the contribution of an iron genetic modifier, hfe, to
TRANSCRIPT
The Pennsylvania State University
The Graduate School
College of Medicine
THE CONTRIBUTION OF AN IRON GENETIC MODIFIER, HFE,
TO ALZHEIMER’S DISEASE
A Dissertation in
Neuroscience
by
Eric Christopher Hall II
© 2009 Eric C. Hall II
Submitted in Partial Fulfillment of the Requirements
for the Degree of
Doctor of Philosophy
December 2009
The dissertation of Eric Christopher Hall II was reviewed and approved* by the following:
James R. Connor University Distinguished Professor and Vice Chair of Neurosurgery Dissertation Advisor Chair of Committee
Anne M. Andrews Associate Professor of Molecular Toxicology and Biomedical Sciences Jack T. Rogers Associate Professor of Psychiatry and Neuroscience
Jiyue Zhu Assistant Professor of Cellular and Molecular Physiology
Robert J. Milner Professor of Neural and Behavioral Sciences Director, Graduate Program in Neuroscience
*Signatures are on file in the Graduate School
iii
ABSTRACT
Alzheimer’s disease (AD) is a neurodegenerative disorder of the human central nervous
system characterized by loss of memory that leads to dementia. The pathological characteristics
of AD include an accumulation of amyloid‐β (Aβ) plaques and aggregated
hyperphosphorylated tau protein, neurofibrillary tangles (NFT), throughout the brain. Multiple
hypotheses have been proposed to explain AD etiology including the loss of brain iron
regulation. Among the areas investigated are the of variant forms of the HFE gene as a risk
factor or disease modifier. HFE is an iron regulatory protein that limits cellular iron uptake. A
mutation in the HFE gene can result in a loss of function. Epidemiological studies have
investigated an association between HFE polymorphisms and AD risk; however these studies
are not all in agreement. In order to determine more clearly how HFE could impact
neurodegenerative processes, experiments in this thesis were designed to investigate the
cellular contribution of the HFE H63D polymorphism in proposed AD pathogenic pathways.
We utilized a novel stably transfected human neuroblastoma SH‐SY5Y cell model expressing
HFE polymorphisms for our studies.
The thesis focused on three main areas: 1) HFE and amyloid regulation, 2) HFE and tau
phosphorylation and 3) HFE effects on potential drug treatment strategies in AD. To evaluate
amyloid regulation, we determined amyloid precursor protein (APP) synthesis, processing, and
cellular vulnerability to Aβ toxicity in the presence of different forms of HFE. APP levels
increased with HFE expression, although no effect of HFE variants was observed. There was no
iv
change in APP processing as determined by performing spectrofluorometric secretase‐specific
activity assays. We discovered that cells expressing the H63D variant are more sensitive to Aβ
peptide exposure determined by MTT assay [3‐(4,5‐dimethylthiazol‐2‐yl)‐2,5‐diphenyl
tetrazolium bromide]. There was an up‐regulation of the intrinsic apoptotic pathway as
determined by increases in caspase‐9 expression, caspase‐3 activity, and early apoptosis based
on detection of Annexin V. It appeared that these changes were a result of mitochondria
dysfunction in the cells expressing H63D because we also observed an increase in Bax in the
cellular mitochondrial fraction and cytochrome C levels were increased in the cytosolic fraction.
Mitochondria dysfunction and oxidative stress can act as indirect mechanisms to impact tau
phosphorylation. Tau phosphorylation is regulated by a homeostasis of tau kinase and
phosphatase activity that can be impacted by cellular stressors. In cells expressing the H63D
polymorphism, tau phosphorylation was increased at serine‐residues implicated in NFT
generation. There was no change in phosphatase expression in H63D cells; yet there was an
increase in glycogen synthase kinase‐3 beta (GSK‐3β) activity as determined by measuring GSK‐
3β phosphorylation at its serine‐9 residue. The genotype associated alterations in GSK‐3β
activity with HFE expression prompted us to investigate the role of a novel intracellular
regulator of amyloid and tau phosphorylation; the prolyl‐peptidyl cis/trans isomerase, Pin1.
Oxidative stress has been shown to impact Pin1 expression and activity, thus we hypothesized
that H63D cells would have altered Pin1 activity. Total Pin1 expression levels were not affected
by HFE expression; however there was an increase in phosphorylation of Pin1 at its serine‐16
v
residue suggesting a decrease in Pin1 activity in H63D cells. We also found that iron‐mediated
oxidative stress could increase Pin1 phosphorylation.
Overall, these data suggest that gene‐environment interactions are significant in
elucidating disease etiology and identifying therapeutic targets to improve disease outcomes.
Independent of HFE status, we discovered that GSK‐3β could be impacted by cellular iron.
Thus, increased dietary iron intake could result in even higher GSK‐3β activity in individuals
carrying an H63D allele. GSK‐3β is involved in numerous cellular processes including, but not
limited to: protein phosphorylation, mitochondria function, and apoptosis, which were
impacted by expression of the H63D HFE polymorphism. These data suggest GSK‐3β as a
potential pharmacological target that could greatly improve cellular function in AD patients,
especially those carrying the H63D allele.
vi
TABLE OF CONTENTS
LIST OF FIGURES ................................................................................................................. ix
LIST OF ABBREVIATIONS ................................................................................................. xi
ACKNOWLEDGEMENTS ................................................................................................... xv
Chapter 1 ALZHEIMER’S DISEASE AND CONSEQUENCES OF IRON IMBALANCE ……………………………………………………………………………...1
Introduction ................................................................................................................... 1 Alzheimer’s disease pathway mechanisms ............................................................... 3
Amyloid Cascade ..................................................................................................... 3 Neurofibrillary Tangles…………………………………………………………...7 Neuroinflammation……………………………………………………………....12 Mitochondria Dysfunction…………………………………………………….....14 Oxidative Stress…………………………………………………………………...16
Iron dyshomeostasis and AD……………………………………………......……....17 Iron and Amyloid………………………………………………………………....18 Iron and Neurofibrillary Tangles………………………………………………..19 Iron and Oxidative Stress, Mitochondria dysfunction, and inflammation….20
Iron Regulation……………………………………………………………………….22 HFE…………………………………………………………………………………….26 HFE polymorphisms and AD……………………………………………………29 Pin1…………………………………………………………………………………….41 Pin1 and AD……………………………………………………………………….42 Pin1 polymorphisms and AD……………………………………………………46 Summary……………………………………………………………………………....49 References for Chapter 1……………………………………………………………..51
Chapter 2 THE INFLUENCE OF HFE POLYMORPHISMS ON AMYLOID REGULATION IN SH-SY5Y CELLS…………………………………………………………………..79
Abstract……………………………………………………………………………......79 Introduction…………………………………………………………………………...80 Materials and Methods……………………………………………………………....82
Reagents…………………………………………………………………………..82 Cell Culture……………………………………………………………………...83 Western Blot…………………………………………………………………......83 ELISA assay………………………………………………………………….......84 Secretase Activity assays……………………………………………………….84
vii
Statistical analysis………………………………………………………………85 Results…………………………………………………………………………………85 Discussion……………………………………………………………………………..88 Figures…………………………………………………………………………………93 References for Chapter 2……………………………………………………………..99
Chapter 3 THE H63D GENE VARIANT PROMOTES ACTIVATION OF THE INTRINSIC APOPTOSIS PATHWAY WITH Aβ EXPOSURE……………………..……........104
Abstract………………………………………………………………………………104 Introduction………………………………………………………………………….105 Materials and Methods…………………………………………………………......107 Reagents…………………………………………………………………………..107 Cell Culture……………………………………………………………………….109 Cell Viability Assay……………………………………………………………....109 Preparation of cell lysates………………………………………………………..110 Immunoblotting…………………………………………………………………..110 Caspase‐3 activity assay………………………………………………………….111 Annexin V/PI assay……………………………………………………………….111 Statistical analysis…………………………………………………………………112 Results………………………………………………………………………………....112 Discussion……………………………………………………………………………..117 Figures.………………………………………………………………………………...122 References for Chapter 3……………………………………………………………..129
Chapter 4 EXPRESSION OF THE HFE ALLELIC VARIANT H63D IN SH-SY5Y CELLS AFFECTS TAU PHOSPHORYLATION ..................................................................... 139
Abstract………………………………………………………………………………...139 Introduction………………………………………………………………………..…..140 Materials and Methods……………………………………………………………….142 Reagents………………………………………………………………………….....142 Cell Culture………………………………………………………………………....143 ELISA assays………………………………………………………………………..144 Western Blot………………………………………………………………………...144 Statistical analysis…………………………………………………………………..145 Results…………………………………………………………………………………..145 Discussion………………………………………………………………………………149 Figures…………………………………………………………………………………..153 References for Chapter 4…………………………………………………………..…..162
viii
Chapter 5 PROLYL-PEPTIDYL ISOMERASE, PIN1, ACTIVITY IS COMPROMISED IN ASSOCIATION WITH THE EXPRESSION OF THE HFE POLYMORPHIC ALLELE, H63D…………………………………………………………………………………….167
Abstract…………………………………………………………………………………167 Introduction………………………………………………………………...…………..168 Materials and Methods…………………………………………………………….….171 Reagents……………………………………………………………………………..171 Cell Culture………………………………………………………………………....171 Cell lysate preparation…………………………………………………….……….172 ELISA assays………………………………………………………………………..172 Western blot………………………………………………………………………...173 Statistical analysis…………………………………………………..………….…..174 Results……………………………………………………………………...…………..174 Discussion………………………………………………………………………….......176 Figures……………………………………………………………………………….....180 References for Chapter 5………………………………………………………….…..186
Chapter 6 HFE IN PATHOGENIC DISCOVERY FOR ALZHEIMER’S DISEASE: THERAPEUTIC IMPLICATIONS ............................................................................... 192
Introduction…………………………………………………………………….……..192 What’s the value of current AD drugs?.....................................................................194 HFE, APP, and cell signaling; is there a link?...........................................................195 The impact of GSK‐3β drug discovery……………………………………………...198 Is iron removal the key to defeating AD pathology……………………………….199 Utility of an H63D mouse model……………………………………………………202 Conclusion……………………………………………………………………………..204 References for Chapter 6……………………………………………………………...206
ix
LIST OF FIGURES
1.1. Amyloid Cascade…………………………………………………………………............6
1.2. Neurofibrillary tangle formaion………………………………………………………..11
1.3. Iron uptake and HFE function…………………………………………………….........25
1.4. Pin1 function in Alzheimer’s disease…………………………………………………..45
2.1. Total APP levels in HFE stably transfected SH‐SY5Y cells. ..................................... …93
2.2. Amyloid secretase activity in HFE stably transfected SH‐SY5Y cells .................... …94
2.3. Cellular iron effects on amyloid secretase activity……………………………………95
2.4. Impact of cellular iron exposure on amyloid secretase activity in H63D variant
cells..............................................................................................................................................97
2.5. C83 APP levels in HFE stably transfected SH‐SY5Y cells……………………………98
3.1. Measurement of cell viability………………………………………………….……….122
3.2. The effect of Aβ25‐35 treatment on induction of Bax translocation to mitochondria.123
3.3. The effect of Aβ25‐35 treatment on Cytochrome c release……………………………..124
3.4. The effect of Aβ25‐35 treatment on induction of caspase‐9 activation………………..125
3.5. The effect of Aβ25‐35 treatment on induction of caspase‐3 activity…………………..126
3.6. The effect of Aβ25‐35 treatment on induction of caspase‐8 activation………………..127
3.7. The effect of Aβ25‐35 treatment on apoptosis………………………………………...…128
4.1 Tau phosphorylation in HFE stably transfected SH‐SY5Y cells………………..…....153
4.2. Kinase and Phosphatase expression in SH‐SY5Y……………...……………………...155
4.3. Inhibition of tau phosphorylation…………………………………………..…………..158
x
4.4. Impact of cellular iron on GSK‐3β activity and tau phosphorylation………….....…159
4.5. Tau phosphorylation in SH‐SY5Y cells with Trolox treatment………………………161
5.1. Pin1 expression and activity……………………………………………………………..180
5.2. Cellular iron effects on Pin1 expression and activity………………………………….181
5.3. Iron chelation and Pin1 activity……………………………………………………...….183
5.4. Trolox treatment and Pin1 activity………………………………………………….......184
5.5. HFE effects on Alzheimer’s disease Pin1 substrates…………..…………………...….185
6.1. Cellular impact of H63D HFE on AD pathways………………………………...…….193
xi
LIST OF ABBREVIATIONS
aa amino acid
Aβ amyloid‐beta
AD Alzheimer’s disease
ADAM a disintegrin and metalloprotease
AKT protein kinase B
ALS amyotrophic lateral sclerosis
ApoE Apolipoprotein E
APH anterior pharynx defective protein
APP amyloid precursor protein
ATP adenosine triphosphate
BACE β‐amyloid converting enzyme
BBB blood‐brain barrier
Cdk cyclin‐dependant kinase
CNS central nervous system
DFO desferrioxamine
DMEM Dulbecco’s modified Eagle’s medium
DMT divalent metal transporter
DNA deoxyribonucleic acid
DS Down syndrome
ER endoplasmic reticulum
xii
ERK extracellular signal‐regulated kinase
FAC ferric ammonium citrate
FAD familial Alzheimer’s disease
Fe2+ ferrous iron
Fe3+ ferric iron
GFAP glial fibrillary acidic protein
GI gastrointestinal
GM‐CSF granulocyte‐macrophage colony stimulating factor
GSK glycogen synthase kinase
H2O2 hydrogen peroxide
HFE protein product of HFE gene
HH hereditary hemochromatosis
HRP horse radish peroxidase
IL interleukin
IRE iron responsive element
IRP iron regulatory protein
kD kilodalton
LIP labile iron pool
LPS lipopolysaccharides
M‐CSF macrophage colony stimulating factor
MCI mild cognitive impairment
xiii
MHC major histocompatibility complex
MPT mitochondrial permeability transition pore
mRNA messenger ribonucleic acid
MRI magnetic resonance imaging
MS multiple sclerosis
MTT [3‐(4,5‐dimethylthiazol‐2‐yl)‐2,5‐diphenyl tetrazolium bromide]
NeuN neuron‐specific nuclear protein
NFT neurofibrillary tangles
NMDA N‐methyl D‐aspartate
O2 oxygen
OH∙ hydroxyl radical
PD Parkinson’s disease
PDH pyruvate dehydrogenase
PEN‐2 presenilin enhancer‐2
PS‐1/2 presenelin‐1 or presenilin‐2
PVDF polyvinylidene fluoride
RIPA radioimmunoprecipation assay
ROS reactive oxygen species
sAPP soluble amyloid precursor protein
SH‐SY5Y human neuroblastoma cells
TCA tricarboxylic acid
xiv
Tf transferrin
TfR transferrin receptor
TLR toll‐like receptor
TNF tumor necrosis factor
VDAC voltage‐dependant anion channel
WT wild type
xv
ACKNOWLEDGEMENTS
To my doctoral thesis advisor, Dr. Connor; thank you for your guidance and challenging
me along my journey to become a better scientist. You have led by example in showing the
dedication and commitment required to be a successful scientist, which I admire. Thank you to
my thesis committee members: Drs. Anne Andrews, Robert Milner, Jack Rogers, and Jiyue Zhu.
I appreciate your willingness to serve and the sacrifice of your time in helping me to achieve my
goal.
To my wife, Ayanna, your tremendous support and love over the past few years has
been unrivaled; I appreciate all that you are in my life. Thanks Mom for your love and the
sacrifices made to position me to succeed. Thanks Dad for your support, friendship, and love
throughout my life. To my brother, Ian, thanks for all of your support and keeping me
connected to sports when I was swamped in science. To my grandmothers, thanks for inspiring
and believing in me since I was a young boy to achieve great things. There is still more to come,
I’m still climbing. Thank you to a host of aunts, uncles, cousins, friends, teachers and mentors
who were instrumental in keeping me on the path to succeed. To my Pastor, Dr. Ashe, I
appreciate your support and for showing me what it takes to be a mighty man of valor in my
marriage, relationships, and professional endeavors. Last, but not least, I want to acknowledge
my Lord and Savior, Jesus Christ, for His love and marvelous sacrifice that has enabled me to
achieve this goal and realize much more.
Chapter 1
Alzheimer’s Disease and Consequences of Iron Imbalance
Introduction
Alzheimer’s disease (AD) is a neurodegenerative disorder of the human central nervous
system primarily seen in an elderly population of patients suffering from a loss of memory that
leads to dementia. AD was first discovered by Louis Alzheimer in 1906, in which he described a
patient as having a progressive dementing illness (Alzheimer et al., 1995; Hardy, 2006).
Alzheimer’s disease is progressive and clinically defined by stages based on the patient’s
memory recall, the ability to learn new concepts, and completing assigned tasks to test
executive brain function (Salloway and Correia, 2009). The diagnosis of AD patients is variable
due to its unknown disease etiology. Currently, the most accurate diagnosis is confirmed upon
death. AD is defined pathologically by the accumulation of extracellular amyloid‐beta (Aβ)
plaques and intracellular neurofibrillary tangles (NFT) throughout the brain cortices along with
the loss of neuronal cells, especially in the hippocampus (Masters et al., 1985; Selkoe, 1989).
The etiology of AD appears to be impacted by multiple factors including, but not limited
to: genetic mutations, protein alterations, and environmental factors. Early‐onset AD is
attributed to patients under the age of 65 years who develop AD symptoms as early as 30 years
2
old primarily due to mutations in the genes that code for amyloid precursor protein (APP),
presenilin‐1 (PS1), and/or presenilin‐2 (PS2), which account for 5‐10% of total AD cases (Goate
et al., 1991; Selkoe, 1996a; Thinakaran, 1999). APP is located on chromosome 21 and the
presenilin genes, PS1 and PS2 are located on chromosome 14 and 1, respectively (Thinakaran,
1999).
Late‐onset or sporadic AD occurs in patients 65 years and older, which accounts for 90‐
95% of AD prevalence (Tanzi and Bertram, 2005). The sporadic AD cases are the most elusive in
terms of disease etiology, diagnosis, and treatment. Environmental factors including diet and
toxin exposure may greatly influence sporadic AD onset, which can be further impacted by
genetic modifiers (Tanzi and Bertram, 2005). Apolipoprotein E (ApoE) is a candidate gene for
late‐onset AD and is located on chromosome 19 (Tanzi and Bertram, 2005). Polymorphisms in
the ApoE gene have been associated with increased late‐onset AD risk when the ApoE4 allele is
expressed (Petersen et al., 1995; Tanzi and Bertram, 2005). However, ApoE4 is not a causative
gene and due to the unknown cause(s) of the majority of late‐onset AD cases, therapeutic
efficacy and options are limited.
Current therapies temporarily improve memory loss primarily by treating symptoms of
AD, but not the pathology based on the continued progression of the disease. The current AD
therapy options are cholinesterase inhibitors: donepezil (Aricept ®), galantamine (Razadyne ®),
and rivastigmine (Exelon ®), and a glutamate inhibitor via acting on NMDA receptors:
memantine (Namenda ®) (Shah et al., 2008). All of these drugs act on indirect mechanisms
implicated in AD and do not impact the amyloid and tau pathological markers. Since, AD
3
appears to be a multi‐factorial neurodegenerative disease based on onset, symptoms, and
severity; it would be valuable to further classify AD patients for improved therapeutic
outcomes. This goal may be attained by identifying genetic and environmental factors that
influence known pathological markers.
Alzheimer’s disease pathology mechanisms
Alzheimer’s disease (AD) is a progressive neurodegenerative disease that leads to
learning and memory abnormalities. The disease etiology has eluded medical science experts
for over a century, yet the pathology of AD is clear. Numerous histological pathology studies
have revealed the accumulation of Aβ and NFTs to be the cellular hallmarks of an AD brain
(Glenner and Wong, 1984; Masters et al., 1985; Mattson et al., 1999). Hence, investigators have
focused on discovering the molecular and cellular mechanisms that yield the amyloid and tau
protein pathology markers (Hardy and Allsop, 1991; Selkoe, 1994, 1996a). Ironically, the roles of
these proteins in AD onset are not clearly defined, but much evidence suggests that their effects
are toxic and detrimental to neuronal cells.
‘Amyloid Cascade’
A popular hypothesis regarding AD pathology is the ‘Amyloid Cascade’, which
suggests that increased amyloid production, aggregation, and plaque formation leads to
neuronal loss (Hardy and Higgins, 1992; Hardy and Selkoe, 2002). Amyloid was first isolated
from amyloid filaments in the AD brain by Glenner and Wong in 1984 (Glenner and Wong,
4
1984). It was described as the key protein in the hallmark Aβ plaque of AD brains by Masters
and colleagues in 1985 (Masters et al., 1985). Amyloid precursor protein (APP) is a 110 kDa
transmembrane protein that exists in different isoforms; 695 amino acids (aa) (APP695), 751 aa
(APP751), and 771 aa (APP771) (Golde et al., 1990). APP695 is the predominant isoform and is
highly expressed in neurons (Golde et al., 1990). APP has an extramembranous N‐terminal,
transmembrane sequence, and a cytoplasmic C‐terminal tail (Kang et al., 1987; Selkoe, 1994,
1996b).
APP is processed in one of two cellular pathways: non‐amyloidogenic (α‐secretase) or
amyloidogenic (β‐secretase) (Selkoe et al., 1996; Hardy and Selkoe, 2002). An illustration of this
process can be seen in Figure 1.1. In the non‐amyloidogenic pathway, APP is cleaved by a
disintegrin and metalloprotease (ADAM) within the Aβ sequence of APP (687‐ 688 aa) to
generate soluble APPα (Selkoe et al., 1996; Buxbaum et al., 1998). The ADAM 10 and ADAM 17
proteins compromise the α‐secretase complex (Selkoe et al., 1996; Buxbaum et al., 1998). Soluble
APPα has been shown to offer cellular neuroprotection by improving glutamate transport, cell
adhesion, positive modulation of K+ channels, increased resistance to oxidative stress, increased
NFκB transcription, mitochondrial protection, cholesterol metabolism, synaptotrophic effect
and enhancing cell adhesion (Rogers and Lahiri, 2004; Hardy, 2009). Also, a C‐terminal
fragment, C83, is retained after α‐secretase cleavage (Selkoe et al., 1996; Buxbaum et al., 1998;
Hardy, 2009).
Alternatively, APP can be processed to generate the Aβ peptide in the amyloidogenic
pathway. APP is cleaved at aa 671 by β‐secretase converting enzyme (BACE) (Selkoe et al., 1996;
5
Vassar et al., 1999; Hardy and Selkoe, 2002; Hardy, 2009). There are two isoforms of BACE,
BACE1 and BACE2; BACE1 can initiate the generation of the Aβ peptide by yielding a C‐
terminal fragment of APP containing the Aβ sequence that is 99 amino acids long (C99) (Vassar
et al., 1999; Hardy, 2006, 2009). The cellular function of BACE2 is not known (Vassar et al., 1999;
Hardy, 2006, 2009).
Another enzyme complex known as gamma (γ) secretase cleaves the remaining APP
fragments from the non‐amyloidogenic and amyloidogenic pathways. The γ‐secretase enzyme
complex consists of presenilin‐1 (PS1), presenilin‐2 (PS2), anterior pharynx defective (APH),
presenilin enhancer‐2 (PEN‐2), and Nicastrin (Edbauer et al., 2003; Hardy, 2006). The γ‐
secretase complex can act in the non‐amyloidogenic pathway by cleaving the remaining C83
fragment of APP into P3 and P6 fragments (Edbauer et al., 2003; Hardy, 2006). Also, Aβ
generation is completed by γ‐secretase activity by cleaving the C99 protein fragment at either aa
711 or aa 713 that completes the production of Aβ into mostly a 40 or 42 amino acid length
peptide and a P6 fragment (Vassar et al., 1999; Hardy, 2006, 2009).
6
Figure 1.1: Amyloid Cascade. This diagram illustrates the cellular processing of amyloid precursor protein in either the non‐amyloid generating or amyloid generating (Aβ) cellular pathways.
7
The Aβ peptides generated from the amyloidogenic pathway make up the key
components of the senile Aβ plaque. According to the ‘Amyloid Cascade’, Aβ peptide
production and subsequent plaque formation lead to the other pathological hallmark of AD,
NFT (Hardy and Higgins, 1992; Hardy, 2009). Aβ has been shown to induce tau
phosphorylation in many models supporting the idea that Aβ generation precedes NFT
formation (Greenberg et al., 1994; Adalbert et al., 2007; De Felice et al., 2008).
Neurofibrillary tangles
The neurofibrillary tangles (NFT) found in AD brains consist of paired helical tau
filaments or straight tau filaments that are hyperphosphorylated resulting in a tauopathy
disease (Lee and Trojanowski, 1999; Lee et al., 2001; Lee et al., 2005; Adalbert et al., 2007). Paired
helical filaments were first discovered in NFT of the AD brain by electron microscopy in 1963
(Kidd, 1963). Tau protein was later discovered as the major component of NFT in 1986 by
Grundke‐Iqbal and colleagues (Grundke‐Iqbal et al., 1986; Iqbal and Grundke‐Iqbal, 2006). The
paired helical filaments of tau in AD brains are phosphorylated at more than 30 residues, in
which most are serine‐threonine‐proline residues (Morishima‐Kawashima et al., 1995; Buee et
al., 2000; Wang et al., 2007a). Tau phosphorylation is directly regulated by kinase and
phosphatase activity (Ballatore et al., 2007; Wang et al., 2007a). Indirect mechanisms proposed
to contribute to NFT pathology are oxidative stress, Aβ toxicity, and inflammation (Ballatore et
al., 2007). Several cellular kinases have been shown to phosphorylate tau, but glycogen synthase
8
kinase 3‐beta (GSK‐3β) and cyclin‐dependant kinase 5 (Cdk‐5) are implicated in tau
hyperphosphorylation leading to NFT formation (Lee et al., 2001; Ballatore et al., 2007;
Mazanetz and Fischer, 2007). Thus, I will focus on reviewing the contributions of these kinases
to the tau pathology found in AD.
Glycogen synthase kinase 3‐beta (GSK‐3β) was first discovered by Itarte and colleagues
in 1979 and found to be involved in regulating cellular glycogen metabolism (Itarte and Huang,
1979). Cellular functions of GSK‐3β have been further described and it regulates numerous
cellular proteins including metabolic and signaling proteins, structural proteins, and
transcription factors (Grimes and Jope, 2001; Juhaszova et al., 2004; Huang and Klein, 2006). The
role of GSK‐3β in tau phosphorylation was not discovered until 1988 (Ishiguro et al., 1988) and
it became recognized as tau kinase I until it was later discovered that it was the same protein as
GSK‐3β (Hanger et al., 1992; Ishiguro et al., 1993; Goedert et al., 1995). The activity of GSK‐3β is
regulated by its phosphorylation state at serine‐9 and tyrosine‐216 residues, which can be
regulated by AKT (Grimes and Jope, 2001; De Sarno et al., 2002; Juhaszova et al., 2004).
Transgenic mouse models over‐expressing GSK‐3β display tau hyperphosphorylation and have
behavioral abnormalities (Lucas et al., 2001; Hernandez et al., 2002) suggesting GSK‐3β is a
significant protein contributing to the NFT pathology found in AD brains.
Cyclin‐dependant kinase‐5 (Cdk‐5) is a member of the cyclin‐dependant kinase family of
proteins involved in regulating the cell cycle (Kesavapany et al., 2003). Active Cdk‐5 is only
found in the central nervous system, despite being expressed in other cellular tissues (Nikolic et
al., 1996; Tang and Wang, 1996; Kesavapany et al., 2003). Unlike most cyclin‐dependant kinases,
9
which are activated by associating with cyclin, Cdk‐5 activity is regulated by calcium‐
dependant protease calpain, which cleaves p35 and p39 proteins to p25 and p29, which activate
Cdk‐5 (Lew et al., 1994; Patzke and Tsai, 2002; Kesavapany et al., 2003). Baumann and
colleagues were the first group to discover that Cdk‐5 could phosphorylate tau in AD brains
(Baumann et al., 1993). Since then, numerous studies have gone on to investigate Cdk‐5 in cell
and animal models to evaluate its role in phosphorylating tau. A key study using p25
transgenic mice showed that Cdk‐5 is central to the development of tau aggregation and tangle‐
like formation (Noble et al., 2003). The authors also demonstrated that Cdk‐5 and GSK‐3β co‐
localized and GSK‐3β activity increased as well (Noble et al., 2003). However, a later
investigation found that over‐activation of Cdk‐5 could inhibit GSK‐3β activity in p25
transgenic mice (Plattner et al., 2006). Plattner and colleagues suggested that GSK‐3β is the key
player in tau hyperphosphorylation and Cdk‐5 acts as a modulator based on its ability to inhibit
GSK‐3β activity (Plattner et al., 2006). Additionally, Cdk‐5 has been shown to prime tau
phosphorylation sites for GSK‐3β mediated tau phosphorylation (Sengupta et al., 1997; Li et al.,
2006; Li and Paudel, 2006). Thus, the role of Cdk‐5 continues to be debated regarding its role in
phosphorylating tau in AD.
The first cell culture model of tau hyperphosphorylation demonstrated the initial
neurofibrillary‐like changes in SH‐SY5Y cells by discovering the involvement of phosphatases
(Tanaka et al., 1995; Tanaka et al., 1998; Iqbal and Grundke‐Iqbal, 2006). This was achieved by
culturing the cells in low serum conditions and subsequently inhibiting phosphatases by
okadaic acid (Tanaka et al., 1995; Tanaka et al., 1998; Iqbal and Grundke‐Iqbal, 2006). There are
10
a number of phosphatases (PP): PP1, PP2A, PP2B, PP2C, PP4, PP5, PP6, and PP7 that have been
identified since these key experiments, but their roles under physiological conditions are not as
clear as the kinase activity in regulating tau phosphorylation (Mansuy and Shenolikar, 2006;
Ballatore et al., 2007). The PP2‐A and PP‐1 have been shown to regulate more than 90% of
serine/threonine protein activity in mammalian cells, which suggest that these are the
phosphatases to monitor in disease (Iqbal and Grundke‐Iqbal, 2006; Mansuy and Shenolikar,
2006). Thus, an imbalance of kinase and phosphatase activity can impact NFT generation. A
diagram is provided to highlight the tau residues that are critical to NFT formation (Figure 1.2).
Indirect mechanisms shown to alter the functions of kinases and phosphatases in regulating tau
hyperphosphorylation will be discussed in the proceeding sections.
11
Figure 1.2: Neurofibrillary tangle formation. This diagram illustrates the cellular homeostasis of regulating tau phosphorylation by kinases and phosphatases. Increased phosphorylation of tau at the residues shown increases the formation of neurofibrillary tangles in Alzheimer’s disease brains (Pei et al., 2003; Wang et al., 2007a).
12
Neuroinflammation
The excessive accumulation of Aβ peptides and plaques along with NFT formation can
lead to neuroinflammation, which can cause neurodegeneration through its cytotoxic effects.
Neuroinflammation is another leading hypothesis of AD pathogenesis as microglia dysfunction
has been proposed to be a detrimental factor in the development of AD (Eikelenboom and van
Gool, 2004; Streit, 2004; Streit et al., 2004). Microglia are the resident macrophages of the innate
immune system in the CNS (Streit, 2004; Streit et al., 2004; Tahara et al., 2006). The microglia
found surrounding Aβ plaques in AD brains suggests that Aβ is toxic and the numerous
proinflammatory molecules released indicate an inflammatory process taking place leading to
cell death (Bamberger and Landreth, 2001).
Microglia can become activated by signals such as major histocompatibility complex
(MHC) class I and II proteins, APP, interleukins (IL): IL‐1, IL‐2, IL‐3, IL‐6, mitogens M‐CSF and
GM‐CSF, and TNF‐α that are released in response to cell injury or death (Kreutzberg, 1996).
Proinflammatory cytokines released by microglia such as IL‐1 and IL‐6 can stimulate astrocytes
to secrete more cytokines, as well to activate more microglia for cellular defense (Bamberger
and Landreth, 2001, 2002). Ironically, cytokines have been shown to stimulate APP synthesis
and tau production suggesting that microglia can close the loop in a feedback cycle that drives
the AD pathological process (Hayes et al., 2002). Nonetheless, the presence of microglia and
astrocytes neighboring Aβ plaques in AD brains suggests that these glial cells are scavenging
13
the pathological markers (Bamberger and Landreth, 2001; Combs et al., 2001; Bamberger and
Landreth, 2002; Hayes et al., 2002).
There are three scavenger receptors expressed on microglia that can bind Aβ known as
class A (types I, II, and III), class B (CD36 and SR‐BI), and mucin‐like receptors (Bamberger and
Landreth, 2001; Bamberger et al., 2003). Additionally, toll‐like receptors (TLR) have been
identified on microglia to function in the clearance of Aβ (Tahara et al., 2006). Through AD
transgenic mouse studies of TLR4 mutations, Tahara and colleagues demonstrated that TLR4
plays a significant role in the clearance of Aβ peptides (Tahara et al., 2006). They confirmed
their findings using a LPS‐activated mouse microglia BV‐2 cell line showing that Aβ peptide
exposure increased TLR2, TLR4 and TLR9 expression (Tahara et al., 2006).
Astrocytes have also been shown to surround Aβ plagues and NFT based on be glial
fibrillary acidic protein (GFAP)‐immunostaining (Wisniewski and Wegiel, 1991; Nagele et al.,
2004). Aβ uptake by astrocytes is supported by evidence of Aβ staining in astrocytes of AD
brains (Wisniewski and Wegiel, 1991; Nagele et al., 2004). In the brain, apolipoprotein E (ApoE)
is primarily produced and secreted by astrocytes that function to regulate the transport of
cholesterol (Weisgraber, 1994; Weisgraber et al., 1994; Meda et al., 2001). ApoE can also bind Aβ
to facilitate its degradation by glial cells in the nervous system (Meda et al., 2001; Nagele et al.,
2004). Notably, ApoE is compromised of three isoforms; E2, E3, and E4; E4 expression disrupts
the functions of ApoE (Mahley, 1988). The ApoE4 polymorphism has been shown to be strongly
associated with sporadic AD onset (Tanzi and Bertram, 2005). Thus, an ApoE4 allele may result
in greater inflammation because ApoE cannot aid in clearing Aβ.
14
Additionally, inflammation has been shown to lead to cell death by activating apoptotic
mechanisms (Bamberger and Landreth, 2002). GSK‐3β activity has been shown to be important
in inflammatory cell differentiation, inflammatory cell migration, and proinflammatory
cytokine release (Woodgett and Ohashi, 2005; Jope et al., 2007; Rodionova et al., 2007; Hooper et
al., 2008). These findings advocate for future investigations of GSK‐3β in inflammatory
responses and pathways. The over activation of glial cells by AD pathological markers
implicates neuroinflammation as a key contributor to exacerbating pathology, however, it
would not appear as a causative agent for AD development.
Mitochondria dysfunction
As a cellular organelle, mitochondria are recognized as the ‘powerhouse’ of cells because
of the abundant energy they produce in the form of adenosine triphosphate (ATP) (Mattson et
al., 2008). Numerous mitochondrial proteins have been found to be altered in AD brains
(Castellani et al., 2002; Sullivan and Brown, 2005). It is known that mitochondria exhibit more
dysfunction through aging, but the specific contribution of mitochondria dysfunction to AD
pathology is not as clear (Fukui and Moraes, 2008).
The most compelling finding that mitochondrial dysfunction could cause the
neurodegeneration found in AD is the role of mitochondrial permeability transition pores. The
opening of these mitochondrial inner membrane channels results in the loss of the
electrochemical gradient, caspase activation, and subsequent apoptosis (Wallace, 1999;
Castellani et al., 2002). GSK‐3β, primarily regarded as a tau kinase, has been shown to regulate
15
the activation of the mitochondrial permeability transition (Juhaszova et al., 2004; Murphy,
2004) and apoptosis in numerous studies (Desagher and Martinou, 2000; Beurel and Jope, 2006).
The intrinsic apoptosis pathway has been shown to be mediated by mitochondria
dysfunction and GSK‐3β over activation leading to apoptosis (King and Jope, 2005). A study by
King and Jope demonstrated that GSK‐3β is an intermediate between mitochondria dysfunction
via altered bax oligomerization and apoptosis by caspase‐3 activation based on lithium
inhibition of GSK‐3β (King and Jope, 2005). Additionally, GSK‐3β has been shown to
phosphorylate bax and increase its translocation into mitochondria (Linseman et al., 2004).
When bax oligomerizes at the mitochondrial membrane, a pore is formed that releases
cytochrome c that leads to cytochrome c interacting with Apaf‐1 and pro‐caspase 9, generating a
functional apoptosome that leads to the activation of downstream executioner caspases (Zou et
al., 1999; Linseman et al., 2004). Kluck and colleagues demonstrated that bcl‐2, an inhibitor of
apoptosis, regulates cytochrome c translocation and the release of cytochrome c impacts bcl‐2
function without altering the mitochondrial membrane potential (Kluck et al., 1997). A
deficiency in cytochrome c oxidation has consistently been found in AD brain tissue suggesting
mitochondrial dysfunction (Kish et al., 1992; Mutisya et al., 1994; Castellani et al., 2002).
Furthermore, a yeast two‐hybrid screen of a human brain cDNA library was used to
search for GSK‐3β interacting proteins and pyruvate dehydrogenase (PDH) was identified
(Hoshi et al., 1996). Since PDH is found in the mitochondria, an immunocytochemical study
was done to confirm the presence of GSK‐3β in the mitochondria (Hoshi et al., 1996). GSK‐3β
was shown to phosphorylate PDH implying that GSK‐3β can inactivate PDH. This was a
16
significant finding because of the essential function of PDH converting pyruvate into acetyl
CoA in the TCA cycle and acetylcholine synthesis in cholinergic neurons (Hoshi et al., 1996).
Collectively, GSK‐3β activity appears to impact mitochondrial function through a variety of
mechanisms and impacts other cellular mechanisms implicated in AD pathogenesis.
Oxidative stress
A leading hypothesis for the normal aging seen in humans has been attributed to the
buildup of cellular oxidative stress over time resulting in ‘aging’ or getting older (Terman and
Brunk, 2006). Oxidative stress occurs in a cellular environment when pro‐oxidants exceed
antioxidants resulting in altered cellular metabolism. Oxidative stress can lead to the
production of reactive oxygen species (ROS) in the mitochondria (Sullivan and Brown, 2005;
Fukui and Moraes, 2008) leading to dysfunctional mitochondria mechanisms discussed in the
preceding section. Protein oxidation, lipid peroxidation, and protein carbonyls serve as markers
of cellular stress (Sullivan and Brown, 2005). The generation of hydroxyl radicals and
superoxide can cause DNA damage through deletions and mutations leading to disrupted cell
homeostasis (Sullivan and Brown, 2005). There are numerous antioxidants that the cell
possesses to combat ROS including glutathione, glutathione peroxidase, glutathione
transferases, heme oxygenase, and superoxide dismutases (Sullivan and Brown, 2005).
Oxidative stress has been implicated as a type of cellular disruption that can lead to the
onset of AD. Increased indices of oxidative stress have been found in MCI and AD brains
(Markesbery et al., 2005; Lovell and Markesbery, 2007; Zhu et al., 2007). Protein oxidation and
17
enhanced lipid peroxidation were seen upon immunohistological analysis of AD brains by
Mattson and colleagues (Mattson et al., 1999). Additionally, numerous studies have shown
oxidative stress to impact AD pathology (Sullivan and Brown, 2005; Fukui and Moraes, 2008).
There is an idea that cellular oxidative stress precedes all the cellular mechanisms implicated in
contributing to AD pathology (Maccioni et al., 2009).
There is difficulty in determining whether oxidative stress is a causal agent in AD
pathogenesis because cellular stress can impact virtually all cellular mechanisms discussed in
the preceding sections. The capability of cellular mechanisms to respond to cell stress probably
determines the impact of oxidative stress causing AD pathology. In the next section of this
chapter, I will evaluate the contribution of mismanaged cellular iron levels and iron‐mediated
oxidative stress that can impact numerous AD pathogenic mechanisms. The ability of iron to
directly impact numerous proteins implicated in AD and the contribution of iron‐mediated
oxidative stress may serve as an additional mechanism to impact the etiology of AD.
Iron dyshomeostasis and AD
In AD, there are changes in the cellular environment and modifications in specific
proteins that can influence the pathological characteristics of the disease. Many of these proteins
have been discussed in previous sections of this chapter and the next section of this chapter will
highlight the importance of brain iron homeostasis in the context of AD pathology. It is
important to emphasize that brain iron levels have been shown to increase with aging (Hallgren
and Sourander, 1958; Connor and Menzies, 1995), which has been verified with magnetic
18
resonance imaging studies (Bartzokis et al., 1997; Bartzokis et al., 2004). Elevated levels of iron
have been observed throughout the cerebral cortex and hippocampus of the AD brain (Connor
et al., 1992a; Connor et al., 1992b; Loeffler et al., 1995; Mattson et al., 1999; Sayre et al., 2000;
Quintana et al., 2006).
Iron and Amyloid
The discovery of a functional iron responsive element (IRE) in the 5’untranslated region
of APP mRNA supported the notion of APP being a metalloprotein tightly regulated by
intracellular iron levels affecting its synthesis (Rogers et al., 2002). Thus, it would be logical that
increasing iron levels throughout human aging could lead to an increase in APP substrate for
further generation of Aβ. Moreover, iron levels have been shown to regulate APP processing
through its effects on α‐secretase activity at the cell surface (Bodovitz et al., 1995). It is important
to note that increased iron levels lead to increased production of soluble APPα, but an
accelerated production of this neuroprotective peptide can lead to it being rapidly broken down
by the cell contributing to oxidative stress (Bodovitz et al., 1995). The complexity of such a
system regulating APP processing relative to cellular iron status is very valuable in pursuing
the underlying influence of iron dyshomeostasis throughout aging.
Numerous metal binding sites have been identified on APP and Aβ, implying that
biometals can directly influence amyloid homeostasis (Multhaup, 1997; Barnham et al., 2003;
Barnham and Bush, 2008). Aβ has been shown to be aggregated with numerous biometals,
including iron (Bush, 2003), suggesting that increased accumulation of iron could lead to
19
elevated Aβ aggregation driving AD amyloid pathology via senile Aβ plaque formation.
Potentially, this could place biometal homeostasis at the center of the ‘Amyloid Cascade’. In
contrary to this hypothesis, there are theories regarding APP and Aβ as protective proteins,
which attempt to rescue cells from metal‐induced oxidative stress and toxicity in AD cases
through sequestering biometals (Lee et al., 2005; Castellani et al., 2006). The work of our
laboratory and others revealing iron deposits in senile Aβ plaques and NFTs throughout the
brain implicates a connection among iron dyshomeostasis and AD hallmark characteristics
(Connor et al., 1992a; Connor et al., 1992b; Loeffler et al., 1995; Sayre et al., 2000; Quintana et al.,
2006). The continued investigation of cellular mechanisms impacted by iron and its relationship
with AD pathological markers is warranted.
Iron and Neurofibrillary tangles
Numerous histological studies have shown the accumulation of iron and other biometals
in NFT in AD brains (Good et al., 1992; Smith et al., 1997; Mattson et al., 1999; Shin et al., 2003).
This provides strong circumstantial evidence that iron may be involved in generating NFT
pathology similar to Aβ plaque deposition discussed earlier. A study by Yamamoto and
colleagues demonstrated that iron could induce the aggregation of hyperphosphorylated tau in
post‐mortem AD brains (Yamamoto et al., 2002). In hippocampal cell cultures, iron was shown
to result in a significant decrease in p25, the activator of Cdk‐5, leading to a reduction in tau
phosphorylation suggesting iron could limit NFT formation (Egana et al., 2003). The majority of
these findings demonstrate that cellular iron is associated with NFT; however, the literature
20
lacks mechanistic evidence of iron impacting tau phosphorylation beyond the study by Egana
and colleagues.
Iron and Oxidative Stress, Mitochondria Dysfunction, and Inflammation
It is known that brain iron levels become elevated during the aging process and ferritin,
an iron storage protein, increases to aid in managing intracellular iron levels (Hallgren and
Sourander, 1958; Connor and Menzies, 1995; Thompson et al., 2001). However, this
compensation is not sufficient in AD resulting in an altered iron to ferritin ratio whereby ferritin
levels do not increase in parallel with increasing iron levels (Hallgren and Sourander, 1958;
Connor et al., 1992a; Connor and Menzies, 1995). Thus, such iron mismanagement may cause or
exacerbate AD via oxidative stress. There are numerous factors that allow iron to play a role in
initiating oxidative stress. Any altered activity in the expression or function of iron management
proteins can lead to iron dyshomeostasis whereby oxidative stress mechanisms can emerge to
disrupt biological systems. Iron levels can accumulate throughout the brain potentially
accelerating the aging process as a result of improper iron regulation. The redox‐active state of
iron in a cell can influence oxidative stress as seen in chemical reactions. The Fenton reaction
generates hydroxyl radicals (OH∙) through altered iron regulation leading to the oxidation of
proteins, lipids, and DNA (Thompson et al., 2001). Free radical accumulation via the Fenton
reaction leads to inflammation, which causes cellular stress.
Fenton reaction: Fe2+ + H2O2 → OH• + OH‐
21
It is thought that the breakdown of various biological mechanisms is due to oxidative
stress, which may serve as the catalyst for neurodegeneration. Therefore, iron dyshomeostasis
implicated to be causative of most oxidative stress may accelerate the aging process leading to
the loss of neuronal cells and compromising many cellular functions resulting in complex
disease pathologies. Energy disruption in the form of ATP production and oxygen transport can
become altered through mitochondrial dysfunction where iron plays an important role as well.
The human brain consumes approximately 25% of the total oxygen and mitochondria
are major oxygen consumers (Thompson et al., 2001; Mattson et al., 2008), which makes the
mitochondria critical organelles throughout the nervous system. It is important to note that iron
is required for oxygen delivery and if this process is disrupted, this may lead to altered
mitochondrial function (Thompson et al., 2001). Mitochondria also serve as the substrate for
iron to incorporate into heme to deliver oxygen (Thompson et al., 2001). With mitochondria
serving as the primary site of oxygen consumption for oxidative phosphorylation, oxidative
stress can directly lead to the production of ROS including hydroxyl radicals and nitric oxide
formation within cells (Thompson et al., 2001). ROS have been shown to disrupt numerous
cellular functions and accelerate cellular senescence; probably accelerating the
neurodegenerative process.
Additionally, iron is required in the mitochondria to form iron‐sulfur (Fe‐S) clusters,
which are integrated into proteins involved with many cellular functions (Rouault and Tong,
2005). A key function of Fe‐S clusters is regulating gene expression and transcription, which
have been shown to be critical components within generating the basis of neurodegenerative
22
diseases (Rouault and Tong, 2005). The necessity of iron regulation in the mitochondria to
protect against oxidative stress while allowing iron utilization in Fe‐S clusters and heme
synthesis emphasizes the challenge and importance of mitochondrial iron homeostasis. It is
possible that mitochondrial dysfunction may serve as a common pathway involved in
neurodegeneration as this is the site where the majority of energy is produced along with ROS
generation. Thus, mitochondrial iron regulation is critical to cellular function and may itself
present a rational therapeutic target for ensuring normal cellular function.
Iron regulation
Since I have discussed how iron can impact the cellular pathways implicated in AD
pathogenesis, it is important to understand how iron is regulated. Iron exists as two‐redox
active forms: ferrous (Fe2+) and ferric (Fe3+), which contribute to make iron the most common
biological transition metal in mammalian species (Andrews, 1999a; Andrews et al., 1999;
Thompson et al., 2001). Importantly, iron serves as a cofactor for many biological functions
including DNA synthesis via the ribonucleotide reductase enzyme, ATP production through
interacting with succinate dehydrogenase and aconitase of the TCA cycle, myelin production
via lipid and cholesterol biosynthesis in oligodendrocytes, monoamine neurotransmitter
synthesis, and most importantly in the hemoglobin delivery of oxygen (O2) throughout the
body. The brain is responsible for 25% O2 consumption (Pinero et al., 2000; Thompson et al.,
2001). Disruption of these mechanisms due to improper regulation of iron can lead to altered
gene expression, energy disruption, compromised electrical and chemical neurotransmission,
23
and interrupted oxygen transport. To ensure proper function of these mechanisms, iron is
essential and its levels must be properly managed or regulated.
Iron is typically obtained through one’s diet, where absorption occurs to meet internal
requirements for numerous biological processes within the gastrointestinal (GI) tract. The gut
lumen possesses duodenal crypt cells that monitor an individual’s body iron requirements and
absorb iron as needed (Andrews, 1999a; Andrews et al., 1999). Iron is absorbed via enterocytes,
which deliver iron directly into the blood stream bound to an iron transport protein or store it
in an intracellular iron storage protein (Andrews, 1999a; Andrews et al., 1999). Cellular iron
levels are tightly regulated by iron regulatory protein (IRP) activity where IRPs interact with an
iron responsive element (IRE) in the messenger RNA (mRNA) sequences of cellular iron
management proteins like transferrin (Tf), transferrin receptor (TfR), ferritin, and divalent metal
transporter 1 (DMT1) (Andrews, 1999b; Andrews et al., 1999; Thompson et al., 2001).
There are two IRP proteins known as IRP1 and IRP2, respectively. They bind to IREs,
which are mRNA stem loops that post‐transcriptionally regulate cellular iron homeostasis
(Pinero et al., 2000; Thompson et al., 2001; Rouault and Tong, 2005). Transferrin and ferritin are
the iron transport and storage proteins, respectively. Transferrin binds two molecules of iron
through its two binding domains forming a complex that binds the transferrin receptor.
Endocytosis of this complex occurs via clathrin‐coated vesicles and an acidic pH is achieved
within the endosome to release iron through the divalent metal transporter 1 to ferritin or the
mitochondria (Zecca et al., 2004). There is also another protein complex of HFE and β2
microglobulin that can bind the transferrin complex to regulate iron uptake as well (Feder et al.,
24
1998). A schematic of how HFE influences iron uptake is provided in Figure 1.3. Ferroportin is
a transmembrane exporting protein that transports ferrous iron to ferritin (Dunn et al., 2007).
Ceruloplasmin is another iron regulatory protein that has been shown to load iron into ferritin
as well (Thompson et al., 2001). Ferritin is a 24‐mer subunit protein capable of storing 4500 iron
atoms, which has two types of subunits: (H) and (L) chains; H‐chain oxidizes ferrous (Fe2+) iron
to ferric (Fe3+) iron and the L‐chain contains an iron nucleation site (Lawson et al., 1989; Arosio
and Levi, 2002). There is also a labile iron pool (LIP) that exists in the cell cytosol that contains
free or redox active iron that is readily‐available to be used by the cell for required cellular
mechanisms or cell damaging effects (i.e. Fenton reaction) (Petrat et al., 2000; Kruszewski and
Iwanenko, 2003). Appropriate iron regulation and management are necessary to maintain
normal cellular functions throughout various organ systems in the human body. If iron
homeostasis is not achieved, biological systems are susceptible to oxidative stress potentially
compromising function.
25
Figure 1.3: Iron uptake and HFE function. This diagram illustrates the role of HFE variant
proteins regulating cellular iron uptake at the membrane. Wild type HFE limits the amount of
iron whereas H63D and C282Y HFE do not limit iron uptake. This figure was modified and
adapted from (Connor and Lee, 2006).
26
HFE
Since iron management protein expression is critical to cellular homeostasis, examining
the impact of alterations in such proteins is important, especially in the context of
neurodegenerative disease risk. A key iron regulatory protein was discovered in an iron
overloading disorder known as hereditary hemochromatosis (HH) and known as HFE (name
used to recognize the gene and protein). The HFE gene was mapped to chromosome 6 by Simon
and colleagues in 1975 (Feder et al., 1996). The HFE gene is the most common genetic variation
(1:200) among individuals of European Caucasian descent, which occupies a locus at
chromosome 6p21.3 (Merryweather‐Clarke et al., 2000; Adams et al., 2005).
The most prevalent HFE polymorphisms throughout the world are H63D, C282Y, and
S65C variants (Merryweather‐Clarke et al., 2000; Le Gac et al., 2001). The H63D variant is
caused by a substitution of histidine residue at the 63rd position to aspartic acid resulting in a
minor structural change to HFE (Feder et al., 1996; Dupradeau et al., 2000). The C282Y variant
was discovered due to its primary association with the pathological condition HH, where HH
patients are homozygous for C282Y (Hanson et al., 2001). HH results in increased iron levels
throughout the human body because of excessive absorption of dietary iron leading to the toxic
release of iron (Andrews, 1999a; Levy et al., 1999). C282Y results in an amino acid change at the
282nd position where a cysteine residue replaces the tyrosine residue (Feder et al., 1996). This
results in a significant change in HFE protein structure as the mutation prevents the formation
of disulphide bridges and impacts HFE cellular function (Feder et al., 1996). The S65C missense
27
mutation leads to a serine to cysteine substitution at the 65th amino acid based on the 193 A>T
change and has a population frequency ranging from 1.6 to 5.5% in Caucasians (Camaschella et
al., 2002). S65C polymorphisms have not been found to cause physiological changes as this
mutation has no evidence of affecting HFE protein structure (Le Gac et al., 2001).
HFE encodes for an iron management protein of the same name that is recognized as a
non‐classical major histocompatibility complex (MHC) class 1‐like molecule because it is not
involved in antigen presentation like other MHC class 1 proteins (Ehrlich and Lemonnier, 2000).
Protein crystallography studies of HFE revealed homology to MHC class 1 proteins; they share
37% homology in their ectodomains (Lebron et al., 1998). HFE is a 49 kDa protein that consists
of an α1‐α2 superdomain that forms a platform consisting of an eight‐strand anti‐parallel β
sheet positioned on top of an Ig‐like domain α3 and beta‐2‐microglobulin (β2m) (Lebron et al.,
1998). The narrow MHC class I peptide groove of HFE prevents its function in antigen
presentation (Feder et al., 1997). However, a study by Rohrlich and colleagues showed direct
recognition of human HFE by αβ cytolytic T cells without antigen‐presenting function
suggesting HFE may still function in specific immune responses (Rohrlich et al., 2005).
Furthermore, a study by Roy and colleagues suggested that HFE is involved in the immune
response by treating HFE ‐/‐ mice with lipopolysaccharides (LPS) to generate an inflammatory
response that led to hyposideremia, low serum iron levels, due to insufficient hepcidin
expression (Roy et al., 2004).
The most‐recognizable reported function of HFE protein is its role in regulating cellular
iron uptake. The interactions between HFE and transferrin receptor (TfR) are thought to differ
28
based on mutations in the HFE protein (Lebron and Bjorkman, 1999; Lebron et al., 1999;
Ramalingam et al., 2000). Wild‐type HFE is thought to restrict TfR binding so that only one
mole of Tf can bind. The H63D allele HFE protein migrates to the membrane to form a complex
with TfR, but does not restrict Tf binding resulting in greater iron uptake because it does not
decrease the affinity for diferric transferrin (Feder et al., 1997; Waheed et al., 2002). The C282Y
HFE protein does not reach the membrane due to incorrect folding of the HFE α3 domain
preventing β2m binding, and hence, does not form a complex with TfR resulting in increased
cellular iron uptake (Feder et al., 1997; Waheed et al., 1999; Ramalingam et al., 2000; Waheed et
al., 2002). This was further confirmed based on data showing that the C282Y HFE protein is
retained in the trans‐Golgi complex and endoplasmic reticulum (Feder et al., 1997; Waheed et
al., 1999; Ramalingam et al., 2000; Waheed et al., 2002). Ultimately, it appears that the known
HFE mutations lead to increased cellular iron levels.
In the brain, HFE expression is present in the choroid plexus, blood vessels, and
ependymal cells (Connor et al., 2001), which line the ventricles of the brain. This would further
implicate HFE in regulating iron uptake based on concurrent expression of TfR within the
vascular endothelium of the brain. The HFE protein is expressed in neurons surrounding Aβ
plaques and NFTs in AD brains (Connor et al., 2001; Connor and Lee, 2006). Since HFE
expression was minimal in control tissue compared to AD brains, the idea that HFE may be
induced was proposed (Connor et al., 2001). Evidence of HFE induction was further seen in a
BV‐2 microglia cell line when exposed to toxic agents (Lee and Connor, 2005). There is evidence
of HFE function in innate immunity (Ehrlich and Lemonnier, 2000; Salter‐Cid et al., 2000; de
29
Almeida et al., 2005; Porto and De Sousa, 2007), which suggests that it has diverse functions and
may play a significant role in regulating cellular iron and cellular stress. Genome‐wide studies
have identified a genetic locus on chromosome 6 as a risk factor for sporadic AD and HFE is a
candidate gene (Tanzi and Bertram, 2005).
HFE Polymorphisms and AD
HFE variants have been examined in association with neurological diseases including
Alzheimer disease (AD) (Connor and Lee, 2006), amyotrophic lateral sclerosis (ALS) (Wang et
al., 2004; Yen et al., 2004; Goodall et al., 2005; Restagno et al., 2007; Sutedja et al., 2007),
Parkinsons disease (PD) (Borie et al., 2002; Dekker et al., 2003; Guerreiro et al., 2006), multiple
sclerosis (MS) (Rubio et al., 2004; Ristic et al., 2005; Ramagopalan et al., 2008), and stroke
(Ellervik et al., 2007). The presence of the HFE polymorphisms, H63D and C282Y, has been
evaluated in studies to determine their effects on disease risk. All studies investigating AD and
HFE polymorphism associations examined the frequency of HFE polymorphisms between AD
patients and healthy participants, and/or neurological disease controls (neurological pathology,
but not AD). Additionally, some of the studies examined the impact of HFE polymorphisms on
AD age at onset, effects of gender, and additional gene variants that may influence AD risk as
well. It is important to note the diverse populations of patients these studies cover
geographically, which should be considered in the context of gene‐environment interactions.
There are thirteen genetic association studies that were conducted over eight years
investigating the role of HFE polymorphisms and AD risk. I will examine each study
30
individually and summarize the significant findings in chronological order of publication. A
meta‐analysis of HFE polymorphisms and AD risk can be found at (http://www.alzgene.org). In
summary, the majority of epidemiological studies did not find an overall association of HFE
polymorphisms with AD development. This is not alarming because my interest is centered on
how an HFE polymorphism such as H63D may modify AD onset and progression. Eight studies
showed a positive association between age at AD onset and HFE polymorphisms with or
without ApoE4 polymorphisms, three studies did not evaluate age at onset, and the two studies
that did not show an effect were not clear on their methodology for not finding an effect. These
findings are encouraging because it shows that HFE polymorphisms can impact AD onset and
warrant further investigation. Thus, it is logical to pursue the potential functional consequences
of expressing HFE polymorphisms such as H63D on cellular mechanisms implicated in AD
pathogenesis to discover possible causes for increased age of AD onset with HFE
polymorphisms.
In the year 2000, Moalem and colleagues published the first study investigating the
possible association of HFE polymorphisms with AD (Moalem et al., 2000). This study
examined familial AD cases, adult Down syndrome (DS) patients, and younger and healthy
adult controls. All samples were Caucasians from Toronto, Canada, except samples from DS
cases, which were obtained from New York. The study also considered the risk of the ApoE
gene variant and gender determinants on AD risk. H63D had a frequency of 0.38 in male AD
cases compared to 0.15 in younger controls and 0.21 in older controls. There was a 0.14
frequency of H63D in female AD patients and a much higher frequencies in female controls of
31
0.50 and 0.27 in the younger and older controls, respectively. ApoE4 frequency was 0.83 in AD
males and 0.79 in AD females compared to control frequencies of approximately 0.30, which is
consistent with ApoE4 being associated with AD. 41.7% of AD cases were HFE polymorphism‐
positive (H63D and/or C282Y) and ApoE4‐postive compared to 16.6% of cases of HFE
polymorphism‐positive and ApoE4‐negative. The authors conclude that HFE polymorphisms
are associated with AD in males, but are somewhat protective in females. They also note the
over‐representation of male AD cases that are H63D‐postive and ApoE4‐positive suggesting
this combination to increase AD risk.
Sampierto and colleagues published the first study to examine the role of HFE
polymorphisms on the age at onset of AD and the frequency of HFE variants in AD cases
(Sampietro et al., 2001). An Italian population of subjects was investigated in this study
consisting of sporadic AD cases and age‐matched healthy controls. The frequency of H63D in
AD cases was 0.11 and 0.14 in controls showing that there was not an association with this HFE
polymorphism and AD. ApoE4 expression was also determined and found to have a frequency
of 0.24 in AD versus an allelic frequency of 0.09 in control patients. The authors’ state there is no
effect of sex on the distribution of HFE and ApoE polymorphisms in the patients (data not
shown). All patient groups were stratified according to 60‐69 years age at onset, 70‐79, and
greater than 80. The frequency of the H63D allele was 0.22 in the 60‐69 AD group (n = 23), which
was almost double the 0.12 frequency of H63D in the 70‐79 AD group (n = 47), and five times
more than the 0.04 frequency of the greater than 80 years of age AD group (n=37). The authors
note that they could not determine an increased AD risk with ApoE4 and H63D, but noted that
32
the oldest mean age at onset of 77.3 years was observed in a subset of 51 patients that were
wild‐type HFE positive and ApoE4 negative providing indirect evidence that a synergistic effect
may still exist. In summary, the H63D polymorphism increased the age at onset by five years in
AD cases.
Another study investigated HFE polymorphisms and AD risk along with the frequency
of the C2 transferrin allele in a Spanish population (Lleo et al., 2002). The frequency of the H63D
allele was 0.26 in AD patients and 0.20 in healthy controls, C2 allele frequency was 0.17 and 0.18
in controls, and C282Y frequency was 0.02 in both groups. Thus, there was not an association
between HFE polymorphisms and AD risk. The authors noted that there was an increase in
H63D in male AD patients (53.6%) versus 33.3% in control patients (data not shown). They also
commented that ApoE frequency or age at onset did not yield any significant difference (data
not shown). I agree with the authors that HFE polymorphisms are not associated with AD in
their patient population, but they do not explicitly state that HFE polymorphisms do not
influence age at onset, nor do they provide their methodology for such a conclusion.
A study by Combarros and colleagues investigated the role of HFE polymorphisms in
AD risk, age at onset, and gender contributions (Combarros et al., 2003). The AD patients and
healthy controls were from a geographical region in northern Spain similar to the Lleo et al.
study discussed above. The frequency of the H63D allele was 0.27 in all patients indicating that
this HFE polymorphism is not associated with AD. However, the authors evaluated the impact
of H63D polymorphisms on age at onset. The age groups were stratified accordingly to the
Sampierto et al. study and they did not find the expression of the H63D allele alone was not
33
able to impact age at onset. They also mentioned that there was no difference in the distribution
of H63D in males and females (data not shown). Importantly, they evaluated ApoE4 status and
found that H63D and ApoE4 positive patients increased the age at onset by at least four years. I
conclude that there may be a synergistic effect of H63D and ApoE4 increasing AD age at onset
despite the lack of association of HFE polymorphisms and AD risk.
The first North American study investigating HFE polymorphisms and AD association
was led by Pulliam and colleagues (Pulliam et al., 2003). AD, mild cognitive impairment (MCI),
and neurological high pathology and low pathology control patients from Kentucky were
evaluated. This study is challenging to evaluate because the data are presented as HFE
mutations (could be H63D, C282Y, or S65C variants), which does not allow a detailed analysis
of HFE variants for an association with AD. The authors state the frequency of HFE
polymorphisms in the entire study population (not group‐specific) to be 0.183 for H63D, 0.073
for C282Y, 0.034 for S65C, and 0.710 for the wild type HFE variant, which fit the expected
Hardy‐Weinberg frequency. They found an increased proportion of homozygous or compound
heterozygous HFE polymorphisms in the AD group compared to the low pathology controls
suggesting an increased association of HFE variants with AD. The group also evaluated ApoE4
and the found the presence of ApoE4 and an HFE mutant to significantly increase the risk of
cognitive impairment based on mental status test battery and neurological examination. Since
HFE polymorphisms are associated with increased oxidative stress, this group measured CSF
F2‐Isoprostance levels to measure brain lipid peroxidation as indices of oxidative stress in the
AD group as a mechanism contributing to the HFE polymorphism association with AD. CSF F2‐
34
Isoprostance levels were approximately 75 pg/ml in patients with HFE mutations and 50 pg/ml
in wild type HFE carrying individuals confirming the increased cellular stress associated with
the HFE mutants. The authors conclude that HFE mutations were significantly increased with
AD and cognitively impaired subjects. This study showed an association of HFE
polymorphisms with AD; however, the specific contributions of the HFE variants were not
reported. Also, there was not an evaluation of age at onset provided either.
An Italian group evaluated the association between AD and HFE polymorphisms to
confirm and extend the findings of the studies presented above. The H63D, C282Y, S65C HFE
variants and the ApoE4 polymorphism were evaluated in AD patients and healthy controls
from Northern Italy (Candore et al., 2003). In the AD patient population, C282Y had an allelic
frequency of 0.2%, H63D‐13.8%, and S65C‐1.6% compared to similar values for the control
population: C282Y‐0.3%, H63D‐11.2%, and S65C‐2.0% implying there is not an association
between HFE polymorphisms and AD risk. The group also notes that gender was not a factor
during analyses either. Upon evaluating age at AD onset, no significant differences were found
based on HFE polymorphisms or gender. I would argue that it would be difficult to confirm this
finding when the data were not stratified into age ranges like the Sampierto et al. (2000) and
Combarros et al. (2003) studies to be evaluated. To further support my viewpoint, the
variability of age range was quite different in AD patients with wild type HFE (67.93 ± 2.54)
compared to H63D HFE (67.79 ± 9.14), which argues for the age at onset stratification by decade
(i.e. 60‐69 years, 70‐79 years, and 80 years and above). Also, Candore and colleagues state
35
ApoE4 status did not impact AD risk in combination with H63D, but no ApoE data is provided
to assess a potential impact on the age at AD onset.
HFE polymorphisms were evaluated as genetic modifiers in AD, MCI, and elderly
healthy controls from Montreal, Canada in a study by Berlin and colleagues (Berlin et al., 2004).
ApoE and HFE variants H63D and C282Y frequency was determined. H63D was found in 33%
AD patients, 26% in MCI, and 34% in normal elderly controls showing no association between
H63D variants and AD. The C282Y HFE variant did not shown an association either as it was
found in 5% of AD patients, 6% MCI, and 10% in the control patients. ApoE4 allele frequency
was 54% in AD, 31% in MCI, and 21% in normal controls. There were no gender effects.
Additionally, this group displayed an excellent effort in investigating the effect of HFE
polymorphisms on age at AD onset. They presented a Kaplan‐Meier survival curve showing the
effect of H63D on age at AD onset based on symptoms first reported by relatives. An overall
statistically significant difference was not shown based on median age, but if one observes the
age ranges from 55 to 70 years there is an increase in AD age at onset in H63D homozygotes
based on the shift of the survival curve. Furthermore, the authors describe the impact of H63D
HFE on the onset of AD and MCI subjects combined that revealed an over‐representation of
H63D in presentation of cognitive impairment in patients ranging from 55 to 75 years similar to
the age at onset effect I observed. The authors state that there were no associations between
ApoE4 and the HFE polymorphisms (data not shown). It would be of value if the HFE/ApoE4
patient data were accessible to assess the impact of age at AD onset, not just the AD association
analysis or age at AD onset independently, especially given my viewpoint on the age at onset
36
data in their study with the H63D homozygotes. Nonetheless, the authors conclude that HFE
polymorphisms do not increase AD risk, but suggest that the trends of accelerated cognitive
decline in H63D homozygotes they observed should be investigated further.
Robson and colleagues investigated the transferrin C2 allele, HFE polymorphisms, and
ApoE4 variant expression on the development of AD (Robson et al., 2004). AD, MCI, and
control patients were Caucasians from the Oxford region previously evaluated as a cohort of the
Oxford Project to Investigate Memory and Ageing (OPTIMA) study (Petersen et al., 1999). The
transferrin C2 allele was found in 42.4% AD, 28.9% MCI, and 33.5% control patients. C282Y
variants were expressed in 15.7% AD, 24.6% MCI, and 11.9% control patients. The H63D variant
was found in 27.8% of AD cases, 24.6% MCI, and 27.9% in control patients. Thus, none of the
HFE polymorphisms or the C2 allele was associated with AD risk. The Robson group
performed logistic regression and synergy factor analysis to examine the interactions of gene
variants in AD risk. They discovered an association of AD with both the expression of the C2
allele and C282Y variant suggesting a five times greater risk of developing AD. The authors
comment that either the C282Y or C2 variant affects age of AD onset based on the overall mean
age (data not shown). Again, I would emphasize that their conclusion cannot be confirmed
based on their methodology of comparing the mean age and not evaluating independent age
ranges. ApoE4 allele frequency data was not shown, but the authors state ApoE4 only increased
risk of C282Y/C2 bi‐carriers making them tri‐carriers increasing overall AD risk. They also state
that HFE polymorphism bi‐carriers of H63D and C282Y along with C2 allele expression
increases AD risk. The only issue I have with these conclusions is the frequency of these tri‐
37
carriers throughout the population and how relevant it could be. I conclude that there is not an
association of HFE polymorphisms and AD in this Northern European population, but the age
at onset effects cannot be fully evaluated with the data provided.
HFE polymorphisms and AD risk were investigated in a Portuguese population by
Guerreiro and colleagues (Guerreiro et al., 2006). They also examined the potential association
of HFE polymorphisms and Parkinson’s disease, which will not be discussed as the focus of this
discussion is with AD risk. The HFE polymorphisms, H63D and C282Y, were evaluated in AD,
MCI, and healthy controls. The H63D allele was found in 34.6% of AD, 38.9% MCI, and 35.6%
control patients supporting the notion that H63D is not associated with AD risk. C282Y was
present in 4.6% AD cases, 5.5% MCI, and 4.3% in controls suggesting no effect on AD
development. This group also investigated the impact of HFE polymorphisms on age at AD
onset using Kaplan‐Meier survival curves. They failed to find a statistically significant overall
association between HFE polymorphisms and age at onset. I would like to point out the
trending increase of age at AD onset in H63D homozygotes at approximately 70 years of age
and younger similar to the previous findings of Moalem et al. (2000), Berlin et al. (2004), and
Combarrros et al. (2003). They authors also commented that stratifying the effects of HFE
polymorphisms with ApoE4 variants did not reveal any significant differences in AD
frequency, but they did not comment on a possible synergistic effect of an HFE polymorphism
and ApoE4 on age at AD onset.
A population‐based cohort study of almost 8,000 persons aged 55 years or older known
as the Rotterdam Study in the Netherlands was used to select a random group of patients to
38
evaluate AD risk with HFE polymorphisms and the ApoE4 allele (Alizadeh et al., 2009). The
data are reported as HFE carriers, which could consist of heterozygosity or homozygosity for
either the H63D or C282Y variant. Hence, there were 20.8% of AD men with a HFE
polymorphism versus 28.8% of control men and 15.8% of AD women with a HFE
polymorphism compared to 26.5% of control women. Based on these data, HFE polymorphisms
are not associated with AD risk. Interestingly, the group evaluated the bi‐carrier effects of
having an HFE polymorphism with an ApoE4 polymorphism and assessed AD association.
There were 18.1% of AD men that were bi‐carriers for HFE and ApoE4 variants compared to
10.7% of control men leading to a trending, but not a statistically significant increased risk for
AD. 16.9% AD women expressed an HFE polymorphism and ApoE4 with respect to 11.2 % in
the women control patients, which was not statistically significant regarding an association with
AD development. Alizadeh and colleagues commented that the age at AD onset was earlier in
H63D homozygotes compared to wild‐type HFE carriers, which is consistent with previous
studies and my perspective on the trending data in earlier studies. Furthermore, they stated that
AD men who were bi‐carriers for H63D and ApoE4 (73.2 ± 2.1) displayed an average of 5.5
years earlier at AD age onset with respect to AD non‐carriers (78.7 ± 1.6). The authors conclude
by stating that HFE polymorphisms are not strong determinants of AD, but the H63D variant
may accelerate the age of AD onset.
Blazquez and colleagues evaluated HFE polymorphisms, transferrin C2 allele, and
ApoE4 variants with AD susceptibility in an area known as Basque Country in Spain (Blazquez
et al., 2007). The H63D polymorphism was present in 18% of AD patients and 29.9% of control
39
patients, which prompted the authors to suggest that the H63D variant may serve a protective
role in preventing AD. Since, no other group has shown or suggested such an idea; the authors
suggest that genetic linkage disequilibrium may be responsible for such an effect. I believe these
data are informative from the standpoint of geographic region and possible environment agents
that may be different from other regions. The C282Y variant was found in 4.5% of AD patients
and 3.3% of control patients, which is consistent with many of the other studies. The C2 allele
was present in 14.5% of AD patients and 15.0% of control subjects. ApoE4 polymorphisms were
increased in AD patients (27.2%) with respect to control patients (10.5%). The authors suggest
that an increase of age at AD onset was not affected by the H63D variant as shown by others,
which would be consistent with their H63D and AD association data. Gender was not
considered in their analysis of data either which could provide additional information as well
given the data they obtained that was inconsistent with all other studies of HFE polymorphisms
and AD. Nonetheless, this study does not support an association of HFE polymorphisms and
AD risk nor an effect on age at AD onset.
An investigation by Avila‐Gomez and colleagues assessed the association of HFE
polymorphisms with early‐onset familial AD cases (Avila‐Gomez et al., 2008). The familial AD
patients in this study had a mutation in the preseninlin‐1 gene known as E280A were evaluated
in comparison to non‐demented controls that were not relatives of the AD patients. H63D allele
frequency was 18.1% in AD patients and 16.14% in control subjects, which is very similar
indicating that there is not association with AD. The allelic frequency of the C282Y variant was
non‐existent in AD patients (0%) and 0.23% in the control patients. The authors evaluated the
40
age at onset by comparing the median age of AD patients stratified by H63D presence and did
not find an effect. The age range of the familial AD patients was 33‐55 years, so I would
recommend evaluating an HFE effect by examining the patients in five year subgroups (i.e. 35
years and younger, 36‐40 yrs, etc.) to better evaluate a possible effect. Nevertheless, the authors
noted that the AD patient population in Colombia from the Antioquia region studied within
this investigation is rather large given that the community is a genetic isolate population
(Arcos‐Burgos and Muenke, 2002). This notion introduces the idea that the environmental
changes that could impact HFE polymorphisms in AD risk may be absent in such an isolated
environment.
It is also noteworthy to mention a study by Percy and colleagues examining the role of
ApoE4 and H63D variants in sporadic AD cases with folate‐supplementation in Ontario,
Canada (Percy et al., 2008). Folic acid fortification has been mandatory since 1998 and the
authors investigated HFE polymorphism effects on blood markers of iron, red cell folates, and
serum B12 levels. They found the H63D variant to be associated with lowering red cell folate
concentration, which they note that this is probably due to the increased oxidative stress
associated with the H63D allele. The group also examined the synergistic effect of ApoE4 and
H63D polymorphisms on AD risk and found gender differences. In females, they found that
ApoE4 predisposition to AD was increased in combination with the H63D allele. In males, they
found the predisposition to AD with ApoE4 expression to be lowered with expression of the
H63D allele. In contrast, Moalem et al. (2000) found that bi‐carriers of H63D and ApoE4
polymorphisms increased AD risk. It is also important to note that this was a pilot study with
41
only twenty AD males, which increases the likelihood that this may be a false negative finding
due to a low sample number. Nevertheless, this group’s findings that H63D lowers folate
concentration due to oxidative stress supports the cellular impact that H63D may have in
increasing neurodegenerative risk.
Pin1
Pin1 has garnered much attention for its role in AD and other diseased states (Wulf et
al., 2005; Balastik et al., 2007; Takahashi et al., 2008). The discovery of Pin1, a peptidyl‐prolyl
cis/trans isomerase (PPI), by Lu and colleagues was a pivotal finding with respect to known PPI
functions because Pin1 was shown to regulate mitosis (Lu et al., 1996). PPI proteins including
Pin1 have been shown to be involved in protein assembly, protein folding, protein localization,
protein‐protein interactions, protein phosphorylation, and enzymatic activity (Wulf et al., 2005;
Lu and Zhou, 2007). There are numerous Pin1 substrate proteins that are involved in mitosis,
cytoskeleton structure, transcription, and cell cycle regulation; for a thorough review see (Lu et
al., 2002a; Wulf et al., 2005; Lu and Zhou, 2007). The nature of Pin1 interactions with its cellular
targets is a unique mechanism that is required for cellular homeostasis.
Protein substrate binding of Pin1 occurs through its N‐terminal WW domain (Lu et al.,
1999a; Lu et al., 2002b). Pin1 function is impacted by its phosphorylation state at its serine 16
residue, which is located at the center of the phosphorylated serine/threonine‐proline binding
pocket (Verdecia et al., 2000; Lu et al., 2002b). Thus, if there is increased phosphorylation of Pin1
at its serine 16 residue, Pin1 activity is impaired. The phosphorylation of serine/threonine‐
42
proline motifs of Pin1 substrate proteins is required for Pin1 to bind its targeted protein
substrates and facilitate the cis to trans isomerization to regulate the respective cellular
mechanisms (Schutkowski et al., 1998; Lu et al., 1999a). The cis and trans orientations of proteins
can impact their cellular function based on their protein interactions. Brown and colleagues
discussed how most cellular kinases and phosphates are trans‐specific (Brown et al., 1999),
which suggest that if a cis protein conformation is needed to regulate a specific mechanism a
PPI such as Pin1 is essential to the cell. Some conformational consequences include alterations
in protein dephosphorylation, enzymatic activity, and protein interactions (Lu et al., 2002a).
Pin1 and AD
Pin1 was implicated in AD when Lu and colleagues discovered that Pin1 co‐purified
with paired helical filaments of tau in AD brains and soluble Pin1 levels were minimal (Lu et
al., 1999b). They discovered that Pin1 binds the serine/threonine‐proline motif of tau at the
threonine 231 residue to complex with the NFT pathology found in the AD brain. The authors
suggested that Pin1 may be trapped in the tangles of phosphorylated tau protein resulting in
the depletion of soluble Pin1 levels. The significance of these findings was that an essential cell
cycle regulator also appeared to dephosphorylate a hallmark AD marker, but may be ineffective
due to AD pathology. This group concluded that a depletion of Pin1 as seen in AD could induce
mitotic arrest and promote apoptotic cell death contributing to neurodegeneration, yet
restoration of Pin1 can promote microtubule assembly along with tau binding to microtubules
43
(Lu et al., 1999b). Thus, there is evidence that Pin1 may be able to act as a cellular defense
mechanism to combat tau pathology of AD.
To assess the role of Pin1 in an age‐dependant manner as it relates to the aging
associated with AD, a Pin1 knockout mouse model was utilized (Liou et al., 2003). Previous
studies have shown these Pin1 knockout mice to develop testicular atrophy, body weight loss,
mammary gland proliferative impairment, and retinal degeneration based on a significant
decrease in cyclin D1 indicative of cell cycle disruption (Fujimori et al., 1999; Liou et al., 2002).
Since many of the abnormalities found in the mice appeared to be impacted by age (Liou et al.,
2002), Pin1 knockout mice seemed suitable to investigate with respect to pathological
abnormalities found in AD. Therefore, Pin1 knockout mice were examined at 2‐3 months, 4‐8
months, and 9‐14 months of age to characterize their phenotype. As the Pin1 knockout mice
grew older they developed hunched posture and reduced mobility based on their ability to
remain on a hanging bar (Liou et al., 2003). A significant decrease of neuron‐specific nuclear
protein in the parietal cortex, autophagic vacuoles in neurons, and degenerating lysosomes of
older Pin1 knockout mice showed there was age‐dependent neuronal death.
Hyperphosphorylated tau and conformations of NFT were found in the older Pin1 knockout
mice (Liou et al., 2003). The authors suggest cross‐breeding this model with AD transgenic
models to recapitulate the pathology seen in humans as this was the first study to show
endogenous mouse tau to form NFT pathology (Liou et al., 2003).
The role of Pin1 in AD pathology was strengthened when Pin1 was shown to regulate
APP processing and Aβ production (Pastorino et al., 2006). Pin1 had never been shown before
44
to bind APP although previous studies found APP to have a serine/threonine‐proline directed
motif and increased phosphorylation of APP at threonine 668 during mitosis (Suzuki et al.,
1994; da Cruz e Silva and da Cruz e Silva, 2003; Lee et al., 2003). Therefore, Pastorino and
colleagues sought to determine the role of Pin1 in regulating APP phosphorylation. They
showed that Pin1 binds APP at the threonine 668 residue and co‐localizes with APP at the
plasma membrane (Pastorino et al., 2006). Pin1 was found to impact amyloid processing by
decreasing soluble APPα levels and increasing Aβ generation in Pin1 ‐/‐ breast cancer cells
(Pastorino et al., 2006). These data were confirmed by cross‐breeding Pin1 knockout mice with
Tg2576 mice (overexpress Swedish APP mutation; K670N/M671L). The collective findings of
Pin1 effects in AD pathways involving amyloid and tau implicate it as novel intracellular
regulator that should be targeted for therapeutic intervention. A schematic of how Pin1 exerts
its affect on amyloid and tau proteins confirmation to impact AD pathology is provided (Figure
1.4).
45
Figure 1.4: Pin1 function in Alzheimer’s disease. This schematic shows the role of Pin1 on amyloid and tau proteins in healthy neurons (A) and AD neurons (B). Pin1 catalyzes the cis to trans protein confirmation in healthy neurons to prevent pathogenesis. In AD, Pin1 is unable to facilitate these conformational changes resulting in the AD pathology characteristics of Aβ and NFT generation (Balastik et al., 2007).
46
Pin1 polymorphisms and AD
The Pin1 gene maps to human chromosome 19p13.2 (Lu et al., 1996; Campbell et al.,
1997). Thus, the discovery of a novel late‐onset AD locust on human chromosome 19p13.2
prompted an investigation of Pin1 variants regarding AD risk (Wijsman et al., 2004). Armed
with known evidence of Pin1 impacting proposed AD pathogenesis mechanisms (Liou et al.,
2003; Pastorino et al., 2006; Balastik et al., 2007); genetic variants in Pin1 emerged to be a
candidate gene for sporadic AD. For this reason, four genetic studies have examined Pin1
polymorphism since 2005 (Poli et al., 2005; Lambert et al., 2006; Nowotny et al., 2007; Segat et
al., 2007).
The first published study on Pin1 gene variants and AD risk was performed by Poli and
colleagues (Poli et al., 2005). Their subject population was an Italian group of AD patients (n
=120) and age‐matched controls (n =134). At the time there was no information available on Pin1
allele frequency in the population. This group identified single nucleotide polymorphisms in
exon 2 (c+99G>A) and exon 3 (c+370G>T) in the entire study population. There were four AD
patients and two controls that possessed the +99G>A polymorphism. The +370G>T
polymorphism was found in two AD patients and three controls. Thus, an association of Pin1
variants and AD was not found in this study. The authors state that the low number and
frequency of DNA sequence variations in Pin1 suggest the Pin1 gene is highly conserved among
humans (Poli et al., 2005). ApoE genotyping was performed as well, but due to the low
frequency of Pin1 variants, any interaction with ApoE was not observed. Since this was the first
47
study investigation an association between Pin1 and AD, these data were very meaningful,
especially in identifying single nucleotide polymorphisms.
A similar study was conducted by Segat and colleagues examining Pin1 polymorphisms
and AD risk (Segat et al., 2007). This study utilized an Italian population of patients as well,
which included AD patients (n =111) and age‐matched controls (n =73). Two new single
nucleotide polymorphisms were identified in the promoter region of Pin1 (‐842G/C and ‐
667C/T). A significant increase in the ‐842G/C Pin1 polymorphism was found in 30% of AD
patients compared to 12% in the control subjects (Segat et al., 2007) supporting an association
with AD risk. The ‐667C/T Pin1 polymorphism was found in 58.5% of AD patients and 48% of
controls, suggesting there was not an association with this polymorphism and AD (Segat et al.,
2007). This was the first study showing an association of a Pin1 polymorphism and AD risk by
identifying additional Pin1 polymorphisms in its promoter region. Furthermore, the c+99G/A
and c+370G/T were not found in many carriers in the study similar to the Poli study (Poli et al.,
2005).
A third investigation performed by Lambert and colleagues did not find an association
with Pin1 polymorphisms and AD risk (Lambert et al., 2006). There was an extremely large
group of French‐population based patients examined in this study that included 601 AD
patients and 655 controls. The Pin1 polymorphisms, ‐842G/C and ‐667T/C, expression was
determined among the study participants. The ‐842G/C polymorphism was found in 17% of AD
patients with respect to 20% in the control group suggesting that there was not association with
AD (Lambert et al., 2006). 51% of AD patients carried the ‐667T/C polymorphism compared to
48
53% in control subjects further supporting that Pin1 polymorphisms were not associated with
AD (Lambert et al., 2006). Furthermore, the group evaluated Pin1 polymorphisms with age of
AD onset and did not find an effect (data not shown). The methodology or data were not
provided to evaluate if there was an association with AD onset or not.
The most recent study evaluating Pin1 gene variants and AD risk was led by Nowotny
and colleagues (Nowotny et al., 2007). This group analyzed a total of six Pin1 polymorphisms in
750 AD patients and 658 controls in an American population of Caucasians from the WashU‐
ADRC study (Morris et al., 1997). There were no significant differences found in the allelic and
genotype frequencies of the six Pin1 polymorphisms in AD patients (Nowotny et al., 2007).
Moreover, no differences were found when study participants were stratified for ApoE4
expression (Nowotny et al., 2007). This group went on to perform a meta‐analysis of their data
and the two previous studies that examined Pin1 promoter polymorphisms (Lambert et al.,
2006; Segat et al., 2007), ‐842G/C and ‐667C/T, and did not find an association with AD risk.
They conclude by stating that rare Pin1 polymorphisms that increase Pin1 expression may exert
an effect on late‐onset AD development (Nowotny et al., 2007).
Similar to the HFE polymorphisms and AD epidemiological studies, the majority of
genetic analyses of Pin1 polymorphisms and AD risk have only focused on an association with
AD. Pin1 polymorphisms may be putative risk factors or genetic modifiers that impact age of
AD onset such as the H63D HFE variant. Only one out of four studies attempted to examine the
effect of Pin1 polymorphisms and the age of AD onset (Lambert et al., 2006). Oxidative stress
can impact Pin1 expression and function (Butterfield et al., 2006; Sultana et al., 2006; Kap et al.,
49
2007), thus consideration of Pin1 polymorphisms in the context of gene‐environment
interactions is warranted since many disease causing agents are unknown. Moreover, the
cellular effects of expressing Pin1 polymorphisms need to be elucidated with respect to Pin1
function. Unlike HFE, Pin1 has been widely studied and has multiple cellular functions that
may impact disease risk (Hamdane et al., 2002; Lim and Lu, 2005; Lu and Zhou, 2007; Wang et
al., 2007b). Thus, identifying Pin1 polymorphisms that have increased incidence in diseases
such as AD may impact disease onset, disease progression, and therapeutic outcomes.
Summary
The challenge of discovering the etiology of AD has proved to be an insurmountable
task. Cellular pathological hallmarks of AD have centered primarily on Aβ plaque deposits and
neurofibrillary tangles, which were initially discovered and described by Louis Alzheimer in
1906. It has been over a century and post‐mortem analysis of AD brains have not revealed any
other abundant protein markers, which strengthens the argument that amyloid and tau are
involved in AD development and progression. Research efforts have succeeded in identifying
familial cases of AD with genetic variants in the APP, PS1, and PS2 genes; however, only 5‐10%
of the AD patients possess such a mutation. Thus, approximately 90% of the AD cases remain
unresolved and are recognized as sporadic regarding their etiology of AD.
Since the large number of late‐onset or sporadic AD cases are poorly defined, it is
important to consider how gene variations and/or environmental factors can impact amyloid
and tau pathogenic pathways. The ApoE4 polymorphism is the strongest genetic risk factor
50
associated with late‐onset AD, but it is not a causative agent. A cellular environmental factor
such as iron can significantly impact many of the proposed AD pathogenic pathways if cellular
iron homeostasis is not achieved. Genetic variants such as HFE polymorphisms may alter the
intracellular levels of iron that can accelerate the disruption of cellular mechanisms leading to
neurodegeneration seen in AD and other neurological diseases. The central working hypothesis
of the thesis research in the following chapters is that the H63D HFE polymorphism promotes
AD pathogenic pathways primarily mediated through iron mismanagement. By investigating
the cellular contribution of a putative risk factor for AD onset and progression, a unique
opportunity to identify therapeutic targets and predict therapeutic outcomes may be achieved
to elucidate the disease etiology.
51
References
Adalbert R, Gilley J, Coleman MP (2007) Abeta, tau and ApoE4 in Alzheimerʹs disease: the
axonal connection. Trends Mol Med 13:135‐142.
Adams PC, Reboussin DM, Barton JC, McLaren CE, Eckfeldt JH, McLaren GD, Dawkins FW,
Acton RT, Harris EL, Gordeuk VR, Leiendecker‐Foster C, Speechley M, Snively BM,
Holup JL, Thomson E, Sholinsky P (2005) Hemochromatosis and iron‐overload
screening in a racially diverse population. N Engl J Med 352:1769‐1778.
Alizadeh BZ, Njajou OT, Millan MR, Hofman A, Breteler MM, van Duijn CM (2009) HFE
variants, APOE and Alzheimerʹs disease: findings from the population‐based Rotterdam
study. Neurobiol Aging 30:330‐332.
Alzheimer A, Stelzmann RA, Schnitzlein HN, Murtagh FR (1995) An English translation of
Alzheimerʹs 1907 paper, ʺUber eine eigenartige Erkankung der Hirnrindeʺ. Clin Anat
8:429‐431.
Andrews NC (1999a) Disorders of iron metabolism. N Engl J Med 341:1986‐1995.
Andrews NC (1999b) The iron transporter DMT1. Int J Biochem Cell Biol 31:991‐994.
Andrews NC, Fleming MD, Levy JE (1999) Molecular insights into mechanisms of iron
transport. Curr Opin Hematol 6:61‐64.
Arcos‐Burgos M, Muenke M (2002) Genetics of population isolates. Clin Genet 61:233‐247.
Arosio P, Levi S (2002) Ferritin, iron homeostasis, and oxidative damage. Free Radic Biol Med
33:457‐463.
52
Avila‐Gomez IC, Jimenez‐Del‐Rio M, Lopera‐Restrepo F, Velez‐Pardo C (2008) Association
between HFE 187 C>G (H63D) mutation and early‐onset familial Alzheimerʹs disease
PSEN‐1 839A>C (E280A) mutation. Ann Hematol 87:671‐673.
Balastik M, Lim J, Pastorino L, Lu KP (2007) Pin1 in Alzheimerʹs disease: multiple substrates,
one regulatory mechanism? Biochim Biophys Acta 1772:422‐429.
Ballatore C, Lee VM, Trojanowski JQ (2007) Tau‐mediated neurodegeneration in Alzheimerʹs
disease and related disorders. Nat Rev Neurosci 8:663‐672.
Bamberger ME, Landreth GE (2001) Microglial interaction with beta‐amyloid: implications for
the pathogenesis of Alzheimerʹs disease. Microsc Res Tech 54:59‐70.
Bamberger ME, Landreth GE (2002) Inflammation, apoptosis, and Alzheimerʹs disease.
Neuroscientist 8:276‐283.
Bamberger ME, Harris ME, McDonald DR, Husemann J, Landreth GE (2003) A cell surface
receptor complex for fibrillar beta‐amyloid mediates microglial activation. J Neurosci
23:2665‐2674.
Barnham KJ, Bush AI (2008) Metals in Alzheimerʹs and Parkinsonʹs diseases. Curr Opin Chem
Biol 12:222‐228.
Barnham KJ, McKinstry WJ, Multhaup G, Galatis D, Morton CJ, Curtain CC, Williamson NA,
White AR, Hinds MG, Norton RS, Beyreuther K, Masters CL, Parker MW, Cappai R
(2003) Structure of the Alzheimerʹs disease amyloid precursor protein copper binding
domain. A regulator of neuronal copper homeostasis. J Biol Chem 278:17401‐17407.
53
Bartzokis G, Tishler TA, Shin IS, Lu PH, Cummings JL (2004) Brain ferritin iron as a risk factor
for age at onset in neurodegenerative diseases. Ann N Y Acad Sci 1012:224‐236.
Bartzokis G, Beckson M, Hance DB, Marx P, Foster JA, Marder SR (1997) MR evaluation of age‐
related increase of brain iron in young adult and older normal males. Magn Reson
Imaging 15:29‐35.
Baumann K, Mandelkow EM, Biernat J, Piwnica‐Worms H, Mandelkow E (1993) Abnormal
Alzheimer‐like phosphorylation of tau‐protein by cyclin‐dependent kinases cdk2 and
cdk5. FEBS Lett 336:417‐424.
Berlin D, Chong G, Chertkow H, Bergman H, Phillips NA, Schipper HM (2004) Evaluation of
HFE (hemochromatosis) mutations as genetic modifiers in sporadic AD and MCI.
Neurobiol Aging 25:465‐474.
Beurel E, Jope RS (2006) The paradoxical pro‐ and anti‐apoptotic actions of GSK3 in the intrinsic
and extrinsic apoptosis signaling pathways. Prog Neurobiol 79:173‐189.
Blazquez L, De Juan D, Ruiz‐Martinez J, Emparanza JI, Saenz A, Otaegui D, Sistiaga A,
Martinez‐Lage P, Lamet I, Samaranch L, Buiza C, Etxeberria I, Arriola E, Cuadrado E,
Urdaneta E, Yanguas J, Lopez de Munain A (2007) Genes related to iron metabolism and
susceptibility to Alzheimerʹs disease in Basque population. Neurobiol Aging 28:1941‐
1943.
Bodovitz S, Falduto MT, Frail DE, Klein WL (1995) Iron levels modulate alpha‐secretase
cleavage of amyloid precursor protein. J Neurochem 64:307‐315.
54
Borie C, Gasparini F, Verpillat P, Bonnet AM, Agid Y, Hetet G, Brice A, Durr A, Grandchamp B
(2002) Association study between iron‐related genes polymorphisms and Parkinsonʹs
disease. J Neurol 249:801‐804.
Brown NR, Noble ME, Endicott JA, Johnson LN (1999) The structural basis for specificity of
substrate and recruitment peptides for cyclin‐dependent kinases. Nat Cell Biol 1:438‐443.
Buee L, Bussiere T, Buee‐Scherrer V, Delacourte A, Hof PR (2000) Tau protein isoforms,
phosphorylation and role in neurodegenerative disorders. Brain Res Brain Res Rev
33:95‐130.
Bush AI (2003) The metallobiology of Alzheimerʹs disease. Trends Neurosci 26:207‐214.
Butterfield DA, Poon HF, St Clair D, Keller JN, Pierce WM, Klein JB, Markesbery WR (2006)
Redox proteomics identification of oxidatively modified hippocampal proteins in mild
cognitive impairment: insights into the development of Alzheimerʹs disease. Neurobiol
Dis 22:223‐232.
Buxbaum JD, Liu KN, Luo Y, Slack JL, Stocking KL, Peschon JJ, Johnson RS, Castner BJ, Cerretti
DP, Black RA (1998) Evidence that tumor necrosis factor alpha converting enzyme is
involved in regulated alpha‐secretase cleavage of the Alzheimer amyloid protein
precursor. J Biol Chem 273:27765‐27767.
Camaschella C, Roetto A, De Gobbi M (2002) Genetic haemochromatosis: genes and mutations
associated with iron loading. Best Pract Res Clin Haematol 15:261‐276.
55
Campbell HD, Webb GC, Fountain S, Young IG (1997) The human PIN1 peptidyl‐prolyl
cis/trans isomerase gene maps to human chromosome 19p13 and the closely related
PIN1L gene to 1p31. Genomics 44:157‐162.
Candore G, Licastro F, Chiappelli M, Franceschi C, Lio D, Rita Balistreri C, Piazza G, Colonna‐
Romano G, Grimaldi LM, Caruso C (2003) Association between the HFE mutations and
unsuccessful ageing: a study in Alzheimerʹs disease patients from Northern Italy. Mech
Ageing Dev 124:525‐528.
Castellani R, Hirai K, Aliev G, Drew KL, Nunomura A, Takeda A, Cash AD, Obrenovich ME,
Perry G, Smith MA (2002) Role of mitochondrial dysfunction in Alzheimerʹs disease. J
Neurosci Res 70:357‐360.
Castellani RJ, Lee HG, Zhu X, Nunomura A, Perry G, Smith MA (2006) Neuropathology of
Alzheimer disease: pathognomonic but not pathogenic. Acta Neuropathol (Berl) 111:503‐
509.
Combarros O, Garcia‐Roman M, Fontalba A, Fernandez‐Luna JL, Llorca J, Infante J, Berciano J
(2003) Interaction of the H63D mutation in the hemochromatosis gene with the
apolipoprotein E epsilon 4 allele modulates age at onset of Alzheimerʹs disease. Dement
Geriatr Cogn Disord 15:151‐154.
Combs CK, Karlo JC, Kao SC, Landreth GE (2001) beta‐Amyloid stimulation of microglia and
monocytes results in TNFalpha‐dependent expression of inducible nitric oxide synthase
and neuronal apoptosis. J Neurosci 21:1179‐1188.
56
Connor JR, Menzies SL (1995) Cellular management of iron in the brain. J Neurol Sci 134
Suppl:33‐44.
Connor JR, Lee SY (2006) HFE mutations and Alzheimerʹs disease. J Alzheimers Dis 10:267‐276.
Connor JR, Menzies SL, St Martin SM, Mufson EJ (1992a) A histochemical study of iron,
transferrin, and ferritin in Alzheimerʹs diseased brains. J Neurosci Res 31:75‐83.
Connor JR, Snyder BS, Beard JL, Fine RE, Mufson EJ (1992b) Regional distribution of iron and
iron‐regulatory proteins in the brain in aging and Alzheimerʹs disease. J Neurosci Res
31:327‐335.
Connor JR, Milward EA, Moalem S, Sampietro M, Boyer P, Percy ME, Vergani C, Scott RJ,
Chorney M (2001) Is hemochromatosis a risk factor for Alzheimerʹs disease? J
Alzheimers Dis 3:471‐477.
da Cruz e Silva EF, da Cruz e Silva OA (2003) Protein phosphorylation and APP metabolism.
Neurochem Res 28:1553‐1561.
de Almeida SF, Carvalho IF, Cardoso CS, Cordeiro JV, Azevedo JE, Neefjes J, de Sousa M (2005)
HFE cross‐talks with the MHC class I antigen presentation pathway. Blood 106:971‐977.
De Felice FG, Wu D, Lambert MP, Fernandez SJ, Velasco PT, Lacor PN, Bigio EH, Jerecic J,
Acton PJ, Shughrue PJ, Chen‐Dodson E, Kinney GG, Klein WL (2008) Alzheimerʹs
disease‐type neuronal tau hyperphosphorylation induced by A beta oligomers.
Neurobiol Aging 29:1334‐1347.
De Sarno P, Li X, Jope RS (2002) Regulation of Akt and glycogen synthase kinase‐3 beta
phosphorylation by sodium valproate and lithium. Neuropharmacology 43:1158‐1164.
57
Dekker MC, Giesbergen PC, Njajou OT, van Swieten JC, Hofman A, Breteler MM, van Duijn
CM (2003) Mutations in the hemochromatosis gene (HFE), Parkinsonʹs disease and
parkinsonism. Neurosci Lett 348:117‐119.
Desagher S, Martinou JC (2000) Mitochondria as the central control point of apoptosis. Trends
Cell Biol 10:369‐377.
Dunn LL, Rahmanto YS, Richardson DR (2007) Iron uptake and metabolism in the new
millennium. Trends Cell Biol 17:93‐100.
Dupradeau F, Altenberg‐Greulich B, Warin R, Fuentes V, Monti J, Rochette J (2000) A 3‐
dimensional model building by homology of the HFE protein: molecular consequences
and application to antibody development. Biochim Biophys Acta 1481:213‐221.
Edbauer D, Winkler E, Regula JT, Pesold B, Steiner H, Haass C (2003) Reconstitution of gamma‐
secretase activity. Nat Cell Biol 5:486‐488.
Egana JT, Zambrano C, Nunez MT, Gonzalez‐Billault C, Maccioni RB (2003) Iron‐induced
oxidative stress modify tau phosphorylation patterns in hippocampal cell cultures.
Biometals 16:215‐223.
Ehrlich R, Lemonnier FA (2000) HFE‐‐a novel nonclassical class I molecule that is involved in
iron metabolism. Immunity 13:585‐588.
Eikelenboom P, van Gool WA (2004) Neuroinflammatory perspectives on the two faces of
Alzheimerʹs disease. J Neural Transm 111:281‐294.
58
Ellervik C, Tybjaerg‐Hansen A, Appleyard M, Sillesen H, Boysen G, Nordestgaard BG (2007)
Hereditary hemochromatosis genotypes and risk of ischemic stroke. Neurology 68:1025‐
1031.
Feder JN, Penny DM, Irrinki A, Lee VK, Lebron JA, Watson N, Tsuchihashi Z, Sigal E, Bjorkman
PJ, Schatzman RC (1998) The hemochromatosis gene product complexes with the
transferrin receptor and lowers its affinity for ligand binding. Proc Natl Acad Sci U S A
95:1472‐1477.
Feder JN, Tsuchihashi Z, Irrinki A, Lee VK, Mapa FA, Morikang E, Prass CE, Starnes SM, Wolff
RK, Parkkila S, Sly WS, Schatzman RC (1997) The hemochromatosis founder mutation in
HLA‐H disrupts beta2‐microglobulin interaction and cell surface expression. J Biol
Chem 272:14025‐14028.
Feder JN et al. (1996) A novel MHC class I‐like gene is mutated in patients with hereditary
haemochromatosis. Nat Genet 13:399‐408.
Fujimori F, Takahashi K, Uchida C, Uchida T (1999) Mice lacking Pin1 develop normally, but
are defective in entering cell cycle from G(0) arrest. Biochem Biophys Res Commun
265:658‐663.
Fukui H, Moraes CT (2008) The mitochondrial impairment, oxidative stress and
neurodegeneration connection: reality or just an attractive hypothesis? Trends Neurosci
31:251‐256.
59
Glenner GG, Wong CW (1984) Alzheimerʹs disease: initial report of the purification and
characterization of a novel cerebrovascular amyloid protein. Biochem Biophys Res
Commun 120:885‐890.
Goate A, Chartier‐Harlin MC, Mullan M, Brown J, Crawford F, Fidani L, Giuffra L, Haynes A,
Irving N, James L, et al. (1991) Segregation of a missense mutation in the amyloid
precursor protein gene with familial Alzheimerʹs disease. Nature 349:704‐706.
Goedert M, Spillantini MG, Jakes R, Crowther RA, Vanmechelen E, Probst A, Gotz J, Burki K,
Cohen P (1995) Molecular dissection of the paired helical filament. Neurobiol Aging
16:325‐334.
Golde TE, Estus S, Usiak M, Younkin LH, Younkin SG (1990) Expression of beta amyloid
protein precursor mRNAs: recognition of a novel alternatively spliced form and
quantitation in Alzheimerʹs disease using PCR. Neuron 4:253‐267.
Good PF, Perl DP, Bierer LM, Schmeidler J (1992) Selective accumulation of aluminum and iron
in the neurofibrillary tangles of Alzheimerʹs disease: a laser microprobe (LAMMA)
study. Ann Neurol 31:286‐292.
Goodall EF, Greenway MJ, van Marion I, Carroll CB, Hardiman O, Morrison KE (2005)
Association of the H63D polymorphism in the hemochromatosis gene with sporadic
ALS. Neurology 65:934‐937.
Greenberg SM, Koo EH, Selkoe DJ, Qiu WQ, Kosik KS (1994) Secreted beta‐amyloid precursor
protein stimulates mitogen‐activated protein kinase and enhances tau phosphorylation.
Proc Natl Acad Sci U S A 91:7104‐7108.
60
Grimes CA, Jope RS (2001) The multifaceted roles of glycogen synthase kinase 3beta in cellular
signaling. Prog Neurobiol 65:391‐426.
Grundke‐Iqbal I, Iqbal K, Quinlan M, Tung YC, Zaidi MS, Wisniewski HM (1986) Microtubule‐
associated protein tau. A component of Alzheimer paired helical filaments. J Biol Chem
261:6084‐6089.
Guerreiro RJ, Bras JM, Santana I, Januario C, Santiago B, Morgadinho AS, Ribeiro MH, Hardy J,
Singleton A, Oliveira C (2006) Association of HFE common mutations with Parkinsonʹs
disease, Alzheimerʹs disease and mild cognitive impairment in a Portuguese cohort.
BMC Neurol 6:24.
Hallgren B, Sourander P (1958) The effect of age on the non‐haemin iron in the human brain. J
Neurochem 3:41‐51.
Hamdane M, Smet C, Sambo AV, Leroy A, Wieruszeski JM, Delobel P, Maurage CA, Ghestem
A, Wintjens R, Begard S, Sergeant N, Delacourte A, Horvath D, Landrieu I, Lippens G,
Buee L (2002) Pin1: a therapeutic target in Alzheimer neurodegeneration. J Mol Neurosci
19:275‐287.
Hanger DP, Hughes K, Woodgett JR, Brion JP, Anderton BH (1992) Glycogen synthase kinase‐3
induces Alzheimerʹs disease‐like phosphorylation of tau: generation of paired helical
filament epitopes and neuronal localisation of the kinase. Neurosci Lett 147:58‐62.
Hanson EH, Imperatore G, Burke W (2001) HFE gene and hereditary hemochromatosis: a HuGE
review. Human Genome Epidemiology. Am J Epidemiol 154:193‐206.
Hardy J (2006) A hundred years of Alzheimerʹs disease research. Neuron 52:3‐13.
61
Hardy J (2009) The amyloid hypothesis for Alzheimerʹs disease: a critical reappraisal. J
Neurochem.
Hardy J, Allsop D (1991) Amyloid deposition as the central event in the aetiology of
Alzheimerʹs disease. Trends Pharmacol Sci 12:383‐388.
Hardy J, Selkoe DJ (2002) The amyloid hypothesis of Alzheimerʹs disease: progress and
problems on the road to therapeutics. Science 297:353‐356.
Hardy JA, Higgins GA (1992) Alzheimerʹs disease: the amyloid cascade hypothesis. Science
256:184‐185.
Hayes A, Thaker U, Iwatsubo T, Pickering‐Brown SM, Mann DM (2002) Pathological
relationships between microglial cell activity and tau and amyloid beta protein in
patients with Alzheimerʹs disease. Neurosci Lett 331:171‐174.
Hernandez F, Borrell J, Guaza C, Avila J, Lucas JJ (2002) Spatial learning deficit in transgenic
mice that conditionally over‐express GSK‐3beta in the brain but do not form tau
filaments. J Neurochem 83:1529‐1533.
Hooper C, Killick R, Lovestone S (2008) The GSK3 hypothesis of Alzheimerʹs disease. J
Neurochem 104:1433‐1439.
Hoshi M, Takashima A, Noguchi K, Murayama M, Sato M, Kondo S, Saitoh Y, Ishiguro K,
Hoshino T, Imahori K (1996) Regulation of mitochondrial pyruvate dehydrogenase
activity by tau protein kinase I/glycogen synthase kinase 3beta in brain. Proc Natl Acad
Sci U S A 93:2719‐2723.
62
Huang HC, Klein PS (2006) Multiple roles for glycogen synthase kinase‐3 as a drug target in
Alzheimerʹs disease. Curr Drug Targets 7:1389‐1397.
Iqbal K, Grundke‐Iqbal I (2006) Discoveries of tau, abnormally hyperphosphorylated tau and
others of neurofibrillary degeneration: a personal historical perspective. J Alzheimers
Dis 9:219‐242.
Ishiguro K, Ihara Y, Uchida T, Imahori K (1988) A novel tubulin‐dependent protein kinase
forming a paired helical filament epitope on tau. J Biochem 104:319‐321.
Ishiguro K, Shiratsuchi A, Sato S, Omori A, Arioka M, Kobayashi S, Uchida T, Imahori K (1993)
Glycogen synthase kinase 3 beta is identical to tau protein kinase I generating several
epitopes of paired helical filaments. FEBS Lett 325:167‐172.
Itarte E, Huang KP (1979) Purification and properties of cyclic AMP‐independent glycogen
synthase kinase 1 from rabbit skeletal muscle. J Biol Chem 254:4052‐4057.
Jope RS, Yuskaitis CJ, Beurel E (2007) Glycogen synthase kinase‐3 (GSK3): inflammation,
diseases, and therapeutics. Neurochem Res 32:577‐595.
Juhaszova M, Zorov DB, Kim SH, Pepe S, Fu Q, Fishbein KW, Ziman BD, Wang S, Ytrehus K,
Antos CL, Olson EN, Sollott SJ (2004) Glycogen synthase kinase‐3beta mediates
convergence of protection signaling to inhibit the mitochondrial permeability transition
pore. J Clin Invest 113:1535‐1549.
Kang J, Lemaire HG, Unterbeck A, Salbaum JM, Masters CL, Grzeschik KH, Multhaup G,
Beyreuther K, Muller‐Hill B (1987) The precursor of Alzheimerʹs disease amyloid A4
protein resembles a cell‐surface receptor. Nature 325:733‐736.
63
Kap YS, Hoozemans JJ, Bodewes AJ, Zwart R, Meijer OC, Baas F, Scheper W (2007) Pin1 levels
are downregulated during ER stress in human neuroblastoma cells. Neurogenetics 8:21‐
27.
Kesavapany S, Lau KF, Ackerley S, Banner SJ, Shemilt SJ, Cooper JD, Leigh PN, Shaw CE,
McLoughlin DM, Miller CC (2003) Identification of a novel, membrane‐associated
neuronal kinase, cyclin‐dependent kinase 5/p35‐regulated kinase. J Neurosci 23:4975‐
4983.
Kidd M (1963) Paired helical filaments in electron microscopy of Alzheimerʹs disease. Nature
197:192‐193.
King TD, Jope RS (2005) Inhibition of glycogen synthase kinase‐3 protects cells from intrinsic
but not extrinsic oxidative stress. Neuroreport 16:597‐601.
Kish SJ, Bergeron C, Rajput A, Dozic S, Mastrogiacomo F, Chang LJ, Wilson JM, DiStefano LM,
Nobrega JN (1992) Brain cytochrome oxidase in Alzheimerʹs disease. J Neurochem
59:776‐779.
Kluck RM, Bossy‐Wetzel E, Green DR, Newmeyer DD (1997) The release of cytochrome c from
mitochondria: a primary site for Bcl‐2 regulation of apoptosis. Science 275:1132‐1136.
Kreutzberg GW (1996) Microglia: a sensor for pathological events in the CNS. Trends Neurosci
19:312‐318.
Kruszewski M, Iwanenko T (2003) Labile iron pool correlates with iron content in the nucleus
and the formation of oxidative DNA damage in mouse lymphoma L5178Y cell lines.
Acta Biochim Pol 50:211‐215.
64
Lambert JC, Bensemain F, Chapuis J, Cottel D, Amouyel P (2006) Association study of the PIN1
gene with Alzheimerʹs disease. Neurosci Lett 402:259‐261.
Lawson DM, Treffry A, Artymiuk PJ, Harrison PM, Yewdall SJ, Luzzago A, Cesareni G, Levi S,
Arosio P (1989) Identification of the ferroxidase centre in ferritin. FEBS Lett 254:207‐210.
Le Gac G, Mura C, Ferec C (2001) Complete scanning of the hereditary hemochromatosis gene
(HFE) by use of denaturing HPLC. Clin Chem 47:1633‐1640.
Lebron JA, Bjorkman PJ (1999) The transferrin receptor binding site on HFE, the class I MHC‐
related protein mutated in hereditary hemochromatosis. J Mol Biol 289:1109‐1118.
Lebron JA, West AP, Jr., Bjorkman PJ (1999) The hemochromatosis protein HFE competes with
transferrin for binding to the transferrin receptor. J Mol Biol 294:239‐245.
Lebron JA, Bennett MJ, Vaughn DE, Chirino AJ, Snow PM, Mintier GA, Feder JN, Bjorkman PJ
(1998) Crystal structure of the hemochromatosis protein HFE and characterization of its
interaction with transferrin receptor. Cell 93:111‐123.
Lee HG, Castellani RJ, Zhu X, Perry G, Smith MA (2005) Amyloid‐beta in Alzheimerʹs disease:
the horse or the cart? Pathogenic or protective? Int J Exp Pathol 86:133‐138.
Lee MS, Kao SC, Lemere CA, Xia W, Tseng HC, Zhou Y, Neve R, Ahlijanian MK, Tsai LH (2003)
APP processing is regulated by cytoplasmic phosphorylation. J Cell Biol 163:83‐95.
Lee SY, Connor JR (2005) Regulation of Hfe by stress factors in BV‐2 cells. Neurobiol Aging
26:803‐812.
Lee VM, Trojanowski JQ (1999) Neurodegenerative tauopathies: human disease and transgenic
mouse models. Neuron 24:507‐510.
65
Lee VM, Goedert M, Trojanowski JQ (2001) Neurodegenerative tauopathies. Annu Rev
Neurosci 24:1121‐1159.
Levy JE, Montross LK, Cohen DE, Fleming MD, Andrews NC (1999) The C282Y mutation
causing hereditary hemochromatosis does not produce a null allele. Blood 94:9‐11.
Lew J, Huang QQ, Qi Z, Winkfein RJ, Aebersold R, Hunt T, Wang JH (1994) A brain‐specific
activator of cyclin‐dependent kinase 5. Nature 371:423‐426.
Li T, Paudel HK (2006) Glycogen synthase kinase 3beta phosphorylates Alzheimerʹs disease‐
specific Ser396 of microtubule‐associated protein tau by a sequential mechanism.
Biochemistry 45:3125‐3133.
Li T, Hawkes C, Qureshi HY, Kar S, Paudel HK (2006) Cyclin‐dependent protein kinase 5
primes microtubule‐associated protein tau site‐specifically for glycogen synthase kinase
3beta. Biochemistry 45:3134‐3145.
Lim J, Lu KP (2005) Pinning down phosphorylated tau and tauopathies. Biochim Biophys Acta
1739:311‐322.
Linseman DA, Butts BD, Precht TA, Phelps RA, Le SS, Laessig TA, Bouchard RJ, Florez‐
McClure ML, Heidenreich KA (2004) Glycogen synthase kinase‐3beta phosphorylates
Bax and promotes its mitochondrial localization during neuronal apoptosis. J Neurosci
24:9993‐10002.
Liou YC, Ryo A, Huang HK, Lu PJ, Bronson R, Fujimori F, Uchida T, Hunter T, Lu KP (2002)
Loss of Pin1 function in the mouse causes phenotypes resembling cyclin D1‐null
phenotypes. Proc Natl Acad Sci U S A 99:1335‐1340.
66
Liou YC, Sun A, Ryo A, Zhou XZ, Yu ZX, Huang HK, Uchida T, Bronson R, Bing G, Li X,
Hunter T, Lu KP (2003) Role of the prolyl isomerase Pin1 in protecting against age‐
dependent neurodegeneration. Nature 424:556‐561.
Lleo A, Blesa R, Angelopoulos C, Pastor‐Rubio P, Villa M, Oliva R, Bufill E (2002) Transferrin
C2 allele, haemochromatosis gene mutations, and risk for Alzheimerʹs disease. J Neurol
Neurosurg Psychiatry 72:820‐821.
Loeffler DA, Connor JR, Juneau PL, Snyder BS, Kanaley L, DeMaggio AJ, Nguyen H, Brickman
CM, LeWitt PA (1995) Transferrin and iron in normal, Alzheimerʹs disease, and
Parkinsonʹs disease brain regions. J Neurochem 65:710‐724.
Lovell MA, Markesbery WR (2007) Oxidative damage in mild cognitive impairment and early
Alzheimerʹs disease. J Neurosci Res 85:3036‐3040.
Lu KP, Zhou XZ (2007) The prolyl isomerase PIN1: a pivotal new twist in phosphorylation
signalling and disease. Nat Rev Mol Cell Biol 8:904‐916.
Lu KP, Hanes SD, Hunter T (1996) A human peptidyl‐prolyl isomerase essential for regulation
of mitosis. Nature 380:544‐547.
Lu KP, Liou YC, Zhou XZ (2002a) Pinning down proline‐directed phosphorylation signaling.
Trends Cell Biol 12:164‐172.
Lu PJ, Zhou XZ, Shen M, Lu KP (1999a) Function of WW domains as phosphoserine‐ or
phosphothreonine‐binding modules. Science 283:1325‐1328.
Lu PJ, Wulf G, Zhou XZ, Davies P, Lu KP (1999b) The prolyl isomerase Pin1 restores the
function of Alzheimer‐associated phosphorylated tau protein. Nature 399:784‐788.
67
Lu PJ, Zhou XZ, Liou YC, Noel JP, Lu KP (2002b) Critical role of WW domain phosphorylation
in regulating phosphoserine binding activity and Pin1 function. J Biol Chem 277:2381‐
2384.
Lucas JJ, Hernandez F, Gomez‐Ramos P, Moran MA, Hen R, Avila J (2001) Decreased nuclear
beta‐catenin, tau hyperphosphorylation and neurodegeneration in GSK‐3beta
conditional transgenic mice. Embo J 20:27‐39.
Maccioni RB, Rojo LE, Fernandez JA, Kuljis RO (2009) The role of neuroimmunomodulation in
Alzheimerʹs disease. Ann N Y Acad Sci 1153:240‐246.
Mahley RW (1988) Apolipoprotein E: cholesterol transport protein with expanding role in cell
biology. Science 240:622‐630.
Mansuy IM, Shenolikar S (2006) Protein serine/threonine phosphatases in neuronal plasticity
and disorders of learning and memory. Trends Neurosci 29:679‐686.
Markesbery WR, Kryscio RJ, Lovell MA, Morrow JD (2005) Lipid peroxidation is an early event
in the brain in amnestic mild cognitive impairment. Ann Neurol 58:730‐735.
Masters CL, Simms G, Weinman NA, Multhaup G, McDonald BL, Beyreuther K (1985) Amyloid
plaque core protein in Alzheimer disease and Down syndrome. Proc Natl Acad Sci U S
A 82:4245‐4249.
Mattson MP, Gleichmann M, Cheng A (2008) Mitochondria in neuroplasticity and neurological
disorders. Neuron 60:748‐766.
68
Mattson MP, Pedersen WA, Duan W, Culmsee C, Camandola S (1999) Cellular and molecular
mechanisms underlying perturbed energy metabolism and neuronal degeneration in
Alzheimerʹs and Parkinsonʹs diseases. Ann N Y Acad Sci 893:154‐175.
Mazanetz MP, Fischer PM (2007) Untangling tau hyperphosphorylation in drug design for
neurodegenerative diseases. Nat Rev Drug Discov 6:464‐479.
Meda L, Baron P, Scarlato G (2001) Glial activation in Alzheimerʹs disease: the role of Abeta and
its associated proteins. Neurobiol Aging 22:885‐893.
Merryweather‐Clarke AT, Pointon JJ, Jouanolle AM, Rochette J, Robson KJ (2000) Geography of
HFE C282Y and H63D mutations. Genet Test 4:183‐198.
Moalem S, Percy ME, Andrews DF, Kruck TP, Wong S, Dalton AJ, Mehta P, Fedor B, Warren
AC (2000) Are hereditary hemochromatosis mutations involved in Alzheimer disease?
Am J Med Genet 93:58‐66.
Morishima‐Kawashima M, Hasegawa M, Takio K, Suzuki M, Yoshida H, Titani K, Ihara Y
(1995) Proline‐directed and non‐proline‐directed phosphorylation of PHF‐tau. J Biol
Chem 270:823‐829.
Morris JC, Ernesto C, Schafer K, Coats M, Leon S, Sano M, Thal LJ, Woodbury P (1997) Clinical
dementia rating training and reliability in multicenter studies: the Alzheimerʹs Disease
Cooperative Study experience. Neurology 48:1508‐1510.
Multhaup G (1997) Amyloid precursor protein, copper and Alzheimerʹs disease. Biomed
Pharmacother 51:105‐111.
69
Murphy E (2004) Inhibit GSK‐3beta or thereʹs heartbreak dead ahead. J Clin Invest 113:1526‐
1528.
Mutisya EM, Bowling AC, Beal MF (1994) Cortical cytochrome oxidase activity is reduced in
Alzheimerʹs disease. J Neurochem 63:2179‐2184.
Nagele RG, Wegiel J, Venkataraman V, Imaki H, Wang KC (2004) Contribution of glial cells to
the development of amyloid plaques in Alzheimerʹs disease. Neurobiol Aging 25:663‐
674.
Nikolic M, Dudek H, Kwon YT, Ramos YF, Tsai LH (1996) The cdk5/p35 kinase is essential for
neurite outgrowth during neuronal differentiation. Genes Dev 10:816‐825.
Noble W, Olm V, Takata K, Casey E, Mary O, Meyerson J, Gaynor K, LaFrancois J, Wang L,
Kondo T, Davies P, Burns M, Veeranna, Nixon R, Dickson D, Matsuoka Y, Ahlijanian M,
Lau LF, Duff K (2003) Cdk5 is a key factor in tau aggregation and tangle formation in
vivo. Neuron 38:555‐565.
Nowotny P, Bertelsen S, Hinrichs AL, Kauwe JS, Mayo K, Jacquart S, Morris JC, Goate A (2007)
Association studies between common variants in prolyl isomerase Pin1 and the risk for
late‐onset Alzheimerʹs disease. Neurosci Lett 419:15‐17.
Pastorino L, Sun A, Lu PJ, Zhou XZ, Balastik M, Finn G, Wulf G, Lim J, Li SH, Li X, Xia W,
Nicholson LK, Lu KP (2006) The prolyl isomerase Pin1 regulates amyloid precursor
protein processing and amyloid‐beta production. Nature 440:528‐534.
Patzke H, Tsai LH (2002) Calpain‐mediated cleavage of the cyclin‐dependent kinase‐5 activator
p39 to p29. J Biol Chem 277:8054‐8060.
70
Pei JJ, Gong CX, An WL, Winblad B, Cowburn RF, Grundke‐Iqbal I, Iqbal K (2003) Okadaic‐
acid‐induced inhibition of protein phosphatase 2A produces activation of mitogen‐
activated protein kinases ERK1/2, MEK1/2, and p70 S6, similar to that in Alzheimerʹs
disease. Am J Pathol 163:845‐858.
Percy M et al. (2008) Involvement of ApoE E4 and H63D in sporadic Alzheimerʹs disease in a
folate‐supplemented Ontario population. J Alzheimers Dis 14:69‐84.
Petersen RC, Smith GE, Waring SC, Ivnik RJ, Tangalos EG, Kokmen E (1999) Mild cognitive
impairment: clinical characterization and outcome. Arch Neurol 56:303‐308.
Petersen RC, Smith GE, Ivnik RJ, Tangalos EG, Schaid DJ, Thibodeau SN, Kokmen E, Waring
SC, Kurland LT (1995) Apolipoprotein E status as a predictor of the development of
Alzheimerʹs disease in memory‐impaired individuals. JAMA 273:1274‐1278.
Petrat F, de Groot H, Rauen U (2000) Determination of the chelatable iron pool of single intact
cells by laser scanning microscopy. Arch Biochem Biophys 376:74‐81.
Pinero DJ, Hu J, Connor JR (2000) Alterations in the interaction between iron regulatory
proteins and their iron responsive element in normal and Alzheimerʹs diseased brains.
Cell Mol Biol (Noisy‐le‐grand) 46:761‐776.
Plattner F, Angelo M, Giese KP (2006) The roles of cyclin‐dependent kinase 5 and glycogen
synthase kinase 3 in tau hyperphosphorylation. J Biol Chem 281:25457‐25465.
Poli M, Gatta LB, Dominici R, Lovati C, Mariani C, Albertini A, Finazzi D (2005) DNA sequence
variations in the prolyl isomerase Pin1 gene and Alzheimerʹs disease. Neurosci Lett
389:66‐70.
71
Porto G, De Sousa M (2007) Iron overload and immunity. World J Gastroenterol 13:4707‐4715.
Pulliam JF, Jennings CD, Kryscio RJ, Davis DG, Wilson D, Montine TJ, Schmitt FA, Markesbery
WR (2003) Association of HFE mutations with neurodegeneration and oxidative stress in
Alzheimerʹs disease and correlation with APOE. Am J Med Genet B Neuropsychiatr
Genet 119B:48‐53.
Quintana C, Bellefqih S, Laval JY, Guerquin‐Kern JL, Wu TD, Avila J, Ferrer I, Arranz R, Patino
C (2006) Study of the localization of iron, ferritin, and hemosiderin in Alzheimerʹs
disease hippocampus by analytical microscopy at the subcellular level. J Struct Biol
153:42‐54.
Ramagopalan SV, Cukjati M, Cernilec M, DeLuca GC, Dyment DA, Degenhardt A, Sadovnick
AD, Serbec VC, Ebers GC, Duquette P (2008) Mutations in the hemochromatosis gene
and the clinical outcome of multiple sclerosis. J Neuroimmunol 203:104‐107.
Ramalingam TS, West AP, Jr., Lebron JA, Nangiana JS, Hogan TH, Enns CA, Bjorkman PJ (2000)
Binding to the transferrin receptor is required for endocytosis of HFE and regulation of
iron homeostasis. Nat Cell Biol 2:953‐957.
Restagno G, Lombardo F, Ghiglione P, Calvo A, Cocco E, Sbaiz L, Mutani R, Chio A (2007) HFE
H63D polymorphism is increased in patients with amyotrophic lateral sclerosis of Italian
origin. J Neurol Neurosurg Psychiatry 78:327.
Ristic S, Lovrecic L, Brajenovic‐Milic B, Starcevic‐Cizmarevic N, Jazbec SS, Sepcic J, Kapovic M,
Peterlin B (2005) Mutations in the hemochromatosis gene (HFE) and multiple sclerosis.
Neurosci Lett 383:301‐304.
72
Robson KJ, Lehmann DJ, Wimhurst VL, Livesey KJ, Combrinck M, Merryweather‐Clarke AT,
Warden DR, Smith AD (2004) Synergy between the C2 allele of transferrin and the
C282Y allele of the haemochromatosis gene (HFE) as risk factors for developing
Alzheimerʹs disease. J Med Genet 41:261‐265.
Rodionova E, Conzelmann M, Maraskovsky E, Hess M, Kirsch M, Giese T, Ho AD, Zoller M,
Dreger P, Luft T (2007) GSK‐3 mediates differentiation and activation of
proinflammatory dendritic cells. Blood 109:1584‐1592.
Rogers JT, Lahiri DK (2004) Metal and inflammatory targets for Alzheimerʹs disease. Curr Drug
Targets 5:535‐551.
Rogers JT, Randall JD, Cahill CM, Eder PS, Huang X, Gunshin H, Leiter L, McPhee J, Sarang SS,
Utsuki T, Greig NH, Lahiri DK, Tanzi RE, Bush AI, Giordano T, Gullans SR (2002) An
iron‐responsive element type II in the 5ʹ‐untranslated region of the Alzheimerʹs amyloid
precursor protein transcript. J Biol Chem 277:45518‐45528.
Rohrlich PS, Fazilleau N, Ginhoux F, Firat H, Michel F, Cochet M, Laham N, Roth MP, Pascolo
S, Nato F, Coppin H, Charneau P, Danos O, Acuto O, Ehrlich R, Kanellopoulos J,
Lemonnier FA (2005) Direct recognition by alphabeta cytolytic T cells of Hfe, a MHC
class Ib molecule without antigen‐presenting function. Proc Natl Acad Sci U S A
102:12855‐12860.
Rouault TA, Tong WH (2005) Iron‐sulphur cluster biogenesis and mitochondrial iron
homeostasis. Nat Rev Mol Cell Biol 6:345‐351.
73
Roy CN, Custodio AO, de Graaf J, Schneider S, Akpan I, Montross LK, Sanchez M, Gaudino A,
Hentze MW, Andrews NC, Muckenthaler MU (2004) An Hfe‐dependent pathway
mediates hyposideremia in response to lipopolysaccharide‐induced inflammation in
mice. Nat Genet 36:481‐485.
Rubio JP, Bahlo M, Tubridy N, Stankovich J, Burfoot R, Butzkueven H, Chapman C, Johnson L,
Marriott M, Mraz G, Tait B, Wilkinson C, Taylor B, Speed TP, Foote SJ, Kilpatrick TJ
(2004) Extended haplotype analysis in the HLA complex reveals an increased frequency
of the HFE‐C282Y mutation in individuals with multiple sclerosis. Hum Genet 114:573‐
580.
Salloway S, Correia S (2009) Alzheimer disease: time to improve its diagnosis and treatment.
Cleve Clin J Med 76:49‐58.
Salter‐Cid L, Peterson PA, Yang Y (2000) The major histocompatibility complex‐encoded HFE in
iron homeostasis and immune function. Immunol Res 22:43‐59.
Sampietro M, Caputo L, Casatta A, Meregalli M, Pellagatti A, Tagliabue J, Annoni G, Vergani C
(2001) The hemochromatosis gene affects the age of onset of sporadic Alzheimerʹs
disease. Neurobiol Aging 22:563‐568.
Sayre LM, Perry G, Harris PL, Liu Y, Schubert KA, Smith MA (2000) In situ oxidative catalysis
by neurofibrillary tangles and senile plaques in Alzheimerʹs disease: a central role for
bound transition metals. J Neurochem 74:270‐279.
74
Schutkowski M, Bernhardt A, Zhou XZ, Shen M, Reimer U, Rahfeld JU, Lu KP, Fischer G (1998)
Role of phosphorylation in determining the backbone dynamics of the serine/threonine‐
proline motif and Pin1 substrate recognition. Biochemistry 37:5566‐5575.
Segat L, Pontillo A, Annoni G, Trabattoni D, Vergani C, Clerici M, Arosio B, Crovella S (2007)
PIN1 promoter polymorphisms are associated with Alzheimerʹs disease. Neurobiol
Aging 28:69‐74.
Selkoe DJ (1989) Molecular pathology of amyloidogenic proteins and the role of vascular
amyloidosis in Alzheimerʹs disease. Neurobiol Aging 10:387‐395.
Selkoe DJ (1994) Normal and abnormal biology of the beta‐amyloid precursor protein. Annu
Rev Neurosci 17:489‐517.
Selkoe DJ (1996a) Amyloid beta‐protein and the genetics of Alzheimerʹs disease. J Biol Chem
271:18295‐18298.
Selkoe DJ (1996b) Cell biology of the beta‐amyloid precursor protein and the genetics of
Alzheimerʹs disease. Cold Spring Harb Symp Quant Biol 61:587‐596.
Selkoe DJ, Yamazaki T, Citron M, Podlisny MB, Koo EH, Teplow DB, Haass C (1996) The role of
APP processing and trafficking pathways in the formation of amyloid beta‐protein. Ann
N Y Acad Sci 777:57‐64.
Sengupta A, Wu Q, Grundke‐Iqbal I, Iqbal K, Singh TJ (1997) Potentiation of GSK‐3‐catalyzed
Alzheimer‐like phosphorylation of human tau by cdk5. Mol Cell Biochem 167:99‐105.
Shah RS, Lee HG, Xiongwei Z, Perry G, Smith MA, Castellani RJ (2008) Current approaches in
the treatment of Alzheimerʹs disease. Biomed Pharmacother 62:199‐207.
75
Shin RW, Kruck TP, Murayama H, Kitamoto T (2003) A novel trivalent cation chelator Feralex
dissociates binding of aluminum and iron associated with hyperphosphorylated tau of
Alzheimerʹs disease. Brain Res 961:139‐146.
Smith MA, Harris PL, Sayre LM, Perry G (1997) Iron accumulation in Alzheimer disease is a
source of redox‐generated free radicals. Proc Natl Acad Sci U S A 94:9866‐9868.
Streit WJ (2004) Microglia and Alzheimerʹs disease pathogenesis. J Neurosci Res 77:1‐8.
Streit WJ, Mrak RE, Griffin WS (2004) Microglia and neuroinflammation: a pathological
perspective. J Neuroinflammation 1:14.
Sullivan PG, Brown MR (2005) Mitochondrial aging and dysfunction in Alzheimerʹs disease.
Prog Neuropsychopharmacol Biol Psychiatry 29:407‐410.
Sultana R, Boyd‐Kimball D, Poon HF, Cai J, Pierce WM, Klein JB, Markesbery WR, Zhou XZ, Lu
KP, Butterfield DA (2006) Oxidative modification and down‐regulation of Pin1 in
Alzheimerʹs disease hippocampus: A redox proteomics analysis. Neurobiol Aging
27:918‐925.
Sutedja NA, Sinke RJ, Van Vught PW, Van der Linden MW, Wokke JH, Van Duijn CM, Njajou
OT, Van der Schouw YT, Veldink JH, Van den Berg LH (2007) The association between
H63D mutations in HFE and amyotrophic lateral sclerosis in a Dutch population. Arch
Neurol 64:63‐67.
Suzuki T, Oishi M, Marshak DR, Czernik AJ, Nairn AC, Greengard P (1994) Cell cycle‐
dependent regulation of the phosphorylation and metabolism of the Alzheimer amyloid
precursor protein. Embo J 13:1114‐1122.
76
Tahara K, Kim HD, Jin JJ, Maxwell JA, Li L, Fukuchi K (2006) Role of toll‐like receptor signalling
in Abeta uptake and clearance. Brain 129:3006‐3019.
Takahashi K, Uchida C, Shin RW, Shimazaki K, Uchida T (2008) Prolyl isomerase, Pin1: new
findings of post‐translational modifications and physiological substrates in cancer,
asthma and Alzheimerʹs disease. Cell Mol Life Sci 65:359‐375.
Tanaka T, Iqbal K, Trenkner E, Liu DJ, Grundke‐Iqbal I (1995) Abnormally phosphorylated tau
in SY5Y human neuroblastoma cells. FEBS Lett 360:5‐9.
Tanaka T, Zhong J, Iqbal K, Trenkner E, Grundke‐Iqbal I (1998) The regulation of
phosphorylation of tau in SY5Y neuroblastoma cells: the role of protein phosphatases.
FEBS Lett 426:248‐254.
Tang D, Wang JH (1996) Cyclin‐dependent kinase 5 (Cdk5) and neuron‐specific Cdk5 activators.
Prog Cell Cycle Res 2:205‐216.
Tanzi RE, Bertram L (2005) Twenty years of the Alzheimerʹs disease amyloid hypothesis: a
genetic perspective. Cell 120:545‐555.
Terman A, Brunk UT (2006) Oxidative stress, accumulation of biological ʹgarbageʹ, and aging.
Antioxid Redox Signal 8:197‐204.
Thinakaran G (1999) The role of presenilins in Alzheimerʹs disease. J Clin Invest 104:1321‐1327.
Thompson KJ, Shoham S, Connor JR (2001) Iron and neurodegenerative disorders. Brain Res
Bull 55:155‐164.
Vassar R et al. (1999) Beta‐secretase cleavage of Alzheimerʹs amyloid precursor protein by the
transmembrane aspartic protease BACE. Science 286:735‐741.
77
Verdecia MA, Bowman ME, Lu KP, Hunter T, Noel JP (2000) Structural basis for phosphoserine‐
proline recognition by group IV WW domains. Nat Struct Biol 7:639‐643.
Waheed A, Parkkila S, Saarnio J, Fleming RE, Zhou XY, Tomatsu S, Britton RS, Bacon BR, Sly
WS (1999) Association of HFE protein with transferrin receptor in crypt enterocytes of
human duodenum. Proc Natl Acad Sci U S A 96:1579‐1584.
Waheed A, Grubb JH, Zhou XY, Tomatsu S, Fleming RE, Costaldi ME, Britton RS, Bacon BR, Sly
WS (2002) Regulation of transferrin‐mediated iron uptake by HFE, the protein defective
in hereditary hemochromatosis. Proc Natl Acad Sci U S A 99:3117‐3122.
Wallace DC (1999) Mitochondrial diseases in man and mouse. Science 283:1482‐1488.
Wang JZ, Grundke‐Iqbal I, Iqbal K (2007a) Kinases and phosphatases and tau sites involved in
Alzheimer neurofibrillary degeneration. Eur J Neurosci 25:59‐68.
Wang S, Simon BP, Bennett DA, Schneider JA, Malter JS, Wang DS (2007b) The significance of
Pin1 in the development of Alzheimerʹs disease. J Alzheimers Dis 11:13‐23.
Wang XS, Lee S, Simmons Z, Boyer P, Scott K, Liu W, Connor J (2004) Increased incidence of the
Hfe mutation in amyotrophic lateral sclerosis and related cellular consequences. J
Neurol Sci 227:27‐33.
Weisgraber KH (1994) Apolipoprotein E: structure‐function relationships. Adv Protein Chem
45:249‐302.
Weisgraber KH, Roses AD, Strittmatter WJ (1994) The role of apolipoprotein E in the nervous
system. Curr Opin Lipidol 5:110‐116.
78
Wijsman EM, Daw EW, Yu CE, Payami H, Steinbart EJ, Nochlin D, Conlon EM, Bird TD,
Schellenberg GD (2004) Evidence for a novel late‐onset Alzheimer disease locus on
chromosome 19p13.2. Am J Hum Genet 75:398‐409.
Wisniewski HM, Wegiel J (1991) Spatial relationships between astrocytes and classical plaque
components. Neurobiol Aging 12:593‐600.
Woodgett JR, Ohashi PS (2005) GSK3: an in‐Toll‐erant protein kinase? Nat Immunol 6:751‐752.
Wulf G, Finn G, Suizu F, Lu KP (2005) Phosphorylation‐specific prolyl isomerization: is there an
underlying theme? Nat Cell Biol 7:435‐441.
Yamamoto A, Shin RW, Hasegawa K, Naiki H, Sato H, Yoshimasu F, Kitamoto T (2002) Iron
(III) induces aggregation of hyperphosphorylated tau and its reduction to iron (II)
reverses the aggregation: implications in the formation of neurofibrillary tangles of
Alzheimerʹs disease. J Neurochem 82:1137‐1147.
Yen AA, Simpson EP, Henkel JS, Beers DR, Appel SH (2004) HFE mutations are not strongly
associated with sporadic ALS. Neurology 62:1611‐1612.
Zecca L, Youdim MB, Riederer P, Connor JR, Crichton RR (2004) Iron, brain ageing and
neurodegenerative disorders. Nat Rev Neurosci 5:863‐873.
Zhu X, Su B, Wang X, Smith MA, Perry G (2007) Causes of oxidative stress in Alzheimer
disease. Cell Mol Life Sci 64:2202‐2210.
Zou H, Li Y, Liu X, Wang X (1999) An APAF‐1.cytochrome c multimeric complex is a functional
apoptosome that activates procaspase‐9. J Biol Chem 274:11549‐11556.
Chapter 2
The Influence of HFE Polymorphisms on Amyloid Regulation in
Human Neuroblastoma SH‐SY5Y cells
Abstract
Elevated levels of cellular iron are routinely seen in Alzheimer’s disease (AD)
brains. HFE is the most common genetic polymorphism in Caucasians and its protein is
reported to regulate cellular iron uptake. Several studies have shown an association
between HFE polymorphisms and AD where some studies suggest that HFE
polymorphisms decrease the age of onset of AD. The H63D HFE polymorphism is
associated with altered iron homeostasis. HFE has been detected in cells associated with
amyloid plaques, one of the hallmarks of AD. Thus, we evaluated the role of HFE
polymorphisms in regulating amyloid cellular mechanisms. We created a stably
transfected human neuroblastoma SH‐SY5Y cell model expressing HFE polymorphisms
to investigate their effects on AD amyloid pathways. The SH‐SY5Y cell line does not
express detectable levels of HFE (Wang et al., 2004; Lee et al., 2006). Upon evaluating
APP synthesis, we found total APP levels to be increased in cells expressing WT and
H63D HFE polymorphisms compared to the vector transfection control, but APP levels
did not differ between cell lines expressing WT and H63D polymorphisms. APP is
processed by secretase complexes and we determined the effects of HFE polymorphisms
80
on secretase activity. There were no differences in α‐, β‐, or γ‐secretase activity among
the cell lines. Aβ production was indirectly determined by measuring APP C‐terminal
fragments and no differences were found. Our findings provide evidence that HFE itself
may alter APP levels regardless of the alleles expressed and we conclude that HFE
polymorphisms do not impact APP processing.
Introduction
HFE has generated interest as a genetic modifier of neurodegenerative disease
risk because of the iron mismanagement associated with neurological diseases (Bush,
2003; Huang et al., 2004; Zecca et al., 2004). In Alzheimer disease (AD) cases, genetic
association studies have supported or not supported an increased risk of AD onset with
HFE polymorphisms (Connor and Lee, 2006). HFE is a major histocompatibility class 1‐
like protein that has been shown to regulate intracellular iron uptake by complexing
with transferrin receptors (Feder et al., 1998). It is known that brain iron levels increase
throughout aging and there is increased brain iron in AD patients (Connor et al., 1992b;
Connor and Menzies, 1995; Smith et al., 1997). Therefore, HFE polymorphisms could
impact AD risk based on alterations in brain iron metabolism, but their impact on AD
pathogenic pathways is unknown.
The underlying cause of AD is not known, but the amyloid hypothesis has been
put forth to explain the pathological hallmarks associated with the disease (Hardy and
Higgins, 1992; Mattson, 1997). Amyloid precursor protein (APP) processing occurs via
81
non‐amyloid producing and amyloid producing secretase activity. The non‐amyloid α‐
secretase pathway involves “a disintegrin and metalloproteinase” (ADAM) that cleaves
within the Aβ sequence of APP and the amyloid producing β‐secretase pathway utilizes
the “beta amyloid converting enzyme” (BACE) (Hardy and Higgins, 1992; Mattson,
1997; Vassar et al., 1999; Nixon, 2005). The gamma secretase complex cleaves the
remaining APP peptide in either pathway to produce soluble APPα and the APP C‐
terminal fragment (CTF) C83 in the non‐amyloidogenic pathway or the Aβ peptide and
APP CTF C99 fragment in the amyloidogenic pathway (Grbovic et al., 2003; Zhang et al.,
2006).
A functional iron responsive element (IRE) was discovered in the 5’untranslated
region of APP, thereby solidifying it as a metalloprotein tightly regulated by
intracellular iron levels affecting its synthesis (Rogers et al., 2002). Cellular iron been
shown to regulate APP processing through its effects on α‐secretase activity at the cell
surface (Bodovitz et al., 1995). Increased activity of BACE 1 correlated with oxidative
stress based on the products of lipid peroxidation in sporadic AD cases using autopsied
brain tissue (Borghi et al., 2006) indicating that iron‐mediated oxidative stress may
impact Aβ generation. The work of our laboratory and others revealing iron deposits in
senile plaques and neurofibrillary tangles throughout the brain further implicate a
connection of iron dyshomeostasis and AD hallmark characteristics (Connor et al.,
1992a; Sayre et al., 2000; Connor et al., 2001; Quintana et al., 2006). Iron has also been
shown to be a key molecule in the aggregation of Aβ impacting subsequent plaque
82
formation (Multhaup, 1997; Barnham et al., 2003; Liu et al., 2006). The expression of the
H63D polymorphism may underlie these cellular iron and oxidative stress effects on
amyloid proteins.
We have developed a stably transfected HFE neuroblastoma SH‐SY5Y cell line to
test the hypothesis that the H63D polymorphism will increase APP synthesis and Aβ
production. There are numerous mechanisms that have shown that increased cellular
iron levels to influence amyloid regulation. The H63D polymorphism has been shown to
result in increased cellular iron levels and oxidative stress in our cell line (Wang et al.,
2004; Lee et al., 2006) and in humans (Pulliam et al., 2003). Therefore, we investigated
the role of HFE polymorphisms on cellular amyloid regulation.
Materials and methods
Reagents ‐ Human neuroblastoma SH‐SY5Y cell lines were obtained from American Type
Culture Collection (Manassas, VA, USA). Cell culture reagents including DMEM/F12,
DMEM, pen/strep/glutamine and Geneticin were purchased from Invitrogen (Carlsbad,
CA, USA). Fetal bovine serum was purchased from Gemini Bio‐Products (West
Sacramento, CA, USA). DC protein assay was obtained from Bio‐Rad (Hercules, CA,
USA). A polyclonal rabbit anti‐β‐amyloid precursor protein was purchased from Zymed
Laboratories (Carlsbad, CA, USA).
83
Cell culture ‐ Human neuroblastoma SH‐SY5Y cells were stably transfected to express
wild‐type and H63D HFE forms as previously reported along with a vector alone control
(Wang et al., 2004). We have previously reported that these cells were chosen because
endogenous expression of HFE in these cells could not be detected. The transfected cells
were maintained in DMEM/F12 media supplemented with 10% FBS, 1% antibiotics (pen‐
strep‐glutamine), 1x nonessential amino acids, and 1.8g/L sodium bicarbonate. Cells
were differentiated with 10 μM all‐trans retinoic acid (Sigma‐Aldrich) over six days
(Haque et al., 1999). To evaluate cellular iron effects, cells were treated with ferric
ammonium citrate (FAC) or desferrioxamine (DFO) over 48 hours (Lee et al., 2006).
Western blot ‐ Cells lysates were obtained as described above. Twenty‐five μg total
protein was equally separated by electrophoresis in a 4‐20% 12‐well Criterion gel (Bio‐
Rad, Hercules, CA). Protein was then transferred to a nitrocellulose membrane and
blocked for 1 hr at room temperature in TBS‐T with 5% nonfat milk or 1.5% BSA
(phosphorylated protein detection). Membranes were probed with primary antibodies in
TBS‐T with 5% nonfat milk overnight at 4°C. The membranes were incubated
respectively with rabbit polyclonal anti‐β‐amyloid precursor protein (1:500) and β‐actin
(1:2500). HRP‐conjugated secondary antibodies were added in 5% nonfat milk for 1 hr at
room temperature. Protein signals were obtained by chemiluminescence and visualized
by CCD camera. All western blot experiments were repeated at least twice with a
minimum of four different cultures per genotype per experiment, resulting in a total of
84
eight samples for analysis. The bands on the western blot were quantified by
densitometry using Fuji MultiGauge analysis software.
Enzyme linked immunoabsorbant assay (ELISA) ‐ Cells were lysed with RIPA buffer
supplemented with 1% Triton X‐100 and protease inhibitor cocktail (Sigma Aldrich, St.
Louis, MO). Phosphatase inhibitor (Sigma Aldrich, St. Louis, MO) was included in cell
lysis buffer for phosphorylation protein detection. Cells extracts were spun at 8,000 × g
for 10 min. Total protein levels were determined by Bio‐Rad DC protein assay. A
monoclonal antibody specific for total APP was coated onto the wells of the microtiter
strips provided (Invitrogen (BioSource)). Standards of known total APP protein were
processed to achieve a standard curve to determine the specific amount of total APP
protein in the unknown HFE cell samples. The ELISA assay plate was read at 450nm.
The ELISA experiment was performed using samples in triplicate per genotype and/ or
per treatment at two dilution concentrations along with the known standards for the
specific proteins, resulting in a total of six samples for analysis.
Secretase Activity Assays ‐ The enzymatic activity of the ‐, β‐, and γ‐secretases in the cell
lysates were detected spectrofluorometrically using secretase activity kits (R&D
Systems). The protein sample (50 μL, 200 μg total protein) is added to a secretase specific
( ‐, β‐, or γ‐) APP peptide conjugated to the reporter molecules EDANS and DABCYL;
all samples were run in triplicate. In the un‐cleaved form, the fluorescent emissions from
85
EDANS are quenched by the physical proximity of the DABCYL moiety. Cleavage of the
peptide by the secretase physically separates the EDANS and DABCYL, allowing for the
release of the fluorescent signal. The level of the secretase enzymatic activity in the
sample is proportional to the fluorometric reaction. Excitation (345 nm) and emission
(495 nm) filters were used to determine the secretase activity using a fluorescent
microplate reader (Spectra Max Gemini, Molecular Devices).
Statistical analysis ‐ The data were analyzed by one‐way ANOVA. Differences among the
means were considered statistically significant when the p value was <0.05. If overall p
<0.05, Tukey’s Multiple Comparison post hoc analysis was performed. Data are
presented as the mean ± S.E. GraphPad Prism software (version 4.0) was utilized to
perform the statistical analysis.
Results
We measured total APP levels in human neuroblastoma SH‐SY5Y cells stably
transfected with HFE variants to evaluate the role of HFE in APP production. APP levels
were not different in cells expressing wild type or H63D HFE compared to each other
(Figure 2.1). The absence of HFE in the vector transfection control cells resulted in
approximately a 50% reduction in APP (p<0.01) compared to wild type HFE cells (Figure
1). Also, there was no difference between vector and non‐transfected cells (data not
86
shown). These data suggest that HFE protein expression may play a role in up‐
regulating APP by another mechanism other than cellular iron regulation.
APP conjugated reporter molecules for the specific (α‐, β‐, and γ‐) secretase
complexes were used to determine secretase activity based on the release of a
fluorescence signal when cleavage occurs. Upon performing this experiment, there were
no changes in any of the amyloid secretase activities among the vector, wild type HFE,
or the H63D HFE variant cells (Figure 2.2).
We hypothesized that the cellular iron status would impact α‐secretase activity
based on other studies showing that iron influences α‐secretase activity (Bodovitz et al.,
1995). To evaluate if cellular iron could impact α‐secretase activity, we treated the wild
type HFE cells with iron. A 10 μM dose of iron increased (p<0.001) α‐secretase activity
and 30 μM iron increased (p<0.01) α‐secretase activity as well (Figure 2.3A). Since an
effect was seen, we treated the H63D expressing cells with desferrioxamine (DFO) to see
if α‐secretase activity could be decreased by removing iron. Removing iron did not alter
α‐secretase activity (Figure 2.3A). To determine if cellular iron levels could influence β‐
and γ‐secretase activity, we performed similar experiments since there is not published
literature showing that iron does or does not impact their cellular activity. There were
no changes in β‐secretase activity with either iron exposure or removal of iron (Figure
3B). γ‐secretase activity also was not affected by iron treatment or iron chelation (Figure
2.3C).
87
An additional genetic or environmental factor that can increase cellular iron
intake in cells already expressing the H63D allele may result in additional cellular
consequences based on the cellular stress that may occur. Increased oxidative stress has
been shown to alter β‐ and γ‐secretase activity resulting in increased production of Aβ
peptides (Vassar et al., 1999; Zhang et al., 2006). The H63D HFE cells possess a higher
baseline of oxidative stress with respect to cells expressing WT HFE based on alterations
in mitochondrial membrane potential and lipid peroxidation (Lee et al., 2006). To
determine if β‐ and γ‐secretase activity could be impacted by increased cell stress, we
treated the H63D cells with increasing amounts of iron. Increasing amounts of iron did
not alter the activity of either β‐ or γ‐secretase (Figure 2.4). Alpha‐secretase activity was
measured as well and did not result in a change (Figure 2.4).
Despite HFE polymorphism expression not impacting any of the amyloid
secretase activities (Figure 2.2), total APP levels were increased in cells expressing HFE
protein with respect to the vector control (Figure 2.1). With the increase in APP
expression in WT and H63D HFE cells, we hypothesized that Aβ production would
increase due to more APP availability for cellular processing. To test this hypothesis, we
measured APP C‐terminal fragments as an indirect measure of Aβ generation. C83 is the
fragment remaining after α‐secretase cleavage and the C99 fragment is indicative of β‐
secretase processing of APP (Hardy and Higgins, 1992). The data show that there is not a
statistically significant difference of APP C83 expression among cell lines (Figure 2.5).
APP C99 levels were weakly detected by the western blot method and could not be
88
quantified. In fact, numerous reports have shown that endogenous Aβ production in the
human SH‐SY5Y neuroblastoma cell lines is minimal without transfecting a familial AD
mutation into the cells (Peraus et al., 1997; Takeda et al., 2004).
Discussion
Aging is primarily associated with the onset of AD, but is not required as familial
cases of AD impact younger individuals and there are older people that do not succumb
to the dementia. Thus, there must be additional genetic and/or environmental factors
that can influence sporadic AD onset and progression. Biometals have gained
considerable interest in neurodegenerative disease because they are required to perform
a vast array of cellular functions and impact key pathological proteins (Zecca et al.,
2004). Brain iron levels are known to increase throughout aging (Connor and Menzies,
1995; Bartzokis et al., 2004) and genes that influence cellular iron homeostasis could be
an additional contributor to AD pathology. HFE is an iron regulatory gene that has been
examined in epidemiological studies to determine their association with AD (Connor
and Lee, 2006). Yet, the cellular effects of HFE variants on neuronal cell processes that
could affect AD are not known.
Amyloid homeostasis is thought to be disrupted in AD due to the excessive
accumulation of Aβ plaques found throughout the brain cortices (Masters et al., 1985).
APP yields Aβ peptides that are available to aggregate and generate the hallmark senile
plaques (Hardy and Higgins, 1992; Mattson, 1997). Iron has been shown to directly
89
regulate the production of APP through a novel 5’ IRE in APP mRNA (Rogers et al.,
2002); hence we hypothesized that the HFE variant, H63D, would increase APP
synthesis. Our hypothesis was not supported as there was not a difference in total APP
expression between cells expressing the wild type or H63D allele. Given the IRP/IRE
mechanism requirement for regulating the translation of target proteins like APP, we
considered this. A previous study performed by our group showed that total and active
IRP/IRE binding was not significantly different between the WT and H63D cells (Lee et
al., 2006) suggesting that the difference in cellular iron levels did not impact IRP/IRE
mechanism supporting our findings. It would be valuable to determine APP levels in
another model system expressing different HFE variants to better evaluate the effect of
HFE on APP production.
Furthermore, it is important to note that APP levels increased in WT and H63D
HFE expressing cells compared to the vector (transfection control) cell line that does not
express HFE. The finding that the presence of HFE increased APP is intriguing. This
finding suggests an unknown signaling mechanism that may directly link HFE to APP at
the cellular membrane regardless of cellular iron status. HFE has been suggested to play
a role in innate immunity due to the homology it shares with other class I major
histocompatibility proteins (Rohrlich et al., 2005). A histological analysis of post‐mortem
AD brains revealed pronounced expression of HFE throughout the brain vasculature
and surrounding the hallmark pathological markers Aβ and NFTs (Connor et al., 2001).
These data support the notion that HFE may indeed play a role in immune responses
90
since HFE is not highly expressed in healthy brains. Also, a study in a mouse BV‐2
microglia cell line showed that HFE expression increased with Aβ peptide exposure (Lee
and Connor, 2005) suggesting that HFE protein responds to cellular stress. APP
synthesis has been shown to be impacted by cellular stress as well (Huang et al., 2004);
including the interleukin‐1 acute box in its mRNA (Rogers et al., 1999; Rogers et al.,
2002).
The cellular activity of all the secretases in all of the pathways were determined
in cells expressing the HFE variants. There were no changes in secretase activity for any
of the secretases involved in APP processing. This was surprising given the increased
cellular iron and oxidative stress associated with expressing the H63D variant.
Iron has been shown to impact α‐secretase activity (Bodovitz et al., 1995) and it
would by reasonable to consider that HFE variants should have altered α‐secretase
activity. To further examine the cellular effects of iron on α‐secretase activity, we treated
the WT cells with increasing amounts of iron. Indeed, iron treatments elevated α‐
secretase activity. However, the removal of iron by chelation in cells expressing H63D
did not have an effect on α‐secretase activity. Subsequent experiments treating WT HFE
cells with iron and H63D cells with DFO did not affect either β‐ or γ‐secretase activity.
Moreover, the idea that an environmental factor like diet could impact gene expression
prompted us to treat H63D variant cells with increasing levels of iron. Increasing iron
exposure in cells expressing the H63D polymorphism did not affect α‐, β‐, or γ‐secretase
activity. We conclude that HFE variants do not impact amyloid secretase activity in our
91
cellular model and that cellular iron levels may influence α‐secretase activity based on a
specific cellular iron concentration.
Even though amyloid secretase activity was not altered by the expression of HFE
variants, important knowledge was gained, especially the fact that cellular iron does not
directly impact their function. The increase in APP with wild type and H63D HFE
expression suggests there may be more APP available to be processed to generate in an
in increase in downstream proteins like Aβ. Therefore, we measured the C‐terminal
fragments to determine the impact of secretase activity on APP. C83, the fragment
remaining after α‐secretase cleavage, was not changed in vector or H63D cells compared
to cells expressing the WT HFE variant. C99 expression levels were undetectable by
immunoblotting, which would have indicated the product of β‐secretase cleavage. The
important caveat regarding this data is that an increase in total APP does not necessarily
result in increased levels of soluble APPα or Aβ generation.
The most significant finding from this study is that APP levels increase with HFE
expression regardless of the HFE variant. Our hypothesis that the H63D would result in
increased APP was not supported since the H63D variant did not result in an elevation
in APP synthesis, which highlights that HFE may have another cellular mechanism that
can influence APP production independent of cellular iron regulation. Conversely,
cellular iron is very important to amyloid regulation at the level of APP synthesis via the
5’UTR IRE (Rogers et al., 2002) and Aβ plaque formation through binding Aβ facilitating
fibril formation leading to senile plaque formation (Liu et al., 2006). Future studies
92
should incorporate our findings and integrate them with cellular iron effects on amyloid
regulation shown by others to understand the impact of HFE variants on amyloid
homeostasis.
93
Figure 2.1: Total APP levels in HFE stably transfected SH‐SY5Y cells. Expression of total APP protein was measured in an HFE polymorphism stably transfected SH‐SY5Y cell line by ELISA. Total APP levels were significantly decreased in vector cells (p<0.01) compared to WT HFE expressing cells. There was no difference in total APP between cells expressing the H63D or WT variant. Experiments were performed with a minimum of four different cultures per genotype. One‐way ANOVA was performed to analyze the data, followed by Tukey’s post‐hoc analysis. Data are represented as mean ± S.E.M. The symbol ** (p<0.01) indicates a significance difference from wild type HFE.
Total APP
Vector WT H63D0
25
50
75
100
***
% D
iffer
ence
of c
ontr
ol
94
Figure 2.2: Amyloid secretase activity in HFE stably transfected SH‐SY5Y cells. Alpha (α)‐, beta (β), and gamma (γ)‐secretase enzymatic activity was determined using DNA probe‐based secretase activity assays. α‐Secretase activity was unaffected in any of the cell types compared to cells expressing the WT HFE variant. There was no change in β‐secretase activity either among the vector and HFE variant cells. γ‐secretase activity did not change with expression of HFE expressing wild type or H63D variants. These activity assays were performed with a minimum of four different cultures per genotype. One‐way ANOVA was performed to analyze the data, followed by Tukey’s post‐hoc analysis if applicable. Data are represented as mean ± S.E.M.
Alpha Secretase Activity
Vector WT H63D0
250
500
750
Enzy
mat
ic A
ctiv
ityFl
uore
scen
ce @
495
nm
Beta Secretase Activity
Vector WT H63D0
250
500
750
1000En
zym
atic
Act
ivity
Fluo
rsce
ne @
495
nm
Gamma Secretase Activity
Vector WT H63D0
25
50
75
100
Enzy
mat
ic A
ctiv
ityFl
uors
cene
@ 4
95nm
95
Figure 2.3: Cellular iron effects on amyloid secretase activity. Increasing amounts of ferrous ammonium citrate (FAC) to cells expressing wild type HFE, and increasing amounts of the iron chelator desferrioxamine (DFO) to cells expressing H63D HFE, and measured α‐, β‐, and γ‐secretase activity by secretase‐specific activity assays. Iron treatments increased α‐secretase activity at 10 μM (p<0.01) and 30 μM (p<0.01) compared to the non‐treated baseline WT HFE control levels (A). DFO exposure in H63D variant cells did not impact α‐secretase activity (A). There were no changes in β‐secretase activity with iron treatments in cells expressing the WT HFE variant (B). The removal of iron by chelation in H63D variant cells did not change β‐secretase activity either (B). γ‐secretase activity was not altered by iron or DFO treatments in wild type or H63D HFE variant cells, respectively (C). Experiments were performed with a minimum of four different cultures per genotype. One‐way ANOVA was performed to analyze the data, followed by Tukey’s post‐hoc analysis. Data are represented as mean ± S.E.M. The symbols ** (p<0.01) and *** (p<0.001) indicates a significance difference from the respective non‐treated baseline control.
96
A.
B.
C.
Alpha Secretase Activity
WT
WT 10 µM
iron
WT 30 µM
iron
0
250
500
750 *** **En
zym
atic
Act
ivity
Fluo
rsce
ne @
495
nm
Alpha Secretase Activity
H63D
H63D 5
µM D
FO
H63D 10
µM D
FO0
250
500
750
Enzy
mat
ic A
ctiv
ityFl
uors
cene
@ 4
95nm
Beta Secretase Activity
WT
WT 10 µM
iron
WT 30 µM
iron
0
200
400
600
800
Enzy
mat
ic A
ctiv
ityFl
uors
cene
@ 4
95nm
Beta Secretase Activity
H63D
H63D 5
µM D
FO
H63D 10
µM D
FO0
250
500
750En
zym
atic
Act
ivity
Fluo
rsce
ne @
495
nm
Gamma Secretase Activity
WT
WT 10 µM
iron
WT 30 µM
iron
0
25
50
75
100
Enzy
mat
ic A
ctiv
ityFl
uors
cene
@ 4
95nm
Gamma Secretase Activity
H63D
H63D 5
µM D
FO
H63D 10
µM D
FO0
25
50
75
100
Enzy
mat
ic A
ctiv
ityFl
uors
cene
@ 4
95nm
97
Figure 2.4: Impact of cellular iron exposure on amyloid secretase activity in H63D variant cells. H63D variant cells have a higher baseline of cellular iron compared to cells expressing WT HFE, therefore, additional iron exposure may impact amyloid secretase activity. We further challenged the H63D cells by treating them with iron in the form of ferric ammonium citrate (FAC). There were no effects of FAC on α‐, β‐, or γ‐secretase activity in the cells expressing the H63D variant. Experiments were performed with a minimum of four different cultures per genotype. One‐way ANOVA was performed to analyze the data, followed by Tukey’s post‐hoc analysis if applicable. Data are represented as mean ± S.E.M.
Alpha Secretase Activity
H63D
H63D 10
µM ir
on
H63D 30
µM ir
on0
250
500
750
Enzy
mat
ic A
ctiv
ityFl
uors
cene
@ 4
95nm
Beta Secretase Activity
H63D
H63D 10
µM ir
on
H63D 30
µM ir
on0
250
500
750En
zym
atic
Act
ivity
Fluo
rsce
ne @
495
nm
Gamma Secretase Activity
H63D
H63D 10
µM ir
on
H63D 30
µM ir
on0
25
50
75
100
Enzy
mat
ic A
ctiv
ityFl
uors
cene
@ 4
95nm
98
Figure 2.5: C83 APP levels in HFE stably transfected SH‐SY5Y cells. The intracellular C‐terminal fragment, C83, of APP was measured in an HFE polymorphism stably transfected SH‐SY5Y cell line by western blot. There was a not a statistically significant change in the expression of C83 in vector or H63D variant cells compared to cells expressing the wild type HFE variant. Experiments were performed with a minimum of four different cultures per genotype. One‐way ANOVA was performed to analyze the data, followed by Tukey’s post‐hoc analysis if applicable. Data are represented as mean ± S.E.M. C99 levels were undetectable for quantification.
Vector WT H63D
APP (C83)
β- actin
APP (C99)
β- actin
Vector WT H63D
Undetectable C99 protein
levels for quantification
APP C83
Vector WT H63D0
20
40
60
80
100
120
% D
iffer
ence
of c
ontr
ol
99
References
Barnham KJ, McKinstry WJ, Multhaup G, Galatis D, Morton CJ, Curtain CC, Williamson
NA, White AR, Hinds MG, Norton RS, Beyreuther K, Masters CL, Parker MW,
Cappai R (2003) Structure of the Alzheimerʹs disease amyloid precursor protein
copper binding domain. A regulator of neuronal copper homeostasis. J Biol
Chem 278:17401‐17407.
Bartzokis G, Tishler TA, Shin IS, Lu PH, Cummings JL (2004) Brain ferritin iron as a risk
factor for age at onset in neurodegenerative diseases. Ann N Y Acad Sci
1012:224‐236.
Bodovitz S, Falduto MT, Frail DE, Klein WL (1995) Iron levels modulate alpha‐secretase
cleavage of amyloid precursor protein. J Neurochem 64:307‐315.
Borghi R, Patriarca S, Traverso N, Piccini A, Storace D, Garuti A, Cirmena G, Odetti P,
Tabaton M (2006) The increased activity of BACE1 correlates with oxidative
stress in Alzheimerʹs disease. Neurobiol Aging.
Bush AI (2003) The metallobiology of Alzheimerʹs disease. Trends Neurosci 26:207‐214.
Connor JR, Menzies SL (1995) Cellular management of iron in the brain. J Neurol Sci 134
Suppl:33‐44.
Connor JR, Lee SY (2006) HFE mutations and Alzheimerʹs disease. J Alzheimers Dis
10:267‐276.
Connor JR, Menzies SL, St Martin SM, Mufson EJ (1992a) A histochemical study of iron,
transferrin, and ferritin in Alzheimerʹs diseased brains. J Neurosci Res 31:75‐83.
100
Connor JR, Snyder BS, Beard JL, Fine RE, Mufson EJ (1992b) Regional distribution of
iron and iron‐regulatory proteins in the brain in aging and Alzheimerʹs disease. J
Neurosci Res 31:327‐335.
Connor JR, Milward EA, Moalem S, Sampietro M, Boyer P, Percy ME, Vergani C, Scott
RJ, Chorney M (2001) Is hemochromatosis a risk factor for Alzheimerʹs disease? J
Alzheimers Dis 3:471‐477.
Feder JN, Penny DM, Irrinki A, Lee VK, Lebron JA, Watson N, Tsuchihashi Z, Sigal E,
Bjorkman PJ, Schatzman RC (1998) The hemochromatosis gene product
complexes with the transferrin receptor and lowers its affinity for ligand binding.
Proc Natl Acad Sci U S A 95:1472‐1477.
Grbovic OM, Mathews PM, Jiang Y, Schmidt SD, Dinakar R, Summers‐Terio NB, Ceresa
BP, Nixon RA, Cataldo AM (2003) Rab5‐stimulated up‐regulation of the
endocytic pathway increases intracellular beta‐cleaved amyloid precursor
protein carboxyl‐terminal fragment levels and Abeta production. J Biol Chem
278:31261‐31268.
Haque N, Tanaka T, Iqbal K, Grundke‐Iqbal I (1999) Regulation of expression,
phosphorylation and biological activity of tau during differentiation in SY5Y
cells. Brain Res 838:69‐77.
Hardy JA, Higgins GA (1992) Alzheimerʹs disease: the amyloid cascade hypothesis.
Science 256:184‐185.
101
Huang X, Moir RD, Tanzi RE, Bush AI, Rogers JT (2004) Redox‐active metals, oxidative
stress, and Alzheimerʹs disease pathology. Ann N Y Acad Sci 1012:153‐163.
Lee SY, Connor JR (2005) Regulation of Hfe by stress factors in BV‐2 cells. Neurobiol
Aging 26:803‐812.
Lee SY, Patton SM, Henderson RJ, Connor JR (2006) Consequences of expressing
mutants of the hemochromatosis gene (HFE) into a human neuronal cell line
lacking endogenous HFE. Faseb J.
Liu G, Huang W, Moir RD, Vanderburg CR, Lai B, Peng Z, Tanzi RE, Rogers JT, Huang
X (2006) Metal exposure and Alzheimerʹs pathogenesis. J Struct Biol 155:45‐51.
Masters CL, Simms G, Weinman NA, Multhaup G, McDonald BL, Beyreuther K (1985)
Amyloid plaque core protein in Alzheimer disease and Down syndrome. Proc
Natl Acad Sci U S A 82:4245‐4249.
Mattson MP (1997) Cellular actions of beta‐amyloid precursor protein and its soluble
and fibrillogenic derivatives. Physiol Rev 77:1081‐1132.
Multhaup G (1997) Amyloid precursor protein, copper and Alzheimerʹs disease. Biomed
Pharmacother 51:105‐111.
Nixon RA (2005) Endosome function and dysfunction in Alzheimerʹs disease and other
neurodegenerative diseases. Neurobiol Aging 26:373‐382.
Peraus GC, Masters CL, Beyreuther K (1997) Late compartments of amyloid precursor
protein transport in SY5Y cells are involved in beta‐amyloid secretion. J Neurosci
17:7714‐7724.
102
Pulliam JF, Jennings CD, Kryscio RJ, Davis DG, Wilson D, Montine TJ, Schmitt FA,
Markesbery WR (2003) Association of HFE mutations with neurodegeneration
and oxidative stress in Alzheimerʹs disease and correlation with APOE. Am J
Med Genet B Neuropsychiatr Genet 119B:48‐53.
Quintana C, Bellefqih S, Laval JY, Guerquin‐Kern JL, Wu TD, Avila J, Ferrer I, Arranz R,
Patino C (2006) Study of the localization of iron, ferritin, and hemosiderin in
Alzheimerʹs disease hippocampus by analytical microscopy at the subcellular
level. J Struct Biol 153:42‐54.
Rogers JT, Leiter LM, McPhee J, Cahill CM, Zhan SS, Potter H, Nilsson LN (1999)
Translation of the alzheimer amyloid precursor protein mRNA is up‐regulated
by interleukin‐1 through 5ʹ‐untranslated region sequences. J Biol Chem 274:6421‐
6431.
Rogers JT, Randall JD, Eder PS, Huang X, Bush AI, Tanzi RE, Venti A, Payton SM,
Giordano T, Nagano S, Cahill CM, Moir R, Lahiri DK, Greig N, Sarang SS,
Gullans SR (2002) Alzheimerʹs disease drug discovery targeted to the APP
mRNA 5ʹuntranslated region. J Mol Neurosci 19:77‐82.
Rohrlich PS, Fazilleau N, Ginhoux F, Firat H, Michel F, Cochet M, Laham N, Roth MP,
Pascolo S, Nato F, Coppin H, Charneau P, Danos O, Acuto O, Ehrlich R,
Kanellopoulos J, Lemonnier FA (2005) Direct recognition by alphabeta cytolytic T
cells of Hfe, a MHC class Ib molecule without antigen‐presenting function. Proc
Natl Acad Sci U S A 102:12855‐12860.
103
Sayre LM, Perry G, Harris PL, Liu Y, Schubert KA, Smith MA (2000) In situ oxidative
catalysis by neurofibrillary tangles and senile plaques in Alzheimerʹs disease: a
central role for bound transition metals. J Neurochem 74:270‐279.
Smith MA, Harris PL, Sayre LM, Perry G (1997) Iron accumulation in Alzheimer disease
is a source of redox‐generated free radicals. Proc Natl Acad Sci U S A 94:9866‐
9868.
Takeda K, Araki W, Tabira T (2004) Enhanced generation of intracellular Abeta42
amyloid peptide by mutation of presenilins PS1 and PS2. Eur J Neurosci 19:258‐
264.
Vassar R et al. (1999) Beta‐secretase cleavage of Alzheimerʹs amyloid precursor protein
by the transmembrane aspartic protease BACE. Science 286:735‐741.
Wang XS, Lee S, Simmons Z, Boyer P, Scott K, Liu W, Connor J (2004) Increased
incidence of the Hfe mutation in amyotrophic lateral sclerosis and related
cellular consequences. J Neurol Sci 227:27‐33.
Zecca L, Youdim MB, Riederer P, Connor JR, Crichton RR (2004) Iron, brain ageing and
neurodegenerative disorders. Nat Rev Neurosci 5:863‐873.
Zhang M, Haapasalo A, Kim DY, Ingano LA, Pettingell WH, Kovacs DM (2006)
Presenilin/gamma‐secretase activity regulates protein clearance from the
endocytic recycling compartment. Faseb J 20:1176‐1178.
104
Chapter 3
The H63D HFE Gene Variant Promotes Activation of the Intrinsic
Apoptosis Pathway with Aβ Peptide Exposure
Abstract
Numerous epidemiological studies suggest that the expression of the HFE allelic
variant H63D may be a risk factor or genetic modifier for Alzheimer’s disease (AD). The
H63D variant alters cellular iron homeostasis and increases baseline oxidative stress.
The elevated stress milieu, we have proposed, may alter cellular responses to genetic
and environmental determinants of AD. Accumulation of beta amyloid peptides (Aβ) is
one of the most prominent pathogenic characteristics of AD. Several studies have
demonstrated that Aβ can induce neuronal cell death through apoptosis. In this study,
we provide evidence that an Aβ25–35 fragment, which contains the cytotoxic sequence of
the amyloid peptide, activates the intrinsic apoptosis pathway in SH‐SY5Y human
neuroblastoma cells expressing the HFE allelic variant H63D to a greater extent than in
cells with wild type HFE. Specifically, Aβ25–35 peptide exposure significantly induced Bax
translocation from the cytosol to the mitochondria in H63D expressing cells compared to
WT cells. This translocation was associated with increased cytochrome C release from
mitochondria and an increase in caspase‐9 and caspase‐3 activation in H63D cells.
Consequently, there is increased apoptosis in cells expressing the H63D variant opposed
105
to cells expressing wild type HFE. We also found increased amyloid precursor protein
(APP) and Aβ peptide in the mitochondrial compartment, as well as increased
mitochondrial stress in H63D expressing cells compared to WT cells. These findings
support our hypothesis that the presence of the HFE H63D allele enables factors that
trigger neurodegenerative processes associated with AD and predisposes cells to
cytotoxcity. These studies reveal a mechanism, activation of the mitochondrial apoptotic
pathway, by which HFE H63D could serve as a disease modifier for AD.
Introduction
Alzheimer’s disease (AD) is a neurodegenerative disease that afflicts the elderly
resulting in memory loss and decline in cognitive abilities. AD histopathology is
characterized by the accumulation of amyloid‐β (Aβ) plaques and intracellular
neurofibrillary tangles (Nielsen et al., 1995; Jellinger and Bancher, 1998; Trojanowski and
Lee, 2000; Jellinger, 2002). An important pathological finding of AD is the accumulation
of iron in the same anatomical brain regions characterized by Aβ deposition (Zecca et al.,
2004). Although the pathogenesis of AD is still unclear, several findings suggest that
altered iron metabolism in the brain contributes to changes in amyloid metabolism. Iron
has been shown to impact Aβ plaque formation and amyloid processing, as well as to
enhance neurotoxicity of Aβ (Lovell et al., 1998; Huang et al., 2000; Rottkamp et al.,
2001). Free iron also catalyzes the conversion of superoxide and hydrogen peroxide into
hydroxyl radicals, which promote oxidative stress leading to cell death (Bishop et al.,
106
2002). These observations make a compelling argument that iron status in the brain is a
contributing factor to the pathogenesis of AD. It is logical to propose therefore, that
mutations or variants in genes associated with maintaining iron homeostasis may be
more prevalent in AD. HFE is a protein that is associated with cellular iron uptake. The
mutant H63D protein is associated with elevated intracellular iron levels, which
enhances cell stress (Limdi and Crampton, 2004; Lee et al., 2007b).
Magnetic resonance imaging has also shown increased iron accumulation in the
brains of individuals with hereditary hemochromatosis, who express HFE mutations
(Nielsen et al., 1995; Berg et al., 2000). Several studies have investigated the relationship
between HFE genetic variants and neurodegenerative diseases including AD,
amyotrophic lateral sclerosis, and Parkinson’s disease because loss of brain iron
homeostasis has been associated with proposed mechanisms of neurodegeneration in
each of these disorders (Schenck and Zimmerman, 2004; Wallis et al., 2008). Several
epidemiological studies have found an association between HFE variants and AD,
however some studies did not report a genetic association with AD cases (Moalem et al.,
2000; Sampietro et al., 2001; Candore et al., 2003; Combarros et al., 2003; Pulliam et al.,
2003; Berlin et al., 2004; Connor and Lee, 2006; Guerreiro et al., 2006; Avila‐Gomez et al.,
2008; Percy et al., 2008; Alizadeh et al., 2009). Inconsistency in the populations studied,
and the variability of iron availability in the environment (e.g. soil content, diet, water)
are strongly suggestive of a gene/environment interaction for a risk factor.
107
Thus, we have developed cell culture models to examine the role that HFE might
play in AD pathology mechanisms. Our hypothesis for these studies is that HFE gene
variants create a permissive intracellular milieu that enables pathogenic events. Aβ
toxicity is theorized to play a major role in AD pathogenesis because it is toxic to
neurons (Hardy and Selkoe, 2002). Aβ peptides can increase the generation of reactive
oxygen species (ROS), as well as evoke a cascade of oxidative damage in neurons
(Miranda et al., 2000) resulting in the induction of caspase‐3‐like activity and cell death
by apoptosis (Richardson et al., 1996; Nicotera et al., 1999; Fifre et al., 2006). Moreover, it
has been reported that Aβ induces neuronal apoptosis. Aβ can promote translocation of
pro‐apoptotic members of the Bcl‐2 protein family (for example, Bak and bax) to the
mitochondria. Bcl‐2 family proteins can stimulate the opening of voltage dependent
anion channels, (VDAC), and cause release of cytochrome c into the cytosol. This latter is
a key initiation step in the apoptotic process (Sola et al., 2002) activating caspase‐9 and
effector caspase‐3 (Ferreiro et al., 2007). Finally, in support of the hypothesis that there
will be an HFE association, the addition of Aβ peptide and exogenous iron significantly
promoted neuronal cell death via the apoptotic pathway (Kuperstein and Yavin, 2003;
Wallis et al., 2008).
Materials and Methods
Reagents ‐ Cell culture reagents including DMEM/F12, DMEM, pen/strep/glutamine and
Geneticin were purchased from Invitrogen (Carlsbad, CA, USA). Fetal bovine serum
108
was purchased from Gemini Bio‐Products (West Sacramento, CA, USA). Hydrogen
peroxide/peroxidase assay kit was ordered from Molecular Probes (Eugene, OR, USA).
Reactive oxygen species were from Calboichem (CA, USA), respectively. Mitochondria
activity assay (Cytochrome c oxidase activity assay kit) was purchased from BioChain
Institute (Hayward, CA, USA). Aβ25‐35 and β‐actin monoclonal antibody and beta tubulin
antibody be were ordered from Sigma Co. (St. Louis, MO, USA). Caspase‐9, Caspase‐8,
Bax, cytochrome c, cytochrome oxidase subunit IV antibodies were ordered from Cell
Signaling Technology (Beverly, MA, USA), Santa Cruz Biotechnology (Santa Cruz, CA,
USA), Upstate (Temecula, CA, USA), Clontech (Mountain View, CA, USA) and
Molecular Probes (Eugene, OR, USA) respectively. Amyloid precursor protein antibody
was ordered from Signet (Emeryville, CA, USA).Secondary anti‐rabbit antibody or anti‐
mouse antibody and ECL detection kits were obtained from Amersham Pharmacia
Biotech (Piscataway, NJ, USA). Apoptosis assay kit was purchased from Molecular
Probes. Caspase‐3 Fluorometric assay was ordered from R&D Systems (Minneapolis,
MN, USA). Human beta amyloid 1‐42 colorimetric ELISA kit was purchased from
Biosource (Camarillo, CA, USA). Nitrocellulose membrane was ordered from Pall Life
Sciences (Pensacola, FL, USA), RIPA buffer and phosphates inhibitor (Sigma Aldrich, St.
Louis, MO). Cell fractionation buffer was purchased from Clontech (Mountain View,
CA, USA). BCA protein assay (Pierce Chemical, Rockford, IL, USA). All of the other
chemicals used were purchased from Sigma Co (St. Louis, MO, USA).
109
Cell Culture ‐ SH‐SY5Y human neuroblastoma cell lines were obtained from American
Type Culture Collection (Manassas, VA, USA). Human neuroblastoma SH‐SY5Y cells
were stably transfected to express wild‐type or H63D HFE as previously reported (Lee et
al., 2007b). Vector alone stably transfected cells were used as a control; however the
SHSY5Y cells were chosen because they did not express detectable levels of HFE protein
or mRNA (Lee paper) and thus the appropriate control for the H63D is the wt HFE cells.
The transfected cells were maintained in DMEM/F12 media supplemented with 10%
FBS, 1% antibiotics (pen‐strep‐glutamine), 1x nonessential amino acids, and 1.8g/L
sodium bicarbonate at 37 °C in a 5% CO2 atmosphere.
Cell viability assay ‐ Cells were plated at a density (2x104 cells/well) in 96‐well flat‐
bottomed microtiter plates and then cultured for 48 hours (h). After 48 h, the cells were
exposed to Aβ25‐35 peptides at the concentrations of 10, 20 and 50 μM for 24 h. After 24 h
treatment, 50 μl of 2 mg/ml MTT (Sigma Co St. Louis, MO, USA) was added to 200 μl of
medium present in each well and incubated at 37°C for 2 h. After the incubation period,
an aliquot (220 μL) of the resulting solution was removed from each well followed by
the addition of 150 μL dimethyl sulfoxide. After the precipitate in each well was
resuspended on a microplate mixer for 10 min, an optical density (OD) reading at
540 nm was measured using a plate reader (Spectramax 340PC). All results were
normalized to OD values measured from an identically conditioned well without cell
culture.
110
Preparation of cell lysates ‐ After Aβ treatment (4 h for caspase‐8, bax and cytochrome c
and 9 h for caspase‐9), the cells were washed twice in cold PBS and were then removed
by using cell scraper. The cell suspension was centrifuged at 300 x g for 3 min. The cell
pellets were resuspended in lysis buffer (1% Triton X‐100, 1% phosphates inhibitor, 1%
protease inhibitor cocktail and RIPA buffer) and incubated for 10 min. The lysate were
cleared by centrifugation at 10,000xg for 20 min at 4°C. The supernatant was then
assayed for caspase‐8 and caspase‐9. For evaluation of cytochrome c, bax and
cytochrome c oxidase subunit IV (COX IV) levels, the cells suspension were centrifuged
at 300 x g for 3 min and then the cell pellets were resuspended in a fractionation buffer
mix, left on ice for 10 min and homogenized by 50 strokes in an ice‐cold Dounce
homogenizer. The cells extract was centrifuged at 700 x g for 10 min at 4°C, resulting in a
pellet containing nuclei and supernatant retaining mitochondria and cytosol. The
supernatant was further centrifuged at 10,000 x g for 25 min at 4°C to collect the pellet
enriched‐mitochondria fraction. The resulting supernatant containing the cytosolic
fraction was collected and the enriched mitochondria pellet was lysed by incubation
with fractionation buffer mix. The protein concentrations were determined by BCA
protein assay.
Immunoblotting ‐ For western blot analysis, 50 μg of total protein from each sample was
performed on gradient gels 4–20% (Bio‐Rad, Hercules, CA. USA) and transferred to
111
nitrocellulose membranes. The membranes were blocked with 5% skim milk and
incubated with specific antibodies for caspase‐8 (1:200), cleaved (activated form)
caspase‐9 (1:1000), cytochrome c (1:100), bax (1:2000), cox (1:1000), APP (1:1000) and
beta‐tubulin (1:2000). The membranes were washed with Tris buffered saline containing
0.05% Tween‐20 (TBST) and processed with an HRP‐conjugated anti‐rabbit antibody or
anti‐mouse antibody for ECL detection. The probing for internal control molecules such
as cytochrome c oxidase subunit IV (COX) (1:1000) and actin (1:3000) was always
performed by re‐hybridization after stripping the primary antibodies.
Capase‐3 activity assay ‐ A caspase‐3 fluorometric assay kit was utilized. Briefly, after 8
and 12 h Aβ treatment the cells were collected by centrifugation at 300 x g for 3 min. The
cell pellets were lysed upon addition of the cold lysis buffer (provided by the
manufacturer) and incubated on ice for 10 min. The protein content of the cell lysate
was estimated using a protein determination assay. 100 μg of total protein was then
mixed with 2X reaction buffer (provided by the manufacturer) and 5 μl of caspase‐3
fluorogenic substrate (DEVD‐AFC) in a 96‐well plate. The reaction was incubated at 37
°C for 1.30 h followed by fluorescence analysis using a fluorescent microplate reader
with light excitation at 400nm and light emission at 505 nm.
Cytofluorometric determination of apoptotic cells by Annexin V/PI (propidium iodide)‐staining ‐
HFE cells were exposed to 20 μM Aβ25‐35 for 24 h and then labeled with Annexin V/PI.
112
Briefly, Aβ‐treated and ‐untreated HFE cells (5×105) were washed in PBS and then
incubated with 5 μl of 0.2 μg/ml Annexin V‐FITC and 1 μl of 100 μg/ml propidium
iodide for 15 min at room temperature, prior to flowcytometry analysis. Annexin V+/PI−
cells were identified as early apoptotic, whereas Annexin V+/PI+ cells were classified as
late apoptotic cells.
Statistical analysis ‐ The data were analyzed by one‐way ANOVA. Differences among the
means were considered statistically significant when the p value was <0.05. If overall p
<0.05, Tukey’s Multiple Comparison post hoc analysis was performed. A two‐way
ANOVA with genotype (Vector, WT, H63D) and Aβ25‐35 concentration (10 μM, 20 μM, 50
μM) as grouping factors was performed where appropriate. Bonferroni post tests were
performed if p<0.05. Data are presented as the mean ± S.E.; GraphPad Prism software
(version 4.0) was utilized to perform the statistical analysis.
Results
To evaluate the viability of cells following Aβ25‐35 treatment (toxic fragment of
Aβ), HFE cells were incubated with Aβ25‐35 (0, 10, 20 or 50 μM) for 24 hours. After
incubation, cell viability was measured by an MTT reduction assay. Following Aβ25‐35
treatment, cell viability for each genotype was significantly decreased compared to
untreated control cultures. At 10 μM Aβ25‐35 exposure, vector cells that do not express
HFE had significantly more viable cells compared to wild type HFE and H63D
113
expressing cells (p<0.001) but there was not a significant difference between cells
expressing H63D or WT HFE at this concentration (Figure 3.1). Upon treating the cells
with 20 μM Aβ25‐35, there was a significant decrease in H63D cell viability compared to
WT HFE cells (p<0.05). The vector control was less vulnerable than either WT or H63D
cells (p<0.01). A similar profile of toxicity was seen in cell types at 50 μM Aβ25‐35
exposure. Because there was a significant difference between H63D expressing cells and
WT at 20 μM Aβ, 20 μM was selected as the Aβ25‐35 concentration for subsequent
experiments. Also, a significant interaction of genotype by Aβ25‐35 concentration was
detected (p=0.0005) by two‐way ANOVA.
To investigate potential mediators of apoptosis, we measured Bax translocation
into the mitochondria. Mitochondrial stress and dysfunction have been shown to up‐
regulate Bax translocation (Vila et al., 2001; Perier et al., 2007) and we have shown that
H63D cells have decreased mitochondria membrane potentials (Lee et al., 2007b). Thus,
we tested the hypothesis that Bax translocation will increase with H63D expression and
will be further enhanced by Aβ exposure compared to WT cells. Indeed, the ratio of Bax
found in the mitochondrial fraction compared to the cytosolic fraction was increased in
vector (p<0.01) and H63D (p<0.001) cells compared to WT HFE cells (Figure 3.2). The
purity of both fractions was validated by immunoblotting for marker proteins, using
tubulin and cytochrome c oxidase subunit IV (COX) as markers for cytosol and
mitochondria fractions, respectively. Densitometric analysis of Bax in the mitochondria
was normalized to the level of COX IV (A) and the expression of Bax in the cytosol was
114
normalized to the level of tubulin (B). After a 4 hour exposure to 20 μM Aβ25‐35, the ratio
of Bax in the mitochondria to cytosol was significantly increased in cells expressing
H63D (p<0.001) and vector (p<0.05) compared to WT cells (Figure 3.2). There was a
greater than two‐fold increase in the ratio of mitochondrial to cytosolic Bax in H63D
cells compared to WT, regardless of whether or not the cells were exposed to Aβ.
Based on enhanced Bax translocation to the mitochondria in H63D cells at
baseline and upon Aβ exposure, we evaluated cytochrome C release as another potential
modulator of apoptosis. Cytochrome C is released from the mitochondria under cellular
stress and its release can promote apoptosis similar to Bax (Kluck et al., 1997; Li et al.,
1997; Slee et al., 1999; Desagher and Martinou, 2000). Therefore, cellular fractions of the
mitochondria and cytosol were obtained to evaluate cytochrome C levels among the
various cell types. The data show a two‐fold increase of cytochrome C release in vector
cells (p<0.001) compared to WT and a greater than two‐fold increase of cytochrome C
release in H63D cells (p<0.001) with respect to WT (Figure 3.3). When the cells were
treated with 20 μM Aβ peptide to investigate the cellular response to this stress agent,
there was a greater than two‐fold increase of cytochrome C release in vector cells
(p<0.001) compared to WT. There was a greater than three‐fold increase of cytochrome
C release in H63D cells (p<0.001) versus WT cells (Figure 3.3). It also important to note
that there was three times as much cytochrome C release in H63D cells upon exposure to
Aβ compared to baseline release of cytochrome C; whereas WT cells had only twice as
much cytochrome C release when treated with Aβ.
115
Caspase‐9, an initiator caspase activator of the intrinsic apoptosis pathway, can
be regulated by alterations in mitochondria and is responsive to changes in Bax
translocation and cytochrome C release (Earnshaw et al., 1999; Gervais et al., 1999;
Weidemann et al., 1999). Thus, we evaluated caspase‐9 activation by measuring cleaved
caspase‐9 in our cellular model. There is a two‐fold increase of caspase‐9 activation in
vector (p<0.001) and H63D (p<0.001) cells compared to wild type (Figure 3.4). Numerous
studies have shown that Aβ toxicity can be mediated by activation of caspase‐9, and the
downstream executioner, caspase‐3 (Chan and Mattson, 1999; Allen et al., 2001; Fan et
al., 2005). Our previous experiment showed that cytochrome c release was changed at 4
hours after Aβ25‐35 treatment. Since, caspase‐9 is downstream of cytochrome c, we
expected that caspase‐9 would be changed after 4 hour of Aβ25‐35 treatment. Therefore,
we treated the cells for 9 hours in order to detect caspase‐9. Upon treating the cells with
20 μM Aβ25‐35 for 9 hours, levels of the active form of caspase‐9 increased two‐fold in WT
cells compared to its control expression level, but the levels were still at least 25% less
than those seen in the vector and H63D Aβ‐treated cells (Figure 3.4). Aβ exposure
increased caspase‐9 activation more than 25% in vector and H63D cells compared to
their baseline untreated levels.
Caspase‐3 is an executioner caspase and it has been demonstrated to be a major
protease in apoptosis, especially in response to cellular stress (Earnshaw et al., 1999;
Gervais et al., 1999; Weidemann et al., 1999; Yuan and Yankner, 2000). Given the
preceding data in this study revealing alterations of upstream proteins in the intrinsic
116
apoptotic pathway, we hypothesized that there would be increased caspase‐3 activity in
cells expressing the H63D allele versus wild‐type HFE cells. Caspase‐3 activity is
elevated in vector (p<0.001) and H63D cells (p<0.01) compared to WT HFE cells (Figure
3.5). Our previous experiment showed that caspase‐9 was changed after 9 hours of Aβ25‐
35 treatment. Since, caspase‐3 is downstream of caspase‐9; we expected that caspase‐3
would be changed after 9 hours of Aβ25‐35 treatment. In this experiment, we treated the
cells for 8 and 12 hours in order to detect caspase‐3 activation. Subsequent treatment of
the cells with Aβ25‐35 led to an increase of caspase‐3 activation for all cell types compared
to baseline. At 8 hours of Aβ exposure, caspase‐3 activity increased in vector (p<0.001)
and H63D (p<0.001) cells compared to WT HFE cells (Figure 3.5). It is also important to
note that there was greater activation of caspase‐3 in H63D cells upon 8 hours exposure
to Aβ compared to its baseline than was seen with either vector or WT cells. Exposure
for 12 hours to Aβ exposure did not result in any additional increase in caspase‐3
activity than seen at 8 hours exposure.
The initial findings of Aβ exposure leading to a greater loss of viable cells in WT
HFE and H63D HFE compared to the vector control (Figure 1.1) is not fully explained by
activation of the apoptotic intrinsic pathway. Therefore, the extrinsic pathway may be
involved. Caspase‐8 is the key initiator of the extrinsic apoptosis signaling pathway
(Nijhawan et al., 2000). Thus, we evaluated the expression of activated caspase‐8 and
found WT HFE expression to result in an 11% increase compared to vector and a 45%
increase with respect to H63D expressing cells (Figure 3.6). Moreover, there was a 23%
117
increase of caspase‐8 activation in WT HFE cells compared to H63D and a 17% increase
compared to vector (Figure 3.6).
Previous findings suggest that the Aβ peptide is neurotoxic and pro‐apoptotic
(Yankner et al., 1989; Pike et al., 1991; Loo et al., 1993). To determine the effect of Aβ
peptide on neuronal apoptosis in HFE transfected human neuroblastoma SH‐SY5Y cell
lines, an Annexin V‐FITC/PI flow cytometry assay was performed. In evaluating early
apoptosis, there was a 15% decrease of early apoptotic cells in the vector cells (p<0.001)
compared to WT cells; there was no significant difference between WT and H63D cells
(Figure 3.7). All cell types were exposed to Aβ25‐35 peptide for 24 hours and apoptosis
increased across all genotypes. There were fewer early vector apoptotic (p<0.001) cells
compared to WT HFE early apoptotic cells (Figure 3.7). However, there were more early
apoptotic cells that expressed the H63D (p<0.01) allele compared to cells expressing WT
HFE (Figure 3.7). Importantly, there were more early H63D apoptotic cells than either
vector or WT HFE expressing cells with Aβ peptide exposure.
Discussion
The results of this study indicate that the H63D HFE gene variant is associated
with increased cell stress and mitochondrial dysfunction. Exposure of neuroblastoma
cells to Aβ peptide results in greater cell stress and cell death in those cells carrying
H63D HFE allele. The evidence of elevated oxidative stress associated with the H63D
allele and apparent heightened activity of the intrinsic apoptotic pathway under resting
118
conditions, prompted us to evaluate apoptotic signaling pathways to understand the
mechanisms through which Aβ toxity was mediated. It is well established that Bax
translocation to the mitochondria plays a central role in intrinsic apoptosis signaling
(Sharpe et al., 2004). Bax is a BH3‐ only member of the Bcl‐2 family that mainly resides in
the cytosol in healthy cells and translocates to the mitochondria following exposure to
apoptotic stimuli. In this study, we discovered an increased baseline level of Bax in the
mitochondria in H63D cells compared to WT cells and the levels of Bax in the
mitochondria increased even further with Aβ exposure. It has been reported that Bax
can release cytochrome C by interacting with the mitochondrial permeability transition
(MPT) pore component, in particular the voltage‐dependant anion channels (VDAC). In
addition, Bax also caused mitochondrial alterations typical of MPT, such as loss of
mitochondrial membrane potential (Narita et al., 1998). Therefore, an increased baseline
level of Bax in the mitochondria in H63D cells might be associated with the decreased
mitochondrial membrane potential that we have observed (Lee et al., 2007b).
Glycogen synthase kinase‐3β (GSK‐3β) is a critical activator of neuronal
apoptosis that can phosphorylate Bax and promote its mitochondrial localization during
neuronal apoptosis (Linseman et al., 2004). GSK‐3β activity is increased in cells
expressing the H63D allele and increasing cellular iron can up‐regulate GSK‐3β activity
(Hall et al. in press). Thus, the elevated Bax protein expression in mitochondria in H63D
cells is consistent with increased GSK‐3β activity. The increase in Bax translocation to
119
mitochondria following Aβ exposure may also involve increased GSK‐3β activity
(Linseman et al., 2004).
The increase in Bax in the mitochondria suggests activation of the intrinsic
(mitochondrial) death pathway which is also consistent with the increased cytochrome C
release from mitochondria that was observed in H63D HFE cells. Upon its translocation
to the mitochondria, Bax can form oligomers with other Bax and Bcl‐2 family proteins
generating pores in the mitochondria membrane that stimulate the release of
cytochrome C from the mitochondria (Zong et al., 2001). The subsequent cytosolic
cytochrome C then interacts with Apaf‐1 and pro‐caspase‐9 to form a functional
apoptosome that ultimately activates downstream executioner caspases like caspase‐3
(Zou et al., 1999). Cytochrome C levels, caspase‐9, and caspase‐3 expression were
increased in H63D expressing cells compared to cells expressing wild type HFE and
were further elevated with Aβ peptide exposure. These findings are consistent with the
Bax expression findings and implicate the intrinsic apoptotic pathway to be altered in
cells expressing the H63D variant of HFE.
Additionally, increased cytosolic Ca2+ has been shown to activate the
mitochondrial apoptotic pathway via the opening of the mitochondrial transition pore,
resulting in the release of apoptogenic factors, including cytochrome c (Kroemer et al.,
1998; Rizzuto et al., 1998; Csordas et al., 1999; Duchen, 2000). We found that H63D HFE
variant is associated with elevated cytosolic Ca2+ levels (Mitchell et al. in press), which
may be a mechanism for increased cytochrome C release with expression of the H63D
120
HFE variant. Moreover, it has been reported that Aβ induces the release of Ca2+ from the
endoplasmic reticulum leading to activation of the mitochondrial apoptotic pathway
resulting in a disruption of mitochondrial membrane potential and Bax translocation
into the mitochondria (Ferreiro et al., 2008). This mechanism might also contribute to the
translocation of Bax into the mitochondria and enhanced cytochrome C release under
the H63D HFE variant.
To complete our analysis of the apoptotic pathways, we measured caspase‐8
activation to evaluate the extrinsic apoptotic pathway. Cells expressing H63D HFE
displayed a reduction in caspase‐8 activation compared to WT HFE cells. Aβ25‐35
treatment also resulted in a decrease of caspase‐8 expression in the H63D cells compared
to cells expressing WT HFE. Importantly, GSK‐3β activity has been reported to influence
the two major pathways of apoptosis, but in opposite directions; directly enhancing the
intrinsic pathway while inhibiting the extrinsic pathway (Beurel and Jope, 2006).
Collectively, these findings support the idea that Aβ exposure further elevates GSK‐3β
activity in H63D cells leading to an up‐regulation of the intrinsic apoptotic pathway and
inhibition of the extrinsic apoptotic pathway resulting in a greater rate of apoptosis.
In summary, the present study demonstrates that Aβ peptide exposure induces
apoptosis in the H63D cells by promoting the translocation of Bax to the mitochondria
leading to activation of the mitochondrial intrinsic pathway, which subsequently causes
cytochrome c release and the activation of caspase‐9 and caspase‐3. The results of this
study suggest there is greater vulnerability of H63D cells to Aβ toxicity due to a pre‐
121
existing up‐regulation of the intrinsic apoptotic pathway. Our findings suggest that
mitochondria dysfunction associated with the H63D variant is a key finding that might
accelerate the onset and/or exacerbation of neurodegeneration. The cellular response to
an environmental toxin like Aβ may impact AD progression and the expression of HFE
variants can impact neuronal cell survival. Thus, over the course of a lifetime, a
difference in cell sensitivity to Aβ exposure might greatly influence AD pathogenesis
and subsequent cognitively ability due to mitochondria dysfunction and cellular
oxidative stress associated with the H63D HFE allele. This observation provides further
support for our conceptual framework for studies into HFE polymorphisms in
neurodegenerative diseases and our overall working hypothesis that the H63D allelic
variant establishes a permissive milieu in which environmental factors can promote
neurodegeneration.
122
Figure 3.1: Measurement of cell viability. HFE cells were treated with Aβ25‐35 (0, 10, 20 or 50 μM) for 24 h. After incubation, cell viability was measured with an MTT assay. The data represent the mean ±SEM of at least three independent experiments. Two‐way ANOVA with genotype and Aβ25‐35 concentration as grouping factors. *P<0.05 when compared with WT; **P<0.01 when compared with WT; **P<0.01 when compared with WT.
25-35
β
0µM A
25-35
β
10µM
A 25
-35
β
20µM
A 25
-35
β
50µM
A
0102030405060708090
100110
VectorWTH63D**
******
Cel
l Via
bilit
y (%
of W
T co
ntro
l)
*
123
Figure 3.2: The effect of Aβ25‐35 treatment on induction of Bax translocation to mitochondria. Cells were treated with 20 μM Aβ25‐35 for 4 h, then the cells were fractionated to cytosolic and mitochondria‐enriched fractions. Mitochondria and cytosolic fractions were immunoblotted with protein marker for each fraction, tubulin for cytosol and COX IV for mitochondria, to verify purity of fractions. The levels of Bax in cytosol and mitochondria were analyzed by western blotting. Densitometric analysis of the level of Bax in mitochondria was normalized to the level of COX IV (A) and the level of Bax in cytosol was normalized to the level of tubulin (B) (the product is the ratio between A and B). The data represent the mean ±SEM of at least three independent experiments.*P<0.05 when compared with WT; **P<0.01 when compared with WT; ***P<0.001 when compared with WT.
Vec WT H63
D β
Vec A
β
WT Aβ
H63D A
0
1
2
3
4
5
6
7
8
*****
***
Rat
io o
f Bax
(mito
chon
dria
/cyt
osol
)
*
124
Figure 3.3: The effect of Aβ25‐35 treatment on Cytochrome c release. Cells were treated with 20 μM Aβ25‐35 for 4 h. subsequently, cytosolic and mitochondria‐enriched fractions were prepared. Mitochondria and cytosolic fractions were immunoblotted with protein marker for each fraction, tubulin for cytosol and COX IV for mitochondria, to verify purity of fractions. Cytochrome c level in cytosol fraction) and mitochondria were measured by western blotting. Densitometric analysis of the level of cytochrome c in cytosol was normalized to the level of tubulin (A) and the level of cytochrome c in mitochondria was normalized to the level of COX IV (B) (the product is the ratio between A and B. The data represent the mean ±SEM of at least three independent experiments. ***P<0.001 when compared with WT.
Vec WT H63
D β
Vec A
β
WT Aβ
H63D A
0
1
2
3
4
5
*** ***
******
Rat
io o
f Cyt
ochr
ome
C(c
ytos
ol/m
itoch
ondr
ia)
125
Figure 3.4: The effect of Aβ25‐35 treatment on induction of caspase‐9 activation. Cells were treated with 20 μM Aβ25‐35 for 9 h. Cells were collected and the levels of active form of caspase‐9 were analyzed by immunoblotting. Densitometric analysis of the level of active caspase‐9 was normalized to the level of actin (product: actin, ratio from each treatment is further compared with WT control). The data represent the mean ±SEM of at least three independent experiments. ***P<0.001 when compared with WT.
Vec WT H63
D β
Vec A
β
WT Aβ
H63D A
0.0
0.5
1.0
1.5
2.0
2.5
3.0
*** ***
*** ***
% D
iffer
ence
126
Figure 3.5: The effect of Aβ25‐35 treatment on induction of caspase‐3 activity. Cells were treated for 8, and 12h with 20 μM of Aβ25‐35. Cells were collected and the caspase‐3 activity was determined using a caspase‐3 flourometric assay. The data represent the mean ±SEM of at least three independent experiments. **P<0.01 when compared with WT and ***P<0.001 when compared with WT.
Caspase-3 Activity
Vecto
rW
T H63
D 8
hβ
Vec A
8hβ
WT A
8h
β
H63D A
12h
β
Vec A
12
h
β
WT A
12
h
β
H63D A
0
300
600
900
1200
*****
*** *** *** ***
Enzy
mat
ic A
ctiv
ityFl
uore
scen
ce @
505
nm
127
Figure 3.6: The effect of Aβ25‐35 treatment on induction of caspase‐8 activation. Cells were treated with 20 μM Aβ25‐35 for 4 h. Cells were collected and the levels of active form of caspase‐8 were analyzed by immunoblotting. Densitometric analysis of the level of active caspase‐8 was normalized to the level of actin (product: actin, ratio from each treatment is further compared with WT control). The data represent the mean ±SEM of at least three independent experiments. **P<0.01 when compared with WT and ***P<0.001 when compared with WT.
Vec WT H63D β
Vec A
β
WT Aβ
H63D A
0.0
0.5
1.0
***** **
% D
iffer
ence
128
Figure 3.7: The effect of Aβ25‐35 treatment on apoptosis. Cells were treated with 20 μM Aβ25‐35 for 24 hr as described in Materials and Methods. Cells were collected and apoptosis assay were analyzed by using Annexin V‐FITC/PI flow cytometry assay. The data represent the mean ±SEM of at least three independent experiments.*P<0.05 when compared with WT; **P<0.01 when compared with WT; ***P<0.001 when compared with WT.
Vec WT
H63D β
Vec A
β
WT A
β
H63D A
0
10
20
30
40
50
60
70
80
******
**
Early
Apo
ptos
is (%
)
129
References
Alizadeh BZ, Njajou OT, Millan MR, Hofman A, Breteler MM, van Duijn CM (2009) HFE
variants, APOE and Alzheimerʹs disease: findings from the population‐based
Rotterdam study. Neurobiol Aging 30:330‐332.
Allen JW, Eldadah BA, Huang X, Knoblach SM, Faden AI (2001) Multiple caspases are
involved in beta‐amyloid‐induced neuronal apoptosis. J Neurosci Res 65:45‐53.
Anandatheerthavarada HK, Biswas G, Robin MA, Avadhani NG (2003) Mitochondrial
targeting and a novel transmembrane arrest of Alzheimerʹs amyloid precursor
protein impairs mitochondrial function in neuronal cells. J Cell Biol 161:41‐54.
Avila‐Gomez IC, Jimenez‐Del‐Rio M, Lopera‐Restrepo F, Velez‐Pardo C (2008)
Association between HFE 187 C>G (H63D) mutation and early‐onset familial
Alzheimerʹs disease PSEN‐1 839A>C (E280A) mutation. Ann Hematol 87:671‐673.
Berg D, Hoggenmuller U, Hofmann E, Fischer R, Kraus M, Scheurlen M, Becker G (2000)
The basal ganglia in haemochromatosis. Neuroradiology 42:9‐13.
Berlin D, Chong G, Chertkow H, Bergman H, Phillips NA, Schipper HM (2004)
Evaluation of HFE (hemochromatosis) mutations as genetic modifiers in sporadic
AD and MCI. Neurobiol Aging 25:465‐474.
Beurel E, Jope RS (2006) The paradoxical pro‐ and anti‐apoptotic actions of GSK3 in the
intrinsic and extrinsic apoptosis signaling pathways. Prog Neurobiol 79:173‐189.
Bishop GM, Robinson SR, Liu Q, Perry G, Atwood CS, Smith MA (2002) Iron: a
pathological mediator of Alzheimer disease? Dev Neurosci 24:184‐187.
130
Candore G, Licastro F, Chiappelli M, Franceschi C, Lio D, Rita Balistreri C, Piazza G,
Colonna‐Romano G, Grimaldi LM, Caruso C (2003) Association between the HFE
mutations and unsuccessful ageing: a study in Alzheimerʹs disease patients from
Northern Italy. Mech Ageing Dev 124:525‐528.
Cardoso SM, Santos S, Swerdlow RH, Oliveira CR (2001) Functional mitochondria are
required for amyloid beta‐mediated neurotoxicity. Faseb J 15:1439‐1441.
Chan SL, Mattson MP (1999) Caspase and calpain substrates: roles in synaptic plasticity
and cell death. J Neurosci Res 58:167‐190.
Combarros O, Garcia‐Roman M, Fontalba A, Fernandez‐Luna JL, Llorca J, Infante J,
Berciano J (2003) Interaction of the H63D mutation in the hemochromatosis gene
with the apolipoprotein E epsilon 4 allele modulates age at onset of Alzheimerʹs
disease. Dement Geriatr Cogn Disord 15:151‐154.
Connor JR, Lee SY (2006) HFE mutations and Alzheimerʹs disease. J Alzheimers Dis
10:267‐276.
Csordas G, Thomas AP, Hajnoczky G (1999) Quasi‐synaptic calcium signal transmission
between endoplasmic reticulum and mitochondria. Embo J 18:96‐108.
Desagher S, Martinou JC (2000) Mitochondria as the central control point of apoptosis.
Trends Cell Biol 10:369‐377.
Devi L, Prabhu BM, Galati DF, Avadhani NG, Anandatheerthavarada HK (2006)
Accumulation of amyloid precursor protein in the mitochondrial import
131
channels of human Alzheimerʹs disease brain is associated with mitochondrial
dysfunction. J Neurosci 26:9057‐9068.
Duchen MR (2000) Mitochondria and calcium: from cell signalling to cell death. J Physiol
529 Pt 1:57‐68.
Earnshaw WC, Martins LM, Kaufmann SH (1999) Mammalian caspases: structure,
activation, substrates, and functions during apoptosis. Annu Rev Biochem
68:383‐424.
Fan TJ, Han LH, Cong RS, Liang J (2005) Caspase family proteases and apoptosis. Acta
Biochim Biophys Sin (Shanghai) 37:719‐727.
Ferguson M, Mockett RJ, Shen Y, Orr WC, Sohal RS (2005) Age‐associated decline in
mitochondrial respiration and electron transport in Drosophila melanogaster.
Biochem J 390:501‐511.
Ferreiro E, Oliveira CR, Pereira CM (2008) The release of calcium from the endoplasmic
reticulum induced by amyloid‐beta and prion peptides activates the
mitochondrial apoptotic pathway. Neurobiol Dis 30:331‐342.
Ferreiro E, Eufrasio A, Pereira C, Oliveira CR, Rego AC (2007) Bcl‐2 overexpression
protects against amyloid‐beta and prion toxicity in GT1‐7 neural cells. J
Alzheimers Dis 12:223‐228.
Fifre A, Sponne I, Koziel V, Kriem B, Yen Potin FT, Bihain BE, Olivier JL, Oster T, Pillot
T (2006) Microtubule‐associated protein MAP1A, MAP1B, and MAP2 proteolysis
132
during soluble amyloid beta‐peptide‐induced neuronal apoptosis. Synergistic
involvement of calpain and caspase‐3. J Biol Chem 281:229‐240.
Gervais FG, Xu D, Robertson GS, Vaillancourt JP, Zhu Y, Huang J, LeBlanc A, Smith D,
Rigby M, Shearman MS, Clarke EE, Zheng H, Van Der Ploeg LH, Ruffolo SC,
Thornberry NA, Xanthoudakis S, Zamboni RJ, Roy S, Nicholson DW (1999)
Involvement of caspases in proteolytic cleavage of Alzheimerʹs amyloid‐beta
precursor protein and amyloidogenic A beta peptide formation. Cell 97:395‐406.
Guerreiro RJ, Bras JM, Santana I, Januario C, Santiago B, Morgadinho AS, Ribeiro MH,
Hardy J, Singleton A, Oliveira C (2006) Association of HFE common mutations
with Parkinsonʹs disease, Alzheimerʹs disease and mild cognitive impairment in
a Portuguese cohort. BMC Neurol 6:24.
Hardy J, Selkoe DJ (2002) The amyloid hypothesis of Alzheimerʹs disease: progress and
problems on the road to therapeutics. Science 297:353‐356.
Huang X, Cuajungco MP, Atwood CS, Moir RD, Tanzi RE, Bush AI (2000) Alzheimerʹs
disease, beta‐amyloid protein and zinc. J Nutr 130:1488S‐1492S.
Jellinger KA (2002) Alzheimer disease and cerebrovascular pathology: an update. J
Neural Transm 109:813‐836.
Jellinger KA, Bancher C (1998) Neuropathology of Alzheimerʹs disease: a critical update.
J Neural Transm Suppl 54:77‐95.
133
Kluck RM, Bossy‐Wetzel E, Green DR, Newmeyer DD (1997) The release of cytochrome
c from mitochondria: a primary site for Bcl‐2 regulation of apoptosis. Science
275:1132‐1136.
Kroemer G, Dallaporta B, Resche‐Rigon M (1998) The mitochondrial death/life regulator
in apoptosis and necrosis. Annu Rev Physiol 60:619‐642.
Kuperstein F, Yavin E (2003) Pro‐apoptotic signaling in neuronal cells following iron
and amyloid beta peptide neurotoxicity. J Neurochem 86:114‐125.
Lee KY, Koh SH, Noh MY, Park KW, Lee YJ, Kim SH (2007a) Glycogen synthase kinase‐
3beta activity plays very important roles in determining the fate of oxidative
stress‐inflicted neuronal cells. Brain Res 1129:89‐99.
Lee SY, Patton SM, Henderson RJ, Connor JR (2007b) Consequences of expressing
mutants of the hemochromatosis gene (HFE) into a human neuronal cell line
lacking endogenous HFE. Faseb J 21:564‐576.
Li P, Nijhawan D, Budihardjo I, Srinivasula SM, Ahmad M, Alnemri ES, Wang X (1997)
Cytochrome c and dATP‐dependent formation of Apaf‐1/caspase‐9 complex
initiates an apoptotic protease cascade. Cell 91:479‐489.
Limdi JK, Crampton JR (2004) Hereditary haemochromatosis. Qjm 97:315‐324.
Linseman DA, Butts BD, Precht TA, Phelps RA, Le SS, Laessig TA, Bouchard RJ, Florez‐
McClure ML, Heidenreich KA (2004) Glycogen synthase kinase‐3beta
phosphorylates Bax and promotes its mitochondrial localization during neuronal
apoptosis. J Neurosci 24:9993‐10002.
134
Loo DT, Copani A, Pike CJ, Whittemore ER, Walencewicz AJ, Cotman CW (1993)
Apoptosis is induced by beta‐amyloid in cultured central nervous system
neurons. Proc Natl Acad Sci U S A 90:7951‐7955.
Lovell MA, Robertson JD, Teesdale WJ, Campbell JL, Markesbery WR (1998) Copper,
iron and zinc in Alzheimerʹs disease senile plaques. J Neurol Sci 158:47‐52.
Lustbader JW et al. (2004) ABAD directly links Abeta to mitochondrial toxicity in
Alzheimerʹs disease. Science 304:448‐452.
Miranda S, Opazo C, Larrondo LF, Munoz FJ, Ruiz F, Leighton F, Inestrosa NC (2000)
The role of oxidative stress in the toxicity induced by amyloid beta‐peptide in
Alzheimerʹs disease. Prog Neurobiol 62:633‐648.
Moalem S, Percy ME, Andrews DF, Kruck TP, Wong S, Dalton AJ, Mehta P, Fedor B,
Warren AC (2000) Are hereditary hemochromatosis mutations involved in
Alzheimer disease? Am J Med Genet 93:58‐66.
Narita M, Shimizu S, Ito T, Chittenden T, Lutz RJ, Matsuda H, Tsujimoto Y (1998) Bax
interacts with the permeability transition pore to induce permeability transition
and cytochrome c release in isolated mitochondria. Proc Natl Acad Sci U S A
95:14681‐14686.
Nicotera P, Leist M, Manzo L (1999) Neuronal cell death: a demise with different shapes.
Trends Pharmacol Sci 20:46‐51.
135
Nielsen JE, Jensen LN, Krabbe K (1995) Hereditary haemochromatosis: a case of iron
accumulation in the basal ganglia associated with a parkinsonian syndrome. J
Neurol Neurosurg Psychiatry 59:318‐321.
Nijhawan D, Honarpour N, Wang X (2000) Apoptosis in neural development and
disease. Annu Rev Neurosci 23:73‐87.
Percy M et al. (2008) Involvement of ApoE E4 and H63D in sporadic Alzheimerʹs disease
in a folate‐supplemented Ontario population. J Alzheimers Dis 14:69‐84.
Perier C, Bove J, Wu DC, Dehay B, Choi DK, Jackson‐Lewis V, Rathke‐Hartlieb S,
Bouillet P, Strasser A, Schulz JB, Przedborski S, Vila M (2007) Two molecular
pathways initiate mitochondria‐dependent dopaminergic neurodegeneration in
experimental Parkinsonʹs disease. Proc Natl Acad Sci U S A 104:8161‐8166.
Pike CJ, Walencewicz AJ, Glabe CG, Cotman CW (1991) In vitro aging of beta‐amyloid
protein causes peptide aggregation and neurotoxicity. Brain Res 563:311‐314.
Pulliam JF, Jennings CD, Kryscio RJ, Davis DG, Wilson D, Montine TJ, Schmitt FA,
Markesbery WR (2003) Association of HFE mutations with neurodegeneration
and oxidative stress in Alzheimerʹs disease and correlation with APOE. Am J
Med Genet B Neuropsychiatr Genet 119B:48‐53.
Richardson JS, Zhou Y, Kumar U (1996) Free radicals in the neurotoxic actions of beta‐
amyloid. Ann N Y Acad Sci 777:362‐367.
136
Rizzuto R, Pinton P, Carrington W, Fay FS, Fogarty KE, Lifshitz LM, Tuft RA, Pozzan T
(1998) Close contacts with the endoplasmic reticulum as determinants of
mitochondrial Ca2+ responses. Science 280:1763‐1766.
Rottkamp CA, Raina AK, Zhu X, Gaier E, Bush AI, Atwood CS, Chevion M, Perry G,
Smith MA (2001) Redox‐active iron mediates amyloid‐beta toxicity. Free Radic
Biol Med 30:447‐450.
Sampietro M, Caputo L, Casatta A, Meregalli M, Pellagatti A, Tagliabue J, Annoni G,
Vergani C (2001) The hemochromatosis gene affects the age of onset of sporadic
Alzheimerʹs disease. Neurobiol Aging 22:563‐568.
Schenck JF, Zimmerman EA (2004) High‐field magnetic resonance imaging of brain iron:
birth of a biomarker? NMR Biomed 17:433‐445.
Sharpe JC, Arnoult D, Youle RJ (2004) Control of mitochondrial permeability by Bcl‐2
family members. Biochim Biophys Acta 1644:107‐113.
Slee EA, Harte MT, Kluck RM, Wolf BB, Casiano CA, Newmeyer DD, Wang HG, Reed
JC, Nicholson DW, Alnemri ES, Green DR, Martin SJ (1999) Ordering the
cytochrome c‐initiated caspase cascade: hierarchical activation of caspases‐2, ‐3, ‐
6, ‐7, ‐8, and ‐10 in a caspase‐9‐dependent manner. J Cell Biol 144:281‐292.
Sola S, Diogenes MJ, Brites D, Rodrigues CM (2002) [Release of cytochrome C with the
interaction of bilirubin, amyloid beta‐peptide and glycochenodeoxycholate from
isolated mitochondria]. Acta Med Port 15:269‐275.
137
Sullivan PG, Brown MR (2005) Mitochondrial aging and dysfunction in Alzheimerʹs
disease. Prog Neuropsychopharmacol Biol Psychiatry 29:407‐410.
Trojanowski JQ, Lee VM (2000) ʺFatal attractionsʺ of proteins. A comprehensive
hypothetical mechanism underlying Alzheimerʹs disease and other
neurodegenerative disorders. Ann N Y Acad Sci 924:62‐67.
Vila M, Jackson‐Lewis V, Vukosavic S, Djaldetti R, Liberatore G, Offen D, Korsmeyer SJ,
Przedborski S (2001) Bax ablation prevents dopaminergic neurodegeneration in
the 1‐methyl‐ 4‐phenyl‐1,2,3,6‐tetrahydropyridine mouse model of Parkinsonʹs
disease. Proc Natl Acad Sci U S A 98:2837‐2842.
Wallis LI, Paley MN, Graham JM, Grunewald RA, Wignall EL, Joy HM, Griffiths PD
(2008) MRI assessment of basal ganglia iron deposition in Parkinsonʹs disease. J
Magn Reson Imaging 28:1061‐1067.
Weidemann A, Paliga K, Durrwang U, Reinhard FB, Schuckert O, Evin G, Masters CL
(1999) Proteolytic processing of the Alzheimerʹs disease amyloid precursor
protein within its cytoplasmic domain by caspase‐like proteases. J Biol Chem
274:5823‐5829.
Yankner BA, Dawes LR, Fisher S, Villa‐Komaroff L, Oster‐Granite ML, Neve RL (1989)
Neurotoxicity of a fragment of the amyloid precursor associated with
Alzheimerʹs disease. Science 245:417‐420.
Yuan J, Yankner BA (2000) Apoptosis in the nervous system. Nature 407:802‐809.
138
Zecca L, Youdim MB, Riederer P, Connor JR, Crichton RR (2004) Iron, brain ageing and
neurodegenerative disorders. Nat Rev Neurosci 5:863‐873.
Zong WX, Lindsten T, Ross AJ, MacGregor GR, Thompson CB (2001) BH3‐only proteins
that bind pro‐survival Bcl‐2 family members fail to induce apoptosis in the
absence of Bax and Bak. Genes Dev 15:1481‐1486.
Zou H, Li Y, Liu X, Wang X (1999) An APAF‐1.cytochrome c multimeric complex is a
functional apoptosome that activates procaspase‐9. J Biol Chem 274:11549‐11556.
Chapter 4
Expression of the HFE Allelic Variant H63D in SH-SY5Y Cells Affects Tau
Phosphorylation
Abstract
A number of genetic association studies have appeared that address HFE gene variants
in neurodegenerative disorders. However, the cellular impact of HFE in the nervous system has
received little attention. To begin to address the role of the HFE allelic variants on cellular
events associated with neurodegeneration, we examined the hypothesis that HFE
polymorphisms are associated with alterations in tau phosphorylation in a human
neuroblastoma cell line (SH‐SY5Y). The results show that in a cell culture model, the H63D
allele is associated with increased tau phosphorylation. The mechanisms responsible for these
changes appear related to increased glycogen synthase kinase (GSK)‐3β activity. GSK‐3β
activity is up‐regulated in the cells expressing H63D HFE and can be modified by the addition
of iron or treatment with an iron chelator in SH‐SY5Y cells expressing wild type HFE. Oxidative
stress, also associated with elevated cellular iron, is associated with increased tau
phosphorylation at the same sites as seen in H63D cells and treatment with Trolox, an anti‐
oxidant, lowered tau phosphorylation. These results suggest H63D HFE increases tau
phosphorylation via GSK‐3β activity and iron‐mediated oxidative stress.
140
Introduction
Numerous neurodegenerative diseases are associated with loss of brain iron
homeostasis (Zecca et al., 2004). An imbalance of iron leads to an increase in oxidative stress
which has been tied to cell death and synapse loss (Bush, 2003). Mutations that affect proteins
that regulate cellular iron uptake can be expected to increase oxidative stress and impact
neurodegeneration. Recently, there has been substantial interest in the HFE gene, specifically
the H63D variant as a risk factor or disease modifier in many neurodegenerative diseases
including Alzheimer’s disease (AD) (reviewed in Connor and Lee, 2006; Percy et al., 2008),
amyotrophic lateral sclerosis (ALS) (Wang et al., 2004; Yen et al., 2004; Goodall et al., 2005;
Restagno et al., 2007; Sutedja et al., 2007), Parkinson’s disease (PD) (Borie et al., 2002; Dekker et
al., 2003; Guerreiro et al., 2006) and ischemic stroke (Ellervik et al., 2007). HFE has been studied
within the paradigm of the iron overload disorder hereditary hemochromatosis, but the
frequency of HFE gene variants are much more prevalent than this disease (Waalen et al., 2005).
The HFE protein has been shown to form a complex with transferrin receptors (TfR) at the
cellular membrane and decrease its affinity for transferrin (Tf) resulting in only one transferrin
complexing with the transferrin receptor and thus limiting cellular iron uptake (Feder et al.,
1998). The mutant form of the protein (H63D, rs1799945; www.alzgene.org), fails to complex
with the TfR and subsequently is associated with increased cellular iron (Feder et al., 1998).
Thus, even in the absence of clinical disease, we propose that chronic subclinical iron loading at
the cellular level over many years can provide a permissive milieu that enables
neurodegenerative disease processes, and thus the HFE mutations are a disease modifier. In
141
addition, the function of HFE may expand beyond limiting iron uptake or there may be a gain
of function associated with the mutant protein (Lee et al., 2006).
As mentioned, a host of epidemiological studies have investigated the genetic
association of HFE polymorphisms and AD, ALS, and PD. A consistent underlying pathological
characteristic of these diseases is the hyperphosphorylation of tau. Tau is a microtubule
associated protein primarily responsible for binding microtubules such as tubulin to stabilize
the neuronal cytoskeleton (Lee et al., 2001; Iqbal and Grundke‐Iqbal, 2006). Cytoskeletal
stabilization is vital to axonal transport ensuring adequate neuronal communication. The
hyperphosphorylation of tau by various kinases, primarily glycogen synthase kinase 3β (GSK‐
3β), cyclin‐dependant kinase 5 (Cdk5), and extracellular regulating kinase 2 (ERK‐2) is thought
to be responsible for the pathological lesions containing hyperphosphorylated tau seen in AD,
ALS, and PD (Lee et al., 2001; Mazanetz and Fischer, 2007). Protein phosphatases PP1, PP2A,
and PP2B are responsible for tau dephosphorylation (Lee et al., 2001). Thus, the opposing
actions of kinases and phosphatases provide a balance to achieve a tau phosphorylation
homeostasis.
The impact of iron status, and specifically the presence of the HFE gene variants on this
homeostatic mechanism, may provide insights into how the HFE gene variants contribute to
neurodegenerative disease. Although it has traditionally been thought that the brain is
protected from iron overload, HFE protein is expressed on the blood vessels, choroid plexus
and ependymal surface in the brain and is therefore in a position to impact iron uptake into the
brain (Connor et al., 2001). Patients expressing an HFE polymorphism were found to have a
142
greater amount of oxidative stress determined by their cerebrospinal fluid (CSF) F2‐isoprostane
levels (Pulliam et al., 2003) and patients with ALS who have an HFE mutation have a different
CSF biomarker profile than those with wild type HFE (Mitchell et al., 2008). Also, magnetic
resonance imaging (MRI) has shown increased brain iron levels in patients possessing HFE
polymorphisms (Bartzokis et al., 2006, 2007). Thus, given the frequency of HFE polymorphisms
in HFE and neurodegenerative diseases, the presence of HFE in the brain, and the recent data
that HFE mutations alter brain iron status and oxidative stress, the development of a cellular
model to interrogate the effects of HFE gene variants is timely.
We have developed a stably transfected HFE neuroblastoma SH‐SY5Y cell line to test the
hypothesis that the H63D polymorphism will increase tau phosphorylation. There are
numerous indirect mechanisms that have been shown to influence tau hyperphosphorylation
including, but not limited to: oxidative stress and glutamate toxicity, both of which we have
shown to be altered in our cellular model (Wang et al., 2004; Lee et al., 2006). Thus, in the study
herein, we sought to directly examine tau phosphorylation in our cell model and the associated
mechanisms.
Materials and methods
Reagents ‐ Cell culture reagents including DMEM/F12, DMEM, pen/strep/glutamine and
Geneticin were purchased from Invitrogen (Carlsbad, CA, USA). Fetal bovine serum was
purchased from Gemini Bio‐Products (West Sacramento, CA, USA). DC protein assay was
obtained from Bio‐Rad (Hercules, CA, USA). Mouse monoclonal Tau (Tau‐5) and rabbit
143
polyclonal anti‐tau (serine 404) was purchased from Biosource International (Camarillo, CA,
USA). Mouse monoclonal PHF‐Tau (AT8; serine 202) was purchased from Pierce Biotechnology
(Rockford, IL, USA). Rabbit polyclonal AKT was purchased from Cell Signaling Technology
(Beverly, MA, USA). Mouse monoclonal Cdk5 (J‐3), rabbit polyclonal p35/p25 (C‐19), PP1 (E‐9),
and PP2A (FL‐309) were purchased from Santa Cruz Biotechnology (Santa Cruz, CA, USA).
Rabbit polyclonal ERK1/2 and rabbit polyclonal phospho‐ERK1/2 was purchased from
Chemicon International (Temecula, CA, USA). Horseradish peroxidase‐conjugated secondary
antibodies were purchased from GE Healthcare (Princeton, NJ). All other reagents were
purchased from Sigma‐Aldrich (St. Louis, MO).
Cell culture ‐ Human neuroblastoma SH‐SY5Y cell lines were obtained from American Type
Culture Collection (Manassas, VA, USA). Human neuroblastoma SH‐SY5Y cells were stably
transfected to express wild‐type and H63D HFE forms as previously reported along with a
vector alone control (Wang et al., 2004). We have previously reported that these cells were
chosen because endogenous expression of HFE in these cells could not be detected. The
transfected cells were maintained in DMEM/F12 media supplemented with 10% FBS, 1%
antibiotics (pen‐strep‐glutamine), 1x nonessential amino acids, and 1.8g/L sodium bicarbonate.
Cells were differentiated with 10 μM all‐trans retinoic acid (Sigma‐Aldrich) over six days
(Haque et al., 1999). To evaluate cellular iron effects, cells were treated with ferric ammonium
citrate (FAC) or desferrioxamine (DFO) over 48 hours (Lee et al., 2006).
144
Enzyme linked immunoabsorbant assay (ELISA) ‐ Cells were lysed with RIPA buffer supplemented
with 1% Triton X‐100 and protease inhibitor cocktail (Sigma Aldrich, St. Louis, MO).
Phosphatase inhibitor (Sigma Aldrich, St. Louis, MO) was included in cell lysis buffer for
phosphorylation protein detection. Cells extracts were spun at 8,000 × g for 10 min. Total protein
levels were determined by Bio‐Rad DC protein assay. A monoclonal antibody specific for
human tau phosphorylated residues, total tau, total GSK‐3β, and GSK‐3β (ser‐9) was coated
onto the wells of the microtiter strips provided (Invitrogen (BioSource)). Standards of known
total and phosphorylated proteins were processed to achieve a standard curve to determine the
specific amount of total or phosphorylated protein in the unknown HFE cell samples. Twenty‐
five μg total protein of the unknown HFE samples were diluted 1:100 and 1:1000 to achieve the
desired linear range of specific standard protein concentration (pg/ml) as per the
manufacturer’s recommendation. The ELISA assay plates were read at 450nm. All ELISA
experiments were performed using samples in triplicate per genotype and/ or per treatment at
two dilution concentrations along with the known standards for the specific proteins, resulting
in a total of six samples for analysis.
Western blot ‐ Cells lysates were obtained as described above. Twenty‐five to forty μg total
protein was equally separated by electrophoresis in a 4‐20% 12‐well Criterion gel (Bio‐Rad,
Hercules, CA). Protein was then transferred to a nitrocellulose membrane and blocked for 1 hr
at room temperature in TBS‐T with 5% nonfat milk or 1.5% BSA (phosphorylated protein
detection). Membranes were probed with primary antibodies in TBS‐T with 5% nonfat milk
145
overnight at 4°C. The membranes were incubated respectively with Tau‐5 (1:1000), Tau AT8
(1:500), Tau pS404 (1:500), Cdk5 (1:1000), p35/p25 (1:1000), AKT (1:1000), ERK1/2 (1:1000),
pERK1/2 (1:1000), PP1 (1:1000), and PP2A (1:1000). HRP‐conjugated secondary antibodies were
added in 5% nonfat milk for 1 hr at room temperature. Protein signals were obtained by
chemiluminescence and visualized by CCD camera. All western blot experiments were repeated
at least twice with a minimum of four different cultures per genotype per experiment, resulting
in a total of eight samples for analysis. The bands on the western blot were quantified by
densitometry using Fuji MultiGauge analysis software.
Statistical analysis ‐ The Student’s t‐test was used for analyzing HFE variant comparisons when
one variable was being determined. Experimental data where samples were treated with
various agents (i.e. iron, desferrioxamine, or Trolox) and compared with controls were analyzed
by one‐way analysis of variance. If overall p <0.05, Tukey’s Multiple Comparison post hoc
analysis was performed. Differences among the means were considered statistically significant
when the p value was <0.05. Data are presented as the mean ± S.E.; GraphPad Prism software
(version 4.0) was utilized to perform the statistical analysis.
Results
To begin to determine if there is an association of HFE variants with changes in tau and
tau phosphorylation, we determined the expression levels of total tau and phosphorylated tau
in HFE variant cells. No significant differences were detected in total tau protein among H63D‐
146
expressing cells compared to the wild‐type and vector controls (Figure 4.1A). Subsequently, the
major tau phosphorylation sites related to neurofibrillary tangle formation (Lee et al., 2001;
Mazanetz and Fischer, 2007): Thr‐181, Serine‐199, Serine‐202, Thr‐205, Serine‐214, Thr‐231,
Serine‐396, and Serine‐404 were evaluated to obtain a tau phosphorylation profile in cells
expressing the H63D allele.
As shown in Figure 4.1B, tau phosphorylation was unchanged at Serine‐214 between the
cells expressing the wild type or H63D allele. Phosphorylation at the Thr‐181 and Thr‐231 sites
was significantly decreased (p< 0.05) in cells expressing H63D versus wild‐type HFE‐expressing
cells. An increase in tau phosphorylation at the Serine‐199 (p < 0.01), Serine‐202 (p< 0.01),
Serine‐396 (p< 0.001), and Serine‐404 (p<0.05) sites was present in H63D expressing cells
compared to wild‐type expressing cells (Figures 4.1B and 4.1C).
Based on the increase of phosphorylation at the majority of tau phosphorylation sites in
H63D‐expressing cells, we sought to determine the direct mechanisms responsible for these
observations. Protein kinase and phosphatase activity are the primary regulators for
maintaining equilibrium of tau phosphorylation (Lee et al., 2001; Jamsa et al., 2004; Ballatore et
al., 2007; Mazanetz and Fischer, 2007). Therefore, we measured the protein expression and
activity of Cdk5, ERK‐1 and ERK‐2, and GSK‐3β along with phosphatases PP2A and PP1.
Cdk5 protein levels are decreased in H63D cells (p<0.05) as well as its activity (p<0.01)
based on p25 expression levels compared to wild‐type cells (Figure 4.2A). ERK‐2 protein levels,
but not ERK‐1, were increased in H63D‐expressing cells compared to wild‐type cells (p<0.05)
(Figure 4.2A). The activity of ERK‐1 and ERK‐2 was not affected by either the expression of
147
wild‐type or H63D‐expressing forms of HFE (Figure 4.2A). Total GSK‐3β expression levels were
unaffected by the expression of H63D compared with wild‐type cells (Figure 4.2B), but GSK‐3β
activity was increased (p<0.05) based on the dephosphorylation of the serine‐9 residue of GSK‐
3β (Figure 4.2B). Because, GSK‐3β phosphorylation and hence activation is inhibited by AKT,
AKT levels were determined and found to be decreased in H63D cells (p<0.05), which is
consistent with GSK‐3β activity in H63D cells (Figure 4.2C). Protein phosphatase expression
(PP2A and PP1) levels were unchanged in the presence of either HFE allele (Figure 4.2D). PP2B
levels have not been detected in this cell line (Haque et al., 1999).
To validate that increased GSK‐3β activity is responsible for the increased tau
phosphorylation in H63D‐expressing cells, we treated the cells with lithium chloride to decrease
GSK‐3β activity. Lithium treatment decreased tau phosphorylation at the serine‐199 and serine‐
202 in both wild‐type and H63D cells (Figure 4.3). Moreover, tau serine‐396 site was
significantly decreased in H63D cells only (Figure 4.3). Thus, it appears that GSK‐3β is
influencing three of the four tau phosphorylation increases seen in H63D variant cells.
The data on the H63D cells are compelling that GSK‐3β activation is the primary
mechanism for increasing tau phosphorylation at the Serine‐199, Serine‐202, and Serine‐396
sites. To elucidate the consequence of the H63D allele on GSK‐3β activity; the effect of changing
cellular iron status on GSK‐3β was determined. To mimic the effect of increased intracellular
iron with the H63D allele (Lee et al., 2006), wild‐type expressing cells were treated with ferric
ammonium citrate (FAC). Addition of FAC was associated with an increase in GSK‐3β activity
as evidenced by decreased GSK‐3β phosphorylation, 10μM (p<0.01) (Figure 4.4A). To validate
148
our finding that GSK‐3β activity is iron sensitive, we treated the H63D‐expressing cells with an
iron chelator, desferrioxamine (DFO), which produced a dose‐dependent decrease in GSK‐3β
activity, 5μM (p<0.01) and 10μM (p<0.001), (Figure 4.4A).
The differences in baseline tau phosphorylation between the cells expressing H63D and
wt HFE raised the possibility that iron exposure, as an environmental agent, could lead to even
greater levels of tau phosphorylation in the H63D cells. Therefore, we measured tau
phosphorylation after exposure to iron in the form of FAC. Iron exposure increased tau
phosphorylation at serine‐199, serine‐396, serine‐202, and serine‐404 (Figure 4.4B). The majority
of tau phosphorylation sites in wild‐type expressing cells required exposure of 30μM FAC to
reach the baseline level of tau phosphorylation seen in the H63D‐expressing cells. The already
elevated levels of tau phosphorylation in the untreated H63D cells increased by 29 and 33% at
the serine 199 and serine 396 sites, respectively with further FAC exposure (Figure 4.4B). The
data from the wt cells indicate that the increases in tau phosphorylation in H63D‐expressing
cells can be attributed to increased cellular iron levels leading to increased GSK‐3β activity.
Increases in intracellular iron are commonly associated with increased oxidative stress,
and we have reported an increase in oxidative stress in the cells expressing the H63D allele
(Wang et al., 2004; Lee et al., 2006). Thus, an additional series of experiments were designed to
determine if oxidative stress can contribute to the tau phosphorylation changes seen based on
HFE genotype. Hence, we treated the cells with Trolox, a vitamin E analog to decrease cellular
oxidative stress. Trolox treatment was associated with decreased tau phosphorylation at serine‐
199 (p<0.05), serine‐202 (p<0.05), and serine‐396 (p<0.01) (Figure 4.5). Tau phosphorylation at
149
serine‐404 was not affected by Trolox treatment. These data further support the concept of
increased oxidative stress in H63D‐expressing cells, leading to increased tau phosphorylation.
Discussion
H63D gene variants are among the most common gene variants in the human
population; with estimates of frequency ranging from 6 to 26% differing by population ethnicity
(Adams et al., 2005). Therefore, if the presence of the HFE H63D variant alters baseline cell
stress levels, then disease agents, whether genetic or environmental, can expect to find a
permissive milieu for initiation and progression of disease processes in a significant percentage
of the general population. Because of the multiple sources of iron in the environment, including
dietary exposure, the study of HFE polymorphisms and adult onset neurodegenerative disease
is an attractive model for gene and environment interaction.
Tau hyperphosphorylation is a common pathological pathway in neurodegenerative
disorders reported to have an association with HFE gene variants which provided the impetus
for the cellular studies herein. There is an overall increase in total tau with specific increases in
tau phosphorylation at the serine‐199, serine‐202, serine‐396 and serine‐404 sites in cells
expressing the H63D gene variant compared to wild type. GSK‐3β and Cdk5 are the primary
kinases responsible for these phosphorylation sites mentioned. Therefore, GSK‐3β and Cdk5
expression and activity was evaluated as a possible mechanism for the increased tau
phosphorylation. Neither GSK‐3β nor Cdk5 total protein expression was increased; but GSK‐3β
activity was increased as determined by decreased phosphorylation at its serine 9 residue
(Grimes and Jope, 2001). The increase in GSK‐3β activity is consistent with increased tau
150
phosphorylation at the serine‐199, serine‐202, and serine‐396 residues. Cdk‐5 activity was
decreased as evidenced by decreased expression of its cofactor, p25. The decrease in Cdk‐5
activity may explain the decreased tau phosphorylation at threonine‐181 and threonine‐231
because Cdk‐5 is an important priming kinase for these residues (Li et al., 2006; Sengupta et al.,
2006).
Furthermore, Cdk5 has been shown to inhibit GSK‐3β activity (Plattner et al., 2006).
Thus, the data from our model are consistent and suggest that decreased expression of Cdk5
may allow for increased GSK‐3β activity. To further support that increased GSK‐3β activity
caused the increased tau phosphorylation in H63D expressing cells, we treated the cells with
lithium chloride. Lithium chloride inhibits GSK‐3β activity and subsequently tau
phosphorylation (Jamsa et al., 2004). Indeed this compound has been used clinically in some
neurodegenerative diseases in which tau phosphorylation is part of the pathogenic profile
(Fornai et al., 2008). In our study, lithium treatment decreased tau phosphorylation at the
serine‐199, serine‐202, and serine‐396 sites.
The primary phenotype of expressing the H63D allele is thought to be an increase in the
labile iron pool (Feder et al., 1998; Lee et al., 2006) and an associated potential increase in
oxidative stress (Pulliam et al., 2003; Lee et al., 2006); both of which occur in our model.
Therefore, we determined if cellular iron levels and oxidative stress were influencing tau
phosphorylation. Because GSK‐3β dephosphorylation levels and subsequent activation appears
to be the primary direct mechanism responsible for increased tau phosphorylation sites seen in
our model, we determined if GSK‐3β expression and activity are iron‐sensitive. Indeed, iron
151
treatment increased GSK‐3β activity and treating the cells with an iron chelator decreased GSK‐
3β activity. GSK‐3β has been shown to be regulated by Cdk5 and a previous study showed that
increasing iron levels are associated with decreased Cdk5 expression (Egana et al., 2003). Thus,
it is consistent with our findings that the H63D expressing cells have higher amounts of labile
iron and increased GSK‐3β activity.
An additional mechanism that can lead to tau hyperphosphorylation is oxidative stress
(Ballatore et al., 2007). Iron imbalance can contribute to cellular oxidative stress primarily
through the classical Fenton chemistry reaction producing hydroxyl radicals (Arosio and Levi,
2002; Zecca et al., 2004). The H63D allelic variant of HFE has been associated with evidence of
increased oxidative stress in patients carrying the mutant allele (Pulliam et al., 2003) and in our
cellular model based on increased protein oxidation and decreased mitochondrial membrane
potential (Lee et al., 2006). Hence, we evaluated if oxidative stress could be contributing to the
increased tau phosphorylation seen in the H63D cells. Treatment with the anti‐oxidant, Trolox,
a vitamin‐E analog; decreased tau phosphorylation indicating that iron‐mediated oxidative
stress could be driving the increase in tau phosphorylation seen in the presence of H63D HFE.
In summary, there is a fundamental difference in the baseline of tau phosphorylation in
cells that carry the H63D variant of the HFE gene. This observation is consistent with the
conceptual framework for this line of research that HFE gene variants are a risk factor or disease
modifier in neurodegenerative diseases. We do not propose that the H63D gene is a causative
agent in neurodegenerative disease, but rather is a major factor in establishing a permissive
intracellular milieu that enables that causative agent. In this case, H63D allelic variants promote
152
tau phosphorylation by nature of increased intracellular iron and oxidative stress. Additionally,
environmental factors that increase iron exposure or oxidative stress could be anticipated to
exacerbate the hyperphosphorylation and effects on cellular enzyme activity related to tau
phosphorylation as was demonstrated in this study by the addition of iron to the H63D cells.
These data further suggest that a combination of environmental factors and the H63D gene
variants could explain in inconsistent findings on the association between H63D and
Alzheimer’s disease. Furthermore, based on the fundamental cell differences in extent of tau
phosphorylation, stratification of patients according to HFE polymorphisms may be useful
when evaluating treatment strategies.
Acknowledgements
Copyrighted material from this chapter appears in the Neurobiology of Aging journal
(Hall et al. Expression of the HFE allelic variant H63D in SH‐SY5Y cells affects tau
phosphorylation at serine residues. Neurobiol Aging 2009).
153
Total Tau
Vector WT H63D0
1.0×105
2.0×105
3.0×105
4.0×105
tau
pg/m
l
Tau pT181
Vector WT H63D0
1.0×104
2.0×104
p< 0.05
p< 0.01
p< 0.001
tau
pg/m
l
Tau pS199
Vector WT H63D0
2.5×104
5.0×104
7.5×104
1.0×105
p< 0.01
tau
pg/m
l
p< 0.05
Tau pT231
Vector WT H63D0
2.0×104
4.0×104
6.0×104
p< 0.05
tau
pg/m
l
p< 0.01
Figure 4.1: Tau Phosphorylation in HFE stably transfected SH‐SY5Y cells. Expression of total tau protein was measured in an HFE polymorphism stably transfected SH‐SY5Y cell line by ELISA (A); no significant differences were found. Specific tau phosphorylation sites were determined by ELISA (B) and western blot (C). This figure shows that tau phosphorylation patterns differ in comparing wild type and H63D HFE cell expression. Tau phosphorylation was found to be increased in H63D cells at the serine‐199 (p<0.01), serine‐202 (p<0.01), serine 396 (p<0.001), and serine 404 (p<0.05) residues. Decreased phosphorylation was found in cells expressing H63D HFE at the threonine‐181 (p<0.05) and threonine‐231 (p<0.05) sites as well. Experiments were performed with a minimum of four different cultures per genotype. Representative western blot images are shown with graphs displaying differences in expression determined by densitometric analysis. One‐way ANOVA was performed to analyze the data, followed by Tukey’s post‐hoc analysis. Data are represented as mean ± S.E. A.
B.
Tau pS214
Vector WT H63D0
2.0×103
4.0×103
6.0×103
p< 0.05p< 0.05
tau
pg/m
l
154
Tau pS404
Vector WT H63D0
4.0×100
8.0×100
p< 0.05p< 0.01
p< 0.001
AU
B.
C.
Tau pS396
Vector WT H63D0
5.0×104
1.0×105
1.5×105
p< 0.001p< 0.01
tau
pg/m
l
pS202
β-actin
Vector WT H63D
pS404
Vector WT H63D
β-actin
Tau pS202
Vector WT H63D0
4.0×105
8.0×105
p< 0.01p< 0.01
AU
155
ERK 1
Vector WT H63D0
4.0×105
8.0×105
1.2×106
1.6×106
AU
ERK 2
Vector WT H63D0
1.0×106
2.0×106
3.0×106 p< 0.05
p< 0.05
AU
Figure 4.2: Kinase and Phosphatase expression in SH‐SY5Y cells. Expression of major kinases shown to impact tau phosphorylation was measured by western blot and ELISA. In figure A, Cdk5 expression (p<0.05) and activity levels based on p25 expression levels (p<0.01) were decreased in cells expressing H63D compared to wild type expressing cells (A). ERK 1 and ERK 2 expression levels where unchanged and increased (p<0.05), respectively. ERK 1 and ERK 2 activity were found to be unchanged based on phosphorylated ERK 1 and ERK 2 expression levels (A). Total GSK‐3β expression levels and GSK‐3β activity determined by dephosphorylation of the serine‐9 site was measured by ELISA (B). Total GSK‐3β levels were unchanged. GSK‐3β serine‐9 levels were decreased in H63D expressing cells compared to wild type (p<0.05) indicating increased GSK‐3β activity. AKT is an upstream kinase of GSK‐3β that has been shown to modulate GSK‐3β activity. Total AKT expression levels were measured by western blot (C). Total AKT levels were decreased (p<0.05) in H63D expressing cells. Phosphatase expression was measured by western blot (D). PP1 and PP2A expression levels were found unchanged. One‐way ANOVA was performed to analyze the data, followed by Tukey’s post‐hoc analysis. Data are represented as mean ± S.E.
A.
Cdk-5
Vector WT H63D
p35
p25
Vector WT H63D
CDK5
Vector WT H63D0
3.0×105
6.0×105
9.0×105
p< 0.05p< 0.001
AU
p25
Vector WT H63D0
1.0×105
2.0×105
3.0×105p< 0.01
p< 0.001
AU
β-actinβ-actin
ERK 1
ERK 2
Vector WT H63D
β-actin
156
pERK1
Vector WT H63D0
5.0×105
1.0×106
AU
pERK2
Vector WT H63D0
5.0×105
1.0×106
AU
A.
B.
C.
pERK 1
pERK 2
Vector WT H63D
β-actin
Total GSK-3beta
Vector WT H63D0.00
0.15
0.30
p< 0.01
OD
450
nm
p< 0.01
GSK-3beta (ser-9)
Vector WT H63D0
1.5×10-1
3.0×10-1
4.5×10-1
p< 0.05
p< 0.01O
D 4
50 n
m
AKT
Vector WT H63D
β-actin
Total AKT
Vector WT H63D0
2.0×106
4.0×106
p< 0.05p< 0.01
AU
157
PP2A
Vector WT H63D0
2.0×106
4.0×106
AU
D.
PP1
Vector WT H63D
β-actin
PP2A
β-actin
Vector WT H63D
PP1
Vector WT H63D0
2.0×106
4.0×106
AU
158
Tau pS202
WTH63
D
WT 10mM
H63D 10
mM
WT 20mM
H63D 20
mM0.0
0.5
1.0
1.5 b
a, ^ ^^ *%
Diff
eren
ce
Tau pS404
WTH63
D
WT 10mM
H63D 10
mM
WT 20mM
H63D 20
mM0.0
0.5
1.0
% D
iffer
ence
Figure 4.3: Inhibition of tau phosphorylation. GSK‐3β inhibition was performed by treating the cells with increasing amounts of lithium chloride. The tau phosphorylation sites found to be increased in H63D cells were treated based on evidence of GSK‐3β being shown to be the primary kinase responsible for increased tau phosphorylation in our cell model system. Lithium treatment decreased tau phosphorylation at the serine‐199, serine‐202, and serine‐396 sites determined by ELISA and western blot. The phosphorylation at the serine‐404 site was unaffected. Lithium treatment decreased tau phosphorylation at the serine‐396 site only in H63D cells. One‐way ANOVA was performed to analyze the data, followed by Tukey’s post‐hoc analysis. Data are represented as mean ± S.E. The letters a (p<0.05), b (p<0.01), and c (p<0.001) indicate significance from baseline wild type HFE. The symbols # (p<0.05), * (p<0.01), and ^ (p<0.001) indicate significance from wild type HFE.
Tau pS199
WT H63
D
WT 10m
M
H63D 10
mM
WT 20m
M
H63D 20
mM0.0
0.5
1.0
1.5
2.0
b, ^
b
*#
a, ^% D
iffer
ence
Tau pS396
WT
H63D
WT 1
0mM
H63D 10
mM
WT 2
0mM
H63D 20
mM0.0
0.5
1.0
1.5
2.0a
c
#
% D
iffer
ence
159
GSK-3beta (ser-9)
WT-contro
l
WT-10µM
iron
WT-30µM
iron
0.0
0.5
1.0
p< 0.01
% D
iffer
ence
Tau pS202
WT
H63D
WT 10
µM
H63D 10
µM
WT 30
µM
H63D 30
µM
0.0
0.5
1.0
ab b
% D
iffer
ence
Figure 4.4: Impact of cellular iron on GSK‐3β activity and tau phosphorylation. We added increasing amounts of ferrous ammonium citrate to cells expressing wild type HFE, and increasing amounts of the iron chelator desferrioxamine (DFO) to cells expressing H63D HFE, and measured GSK‐3β activity (A). GSK‐3 activity increased with iron treatment (p<0.01) and decreased GSK‐3β activity with DFO treatment (p<0.001) determined by ELISA. Tau phosphorylation sites increased with iron treatments (10μM, 30μM) at the serine‐199, serine‐396, serine‐202, and serine‐404 determined by ELISA and western blot (B). One‐way ANOVA was performed to analyze the data, followed by Tukey’s post‐hoc analysis. Data are represented as mean ± S.E. The letters a (p<0.05), b (p<0.01), and c (p<0.001) indicate significance from baseline wild type HFE. The symbols # (p<0.05), * (p<0.01), and ^ (p<0.001) indicate significance from baseline H63D.
A.
B.
GSK-3beta (ser-9)
H63D-co
ntrol
H63D-5µ
M DFO
H63D-10
µM D
FO0.0
0.5
1.0
1.5 p< 0.01
p< 0.001
% D
iffer
ence
Tau pS199
WT
H63D
WT 1
0µM
H63D 10
µM
WT 3
0µM
H63D 30
µM0.0
0.5
1.0
1.5
2.0b
a
c, #c, *
% D
iffer
ence
Tau pS396
WT
H63D
WT 1
0µM
H63D 10
µM
WT 3
0µM
H63D 30
µM0.0
0.5
1.0
1.5
2.0
2.5
3.0
b
c, * c, #
a a
% D
iffer
ence
Tau pS404
WT
H63D
WT 1
0µM
H63D 10
µM
WT 3
0µM
H63D 30
µM0.0
0.5
1.0
1.5a
b, #
% D
iffer
ence
B.
161
Tau pS202
WTH63
D
WT trolox
H63D tro
lox0.0
0.5
1.0
1.5
*
a#
% D
iffer
ence
Tau pS404
WTH63
D
WT trolox
H63D tr
olox0.0
0.5
1.0
1.5a a
% D
iffer
ence
Figure 4.5: Tau Phosphorylation in SH‐SY5Y cells with Trolox treatment. Increased tau phosphorylation sites were treated for 72 hours with 200 μM Trolox, a water‐soluble vitamin E analog to assess the effect of oxidative stress on tau phosphorylation. Trolox treatment decreased tau phosphorylation at serine‐199 (p<0.05), serine‐202 (p<0.05), and serine‐396 (p<0.01) in H63D cells compared to wild‐type determined by ELISA and western blot. One‐way ANOVA was performed to analyze the data, followed by Tukey’s post‐hoc analysis. Data are represented as mean ± S.E. The letters a (p<0.05), b (p<0.01), and c (p<0.001) indicate significance from baseline wild type HFE. The symbols # (p<0.05), * (p<0.01), and ^ (p<0.001) indicate significance from baseline H63D.
Tau pS199
WT
H63D
WT tro
lox
H63D tro
lox0.0
0.5
1.0
1.5a
* #
% D
iffer
ence
Tau pS396
WT
H63D
WT tro
lox
H63D tro
lox0.0
0.5
1.0
1.5a
*#
% D
iffer
ence
162
References
Adams, P. C., D. M. Reboussin, et al. (2005). ʺHemochromatosis and iron‐overload screening in
a racially diverse population.ʺ N Engl J Med 352(17): 1769‐78.
Arosio, P. and S. Levi (2002). ʺFerritin, iron homeostasis, and oxidative damage.ʺ Free Radic Biol
Med 33(4): 457‐63.
Ballatore, C., V. M. Lee, et al. (2007). ʺTau‐mediated neurodegeneration in Alzheimerʹs disease
and related disorders.ʺ Nat Rev Neurosci 8(9): 663‐72.
Bartzokis, G., T. A. Tishler, et al. (2006). ʺBrain ferritin iron may influence age‐ and gender‐
related risks of neurodegeneration.ʺ Neurobiol Aging.
Bartzokis, G., T. A. Tishler, et al. (2007). ʺBrain ferritin iron may influence age‐ and gender‐
related risks of neurodegeneration.ʺ Neurobiol Aging 28(3): 414‐23.
Borie, C., F. Gasparini, et al. (2002). ʺAssociation study between iron‐related genes
polymorphisms and Parkinsonʹs disease.ʺ J Neurol 249(7): 801‐4.
Bush, A. I. (2003). ʺThe metallobiology of Alzheimerʹs disease.ʺ Trends Neurosci 26(4): 207‐14.
Connor, J. R. and S. Y. Lee (2006). ʺHFE mutations and Alzheimerʹs disease.ʺ J Alzheimers Dis
10(2‐3): 267‐76.
Connor, J. R., E. A. Milward, et al. (2001). ʺIs hemochromatosis a risk factor for Alzheimerʹs
disease?ʺ J Alzheimers Dis 3(5): 471‐477.
163
De Sarno, P., X. Li, et al. (2002). ʺRegulation of Akt and glycogen synthase kinase‐3 beta
phosphorylation by sodium valproate and lithium.ʺ Neuropharmacology 43(7): 1158‐64.
Dekker, M. C., P. C. Giesbergen, et al. (2003). ʺMutations in the hemochromatosis gene (HFE),
Parkinsonʹs disease and parkinsonism.ʺ Neurosci Lett 348(2): 117‐9.
Egana, J. T., C. Zambrano, et al. (2003). ʺIron‐induced oxidative stress modify tau
phosphorylation patterns in hippocampal cell cultures.ʺ Biometals 16(1): 215‐23.
Ellervik, C., A. Tybjaerg‐Hansen, et al. (2007). ʺHereditary hemochromatosis genotypes and risk
of ischemic stroke.ʺ Neurology 68(13): 1025‐31.
Feder, J. N., D. M. Penny, et al. (1998). ʺThe hemochromatosis gene product complexes with the
transferrin receptor and lowers its affinity for ligand binding.ʺ Proc Natl Acad Sci U S A
95(4): 1472‐7.
Fornai, F., P. Longone, et al. (2008). ʺLithium delays progression of amyotrophic lateral
sclerosis.ʺ Proc Natl Acad Sci U S A 105(6): 2052‐7.
Goodall, E. F., M. J. Greenway, et al. (2005). ʺAssociation of the H63D polymorphism in the
hemochromatosis gene with sporadic ALS.ʺ Neurology 65(6): 934‐7.
Grimes, C. A. and R. S. Jope (2001). ʺThe multifaceted roles of glycogen synthase kinase 3beta in
cellular signaling.ʺ Prog Neurobiol 65(4): 391‐426.
164
Guerreiro, R. J., J. M. Bras, et al. (2006). ʺAssociation of HFE common mutations with
Parkinsonʹs disease, Alzheimerʹs disease and mild cognitive impairment in a Portuguese
cohort.ʺ BMC Neurol 6: 24.
Haque, N., T. Tanaka, et al. (1999). ʺRegulation of expression, phosphorylation and biological
activity of tau during differentiation in SY5Y cells.ʺ Brain Res 838(1‐2): 69‐77.
Iqbal, K. and I. Grundke‐Iqbal (2006). ʺDiscoveries of tau, abnormally hyperphosphorylated tau
and others of neurofibrillary degeneration: a personal historical perspective.ʺ J
Alzheimers Dis 9(3 Suppl): 219‐42.
Jamsa, A., K. Hasslund, et al. (2004). ʺThe retinoic acid and brain‐derived neurotrophic factor
differentiated SH‐SY5Y cell line as a model for Alzheimerʹs disease‐like tau
phosphorylation.ʺ Biochem Biophys Res Commun 319(3): 993‐1000.
Lee, S. Y., S. M. Patton, et al. (2006). ʺConsequences of expressing mutants of the
hemochromatosis gene (HFE) into a human neuronal cell line lacking endogenous HFE.ʺ
Faseb J.
Lee, V. M., M. Goedert, et al. (2001). ʺNeurodegenerative tauopathies.ʺ Annu Rev Neurosci 24:
1121‐59.
Li, T., C. Hawkes, et al. (2006). ʺCyclin‐dependent protein kinase 5 primes microtubule‐
associated protein tau site‐specifically for glycogen synthase kinase 3beta.ʺ Biochemistry
45(10): 3134‐45.
165
Mazanetz, M. P. and P. M. Fischer (2007). ʺUntangling tau hyperphosphorylation in drug design
for neurodegenerative diseases.ʺ Nat Rev Drug Discov 6(6): 464‐79.
Mitchell, R., W. Freeman, et al. (2008). ʺA CSF Biomarker Panel for Identification of Patients
with Amyotrophic Lateral Sclerosis.ʺ Neurology (in press).
Percy, M., S. Moalem, et al. (2008). ʺInvolvement of ApoE E4 and H63D in sporadic Alzheimerʹs
disease in a folate‐supplemented Ontario population.ʺ J Alzheimers Dis 14(1): 69‐84.
Plattner, F., M. Angelo, et al. (2006). ʺThe roles of cyclin‐dependent kinase 5 and glycogen
synthase kinase 3 in tau hyperphosphorylation.ʺ J Biol Chem 281(35): 25457‐65.
Pulliam, J. F., C. D. Jennings, et al. (2003). ʺAssociation of HFE mutations with
neurodegeneration and oxidative stress in Alzheimerʹs disease and correlation with
APOE.ʺ Am J Med Genet B Neuropsychiatr Genet 119B(1): 48‐53.
Restagno, G., F. Lombardo, et al. (2007). ʺHFE H63D polymorphism is increased in patients with
amyotrophic lateral sclerosis of Italian origin.ʺ J Neurol Neurosurg Psychiatry 78(3): 327.
Sengupta, A., M. Novak, et al. (2006). ʺRegulation of phosphorylation of tau by cyclin‐
dependent kinase 5 and glycogen synthase kinase‐3 at substrate level.ʺ FEBS Lett
580(25): 5925‐33.
Sutedja, N. A., R. J. Sinke, et al. (2007). ʺThe association between H63D mutations in HFE and
amyotrophic lateral sclerosis in a Dutch population.ʺ Arch Neurol 64(1): 63‐7.
166
Waalen, J., B. G. Nordestgaard, et al. (2005). ʺThe penetrance of hereditary hemochromatosis.ʺ
Best Pract Res Clin Haematol 18(2): 203‐20.
Wang, X. S., S. Lee, et al. (2004). ʺIncreased incidence of the Hfe mutation in amyotrophic lateral
sclerosis and related cellular consequences.ʺ J Neurol Sci 227(1): 27‐33.
Yen, A. A., E. P. Simpson, et al. (2004). ʺHFE mutations are not strongly associated with
sporadic ALS.ʺ Neurology 62(9): 1611‐2.
Zecca, L., M. B. Youdim, et al. (2004). ʺIron, brain ageing and neurodegenerative disorders.ʺ Nat
Rev Neurosci 5(11): 863‐73.
167
Chapter 5
Prolyl‐Peptidyl Isomerase, Pin1, Activity is Compromised in Association with
the Expression of the HFE Polymorphic Allele, H63D
Abstract
There is substantial interest in HFE gene variants as putative risk factors in
neurodegenerative diseases such as Alzheimer disease (AD). Previous studies in cell models
have shown the H63D HFE variant to result in increased cellular iron, oxidative stress,
glutamate dyshomeostasis, and an increase in tau phosphorylation; all processes thought to
contribute to AD pathology. Pin1 is a prolyl‐peptidyl cis/trans isomerase that regulates the
dephosphorylation of the amyloid and tau proteins implicated in AD. Previous reports have
shown Pin1 function to be altered by oxidative stress, thus it was logical to hypothesize that
expressing the H63D polymorphism would decrease Pin1 activity. To test our hypothesis, we
utilized stably transfected human neuroblastoma SH‐SY5Y cell lines expressing the different
HFE polymorphisms. Under resting conditions, total Pin1 levels were unchanged between the
wild type and H63D HFE cells, yet there was a significant decrease in Pin1 activity in H63D
variant cells as determined by phosphorylation of Pin1 at its serine 16 residue. To evaluate
whether cellular iron status could influence Pin1, we treated the WT HFE cells with exogenous
iron in the form of ferric ammonium citrate and found Pin1 phosphorylation increased with
increasing levels of iron. Iron exposure to H63D variant cells did not impact Pin1
phosphorylation. Removal of iron with desferrioxamine increased Pin1 activity in H63D cells,
168
but had no effect in cells expressing WT HFE. Collectively, these data suggest that Pin1 can be
altered by cellular iron and that H63D gene variants alter the activity of Pin1 in a direction
predicted by changes in iron status. The data also indicate there is a cellular equilibrium
mechanism of Pin1 activity in cells that is reset by the presence of H63D. The HFE H63D cells
have been shown to be more susceptible to oxidative stress, which can also impact Pin1
function. Therefore we treated the HFE cells with the antioxidant Trolox. Pin1 phosphorylation
decreased in H63D cells and was unaffected in WT HFE cells with Trolox treatment. These data
further support that H63D impacts Pin1 function in the cell. Thus, we have shown another
cellular mechanism that HFE polymorphisms can influence regarding their putative role as risk
factors for neurodegenerative diseases.
Introduction
Alzheimer disease (AD) is a neurodegenerative disease that results in cognitive
deficiencies presumably resulting from neuropathological changes including accumulation of
amyloid‐beta (Aβ) plaques and neurofibrillary tangles (NFT) (Selkoe, 1989; Mandelkow and
Mandelkow, 1993; Trojanowski et al., 1995). Numerous studies have discovered that biometals
such as iron, copper, zinc, and aluminum can directly impact AD pathological markers (Bush,
2003; Huang et al., 2004; Zecca et al., 2004). Iron has been shown to directly regulate amyloid
precursor protein (APP) synthesis via an iron responsive element in the 5’ UTR of APP mRNA
(Rogers et al., 2002). Furthermore, iron and other metals have been found in neuritic plaques
and also influence amyloid‐beta (Aβ) aggregation (Connor et al., 1992a; Connor et al., 1992b;
169
Huang et al., 2004; Liu et al., 2006) and metal chelators have been shown to be effective in
removing Aβ plaques in a AD transgenic mouse model (Cherny et al., 2001). The aggregation of
tau protein leading to NFT pathology is also impacted by iron (Sayre et al., 2000; Egana et al.,
2003). It is clear from these data that metal homeostasis in the brain is essential for healthy brain
aging and that identification of genetic and environmental factors that may disrupt the
biometal homeostasis is critical to identifying pathogenic mechanisms leading to
neurodegenerative processes.
One such factor that combines gene and environment interaction is HFE. The gene is
located on chromosome 6 and has been investigated in numerous genetics studies as a possible
risk factor for developing or modifying AD onset (Connor and Lee, 2006). HFE protein is a
major histocompatibility class‐1 like molecule that is reported to be involved in iron regulation
(Feder et al., 1998) and innate immunity (Ehrlich and Lemonnier, 2000; Rohrlich et al., 2005).
One of the reported functions of HFE protein is to complex with transferrin receptor (TfR) at the
cell membrane to decrease TfR affinity for iron uptake (Feder et al., 1998). When the H63D HFE
variant is expressed, the ability to limit iron uptake is lost, resulting in increased cellular iron
(Feder et al., 1998; Lee et al., 2007). The elevated cellular iron levels associated with the H63D
can lead to oxidative stress (Pulliam et al., 2003; Lee et al., 2007) and exacerbate the
inflammatory response of macrophages (Mitchell et al., 2009). Recently, we have reported that
cells carrying the H63D HFE have alterations in glutamate homeostasis and tau
phosphorylation (Hall II et al., in press; Mitchell et al., in press). These proposed functions of
170
HFE and the data on H63D HFE are directly relevant to AD pathogenesis mechanisms that
involve iron, oxidative stress, and neuroinflammation.
The prolyl‐peptidyl isomerase Pin1 can affect numerous cellular mechanisms, such as
protein localization, protein interactions, protein dephosphorylation, transcription activity,
enzymatic activity, protein stability, and cell cycle regulation (Wulf et al., 2005; Lu and Zhou,
2007; Takahashi et al., 2008). Pin1 has two domains, an N‐terminal WW domain and a C‐
terminal PPIase domain, that represent its unique ability to bind specific serine/threonine‐
proline substrates and alter their conformation, respectively (Lu et al., 2002; Lu and Zhou, 2007).
The ability of Pin1 to alter the cis‐ and trans‐confirmation of its substrate proteins is the direct
mechanism that affects the phosphorylation status of its targeted proteins (Lu et al., 2002; Lu
and Zhou, 2007). Pin1 interacts with protein phosphatase 2A (PP2A) to dephosphorylate its
protein substrates (Hamdane et al., 2006; Balastik et al., 2007; Lu and Zhou, 2007).
A study investigating mild cognitively impaired (MCI) patients found Pin1 to be an
oxidatively modified protein that could impact future AD development (Butterfield et al., 2006).
Pin1 can regulate the phosphorylation of APP and tau at the Threonine 668 and Threonine 231
residues, respectively (Hamdane et al., 2002; Hamdane et al., 2006; Pastorino et al., 2006;
Balastik et al., 2007). A deficiency in Pin1 expression and/or activity appears to lead to the
accumulation of Aβ fragments and NFT formation suggesting this enzyme could be a key
regulator of proteins involved in AD pathology (Liou et al., 2003; Hamdane et al., 2006;
Pastorino et al., 2006). Thus, we hypothesized that cells expressing H63D would have
171
decreased Pin1 activity due to the increased oxidative stress associated with expressing the
H63D allele.
Materials and methods
Reagents ‐ Cell culture reagents including DMEM/F12, DMEM, pen/strep/glutamine and
Geneticin were purchased from Invitrogen (Carlsbad, CA, USA). Fetal bovine serum was
purchased from Gemini Bio‐Products (West Sacramento, CA, USA). DC protein assay was
obtained from Bio‐Rad (Hercules, CA, USA). A rabbit polyclonal Pin1 antibody was
purchased from Santa Cruz Biotechnology (Santa Cruz, CA, USA). A rabbit polyclonal Pin1
(serine 16) antibody was purchased from Cell Signaling Technology (Beverly, MA, USA).
Cell culture ‐ Human neuroblastoma SH‐SY5Y cell lines were obtained from American Type
Culture Collection (Manassas, VA, USA). Human neuroblastoma SH‐SY5Y cells were stably
transfected to express wild‐type and H63D HFE forms as previously reported along with a
vector alone control (Wang et al., 2004). We have previously reported that these cells were
chosen because endogenous expression of HFE could not be detected. The transfected cells
were maintained in DMEM/F12 media supplemented with 10% FBS, 1% antibiotics (pen‐strep‐
glutamine), 1x nonessential amino acids, and 1.8g/L sodium bicarbonate. Cells were
differentiated with 10 μM all‐trans retinoic acid (Sigma‐Aldrich, St. Louis, MO, USA) over six
days (Haque et al., 1999). To evaluate cellular iron effects, cells were treated with ferric
172
ammonium citrate (FAC) or desferrioxamine (DFO) over 48 hours (Sigma‐Aldrich, St. Louis,
MO, USA) (Lee et al., 2007).
Cell lysate preparation ‐ Cells were lysed with RIPA buffer supplemented with 1% Triton X‐100
and protease inhibitor cocktail (Sigma Aldrich, St. Louis, MO). Phosphatase inhibitor (Sigma
Aldrich, St. Louis, MO) was included in cell lysis buffer for phosphorylation protein detection.
Cells extracts were spun at 8,000 × g for 10 min. Total protein levels were determined by Bio‐
Rad DC protein assay.
Enzyme linked immunoabsorbant assay (ELISA) ‐ The phosphorylation of APP at threonine 668 was
determined using a DuoSet IC ELISA assay (R & D Systems, Minneapolis, MN, USA). An
immobilized capture antibody for multiple APP isoforms that binds phosphorylated and
unphosphorylated protein was used to coat the wells of a 96 well microplate overnight.
Phosphorylated APP threonine 668 protein standards were added to achieve a standard curve
to determine the specific amount of phosphorylated APP in unknown HFE cell samples. A
biotynalated detection antibody recognizing APP threonine 668 was used to detect
phosphorylated APP using standard streptavidin‐HRP. The ELISA assay plates were read at
450nm and 540nm to correct for optical imperfections. All solutions used throughout were
provided or prepared according to the manufacturer’s recommendations. This ELISA
experiment was performed using samples in triplicate per genotype at two dilution
173
concentrations along with the known standards for the phosphorylated APP threonine 668
proteins, resulting in a total of six samples for analysis.
A monoclonal antibody specific for human tau phosphorylated residue threonine 231
was coated onto the wells of the microtiter strips provided (Invitrogen, Carlsbad, CA, USA).
Standards of known phosphorylated tau threonine 231 proteins were processed to achieve a
standard curve to determine the specific amount of phosphorylated protein in the unknown
HFE cell samples. The ELISA assay plate was read at 450nm. This ELISA experiment was
performed using samples in triplicate per genotype at two dilution concentrations along with
the known standards for the phosphorylated tau threonine 231 proteins, resulting in a total of
six samples for analysis.
Western blot ‐ Cells lysates were obtained as described above. Twenty‐five to forty μg total
protein was equally separated by electrophoresis in a 4‐20% 12‐well Criterion gel (Bio‐Rad,
Hercules, CA). Protein was then transferred to a nitrocellulose membrane and blocked for 1 hr
at room temperature in TBS‐T with 5% nonfat milk or 1.5% BSA (phosphorylated protein
detection). Membranes were probed with primary antibodies in TBS‐T with 5% nonfat milk
overnight at 4°C. The membranes were incubated with antibodies specific for total Pin1 (1:1000),
Pin1 serine‐16 (1:500), and β‐actin (1:5000). HRP‐conjugated secondary antibodies were added
in 5% nonfat milk for 1 hr at room temperature. Protein signals were obtained by
chemiluminescence and visualized by CCD camera. All western blot experiments were repeated
at least twice with a minimum of four different cultures per genotype per experiment, resulting
174
in a total of eight samples for analysis. The bands on the western blot were quantified by
densitometry using Fuji MultiGauge analysis software.
Statistical analysis ‐ The Student’s t‐test was used for analyzing HFE variant comparisons when
one variable was being determined. Experimental data where samples were treated with
various agents (i.e. iron or desferrioxamine) and compared with controls were analyzed by one‐
way analysis of variance. Differences among the means were considered statistically significant
when the p value was <0.05. If overall p <0.05, Tukey’s Multiple Comparison post hoc analysis
was performed. Data are presented as the mean ± S.E.; GraphPad Prism software (version 4.0)
was utilized to perform the statistical analysis.
Results
The first study was designed to examine total Pin 1 protein expression as a function of
genotype. No significant differences were found (Figure 1). The phosphorylation of Pin1 at its
serine 16 residue, a measure of Pin1 activity, was significantly increased (30%) in cells
expressing the H63D variant (p<0.01) compared to the vector and wild type HFE cells (Figure
5.1). These data would suggest that there is approximately a 30% decrease in the activity of Pin1
in H63D cells.
Because HFE is involved in regulating cellular iron status and we have shown that there
is more iron in the labile iron pool in cells carrying the H63D variant, the second study was to
determine if iron could impact Pin1 phosphorylation status. Upon, treating the wild type HFE
175
cells with ferric ammonium citrate (FAC), total Pin1 protein expression was unaffected (Figure
5.2A) but the phosphorylation of Pin1 at serine 16 increased in a dose dependent manner (10
μM FAC, (p<0.01); 30 μM FAC, p<0.001) compared to the non‐treated wild type control (Figure
5.2B). However, differences in Pin1 phosphorylation in the H63D carrying cells treated with
FAC did not reach statistical significance at p<0.05 cutoff at the same concentrations of iron
provided to WT cells (Figure 5.2B).
To continue to evaluate the sensitivity of Pin1 phosphorylation to iron availability, the
WT and H63D cells were treated with the iron chelator desferrioxamine (DFO). Pin1
phosphorylation was unaffected in the wt HFE cells by iron chelation but was decreased with
10 μM DFO treatment (p<0.01) in the H63D HFE cell lines (Figure 5.3).
Increased cellular iron can result in oxidative stress. There is evidence of oxidative stress
impacting the cellular activity of Pin1 (Butterfield et al., 2006; Sultana et al., 2006; Kap et al.,
2007) and we have shown increased indices of oxidative stress in H63D expressing cells (Lee et
al., 2007). Therefore, we treated the cells expressing the HFE variants with Trolox, a vitamin E
analog to decrease cellular oxidative stress. Pin1 phosphorylation decreased approximately 20%
with Trolox treatment in H63D cells (p=0.0257), but was unchanged in the control wt expressing
cells treated with Trolox (Figure 5.4).
Pin1 has been shown to regulate the phosphorylation of AD related proteins (Hamdane
et al., 2006; Pastorino et al., 2006; Balastik et al., 2007). Therefore, APP phosphorylation at its
threonine 668 residue was measured in HFE cells by enzyme‐linked immunoabsorbant assay
(ELISA). APP phosphorylation was increased in vector cells by approximately 65% (p<0.01)
176
compared to cells expressing wild type HFE (Figure 4). Cells expressing the H63D variant had
50% less APP threonine 668 phosphorylation (p<0.05) with respect to WT HFE cells (Figure 5).
Tau threonine 231 phosphorylation was evaluated by ELISA and was decreased in vector cells
(p<0.01) by 42% and H63D expressing cells (p<0.05) by 33% compared to cells expressing WT
HFE (Figure 5.5).
Discussion
The H63D variant of HFE gene is under examination as a risk factor for
neurodegenerative diseases. We have established a cell model in which to directly examine the
impact of HFE polymorphisms on an otherwise homogenous genetic background in a
controlled culture condition. The controlled conditions are essential to understanding the
contribution of HFE gene variants to neurodegenerative disease because of the likelihood of
gene/environment interaction given the availability of iron in the environment. Our working
hypothesis for H63D HFE gene variants and neurodegenerative disorders is that the H63D
allelic variants does not in itself cause disease but creates a permissive or enabling cellular
milieu for pathogenic agents. We have previously reported that expression of the H63D HFE
variant in stably transfected SH‐SY5Y cells results in increased cellular stress (Lee et al., 2007),
altered glutamate homeostasis (Mitchell et al., in press), and increased tau phosphorylation
(Hall II et al., in press). Oxidative stress and glutamate excitotoxicity are indirectly thought to
contribute to neurodegeneration in AD whereas tau phosphorylation is more directly
implicated as part of the pathogenesis of AD (Doble, 1999; Ballatore et al., 2007). In this study,
177
we extend these observations to show that Pin1, an enzyme responsible for regulating
phosphorylation of amyloid and tau, is altered in cells expressing H63D HFE.
H63D HFE cells have more labile iron in the cytosol (Lee et al., 2007; Mitchell et al., 2009)
than wt HFE cells. Therefore, we treated the wt cells with increasing amounts of iron which
resulted in a dose‐dependent increase in phosphorylation of Pin1 at its serine 16 residue. We
also challenged the H63D HFE cells with more iron to determine if we could further increase
the phosphorylation of Pin1 but increasing amounts of iron did not significantly increase Pin1
phosphorylation. To further evaluate the sensitivity of Pin1 to iron and the influence of
genotype on the sensitivity, we limited iron via chelation. Chelating iron resulted in a reduction
of Pin1 phosphorylation in the H63D HFE cells but not in the WT HFE expressing cells. These
findings suggest that there is a threshold for Pin1 activity that is maintained in the cells and that
the H63D genotype has pushed the activity to a higher level that can be achieved by elevating
iron levels in the WT cells.
The iron effect on Pin1 could be indirect due to oxidative stress through the Fenton
reaction (Thompson et al., 2001; Bush, 2003). To determine the role of oxidative stress, we
treated the HFE polymorphism carrying cells with the antioxidant Trolox, a vitamin E analog.
In the WT HFE cells, there was not a change in Pin1 phosphorylation. Trolox treatment in the
H63D cells resulted in an a decrease in Pin1 serine 16 phosphorylation indicating that Pin1
activity is effected under resulting conditions in the presence of H63D HFE by oxidative stress and
can be improved by antioxidant treatment, whereas the treatment has no effect on the wt cells.
178
Our data are consistent with the findings of other groups (Butterfield et al., 2006; Sultana et al.,
2006; Kap et al., 2007) showing that Pin1 can be impacted by oxidative stress.
To further evaluate a potential genetic influence on Pin1, we examined the impact of the
H63D variant on Pin1 expression and function. Total Pin1 expression levels were found to be
unchanged between the HFE polymorphisms and the transfection vector control. Pin1 activity
was significantly decreased in cells that expressed the H63D variant compared to wild type
HFE cells based on an increase in phosphorylation of Pin1 at serine 16 shown in the results
section. This finding is consistent with the idea of the H63D variant being associated with iron‐
mediated oxidative stress that can disrupt the activity of Pin1.
The reduction in Pin1 activity in association with the H63D HFE mutation could lead to
cellular changes associated with AD, specifically the inability to dephosphorylate APP and tau
proteins (Hamdane et al., 2002; Etzkorn, 2006; Balastik et al., 2007). Pin1 is an intracellular
regulator of amyloid and tau protein phosphorylation at the APP threonine 668 and tau
threonine 231 amino acid residues, which appear to be important to the pathological generation
of Aβ plaques and tangles (Liou et al., 2003; Hamdane et al., 2006; Pastorino et al., 2006). To
evaluate the consequence of altered Pin1 function in H63D variant cells, we determined the
phosphorylation of APP threonine 668 and tau threonine 231 levels in HFE expressing cells.
Surprisingly, we found a significant decrease in APP phosphorylation in H63D cells compared
to cells expressing wild type HFE. Furthermore, a significant reduction at the tau threonine 231
residue occurred in the H63D expressing cells, consistent with previous findings (Hall II et al.,
in press). These data are not consistent with the reduction in Pin1 activity in H63D variant cells.
179
The apparent inconsistency may be explained by our previous report of a reduction in cdk‐5
expression and activity in H63D polymorphism expressing cells (Hall II et al., in press). Cdk‐5
has been shown to regulate the phosphorylation of these proteins at the threonine 668 and
threonine 231 specific sites of APP and tau, respectively (Iijima et al., 2000; Liu et al., 2003;
Ryder et al., 2003; Li et al., 2006; Ballatore et al., 2007). Furthermore, the decreased Pin1 activity
data in the H63D cells are consistent with the increased GSK‐3β activity associated with
expressing the H63D variant (Hall II et al., in press). Min and colleagues showed that lithium
inhibition of GSK‐3β resulted in an increase in Pin1 activation suggesting that regulating GSK‐
3β may affect Pin1’s ability to dephosphorylate its substrates such as tau (Min et al., 2005).
These data, in association with our findings that Trolox effects on Pin1 in this study are
affected by H63D HFE expression are compelling evidence that the HFE polymorphism should
be considered when evaluating treatment strategies in neurodegenerative diseases. We
conclude that the discovery of HFE, implicated as putative risk factor for neurodegenerative
disease such as AD can impact Pin1 is clinically meaningful and given the abundance of iron in
the diet and environment, further investigation is warranted into the gene‐environment
interaction between HFE polymorphisms and iron.
180
Figure 5.1: Pin1 expression and activity. Expression of total Pin1 protein and Pin1 activity determined by its serine 16 phosphorylation was measured in an HFE polymorphism stably transfected SH‐SY5Y cell line by western blot. The data show that there were no significant differences in total Pin1 protein levels. Pin1 phosphorylation was significantly increased in H63D expressing cells (p<0.01) compared to cells containing wild type HFE. Experiments were performed with a minimum of four different cultures per genotype. Representative western blot images are shown with graphs displaying differences in expression determined by densitometric analysis. One‐way ANOVA was performed to analyze the data followed by Tukey’s post‐hoc analysis. Data are represented as mean ± S.E. The symbol ** (p<0.01) indicates a significance difference from wild type HFE.
Pin1 (ser-16)
Vector WT H63D0
25
50
75
100
125
150 **
% D
iffer
ence
Pin1 Pin1 (ser 16)
β-actin β-actin
Total Pin1
Vector WT H63D0
25
50
75
100
% D
iffer
ence
Vector WT H63D Vector WT H63D
181
Figure 5.2: Cellular iron effects on Pin1 expression and activity. We added increasing amounts of ferrous ammonium citrate (FAC) to cells expressing wild type and H63D HFE. Total Pin1 protein levels were not changed with iron treatments (A). Pin1 activity decreased with increasing amounts of iron in a dose‐dependent fashion in wild type cells (10 μM, p<0.01) and (30 μM, p<0.001) as indicated by an increase in Pin1 serine 16 phosphorylation (B). Additionally, we further challenged the H63D cells by treating them with iron in the form of FAC. H63D expressing cells did not achieve a statistical significant increase in Pin1 serine 16 phosphorylation with iron treatments (B). Experiments were performed with a minimum of four different cultures per genotype. Representative western blot images are shown with graphs displaying differences in expression determined by densitometric analysis. One‐way ANOVA was performed to analyze the data followed by Tukey’s post‐hoc analysis. Data are represented as mean ± S.E. The symbols ** (p<0.01) and *** (p<0.001) indicate significance from the respective non‐treated group. A.
Pin1 Pin1
H63D 10 µM 30 µM WT 10 µM 30 µM
β-actin β-actin
Total Pin1
WT 0µM iro
n
WT 10µM
iron
WT 30µM
iron
0
25
50
75
100
% D
iffer
ence
Total Pin1
H63D 0µ
M iron
H63D 10
µM iro
n
H63D 30
µM iro
n0
25
50
75
100
% D
iffer
ence
182
Pin1 (ser-16)
H63D 0µ
M iron
H63D 10
µM iro
n
H63D 30
µM iro
n0
25
50
75
100
125
% D
iffer
ence
B.
WT 10 µM 30 µM
Pin1 (ser 16)
H63D 10 µM 30 µM
β-actin β-actin
Pin1 (ser 16)
Pin1 (ser-16)
WT 0 µM
iron
WT 10 µM
iron
WT 30 µM
iron
0
25
50
75
100
125
150
175
*****
% D
iffer
ence
183
Pin1 (ser-16)
H63D 0
µM DFO
H63D 5
µM DFO
H63D 10
µM DFO0
25
50
75
100
**
% D
iffer
ence
Figure 5.3: Iron chelation and Pin1 activity. Increasing amounts of the iron chelator desferrioxamine (DFO) were added to cells expressing wild type and H63D HFE. Pin1 activity was measured by western blot via phosphorylation of Pin1 at serine 16. Pin1 activity was not altered when WT cells were treated with DFO. DFO treatment of H63D cells resulted in an increase in Pin1 activity as indicated by a decrease in phosphorylation of Pin1 at serine 16. At 5 μM DFO, there was not a significant difference compared to the non‐treated H63D group. Upon treating the H63D cells with 10 μM DFO, there was a significant decrease (p<0.01) in Pin1 phosphorylation. Experiments were performed with a minimum of four different cultures per genotype. Representative western blot images are shown with graphs displaying differences in expression determined by densitometric analysis. One‐way ANOVA was performed to analyze the data followed by Tukey’s post‐hoc analysis. Data are represented as mean ± S.E. The symbol ** (p<0.01) indicate significance from the respective non‐treated group.
WT 5 µM 10 µM
Pin1 (ser 16) Pin1 (ser 16)
H63D 5 µM 10 µM
β-actin β-actin
PIn1 (ser-16)
WT 0µM D
FO
WT 5µM D
FO
WT 10µM
DFO
0
25
50
75
100
% D
iffer
ence
184
Pin1 (ser-16)
H63D H63D Trolox0
25
50
75
100
*
% D
iffer
ence
Figure 5.4: Trolox treatment and Pin1 activity. HFE polymorphism stably transfected SH‐SY5Y cells expressing wild type and H63D variant were treated for 72 hours with 200 μM Trolox, a water‐soluble vitamin E analog to assess the effect of oxidative stress on Pin1 activity determined by western blot. Trolox treatment had no effect on WT HFE cells. There was an increase in Pin1 activity upon treating H63D cells with Trolox as evidenced by a decrease in Pin1 serine 16 phosphorylation (p=0.0257). Student’s t‐test was performed to analyze the data; data are represented as mean ± S.E. The symbol * (p<0.05) indicates a significance difference from baseline H63D HFE.
Pin1 (ser 16) Pin1 (ser 16)
β-actin β-actin
WT Trolox H63D Trolox
Pin1 (ser-16)
WT WT Trolox0
25
50
75
100
% D
iffer
ence
185
Figure 5.5: HFE effects on Alzheimer disease Pin1 substrates. The phosphorylation of amyloid and tau proteins at threonine 668 (t668) and threonine 231 (t231), respectively, has been shown to be impacted by Pin1 activity. APP t668 protein levels were increased in vector cells (p<0.01) compared to wild type HFE cells determined by an ELISA assay. APP t668 levels were decreased in H63D expressing cells (p<0.05) compared to WT HFE cells. Tau t231 protein levels were decreased in vector (p<0.01) and H63D cells (p<0.01) with respect to cells expressing wild type HFE. Experiments were performed with a minimum of four different cultures per genotype. One‐way ANOVA was performed to analyze the data followed by Tukey’s post‐hoc analysis. Data are represented as mean ± S.E. The symbols * (p<0.05) and ** (p<0.01) indicates a significance difference from wild type HFE.
Tau pT231
Vector WT H63D0
20
40
60
80
100
** *
% D
iffer
ence
APP pT668
Vec WT H63D0
50
100
150
200**
*% D
iffer
ence
186
References
Balastik M, Lim J, Pastorino L, Lu KP (2007) Pin1 in Alzheimerʹs disease: multiple substrates,
one regulatory mechanism? Biochim Biophys Acta 1772:422‐429.
Ballatore C, Lee VM, Trojanowski JQ (2007) Tau‐mediated neurodegeneration in Alzheimerʹs
disease and related disorders. Nat Rev Neurosci 8:663‐672.
Bush AI (2003) The metallobiology of Alzheimerʹs disease. Trends Neurosci 26:207‐214.
Butterfield DA, Poon HF, St Clair D, Keller JN, Pierce WM, Klein JB, Markesbery WR (2006)
Redox proteomics identification of oxidatively modified hippocampal proteins in mild
cognitive impairment: insights into the development of Alzheimerʹs disease. Neurobiol
Dis 22:223‐232.
Cherny RA, Atwood CS, Xilinas ME, Gray DN, Jones WD, McLean CA, Barnham KJ, Volitakis I,
Fraser FW, Kim Y, Huang X, Goldstein LE, Moir RD, Lim JT, Beyreuther K, Zheng H,
Tanzi RE, Masters CL, Bush AI (2001) Treatment with a copper‐zinc chelator markedly
and rapidly inhibits beta‐amyloid accumulation in Alzheimerʹs disease transgenic mice.
Neuron 30:665‐676.
Connor JR, Lee SY (2006) HFE mutations and Alzheimerʹs disease. J Alzheimers Dis 10:267‐276.
Connor JR, Menzies SL, St Martin SM, Mufson EJ (1992a) A histochemical study of iron,
transferrin, and ferritin in Alzheimerʹs diseased brains. J Neurosci Res 31:75‐83.
Connor JR, Snyder BS, Beard JL, Fine RE, Mufson EJ (1992b) Regional distribution of iron and
iron‐regulatory proteins in the brain in aging and Alzheimerʹs disease. J Neurosci Res
31:327‐335.
187
Doble A (1999) The role of excitotoxicity in neurodegenerative disease: implications for therapy.
Pharmacol Ther 81:163‐221.
Egana JT, Zambrano C, Nunez MT, Gonzalez‐Billault C, Maccioni RB (2003) Iron‐induced
oxidative stress modify tau phosphorylation patterns in hippocampal cell cultures.
Biometals 16:215‐223.
Ehrlich R, Lemonnier FA (2000) HFE‐‐a novel nonclassical class I molecule that is involved in
iron metabolism. Immunity 13:585‐588.
Etzkorn FA (2006) Pin1 flips Alzheimerʹs switch. ACS Chem Biol 1:214‐216.
Feder JN, Penny DM, Irrinki A, Lee VK, Lebron JA, Watson N, Tsuchihashi Z, Sigal E, Bjorkman
PJ, Schatzman RC (1998) The hemochromatosis gene product complexes with the
transferrin receptor and lowers its affinity for ligand binding. Proc Natl Acad Sci U S A
95:1472‐1477.
Hall II E, Lee S, Mairue N, Simmons Z, Connor J (in press) Expression of the HFE allelic variant
H63D in SH‐SY5Y cells affects tau phosphorylation at serine residues. Neurobiology of
Aging in press.
Hamdane M, Dourlen P, Bretteville A, Sambo AV, Ferreira S, Ando K, Kerdraon O, Begard S,
Geay L, Lippens G, Sergeant N, Delacourte A, Maurage CA, Galas MC, Buee L (2006)
Pin1 allows for differential Tau dephosphorylation in neuronal cells. Mol Cell Neurosci
32:155‐160.
Hamdane M, Smet C, Sambo AV, Leroy A, Wieruszeski JM, Delobel P, Maurage CA, Ghestem
A, Wintjens R, Begard S, Sergeant N, Delacourte A, Horvath D, Landrieu I, Lippens G,
188
Buee L (2002) Pin1: a therapeutic target in Alzheimer neurodegeneration. J Mol Neurosci
19:275‐287.
Haque N, Tanaka T, Iqbal K, Grundke‐Iqbal I (1999) Regulation of expression, phosphorylation
and biological activity of tau during differentiation in SY5Y cells. Brain Res 838:69‐77.
Huang X, Moir RD, Tanzi RE, Bush AI, Rogers JT (2004) Redox‐active metals, oxidative stress,
and Alzheimerʹs disease pathology. Ann N Y Acad Sci 1012:153‐163.
Iijima K, Ando K, Takeda S, Satoh Y, Seki T, Itohara S, Greengard P, Kirino Y, Nairn AC, Suzuki
T (2000) Neuron‐specific phosphorylation of Alzheimerʹs beta‐amyloid precursor
protein by cyclin‐dependent kinase 5. J Neurochem 75:1085‐1091.
Kap YS, Hoozemans JJ, Bodewes AJ, Zwart R, Meijer OC, Baas F, Scheper W (2007) Pin1 levels
are downregulated during ER stress in human neuroblastoma cells. Neurogenetics 8:21‐
27.
Lee SY, Patton SM, Henderson RJ, Connor JR (2007) Consequences of expressing mutants of the
hemochromatosis gene (HFE) into a human neuronal cell line lacking endogenous HFE.
Faseb J.
Li T, Hawkes C, Qureshi HY, Kar S, Paudel HK (2006) Cyclin‐dependent protein kinase 5
primes microtubule‐associated protein tau site‐specifically for glycogen synthase kinase
3beta. Biochemistry 45:3134‐3145.
Liou YC, Sun A, Ryo A, Zhou XZ, Yu ZX, Huang HK, Uchida T, Bronson R, Bing G, Li X,
Hunter T, Lu KP (2003) Role of the prolyl isomerase Pin1 in protecting against age‐
dependent neurodegeneration. Nature 424:556‐561.
189
Liu F, Su Y, Li B, Zhou Y, Ryder J, Gonzalez‐DeWhitt P, May PC, Ni B (2003) Regulation of
amyloid precursor protein (APP) phosphorylation and processing by p35/Cdk5 and
p25/Cdk5. FEBS Lett 547:193‐196.
Liu G, Huang W, Moir RD, Vanderburg CR, Lai B, Peng Z, Tanzi RE, Rogers JT, Huang X (2006)
Metal exposure and Alzheimerʹs pathogenesis. J Struct Biol 155:45‐51.
Lu KP, Zhou XZ (2007) The prolyl isomerase PIN1: a pivotal new twist in phosphorylation
signalling and disease. Nat Rev Mol Cell Biol 8:904‐916.
Lu KP, Liou YC, Zhou XZ (2002) Pinning down proline‐directed phosphorylation signaling.
Trends Cell Biol 12:164‐172.
Mandelkow EM, Mandelkow E (1993) Tau as a marker for Alzheimerʹs disease. Trends Biochem
Sci 18:480‐483.
Min SH, Cho JS, Oh JH, Shim SB, Hwang DY, Lee SH, Jee SW, Lim HJ, Kim MY, Sheen YY, Kim
YK (2005) Tau and GSK3beta dephosphorylations are required for regulating Pin1
phosphorylation. Neurochem Res 30:955‐961.
Mitchell R, Lee S, Simmons Z, Connor J (in press) Impact of HFE Polymorphisms on Cellular
Glutamate Regulation. Neurobiology of Aging in press.
Mitchell RM, Lee SY, Randazzo WT, Simmons Z, Connor JR (2009) Influence of HFE variants
and cellular iron on monocyte chemoattractant protein‐1. J Neuroinflammation 6:6.
Pastorino L, Sun A, Lu PJ, Zhou XZ, Balastik M, Finn G, Wulf G, Lim J, Li SH, Li X, Xia W,
Nicholson LK, Lu KP (2006) The prolyl isomerase Pin1 regulates amyloid precursor
protein processing and amyloid‐beta production. Nature 440:528‐534.
190
Pulliam JF, Jennings CD, Kryscio RJ, Davis DG, Wilson D, Montine TJ, Schmitt FA, Markesbery
WR (2003) Association of HFE mutations with neurodegeneration and oxidative stress in
Alzheimerʹs disease and correlation with APOE. Am J Med Genet B Neuropsychiatr
Genet 119B:48‐53.
Rogers JT, Randall JD, Cahill CM, Eder PS, Huang X, Gunshin H, Leiter L, McPhee J, Sarang SS,
Utsuki T, Greig NH, Lahiri DK, Tanzi RE, Bush AI, Giordano T, Gullans SR (2002) An
iron‐responsive element type II in the 5ʹ‐untranslated region of the Alzheimerʹs amyloid
precursor protein transcript. J Biol Chem 277:45518‐45528.
Rohrlich PS, Fazilleau N, Ginhoux F, Firat H, Michel F, Cochet M, Laham N, Roth MP, Pascolo
S, Nato F, Coppin H, Charneau P, Danos O, Acuto O, Ehrlich R, Kanellopoulos J,
Lemonnier FA (2005) Direct recognition by alphabeta cytolytic T cells of Hfe, a MHC
class Ib molecule without antigen‐presenting function. Proc Natl Acad Sci U S A
102:12855‐12860.
Ryder J, Su Y, Liu F, Li B, Zhou Y, Ni B (2003) Divergent roles of GSK3 and CDK5 in APP
processing. Biochem Biophys Res Commun 312:922‐929.
Sayre LM, Perry G, Harris PL, Liu Y, Schubert KA, Smith MA (2000) In situ oxidative catalysis
by neurofibrillary tangles and senile plaques in Alzheimerʹs disease: a central role for
bound transition metals. J Neurochem 74:270‐279.
Selkoe DJ (1989) Molecular pathology of amyloidogenic proteins and the role of vascular
amyloidosis in Alzheimerʹs disease. Neurobiol Aging 10:387‐395.
191
Sultana R, Boyd‐Kimball D, Poon HF, Cai J, Pierce WM, Klein JB, Markesbery WR, Zhou XZ, Lu
KP, Butterfield DA (2006) Oxidative modification and down‐regulation of Pin1 in
Alzheimerʹs disease hippocampus: A redox proteomics analysis. Neurobiol Aging
27:918‐925.
Takahashi K, Uchida C, Shin RW, Shimazaki K, Uchida T (2008) Prolyl isomerase, Pin1: new
findings of post‐translational modifications and physiological substrates in cancer,
asthma and Alzheimerʹs disease. Cell Mol Life Sci 65:359‐375.
Thompson KJ, Shoham S, Connor JR (2001) Iron and neurodegenerative disorders. Brain Res
Bull 55:155‐164.
Trojanowski JQ, Shin RW, Schmidt ML, Lee VM (1995) Relationship between plaques, tangles,
and dystrophic processes in Alzheimerʹs disease. Neurobiol Aging 16:335‐340;
discussion 341‐335.
Wang XS, Lee S, Simmons Z, Boyer P, Scott K, Liu W, Connor J (2004) Increased incidence of the
Hfe mutation in amyotrophic lateral sclerosis and related cellular consequences. J
Neurol Sci 227:27‐33.
Wulf G, Finn G, Suizu F, Lu KP (2005) Phosphorylation‐specific prolyl isomerization: is there an
underlying theme? Nat Cell Biol 7:435‐441.
Zecca L, Youdim MB, Riederer P, Connor JR, Crichton RR (2004) Iron, brain ageing and
neurodegenerative disorders. Nat Rev Neurosci 5:863‐873.
192
Chapter 6
HFE in Pathogenic Discovery for Alzheimer’s Disease: Therapeutic
Implications
Introduction
This thesis has focused on the HFE gene and the contribution of the H63D HFE variant
as a putative risk factor for Alzheimer disease (AD). A schematic of the thesis findings with
respect to the pathogenic pathways implicated in AD is provided (Figure 6.1). In the first aim of
this thesis, we described the role of HFE variants on amyloid regulation. We discovered that
APP levels increased with HFE expression and Aβ peptide exposure led to increased apoptosis
with the H63D polymorphism presumably due to mitochondria dysfunction. The second thesis
objective focused on the involvement of HFE polymorphisms in tau phosphorylation. Elevated
GSK‐3β activity with expression of the H63D HFE variant resulted in increased tau
phosphorylation. The third aim of the thesis explored the impact of HFE polymorphisms on
Pin1 expression and activity. Pin1 phosphorylation was compromised with expression of the
H63D polymorphism.
There is an urgent need to eliminate this debilitating disease as its impact transcends
patients, families, and communities alike around the world. The greatest difficulty may lie in
the unknown pathogenesis of AD, which limits the pharmacological options to substantially
disrupt this progressive disease. Yet, current knowledge of biological mechanisms altered in
AD must be applied to generate new medicines in a timely manner as time is of the essence.
193
Figure 6.1: Cellular impact of H63D HFE on AD pathways. This is a schematic illustrating the
role of H63D HFE on amyloid homeostasis, Aβ toxicity effects, mitochondrial dysfunction, tau
phosphorylation, Pin1 regulation, apoptosis, and ultimately neurodegeneration.
194
What’s the value of current AD drugs?
There are four FDA‐approved drugs indicated for the treatment of AD available today;
cholinesterase inhibitors: donepezil (Aricept ®), galantamine (Razadyne ®), and rivastigmine
(Exelon ®) and glutamate inhibitor: memantine (Namenda ®) (Shah et al., 2008). Cholinergic
dysfunction has been associated with cognitive impairment (Farlow, 1998; Farlow and Evans,
1998; Shah et al., 2008). Thus, the cholinergic drugs are designed to improve cholinergic
homeostasis through a variety of cellular mechanisms (Shah et al., 2008). Glutamate
excitotoxicity has been shown to impact cellular function in neurodegenerative diseases (Doble,
1999) and there is indirect evidence that glutamate dyshomeostasis can disrupt tau pathology in
AD (Ballatore et al., 2007). As a glutamate‐gated NMDA channel, Namenda has been effective
in treating advanced stage cases of AD (Shah et al., 2008). Collectively, these medications are the
only prescribed medications for AD and have been shown to provide therapeutic benefit by
limiting cognitive decline.
However, the benefits are short‐lived, usually lasting up to six months before patients
succumb to the dementia due to neuron loss (Giacobini, 2001, 2002). Hence, it would seem that
these drugs are not impacting AD pathogenesis as the disease continues to run its course.
Nevertheless, the drugs are effective in limiting cognitive decline if only for a limited time.
Thus, the following discussion in this chapter intends to utilize the significant findings of this
thesis to propose therapeutic interventions in AD and promote the significance of gene‐
environment interactions in pathogenic discovery.
195
HFE and APP; is there a link?
In the first aim of this thesis, HFE expression resulted in an increase of total APP
compared to the vector control (no HFE expression), but APP levels did not differ between the
WT and H63D polymorphisms. Similar to most iron management proteins (not including HFE),
APP is regulated by iron regulatory protein (IRP) activity (Rogers et al., 2002). IRP 1 and IRP 2
bind to iron responsive elements (IRE) in the mRNA of targeted proteins based on
environmental conditions of cellular iron levels to control the expression of the iron
management proteins (Connor et al., 1992b; Thompson et al., 2001). Previous data showed that
the active and total IRP/IRE binding was not significantly different in our model based on HFE
polymorphism expression (Lee et al., 2007) suggesting that this may impact IRP/IRE interaction
in the 5’ UTR of APP mRNA (Rogers et al., 2002). Nonetheless, our HFE APP data highlight the
importance of investigating HFE beyond its iron regulatory role and to further consider its
function since its expression elevated APP. HFE has been thought to play a unique role in
innate immunity (de Almeida et al., 2005; Porto and De Sousa, 2007).
It is known that major histocompatibility complex (MHC) class 1 molecules are involved
in antigen recognition and receptor interaction (Fehlmann et al., 1985; Ehrlich and Lemonnier,
2000). HFE, a MHC class 1‐like protein, and MHC class‐1 molecules share 37% homology
(Lebron et al., 1998). HFE has a fold in its α1 and α2 domain helices that constricts its groove to
bind peptides that is thought to disrupt its ability in antigen recognition (Feder et al., 1998;
Ehrlich and Lemonnier, 2000). Yet, a study by Rohrlich and colleagues showed direct
recognition of human HFE by αβ cytolytic T cells without antigen‐presenting function (Rohrlich
196
et al., 2005). Receptor interaction is another function of MHC class 1 molecules that HFE may
possess, even though HFE presumably doesn’t interact with T cell antigen receptors due to
constriction of its peptide binding groove. HFE has been reported to bind beta‐2 (β2)
microglobulin and complex with transferrin receptor at the cell membrane (Bennett et al., 2000).
These protein binding interactions of HFE may be altered with HFE variants as binding with
the transferrin receptor complex are modified with H63D HFE (Waheed et al., 1997; Feder et al.,
1998; Waheed et al., 1999; Waheed et al., 2002). There is also evidence of APP being involved in
receptor interaction/binding.
The cellular function of APP, a type 1 transmembrane protein, is not fully known
(Thinakaran and Koo, 2008; Hardy, 2009). The cytoplasmic domain sequence of APP was shown
to promote transferrin internalization by transplanting APP cytoplasmic sequences into the
cytoplasmic domain of transferrin receptor (Lai et al., 1995). The two cytoplasmic sequence
motifs in APP are related to lipoprotein receptor internalization and tyrosine‐based
internalization (Lai et al., 1995), which imply that APP may act as an integral membrane
receptor. Furthermore, the APP cytoplasmic domain sequence has been shown to serve as a
docking site for numerous proteins involved in cell signal transduction (Venezia et al., 2004a;
Venezia et al., 2004b; Venezia et al., 2007).
The capabilities of HFE and APP may play a role in these proteins responding to
external stimuli. In AD brains, HFE expression was found to be induced throughout the brain
vasculature and in the surrounding plaques and tangles (Connor et al., 2001). These data
suggest HFE may function in the innate immunity response to AD pathology. Our data
197
showing an increase in total APP expression with HFE expression may be due to an unknown
mechanism that occurs with the ‘induction’ of HFE at the cell membrane that may involve
crosstalk with APP. APP is expressed on the plasma membrane as well and can respond to
oxidative stress by up‐regulating its expression as well (Huang et al., 2004). Furthermore, there
is an interleukin‐1 (IL‐1) translational enhancer in APP mRNA establishing that APP translation
can be influenced by IL‐1 release from immune cells in the brain (Rogers et al., 1999;
Bandyopadhyay et al., 2006). Such reported mechanisms of APP and others have suggested that
APP may also be involved in innate immunity (Campbell, 2001).
So is there a link between HFE and APP that goes beyond iron regulation in response to
cellular environmental challenges? The MHC class 1‐like properties of HFE suggest it may have
an immune function, HFE protein interactions imply it possesses the ability to complex with
other proteins, and HFE can impact cell signal transduction pathways as a signaling molecule
based on its induction. APP has been shown to be involved in cell signaling through its
cytoplasmic domain sequence, APP synthesis can regulated by iron and cytokines, and it has
been shown to possess neurotrophic characteristics that propose it may be involved in immune
response. Iron regulation has been shown to impact immunological mechanisms (Ehrlich and
Lemonnier, 2000; Salter‐Cid et al., 2000; Porto and De Sousa, 2007; Maccioni et al., 2009) HFE
and APP may orchestrate a convergence of iron and immunological pathways in neurons that
substantially impacts AD pathogenesis.
198
The impact of GSK‐3β drug discovery
In the latter part of aim 1, we found that H63D cells were more vulnerable to Aβ toxicity
because of mitochondria dysfunction and demonstrated that this lead to increased apoptosis.
The mitochondrial dysfunction discovered may be a result of the increased GSK‐3β activity in
cells expressing the H63D variant. GSK‐3β has been shown to regulate the opening of the
mitochondrial permeability transition pores, where increased opening of these channels can
lead to a loss of the electrochemical gradient and result in apoptosis (Juhaszova et al., 2004;
Murphy, 2004). Moreover, in the second aim of this thesis, tau phosphorylation increased at the
serine (ser)‐199, ser‐202, ser‐396, and ser‐404 sites implicated in NFT generation. The increased
GSK‐3β activity associated with the H63D was found to be responsible for the increase in tau
phosphorylation. The diverse role of GSK‐3β functions in amyloid processing, tau
phosphorylation, mitochondria function, and apoptosis‐promoting mechanisms makes it a very
attractive therapeutic target for neurodegenerative diseases (Grimes and Jope, 2001; Juhaszova
et al., 2004; Huang and Klein, 2006; Uemura et al., 2007). Drug discovery for GSK‐3β inhibition
has gained momentum in many neurological diseases, diabetes, and cancer, but there are
limitations to specificity, efficacy, and therapeutic delivery (Martinez et al., 2002; Cohen and
Goedert, 2004; Avila and Hernandez, 2007; Mazanetz and Fischer, 2007; Martinez, 2008).
While drug development continues for novel GSK‐3β drugs to combat AD, lithium
treatments should be considered for AD treatment since it’s FDA‐approved for other
indications and is readily‐available. Lithium has been shown to be a selective inhibitor of GSK‐
3β (Klein and Melton, 1996; Hong et al., 1997; De Sarno et al., 2002). Numerous studies have
199
shown lithium to impact amyloid regulation and/or tau phosphorylation in cell and animal
models (Munoz‐Montano et al., 1997; Alvarez et al., 1999; Noble et al., 2005; Caccamo et al.,
2007). Research efforts in this thesis have also shown that lithium exposure could reduce tau
phosphorylation in cells expressing the H63D variant. Recently, a 10‐week clinical study found
that lithium treatment increased brain derived neurotrophic factor levels in AD patients and
improved their cognition suggesting longer studies are needed to see the long‐term benefits of
lithium (Leyhe et al., 2009). I would also propose that lithium AD clinical trials should stratify
outcomes based on HFE polymorphisms since baseline GSK‐3β activity is different.
Is iron removal the key to defeating AD pathology?
Cellular iron mismanagement may cause or exacerbate AD through a variety of cellular
mechanisms (Qian and Shen, 2001; Thompson et al., 2001; Bush, 2003; Zecca et al., 2004). Iron
has been shown to regulate APP synthesis and processing (Bodovitz et al., 1995; Rogers et al.,
2002) and tau phosphorylation based on the experiments performed in this thesis. We also
know that iron accumulates in AD brains and is thought to contribute to senile plaques and
NFT (Connor et al., 1992a; Connor et al., 1992b; Sayre et al., 2000; Quintana et al., 2006). Iron and
other biometals can aggregate Aβ peptides (Bush, 2003) and tau aggregation (Yamamoto et al.,
2002); which emphasize the potential utility of chelators to remove iron and inhibit AD
pathology markers. Henceforth, I will review chelator compounds that have been shown to be
effective, are actively being evaluated in trials, and a prospective chelator that has not been fully
characterized.
200
The iron chelator, desferrioxamine (DFO) has proved to be effective in cellular models,
animal models, and human trials to alter AD pathology and improve cognitive performance
(Crapper McLachlan et al., 1991; Savory et al., 1998; Huang et al., 2004). DFO can reduce APP
expression through the 5’ UTR mRNA of APP (Huang et al., 2004; Venti et al., 2004) further
implicating it as a viable therapeutic option that may be more effective with drug targeting
technology. Currently, DFO is FDA‐approved to treat iron overload disorders (Cahill et al.,
2008). Some limitations of DFO therapy include the route of administration, intramuscular or
intravenous injections twice daily, that can lead to inflammation and poor absorption in the GI
tract, which leads to rapid degradation and excretion (Crapper McLachlan et al., 1991; Savory et
al., 1998; Cahill et al., 2008).
Clioquinol (CQ) is a member of the quinoline drug family. It is a hydrophobic
compound that can cross the BBB and chelate iron and other metals (Lynch et al., 2000; Cherny
et al., 2001; Finefrock et al., 2003; Kaur et al., 2003). CQ has been shown to inhibit Aβ
aggregation and H2O2 formation, thereby decreasing cellular stress via iron and copper
chelation (Bush, 2003; Huang et al., 2004). Also, CQ has proven to be efficacious in decreasing
Aβ deposits in cell culture and an AD transgenic mouse model (Cherny et al., 2001; Huang et
al., 2004). In a Phase II clinical trial evaluating CQ for AD patient treatment, CQ was found to
improve cognition, but was withdrawn because of drug quality preparation (Ritchie et al., 2003;
Cahill et al., 2008).
An improved second generation quinoline, an 8‐hydroxy quinoline analog recognized as
PBT‐2 has been developed to overcome the limitations of CQ (Adlard et al., 2008; Barnham and
201
Bush, 2008). Adlard and colleagues demonstrated that PBT‐2 was more effective than CQ in
treating AD transgenic mice by substantially decreasing brain Aβ levels within days and
improving spatial learning and memory. Recently, PBT‐2 has been evaluated for safety and
efficacy in a double‐blind, randomized, placebo‐controlled phase IIa clinical trial for treating
AD patients (Lannfelt et al., 2008). The safety of PBT‐2 was favorable and it lowered CSF Aβ42,
but there was no effect on plasma AD biomarkers (Aβ40 and Aβ42) (Lannfelt et al., 2008).
Cognitive performance improved with PBT‐2 based on two executive function component tests
of a neuropsychological test battery (Lannfelt et al., 2008). Further testing in clinical trials will
determine if PBT‐2 emerges as additional therapy to treat AD.
Another chelator has been shown to be effective in removing iron and aluminum from
hyperphosphorylated tau in AD (Shin et al., 2003). This trivalent chelator known as Feralex was
synthesized using three naturally occurring products: glycine, maltol, and glucosamine (Kruck
and Burrow, 2002). Upon comparing Feralex to DFO, the authors tested these compounds on
AD brain sections and found no differences in the ability to chelate iron, but Feralex also
removed aluminum (Shin et al., 2003). They conclude that Feralex may provide greater
medicinal effects in treating AD patients (Shin et al., 2003), however, this compound has yet to
be tested in an animal model. The potential toxic effects of Feralex and its ability to penetrate
the BBB have yet to be determine and will warrant whether or not it can become an effective
therapy for AD.
It is clear that the chelators discussed show cellular and cognitive benefits; however, the
absorption, toxicity, and ability to cross the BBB are the significant hurdles. Iron is essential for
202
numerous physiological functions throughout the body and excess removal can disrupt cellular
function, which emphasizes the importance of targeting chelators to their desired cellular
destination. Thus, the pursuit for finding effective chelators to impact biometals, amyloid, and
tau homeostasis continues for AD intervention. PBT‐2 appears promising because it has
superior function with respect to CQ and does not have the safety issues CQ presented. It
would be interesting to evaluate the PBT‐2 compound on Pin1and GSK‐3β function since they
can regulate both AD pathological hallmarks (Butterfield et al., 2006; Balastik et al., 2007). In the
third aim of this thesis work we found that DFO treatment improved Pin1 function and the
ability of PBT‐2 to disrupt existing plaque formation and possibly chelate iron (data not shown)
may be another mechanism of therapeutic intervention for AD. Also, the novel finding that
GSK‐3β activity can be impacted by cellular iron discovered in this thesis work further
emphasizes the importance of brain iron homeostasis in managing AD risk.
Utility of an H63D mouse model
To better examine the functional contributions of gene variants associated with AD risk,
like HFE; improved model systems need to be developed. Recently, our group developed a
mouse model to investigate the contribution of the H63D variant to neurodegenerative diseases,
an H67D knock‐in (homologous to human H63D). The model developed is very similar to the
transgenic line developed by Tomatsu and colleagues that became unavailable (Tomatsu et al.,
2003). Our research group plans to substantiate the findings of my thesis and other published
work by our laboratory (Mitchell et al., 2009; Mitchell et al., in press), further examine
203
pathological cellular pathways, and evaluate therapeutic compounds with this mouse model.
The H67D mice can be utilized to investigate the numerous cellular effects of the H63D
variant in the AD pathogenic pathways discussed in the preceding sections of this chapter. As
an example, I will describe my strategy for investigating tau phosphorylation and
neurodegeneration. I hypothesize that homozygous H67D mice will have increased GSK‐3β
activity and decreased Pin1 function, which will lead to NFT‐like pathology and cell death. This
will be determined by evaluating the mice at 4, 8, 12, and 16 months of age to monitor the
impact of aging on the development of pathology and neurodegeneration compared to wild
type HFE and heterozygous H67D littermates. Morris water‐maze and rotorod behavioral
paradigms should be performed to assess learning and memory along with motor coordination,
respectively. Upon sacrifice at the time points, the brain should be removed and each
hemisphere should be used for either immunohistochemical or biochemical analysis. GSK‐3β
expression and phosphorylation, Pin1 expression and phosphorylation, total tau expression and
phosphorylation (PHF‐1 Ab: ser‐396 and ser‐404; AT8 Ab: ser‐199 and ser‐202; AT180 Ab: thr‐
231), and caspase‐3 expression should be determined using both methods with the appropriate
antibodies. NFT‐like pathology would be determined by performing Thioflavin‐S and Gallyas
silver staining, which bind tau filament aggregates; positive staining with either marker detects
NFTs (Liou et al., 2003). NeuN (neuron‐specific nuclear protein) histological staining should be
performed and the NeuN positive neurons should be counted; a reduction in staining would
indicate neuronal death (Liou et al., 2003).
204
Previous studies of GSK‐3β transgenic mice and Pin1 knockout mice have been shown
to result in tau pathology independent of each other (Lucas et al., 2001; Hernandez et al., 2002;
Liou et al., 2003). Thus, the elevation in GSK‐3β activity and reduction in Pin1 function
associated with expressing the H63D variant, which alters cellular iron regulation, should result
in a model of neurodegeneration. If my hypothesis is not supported, I would propose to feed
the mice iron via oral gavage (Kaur et al., 2007) to invoke a gene‐environment interaction that
should result in increased cellular stress to alter the proposed pathogenic mechanisms outlined.
Future investigations with the H67D mouse model could test therapeutic compounds to combat
the pathology proposed.
Conclusion
In summary, this dissertation has established the importance of the role of the H63D
variant in the hallmark AD pathogenic pathways and highlighted therapeutic targets to alter
AD pathology. We have proposed that H63D does not cause AD, but acts as a putative risk
factor that impacts disease onset and progression through altering numerous cellular
mechanisms implicated in AD. Independent of HFE status, we discovered that GSK‐3β could be
impacted by cellular iron. Regulating brain iron levels continues to be paramount because iron
dyshomeostasis can impact numerous biological mechanisms resulting in a profile of complex
pathological changes in AD. We have shown that a single genetic variation can modify AD
associated cellular pathways in a controlled environment, which emphasizes the importance of
identifying causative agents that trigger AD. The continued development of genetic,
205
biochemical, and imaging biomarker profiles will significantly aid in identifying and treating
AD patients.
206
References:
Adlard PA et al. (2008) Rapid restoration of cognition in Alzheimerʹs transgenic mice with 8‐
hydroxy quinoline analogs is associated with decreased interstitial Abeta. Neuron 59:43‐
55.
Alvarez G, Munoz‐Montano JR, Satrustegui J, Avila J, Bogonez E, Diaz‐Nido J (1999) Lithium
protects cultured neurons against beta‐amyloid‐induced neurodegeneration. FEBS Lett
453:260‐264.
Avila J, Hernandez F (2007) GSK‐3 inhibitors for Alzheimerʹs disease. Expert Rev Neurother
7:1527‐1533.
Balastik M, Lim J, Pastorino L, Lu KP (2007) Pin1 in Alzheimerʹs disease: multiple substrates,
one regulatory mechanism? Biochim Biophys Acta 1772:422‐429.
Ballatore C, Lee VM, Trojanowski JQ (2007) Tau‐mediated neurodegeneration in Alzheimerʹs
disease and related disorders. Nat Rev Neurosci 8:663‐672.
Bandyopadhyay S, Hartley DM, Cahill CM, Lahiri DK, Chattopadhyay N, Rogers JT (2006)
Interleukin‐1alpha stimulates non‐amyloidogenic pathway by alpha‐secretase (ADAM‐
10 and ADAM‐17) cleavage of APP in human astrocytic cells involving p38 MAP kinase.
J Neurosci Res 84:106‐118.
Barnham KJ, Bush AI (2008) Metals in Alzheimerʹs and Parkinsonʹs diseases. Curr Opin Chem
Biol 12:222‐228.
207
Bennett MJ, Lebron JA, Bjorkman PJ (2000) Crystal structure of the hereditary
haemochromatosis protein HFE complexed with transferrin receptor. Nature 403:46‐53.
Bodovitz S, Falduto MT, Frail DE, Klein WL (1995) Iron levels modulate alpha‐secretase
cleavage of amyloid precursor protein. J Neurochem 64:307‐315.
Bush AI (2003) The metallobiology of Alzheimerʹs disease. Trends Neurosci 26:207‐214.
Butterfield DA, Abdul HM, Opii W, Newman SF, Joshi G, Ansari MA, Sultana R (2006) Pin1 in
Alzheimerʹs disease. J Neurochem 98:1697‐1706.
Caccamo A, Oddo S, Tran LX, LaFerla FM (2007) Lithium reduces tau phosphorylation but not
A beta or working memory deficits in a transgenic model with both plaques and tangles.
Am J Pathol 170:1669‐1675.
Cahill CM, Lahiri DK, Huang X, Rogers JT (2008) Amyloid precursor protein and alpha
synuclein translation, implications for iron and inflammation in neurodegenerative
diseases. Biochim Biophys Acta.
Campbell A (2001) beta‐amyloid: friend or foe. Med Hypotheses 56:388‐391.
Cherny RA, Atwood CS, Xilinas ME, Gray DN, Jones WD, McLean CA, Barnham KJ, Volitakis I,
Fraser FW, Kim Y, Huang X, Goldstein LE, Moir RD, Lim JT, Beyreuther K, Zheng H,
Tanzi RE, Masters CL, Bush AI (2001) Treatment with a copper‐zinc chelator markedly
and rapidly inhibits beta‐amyloid accumulation in Alzheimerʹs disease transgenic mice.
Neuron 30:665‐676.
Cohen P, Goedert M (2004) GSK3 inhibitors: development and therapeutic potential. Nat Rev
Drug Discov 3:479‐487.
208
Connor JR, Menzies SL, St Martin SM, Mufson EJ (1992a) A histochemical study of iron,
transferrin, and ferritin in Alzheimerʹs diseased brains. J Neurosci Res 31:75‐83.
Connor JR, Snyder BS, Beard JL, Fine RE, Mufson EJ (1992b) Regional distribution of iron and
iron‐regulatory proteins in the brain in aging and Alzheimerʹs disease. J Neurosci Res
31:327‐335.
Connor JR, Milward EA, Moalem S, Sampietro M, Boyer P, Percy ME, Vergani C, Scott RJ,
Chorney M (2001) Is hemochromatosis a risk factor for Alzheimerʹs disease? J
Alzheimers Dis 3:471‐477.
Crapper McLachlan DR, Dalton AJ, Kruck TP, Bell MY, Smith WL, Kalow W, Andrews DF
(1991) Intramuscular desferrioxamine in patients with Alzheimerʹs disease. Lancet
337:1304‐1308.
de Almeida SF, Carvalho IF, Cardoso CS, Cordeiro JV, Azevedo JE, Neefjes J, de Sousa M (2005)
HFE cross‐talks with the MHC class I antigen presentation pathway. Blood 106:971‐977.
De Sarno P, Li X, Jope RS (2002) Regulation of Akt and glycogen synthase kinase‐3 beta
phosphorylation by sodium valproate and lithium. Neuropharmacology 43:1158‐1164.
Doble A (1999) The role of excitotoxicity in neurodegenerative disease: implications for therapy.
Pharmacol Ther 81:163‐221.
Ehrlich R, Lemonnier FA (2000) HFE‐‐a novel nonclassical class I molecule that is involved in
iron metabolism. Immunity 13:585‐588.
Farlow MR (1998) New treatments in Alzheimer disease and the continued need for placebo‐
controlled trials. Arch Neurol 55:1396‐1398.
209
Farlow MR, Evans RM (1998) Pharmacologic treatment of cognition in Alzheimerʹs dementia.
Neurology 51:S36‐44; discussion S65‐37.
Feder JN, Penny DM, Irrinki A, Lee VK, Lebron JA, Watson N, Tsuchihashi Z, Sigal E, Bjorkman
PJ, Schatzman RC (1998) The hemochromatosis gene product complexes with the
transferrin receptor and lowers its affinity for ligand binding. Proc Natl Acad Sci U S A
95:1472‐1477.
Fehlmann M, Peyron JF, Samson M, Van Obberghen E, Brandenburg D, Brossette N (1985)
Molecular association between major histocompatibility complex class I antigens and
insulin receptors in mouse liver membranes. Proc Natl Acad Sci U S A 82:8634‐8637.
Finefrock AE, Bush AI, Doraiswamy PM (2003) Current status of metals as therapeutic targets in
Alzheimerʹs disease. J Am Geriatr Soc 51:1143‐1148.
Giacobini E (2001) Do cholinesterase inhibitors have disease‐modifying effects in Alzheimerʹs
disease? CNS Drugs 15:85‐91.
Giacobini E (2002) Long‐term stabilizing effect of cholinesterase inhibitors in the therapy of
Alzheimerʹ disease. J Neural Transm Suppl:181‐187.
Grimes CA, Jope RS (2001) The multifaceted roles of glycogen synthase kinase 3beta in cellular
signaling. Prog Neurobiol 65:391‐426.
Hardy J (2009) The amyloid hypothesis for Alzheimerʹs disease: a critical reappraisal. J
Neurochem.
210
Hernandez F, Borrell J, Guaza C, Avila J, Lucas JJ (2002) Spatial learning deficit in transgenic
mice that conditionally over‐express GSK‐3beta in the brain but do not form tau
filaments. J Neurochem 83:1529‐1533.
Hong M, Chen DC, Klein PS, Lee VM (1997) Lithium reduces tau phosphorylation by inhibition
of glycogen synthase kinase‐3. J Biol Chem 272:25326‐25332.
Huang HC, Klein PS (2006) Multiple roles for glycogen synthase kinase‐3 as a drug target in
Alzheimerʹs disease. Curr Drug Targets 7:1389‐1397.
Huang X, Moir RD, Tanzi RE, Bush AI, Rogers JT (2004) Redox‐active metals, oxidative stress,
and Alzheimerʹs disease pathology. Ann N Y Acad Sci 1012:153‐163.
Juhaszova M, Zorov DB, Kim SH, Pepe S, Fu Q, Fishbein KW, Ziman BD, Wang S, Ytrehus K,
Antos CL, Olson EN, Sollott SJ (2004) Glycogen synthase kinase‐3beta mediates
convergence of protection signaling to inhibit the mitochondrial permeability transition
pore. J Clin Invest 113:1535‐1549.
Kaur D, Peng J, Chinta SJ, Rajagopalan S, Di Monte DA, Cherny RA, Andersen JK (2007)
Increased murine neonatal iron intake results in Parkinson‐like neurodegeneration with
age. Neurobiol Aging 28:907‐913.
Kaur D, Yantiri F, Rajagopalan S, Kumar J, Mo JQ, Boonplueang R, Viswanath V, Jacobs R, Yang
L, Beal MF, DiMonte D, Volitaskis I, Ellerby L, Cherny RA, Bush AI, Andersen JK (2003)
Genetic or pharmacological iron chelation prevents MPTP‐induced neurotoxicity in
vivo: a novel therapy for Parkinsonʹs disease. Neuron 37:899‐909.
211
Klein PS, Melton DA (1996) A molecular mechanism for the effect of lithium on development.
Proc Natl Acad Sci U S A 93:8455‐8459.
Kruck TP, Burrow TE (2002) Synthesis of feralex a novel aluminum/iron chelating compound. J
Inorg Biochem 88:19‐24.
Lai A, Sisodia SS, Trowbridge IS (1995) Characterization of sorting signals in the beta‐amyloid
precursor protein cytoplasmic domain. J Biol Chem 270:3565‐3573.
Lannfelt L, Blennow K, Zetterberg H, Batsman S, Ames D, Harrison J, Masters CL, Targum S,
Bush AI, Murdoch R, Wilson J, Ritchie CW (2008) Safety, efficacy, and biomarker
findings of PBT2 in targeting Abeta as a modifying therapy for Alzheimerʹs disease: a
phase IIa, double‐blind, randomised, placebo‐controlled trial. Lancet Neurol 7:779‐786.
Lebron JA, Bennett MJ, Vaughn DE, Chirino AJ, Snow PM, Mintier GA, Feder JN, Bjorkman PJ
(1998) Crystal structure of the hemochromatosis protein HFE and characterization of its
interaction with transferrin receptor. Cell 93:111‐123.
Lee SY, Patton SM, Henderson RJ, Connor JR (2007) Consequences of expressing mutants of the
hemochromatosis gene (HFE) into a human neuronal cell line lacking endogenous HFE.
Faseb J.
Leyhe T, Eschweiler GW, Stransky E, Gasser T, Annas P, Basun H, Laske C (2009) Increase of
BDNF serum concentration in lithium treated patients with early Alzheimerʹs disease. J
Alzheimers Dis 16:649‐656.
212
Liou YC, Sun A, Ryo A, Zhou XZ, Yu ZX, Huang HK, Uchida T, Bronson R, Bing G, Li X,
Hunter T, Lu KP (2003) Role of the prolyl isomerase Pin1 in protecting against age‐
dependent neurodegeneration. Nature 424:556‐561.
Lucas JJ, Hernandez F, Gomez‐Ramos P, Moran MA, Hen R, Avila J (2001) Decreased nuclear
beta‐catenin, tau hyperphosphorylation and neurodegeneration in GSK‐3beta
conditional transgenic mice. Embo J 20:27‐39.
Lynch T, Cherny RA, Bush AI (2000) Oxidative processes in Alzheimerʹs disease: the role of
abeta‐metal interactions. Exp Gerontol 35:445‐451.
Maccioni RB, Rojo LE, Fernandez JA, Kuljis RO (2009) The role of neuroimmunomodulation in
Alzheimerʹs disease. Ann N Y Acad Sci 1153:240‐246.
Martinez A (2008) Preclinical efficacy on GSK‐3 inhibitors: towards a future generation of
powerful drugs. Med Res Rev 28:773‐796.
Martinez A, Castro A, Dorronsoro I, Alonso M (2002) Glycogen synthase kinase 3 (GSK‐3)
inhibitors as new promising drugs for diabetes, neurodegeneration, cancer, and
inflammation. Med Res Rev 22:373‐384.
Mazanetz MP, Fischer PM (2007) Untangling tau hyperphosphorylation in drug design for
neurodegenerative diseases. Nat Rev Drug Discov 6:464‐479.
Mitchell R, Lee S, Simmons Z, Connor J (in press) Impact of HFE Polymorphisms on Cellular
Glutamate Regulation. Neurobiology of Aging in press.
Mitchell RM, Lee SY, Randazzo WT, Simmons Z, Connor JR (2009) Influence of HFE variants
and cellular iron on monocyte chemoattractant protein‐1. J Neuroinflammation 6:6.
213
Munoz‐Montano JR, Moreno FJ, Avila J, Diaz‐Nido J (1997) Lithium inhibits Alzheimerʹs
disease‐like tau protein phosphorylation in neurons. FEBS Lett 411:183‐188.
Murphy E (2004) Inhibit GSK‐3beta or thereʹs heartbreak dead ahead. J Clin Invest 113:1526‐
1528.
Noble W, Planel E, Zehr C, Olm V, Meyerson J, Suleman F, Gaynor K, Wang L, LaFrancois J,
Feinstein B, Burns M, Krishnamurthy P, Wen Y, Bhat R, Lewis J, Dickson D, Duff K
(2005) Inhibition of glycogen synthase kinase‐3 by lithium correlates with reduced
tauopathy and degeneration in vivo. Proc Natl Acad Sci U S A 102:6990‐6995.
Porto G, De Sousa M (2007) Iron overload and immunity. World J Gastroenterol 13:4707‐4715.
Qian ZM, Shen X (2001) Brain iron transport and neurodegeneration. Trends Mol Med 7:103‐
108.
Quintana C, Bellefqih S, Laval JY, Guerquin‐Kern JL, Wu TD, Avila J, Ferrer I, Arranz R, Patino
C (2006) Study of the localization of iron, ferritin, and hemosiderin in Alzheimerʹs
disease hippocampus by analytical microscopy at the subcellular level. J Struct Biol
153:42‐54.
Ritchie CW, Bush AI, Mackinnon A, Macfarlane S, Mastwyk M, MacGregor L, Kiers L, Cherny
R, Li QX, Tammer A, Carrington D, Mavros C, Volitakis I, Xilinas M, Ames D, Davis S,
Beyreuther K, Tanzi RE, Masters CL (2003) Metal‐protein attenuation with
iodochlorhydroxyquin (clioquinol) targeting Abeta amyloid deposition and toxicity in
Alzheimer disease: a pilot phase 2 clinical trial. Arch Neurol 60:1685‐1691.
214
Rogers JT, Leiter LM, McPhee J, Cahill CM, Zhan SS, Potter H, Nilsson LN (1999) Translation of
the alzheimer amyloid precursor protein mRNA is up‐regulated by interleukin‐1
through 5ʹ‐untranslated region sequences. J Biol Chem 274:6421‐6431.
Rogers JT, Randall JD, Cahill CM, Eder PS, Huang X, Gunshin H, Leiter L, McPhee J, Sarang SS,
Utsuki T, Greig NH, Lahiri DK, Tanzi RE, Bush AI, Giordano T, Gullans SR (2002) An
iron‐responsive element type II in the 5ʹ‐untranslated region of the Alzheimerʹs amyloid
precursor protein transcript. J Biol Chem 277:45518‐45528.
Rohrlich PS, Fazilleau N, Ginhoux F, Firat H, Michel F, Cochet M, Laham N, Roth MP, Pascolo
S, Nato F, Coppin H, Charneau P, Danos O, Acuto O, Ehrlich R, Kanellopoulos J,
Lemonnier FA (2005) Direct recognition by alphabeta cytolytic T cells of Hfe, a MHC
class Ib molecule without antigen‐presenting function. Proc Natl Acad Sci U S A
102:12855‐12860.
Salter‐Cid L, Peterson PA, Yang Y (2000) The major histocompatibility complex‐encoded HFE in
iron homeostasis and immune function. Immunol Res 22:43‐59.
Savory J, Huang Y, Wills MR, Herman MM (1998) Reversal by desferrioxamine of tau protein
aggregates following two days of treatment in aluminum‐induced neurofibrillary
degeneration in rabbit: implications for clinical trials in Alzheimerʹs disease.
Neurotoxicology 19:209‐214.
Sayre LM, Perry G, Harris PL, Liu Y, Schubert KA, Smith MA (2000) In situ oxidative catalysis
by neurofibrillary tangles and senile plaques in Alzheimerʹs disease: a central role for
bound transition metals. J Neurochem 74:270‐279.
215
Shah RS, Lee HG, Xiongwei Z, Perry G, Smith MA, Castellani RJ (2008) Current approaches in
the treatment of Alzheimerʹs disease. Biomed Pharmacother 62:199‐207.
Shin RW, Kruck TP, Murayama H, Kitamoto T (2003) A novel trivalent cation chelator Feralex
dissociates binding of aluminum and iron associated with hyperphosphorylated tau of
Alzheimerʹs disease. Brain Res 961:139‐146.
Thinakaran G, Koo EH (2008) Amyloid precursor protein trafficking, processing, and function. J
Biol Chem 283:29615‐29619.
Thompson KJ, Shoham S, Connor JR (2001) Iron and neurodegenerative disorders. Brain Res
Bull 55:155‐164.
Tomatsu S, Orii KO, Fleming RE, Holden CC, Waheed A, Britton RS, Gutierrez MA, Velez‐
Castrillon S, Bacon BR, Sly WS (2003) Contribution of the H63D mutation in HFE to
murine hereditary hemochromatosis. Proc Natl Acad Sci U S A 100:15788‐15793.
Uemura K, Kuzuya A, Shimozono Y, Aoyagi N, Ando K, Shimohama S, Kinoshita A (2007)
GSK3beta activity modifies the localization and function of presenilin 1. J Biol Chem
282:15823‐15832.
Venezia V, Nizzari M, Carlo P, Corsaro A, Florio T, Russo C (2007) Amyloid precursor protein
and presenilin involvement in cell signaling. Neurodegener Dis 4:101‐111.
Venezia V, Russo C, Repetto E, Nizzari M, Violani E, Carlo P, Marchetti B, Schettini G (2004a)
Apoptotic cell death and amyloid precursor protein signaling in neuroblastoma SH‐
SY5Y cells. Ann N Y Acad Sci 1030:339‐347.
216
Venezia V, Russo C, Repetto E, Salis S, Dolcini V, Genova F, Nizzari M, Mueller U, Schettini G
(2004b) Apoptotic cell death influences the signaling activity of the amyloid precursor
protein through ShcA and Grb2 adaptor proteins in neuroblastoma SH‐SY5Y cells. J
Neurochem 90:1359‐1370.
Venti A, Giordano T, Eder P, Bush AI, Lahiri DK, Greig NH, Rogers JT (2004) The integrated
role of desferrioxamine and phenserine targeted to an iron‐responsive element in the
APP‐mRNA 5ʹ‐untranslated region. Ann N Y Acad Sci 1035:34‐48.
Waheed A, Parkkila S, Saarnio J, Fleming RE, Zhou XY, Tomatsu S, Britton RS, Bacon BR, Sly
WS (1999) Association of HFE protein with transferrin receptor in crypt enterocytes of
human duodenum. Proc Natl Acad Sci U S A 96:1579‐1584.
Waheed A, Grubb JH, Zhou XY, Tomatsu S, Fleming RE, Costaldi ME, Britton RS, Bacon BR, Sly
WS (2002) Regulation of transferrin‐mediated iron uptake by HFE, the protein defective
in hereditary hemochromatosis. Proc Natl Acad Sci U S A 99:3117‐3122.
Waheed A, Parkkila S, Zhou XY, Tomatsu S, Tsuchihashi Z, Feder JN, Schatzman RC, Britton
RS, Bacon BR, Sly WS (1997) Hereditary hemochromatosis: effects of C282Y and H63D
mutations on association with beta2‐microglobulin, intracellular processing, and cell
surface expression of the HFE protein in COS‐7 cells. Proc Natl Acad Sci U S A 94:12384‐
12389.
Yamamoto A, Shin RW, Hasegawa K, Naiki H, Sato H, Yoshimasu F, Kitamoto T (2002) Iron
(III) induces aggregation of hyperphosphorylated tau and its reduction to iron (II)
217
reverses the aggregation: implications in the formation of neurofibrillary tangles of
Alzheimerʹs disease. J Neurochem 82:1137‐1147.
Zecca L, Youdim MB, Riederer P, Connor JR, Crichton RR (2004) Iron, brain ageing and
neurodegenerative disorders. Nat Rev Neurosci 5:863‐873.
VITA
Eric Christopher Hall II
EDUCATION The Pennsylvania State University, College of Medicine, Hershey, PA PhD in Neuroscience Program Completed in June 2009. Xavier University of Louisiana, New Orleans, LA Bachelor of Science in Biology, Minor in Chemistry Completed in May 2004. Southfield Christian High School, Southfield, MI Completed in June 2000. TEACHING EXPERIENCE Teaching assistant in Physiological Psychology and Physiological Lab at Lebanon
Valley College. Fall Semester, 2005.
ACTIVITIES AND LEADERSHIP AT PSU Institutional Review Board, Member, 2005-2009 Graduate Student Association – Vice-President, 2005-2006; President, 2006-2007
Penn State Hershey Diversity Office Task Force, Member; 2007-2008 Children’s Miracle Network Fundraising Committee, Member; 2006-2007 Intramural Basketball League-Penn State Hershey, Director; 2004-2005
HONORS Neuroscience Scholars Fellowship, Society for Neuroscience; 2007-2009 PSU College of Medicine Alumni Society Endowed Scholarship; 2007 Minority Leadership Fellowship, Biotechnology Institute; 2006 PSU Schreyer Institute, Teaching Certificate; 2006 Minority Scholars Fellowship, Biotechnology Institute; 2005 Huck Institutes of the Life Sciences Fellowship, PSU; 2004-2006 PUBLICATIONS
Hall II EC, Lee SY, Mairue N, Simmons Z, and Connor JR (2009). Expression of the HFE allelic variant H63D in SH-SY5Y cells affects tau phosphorylation. In Press. Neurobiology of Aging
Mairue N, Hall II EC, Lee SY, Cheepsunthorn P, and Connor JR (2009). The HFE
polymorphism H63D enhances activation of the intrinsic pathway of apoptosis. Submitted.
Hall II EC, Lee SY, and Connor JR (2009). Pin1 activity is compromised when
expressing the H63D allele of the HFE gene. Submitted.