosphate-buffered saline (PBS). Protein purity was >95% by Coomassie Blue
staining. Protein concentration was calculated from absorbance at 280 nm using
Beer’s Law (A = ε
M
cl; where ε
M
= 82,330 M
-1
cm
-1
).
2.2.2. Enzyme-Linked ImmunoSorbent Assay (ELISA)
An ELISA was performed to assess the quality of recombinant vitronectin and
compare the immunoreactivities of plasma-purified and recombinant vitronectin to a
polyclonal anti-vitronectin antibody. Plasma-purified and recombinant vitronectin
were diluted in 0.1 M sodium bicarbonate, pH 9. Indicated amounts of protein were
added to a 96-well microplate and allowed to coat for 2 hours at 37
o
C. The plate was
washed three times with PBS containing 0.01% Tween 20 (PBS-T), blocked for 2
hours at 37
o
C in 3% BSA/ PBS-T, and washed three times in PBS-T. Primary anti-
vitronectin antibody (diluted 1:10,000 in 3% BSA/ PBS-T; Santa Cruz
Biotechnology, Santa Cruz, CA) was added to each well for 1 hour at 37 °C, washed
three times in PBS-T, followed by incubation in horseradish peroxidase-conjugated
anti-rabbit IgG (diluted 1:10,000 in 3% BSA/ PBS-T; Promega, Madison, WI) for 1
hour at 37 °C. After three washes in PBS-T, the plate was developed with 3, 3, 5, 5 -
tetramethylbenzidine (KPL, Gaithersburg, MD). The reaction was stopped by the
addition of 1 N HCl and absorbance was measured at 450 nm. Background

24
absorbance from uncoated wells was subtracted from each experimental condition.
2.2.3. Cell adhesion assay
To ascertain whether recombinant vitronectin promotes cellular attachment like
the native protein, a cell adhesion assay was performed. A previously described cell
adhesion assay (Jin, He et al. 2000) was employed with modifications. A 96-well
plate was coated overnight at 4
o
C with 200 ng/cm
2
BSA, plasma-purified human
vitronectin, or recombinant vitronectin diluted in phosphate-buffered saline (PBS).
Plates were rinsed three times with PBS, blocked for one hour in 1% BSA/ PBS at
room temperature, rinsed again, and allowed to air-dry. Retinal pigment epithelium
(RPE) cells were isolated from human fetal eyes obtained from Advanced Bioscience
Resources, Inc. (Alameda, CA), as described previously (Jin, He et al. 2000), and
grown in complete medium [Dulbecco’s modified Eagle’s medium (DMEM, VWR;
West Chester, PA) supplemented with 2 mM L-glutamine, 100 U/ml penicillin, 100
µg/ml streptomycin, and 10% fetal bovine serum (Invitrogen; Carlsbad, CA)] at
37
o
C in a humidified incubator. When confluency was achieved, RPE cells were
harvested and resuspended in serum-free medium and 2 x 10
4
cells were added to
each well of a coated 96-well plate. After incubation at 37
o
C for one hour, cells
were rinsed twice with PBS and the number of attached cells was estimated
spectrophotometrically using an MTT-based assay. MTT (3-[4, 5-dimethylthiazol-2-
yl]-2, 5-diphenyl tetrazolium bromide) dissolved in DMEM was added to each well
and the cells were incubated at 37
o
C for four hours. Tetrazolium crystals were
dissolved by the addition of MTT solubilization solution (10% Triton X-100, 0.1 N

25
HCl in anhydrous isopropanol), and absorbance was measured at 570 nm. The assay
was performed in triplicate.
2.2.4. Circular dichroism spectroscopy
Due to their inherent molecular asymmetry, biological molecules exhibit
differential absorption of left and right circularly polarized light. This phenomenon,
known as circular dichroism (CD), can be used to determine the secondary structure
of proteins using the peptide bond as a chromophore. Standard CD spectra for α-
helix, β-sheet, and random coil secondary structures are shown in Figure 2.2.
CD spectroscopy was performed as previously described (Jayasinghe and Langen
2004). Briefly, indicated amounts of plasma-purified or recombinant human
vitronectin were diluted in 10 mM sodium phosphate buffer, pH 7.4, and placed in a
1 mm path-length quartz cuvette. CD spectra were recorded from 260-190 nm at a
50 nm/minute scanning speed with a data pitch of 1 nm using a Jasco 815
spectropolarimeter (Jasco Inc., Easton, MD). Ten to fifteen scans were averaged for
each sample. Observed intensity was background subtracted and offset-corrected.
Molar ellipticity [ Ө] was calculated from the corrected intensity using the following
equation:
[ Ө] (deg cm
2
/dmol) = Intensity
# of residues x concentration (M) x path length (mm)

26

19 0 2 00 2 10 2 20 2 30 2 40 2 50 260
-600 0
-400 0
-200 0
0
200 0
400 0
600 0
Molar Ellipticity
W avelength (nm )
rVn
hVn

Figure 2.2 Representative standards for circular dichroism spectroscopy and
secondary structure of plasma-purified and recombinant vitronectin. Left Standard
curves for circular dichroism spectroscopy. Alpha helical structures (black) exhibit
troughs at 222 nm and 209 nm, and positive ellipticity at 192 nm. Beta sheet
structures (red) exhibit negative ellipticity at 218 nm and positive ellipticity at 196
nm. A random coil secondary structure (green) exhibits positive ellipticity at 212 nm
and negative ellipticity at 195 nm. Right Plasma-purified and recombinant
vitronectin have similar secondary structure by circular dichroism. Recombinant
vitronectin (blue squares) and plasma-purified human vitronectin (green circles)
were diluted in 10 mM phosphate buffer (pH 7.4). Circular dichroism spectra were
recorded from 260-190 nm and 10-15 scans were averaged per sample. Molar
ellipticity ( Ө) was calculated using intensity, polypeptide length, molar
concentration, and path length (mm).

27
2.3. Results
2.3.1. Recombinant human vitronectin can be purified from inclusion bodies.
Due to the need for obtaining a homogeneous, cost-effective, and renewable
supply of protein to use in our studies, we constructed a plasmid vector encoding
mature, full-length human vitronectin. The signal peptide was omitted from the
cDNA for prokaryotic expression. Due to the presence of an internal NcoI site,
digestion of the PCR product with NcoI and HindIII restriction enzymes produced
two DNA fragments (Figure 2.3). Expression of the construct in Escherichia coli
DH5 α cells resulted in the formation of inclusion bodies, from which recombinant
vitronectin was purified by heparin affinity chromatography. A representative
chromatogram is shown in Figure 2.4. Recombinant vitronectin eluted from the
column at 30-35% Buffer B, which corresponds to 300-350 mM NaCl. The
molecular weight of the purified protein estimated by SDS-PAGE was
approximately 55 kDa, compared to the theoretical molecular weight of 52.4 kDa
(Figure 2.5). Recombinant vitronectin has a lower molecular weight than plasma-
purified vitronectin, which is post-translationally modified. ELISA demonstrates
that plasma-purified and recombinant vitronectin display similar reactivity to a
polyclonal anti-vitronectin antibody (Figure 2.6). Enhanced absorbance of plasma-
purified vitronectin may be the result of enhanced background reactivity to the
secondary antibody, a phenomenon we consistently observed.
2.3.2. Recombinant vitronectin promotes cell adhesion.
The role of vitronectin in cell adhesion, spreading, and migration is mediated by

28

Figure 2.3 Agarose gel electrophoresis of pSE420 vitronectin inserts digested with
NcoI and HindIII. The resulting products were approximately 1 kb and 0.4 kb in
length, as expected. M, marker. Arrows represent standard markers in kilobases.

29

Figure 2.4 Representative chromatogram from the purification of recombinant
vitronectin expressed in Escherichia coli DH5 α cells. Isolated inclusion bodies were
solubilized in 8 M urea buffer and applied to a heparin affinity column. Proteins
were separated using a linear NaCl gradient. A representative chromatogram is
shown. Vitronectin was eluted from the column at approximately 30-35% Buffer B
(0.3-0.35 M NaCl).

30

Figure 2.5 Affinity-purified recombinant vitronectin extracted from inclusion
bodies. Lane 3 shows 5 μg of vitronectin after purification. Proteins were separated
by SDS-PAGE and stained with Coomassie Brilliant Blue. Lane 1, molecular weight
marker. Lane 2, 5μg bovine serum albumin. Arrowheads represent molecular weight
standards in kilodaltons.

31
the N-terminal RGD tri-peptide (Cherny, Honan et al. 1993). To assess whether
recombinant vitronectin promotes cell attachment, a cell adhesion assay was
performed. Cell culture wells coated with recombinant vitronectin retained 22%
more cells than uncoated wells (Figure 2.7) as estimated by MTT reduction, although
this quantity was not statistically significant (p = 0.08017). Cell adhesion activities
of plasma-purified vitronectin, recombinant vitronectin, BSA (not heat inactivated)
were similar, as compared to PBS (Figure 2.7). BSA was used as an additional
positive control, as co-purified and bound non-albumin proteins were not denatured
by heat-shock. These results demonstrate that the cell attachment activities of
plasma-purified and recombinant vitronectin are comparable.
2.3.3. Recombinant and plasma-purified vitronectin have similar secondary
structure.
Recombinant vitronectin was purified in a fully denatured conformation from
inclusion bodies solubilized in 8M urea followed by heparin affinity chromatography
using the same buffer. To ascertain whether the recombinant protein was properly
refolded after dialysis, spectroscopy was employed. The CD spectrum for
recombinant vitronectin (Figure 2.2, green line) is similar to the spectrum reported
for the native protein, which exhibits a gradual slope from approximately 230 nm,
with a trough near 205 nm (Pitt W.G. 1989). This is characteristic of a protein with a
predominantly beta-sheet and random fold secondary structure, with little alpha-
helical content. Plasma-purified human vitronectin had a similar, but not entirely
identical, CD spectrum (Figure 2.2, blue line), with a trough around 208 nm.

32
Although the shapes are similar, the spectral shift may be due to experimental error,
the presence of aggregates in the plasma-purified protein, or a slight difference in
secondary structure. Based on this result, we conclude that recombinant vitronectin
is correctly folded to an acceptable degree.

2.4. Discussion
We have developed a method for the expression and purification of untagged,
full-length recombinant vitronectin. The vitronectin signal sequence was omitted in
our construct due to our use of a prokaryotic expression system. The pSE420
plasmid was chosen for its ability to function as both a molecular cloning and protein
expression vector. Recombinant human vitronectin expressed in Escherichia coli
DH5 α cells and purified from inclusion bodies is comparable to vitronectin purified
from human plasma, as demonstrated by ELISA, promotion of cell adhesion, and
circular dichroism spectroscopy. Due to the inherent heterogeneity of plasma-
purified vitronectin, and problems associated with the purification process,
recombinant vitronectin is a superior source of protein for our studies.
This established method for the production of recombinant vitronectin will
provide a consistent and reliable source of protein in order to test the hypothesis that
vitronectin is an amyloidogenic protein. Once the portion of vitronectin essential for
fibril formation has been established, this system will be useful in the generation of
mutants for structural studies. For example, cysteine-less and single cysteine
mutants can be synthesized via molecular cloning for use in electron paramagnetic

33
0 100 200 300 400 500
-0.2
0.0
0.2
0.4
0.6
0.8
1.0
1.2
1.4
1.6
1.8
Absorbance
ng/well
recombinant Vn
Biosource Vn

Figure 2.6 ELISA of recombinant vitronectin is comparable to commercially-
available human vitronectin purified from plasma. Wells were coated in triplicate
with indicated amounts of protein. ELISA was performed using polyclonal anti-
vitronectin and HRP-conjugated anti-rabbit antibodies and developed with 3, 3, 5, 5 -
tetramethylbenzidine. Absorbance was measured at 450 nm.

34

0
20
40
60
80
100
120
rVn hVn BSA
Cell adhesion
% of control
PBS

Figure 2.7 Recombinant vitronectin promotes RPE cell attachment. A 96-well plate
was coated with 200 ng/cm
2
of the indicated protein and 2 x 10
4
cells per well were
allowed to attach for one hour. Cell number was estimated spectrophotometrically
using an MTT-based assay. Recombinant vitronectin and plasma-purified
vitronectin have similar cell attachment activities. BSA was used as an additional
positive control. Error bars represent standard error of the mean (n=3).

35
resonance (EPR) experiments. In addition, this system can be used to study the
effects of amino acid substitutions, deletions and other modifications.

Supplement
As an addendum to this study, we purified vitronectin from human serum. Fifty
milliliters of purchased frozen human serum (Sigma-Aldrich; St. Louis, MO) was
thawed at 37
o
C and brought to a total volume of 100 mL containing 8 M urea, 10
mM DTT, 5 mM EDTA, 1 mM PMSF, 1 μM leupeptin, and 100 nM aprotinin, all in
PBS. The mixture was centrifuged at 20,000 rpm for 30 minutes at 4
o
C. The
supernatant was applied to a heparin column equilibrated with the same buffer, and
proteins were eluted with a linear NaCl gradient. Vitronectin-containing fractions
were identified by SDS-PAGE, pooled and dialyzed against PBS. Protein
concentration was calculated using Beer’s Law (A = ε
M
cl; where ε
M
280 nm

= 82,330
M
-1
cm
-1
) and purified protein was lyophilized from PBS in 100 μg aliquots. Figure
2.8 shows a representative chromatogram obtained during the heparin affinity
purification of vitronectin from human serum. The collected fractions were
separated by SDS-PAGE and visualized by Coomassie stain (Figure 2.9).

36

07 0419 Human Vn purification 50mL serum001:1COPY_UV1_280nm 07 0419 Human Vn purification 50mL serum001:1COPY_Conc
07 0419 Human Vn purification 50mL serum001:1COPY_Fractions 07 0419 Human Vn purification 50mL serum001:1COPY_Logbook
-4000
-3500
-3000
-2500
-2000
-1500
-1000
-500
mAU
0 20 40 60 80 100 120 140 ml
1 3 5 7 9 11 14 17 20 23 26 29 32 35 38 41 44 47 50 53 56 59 62 65 68 71 74 77 80 83 85

Figure 2.8 Representative chromatogram from heparin affinity purification of
vitronectin from human serum. Peak fractions were separated by SDS-PAGE and
visualized by Coomassie stain. Human vitronectin eluted from the column at
approximately 20-30% Buffer B.

37

Figure 2.9 Fractions collected during heparin affinity chromatography of vitronectin
purification from human serum. Since all samples analyzed contained vitronectin
protein, fractions were pooled, dialyzed against PBS, and lyophilized. Arrowheads
indicate molecular weight markers (kilodaltons). V, 2.5 μg human vitronectin. FT,
flow-thru fraction. 11-29, fraction numbers.

38
Chapter III. Formation of soluble oligomers and amyloid fibrils by the
multifunctional protein vitronectin
3.1. Introduction/Rationale
Amyloid fibrils were first viewed under electron microscopy in 1959 by Cohen
and Calkins, who observed fibrils of 80-100 Ǻ width and indeterminate length in
ultrathin sections of amyloid-laden tissue (Cohen and Calkins 1959). Solubilization
of isolated fibrils in guanidine hydrochloride and subsequent gel filtration led to
identification of the unique protein associated with each amyloidopathy [reviewed in
(Sipe and Cohen 2000)]. This identification paved the way to in vitro formation of
amyloid fibrils by natural and synthetic amyloid proteins, in addition to the
generation of transgenic animal models of amyloid diseases (Quon, Wang et al.
1991; Wirak, Bayney et al. 1991; Butler, Jang et al. 2004). The structure and
tinctorial properties of ex vivo and in vitro amyloid fibrils are similar (Ferreira,
Vieira et al. 2007), justifying their in vitro biochemical characterization. Different
techniques are employed to form amyloid fibrils in vitro, including incubation in
physiologic buffer, as well as acid, heat, or chemical denaturation. Regardless of
how they are produced, amyloid fibrils are insoluble and protease resistant (Sipe
1994).
According to the amyloid stretch hypothesis, the amyloidogenicity of a protein
can be localized to short stretches of amino acids in key areas (Esteras-Chopo,
Serrano et al. 2005). Historically, the amyloidogenic core of fibrils has been
elucidated through studies of synthetic peptide fragments for proteins such as A β,

39
IAPP, and serum amyloid A (Sipe 1994), or limited protease digestion in the case of
α-synuclein (Miake, Mizusawa et al. 2002). Protease digestion may be followed by
Western blot and/or mass spectrometry to reveal the identity of the peptides.
Interestingly, when a non-destabilizing, six-residue, amyloidogenic peptide
(STVIIE) was inserted into a soluble globular domain of a protein, the protease-
resistant amyloid core included adjacent segments (Esteras-Chopo, Serrano et al.
2005). This suggests that short stretches of amyloidogenic amino acids can pull
additional parts of the protein into fibril formation. Thus, algorithms that aid in
predicting aggregation propensity of a polypeptide may be useful in the identification
of fibril-forming regions in the protein.
Although fibril deposition is a hallmark of amyloidopathies, recent studies
indicate that soluble prefibrillar oligomers may play a major role in the
pathophysiology of disease. The accumulating body of evidence regarding the
significance of A β oligomers clearly exhibits the potential validity of this hypothesis.
Soluble A β content more accurately predicts clinically-established cognitive deficits
in Alzheimer disease patients (McLean, Cherny et al. 1999). Prefibrillar A β has
been identified in human Alzheimer disease brain (Kuo, Emmerling et al. 1996), as
well as in transgenic animal models of Alzheimer disease (Lesne, Koh et al. 2006).
A β oligomers formed in vitro are toxic to cultured cells and inhibit long-term
potentiation in hippocampal slice cultures (Lambert, Barlow et al. 1998; Walsh,
Klyubin et al. 2002). In addition, A β oligomers bind membranes and disrupt
membrane integrity in vitro (Hartley, Walsh et al. 1999; Demuro, Mina et al. 2005;

40
Quist, Doudevski et al. 2005). Of considerable importance, Glabe and colleagues
demonstrated oligomer toxicity of other amyloid proteins, such as IAPP, α-
synuclein, lysozyme, and prion protein, and they developed the conformation-
specific A11 antibody, which attenuated oligomer-mediated cell death (Kayed, Head
et al. 2003).
As stated previously, vitronectin is an appealing candidate for amyloid formation
due to its abundance in serum and tissues, and association with insoluble, disease-
associated plaques. Because the role of vitronectin in amyloid formation within
these age-related, thioflavin-positive aggregates is currently unknown, we aim to
determine whether vitronectin is amyloidogenic using recombinant vitronectin
protein, expressed and purified as described in Chapter II. Because vitronectin
misfolding has not been previously studied, we will need to empirically establish the
optimal conditions for oligomer and fibril formation, and since vitronectin is natively
folded, partial denaturation may be necessary. For example, β
2
-microglobulin forms
fibrils at neutral pH only under destabilizing conditions that expose the
amyloidogenic C-terminus (Jones, Manning et al. 2003). Moreover, some amyloid
proteins require the use of acidic pH and/or denaturing organic solvents, such as
trifluoroethanol or hexafluoroisopropanol, to form oligomers and fibrils (Lambert,
Viola et al. 2001; Bucciantini, Giannoni et al. 2002).
We will monitor the development of vitronectin aggregates by thioflavin T
fluorescence, a more reliable alternative to Congo red (LeVine 1993), and examine
their morphologies by transmission electron microscopy. Immunoreactivity with the

41
A11 antibody will indicate the presence of toxic, nonfibrillar oligomers, and the
cytotoxicity of vitronectin oligomers will be tested in a cell culture paradigm.
Because amyloid oligomers may target the plasma membrane, we will investigate the
effects of vitronectin oligomers on the integrity of synthetic vesicles. Lastly, we aim
to characterize the amyloidogenic core region of vitronectin using protease digestion,
followed by amino acid sequencing, Western blot and mass spectrometry analysis.
In this study, we demonstrate that vitronectin behaves as an amyloid protein in
vitro and soluble nonfibrillar vitronectin oligomers are toxic to cultured cells. These
data suggest that vitronectin misfolding and aggregation may contribute to the
pathophysiology of diseases associated with vitronectin-containing plaques.

3.2. Materials and Methods
3.2.1. Reagents
Plasma-purified vitronectin was purchased from Biosource (Camarillo, CA). The
anti-oligomer A11 antibody was a generous gift from Dr. Charles Glabe (University
of California, Irvine). Production of the polyclonal anti-vitronectin antibody raised
against full-length recombinant vitronectin and the polyclonal M1 antibody, specific
for the C-terminal vitronectin fragment NH
2
-
RRPSRATWLSLFSSEESNLGANNYDDYRMDWLV-COOH, was performed by
Biomer Technology (Hayward, CA). The IgG fraction was enriched from rabbit
serum using Affi-Gel Protein A support (Bio-Rad; Hercules, CA). All other reagents
were purchased from Sigma-Aldrich (St. Louis, MO), unless otherwise specified.

42
3.2.2. Oligomer and fibril preparation and analysis of morphology
Optimal conditions for oligomer and fibril formation were determined empirically
using phosphate buffer, PBS, guanidine hydrochloride, urea, DTT, heparin, acetic
acid, hydrochloric acid (HCl), sodium acetate, trifluoroethanol (TFE), and
1,1,1,3,3,3-hexafluoro-2-propanol (HFIP). Morphology was observed under electron
microscopy. Details on the optimization of vitronectin amyloid formation are listed
in Table 3.1. Treatment with HFIP consistently yielded enriched populations of
oligomers and fibrils with excellent morphology.
Lyophilized vitronectin protein in a siliconized microfuge tube was dissolved in
cold 1,1,1,3,3,3-hexafluoro-2-propanol (HFIP) for 10 minutes at room temperature.
For fibrils, this solution was diluted to 50% HFIP in water and stirred at room
temperature for 7-14 days at a protein concentration of 10-25 µM. Fibril formation
was monitored by electron microscopy. To prepare soluble oligomers, vitronectin in
HFIP was diluted to 20% HFIP in water, with or without 1 mM HCl, and stirred at
room temperature with a vented lid for 3-7 days. The final protein concentration was
approximately 20 µM after the gradual evaporation of HFIP. Oligomer formation
was monitored by electron microscopy and by dot blot using the anti-oligomer
antibody.
3.2.3. Transmission electron microscopy (TEM)
The morphology of oligomers and fibrils was evaluated by TEM. HFIP was
evaporated under a gentle stream of nitrogen. Ten microliters of sample was applied
to a 200-mesh, formvar-coated nickel grid (Electron Microscopy Sciences; Hatfield,

43

Condition Oligomers Protofibrils Fibrils Amorphous
aggregates
Water X X X
5 mM DTT X X
PB X X X X
5 mM DTT / PB X X
PBS X X X X
5mM DTT/ PBS X X
25% TFE/ PBS X X
25% TFE/
NaOAc/ PBS
X
0.5 M Gnd/ PBS X X
1 M Gnd/ PBS X X
2 M Gnd/ PBS X X
5% HFIP X X
10% HFIP X
20% HFIP X X
30% HFIP X
40% HFIP X
50% HFIP X X
20% HFIP/ 1mM
HCl
X X
20% HFIP/ 1mM
HCl/ PBS
X
50% HFIP/ PB X X
50% HFIP/ PBS X X X
4 M urea/ 1%
acetic acid
X X X
200 μg/mL
heparin
X

Table 3.1 Optimization of vitronectin oligomer and fibril formation. Recombinant
vitronectin was prepared at 5-10 μM under the conditions indicated and stirred at
room temperature for 1-4 weeks. Morphology was analyzed by electron microscopy.
Unless otherwise stated, reagents were diluted in water. PB, 10 mM phosphate
buffer, pH 7.4. NaOAc, 25 Mm sodium acetate buffer, pH 5.5. Gnd, guanidine
hydrochloride.

44

PA) for 5 minutes, stained with 3% uranyl acetate for 5 minutes, rinsed, and air-
dried. The grids were examined using a Jeol JEM1200EX microscope at 80 kV.
3.2.4. Thioflavin T fluorescence
Thioflavin T fluorescence was used to ascertain increases in β sheet content,
which is indicative of misfolding and aggregation. For the free dye, the excitation
wavelength is 385 nm and emission wavelength is 445 nm. Thioflavin T interacts
with the cross β structure in an unknown way that results in a new absorption at 450
nm and enhanced emission at 482 nm. Recombinant vitronectin was dissolved in
PBS, 20% HFIP (for oligomers), or 50% HFIP (for fibrils). Samples were diluted to
1 μM with a buffer containing 5 μM thioflavin T in 50 mM glycine buffer, pH 9.
The sample was placed in a quartz cuvette (0.1 cm path-length), excited at 450 nm,
and fluorescence was monitored over 460-550 nm. Background fluorescence of
buffers used to prepare the samples without protein was subtracted from the sample
fluorescence. Data shown is the corrected fluorescence at 482 nm.
3.2.5. Dot blot
Two microliters of each sample were spotted onto nitrocellulose membrane and
allowed to air-dry. Tris-buffered saline (20 mM Tris, 0.8% NaCl, pH 7.4) containing
0.001% Tween-20 (TBST) was used for washing and dilution. The membrane was
blocked for 1 hour with 10% nonfat dried milk dissolved in TBST, washed 3 x 10
minutes, incubated for one hour in primary anti-oligomer antibody (diluted 1:5,000
in 3% BSA/ TBST), washed 3 x 10 minutes, incubated for 30 minutes in HRP-

45
conjugated secondary anti-rabbit antibody (diluted 1:10,000 in 3% BSA/ TBST;
Vector Laboratories, Burlingame, CA), and washed 3 x 10 minutes. The membrane
was developed using enhanced chemiluminescence reagents (Amersham
Biosciences; Piscataway, NJ) and exposed to Hyperfilm (Amersham Biosciences;
Piscataway, NJ). The same procedure was performed for dot blot with anti-
vitronectin antibody (1:10,000) using 0.1% Tween-20 in TBST for washing and
dilution. A control membrane with primary antibody omitted was simultaneously
processed.
3.2.6. Cell culture and cytotoxicity assays
The effects of vitronectin oligomers were tested in a cell culture paradigm. SH-
SY5Y cells were obtained from ATCC (Rockville, MD). Retinal pigment epithelium
(RPE) cells were isolated from human fetal eyes obtained from Advanced Bioscience
Resources, Inc. (Alameda, CA) as described previously (Jin, He et al. 2000). Cells
were maintained in complete medium [Dulbecco’s modified Eagle’s medium
(DMEM, VWR; West Chester, PA) supplemented with 2 mM L-glutamine, 100
U/ml penicillin, 100 µg/ml streptomycin, and 10% fetal bovine serum (Invitrogen;
Carlsbad, CA)] at 37
o
C in a humidified incubator. Fourth-passage cells were seeded
in a 96-well plate at 2 x 10
4
cells per well and grown in complete medium for 3-4
days to approximately 90% confluence. Cells were maintained in serum-free
medium for one day prior to the experiment. On the day of the assay, media was
removed, replaced with the indicated samples diluted in DMEM, and incubated at
37
o
C for 4 hours. MTT (3-[4, 5-dimethylthiazol-2-yl]-2, 5-diphenyl tetrazolium

46
bromide) dissolved in DMEM was added and cells were placed at 37
o
C for an
additional 4 hours. Tetrazolium crystals were dissolved by the addition of MTT
solubilization solution (10% Triton X-100, 0.1 N HCl in anhydrous isopropanol) and
absorbance was measured at 570 nm. Experiments were carried out in triplicate.
3.2.7. Membrane leakage assay
To investigate the effects of vitronectin oligomers on membranes, a leakage assay
was performed. A schematic diagram of the membrane leakage assay is shown in
Figure 3.1. Large unilamellar vesicles (100 nm diameter) containing 90%
phosphatidylcholine and 10% phosphatidylserine (Avanti Polar Lipids; Alabaster,
AL) were extruded in the presence of the fluorophore-quencher pair 8-
aminonapthalene-1,3,6-trisulfonic acid (ANTS) and p-xylene-bis-pyridinium
bromide (DPX). To assess membrane leakage, 250 μl of a solution containing 500
μM lipid vesicles and the sample of interest in diluted in Buffer 1 (10 mM HEPES,
50 mM KCl, 1 mM EDTA, 3 mM sodium azide) was placed in a 2 mm path-length
quartz cuvette. ANTS fluorescence was monitored as a function of time at 520 nm,
with excitation at 353 nm, using a Jasco FP-6500 spectrofluorimeter. Maximum
fluorescence intensity was determined by the addition of 5 μl of 10% Triton X-100.
Intensities were normalized to the intrinsic fluorescence of the vesicles and percent
leakage was estimated by dividing the sample and maximum fluorescence intensity
values. Leakage assays were performed in duplicate. The data represent the results
of three independent experiments.

47

Figure 3.1 A cell-free, fluorescence-based assay to monitor membrane leakage
(adapted from S. Jayasinghe). The structures of the fluorophore 8-aminonaphthalene-
1,3,6 trisulfonic acid (ANTS; λ
ex
= 353 nm, λ
em
= 520 nm) and quencher p-xylene-
bis-pyridinium bromide (DPX) are shown in (A) and (B), respectively. The
fluorophore-quencher pair ANTS-DPX is entrapped in large unilamellar vesicles of
100 nm diameter and ANTS fluorescence is measure over time. The addition of
soluble oligomers disrupts membrane integrity, resulting in the leakage of vesicle
contents, observed as an increase in ANTS fluorescence.

48
3.2.8. Fibril digestion and analysis of protease-resistant fragment
Fibrils were collected by centrifugation and the pellets were digested in a solution
containing 1:25 trypsin (protease-to-protein mass ratio) and 10 mM DTT, all in 50
mM ammonium bicarbonate, pH 8, at 37
o
C. Overnight digests were heat inactivated
at 60
o
C, analyzed by SDS-PAGE, and proteolytic fragments were visualized with
Coomassie Blue staining. The major protease-resistant band was either transferred
onto a PVDF membrane and submitted to the USC/Norris Microchemical Core
Facility for peptide sequencing by N-terminal Edman degradation, or excised and
passively eluted in 50 mM Tris, 50 mM NaCl (pH 8) and submitted to the USC
Proteomics Core Facility for matrix-assisted laser desorption/ionization time-of-
flight (MALDI-TOF) mass spectrometry analysis.
3.2.9. Western blot
Proteins were separated by SDS-PAGE and transferred at 200 mA to a
nitrocellulose membrane at 4
o
C. TBST (0.1% Tween-20) was used for washing and
dilution. The membrane was blocked for 1 hour with 10% nonfat dried milk diluted
in TBST, washed 3 x 10 minutes, incubated for one hour in primary M1 antibody
(diluted 1:5,000 in 3% BSA/ TBST), washed 3 x 10 minutes, incubated for 30
minutes in HRP-conjugated secondary anti-rabbit antibody (diluted 1:10,000 in 3%
BSA/ TBST; Vector Laboratories, Burlingame, CA), and washed 3 x 10 minutes.
The membrane was developed using enhanced chemiluminescence reagents
(Amersham Biosciences; Piscataway, NJ) and exposed to Hyperfilm (Amersham
Biosciences; Piscataway, NJ).

49
3.3. Results
3.3.1. Vitronectin forms soluble nonfibrillar oligomers and amyloid fibrils.
To investigate whether vitronectin has an inherent propensity to behave as an
amyloid protein, we aged plasma-purified and recombinant human vitronectin in
phosphate-buffered saline (PBS). Plasma-purified and recombinant vitronectin in
PBS form spherical and fibrillar structures by TEM (Figure 3.2, left). Because
vitronectin in PBS produces a heterogeneous mixture of aggregates, we empirically
determined the optimal conditions to enrich the fibril and oligomer populations. We
attempted to form oligomers and fibrils using a variety of buffers and solvents (Table
3.1). HFIP, a solvent which enhances the rate of aggregation, consistently and
reproducibly generated oligomers and fibrils that exhibited proper morphology.
Treatment with 50% HFIP promotes the formation of vitronectin fibrils, which
resemble the morphology of typical amyloid fibrils (Figure 3.2, middle). The fibrils
are approximately 3-8 nm in diameter, of varied length, and often twisted. Plasma-
purified and recombinant vitronectin incubated in 20% HFIP and 1 mM HCl adopt
spherical and protofibrillar structures, which range in size from 6-35 nm in diameter
(Figure 3.2, right). Enhanced thioflavin fluorescence of vitronectin oligomers and
fibrils formed using this method correlates with the observed morphologies (Figure
3.3). Recombinant vitronectin shows a small increase in fluorescence after
incubation in PBS for 7 days Figure 3.3, left). This represents the inherent
aggregation propensity of unassembled vitronectin. Vitronectin oligomers and fibrils
show a marked enhancement of thioflavin T fluorescence (Figure 3.3, middle and

50

Figure 3.2 Human vitronectin forms amyloid fibrils and oligomers. Left, Plasma-
purified (hVn) and recombinant human vitronectin (rVn) aged in phosphate-buffered
saline form a heterogeneous mixture of spherical and fibrillar structures by TEM.
Middle, Vitronectin treated with 50% HFIP promotes the formation of typical
amyloid fibrils that are approximately 3-8 nm in diameter. Right, Incubation in 20%
HFIP enriches the population of spherical and protofibrillar oligomers, which range
from 6-35 nm in diameter. Scale bar = 100 nm.

51

Vn in PBS Vn oligomers Vn fibrils
0
50
100
150
200
250
300
350
Fluorescence Units, 482 nm

Figure 3.3 Vitronectin oligomers and fibrils enhance thioflavin T fluorescence.
Recombinant vitronectin was prepared in the indicated buffers as described.
Samples were diluted to 1 μM in a buffer containing 5 μM thioflavin T in 50 mM
glycine buffer, pH 9, excited at 450 nm, and thioflavin T fluorescence was monitored
at 482 nm. Recombinant vitronectin in PBS exhibited a small increase in
fluorescence, while vitronectin oligomers and fibrils displayed marked enhancement
of thioflavin T fluorescence, indicating an increase in β sheet content and the
presence of misfolding and aggregation.

52
right), indicating a large increase in β sheet content. These data suggests the
presence of vitronectin misfolding and aggregation and is typical of amyloid
proteins.
3.3.2. A11-positive vitronectin oligomers are toxic to cultured cells and induce
membrane leakage.
In order to further characterize vitronectin oligomers, we performed a dot blot
assay using the conformation-specific A11 antibody, which recognizes toxic
oligomers formed by a variety of amyloidogenic proteins. Recombinant vitronectin
freshly dissolved in phosphate buffer (Figure 3.4, B) and vitronectin treated with
50% HFIP to induce fibrils (Figure 3.4, D) are not recognized by the A11 antibody.
However, recombinant vitronectin incubated in 20% HFIP exhibits A11
immunoreactivity (Figure 3.4, C), suggesting that the population of soluble
nonfibrillar oligomers is enriched under these conditions. Although plasma-purified
vitronectin is recognized by the A11 antibody (Figure 3.4, A), background reactivity
staining is observed when incubated in secondary antibody alone, indicating the
presence of a contaminant that reacted strongly with the anti-rabbit secondary
antibody (data not shown). The presence of vitronectin in these samples was
confirmed using a primary anti-vitronectin antibody (Figure 3.4, F-I). A β oligomers
(Figure 3.4, E) served as a positive control for the A11 antibody, but were not
recognized by the vitronectin antibody, as expected (Figure 3.4, J).
Recent evidence suggests soluble nonfibrillar oligomers may comprise a toxic
species. Because A11 immunoreactivity has been reported to correlate with

53

Figure 3.4 Vitronectin forms soluble nonfibrillar amyloid oligomers. Samples were
spotted on nitrocellulose membrane and probed with either A11 (A-E) or vitronectin
(F-J) primary antibodies. Recombinant vitronectin incubated in 20% HFIP exhibits
reactivity to the oligomer-specific A11 antibody (C), whereas recombinant
vitronectin freshly dissolved in phosphate buffer (B) or treated with 50% HFIP (D)
do not, suggesting that treatment with 20% HFIP enriches the population of soluble
oligomers. Plasma-purified vitronectin is also recognized by the A11 antibody (A),
although background staining is observed when incubated in secondary antibody
only (data not shown). Reactivity of samples with the vitronectin antibody confirms
the presence of protein (F-I). A β oligomers served as a positive control for the A11
antibody (E), but are not detected by the vitronectin antibody (J), as expected.

54
cytotoxicity and vitronectin oligomers are A11-positive, we examined their effects
on cultured SH-SY5Y and RPE cells. SH-SY5Y neuroblastoma cells and RPE cells
were chosen due to the emerging role of soluble oligomers as pathogenic entities in
neurodegenerative diseases such as AD and AMD. HFIP was removed by
evaporation and indicated concentrations of vitronectin oligomers diluted in serum-
free cell culture medium were added to cells for 4 hours at 37
o
C. Cell viability was
assessed spectrophotometrically using an MTT-based assay. Vitronectin oligomers
were toxic to both SH-SY5Y and RPE cells in a dose-dependent manner, as observed
by impaired MTT reduction (Figure 3.5, white and grey bars). Plasma-purified and
recombinant vitronectin aged in PBS exhibited mild RPE toxicity (Figure 3.5,
hatched bars). Pre-incubation of 5 µM vitronectin oligomers with an equimolar
amount of A11 antibody prior to treatment significantly attenuated the impairment in
MTT reduction, indicating oligomer-specific toxicity (Figure 3.5, black bar) and
demonstrating the ability of the A11 antibody to block the majority of toxic
oligomeric species.
It has been suggested that the plasma membrane is a common target of toxic
nonfibrillar oligomers. To investigate the hypothesis that vitronectin oligomers
disrupt membrane integrity, we employed a fluorescence-based leakage assay using
large unilamellar vesicles of 100 nm diameter loaded with the fluorophore-quencher
pair ANTS-DPX. Indicated concentrations of freshly-dissolved recombinant
vitronectin or recombinant vitronectin oligomers were diluted in Buffer 1 (see
Materials and Methods) containing 500 μM lipid vesicles. ANTS fluorescence was

55

Figure 3.5 Soluble nonfibrillar vitronectin oligomers compromise cell viability.
Vitronectin oligomers are toxic to SH-SH5Y (white bars) and RPE cells (grey bars)
in a dose-dependent manner, as evidenced by impaired MTT reduction. Plasma-
purified and recombinant vitronectin aged in PBS exhibit mild toxicity (hatched
bars). RPE toxicity was significantly attenuated when 5 µM vitronectin oligomers
were pre-incubated with an equimolar amount of A11 antibody prior to treatment
(black bar), although complete rescue was not observed. Cell viability was assessed
spectrophotometrically using an MTT-based assay. Error bars represent standard
error of the mean. (n=3; **p < 0.01)

56
monitored over time at 520 nm ( λ
ex
= 353 nm) and percent leakage was estimated by
dividing the sample and maximum fluorescence intensity values after correcting for
the intrinsic fluorescence of the vesicles. Freshly- dissolved recombinant vitronectin
in phosphate buffer had no appreciable effect on membrane permeability (Figure 3.6,
white bars). Vitronectin oligomers induced dose-dependent leakage of ANTS from
LUVs (Figure 3.6, grey bars), indicating membrane disruption is caused by the
oligomeric conformation.
3.3.3. Trypsin digestion of vitronectin fibrils reveals a protease-resistant core.
To facilitate the isolation and identification of the amyloidogenic core of
vitronectin fibrils, we employed protease digestion followed by peptide sequencing,
immunoblotting, and mass spectrometry analysis. Undigested vitronectin fibrils are
shown in Figure 3.7A (lane 1). It is evident that a large amount of protein was not
resolved by SDS-PAGE and was retained in the wells or did not pick up the
Coomassie stain (Figure 3.7A, asterisk). Trypsin digestion of vitronectin fibrils
produced a major protease-resistant fragment of ~ 5 kDa (Figure 3.7A, lane 2). This
band was not present when soluble recombinant vitronectin was digested (data not
shown). Partial protein sequencing of the first twenty-one amino acids by N-
terminal Edman degradation identified the peptide
AMWLSLFSSEESNLGANNYDD in the excised band, which comprises C-terminal
residues 380-400. The origin of the peptide was confirmed by Western blot with a
C-terminal-specific vitronectin antibody (Figure 3.7A, lane 3). MALDI-TOF mass
spectrometry analysis of the excised fragment identified several vitronectin peptides

57
(Figure 3.7B) and suggests that residues 380-427 may be important for amyloid
formation. The peptides correlate well with C-terminal stretches of increased cross-
beta aggregation propensity as predicted by the TANGO algorithm (Figure 3.7B).

3.4. Discussion
In this study, we provide novel evidence that vitronectin is capable of amyloid
formation. Vitronectin readily aggregates in physiologic buffer. HFIP treatment,
although not required, enriched the individual populations of soluble, spherical
oligomers and amyloid fibrils. Our results establish the inherent amyloidogenicity of
vitronectin and are in agreement with previously published reports regarding the
toxicity of soluble oligomers (Bucciantini, Giannoni et al. 2002). The finding that
HFIP treatment promotes oligomer and fibril formation supports the hypothesis that
partial unfolding or denaturation of the native protein state promotes misfolding
(Chiti and Dobson 2006). Vitronectin oligomers recognized by the conformation-
specific A11 antibody are cytotoxic to cultured SH-SY5Y and RPE cells in a dose-
dependent manner, suggesting they may indeed contribute to the development and
progression of disease. Soluble oligomers formed in vitro from a variety of proteins
appear morphologically similar and exhibit toxicity to cultured cells (Hartley, Walsh
et al. 1999; Janson, Ashley et al. 1999; Chiti, Bucciantini et al. 2001; Baskakov,
Legname et al. 2002; Bucciantini, Giannoni et al. 2002; Chromy, Nowak et al. 2003;
Hoshi, Sato et al. 2003; Kayed, Head et al. 2003; Bucciantini, Calloni et al. 2004;
Zhu, Han et al. 2004), suggesting a common pathogenic mechanism. Preincubation

58
0
10
20
30
40
50
60
70
2.5 μM5 μM

% Leakage
Unassembled
Oligomers
1 μM

Figure 3.6 Vitronectin oligomers induce membrane leakage. Large unilamellar
vesicles (100 nm in diameter) loaded with the fluorophore-quencher pair ANTS-
DPX were incubated with indicated concentrations of protein. Recombinant
vitronectin oligomers incubated with 500 µM LUVs resulted in appreciable leakage
of vesicles in a dose-dependent fashion (grey bars) as compared to recombinant
vitronectin freshly dissolved in phosphate buffer (white bars). Membrane leakage
was assessed by monitoring ANTS fluorescence ( λ
ex
= 353 nm, λ
em
= 520 nm).
Intensities were normalized to the intrinsic fluorescence of the vesicles and percent
leakage was estimated by dividing the sample and maximum fluorescence intensity
values. Maximum intensity was achieved by the addition of 10% Triton X-100.
Data shown are the combined average of three independent experiments (n=2). Error
bars represent the standard error of the mean.

59

Figure 3.7 Vitronectin fibrils contain a protease-resistant core. A) In-solution
trypsin digestion of vitronectin fibrils reveals a protease-resistant band of ~ 5 kDa
(lane 2). Partial sequencing by N-terminal Edman degradation identified the peptide
AMWLSLFSSEESNLGANNYDD, which corresponds to residues 380-400. These
amino acids reside in the C-terminus of the protein and likely comprise a portion the
amyloidogenic core. This result was confirmed by immunoblotting with the C-
terminal-specific M1 antibody (lane 3). Fibrils were digested overnight at 37
o
C in a
solution containing 1:25 trypsin (protease-to-protein mass ratio) and 10 mM DTT
and products were analyzed by SDS-PAGE followed by Coomassie staining.
Arrowheads indicate molecular weight (kDa). Lane 1 contains undigested
vitronectin fibrils, much of which is unresolved. B) Upper panel Several peptides
were identified by MALDI-TOF analysis of the protease-resistant band. The
mass/charge (m/z) values, amino acid sequence and residue numbers are displayed in
the table. Each m/z does not correspond to any other vitronectin peptides and
oxidation had to be invoked for peptide A380-R427. Lower panel Primary sequence
analysis of vitronectin using the TANGO algorithm (Fernandez-Escamilla, Rousseau
et al. 2004) reveals several C-terminal stretches prone to cross-beta aggregation, two
of which correlate well with the peptides A380-R401 (gray) and A380-R427
(hatched) identified by mass spectrometry.

60
of vitronectin oligomers with the A11 antibody prior to treatment significantly
rescued RPE cell viability, demonstrating that the majority of toxic nonfibrillar
oligomers are neutralized by the A11 antibody. Vitronectin oligomers permeabilize
synthetic vesicles in a cell-free assay, which is consistent with studies showing that
soluble amyloid oligomers disrupt membranes (Janson, Ashley et al. 1999; Kayed,
Sokolov et al. 2004; Demuro, Mina et al. 2005; Quist, Doudevski et al. 2005).
However, the precise mechanism of oligomer-mediated toxicity remains
controversial (Lashuel and Lansbury 2006; Jayasinghe and Langen 2007).
Vitronectin fibrils resemble the morphology of typical amyloid fibrils and contain
a protease-resistant domain which likely contains the core residues sufficient for
amyloid formation. A trypsin-resistant band was seen following SDS-PAGE and
Coomassie staining when vitronectin fibrils were digested, but not when soluble
recombinant vitronectin was digested. Sequencing and immunoblotting of the
fragment revealed a C-terminal epitope and mass spectrometry analysis identified
several peptides within residues 380-427, indicating that this region may be
important for amyloid formation. Interestingly, the identified fragments overlap with
two C-terminal regions that have increased cross-beta aggregation propensity, as
calculated by the TANGO algorithm (Fernandez-Escamilla, Rousseau et al. 2004).
The combined data from mass spectrometry and TANGO analysis suggest the
existence of an ordered amyloid core which limits protease accessibility to the
stretch of amino acids between the two predicted aggregation-prone regions. The
putative amyloid core resides within a naturally-occurring 10 kDa fragment (residues

61
380-459), the generation of which is regulated in vivo by a genetic polymorphism
(Figure 3.8). Primary sequence analysis of the 10 kDa vitronectin fragment reveals a
stretch of highly hydrophobic amino acids, which is consistent with the hypothesis
that hydrophobicity is a key factor in aggregation propensity (Otzen, Kristensen et al.
2000; Chiti, Taddei et al. 2002; Fernandez-Escamilla, Rousseau et al. 2004; Pawar,
Dubay et al. 2005; Chiti and Dobson 2006). Further studies are warranted to
precisely delineate the boundaries of the vitronectin amyloidogenic core and to
characterize the in vitro and in vivo significance of the 10 kDa vitronectin fragment.
Vitronectin exists in two conformationally distinct forms in vivo. The majority of
vitronectin in plasma and serum circulates as a non-heparin-binding monomer.
Approximately 2-8% of vitronectin in the blood is in the alternate heparin-binding,
partially unfolded conformation which can self-associate (Izumi, Yamada et al.
1989). This multimeric form is thought to be the predominant conformation of
vitronectin in the extracellular matrix (Preissner, Grulich-Henn et al. 1990).
Chaotropic denaturation, which was used in the purification of recombinant
vitronectin, exposes the heparin-binding site (Tomasini and Mosher 1988), which is
adjacent to a hydrophobic patch. Since vitronectin readily aggregates in its heparin-
binding state, this specific conformation may aid in amyloid formation, a hypothesis
further supported by the observation that hydrophobic interactions appear to drive
functional oligomerization (Hogasen, Mollnes et al. 1992). Although a relatively
small percentage of vitronectin is in an alternate conformation, conditions of high
local concentration via increased synthesis or recruitment, which may occur at

62

Figure 3.8 Domain structure of vitronectin. A genetic polymorphism at residue 381
(asterisk) regulates the level of proteolysis at arginine 379, which produces the two-
chain 65+10 kDa form of vitronectin held together by a disulfide bond. The green
hatched area represents the putative amyloid core, which resides within the 10 kDa
fragment. SMB, somatomedin B domain. Yellow, RGD peptide. HBD, heparin
binding domain. Arrow, proteolytic cleavage site (arginine 379).

63
extravascular sites or in the setting of chronic inflammation, may promote
vitronectin misfolding and amyloid formation. Our results put forth the possibility
that vitronectin misfolding and amyloid formation may play a role in the
pathophysiology of age-related diseases such as AMD, atherosclerosis, and AD.

Supplement
To determine the size and molecular weight of vitronectin oligomers, we
performed size exclusion chromatography. Two gel filtration columns with different
separation ranges of globular proteins were used: Superdex 75 (3000 – 70,000 M
r,
,
Figure 3.9), and Superdex 200 (10,000 – 600,000 M
r
, Figure 3.10). After columns
were equilibrated with 10 mM sodium phosphate buffer, pH 7.4, vitronectin
oligomers were added (100 μg total protein) and allowed to separate. Gel filtration
standards were used to calibrate the void volume and generate an elution profile.
Recombinant vitronectin oligomers were eluted either in the void volume or the
monomeric fraction, and were not able to be isolated by size exclusion
chromatography. It is possible that vitronectin oligomers are inherently unstable and
dissociate in the 10 mM sodium phosphate buffer used for the experiments.
Likewise, the further aggregation of oligomers in phosphate buffer may have
generated high molecular weight species which eluted in the void volume.
To determine the manner by which recombinant vitronectin oligomers induce cell
death, we used an apoptosis detection kit. Cultured RPE cells were treated for 4
hours followed by a colorimetric Terminal deoxynucleotidyl Transferase Biotin-

64
dUTP Nick End Labeling (TUNEL) assay (R&D Systems; Minneapolis, MN). The
data shown in Figure 3.11 demonstrate the absence of a statistically significant
difference between untreated cells and those treated with vitronectin oligomers.
However, an MTT reduction assay performed after the same treatment period of four
hours showed impaired cell viability (data not shown). These results suggest that
oligomer-induced cell death may not occur via an apoptotic mechanism. However, it
is possible that degenerating or dead cells were no longer able to adhere to the tissue
culture plate and were therefore removed by subsequent washing steps.

65

07 0125 100u g Vn ol igomers Su perde x 75 001:1_UV1 _280nm 07 0125 100ug Vn oligo mers Superdex 75001:1_Fracti ons 07 0125 100ug Vn oligomers Superdex 75001:1_Logbook
0.0
10.0
20.0
30.0
40.0
mAU
0.0 5.0 10.0 15.0 20.0 25.0 30.0 35.0 ml
1 2 3 4 5 6 7 8 9 10 12 14 16 18 20 22 24 26 28 30 32 34 36 38 40 42 44 46 48 50 52 54 56 58 Waste

Figure 3.9 Gel filtration of recombinant vitronectin oligomers using the Superdex 75
column. Vitronectin oligomers were eluted almost entirely in the void volume (large
peak), indicating M
r
> 70,000. Effective separation was not achieved, though these
results suggest that the majority of vitronectin was not in monomeric form.

66

07 0125 100u g Vn ol igomers Su perde x 20 0:1_ UV1_280nm 07 0125 100ug Vn oligomers Superdex 200:1_Fractions 07 0125 100ug Vn oligo mers Supe rdex 200:1_Lo gboo k
-5.0
0.0
5.0
10.0
15.0
20.0
25.0
mAU
0.0 5.0 10.0 15.0 20.0 25.0 30.0 ml
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 Waste

Figure 3.10 Gel filtration of recombinant vitronectin oligomers using the Superdex
200 column. Again, most of the vitronectin protein eluted in the void volume,
indicating an M
r
> 600,000, which corresponds to an 11-mer or larger. A small
quantity of protein eluted between 18-20 mL, which correlates to a M
r
between
67,000 and 43,000, and likely contains the monomeric fraction (Mr = 52, 440).

67

0.00
0.05
0.10
0.15
0.20
0.25
0.30
0.35
0.40
0.45
5 μM 1 μM 0.1 μM
OD (450 nm)
untreated
oligomers

Figure 3.11 Colorimetric TUNEL assay of cultured RPE cells treated with
recombinant vitronectin oligomers. No statistically significant difference in labeling
was observed between control and experimental conditions, although an MTT
reduction assay performed after the same treatment period of four hours showed
impaired cell viability (data not shown). Error bars represent the standard error of
the mean (n=3).

68
Chapter IV. Analysis of vitronectin aggregation in human atherosclerotic
plaques
4.1. Introduction/Rationale
Vitronectin has been identified in age-related disease plaques and thioflavin
staining of these insoluble deposits demonstrates an underlying protein misfolding
process (Vallet, Guntern et al. 1992; Anderson, Talaga et al. 2004; Rocken,
Tautenhahn et al. 2006). Due to our findings which demonstrate that vitronectin has
a high propensity to self-associate in vitro, it is likely that vitronectin protein in these
deposits exists in an aggregated state. The aim of this study is to determine whether
vitronectin aggregation occurs in vivo. To confirm our in vitro data and further
support our hypothesis that vitronectin plays a role in the pathophysiology of age-
related diseases, we will investigate whether vitronectin is present in the insoluble
fraction of disease plaques.
Initial isolation of fibrils from human tissue used gentle physical separation,
involving homogenization in saline and low-speed centrifugation (Cohen and
Calkins 1964). A water extraction method was also developed whereby the saline-
insoluble pellet is repeatedly homogenized in water (Pras, Schubert et al. 1968).
Differential sedimentation and solubility have also been employed. In this study, we
will incorporate these methods to extract insoluble material from human tissue.
Ideally, we would prefer to use human ocular drusen for these experiments.
However, we do not have enough drusen material, as we are able collect drusen only
when the donor eyes we receive contain these deposits. In addition, we prefer to

69
optimize this experimental protocol using vitronectin-containing plaques from other
tissues that are readily available. We first attempted to use frontal cortex from
human Alzheimer disease brain, but obtained better results using atherosclerotic
plaques. Human basilar arteries obtained from the USC Alzheimer Disease Research
Center will be used in this study. After homogenizing atherosclerotic plaques in
aqueous buffer, the soluble and insoluble fractions will be separated by
centrifugation and the presence of vitronectin in each fraction will be ascertained by
Western blot. Partitioning of vitronectin to the insoluble fraction of atherosclerotic
plaques would suggest that it is aggregated in these deposits. Trypsin digestion of
the insoluble fraction will also be performed. Detection of the C-terminal, protease-
resistant fragment (described in Chapter 3) following trypsin digestion would
suggest that at least a portion of vitronectin in atherosclerotic plaques is in an
ordered, fibrillar conformation.

4.2. Materials and Methods
4.2.1. Procurement of human tissue samples
Frozen human basilar artery specimens were obtained from the USC Alzheimer
Disease Research Center (ADRC) with IRB approval and stored at -80
o
C.
4.2.2. Preparation of human basilar artery plaques
Samples were thawed at room temperature and rinsed in ice-cold PBS to remove
visible blood. After blotting excess buffer using Whatman filter paper, arteries were
opened lengthwise and atherosclerotic plaques were carefully separated from the

70
vessel wall by microdissection. Plaques were rinsed three times in PBS, blotted, and
weighed. Approximately 150 mg of plaque material was minced with a razor blade,
transferred to a 1.5 mL microcentrifuge tube, and homogenized in 1 mL PBS using a
Polytron PT 1200C (Kinematica; Littau, Switzerland). The homogenate was
distributed evenly between four tubes and centrifuged at 100,000 rpm for one hour at
4
o
C in an Optima TLX ultracentrifuge (Beckman Coulter; Fullerton, CA).
Supernatants were pooled and stored at 4
o
C. Pellets were treated with 100 μl of each
of the following: PBS, 8 M urea, trifluoroacetic acid (TFA), or formic acid. The
samples were incubated at 37
o
C overnight, and centrifuged at 14,000 rpm for 30
minutes at room temperature. Supernatants were collected and protein concentration
was determined using the Bradford assay (BioRad; Hercules, CA). Proteins were
separated by SDS-PAGE and either visualized by Coomassie stain or transferred to a
nitrocellulose membrane for Western blot as previously described in Chapter 3.
To further characterize the solubility of atherosclerotic plaque proteins, we used a
protocol described by Glabe and colleagues (Lesne, Koh et al. 2006) to sequentially
extract extracellular, cytoplasmic, membrane-associated, and insoluble proteins
(Figure 1). Basilar artery samples were thawed and minced as described above,
homogenized in Buffer 1 (50 mM Tris-HCl, pH 7.6, 0.01% NP-40, 150 mM NaCl, 2
mM EDTA, 0..1% SDS, 1 mM PMSF, 1 μM leupeptin, 100 nM aprotinin), and
centrifuged at 3,000 rpm for 5 minutes. The supernatant containing extracellular
proteins was removed. The remaining pellet was homogenized in Buffer 2 (50 mM
Tris-HCl, pH 7.4, 150 mM NaCl, 0.1% Triton X-100), then centrifuged at 13,000

71
rpm for 90 minutes. The supernatant containing cytoplasmic proteins was removed.
The remaining pellet was homogenized in Buffer 3 (50 mM Tris-HCl, pH 7.4, 150
mM NaCl, 0.5% Triton X-100, 3% SDS, 1% deoxycholate, 1 mM PMSF, 1 μM
leupeptin, 100 nM aprotinin), divided evenly among 4 new tubes, and centrifuged at
13,000 rpm for 90 minutes. The supernatant containing membrane-associated
proteins was removed. The remaining pellets were treated with 20 μl 70% formic
acid, 20 μl HFIP, 400 μl 8M urea buffer, or 400 μl of 100 mM ammonium
bicarbonate, pH 8, containing 20 μg trypsin. After one hour, samples containing
formic acid or HFIP were neutralized by the addition of 4 μl 10% SDS, 60 μl 5 N
NaOH, and 336 μl 1 M Tris, pH 8.0. All samples were centrifuged at 13,000 rpm for
90 minutes, and the supernatant containing “insoluble proteins” was removed. Total
protein content was determined using the Bradford assay. Proteins were separated
by SDS-PAGE and either visualized by Coomassie stain or transferred to a
nitrocellulose membrane for Western blot as previously described in Chapter 3.
4.2.3. Trypsin digest
After extraction of proteins from basilar artery plaques in PBS and subsequent
ultracentrifugation, the pellet was treated with a solution containing trypsin (~ 1: 500
protease-to-protein mass ratio) and 10 mM DTT in 100 mM ammonium bicarbonate,
pH 8. Following overnight digestion at 37
o
C, the sample was centrifuged at 14,000
rpm for 30 minutes at room temperature and the supernatant containing material
solubilized by trypsin digestion was collected. Proteins were separated by SDS-
PAGE and visualized by Coomassie stain or transferred to a nitrocellulose membrane

72

Figure 4.1 Protocol for the extraction of extracellular, cytoplasmic, membrane-
associated, and insoluble proteins from human atherosclerotic plaque material
(adapted from Lesne, Koh et al. 2006). Plaques were microdissected from basilar
artery samples and subjected to sequential solubilization and centrifugation steps.
After extraction, a final centrifugation step was performed, and proteins were
separated by SDS-PAGE and either visualized by Coomassie stain or transferred to
nitrocellulose membrane for Western blot.

73
for Western blot as previously described in Chapter 3.
4.2.4. Quantification of vitronectin in plaque material
We used an ELISA-based method to estimate the amount of vitronectin contained
in sequentially extracted plaque fractions. Plasma-purified vitronectin and plaque
fractions were diluted in 0.1 M sodium bicarbonate, pH 9 and allowed to coat for 2
hours at 37
o
C. The plate was washed three times with PBS containing 0.01% Tween
20 (PBS-T), blocked for 2 hours at 37
o
C in 3% BSA/ PBS-T, and washed three times
in PBS-T. Primary anti-vitronectin antibody (diluted 1:10,000 in 3% BSA/ PBS-T;
Santa Cruz Biotechnology, Santa Cruz, CA) was added to each well for 1 hour at 37
°C, washed three times in PBS-T, followed by incubation in horseradish peroxidase-
conjugated anti-rabbit IgG (diluted 1:10,000 in 3% BSA/ PBS-T; Promega, Madison,
WI) for 1 hour at 37 °C. After three washes in PBS-T, the plate was developed with
3, 3, 5, 5 -tetramethylbenzidine (KPL, Gaithersburg, MD). The reaction was stopped
by the addition of 1 M HCl and absorbance was measured at 450 nm. Background
absorbance from uncoated wells was subtracted from each experimental condition.
In order to determine the fraction of vitronectin present in atherosclerotic plaques
relative to the amount of total protein, known quantities of plasma-purified human
vitronectin were used to generate a standard curve by plotting absorbance versus
protein concentration within the linear range of the assay. The data were best
described by the equation y = 0.28613x, where y equals the absorbance at 450 nm
and x equals the quantity of vitronectin in nanograms. This equation was used to
estimate the amount of vitronectin in the basilar artery samples. Total protein

74
content was determined using the Bradford assay. Percent vitronectin was calculated
by dividing the estimated vitronectin quantity by the amount of total protein and
multiplying by 100.

4.3. Results
4.3.1. Vitronectin is present in the insoluble fraction of human atherosclerotic
plaques.
To investigate whether age-related disease deposits contain aggregated
vitronectin, we isolated human basilar artery plaques by microdissection, followed
by homogenization in PBS and subsequent centrifugation to separate the soluble and
insoluble protein fractions. We attempted to liberate additional proteins contained in
the insoluble fraction using acids (to solubilize hydrophobic peptides), chaotropic
denaturants (to dissociate aggregates), detergents (to release membrane-associated
proteins), and enzyme digestion (to cleave accessible peptides). PBS-soluble
proteins, as well as the insoluble fraction resuspended in 8 M urea, 10% SDS, or
digested with trypsin to release additional proteins, are shown visualized by
Coomassie stain in Figure 4.2. Note that in the insoluble fractions, significant
amounts of protein remain in the wells and were not resolved by SDS-PAGE. To
identify which fractions contain vitronectin protein, a Western blot was performed
(Figure 4.3). Vitronectin immunoreactivity was detected in the supernatants
obtained after pellets were treated with 8 M urea, trifluoroacetic acid, formic acid,
10% SDS, and trypsin digestion and centrifuged (Figure 4.3, lanes 3-7), but neither

75
in the initial supernatant fraction nor when the pellet was washed again in PBS
(Figure 4.3, lanes 1 and 2). We consistently observed two major vitronectin bands of
~ 40 kDa and ~ 65 kDa, neither of which corresponds to the molecular weights of
full-length plasma-purified human vitronectin (Figure 4.3, lane Vn).
The amount of vitronectin in the pellet fraction in relation to total protein content
was estimated by ELISA. A standard curve was generated using known quantities of
plasma-purified human vitronectin and plotting absorbance versus protein
concentration within the linear range of the assay. We used the equation y =
0.28613x (where y equals the absorbance at 450 nm and x equals the quantity of
vitronectin in nanograms) to determine the amount of vitronectin in the initial
supernatant, as well as the pellet fraction treated with either 8 M urea or digested
with trypsin. Total protein content was quantified using the Bradford assay. Percent
vitronectin was calculated by dividing the estimated vitronectin quantity by the
amount of total protein and multiplying by 100. As shown in Table 4.1, supernatants
with additional proteins released after the insoluble pellets were treated with 8 M
urea or digested with trypsin and centrifuged, contain more vitronectin than the
supernatant obtained from the initial homogenization of the plaque in PBS,
demonstrating that vitronectin partitions to the insoluble fraction. This suggests that
vitronectin may deposit in an aggregated state within atherosclerotic plaques.
To further characterize the solubility of atherosclerotic plaque proteins, we
employed a previously described protocol used to detect A β assemblies in the mouse

76

Figure 4.2 Soluble and insoluble protein fractions extracted from human basilar
artery plaque. Plaque material was homogenized in PBS, divided into equal portions
and ultracentrifuged. Additional proteins were released from the resulting pellets by
treatment with 8 M urea, 10 % SDS, or trypsin, followed by centrifugation. Proteins
were separated by SDS-PAGE and visualized with Coomassie staining. Vn, 2.5 μg
plasma-purified human vitronectin. SN, supernatant. Ur, 8 M urea. Sds, 10% SDS.
Tr, trypsin. Arrowheads represent molecular weight standards in kilodaltons.

77

Figure 4.3 Vitronectin in present in the insoluble fraction of human basilar artery
plaque. Vitronectin immunoreactivity was not detected in the supernatant fraction
(lane 1), or when the pellet was washed again with PBS (lane 2), suggesting the
preferential partitioning of vitronectin into the PBS-insoluble fractions.
Immunoblotting demonstrates the presence of two vitronectin fragments (~ 40 kDa
and ~ 65 kDa) in the insoluble pellet fraction of a plaque specimen that was
homogenized in PBS, followed by the release of additional proteins from the
resulting pellets by treatment with 8 M urea (lane 3), trifluoroacetic acid (lane 4),
formic acid (lane 5), 10 % SDS (lane 6), or trypsin (lane 7), followed by
centrifugation. Arrowheads represent molecular weight standards in kilodaltons.

78
brain (Lesne, Koh et al. 2006), which involves sequential solubilization and
centrifugation steps. We attempted to extract additional proteins from the final
insoluble pellet using formic acid, HFIP, 8 M urea, and trypsin. After a final
centrifugation step, proteins were separated by SDS-PAGE. Figure 4.4 shows the
sequentially extracted atherosclerotic plaque proteins as visualized by Coomassie
stain. Note that in all conditions, some protein remains in the wells, suggesting the
existence of aggregates that were not resolved by electrophoresis (Figure 4.4, lanes
1-7). Western blot was performed in order to ascertain the presence of vitronectin in
these fractions. Vitronectin is detected in the extracellular, cytoplasmic, membrane-
associated, and insoluble fractions, as seen by Western blot with a polyclonal
antibody (Figure 4.5, left), although there was significantly less vitronectin
immunoreactivity when the pellet was treated with 70% formic acid (Figure 4.5, left,
Lane 4). This may be due to insufficient solubilization, acid hydrolysis of proteins
by formic acid, or improper loading of the sample.
The extracellular, cytoplasmic, membrane-associated, and insoluble fractions
were also analyzed by Western blot using the C-terminal specific M1 anti-vitronectin
antibody. No M1 immunoreactivity was observed in the extracellular and
cytoplasmic fractions (Figure 4.5, right, lanes 1 and 2). Likewise, no
immunoreactivity was observed when additional proteins were liberated from the
insoluble fraction with formic acid, 8 M urea, or trypsin (Figure 4.5, right, lanes 4, 6,
and 7). Vitronectin was detected in the membrane-associated fraction and the
insoluble fraction treated with HFIP (Figure 4.5, right, lanes 3 and 5). This suggests

79

Sample % Vitronectin
Supernatant 0.000982
8 M urea 0.00702
Trypsin digest 0.02446

Table 4.1 Estimated amounts of vitronectin as a fraction of total protein in
atherosclerotic plaques. Known quantities of plasma-purified human vitronectin
were used to generate a standard curve by plotting absorbance versus protein
concentration within the linear range of the assay. The data were best described by
the equation y = 0.28613x, where y equals the absorbance at 450 nm and x equals the
quantity of vitronectin in nanograms. This equation was used to estimate the amount
of vitronectin in the basilar artery samples. Total protein content was determined
using the Bradford assay. Percent vitronectin was calculated by dividing the
estimated vitronectin quantity by the amount of total protein and multiplying by 100.
The data suggest that additional proteins released from the insoluble pellet fractions
after treatment with either 8 M urea or trypsin are enriched in vitronectin, as opposed
to the fraction containing PBS-soluble proteins (supernatant).

80
that these samples may contain peptides composed of the vitronectin C-terminus. It
appears that the 40 kDa band immunoreactive to the M1 antibody (Figure 4.5, right,
lanes 3 and 5) corresponds to the lower molecular weight band seen in the blot
probed with the anti-vitronectin antibody (Figure 4.5, left, lanes 1-7). Additionally,
we can conclude that the 65 kDa band detected by the anti-vitronectin antibody may
not contain the C-terminal peptides recognized by the M1 antibody.

4.4. Discussion
In this study, we show that vitronectin is present in the PBS-insoluble fraction of
human basilar artery plaques, as demonstrated by Western blot. We reliably
observed 40 kDa and 65 kDa vitronectin bands in the insoluble fractions (Figures 4.3
and 4.5), suggesting that the two experiments yielded consistent results. While
neither band corresponds to full-length plasma-purified vitronectin, we can conclude
from the results obtained using the M1 antibody that the higher molecular weight
band does not contain the vitronectin C-terminus, but the lower molecular weight
band does. Using ELISA, we were able to estimate that the percentage of vitronectin
relative to total protein is increased in the insoluble fraction. These results suggest
that vitronectin preferentially segregates into the insoluble fraction and may exist in
an aggregated conformation within atherosclerotic plaques.
The 40 kDa and 65 kDa vitronectin bands observed in Figures 4.3 and 4.5 may
represent amyloidogenic fragments, which can be further characterized by N-
terminal and C-terminal-specific vitronectin antibodies and identified by mass

81

Figure 4.4 Atherosclerotic plaque proteins fractionated according to cellular
compartment (Lesne, Koh et al. 2006). Samples were centrifuged and proteins were
separated by SDS-PAGE and visualized by Coomassie stain. It appears that some
protein remained in the wells and was not completely resolved in each of the
conditions. Vn, 2.5 μg plasma-purified human vitronectin. Lane 1, Extracellular
proteins. Lane 2, Cytoplasmic proteins. Lane 3, Membrane associated proteins.
Lane 4, Insoluble fraction treated with 70% formic acid. Lane 5, Insoluble fraction
treated with HFIP. Lane 6, Insoluble fraction treated with 8 M urea. Lane 7,
Insoluble fraction treated with 20 μg trypsin. Arrowheads represent molecular
weight standards in kilodaltons.

82

Figure 4.5 Vitronectin immunoblot of human atherosclerotic plaque proteins
separated by cellular compartment (Lesne, Koh et al. 2006). Left, vitronectin
immunoblot using a polyclonal anti-vitronectin antibody. Vitronectin is present in
all fractions, although not as abundant when the pellet was treated with 70% formic
acid (Lane 4). The two vitronectin fragments are ~ 40 kDa and ~ 65 kDa, consistent
with the bands detected in Figure 4.3. Right, Vitronectin immunoblot using the C-
terminal-specific M1 anti-vitronectin antibody. Vitronectin was detected in the
membrane-associated fraction, as well as when the pellet was treated with HFIP,
although some protein remained in the wells and was not resolved. The vitronectin
fragments seen in lanes 3 and 5 may correspond to the ~ 40 kDa band seen in the blot
on the left and may contain the vitronectin C-terminus. Lane 1, extracellular-
enriched proteins. Lane 2, Cytoplasmic proteins. Lane 3, Membrane associated
proteins. Lane 4, Insoluble fraction plus 70% formic acid. Lane 5, Insoluble
fraction plus HFIP. Lane 6, Insoluble fraction plus 8 M urea. Lane 7, Insoluble
fraction digested with 20 μg trypsin. Arrowheads represent molecular weight
standards in kilodaltons.

83
spectrometry analysis. An alternative approach is to isolate and purify the
vitronectin fragments using immunoprecipitation with an anti-vitronectin antibody,
followed by amino acid sequencing and/or mass spectrometry. In addition to
identifying the two consistently observed vitronectin fragments, we will examine the
PBS-soluble, extracellular, cytoplasmic, membrane-associated, and insoluble
fractions by electron microscopy. By examining the morphology of the samples, we
can visualize the nature of the aggregates isolated from plaque material at each step.
This will guide further characterization of the sequentially extracted fractions.
In conjunction with repeating these experiments using atherosclerotic plaques
from different patients, we aim to use this protocol using brain tissue from patients
with Alzheimer disease and ultimately, human ocular drusen. We also intend to
identify the amyloidogenic C-terminus of vitronectin within atherosclerotic plaques,
Alzheimer disease brain, and drusen via immunogold labeling using the C-terminal-
specific M1 antibody in conjunction with electron microscopy. The preliminary data
from this study put forth the possibility that vitronectin aggregation occurs in vivo
and warrant additional investigation towards the evaluation of this hypothesis.

84
Concluding Remarks
The aim of this study was to investigate the hypothesis that the multifunctional
protein vitronectin is amyloidogenic and its implications regarding the
pathophysiology of age-related diseases. We first developed a protocol for the
prokaryotic expression and purification of full-length, untagged recombinant
vitronectin that is comparable to plasma-purified human vitronectin as characterized
by ELISA, circular dichroism spectroscopy, and cell adhesion activity. This
provided a renewable, cost-effective, and homogeneous supply of protein to test our
hypothesis. We observed that vitronectin readily aggregates in physiologic buffer,
and subsequently, we empirically determined the optimal conditions for vitronectin
amyloid formation. Soluble prefibrillar oligomers and amyloid fibrils of
recombinant vitronectin exhibit morphologies and biochemical properties similar to
those reported for other amyloid proteins. Vitronectin oligomers are cytotoxic and
permeabilize synthetic vesicles, suggesting a membrane-dependent mechanism.
Vitronectin fibrils contain a C-terminal protease-resistant domain, which likely
contains an ordered amyloid core. This putative core corresponds to two
aggregation-prone regions as predicted by primary sequence analysis using the
TANGO algorithm. Elucidation of the structure of the amyloidogenic core and the
key residues involved amyloid formation require further investigation. Our results
demonstrate that vitronectin has a high propensity to self-associate in vitro,
increasing the likelihood that vitronectin aggregation occurs in vivo. In line with this
hypothesis, we consistently observed two vitronectin fragments in the PBS-insoluble

85
fractions of atherosclerotic plaques from human basilar arteries. To gain more
insight regarding the nature of this distinctive two-band pattern, further
characterization and identification of the fragments is needed. Additionally, we aim
to employ the outlined protocol to extract the insoluble fraction from Alzheimer
disease brain and human ocular drusen, and we fully expect to detect vitronectin in
those samples as well. In conclusion, the inherent amyloidogenic propensity of
vitronectin and toxicity of nonfibrillar vitronectin oligomers suggests that vitronectin
misfolding and aggregation may contribute to the pathophysiology of age-related
diseases.

86
Bibliography
Akiyama, H., T. Kawamata, et al. (1991). "Immunohistochemical localization of
vitronectin, its receptor and beta-3 integrin in Alzheimer brain tissue." J
Neuroimmunol 32(1): 19-28.

Ambati, J., A. Anand, et al. (2003). "An animal model of age-related macular
degeneration in senescent Ccl-2- or Ccr-2-deficient mice." Nat Med 9(11):
1390-7.

Anderson, D. H., G. S. Hageman, et al. (1999). "Vitronectin gene expression in the
adult human retina." Invest Ophthalmol Vis Sci 40(13): 3305-15.

Anderson, D. H., K. C. Talaga, et al. (2004). "Characterization of beta amyloid
assemblies in drusen: the deposits associated with aging and age-related
macular degeneration." Exp Eye Res 78(2): 243-56.

AREDS (2001). "A randomized, placebo-controlled, clinical trial of high-dose
supplementation with vitamins C and E, beta carotene, and zinc for age-
related macular degeneration and vision loss: AREDS report no. 8." Arch
Ophthalmol 119(10): 1417-36.

Barghorn, S., V. Nimmrich, et al. (2005). "Globular amyloid beta-peptide oligomer -
a homogenous and stable neuropathological protein in Alzheimer's disease." J
Neurochem 95(3): 834-47.

Baskakov, I. V., G. Legname, et al. (2002). "Pathway complexity of prion protein
assembly into amyloid." J Biol Chem 277(24): 21140-8.

Bird, A. C., N. M. Bressler, et al. (1995). "An international classification and grading
system for age-related maculopathy and age-related macular degeneration.
The International ARM Epidemiological Study Group." Surv Ophthalmol
39(5): 367-74.

Bittorf, S. V., E. C. Williams, et al. (1993). "Alteration of vitronectin.
Characterization of changes induced by treatment with urea." J Biol Chem
268(33): 24838-46.

Bressler, N. M., S. B. Bressler, et al. (2003). "Potential public health impact of Age-
Related Eye Disease Study results: AREDS report no. 11." Arch Ophthalmol
121(11): 1621-4.

Bucciantini, M., G. Calloni, et al. (2004). "Prefibrillar amyloid protein aggregates
share common features of cytotoxicity." J Biol Chem 279(30): 31374-82.

87
Bucciantini, M., E. Giannoni, et al. (2002). "Inherent toxicity of aggregates implies a
common mechanism for protein misfolding diseases." Nature 416(6880):
507-11.

Butler, A. E., J. Jang, et al. (2004). "Diabetes due to a progressive defect in beta-cell
mass in rats transgenic for human islet amyloid polypeptide (HIP Rat): a new
model for type 2 diabetes." Diabetes 53(6): 1509-16.

Cherny, R. C., M. A. Honan, et al. (1993). "Site-directed mutagenesis of the
arginine-glycine-aspartic acid in vitronectin abolishes cell adhesion." J Biol
Chem 268(13): 9725-9.

Chiti, F., M. Bucciantini, et al. (2001). "Solution conditions can promote formation
of either amyloid protofilaments or mature fibrils from the HypF N-terminal
domain." Protein Sci 10(12): 2541-7.

Chiti, F. and C. M. Dobson (2006). "Protein misfolding, functional amyloid, and
human disease." Annu Rev Biochem 75: 333-66.

Chiti, F., N. Taddei, et al. (2002). "Kinetic partitioning of protein folding and
aggregation." Nat Struct Biol 9(2): 137-43.

Chromy, B. A., R. J. Nowak, et al. (2003). "Self-assembly of Abeta(1-42) into
globular neurotoxins." Biochemistry 42(44): 12749-60.

Cleary, J. P., D. M. Walsh, et al. (2005). "Natural oligomers of the amyloid-beta
protein specifically disrupt cognitive function." Nat Neurosci 8(1): 79-84.

Cohen, A. S. and E. Calkins (1959). "Electron microscopic observations on a fibrous
component in amyloid of diverse origins." Nature 183(4669): 1202-3.

Cohen, A. S. and E. Calkins (1964). "The Isolation of Amyloid Fibrils and a Study of
the Effect of Collagenase and Hyaluronidase." J Cell Biol 21: 481-6.

Crabb, J. W., M. Miyagi, et al. (2002). "Drusen proteome analysis: an approach to
the etiology of age-related macular degeneration." Proc Natl Acad Sci U S A
99(23): 14682-7.

Dahlback, B. and E. R. Podack (1985). "Characterization of human S protein, an
inhibitor of the membrane attack complex of complement. Demonstration of
a free reactive thiol group." Biochemistry 24(9): 2368-74.

Dahlback, K., H. Lofberg, et al. (1987). "Immunohistochemical demonstration of
vitronectin in association with elastin and amyloid deposits in human
kidney." Histochemistry 87(6): 511-5.

88
Dahlback, K., H. Lofberg, et al. (1988). "Immunohistochemical studies on
vitronectin in elastic tissue disorders, cutaneous amyloidosis, lichen ruber
planus and porphyria." Acta Derm Venereol 68(2): 107-15.

Demuro, A., E. Mina, et al. (2005). "Calcium dysregulation and membrane
disruption as a ubiquitous neurotoxic mechanism of soluble amyloid
oligomers." J Biol Chem.

Eikelenboom, P., S. S. Zhan, et al. (1994). "Cellular and substrate adhesion
molecules (integrins) and their ligands in cerebral amyloid plaques in
Alzheimer's disease." Virchows Arch 424(4): 421-7.

Ekmekci, H., H. Sonmez, et al. (2002). "Plasma vitronectin levels in patients with
coronary atherosclerosis are increased and correlate with extent of disease." J
Thromb Thrombolysis 14(3): 221-5.

Esteras-Chopo, A., L. Serrano, et al. (2005). "The amyloid stretch hypothesis:
recruiting proteins toward the dark side." Proc Natl Acad Sci U S A 102(46):
16672-7.

Fernandez-Escamilla, A. M., F. Rousseau, et al. (2004). "Prediction of sequence-
dependent and mutational effects on the aggregation of peptides and
proteins." Nat Biotechnol 22(10): 1302-6.

Ferreira, S. T., M. N. Vieira, et al. (2007). "Soluble protein oligomers as emerging
toxins in alzheimer's and other amyloid diseases." IUBMB Life 59(4): 332-
45.

Friedman, D. S., B. J. O'Colmain, et al. (2004). "Prevalence of age-related macular
degeneration in the United States." Arch Ophthalmol 122(4): 564-72.

Gibson, A. D. and C. B. Peterson (2001). "Full-length and truncated forms of
vitronectin provide insight into effects of proteolytic processing on function."
Biochim Biophys Acta 1545(1-2): 289-304.

Glabe, C. G. (2004). "Conformation-dependent antibodies target diseases of protein
misfolding." Trends Biochem Sci 29(10): 542-7.

Glenner, G. G., E. D. Eanes, et al. (1974). "Beta-pleated sheet fibrils. A comparison
of native amyloid with synthetic protein fibrils." J Histochem Cytochem
22(12): 1141-58.

Guettier, C., N. Hinglais, et al. (1989). "Immunohistochemical localization of S
protein/vitronectin in human atherosclerotic versus arteriosclerotic arteries."
Virchows Arch A Pathol Anat Histopathol 414(4): 309-13.

89
Hageman, G. S., P. J. Luthert, et al. (2001). "An integrated hypothesis that considers
drusen as biomarkers of immune-mediated processes at the RPE-Bruch's
membrane interface in aging and age-related macular degeneration." Prog
Retin Eye Res 20(6): 705-32.

Hageman, G. S., R. F. Mullins, et al. (1999). "Vitronectin is a constituent of ocular
drusen and the vitronectin gene is expressed in human retinal pigmented
epithelial cells." Faseb J 13(3): 477-84.

Hartley, D. M., D. M. Walsh, et al. (1999). "Protofibrillar intermediates of amyloid
beta-protein induce acute electrophysiological changes and progressive
neurotoxicity in cortical neurons." J Neurosci 19(20): 8876-84.

Hogasen, K., T. E. Mollnes, et al. (1992). "Heparin-binding properties of vitronectin
are linked to complex formation as illustrated by in vitro polymerization and
binding to the terminal complement complex." J Biol Chem 267(32): 23076-
82.

Hoshi, M., M. Sato, et al. (2003). "Spherical aggregates of beta-amyloid
(amylospheroid) show high neurotoxicity and activate tau protein kinase
I/glycogen synthase kinase-3beta." Proc Natl Acad Sci U S A 100(11): 6370-
5.

Ishikawa-Sakurai, M. and M. Hayashi (1993). "Two collagen-binding domains of
vitronectin." Cell Struct Funct 18(4): 253-9.

Izumi, M., T. Shimo-Oka, et al. (1988). "Identification of the collagen-binding
domain of vitronectin using monoclonal antibodies." Cell Struct Funct 13(3):
217-25.

Izumi, M., K. M. Yamada, et al. (1989). "Vitronectin exists in two structurally and
functionally distinct forms in human plasma." Biochim Biophys Acta 990(2):
101-8.

Janson, J., R. H. Ashley, et al. (1999). "The mechanism of islet amyloid polypeptide
toxicity is membrane disruption by intermediate-sized toxic amyloid
particles." Diabetes 48(3): 491-8.

Jayasinghe, S. A. and R. Langen (2004). "Identifying structural features of fibrillar
islet amyloid polypeptide using site-directed spin labeling." J Biol Chem
279(46): 48420-5.

Jayasinghe, S. A. and R. Langen (2007). "Membrane interaction of islet amyloid
polypeptide." Biochim Biophys Acta.

90
Jenne, D. and K. K. Stanley (1987). "Nucleotide sequence and organization of the
human S-protein gene: repeating peptide motifs in the "pexin" family and a
model for their evolution." Biochemistry 26(21): 6735-42.

Jin, M., S. He, et al. (2000). "Promotion of adhesion and migration of RPE cells to
provisional extracellular matrices by TNF-alpha." Invest Ophthalmol Vis Sci
41(13): 4324-32.

Johnson, L. V., W. P. Leitner, et al. (2001). "Complement activation and
inflammatory processes in Drusen formation and age related macular
degeneration." Exp Eye Res 73(6): 887-96.

Jones, S., J. Manning, et al. (2003). "Amyloid-forming peptides from beta2-
microglobulin-Insights into the mechanism of fibril formation in vitro." J Mol
Biol 325(2): 249-57.

Kayed, R., E. Head, et al. (2003). "Common structure of soluble amyloid oligomers
implies common mechanism of pathogenesis." Science 300(5618): 486-9.

Kayed, R., Y. Sokolov, et al. (2004). "Permeabilization of lipid bilayers is a common
conformation-dependent activity of soluble amyloid oligomers in protein
misfolding diseases." J Biol Chem 279(45): 46363-6.

Kitagaki-Ogawa, H., T. Yatohgo, et al. (1990). "Diversities in animal vitronectins.
Differences in molecular weight, immunoreactivity and carbohydrate chains."
Biochim Biophys Acta 1033(1): 49-56.

Kobayashi, J., S. Yamada, et al. (1994). "Distribution of vitronectin in plasma and
liver tissue: relationship to chronic liver disease." Hepatology 20(6): 1412-7.

Kuo, Y. M., M. R. Emmerling, et al. (1996). "Water-soluble Abeta (N-40, N-42)
oligomers in normal and Alzheimer disease brains." J Biol Chem 271(8):
4077-81.

Lambert, M. P., A. K. Barlow, et al. (1998). "Diffusible, nonfibrillar ligands derived
from Abeta1-42 are potent central nervous system neurotoxins." Proc Natl
Acad Sci U S A 95(11): 6448-53.

Lambert, M. P., K. L. Viola, et al. (2001). "Vaccination with soluble Abeta
oligomers generates toxicity-neutralizing antibodies." J Neurochem 79(3):
595-605.

91
Lashuel, H. A. and P. T. Lansbury, Jr. (2006). "Are amyloid diseases caused by
protein aggregates that mimic bacterial pore-forming toxins?" Q Rev Biophys
39(2): 167-201.

Lesne, S., M. T. Koh, et al. (2006). "A specific amyloid-beta protein assembly in the
brain impairs memory." Nature 440(7082): 352-7.

LeVine, H., 3rd (1993). "Thioflavine T interaction with synthetic Alzheimer's
disease beta-amyloid peptides: detection of amyloid aggregation in solution."
Protein Sci 2(3): 404-10.

Lue, L. F., Y. M. Kuo, et al. (1999). "Soluble amyloid beta peptide concentration as
a predictor of synaptic change in Alzheimer's disease." Am J Pathol 155(3):
853-62.

Luibl, V., J. M. Isas, et al. (2006). "Drusen deposits associated with aging and age-
related macular degeneration contain nonfibrillar amyloid oligomers." J Clin
Invest 116(2): 378-85.

McGuire, E. A., M. E. Peacock, et al. (1988). "Phosphorylation of vitronectin by a
protein kinase in human plasma. Identification of a unique phosphorylation
site in the heparin-binding domain." J Biol Chem 263(4): 1942-5.

McLean, C. A., R. A. Cherny, et al. (1999). "Soluble pool of Abeta amyloid as a
determinant of severity of neurodegeneration in Alzheimer's disease." Ann
Neurol 46(6): 860-6.

Mehringer, J. H., C. J. Weigel, et al. (1991). "Cyclic AMP-dependent protein kinase
phosphorylates serine378 in vitronectin." Biochem Biophys Res Commun
179(1): 655-60.

Miake, H., H. Mizusawa, et al. (2002). "Biochemical characterization of the core
structure of alpha-synuclein filaments." J Biol Chem 277(21): 19213-9.

Mullins, R. F., S. R. Russell, et al. (2000). "Drusen associated with aging and age-
related macular degeneration contain proteins common to extracellular
deposits associated with atherosclerosis, elastosis, amyloidosis, and dense
deposit disease." Faseb J 14(7): 835-46.

Niculescu, F., H. G. Rus, et al. (1989). "Immunoelectron-microscopic localization of
S-protein/vitronectin in human atherosclerotic wall." Atherosclerosis 78(2-3):
197-203.

92
Novitskaya, V., O. V. Bocharova, et al. (2006). "Amyloid fibrils of mammalian prion
protein are highly toxic to cultured cells and primary neurons." J Biol Chem
281(19): 13828-36.

Ogawa, T., N. Yorioka, et al. (1994). "Immunohistochemical studies of vitronectin,
C5b-9, and vitronectin receptor in membranous nephropathy." Nephron
68(1): 87-96.

Okada, M., K. Yoshioka, et al. (1993). "Immunohistochemical localization of C3d
fragment of complement and S-protein (vitronectin) in normal and diseased
human kidneys: association with the C5b-9 complex and vitronectin
receptor." Virchows Arch A Pathol Anat Histopathol 422(5): 367-73.

Otzen, D. E., O. Kristensen, et al. (2000). "Designed protein tetramer zipped together
with a hydrophobic Alzheimer homology: a structural clue to amyloid
assembly." Proc Natl Acad Sci U S A 97(18): 9907-12.

Ozaki, S., L. V. Johnson, et al. (1999). "The human retina and retinal pigment
epithelium are abundant sources of vitronectin mRNA." Biochem Biophys
Res Commun 258(3): 524-9.

Pawar, A. P., K. F. Dubay, et al. (2005). "Prediction of "aggregation-prone" and
"aggregation-susceptible" regions in proteins associated with
neurodegenerative diseases." J Mol Biol 350(2): 379-92.

Pitt W.G. , F.-H. D. J., Mosher D. F., and Cooper S. L. (1989). "Vitronectin
adsorption on polystyrene and oxidized polystyrene." Journal of Colloid and
Interface Science 129(1): 231-239.

Pras, M., M. Schubert, et al. (1968). "The characterization of soluble amyloid
prepared in water." J Clin Invest 47(4): 924-33.

Preissner, K. T. (1991). "Structure and biological role of vitronectin." Annu Rev Cell
Biol 7: 275-310.

Preissner, K. T., J. Grulich-Henn, et al. (1990). "Structural requirements for the
extracellular interaction of plasminogen activator inhibitor 1 with endothelial
cell matrix-associated vitronectin." J Biol Chem 265(30): 18490-8.

Preissner, K. T. and G. Muller-Berghaus (1987). "Neutralization and binding of
heparin by S protein/vitronectin in the inhibition of factor Xa by antithrombin
III. Involvement of an inducible heparin-binding domain of S
protein/vitronectin." J Biol Chem 262(25): 12247-53.

93
Preissner, K. T., R. Wassmuth, et al. (1985). "Physicochemical characterization of
human S-protein and its function in the blood coagulation system." Biochem
J 231(2): 349-55.

Quist, A., I. Doudevski, et al. (2005). "Amyloid ion channels: a common structural
link for protein-misfolding disease." Proc Natl Acad Sci U S A 102(30):
10427-32.

Quon, D., Y. Wang, et al. (1991). "Formation of beta-amyloid protein deposits in
brains of transgenic mice." Nature 352(6332): 239-41.

Reilly, J. T. and J. R. Nash (1988). "Vitronectin (serum spreading factor): its
localisation in normal and fibrotic tissue." J Clin Pathol 41(12): 1269-72.

Robert, R., M. J. Jacobin-Valat, et al. (2006). "Identification of human scFVs
targeting atherosclerotic lesions: Selection by single round in vivo phage-
display." J Biol Chem.

Rocken, C., J. Tautenhahn, et al. (2006). "Prevalence and pathology of amyloid in
atherosclerotic arteries." Arterioscler Thromb Vasc Biol 26(3): 676-7.

Sanbe, A., H. Osinska, et al. (2004). "Desmin-related cardiomyopathy in transgenic
mice: a cardiac amyloidosis." Proc Natl Acad Sci U S A 101(27): 10132-6.

Sanbe, A., H. Osinska, et al. (2005). "Reversal of amyloid-induced heart disease in
desmin-related cardiomyopathy." Proc Natl Acad Sci U S A 102(38): 13592-
7.

Sarks, J. P., S. H. Sarks, et al. (1994). "Evolution of soft drusen in age-related
macular degeneration." Eye 8 ( Pt 3): 269-83.

Schvartz, I., T. Kreizman, et al. (2002). "The PKA phosphorylation of vitronectin:
effect on conformation and function." Arch Biochem Biophys 397(2): 246-
52.

Schvartz, I., D. Seger, et al. (1999). "Vitronectin." Int J Biochem Cell Biol 31(5):
539-44.

Seger, D. and S. Shaltiel (2000). "Evidence showing that the two-chain form of
vitronectin is produced in the liver by a selective furin cleavage." FEBS Lett
480(2-3): 169-74.

Seiffert, D., G. Ciambrone, et al. (1994). "The somatomedin B domain of vitronectin.
Structural requirements for the binding and stabilization of active type 1
plasminogen activator inhibitor." J Biol Chem 269(4): 2659-66.

94
Seiffert, D., K. Crain, et al. (1994). "Vitronectin gene expression in vivo. Evidence
for extrahepatic synthesis and acute phase regulation." J Biol Chem 269(31):
19836-42.

Seiffert, D., M. Geisterfer, et al. (1995). "IL-6 stimulates vitronectin gene expression
in vivo." J Immunol 155(6): 3180-5.

Sipe, J. D. (1994). "Amyloidosis." Crit Rev Clin Lab Sci 31(4): 325-54.

Sipe, J. D. and A. S. Cohen (2000). "Review: history of the amyloid fibril." J Struct
Biol 130(2-3): 88-98.

Skorstengaard, K., T. Halkier, et al. (1990). "Sequence location of a putative
transglutaminase cross-linking site in human vitronectin." FEBS Lett 262(2):
269-74.

Stanley, K. K. (1986). "Homology with hemopexin suggests a possible scavenging
function for S-protein/vitronectin." FEBS Lett 199(2): 249-53.

Stockmann, A., S. Hess, et al. (1993). "Multimeric vitronectin. Identification and
characterization of conformation-dependent self-association of the adhesive
protein." J Biol Chem 268(30): 22874-82.

Su, H. R. (1996). "S-protein/vitronectin interaction with the C5b and the C8 of the
complement membrane attack complex." Int Arch Allergy Immunol 110(4):
314-7.

Sun, W. H. and D. F. Mosher (1989). "Polymorphism of vitronectin." Blood 73(1):
353-4.

Sunde, M., L. C. Serpell, et al. (1997). "Common core structure of amyloid fibrils by
synchrotron X-ray diffraction." J Mol Biol 273(3): 729-39.

Suzuki, S., A. Oldberg, et al. (1985). "Complete amino acid sequence of human
vitronectin deduced from cDNA. Similarity of cell attachment sites in
vitronectin and fibronectin." Embo J 4(10): 2519-24.

Suzuki, S., M. D. Pierschbacher, et al. (1984). "Domain structure of vitronectin.
Alignment of active sites." J Biol Chem 259(24): 15307-14.

Tollefsen, D. M., C. J. Weigel, et al. (1990). "The presence of methionine or
threonine at position 381 in vitronectin is correlated with proteolytic cleavage
at arginine 379." J Biol Chem 265(17): 9778-81.

95
Tomasini, B. R. and D. F. Mosher (1988). "Conformational states of vitronectin:
preferential expression of an antigenic epitope when vitronectin is covalently
and noncovalently complexed with thrombin-antithrombin III or treated with
urea." Blood 72(3): 903-12.

Tschopp, J., D. Masson, et al. (1988). "The heparin binding domain of S-
protein/vitronectin binds to complement components C7, C8, and C9 and
perforin from cytolytic T-cells and inhibits their lytic activities."
Biochemistry 27(11): 4103-9.

Vallet, P. G., R. Guntern, et al. (1992). "A comparative study of histological and
immunohistochemical methods for neurofibrillary tangles and senile plaques
in Alzheimer's disease." Acta Neuropathol (Berl) 83(2): 170-8.

van Aken, B. E., D. Seiffert, et al. (1997). "Localization of vitronectin in the normal
and atherosclerotic human vessel wall." Histochem Cell Biol 107(4): 313-20.

Walsh, D. M., I. Klyubin, et al. (2002). "Naturally secreted oligomers of amyloid
beta protein potently inhibit hippocampal long-term potentiation in vivo."
Nature 416(6880): 535-9.

Wirak, D. O., R. Bayney, et al. (1991). "Regulatory region of human amyloid
precursor protein (APP) gene promotes neuron-specific gene expression in
the CNS of transgenic mice." Embo J 10(2): 289-96.

Wojciechowski, K., C. H. Chang, et al. (2004). "Expression, production, and
characterization of full-length vitronectin in Escherichia coli." Protein Expr
Purif 36(1): 131-8.

Yatohgo, T., M. Izumi, et al. (1988). "Novel purification of vitronectin from human
plasma by heparin affinity chromatography." Cell Struct Funct 13(4): 281-92.

Zhang, R., L. Barker, et al. (2004). "Mining biomarkers in human sera using
proteomic tools." Proteomics 4(1): 244-56.

Zhu, M., S. Han, et al. (2004). "Annular oligomeric amyloid intermediates observed
by in situ atomic force microscopy." J Biol Chem 279(23): 24452-9.

Abstract (if available)

Abstract Classic amyloidopathies are characterized by the accumulation of misfolded proteins in the form of plaques. Whilst fibril deposition is pathognomonic for amyloidopathies, recent data suggest that prefibrillar species may mediate the development and progression of disease. Soluble amyloid oligomers exhibit similar morphologic and cytotoxic properties, suggesting a shared pathogenic mechanism among amyloid proteins. The oligomer-specific A11 antibody is a useful tool to study amyloid diseases which lack abundant fibril deposition. As part of a collaborative study on the pathophysiology of age-related macular degeneration, we studied protein misfolding in human ocular drusen and demonstrated the presence of prefibrillar oligomers using the A11 antibody. However, the oligomer-forming protein in drusen has not yet been identified. Historically, the amyloidogenic protein is oftentimes the most abundant protein near or within the disease plaque. Vitronectin is one of the most abundant drusen proteins, is contained in all drusen, and is also present within the insoluble deposits associated with Alzheimer disease, atherosclerosis, systemic amyloidoses, and glomerulonephritis. These deposits stain positive with thioflavin, indicating an underlying protein misfolding process. The extent to which vitronectin contributes to amyloid formation within these plaques and the role of vitronectin in the pathophysiology of the aforementioned diseases is currently unknown. The investigation of vitronectin misfolding and aggregation is significant since the formation of oligomers and fibrils is a common ability of amyloid proteins, although they share neither sequence nor native structural homology. In this study, we tested the hypothesis that vitronectin is amyloidogenic.

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

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

PDF

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

PDF

Genetic association studies of age-related macular degeneration from candidate gene to whole genome

PDF

Structure and kinetics of the Orb2 functional amyloid

PDF

Protein phosphatase 2A and annexin A5: modulators of cellular functions

PDF

Overexpression and interaction of Orb2 in S2 insect cells

PDF

Novel synthesis of β-glycosides for SPPS of GLCNAC glycoproteins and study of their site-specific biochemical and biophysical consequences

PDF

Site-specific effects of ubiquitin and ubiquitin-like modifier proteins on α-synuclein aggregation

PDF

The effect of familial mutants of Parkinson's disease on membrane remodeling

PDF

Elements of photoreceptor homeostasis: investigating phenotypic manifestations and susceptibility to photoreceptor degeneration in genetic knockout models for retinal disease

PDF

Uncovering the protective role of protein glycosylation in Parkinson's disease utilizing protein semi-synthesis

PDF

Age related macular degeneration in Latinos: risk factors and impact on quality of life

Asset Metadata

Creator Shin, Thuzar Myo (author)

Core Title Vitronectin misfolding and aggregation: implications for the pathophysiology of age-related diseases

School Keck School of Medicine

Degree Doctor of Philosophy

Degree Program Pathobiology

Publication Date 07/27/2007

Defense Date 06/25/2007

Publisher University of Southern California (original), University of Southern California. Libraries (digital)

Tag age-related macular degeneration,aging,Amyloid,drusen,fibril,OAI-PMH Harvest,recombinant protein

Language English

Advisor Chen, Jeannie (committee chair), Hinton, David R. (committee chair), Langen, Ralf (committee member), Stallcup, Michael R. (committee member), Triche, Timothy J. (committee member)

Creator Email thuzar@usc.edu

Permanent Link (DOI) https://doi.org/10.25549/usctheses-m694

Unique identifier UC1318244

Identifier etd-Shin-20070727 (filename),usctheses-m40 (legacy collection record id),usctheses-c127-521066 (legacy record id),usctheses-m694 (legacy record id)

Legacy Identifier etd-Shin-20070727.pdf

Dmrecord 521066

Document Type Dissertation

Rights Shin, Thuzar Myo

Type texts

Source University of Southern California (contributing entity), University of Southern California Dissertations and Theses (collection)

Repository Name Libraries, University of Southern California

Repository Location Los Angeles, California

Repository Email cisadmin@lib.usc.edu