Effects of apolipoprotein e on n methyl d aspartate receptor signalling the roles of ageing and chicken extract

EFFECTS OF APOLIPOPROTEIN E ISOFORMS
ON N-METHYL-D-ASPARTATE RECEPTOR
SIGNALLING:
THE ROLES OF AGEING AND CHICKEN
EXTRACT

YONG SHAN MAY
(B.Sc. (Hons), UM)

A THESIS SUBMITTED
FOR THE DEGREE OF DOCTOR OF
PHILOSOPHY

DEPARTMENT OF PHYSIOLOGY
YONG LOO LIN SCHOOL OF MEDICINE
NATIONAL UNIVERSITY OF SINGAPORE

2014


DECLARATION
I hereby declare that the thesis is my original work and it has been written by
me in its entity. I have duly acknowledged all the sources of information
which have been used in the thesis.

This thesis has not been submitted for any degree in any university previously.

__________________________
YONG SHAN MAY
23rd January 2014

ACKNOWLEDGEMENTS
I would like to express my gratitude to my supervisor, Dr. Wong Boon Seng
for his guidance and supervision throughout my years of PhD study. He has
taught me to think critically and helped me to grow as an individual. I am truly
grateful to my TAC team members, Dr. Lim Kah Leong and Dr. Ramani, for
the guidance provided in my project. I also would like to thank Dr. Low Chian
Ming for his advice to overcome several obstacles that I experienced. I extent
my gratitude to Dr. Paramjeet Singh (Cerebros Pacific Limited) for providing
the CE powder for my project.

I sincerely thank my previous and current fellow lab mates, Ching Ching, Li
Min, Hong Heng, Ray, Shiau Chen, Mei Li, Ira, Elizabeth, Cynthia, Pei Ling,
Ket Yin, Bei En and past FYP students for making my lab experience a
memorable and an enjoyable one. Special thanks to Irwin, Alvin, Francis,
Vanessa and Li Ren for their camaraderie and support. All of them have been
great mentors and friends, offering me advice and encouragement that helped
me to pull through the difficult times. I will always remember the time that we
have spent together and thank you for making a difference in my life.

I express my heartfelt gratitude to my family members for their support and
unconditional love. Last but not least, I want to thank my boyfriend, Chee Wai,
for being so understanding and supportive during my last two years of study.

i


TABLE OF CONTENTS
ACKNOWLEDGEMENTS ................................................................................ i
TABLE OF CONTENTS ...................................................................................ii

SUMMARY ...................................................................................................... vi
LIST OF TABLES ......................................................................................... viii
LIST OF FIGURES .......................................................................................... ix
ABBREVATIONS ............................................................................................ xi
LIST OF PUBLICATION...………………………………………………...xiv
Chapter 1: Introduction ...................................................................................... 1
1.1. Apolipoprotein E (ApoE) ....................................................................... 1
1.1.1 Characteristics of ApoE ................................................................ 1
1.1.2. Expression and functions of apoE in CNS: astrocytic vs neuronal
apoE........................................................................................................ 2
1.1.3. Differences between apoE isoforms............................................. 4
1.1.4. ApoE isoforms in synaptic plasticity ........................................... 5
1.1.5. ApoE receptors ............................................................................. 7
1.2. Alzheimer’s disease (AD) ...................................................................... 9
1.2.1. AD pathogenesis and progression ................................................ 9
1.2.2. Hippocampus in synaptic plasticity and memory formation...... 10
1.2.3. Ageing and synaptic plasticity ................................................... 12
1.2.4. Genetic risk factor for sporadic AD: apolipoprotein E (apoE)
variants ................................................................................................. 13
1.3. N-methyl-D-aspartate receptor (NMDAR): a key player in learning and
memory ...................................................................................................... 15
1.3.1. Characteristics and expression of NMDAR ............................... 16
1.3.2. Functions of NMDAR ................................................................ 19
1.3.3. Modulation of NMDARs by phosphorylation and
dephosphorylation mechanisms ........................................................... 20
1.3.4. Opposing roles of NMDAR subunits in synaptic plasticity ....... 21
ii


1.3.5. NMDAR in ageing ..................................................................... 22

1.3.6. NMDAR in AD .......................................................................... 23
1.3.7. NMDAR in excitotoxity............................................................. 24
1.4. α-amino-hydroxy-5-methyl-isoxazole-4-propionic acid (AMPAR) .... 25
Chapter 2: Materials And Methods .................................................................. 28
2.1. Common reagents and materials .......................................................... 28
2.2. Animal model ....................................................................................... 28
2.2.1. Tissue preparation for protein analysis ...................................... 29
2.2.2. Preparation of brain homogenates .............................................. 29
2.2.3. Protein quantitation of brain lysates ........................................... 30
2.2.4. Western blotting ......................................................................... 30
2.2.5. Tissue preparation for immunofluorescence .............................. 32
2.2.6. Immunofluorescence .................................................................. 32
2.3. Cell culture ........................................................................................... 35
2.3.1. Immortalization and transfection of apoE-knockout neuronal
cells ...................................................................................................... 35
2.3.2. CE treatment............................................................................... 36
2.3.3. Cell lysis ..................................................................................... 37
2.3.4. Protein quantitation of cell lysates ............................................. 37
2.3.5. Western blotting ......................................................................... 37
2.3.6. Calcium assay............................................................................. 39
2.4. Statistical analysis ................................................................................ 40
Chapter 3: Impaired NMDAR-induced Signalling In Aged ApoE4-KI Mice . 41
3.1. Introduction .......................................................................................... 41
3.1.1. Human apoE gene-knockin (apoE-KI) mice model for
investigating effects of apoE isoforms on NMDAR changes during
ageing. .................................................................................................. 41
3.1.2. Interactions between apoE and N-methyl-D-aspartate receptor
(NMDAR) ............................................................................................ 47
3.1.3. Postsynaptic density (PSD) proteins/ NMDAR-associated
proteins (NAPs) .................................................................................... 48

3.1.3.2. PSD95 ...................................................................................... 50
3.1.3.2. Calcium/calmodulin-dependent protein kinase II (CaMKII) ... 51
3.1.4. Signalling pathways coupled to NMDAR ................................. 53
3.1.4.1. Protein kinase C (PKC) pathway ............................................. 53
iii


3.1.4.2. Protein kinase A (PKA) pathway............................................. 55
3.1.4.3. Ras/mitogen-activated protein kinase (MAPK) pathway ........ 56
3.1.5. CREB in learning and memory .................................................. 59
3.2. Objectives of study ............................................................................... 61
3.3. Results .................................................................................................. 63
3.3.1. Expression of human apoE in brains of huApoE-knockin (apoEKI) mice across three time points......................................................... 65
3.3.2. Expression of apoE receptor and PSD95 in brains of huApoE-KI
mice across three time points ............................................................... 66
3.3.3. NMDAR subunits phosphorylation profile in brains of huApoEKI mice across three time points .......................................................... 68
3.3.4. PKA and PKC signalling profile in brains of huApoE-KI mice
across three time points ........................................................................ 71
3.3.5. GluR1 and αCaMKII expression profile in brains of huApoE-KI
mice across three time points ............................................................... 73
3.3.6. ERK-CREB signalling pathway in brains of huApoE-KI mice
across three time points ........................................................................ 76
3.3.7. Expression of human apoE in neurons and astrocytes of
huApoE- KI mice brains across three time points ............................... 79
3.3.8. Protein expression level of NMDAR subunits in brains of
huApoE-KI mice across three time points ........................................... 85
3.4. Discussion ............................................................................................ 89
3.4.1. ApoE4 isoform downregulates expression level of total apoE but
upregulates neuronal apoE production with increasing age ................. 89
3.4.2. ApoE4-isoforms decreases expression of apoE receptor i.e.

LRP1 and postsynaptic protein PSD95. ............................................... 95
3.4.3. ApoE4 spatio-temporally regulate NMDAR expression and
phosphorylation profiles during ageing. .............................................. 99
3.4.4. Modulation of ERK and CREB activity in apoE4-KI mice is
mediated via PKC but not PKA signalling pathway. ......................... 107
Chapter 4: Impacts of apoE isoforms in cellular responses to chicken extract
(CE) treatment ................................................................................................ 119
4.1. Introduction ........................................................................................ 119
4.1.1. Cyclic nucleotide phosphodiesterase (PDE) and CREB .......... 119
4.1.1. Chicken extract (CE): potential PDE inhibitor ........................ 120
4.1.2. Beneficial effects of CE to mental health ................................ 121
4.2. Objective of study .............................................................................. 124
4.3. Results ................................................................................................ 125
4.3.1. Chronic expression of apoE in huApoE stable cells ................ 125
iv


4.3.2. Effects of CE treatment on expression of human apoE in
huApoE stable cells ............................................................................ 126
4.3.3. Effects of CE treatment on basal intracellular calcium level in
huApoE stable cells ............................................................................ 127
4.3.4. Effects of CE treatment on phosphorylation of NR1 subunit in
huApoE stable cells ............................................................................ 129
4.3.5. Effects of CE treatment on PKA and PKC signalling profile in
huApoE stable cells ............................................................................ 131
4.3.6. Effects of CE treatment on GluR1 and αCaMKII expression
profile in huApoE stable cells ............................................................ 134
4.3.7. Effects of CE treatment on ERK-CREB signalling in huApoE
stable cells .......................................................................................... 136
4.4. Discussion .......................................................................................... 138

4.4.1. Different intracellular calcium responses in huApoE-transfected
neurons and downregulation of apoE4 expression upon CE treatment
............................................................................................................ 138
4.4.2. Upregulation of PKC pathway in huApoE3-transfected neurons
and downregulation of PKA pathway in huApoE4-transfected neurons
upon CE treatment.............................................................................. 141
4.4.3. ERK-CREB signalling in CE-treated huApoE3- and huApoE4transfected neurons............................................................................. 145
4.3 Conclusion........................................................................................... 152
4.4 Future Directions ................................................................................. 154
BIBLIOGRAPHY .......................................................................................... 162
Appendix ........................................................................................................ 225

v


SUMMARY
Apolipoprotein E4 (apoE4) isoform has been shown to accelerate cognitive
decline in human and mouse models during ageing compared to apoE3
isoform. Mice expressing human apoE4 display impaired learning and
memory and glutamatergic neurotransmission. Despite the ongoing studies to
look for preventive measures and therapeutic strategies, researchers have yet
to unravel the complex underlying mechanisms and rectify the pathological
effect of apoE4 in learning and memory.

My study shows that apoE4 regulates expression of NMDAR subunits and its
activity in a temporal and region-specific manner during ageing when
compared with apoE3-knock in (apoE3-KI) mice. Western blotting analyses
show an increased phosphorylation of NR1 subunit particularly at Ser896 in
young apoE4-KI mice. In contrast, this phosphorylation is downregulated in
old apoE4-KI mice when compared with apoE3-KI mice. The tyrosine

phosphorylation of NR2A of apoE4-KI mice is reduced regardless of age
whereas there is no difference in NR2B activity between apoE3- and apoE4KI mice across all time-points. Immunofluorescence studies demonstrate an
increase in NR1 signal intensity in the hippocampus and cortex at week 12
followed by downregulation of its total expression at week 72 in the
hippocampus. Similarly, NR2A subunit expression levels in most hippocampal
subregions and cortex of apoE4-KI mice are always lower than age-matched
apoE3 counterparts. The NR2B expression levels show an unexpected
decrease in the hippocampus at middle age which extend into the cortex at old
age. The age-dependent and region-specific modulation of the NMDAR
subunits is correlated to the source of apoE4 production. This suggests that the
increasing neuronal apoE4 production at old age particularly in the CA3 and
cortical area, exerts a detrimental effect on NMDAR expression due to the
neurotoxicity of apoE4 fragments. Hence, the downregulated phosphorylation
and region-specific expression of NMDAR of apoE4-KI mice may partially
explain their impaired behavioural performances at old age compared to agematched apoE3 counterparts as observed by others.

vi


Analysis on NMDAR-associated proteins and the signalling pathways coupled
to NMDAR activity in apoE4-KI mice show that the postsynaptic density
protein PSD95 and apoE receptor, LRP1, are downregulated in apoE4-KI mice
at all ages. This suggests that a reduced NMDAR functionality mediated via
these proteins. Immunoblotting of Ca2+-sensitive kinases including αCaMKII,
PKCα and PKA-Cα demonstrate an increased αCaMKII and PKCα activation
with expected elevation in their common downstream target, GluR1 phosphoS831 without affecting PKA-Cα in young apoE4-KI mice. In contrast,
phosphorylation of αCaMKII, PKCα and GluR1 Ser831 are downregulated at
old age. The correspondence between signalling profiles of ERK-CREB
versus αCaMKII and ERK-CREB versus PKCα strongly suggest their key
roles played in facilitating ERK and CREB activation.

Second part of my study investigated the effects of chicken extract (CE) on the
NMDAR and its downstream signalling cascades in the context of apoE
isoforms in vitro as this dietary supplement has been shown to improve
cognition in human. Howver, the underlying mechanisms are unclear. Data
from CE treatment of apoE-transfected neurons demonstrate the differential
effects of apoE isoforms on intracellular Ca2+ responses and triggering of the
Ca2+-dependent signalling pathways. In particular, PKA-Cα pathway is
downregulated whilst PKCα pathway is upregulated in CE-treated apoE4 and
apoE3 neurons respectively. Moreover, the basal intracellular Ca2+ level,
αCaMKII and GluR1 S831 are increased in apoE3 neurons whereas the
opposite occurs for mock and apoE4-transfected neurons after treatment.
These might have led to enhanced ERK1/2 and CREB activity in apoE3
neurons but reduced ERK-CREB signalling in mock as well as apoE4 neurons.

(564 words)

vii


LIST OF TABLES
Table 2.1. List of primary antibodies used for immunofluorescence. The
source and the dilution factor used are as shown. ............................................ 34
Table 2.2. Composition of CE compound ....................................................... 37
Table 2.3. List of primary antibodies used for immunoblotting. The source and
the dilution factor used are as shown. .............................................................. 38
Table 2.4. List of secondary antibodies used throughout the study. The
purpose, source and dilution factor used are as shown. ................................... 39
Table 3.1. Samples, brain regions of interest and findings of reviewed articles
.......................................................................................................................... 46

Table 3.2. Recapitulative table on all significant comparisons between
huApoE3 and huApoE4 mouse lines. ............................................................ 117
Table 4.1. Recapitulative table on the findings of CE treatment on mock,
huApoE3 and huApoE4 stable cell lines. ...................................................... 150

viii


LIST OF FIGURES
Figure 1.1. Amino acid sequence of different apoE isoforms .......................... 2
Figure 1.2. Structure of NMDAR subunit ...................................................... 18
Figure 1.3. Model of bidirectional plasticity in AMPAR phosphorylation and
dephosphorylation ............................................................................................ 27
Figure 3.1. Schematic diagram of the interactions between apoE and NAPs,
and NMDAR-coupled signalling cascades. ..................................................... 61
Figure 3.2. Expression level of huApoE in brains of apoE-KI mice models
across three time points i.e. 12, 32 and 72 weeks. ........................................... 65
Figure 3.3. Protein expression level of LRP1 and PSD95 in brains of apoE-KI
mice models across three time points i.e. 12, 32 and 72 weeks. ...................... 66
Figure 3.5. Protein expression level of PKA-Cα and PKCα in brains of apoEKI mice models across three time points i.e. 12, 32 and 72 weeks. ................ 71
Figure 3.6. Protein expression level of GluR1 and αCaMKII in brains of apoEKI mice models across three time points i.e. 12, 32 and 72 weeks. ................ 73
Figure 3.7. Protein expression level of ERK and CREB in brains of apoE-KI
mice models across three time points i.e. 12, 32 and 72 weeks. ...................... 76
Figure 3.8. Cellular expression levels of huApoE in brains of apoE-KI mice
models across three time points i.e. 12, 32 and 72 weeks. .............................. 82
Figure 3.9. Expression level of total NMDAR subunit proteins in brains of
apoE-KI mice models across three time points i.e. 12, 32 and 72 weeks. ....... 87
Figure 3.10. Model of age-dependent regulation of intracellular signalling
pathways by apoE4. ....................................................................................... 118
Figure 4.1. Protein expression level of apoE in mock and apoE-transfected

neurons. .......................................................................................................... 125

ix


Figure 4.2. Protein expression level of apoE in mock and apoE-transfected
neurons. .......................................................................................................... 126
Figure 4.3. Basal intracellular calcium ion concentration in mock and apoEtransfected neurons. ....................................................................................... 127
Figure 4.4. Phosphorylation level of NR1 subunit in mock and apoEtransfected neurons. ....................................................................................... 129
Figure 4.5. Protein expression level of PKA-Cα and PKCα in mock and apoEtransfected neurons. ....................................................................................... 131
Figure 4.6. Protein expression level of GluR1 subunit and αCaMKII in mock
and apoE-transfected neurons. ....................................................................... 134
Figure 4.7. Protein expression level of ERK and CREB in mock and apoEtransfected neurons. ....................................................................................... 136
Figure 4.8. Model of differential regulation of cellular responses to CE
treatment by apoE isoforms. .......................................................................... 151
Supplementary Figure 1. Protein expression levels of GFAP and NeuN in
brains of apoE-KI mice models across three time points i.e. 12, 32 and 72
weeks.............................................................................................................. 225
Supplementary Figure 2. Protein expression level of LP1 in mock and apoEtransfected neurons. ....................................................................................... 225

x


ABBREVIATIONS
AC

Adenylyl cyclase

AD

Alzheimer’s disease

AMPAR

α-amino-3-hydroxy-5-methyl-4-isoxazolepropionic
acid receptor

ApoE

Apolipoprotein E

ApoER2

ApoE receptor 2

APP

Amyloid precursor protein

ATP

Adenosine triphosphate

Bcl2

B-cell lymphoma 2

BDNF

Brain-derived neurotrophic factor

CA

Cornus ammoni

CaMK

Calcium-calmodulin dependent kinase

cAMP

Cyclic adenosine monophosphate

CaN

Calcineurin

CE

Chicken extract

CNS

Central nervous system

CREB

cAMP-responsive element-binding protein

CSF

Cerebrospinal fluid

DAG

Diacylglycerol

DG

Dentate gyrus

EDTA

Ethylenediaminetetraacetic acid

EPSC

Excitatory postsynaptic current

ERK

Extracellular-signal regulated kinase

FDG-PET

Fluorodeoxyglucose positron emission tomography

GFAP

Glial fibrillary acidic protein

GLUT

Glutamate transporter

GRF

Guanine nucleotide Releasing Factor

GTP

Guanosine triphosphate

HDL

High density lipoprotein

HFS

High frequency stimulation

ICD

Intracellular domain

iGluR

Ionotropic glutamate receptor

Ins(1,4,5)P3

Inositol-1,4,5-triphosphate

JNK

Jun-N-terminal kinase

xi


KI

Knockin

KO

Knockout

LDLR

Low density lipoprotein receptor

LFS

Low frequency stimulation

LRP1

LDL-related protein 1

LTD

Long-term depression

LTP

Long-term potentiation

mAchR

Muscarinic acetylcholine receptor

MAGUK

Membrane-associated guanylate kinase

MAP2

Microtubule-associated protein 2

MAPK

Mitogen activated protein kinase

MCI

Mild cognitive impairment

MF

Mossy fibre

mGluR

Metabotropic glutamate receptor

MK801

[5R,10S]-[+]-5-methyl-10,11-dihydro-5H-dibenxo
[a,d]cyclohepten-5,10-imine

mRNA

Messenger ribonucleic acid

MWM

Morris water maze

NMDAR

N-methyl-D-aspartate receptor

NO

Nitric oxide

NRHyper

NMDAR hyperactivity

NRHypo

NMDAR hypoactivity

NSE

Neuron-specific-enolase

Oxy-Hb

Oxyhemoglobin

PDE

Phosphodiesterase

PDZ

PSD95, disc large, zona occludens-1

PI3K

Phosphatidylinositol 3-kinases

PKA-C

Protein kinase A catalytic subunit

PKC

Protein kinase C

PMA

Phorbol 12-myristate 13-acetate

PP1/2A

Phosphatases 1,2A

PSD95

Postsynaptic density protein 95

RSK

Ribosomal protein S6 kinase

SAP

Synapse-associated-protein

SC

Schaeffer collateral

SDS

Sodium dodecyl sulfate

xii


tPA

Tissue plasminogen activator

TR

Targeted replacement

U0126

dicyano-1,4-bis[2-aminophenylthio]butadiene

α2M

Alpha-2 marcoglobulin

xiii


LIST OF PUBLICATION

S. Yong, Q. Ong, B. Siew and B. B. Wong, Food Funct., 2014, DOI:
10.1039/C4FO00428K.

xiv

Chapter 1: Introduction

Chapter 1: Introduction
1.1. Apolipoprotein E (ApoE)
1.1.1 Characteristics of ApoE
ApoE is a 34 kilodalton (kDa) protein bearing 299 amino acids in sequence.
ApoE is characterized by a single nucleotide polymorphism (SNP) on the gene
located at chromosome 19. The majority of the human population is
homozygous for apoE3 with an allelic frequency of 78% among Europeans.
ApoE4 makes up about 14% of the population and the rest are apoE2 (8%). In
a study of apoE polymorphisms and lipid profiles in three ethnic groups i.e.
Malays, Chinese and Indians in the Singapore population, ε3 allele is the most
common (82%) followed by ε4 (10%) and ε2 (8%) (Tan et al., 2003). The
three major isoforms predominantly expressed in the human population are
apoE2, apoE3 and apoE4. These isoforms differ by two amino acids at
position 112 and 158 in which apoE2 has two cysteines (Cys), apoE3 has Cys112 and Arg-158 (arginine) whereas apoE4 has Arg-112 and Arg-158 (Rall et
al., 1982; Weisgraber, 1994). In general, apoE has two structural domains
including a 22kDa N-terminal domain which binds to its receptor and a 10kDa
C-terminal domain which is the lipid-binding site that can interact with other
extracelullar proteins such as Aβ (Mahley and Rall, 2000). In the CNS, apoE
exists in the form of discoidal-shaped high-density lipoprotein (HDL)-like
particles containing phospholipids and cholesterol which is distinct from those
in the peripheral system (DeMattos et al., 2001a; Pitas et al., 1987). It is the
major lipoprotein in the CNS among other apolipoproteins such as apoA, apoC,
apoD and apoJ (also known as clusterin) whilst apoB is absent (Holtzman et
al., 2012). Concentration of apoE in the cerebrospinal fluid (CSF) is 100-200
nM (Riemenschneider et al., 2002) and total concentration in brain extract is
approximately 5 µg/mL (Haass and Selkoe, 2007).

1


Chapter 1: Introduction

Figure 1.1. Amino acid sequence of different apoE isoforms
The three major apoE isoforms differ in their amino acids at positions 112 and
158 which give rise to their distinct properties. Each comprises two structural
domains that bind to apoE receptors at the N-terminal and lipid particles such
as HDL at the C-terminal.

1.1.2. Expression and functions of apoE in CNS: astrocytic vs neuronal
apoE
Under normal conditions, apoE is mainly synthesized by astrocytes but can
also be found in oligodendrocytes, neurons, smooth muscle cells and choroid
plexus in the CNS (Boyles et al., 1985; Herz and Beffert, 2000; Xu et al.,
2006). Astrocyte-derived apoE is involved in the transport of cholesterol into
neurons to regulate synaptic plasticity through lipid homeostasis (Gong et al.,
2002; Mauch et al., 2001). It is antioxidative in nature and helps in the
clearance of Aβ (DeMattos, 2004; Lomnitski et al., 1999; Ye et al., 2005).
Besides that, apoE mediates neurite outgrowth and stabilization of
microtubules. These structural changes affect synapse formation and hence
influence synaptic plasticity (Nathan et al 2002).

2


Chapter 1: Introduction
Under pathological conditions, excitotoxic injury in CNS such as kainic acid
treatment, oxidative stress and even ageing induces neurons to increase apoE

production to initiate the repair process and mediate protection from these
insults (Boschert et al., 1999; Huang, 2010; Mahley, 1988; Roses, 1997; Xu et
al., 2006). However, proteolytical processing of neuronal apoE4 produces
harmful fragments that induce neurodegeneration especially in ε4 carriers
(Brecht et al., 2004; Huang, 2010; Huang et al., 2004; Mahley et al., 2006;
Roses, 1997) which may underlie its correlation to the earlier onset of
Alzheimer’s disease (AD) (Boschert et al., 1999). Both in vitro and in vivo
studies have shown that neuronal apoE4 is more susceptible to cleavage by a
neuron-specific, chymotrypsin-like serine protease generating neurotoxic
fragments which are detrimental to the neurological repair or maintenance
process compared to apoE3 (Huang et al., 2001; Tolar et al., 1999; Tolar et al.,
1997).
The effect of the ε4 allele on the CNS apoE protein levels has been studied in
both AD patient CNS and animal models, reporting controversial results
showing either reduced levels (Beffert et al., 1999; Ong et al., 2014; Poirier,
2005; Riddell et al., 2008), no change (Fryer et al., 2005; Sullivan et al., 2004)
or increases (Fukumoto et al., 2003) in apoE expression compared to that of
apoE4 individuals. The reduced brain apoE level in ε4 carriers (Farrer et al.,
1997; Poirier, 2005) and their predisposition to AD implies that apoE is
required to sustain a certain level of cognitive function perhaps by maintaining
synaptic integrity.

It is suggested that adequate level of apoE may be crucial to regulate brain
homeostasis during ageing as apoE expression increases in the liver in an agedependent manner (Gee et al., 2005). However, it is unclear whether apoE
expression changes in the brain with ageing in human whereas conflicting
observation have been made from animal studies. In rodents, apoE expression
level decreases more than five-fold in the cortex and hypothalamus (Jiang et
al., 2001) but increases in the hippocampus (Terao et al., 2002) of aged mice.
Aged rats also demonstrate elevated glial apoE expression in basal ganglia and
corpus callosum (Morgan et al., 1999). In contrast, a recent study reported

3


Chapter 1: Introduction
there is no alteration in apoE mRNA and protein expression level in the cortex,
hippomcapus and striatum of aging rat brains (Gee et al., 2006).

1.1.3. Differences between apoE isoforms
The replacement of Arg in apoE4 at position 112 has an impact on its structure
and functions making it the least stable and a more pathological isoform
compared to the other two. Arg-112 mediates the interaction between Arg-61
at the N-terminus and Glu-255 (glutamine) at the C-terminus causing the
whole molecule to exist as a partially folded intermediate or molten globule
(Dong et al., 1994; Weisgraber, 1990). This unstable structure denatures at a
lower temperature making it less concentrated in the CNS. In fact there is a
lower level of apoE in brain and serum of AD subjects whereby 40 to 80% of
them have at least one ε4 allele (Farrer et al., 1997). ApoE4 has a lower
cholesterol transport capacity and Aβ clearance ability compared to apoE2 and
apoE3 leading to increased Aβ production (Dodart et al., 2005; Holtzman et al.,
2000). Furthermore, apoE4 mediates some of the detrimental effects such as
tau1 phosphorylation (Mandelkow and Mandelkow 1994), lysosomal leakage,
mitochondrial dysfunction, neurodegeneration and cognitive deficits in AD
(Buttini et al., 2002; Chang et al., 2005; Huang et al., 2001; Ji et al., 2002;
Raber et al., 2002; Risner et al., 2006). ApoE4 has been shown to cause
cytotoxicity in a dose- and time-dependent manner. Significant neurotoxicity
was detected 24 hours after apoE4 treatment and the percentage of cell death
plateaus at 72 hours (Qiu et al., 2003).
Generation of neurotoxic fragments such as apoE4 (Δ1-272, Δ272-299, Δ127242) are harmful to neurons and Neuro-2a cell lines derived from mouse
neuroblastoma.

For

instance,

apoE4

1

(Δ1-272)

is

associated

with

Tau is one of the microtubule-associated proteins that plays a role in the
stabilization of neuronal microtubules and provides the tracks for intracellular
transport. However, when tau undergoes modifications such as phosphorylation,
hyperphosphorylated tau will aggregate into paired helical fragments that coalesce
into neurofibrillary tangles.

4


Chapter 1: Introduction
neurofibrillary tangle-like structures in cortical and hippocampal neurons in
transgenic mice (Harris et al., 2003; Huang et al., 2001). ApoE4 (Δ272-299)
which is present in AD brains peaks at 6-7 month old in mice and impairs
spatial learning and memory (Brecht et al., 2004; Harris et al., 2003).

Furthermore, AD patients also have much higher levels of apoE fragments
compared to non-demented controls (Harris et al., 2003; Huang et al., 2001).
Thus under pathological conditions, it is worse for neurons to express apoE4
than to express no apoE at all, as opposed to apoE2 and apoE3 which are more
stable and more effective in maintenance of neurons (Mahley, 1988; Morrow
et al., 2002). In other words, apoE4 has an adverse gain-of-function compared
to the other two isoforms (Buttini et al., 2000; Huang, 2006; Raber et al.,
2000).

1.1.4. ApoE isoforms in synaptic plasticity
ApoE isoforms differentially modulate synaptic plasticity and learning and
memory. ApoE-targeted replacement mice (huApoE-TR) or human apoEknockin (huApoE-KI) expressing human apoE isoforms have shown that
huApoE3 mice have enhanced LTP compared to huApoE4 mice (Grootendorst
et al., 2005; Trommer et al., 2004). ApoE4 has been postulated to impair
synaptic plasticity by reducing NMDAR-dependent Ca2+ influx mediated by
apoE receptors in a dose-dependent manner. Furthermore, apoE4 may impede
apoE receptor recycling by sequestering the receptors intracellularly such that
they are unable to be expressed on neuronal surfaces and trigger downstream
signalling pathways upon ligand binding (Chen et al., 2010). Exogenously
added apoE4 has been shown to reduce surface expression of apoE receptor 2
(apoER2) in primary neurons which functions as a dual receptor for apoE and
reelin, a ligand that modulates NMDAR functions both in vitro and in vivo
(Beffert et al., 2005). ApoE3 and apoE4 differ from each other in their
influences on neurite extension (DeMattos et al., 1998; DeMattos et al., 2001b;
Fagan et al., 1996; Handelmann et al., 1992; Nathan et al., 1995). In cultured
dorsal root ganglion neurons and Neuro-2a cells, apoE3 together with βVLDL stimulates neurite branching and extension as opposed to apoE4
5


Chapter 1: Introduction

(Holtzman et al., 1995; Nathan et al., 1994). This stimulatory effect on neurite
outgrowth is also seen in rat hippocampal neurons mediated by astrocytederived apoE3 but not apoE4 (Sun et al., 1998). In organotypic hippocampal
slice culture isolated from huApoE-TR mice model, apoE3 can stimulate
neuronal sprouting which is inhibited by apoE4 (Teter et al., 2002). This
contrasting effect of apoE4 is probably due to its facilitation of tubulin
depolymerization in neuronal cells leading to microtubule instability (Nathan
et al., 1994). In addition, in vivo and in vitro studies show apoE4 impedes
synaptogenesis as apoE4 transgenic mice have reduced number of synapses
per neuron (Cambon et al 2000), and lower density of dendritic spines than
apoE3 mice (Dumanis et al., 2009; Ji et al., 2003). Similarly, exogenously
added apoE4 or its proteolytic fragments decreases spine density in primary
cortical neuronal cultures (Brodbeck et al., 2008). These structural deformities
in apoE4 mice perhaps contribute to learning and memory deterioration.

In spite of all these, young apoE4-KI mice display enhanced plasticity
compared to apoE3 mice but this effect disappears with age (Kitamura et al.,
2004). Furthermore, this enhanced LTP may be mediated by postsynaptic
properties as there is no difference in presynaptic release mechanisms between
apoE3 and apoE4-KI animals. Similarly, young human ε4 carriers exhibit
higher intelligence and event-related potential (Yu et al., 2000) suggesting that
apoE isoform modification of LTP may be age-dependent.

Studies that

support the beneficial effect of apoE4 in early life show that school children
with ε4 allele score better in verbal fluency scores and outperform ε2 peers in
Rey Complex Figure Test (RCFT) copy trial (Alexander et al., 2007; Bloss et
al., 2010). One of the interesting observations is that ε4 carriers tend to recruit
task-related

regions

more

intensely

than

non-carriers

and

exhibit

compensatory neural recruitment (Han and Bondi, 2008). This phenomenon
occurs even in non-demented older adults suggesting ε4 carriers undergo
compensatory changes in order to achieve the same performance as noncarriers (Kukolja et al., 2010; Tuminello and Han, 2011). However, the
beneficial role of apoE4 in early life remains controversial as there are
contradictory findings indicating that 5 to 7-year-old ε4 carriers with sleep
apnoea show impaired cognition and middle-aged adults do not differ in
6


Chapter 1: Introduction
cognitive performance (Dennis et al., 2010; Filbey et al., 2010; Gozal et al.,
2007). Furthermore, additional environmental risk factors such as family
history of AD and gender susceptibility may interact with ε4 allele to affect
cognition as 7 to 10 year-old girls have lower spatial memory retention and
lower visual recall scores on Family Pictures test (Acevedo et al., 2010; Bloss
et al., 2008). Perhaps the potential benefit of ε4 allele may be confined to a

very narrow window in early life span and as the detrimental effects of apoE4
surfaces, the compensatory mechanism cannot sustain the premorbid cognitive
performance levels and hence the condition worsens with age.

1.1.5. ApoE receptors
ApoE binds to members of low density lipoprotein receptor (LDLR) family
including very low density lipoprotein receptor (VLDLR), apoER2, LDLrelated protein 1 (LRP1), LDL-related protein-1B (LRP-1B), multiple EGF
repeat-containing protein-7 (MEGF7) and megalin (Brecht et al., 2004;
Holtzman et al., 2012). These receptors have one or more ligand-binding
domains, epidermal growth factor (EGF) and a cytoplasmic tail containing a
consensus amino acid NPxY motif which acts as an interaction site for
intracellular adaptor proteins that are further coupled to downstream
transduction pathways (Beffert et al., 2005; Gotthardt et al., 2000; Hoe et al.,
2006; Trommsdorff et al., 1998). Some of the functions of apoE receptors are
utilization of mitogen-activated protein kinases (MAPK), tyrosine kinases,
lipid kinases and ligand-gated ion channels such as glutamate receptors
(Beffert et al., 2005; Bock and Herz, 2003).

LRP1 is one of the major apoE receptors that regulates CNS apoE levels. It is
a 600 kDa receptor dimer consisting of an extracellular and intracellular
domain which is 515 kDa and 85 kDa respectively. It is expressed
ubiquitously by hepatocytes, neurons, vascular smooth muscle, lung,
macrophages and embryonic tissues (Bu et al., 1994; Ishiguro et al., 1995;
Moestrup et al., 1992). Other than apoE, ligands that bind to LRP1 are
activated α2-macroglobulin (α2M*), amyloid precursor protein (APP), tissue
plasminogen

activator

(tPA),

protease/protease
7

inhibitor

complexes,


Chapter 1: Introduction
lipoprotein lipase and platelet-derived growth factor (PDGF). It plays dual role
in endocytosis and regulation of signal transduction involving calcium
currents, PDGF receptor signalling, phagocytosis of apoptotic cells and
embryonic development (Boucher et al., 2003; Herz and Strickland, 2001;
May et al., 2002; Wang et al., 2003; Yepes et al., 2003). Genetic deletion of
LRP1 causes embryonic lethality and hence tissue-specific LRP1 knockout
animals have been generated which exhibit pronounced tremor, ataxia,
hyperactivity, dystonia and premature death without structural abnormalities
in these animals (Herz et al., 1992; May et al., 2004). They also show
impaired

brain

lipid

metabolism,

age-dependent

synaptic

loss

and

neurodegeneration (Liu et al., 2010). This highlights the crucial role of LRP1
in modulation and turnover of synaptic proteins that contributes to the
anomalies in morphology and phenotype in the absence of LRP1. As an
endocytic receptor, LRP1 facilitates uptake and clearance of neural proteases
such as neuroserpin from synaptic regions, as increased level of proteases will
disrupt the balance between proteases and protease inhibitors leading to
synaptic dysfunction (Makarova et al., 2003).
LRP1 has been linked to AD due to its binding to and clearance of Aβ as well
as APP metabolism (Beffert et al., 1999; Rebeck et al., 1993; VázquezHiguera et al., 2009), mediated via adaptor protein FE65 that binds to NPxY
motif of LPR1 and APP (Kinoshita et al., 2001; Pietrzik et al., 2004;
Trommsdorff et al., 1998). It is observed that decreased LRP1 expression in
amyloid mouse model and brain capillaries of AD brain may contribute to
impaired Aβ clearance (Deane et al., 2004; Van Uden et al., 2002). LRP1
undergoes proteolytic processing whereby the extracellular domain can be
cleaved by β-secretase, BACE1, which also cleaves APP and generates the Aβ
N-terminus, close to the transmembrane region resulting in the shedding of the
extracellular domain (Quinn et al., 1999). The intracellular domain is then
released from the plasma membrane by cleavage involving γ-secretase (May et
al., 2002). This allows the free C-tail to translocate together with its adaptor
proteins to other subcellular localizations. Moreover, the circulating LRP1
may also serve as a peripheral 'sink' for Aβ clearance (Sagare et al., 2007).

8

Chapter 1: Introduction
1.2. Alzheimer’s disease (AD)
AD is the most common cause of dementia among the elderly (>65 years) and
accounts for up to 75 percent of all dementia cases. Global prevalence of
dementia is estimated to be 3.9 % in people aged 60 years and above, and
affects more than 25 million people in the world (Brookmeyer et al., 2007;
Ferri et al., 2005; Wimo et al., 2003). Population ageing has become a global
phenomenon. In both developed and developing countries, the United Nation
Ageing Program and the United States Centers for Disease Control and
Prevention have projected that the aged population (65 years or more) in the
world is expected to rise from 420 million in 2000 to nearly 1 billion by 2030,
which is an increase from 7 to 12 percent of world population (Centers for
Disease Control and Prevention, 2003). Similarly, a recent study reported the
worldwide total number of 26.6 million AD patients will quadruple by the
year 2050 (Brookmeyer et al., 2007). With the incidence rate of 5 million
cases per year and the age-specific prevalence of AD doubling every five
years after the age of 65 years, Alzheimer's disease has rapidly become a
major public health crisis by imposing financial and societal burden as 43% of
AD patients require high level of care such as nursing homes and institutions.
The societal cost of dementia worldwide was estimated to be more than
US$315 billion in 2005 (Wimo et al., 2007). Hence, much effort has been
made to seek for effective therapeutic and preventive strategies as even a 1year delay in the onset of and progression of AD will significantly reduce the
global burden of this disease (Brookmeyer et al., 1998; Brookmeyer et al.,
2007).

1.2.1. AD pathogenesis and progression
AD is a multifactorial disease which is subtyped into two groups i.e. familial
making up 5% of the cases, and sporadic, for the rest. The former, also known
as early-onset AD, is caused by rare autosomal inherited mutations involving
presenillin 1 and 2 (PS1 and PS2), and the amyloid precursor protein gene

(APP), which typically occurs before age 65. The age of onset for the latter is

9