Protective effects of s propargylcysteine (SPRC) on in vitro neuronal damage induced by amyloid beta (25 35 p 1 3

PROTECTIVE EFFECTS OF S-PROPARGYLCYSTEINE (SPRC) ON IN VITRO
NEURONAL DAMAGE INDUCED BY AMYLOID-BETA(25-35)








WONG WAN HUI
B.Sc. (Hons)
National University of Singapore






A THESIS SUBMITTED

FOR THE DEGREE OF DOCTOR OF PHILOSOPHY

DEPARTMENT OF PHARMACOLOGY

NATIONAL UNIVERSITY OF SINGAPORE

2011





DECLARATION

I hereby declare that this thesis is my original work and it has been written
by me in its entirety. I have duly acknowledged all the sources of information
which have been used in the thesis.
This thesis has also not been submitted for any degree in any university previously.




__________________________________
Wong Wan Hui




i

ACKNOWLEDGEMENTS

I would like to thank the following people for making this thesis possible.
Firstly, I would like to convey my heartfelt thanks to my supervisor A/P Zhu Yi-Zhun
for his support and guidance the past seven years in his laboratory. I have learnt much and grew
to be a better researcher with the independence and freedom he has always granted me. Without
his encouragement I would never have embarked on this journey of self-discovery and learning.
I would also like to thank my mentors Dr Wang Hong, Dr Sonja Koh and Dr Wang
Zhong Jing for their patience and guidance. Dr Wang Hong had been my mentor in both
research and life. I will never forget all the times we spent having lunch together and discussing
about a wide range of issues. Dr Sonja Koh had made herself always available to answer my
doubts. Her presence had been inspiring with her creativity and I always felt assured with her
around. Dr Wang Zhong Jing was the first person who introduced me to the world of hydrogen
sulfide and research with his passion. He taught me how to think through problems and to
analyse research questions that made me a scientist today. For this much guidance from such
great scientists, I am most grateful.
Thirdly, I would like to extend my gratitude to the Head of Department and his staff
from the Department of Pharmacology, Yong Loo Lin School of Medicine. I had been in this
department since my undergraduate days. I have been nurtured on this unique field of study by
wonderful and passionate teaching staff. I also had the honour of knowing and working with
most of the staff in our department. Regardless of their positions, various staff from the
department never failed to make me feel a part of the family, and always ready to extend their
ii

help to me as a student and colleague alike. It made my research journey so much more
enjoyable with their support.
To my fellow seniors and juniors from this laboratory throughout these seven years, it
had been a great honour to know each and every one of them. They have given me their
unbridled comments and encouragement through the times at lab meetings. They have also
inspired me with their perseverance and pulled me along. Their friendship and camaraderie
accompanied me on the most lonely days.

I would also like to thank my family members who have given me their unconditional
support through this arduous time. They have never doubted my ability even when I felt lost, and
they have given me the motivation to stand up again despite falling down so many times. Even
through the difficult twelve-hour incubation times when I have to return to office at nights, my
family would encourage me whenever I wanted to give up. This thesis would not have been
possible without them.
I want to specially mention my best friends Shi Ping, Shixin and Shu Min for their
encouragement and belief in me throughout this time. Even though they might not be in the same
field of study, they were willing to lend a listening ear to me whenever I felt down. In fact,
without Shixin’s encouragement in the form of a bamboo story, I would still be that bamboo
farmer who will give up because I cannot see the shoots growing. I will also remember the times
Shumin and I discuss our research topics over coffee, and how she told me to doubt in order to
grow. I am most blessed to have these best friends.
Lastly, I would like to thank my fiancé Isaac for being with me in this journey. He had
brought out the best in me, and allowed me to be myself. He was always patient to listen to my
iii

woes and though he cannot understand the difficult pharmacological terms, he could empathise
with my difficulties. I will never forget the times he painstakingly tried to help me organize my
thoughts for this thesis as he struggled to study for his own exams. I will always be grateful to
his understanding and patience, such that he would wait for me to finish my experiments and
accompany me through those rushed dinners. I am thankful to have him in this journey and will
be happy to have him for the rest of my life.
I am most grateful for this chance to write this thesis and most importantly, a chance to
prove myself as a researcher and a person. I have grown to be more patient and determined from
this experience. And most importantly, I have learnt to pick myself up whenever I fall. This is
something I am sure, will help me through my life.

iv

TABLE OF CONTENTS
Acknowledgements i
Abstract viii
List of Tables x
List of Figures xi

CHAPTER 1: INTRODUCTION 1
1.1 GENERAL INTRODUCTION 1
1.1.1 Alzheimer’s Disease (AD) 1
1.1.2 Aggregated proteins and diseases 2
1.1.3 Self-aggregation properties of amyloid peptides 2
1.1.4 Factors affecting fibril formation 4
1.2 AMYLOID-BETA PEPTIDE (Aβ) 6
1.2.1 Products of sequential cleavage 6
1.2.2 Neurotoxicity of Aβ
25-35
8
1.2.3 Structure of Aβ
25-35
8
1.3 GARLIC AND S-ALLY-L-CYSTEINE (SAC) 10
1.3.1 Aged garlic extract 10
1.3.2 S-ally-L-cysteine 10
1.4 S-PROPARGYL-L-CYSTEINE (SPRC) 13
1.4.1 Chemical properties and pharmacokinetics 13
1.4.2 Cardioprotective effects 14
1.4.3 Neuroprotective effects 16
1.5 HYDROGEN SULFIDE (H
2
S) 17

1.5.1 General properties 17
1.5.2 Synthesis 17
1.5.3 Biological targets of H
2
S 19
1.5.4 H
2
S as a neuromodulator 20
1.5.5 H
2
S in AD 21
1.6 OXIDATIVE STRESS 23
1.6.1 Oxidative balance in the brain 23
1.6.2 Imbalance in the disease state 24
1.6.3 Aβ-induced oxidative stress 25
1.6.4 Antioxidant therapy in AD 29
1.7 INFLAMMATION 31
v

1.7.1 Inflammation in AD patients 31
1.7.2 Role of inflammation in the brain and disease state 32
1.7.3 Inflammatory mediators in AD 35
1.7.4 Therapeutic efficacy of anti-inflammatory drugs 39
1.8 CELL DEATH MECHANISMS 41
1.8.1 Roles of astrocytes in neuroprotection and defense 41
1.8.2 Apoptosis in glial cells 42
1.8.3 The relationship between autophagy and Aβ 47

CHAPTER 2: AIMS AND OBJECTIVES 52

CHAPTER 3: MATERIALS AND METHODS 54
3.1 EXPERIMENTAL PROTOCOLS 54
3.1.1 In vitro study 54
3.1.2 Part II: Oligomeric Aβ
25-35
55
3.1.3 Part III: Fibrillar Aβ
25-35
55
3.2 IN VITRO STUDY 56
3.2.1 Thioflavin T fluorescence 56
3.2.2 Coomassie Blue Staining 57
3.2.3 Free radical scavenging 57
3.24 Statistical analysis 58
3.3 CELL CULTURE STUDIES 59
3.3.1 Chemicals 59
3.3.2 Cell culture 59
3.3.3 MTT assay 60
3.3.4 H
2
S pathway 60
3.3.5 Western blot 61
3.3.6 Reactive oxygen species (ROS) generation 62
3.3.7 Inflammation 64
3.3.8 Cell Death Mechanisms 65
3.3.9 Transmission electron microscopy 67
3.3.10 Statistical analysis 67

vi

CHAPTER 4: PART I - IN VITRO STUDY 68
4.1 RESULTS 68
4.1.1 Aβ
25-35
aggregation with time 68
4.1.2 Aβ
25-35
aggregation with temperature 69
4.1.3 Effects of SPRC on Aβ
25-35
aggregation 70
4.1.4 Effects of SAC on Aβ
25-35
aggregation 72
4.1.5 Effects of sodium hydrosulfide (NaHS) on Aβ
25-35
aggregation 74
4.1.6 Comparison of equimolar concentrations of drugs 76
4.1.7 Effects on radical scavenging 78
4.2 DISCUSSION 80
4.2.1 Aβ
25-35
aggregates with increasing time and temperature 80
4.2.2 Drug treatments disrupt the formation of Aβ
25-35
fibrils 81
4.2.3 SPRC reduces Aβ
25-35
aggregation more effectively than SAC 84
4.2.4 NaHS confer a Type II protection against aggregation 85

4.2.5 Free radicals were scavenged by drugs in solution 87
4.3 SIGNIFICANCE OF PART I 80

CHAPTER 5: PART II - OLIGOMERIC Aβ 91
5.1 RESULTS 91
5.1.1 Effects of aggregated Aβ
25-35
on cell viability 91
5.1.2 Effects of drug pre-treatment on Aβ-induced cytotoxicity 92
5.1.3 Effects on H
2
S pathway 95
5.1.4 Effects on oxidative stress 98
5.1.5 Effects on inflammation 104
5.1.6 Effects on cell death mechanisms 108
5.1.7 Effects on cell morphology 115
5.2 DISCUSSION 122
5.2.1 Oligomeric Aβ
25-35
is toxic to glial cells 122
5.2.2 Pre-treatment of drugs alleviates Aβ-induced injury mediated by the H2S pathway
122
5.2.3 SPRC and SAC can relieve intracellular ROS 125
5.2.4 SPRC restores SOD levels and activities 127
5.2.5 SPRC possesses anti-inflammatory properties different from SAC 129
5.2.6 G1 progression encouraged by SPRC and SAC 133
vii

5.2.7 SPRC reduces DNA damage 135
5.2.8 The autophagic pathway is halted by SPRC and SAC 136

5.2.9 Pre-treated cells had improved cell ultrastructures 137

CHAPTER 6: PART III - FIBRILLAR Aβ 140
6.1 RESULTS 140
6.1.1 Time dependence of aggregated Aβ
25-35
on cell viability 140
6.1.2 Effects of drugs on Aβ-induced cytotoxicity 141
6.1.3 Effects on H
2
S pathway 144
6.1.4 Effects on ROS generation 147
6.1.5 Effects on inflammation 155
6.1.6 Effects on cell death mechanisms 159
6.1.7 Effects on cell morphology 168
6.2 DISCUSSION 176
6.2.1 Higher incubation temperature encourages the formation of Aβ
25-35
fibrils 176
6.2.2 SPRC confers protection after longer pre-treatment at higher dose 178
6.2.3 The H
2
S pathway mediated the cytoprotection by SPRC 180
6.2.4 More oxidative stress was induced by fibrillar Aβ 183
6.2.5 Drug treatments regulated antioxidant enzyme expressions and activities differently
185
6.2.6 SPRC does not involve the H
2
S pathway its anti-inflammatory effects 190
6.2.7 Fibrillar Aβ

25-35
affects autophagy differently from oligomeric Aβ
25-35
194
6.2.8 SPRC modifies the autophagic pathway 195
6.2.9 The ultrastructural changes were restored after drug treatments 197

CHAPTER 7: CONCLUSIONS 199
7.1 SIGNIFICANT CONTRIBUTIONS 199
7.2 FUTURE WORK 203

References 205


viii

ABSTRACT
Alzheimer’s disease (AD) is a neurodegenerative disease characterized by widespread
extracellular deposits of amyloid-beta (Aβ) protein in the brain. Amongst which, the Aβ
25-35

peptide is the shortest fragment which retains the toxicity of the full-length protein. In this study,
this peptide was found to aggregate in a time- and temperature-dependent manner. This
aggregation can be slowed with the co-incubation of S-propargyl-cysteine (SPRC), S-allyl-
cysteine (SAC) or sodium hydrosulfide (NaHS) in the solution, accompanied with decreased
sizes of the Aβ aggregates. Aβ radicalizes in solution to cause detrimental damage even outside
cells. SPRC can scavenge free radicals better than SAC, but less competent than NaHS. The
triple bond in SPRC is more nucleophilic than SAC that can react with the lone pair of electrons
in free radicals. Oligomeric or fibrillar Aβ
25-35

were added to the C6 glioma cell line and the cell
viabilities were compromised. The decline in cell viability was more obvious when treated with
fibrillar Aβ
25-35
, which acted on exacerbating oxidative stress by increasing H
2
O
2
levels,
inflammation and disruption of the autophagic activation. Moreover, the aggregated Aβ
decreased the H
2
S levels produced and the expression of cystathione-β synthase (CBS) in the
cells. Pre-treatments of SPRC and SAC both restored the Aβ-induced reductions in cell
viabilities, but the doses required to restore the cell viabilities were higher in damage by Aβ
fibrils. SPRC mimicked the protection by SAC on glioma cells through its antioxidant nature,
particularly targeting superoxide dismutase (SOD) and glutathione peroxidase (GPx) in both
forms of Aβ injuries. SPRC also decreases pro-inflammatory IL-1β and increases anti-
inflammatory IL-10 to reduce the inflammatory responses evoked by Aβ
25-35
though differently
from SAC. SPRC reduces DNA fragmentation and reverses autophagic activation that preserves
cellular integrity in a similar manner to SAC. These protective effects of SPRC can be due its
cysteine backbone similar to SAC and its endogenous H
2
S-producing nature. When compared to
ix

NaHS that produces H
2

S exogenously, SPRC sustains the protection to the cellular status despite
the form of Aβ. These promising results can suggest SPRC as a novel therapeutic agent for
neurodegenerative diseases.
x

LIST OF TABLES

Table 1: Summary of commonly used therapies demonstrating strong antioxidant properties.
Table 2: Primer sequences for various inflammatory genes used in RT-PCR.
Table 3: Summary of effects of SPRC on various aspects affecting cell survival.
Table 4: Summary of effects of SAC on various cellular aspects.

xi

LIST OF FIGURES
Figure 1: Aggregation properties of peptides and their mechanisms.
Figure 2: Amyloid precursor protein (APP) proteolysis through the non-amyloidogenic or
amyloidogenic pathway.
Figure 3: Chemical structure of S-allyl-cysteine (SAC).
Figure 4: Chemical structure of S-propargyl-cysteine (SPRC).
Figure 5: Comparison of metabolic pathways of cysteine and SPRC.
Figure 6: The production of hydrogen sulfide (H
2
S) in mammalian tissues.
Figure 7: The reactions of detoxifying oxygen radicals in the body.
Figure 8: Release of pro-inflammatory cytokines by the overproduction of Aβ.
Figure 9: Experimental protocol for Part II: Oligomeric Aβ
25-35
.
Figure 10: Experimental protocol for Part III: Fibrillar Aβ

25-35
.

Part I: In Vitro Study
Figure 11: Aggregation of Aβ
25-35
at 37°C over time.
Figure 12: Aggregation of Aβ
25-35
with increasing temperature for 96 hours.
Figure 13: Effects of different concentrations of SPRC on Aβ
25-35
aggregation.
Figure 14: Effects of different concentrations of SAC on Aβ
25-35
aggregation.
Figure 15: Effects of different concentrations of sodium hydrosulfide (NaHS) on Aβ
25-35

aggregation.
Figure 16: Comparison between treatments with different drugs.
Figure 17: Effects of radical scavenging abilities of various drugs.
Figure 18: Size-exclusion chromatography carried out by HPLC to separate samples according to
molecular size.

Part II: Oligomeric Aβ
Figure 19: Effects of aggregated Aβ
25-35
on C6 glioma cell viability.
Figure 20: Effects of SPRC pre-treatment on Aβ-induced cytotoxicity.

Figure 21: Effects of SAC or NaHS pre-treatments on Aβ-induced cytotoxicity.
Figure 22: Comparison between equimolar concentrations of drugs on Aβ-induced cytotoxicity.
Figure 23: Effects of pre-treatment of drugs on H
2
S concentrations in cell medium.
Figure 24: Effects of pre-treatment of drugs on cystathionine-β synthase (CBS) expression in cell
lysates.
Figure 25: Effects of the CBS inhibitor aminooxyacetate (AOAA) on cell viability.
Figure 26: Effects of pre-treatment of drugs on DCF production.
Figure 27: Effects of pre-treatment of drugs on DHE fluorescence.
Figure 28: Effects of pre-treatment of drugs on total superoxide dismutase (SOD) expression and
activities in cell lysates.
Figure 29: Effects of pre-treatment of drugs on catalase expression and activities in cell lysates.
Figure 30: Effects of pre-treatment of drugs on glutathione peroxidase (GPx) expression and
activities in cell lysates.
Figure 31: Effects of pre-treatment of drugs on IL-1β expression.
Figure 32: Effects of pre-treatment of drugs on IL-6 expression.
Figure 33: Effects of pre-treatment of drugs on TNF-α expression.
Figure 34: Effects of pre-treatment of drugs on IL-10 expression of inflammatory factors.
Figure 35: Effects of pre-treatment of drugs on cell cycle analysis.
xii

Figure 36: Effects of pre-treatment of drugs on DNA fragmentation using TUNEL staining.
Figure 37: Effects of pre-treatment of drugs on PARP and pro-caspase 3 expressions in cell
lysates.
Figure 38: Effects of pre-treatment of drugs on autophagy using acridine orange staining.
Figure 39: Effects of pre-treatment of drugs on LC3 expression in cell lysates.
Figure 40: Effects of pre-treatment of drugs on the ultrastructure of the cells using transmission
electron microscopy at lower magnification.
Figure 41: Effects of pre-treatment of drugs on the ultrastructure of the cells using transmission

electron microscopy at higher magnification.

Part III: Fibrillar Aβ
Figure 42: Effects of different times of incubation with aggregated Aβ
25-35
on cell viability.
Figure 43: Effects of SPRC pre-treatment on Aβ-induced cytotoxicity.
Figure 44: Effects of SAC and NaHS pre-treatment on Aβ-induced cytotoxicity.
Figure 45: Comparison between equimolar concentrations of drugs on Aβ-induced cytotoxicity.
Figure 46: Effects of pre-treatment of drugs on H
2
S concentrations in cell medium.
Figure 47: Effects of pre-treatment of drugs on CBS expression in cell lysates.
Figure 48: Effects of the CBS inhibitor AOAA on cell viability.
Figure 49: Effects of pre-treatment of drugs on DCF production.
Figure 50: Effects of pre-treatment of drugs on DHE fluorescence.
Figure 51: Effects of pre-treatment of drugs on SOD expression and activities in cell lysates.
Figure 52: Effects of pre-treatment of drugs on catalase expression and activities in cell lysates.
Figure 53: Effects of pre-treatment of drugs on glutathione peroxidase (GPx) expression and
activities in cell lysates.
Figure 54: Effects of pre-treatment of drugs on IL-1β expression.
Figure 55: Effects of pre-treatment of drugs on IL-6 expression.
Figure 56: Effects of pre-treatment of drugs on TNF-α expression.
Figure 57: Effects of pre-treatment of drugs on IL-10 expression.
Figure 58: Effects of pre-treatment of drugs on cell cycle status.
Figure 59: Effects of pre-treatment of drugs on DNA fragmentation using TUNEL staining.
Figure 60: Effects of pre-treatment of drugs on PARP and pro-caspase 3 expressions in cell
lysates.
Figure 61: Effects of pre-treatment of drugs on autophagy using acridine orange staining.
Figure 62: Effects of pre-treatment of drugs on LC3 expression in cell lysates.

Figure 63: Effects of pre-treatment of drugs on the ultrastructure of the cells using transmission
electron microscopy at lower magnification.
Figure 64: Effects of pre-treatment of drugs on the ultrastructure of the cells using transmission
electron microscopy at higher magnification.


INTRODUCTION
1

CHAPTER 1: INTRODUCTION
1.1 General Introduction
1.1.1 Alzheimer’s disease
Alzheimer’s disease (AD) was first described by Alzheimer about a century ago and is
the most common form of dementia today that accounts for 50-70 percent of all dementia cases.
According to the Alzheimer’s Disease Association (1), AD is the sixth leading cause of death in
the United States. In Singapore, the aging population has brought about an increased prevalence
in the occurrence of AD, where there were about 30,000 affected patients as of 2010 (1). The
worrying trend of an aging population has been highlighted extensively by the Singapore
government and has been directing future policy-making. Amongst which, the dependence of
AD patients on caregivers and long-term care systems is of concern to the healthcare
infrastructure.
AD is a progressive disease clinically characterized by episodic memory problems to a
slow global decline of cognitive functions. End-stage AD patients are usually bedridden and
dependent on custodial care until death occurs on an average of nine years after diagnosis (2).
There is currently no known cure for AD, although several types of treatments targeting different
pathways had been shown to alleviate symptoms.
AD patients typically exhibit cerebral atrophy where neuronal death is the main cause of
the behavioural deficits. Post-mortem analyses of human AD brains reveal senile plaques
composed of extracellular deposits of amyloid-β (Aβ) and neurofibrillary tangles formed by
accumulation of abnormal hyperphosphorylated tau in brain regions responsible for memory and

INTRODUCTION
2

cognition, namely the hippocampus and frontotemporal cortex. The severe neuronal cell death in
such areas significantly correlated to the distribution of Aβ plagues found. However, the extent
of Aβ toxicity does not parallel the clinical severity and thus increases the challenges in drug
discovery for AD.
1.1.2 Aggregated proteins and diseases
Aggregated proteins have been implicated in several neurodegenerative diseases such as
Alzheimer’s, Parkinson’s, prion and motor neuron diseases (3). Protein deposits observed in the
brain were first characterized in the 19
th
century and could confer vastly different symptoms to
affected individuals. Such deposits may differ largely in their amino acid sequences, but form
morphologically similar aggregates termed amyloid in tissues most affected by
neurodegeneration, suggesting that some general properties of affected polypeptides steer them
towards enhanced self-assembly and hence, toxicity.
1.1.3 Self-aggregation properties of amyloid peptides
Amyloid fibrils are rich in β-sheets with a highly ordered cross-β core structure to give
strength and stability. These fibrils exhibit the interesting phenomenon of self-assembly, or “the
spontaneous organization of molecules under thermodynamic equilibrium, towards a state of free
energy via non-covalent interactions such as hydrogen bonding, electrostatic, hydrophobic and
van der Waals” (4). Amyloid peptides are usually soluble; yet under certain conditions these
peptides undergo conformational changes to self-assemble into fibrils. All amyloid fibrils are
defined by three experimental criteria: i) green birefringence upon staining with Congo Red; ii)
fibrillar morphology; and iii) β-sheet secondary structure (4).
INTRODUCTION
3

Figure 1: Aggregation properties of peptides and their mechanisms. (a) Self-aggregation
properties of peptides that undergo a conformational change from the protofibril state to
filaments and fibrils. Adapted from Nilsson, Methods 34 (2004). (b) Amyloid-beta peptides can
display aggregation by folding into aggregates through different proposed mechanisms. Adapted
from Hughes and Dunstan, Modern Biopolymer Science (2009).
The formation of fibrils occurs when conformational changes expose hydrophobic groups
usually hidden, driving the non-specific aggregation of such peptides. The typical self-
aggregation of the amyloid peptide (Figure 1a) starts from the formation of protofibrils (or
oligomers), that extents in length to become filaments. The filamentous materials aggregate and
grow in size to become fibrils. Typically, fibrils are linear, unbranched and of variable length
dependent on the number of protofilaments that make up the fibril. Certain fibrils exhibit helical
shapes where protofilaments twist around each other through both electrostatic and hydrophobic
interactions (5). Such aggregation that occurs from protein misfolding is proposed to be a generic
property of all proteins and peptides (6). The mechanisms of amyloid fibril assembly have been
summarized by Hughes and Dunstan (2009) into three pathways:

(b)
(a)
INTRODUCTION
4

1. Nucleation-dependent polymerization
This theory suggests that the formation of an energetically unfavourable nucleus precedes
fibril formation and elongation. The rate of formation is determined by any factor
influencing the concentration or conformation of the proteins. Once the nucleus is formed,
the addition of monomers becomes thermodynamically favourable and occurs rapidly.
2. Nucleation conformational model
This model proposes that the accumulation of disordered polypeptides with a nucleus
transits these proteins into stable conformations. This conformational conversion to

stability is a rate-limiting step after which monomers could build on quickly.
3. Off pathway model
The nucleus is formed in a parallel, irreversible step after the initial refolding phase
where an amyloidogenic intermediate emerges. The filaments are elongated by addition
of such intermediates at each end, and further associate to form fibrils.
1.1.4 Factors affecting fibril formation
The ideal thermodynamics promoting fibril formation should allow for a small energy
barrier between the folded and unfolded states, as well as an optimal distance for efficient
intermolecular interactions between molecules. Hence, factors that can achieve such favourable
energetics will promote fibril formation (Figure 1b).
The rate of amyloid peptide aggregation is highly dependent on solvent conditions (7).
Firstly, conditions that decrease solubility and/or increase the ease of acquiring the required
secondary structure for polymerization accelerate the aggregation kinetics. Secondly,
aggregation rates are affected by sites available for elongation. Initial lag times and threshold
INTRODUCTION
5

behavior observed in typical aggregation studies suggest the need for available sites, and once
present the aggregation proceeds rapidly. Factors that affect aggregation processes include
temperature, pH, addition of salts, mutations resulting in changes to the amino acid side chains,
shear forces, available interfaces for formation etc. As such, these suggest the delicate balances
necessary to slow the production of amyloid aggregates.

INTRODUCTION
6

1.2 Amyloid-beta peptides are central to Alzheimer’s disease
The amyloid-beta (Aβ) peptide forms aggregates that are a hallmark of Alzheimer’s
disease. In the brain, both soluble and insoluble Aβ exist at differing concentrations. Soluble Aβ
(oligomers) is decreased in the cerebrospinal fluid and plasma of AD patients (8) and insoluble

Aβ (fibrils) exists mainly as pathological plagues in postmortem brain sections (9).
1.2.1 Products of sequential cleavage
The amyloid fibrils found in plaques are made of Aβ peptides containing 39-42 amino
acid residues. The most common Aβ species found in patients are the 40-mer and 42-mer
peptides. The sequential cleavage of the amyloid precursor protein (APP) by three enzymes, α-,
β- and γ-secretases, produces these peptides of different lengths. Of which, the γ-secretase
enzyme has been identified as a complex of enzymes composed of presenilin 1 or 2 (PS1 and
PS2), nicastrin, anterior pharynx defective and presenilin enhancer 2. Familial AD is an
autosomal dominant disease with early onset pathogenesis where patients had been identified
with mutations of three genes APP, PS1 and PS2. These mutations resulted in the early-onset
of AD and affected the stability and metabolism of Aβ peptides, directly linking these gene
products to the disease.
The proteolytic cleavage of APP can be divided into the non-amyloidogenic and
amyloidogenic pathways (Figure 2) (10).
The non-amyloidogenic pathway is the prevalent pathway in cells where APP is cleaved
at a position 83 amino acids from the C terminus by the α-secretase. Three putative candidates of
the enzyme belongs to the family of a disintegrin and metalloprotease (ADAM) have been
identified: ADAM9, ADAM10 and ADAM17. The resultant products are a large N-terminus
INTRODUCTION
7

fragment (sAPPα) that is secreted into the extracellular medium and a 83-amino-acid C-terminus
(C83) that is retained in the membrane. The fragment retained in the membrane is subsequently
cleaved by the γ-secretase, producing a short fragment termed p3. Since these cleavages occur
out of the range of the Aβ peptides, this pathway does not contribute to the production of the
pathological Aβ.

Figure 2: APP proteolysis (adapted from LaFerla et al, Nature Rev Neurosci 8, 2007). APP can
undergo proteolytic processing either the non-amyloidogenic or amyloidogenic pathway. The
amyloidogenic pathway results in the production of pathological Aβ peptides.

The amyloidogenic pathway occurs when the initial proteolysis was carried out by β-
secretase (also known as -site APPcleaving enzyme 1; BACE1) at a position 99 amino acids
from the C-terminus. This 99-mer fragment (C99) is retained in the membrane while the N-
terminus (sAPPβ) is released into the extracellular medium. Subsequent cleavage of the free
fragment by the γ-secretase complex comprising of presenilin 1 or 2, nicastrin, anterior
pharynx defective (APH-1) and presenilin enhancer 2 (PEN2) liberates the intact Aβ peptide.
The major species produced is 40 residues in length (Aβ
1-40
), while a small proportion
(approximately 10%) is the 42 residue variant (Aβ
1-42
). Aβ
1-42
is more hydrophobic and more
prone to fibril formation than the shorter variant.
INTRODUCTION
8

1.2.2 Neurotoxicity of Aβ
25-35

Interestingly, different lengths of Aβ peptides display a range of neurotoxicity in vivo
(11), and the ability of the full-length peptide to affect cognitive processes is best displayed by
the undecapeptide Aβ
25-35
(12). This fragment of sequence GSNKGAIIGLM is the shortest
segment responsible for the observed neurotoxicity of Aβ and is widely accepted as the
functional domain of the full-length Aβ. Administering Aβ
25-35
to rats led to impaired spatial

working memory and the acquisition of passive avoidance reactions (13) that correlated well
with a deficiency in the amount and activity of choline acetyltransferase. Aβ
25-35
is also observed
in neurons of subiculum and entorhinal cortex of AD brains (14).
The ten amino-acid fragment is also the shortest fragment that can exhibit large β-sheet
fibrils in solution. Moreover, some in vitro studies have found that the fragment is not only toxic
upon aging, it also demonstrates strong toxicity when freshly prepared (15), a property dissimilar
to the full-length peptide.
1.2.3 Structure of the Aβ
25-35
peptide
The rapid dissolution of Aβ
25-35
in solution has presented difficulties in unraveling its
three-dimensional monomeric structure by NMR either in water or organic solvents. The actual
monomeric structure is pivotal in the resultant morphology because this affects the conformation
of the intermediate species and subsequent fibril shapes. The peptide can adopt different
monomeric conformations depending on the media. Generally, it was concluded that the peptide
adopts a helical structure in apolar organic solvents and a β-structure (β-turn or β-sheet) in
aqueous buffers. But these trends may be affected by pH, concentration, incubation times,
preparation and purification processes (5). Noted by Wei and Shea (16), the hydrogen-bond-
INTRODUCTION
9

forming groups of the β-turn in Aβ
25-35
are more exposed than the other peptide residues and are
possibly the initial points for rapid aggregation. In particular, the presence of glycine residues at
positions 25, 29 and 33 are proposed to play an important role in formation of β-sheet structures

in the aqueous solution (17). Although the C-terminal hydrophobic portion of the fragment is
crucial for aggregation, substitution of methionine35 with alanine (18) disrupted the required β-
sheet conformation but surprisingly enhanced toxicity. The β-branched residues of isoleucine31
and isoleucine32 have also been implicated in the folding and/or stabilization of the assembled
aggregate structure.
Much aggregation kinetics had been done on Aβ
1-42
and Aβ
1-40
, but little is known about
Aβ
25-35
.

INTRODUCTION
10

1.3 Garlic and S-allyl-cysteine
1.3.1 Aged garlic extract
Garlic (Allium sativum) is a condiment used by many peoples in their daily lives and has
been reputed to have strong medicinal uses. Amongst these, garlic had shown to prevent
atherosclerosis (19), cancer (20) and ischemia-reperfusion injury (21) by its antioxidant and free-
radical scavenging properties. Aged garlic extract is one of the many nutraceuticals that have
been in the market as a complementary supplement to our diets (22, 23). The aged garlic extract
can most prominently protect the frontal brain morphology in the senescence- accelerated mouse
model, most likely through influencing the cognitive and behavioural functions to alleviate
neurodegenerative conditions (24).
Both aged garlic extract and its main active compound, S-allyl-cysteine (SAC) showed
neuroprotective effects in the mouse transgenic model of Alzheimer’s disease (25). Aged garlic
extract and SAC effectively reduced the Alzheimer’s cerebral plaques with a concomitant

increase in α-cleaved sAPPα. This protection reduced the amyloidogenic pathway while steering
towards the secretory pathway, preventing the accumulation of the pathological Aβ. In the same
study, the authors also found that while aged garlic extract and SAC could reduce inflammation
and tau phosphorylation, the efficacy was greater using aged garlic extract.
1.3.2 S-allyl-cysteine
S-allyl-cysteine (SAC) is a major organosulfur component of aged garlic extract. SAC is
a derivative of the amino acid cysteine, where an allyl group is added to the sulfur atom (Figure
3) (26). As an active compound of garlic extract, SAC exhibits beneficial effects similar to the
garlic extract. The antiproliferative effects of SAC on human neuroblastoma cells prompted
INTRODUCTION
11

cancer research using natural active compounds (27). SAC also demonstrates strong antioxidant
properties useful in diabetes (28) and ethanol-induced acute liver injury (29). Moreover, SAC
alleviated symptoms in a mouse model of acetaminophen-induced hepatotoxicity (30) through
anti-inflammatory and anti-oxidative effects.

Figure 3: Chemical structure of SAC.
Interestingly, SAC has anti-amyloidogenic effects and destabilizes the Aβ fibrils in vitro
(31). Rao et al suggested that SAC inhibited Aβ aggregate formation by halting further
oligomerization and breaking preformed Aβ fibrils. The -OH group of the carboxylic group in
SAC was also postulated to enhance hydrophobic interactions by hydrogen bonding to the donor
groups of Aβ, although the actual mechanisms are not known.
Furthermore, SAC ameliorated Aβ-induced cytotoxicity in nerve-growth factor-
differentiated PC12 cells (32). Aβ-induced cytotoxicity was not attenuated by a caspase-3
inhibitor, nor did 4-hydroxynonenal (HNE) play a role in the observed cell death, suggesting that
apoptosis is not pivotal in Aβ-induced cytotoxicity, and the SAC neuroprotection is not via the
anti-apoptotic pathway.
As such, the antioxidant potential of SAC was investigated in other models of
Alzheimer’s disease. SAC improved learning abilities using the radial maze following

intrahippocampal injection of Aβ
25-35
related to its capability to reduce oxidative stress (33).
Pretreatment of SAC just thirty minutes before the peptide injection significantly reduced