drosophila melanogaster as a model organism to study human ... · disease-causing genes. using the...
TRANSCRIPT
Drosophila melanogaster as a model organism to study human
neurodegenerative diseases
by
Kinga Maria Michno
A thesis submitted in conformity with the requirements
for the degree of Doctor of Philosophy
Department of Molecular Genetics
University of Toronto
© Copyright by Kinga Maria Michno (2009)
ii
Drosophila melanogaster as a model organism to study human
neurodegenerative diseases
Kinga Maria Michno
Doctor of Philosophy
Department of Molecular Genetics
University of Toronto
2009
Abstract
A great deal of our current understanding about the biology of neurodegenerative diseases has
come from studying the function of genes linked to inherited forms of these disorders. Work
performed in animal models, vertebrates as well as invertebrates, has been instrumental in
deciphering the cellular, physiological and behavioural deficits arising from the expression of
disease-causing genes. Using the fruit fly, Drosophila melanogaster, as a model we examined
the normal and aberrant function of two genes linked to the onset of neurodegeneration in
humans, presenilin and superoxide dismutase-1. Drosophila is an extremely versatile model and
in many ways is ideal for studying the genetic basis of human disease. The high degree of
genetic conservation coupled with low genetic redundancy make this model particularly well
suited for studying the function of disease causing genes. We demonstrate a novel genetic,
physical and physiological interaction between presenilin and calmodulin and describe how this
interaction impacts a very early cellular defect associated with Alzheimer’s Disease, intracellular
iii
calcium dyshomeostasis. We also describe progressive locomotory deficits in flies expressing
mutant alleles of the superoxide dismutase gene, which have been linked to the onset of familial
amyotrophic lateral sclerosis. Collectively, our work demonstrates that Drosophila can be used
to study the cellular, physiological and behavioural basis of human neurodegenerative diseases
and may provide a model to identify novel therapeutic avenues for neurodegenerative diseases.
iv
Acknowledgments
This thesis is a small part of a much bigger story. The story began many years ago when a young
couple decided to escape the constraints of communism to ensure a successful future for their
daughter—me. Having a graduate degree in Slavic Literature my Mother sacrificed her own
future so that I could pursue mine. For this reason I'd like to begin by thanking my mother for
her steadfast support, commitment and engagement in my education. I am thankful to both my
parents for insisting that I pursue my education and for supporting me both emotionally and
financially along the way. I would also like to thank my Father for being a living example of
how to think outside of the box and for teaching me to always look for creative solutions to
problems.
No matter how bad my day was at the lab, a single thought could make all the built up anxiety
and stress disappear. I would think of my husband Artur, and reflect on how lucky I am to be
able to share my life with such a fantastic individual and loving partner. Artur, your are my
foundation, my rock, my reality check and I couldn't have arrived at this point without your
support and patience—thank you.
Graduate research is much like passing a driving exam on an insane obstacle course. Each of us
gets behind the wheel and eagerly pushes the pedal to the metal; determined to be the first to
cross the finish line without any scratches or bruises. But at some point we all crash, and when
we do we do it those who drive with us who pick up the pieces and encourage us to get back
behind the wheel. It has been a true privilege to work along side my fellow Boulianners; to share
all my good and bad science days. I would also like to thank Tania Alexson who has been my
long-time friend, research colleague and confidant. Tania has helped me to re-analyze my data
from so many different angles my head is still spinning. I would also like to thank Shirley Liu
for being generous with her time in helping me ensure the timely completion of the final
experiment necessary to complete the final data chapter of this thesis.
v
A decisive milestone of this dissertation was a very productive collaboration with Diane
O'Dowd’s laboratory at the University of California. I have profound gratitude to Diane for her
guidance and for welcoming me into her lab. Diane provided me unrestricted access to her
equipment without which a considerable portion of the first data chapter would have been very
difficult to complete. By the same token I would like to thank Jorge Campusano and Betty
Siscero for taking the time to train me and for offering much appreciated technical assistance. I
would also like to acknowledge and thank Joel Levine for helping me design and execute the
locomotory behavioural analysis described in the second chapter.
I have been privileged in having excellent mentorship during my training and in chronological
order I would like to thank my mentors. First, thank you to Paul Hamel who took a chance on an
undergraduate student, completely wet behind the ears. As a master's supervisor Paul was
demanding and relentlessly pushed me until I developed a strong work ethic. I’d like to sincerely
thank my PhD supervisor; Gabrielle Boulianne. Gabrielle welcomed me into her lab even though
I had no experience working with Drosophila. Over the years Gabrielle has made a significant
impact on my scientific development. Gabrielle’s mentorship has been instrumental in helping
me to develop my critical thinking and communication skills. I am also very grateful to
Gabrielle for allowing me a great deal of ownership over my research and for being supportive
during my various collaborations. I would also like to thank my committee members Freda
Miller and Marla Sokolowski. Thanks to Gabrielle, Marla and Freda I always left my
committee meetings motivated and determined to reach my research goals. Thank you for your
insightful suggestions and encouragemen; it has been a true privilege to learn from each of you.
Finally, I’d like to thank the Creator for providing me with sufficient aptitude and resolve to
preservere. Believe; achieve.
vi
Table of Contents
Acknowledgments .......................................................................................................................... iv
Table of Contents ........................................................................................................................... vi
List of Tables ................................................................................................................................. ix
List of Figures ................................................................................................................................. x
List of Abbreviations .................................................................................................................... xii
CHAPTER 1 ................................................................................................................................... 1
1.1 Introduction to human neurodegenerative diseases ............................................................ 1
1.2 Alzheimer’s Disease ........................................................................................................... 2
1.2.1 Genetics of Alzheimer’s Disease ............................................................................ 3
1.2.2 Alzheimer’s Disease aetiology ............................................................................... 8
1.2.3 Therapeutic treatments .......................................................................................... 13
1.3 Amyotrophic Lateral Sclerosis ......................................................................................... 14
1.3.1 Genetics of Amyotrophic Lateral Sclerosis .......................................................... 15
1.3.2 Amyotrophic lateral sclerosis aetiology ................................................................ 17
1.3.3 Therapeutic treatments .......................................................................................... 19
1.4 Drosophila as a model of human neurodegenerative diseases ......................................... 20
1.4.1 Drosophila models of Alzheimer’s Disease ......................................................... 23
1.4.2 Drosophila models of Amyotrophic Lateral Sclerosis ......................................... 25
1.5 Purpose of our studies ....................................................................................................... 27
CHAPTER 2 ................................................................................................................................. 28
2 Analysis of the interaction of presenilin with the intracellular calcium stores machinery ...... 28
2.1 Abstract ............................................................................................................................. 29
2.2 Introduction ....................................................................................................................... 30
vii
2.3 Materials and Methods ...................................................................................................... 32
2.4 Results ............................................................................................................................... 36
2.4.1 Intracellular calcium dynamics in primary cholinergic Drosophila neurons ....... 36
2.4.2 Psn-induces deficits in intracellular calcium stores content ................................. 38
2.4.3 Loss-of-function mutations in Cam suppress Psn-induced wing scalloping ........ 41
2.4.4 Cam suppresses Psn-induced deficits in intracellular calcium stores content ...... 43
2.4.5 Psn and Cam physically interact ........................................................................... 45
2.4.6 Primary neurons expressing wild type Psn have lower incidence of apoptosis ... 47
2.4.7 Cholinergic expression of fAD-mutant Psn results in shortened lifespan. ........... 50
2.5 Discusion ........................................................................................................................... 55
CHAPTER 3 ................................................................................................................................. 60
3.1 Abstract ............................................................................................................................. 61
3.2 Introduction ....................................................................................................................... 62
3.3 Methods and materials ...................................................................................................... 64
3.4 Results ............................................................................................................................... 67
3.4.1 Analysis of transgenic flies expressing human wild type and fALS-SOD1 ......... 67
3.4.2 Ubiquitous expression of fALS-SOD1 gives rise to progressive deficits in
adult fly locomotory activity ................................................................................. 69
3.4.3 Survival analysis of flies expressing wild type or fALS-SOD1 ........................... 77
3.4.4 fALS-SOD1 does not appear to form aggregates in adult flies ............................ 81
3.5 Discussion ......................................................................................................................... 84
CHAPTER 4 ................................................................................................................................. 89
4.1 Discussion ......................................................................................................................... 89
4.1.1 Are aggregates toxic? ............................................................................................ 89
4.1.2 What role does stress play? ................................................................................... 90
4.1.3 Behavioural Genetics in Drosophila ..................................................................... 93
viii
4.1.4 Concluding thoughts ............................................................................................. 94
References ..................................................................................................................................... 95
Appendices .................................................................................................................................. 113
ix
List of Tables
Table 1. Survival Analysis ........................................................................................................... 51
Table 2. Average day time activity of 42-45 day old flies ubiquitously expressing wild type or
fALS-mutant human SOD1. ......................................................................................................... 71
Table 3. Average day time activity of 52-55 day old flies ubiquitously expressing wild type or
fALS-mutant human SOD1. ......................................................................................................... 73
Table 4. Average day time activity of 56-59 day old flies ubiquitously expressing wild type or
fALS-mutant human SOD1. ......................................................................................................... 76
Table 5. Survival Analysis ........................................................................................................... 79
x
List of Figures
Figure 1. Schematic of intramembraneous APP proteolysis. ........................................................ 4
Figure 2. The γ-secretase complex ................................................................................................. 6
Figure 3. Intracellular calcium signalling .................................................................................... 11
Figure 4. The GAL4/UAS system ............................................................................................... 22
Figure 5. Evaluation of calcium content in internal stores in Drosophila primary neuronal
cultures .......................................................................................................................................... 37
Figure 6. Expression of wild type and FAD-mutant Psn. ............................................................ 39
Figure 7. Calcium content in internal calcium stores is affected in cholinergic neurons
expressing Psn. ............................................................................................................................. 40
Figure 8. Psn-induced wing notching is suppressed by loss-of-function mutations in Cam. ...... 42
Figure 9. Psn-induced effects on intracellular calcium stores is suppressed by a loss-of-function
mutation in Cam mutations. .......................................................................................................... 44
Figure 10. Cam binds to full-length as well as the N-terminal fragment of Psn. ........................ 46
Figure 11. Total incidence of apoptosis in cultures expressing wild type or FAD-mutant Psn. . 48
Figure 12. Expression of wild type Psn in cholinergic neurons facilitates cell-autonomous cell
survival. ......................................................................................................................................... 49
Figure 13. Survival analysis of flies expressing wild type Psn in cholinergic neurons. .............. 52
Figure 14. Survival analysis of flies expressing FAD-Psn in cholinergic neurons. .................... 53
Figure 15. Survival analysis for flies expressing FAD-Psn in cholinergic neurons with or
without the loss of a single Cam allele. ........................................................................................ 54
xi
Figure 16. Expression of wild type and fALS-mutant human SOD1. ......................................... 68
Figure 17. Day activity levels of 42-45 day old flies expressing wild type or fALS-SOD1. ...... 70
Figure 18. Day activity levels of 52-55 day old flies expressing wild type or fALS-SOD1. ...... 72
Figure 19. Day activity levels of 56-59 day old flies expressing wild type or fALS-SOD1. ...... 75
Figure 20. Survival analysis of flies ubiquitously expressing fALS-mutant SOD1. ................... 78
Figure 21. Survival analysis of flies ubiquitously expressing wild type SOD1. ......................... 80
Figure 22. Misfolded human SOD1 is not detected in lysates generated from 30 day old flies
expressing human fALS-SOD1. ................................................................................................... 82
Figure 23. Misfolded human SOD1 is not detected in lysates generated from 56-59 day old flies
expressing human fALS-SOD1. ................................................................................................... 83
Figure 24. Misfolded human SOD1 is not detected in lysates generated from 56-59 day old flies
expressing two copies of human fALS-SOD1. ............................................................................. 85
xii
List of Abbreviations
A amyloid peptide
AD Alzheimer’s disease
AICD APP intracellular domain
ALS Amyotrophic Lateral Sclerosis
ANOVA Analysis of variance
APP Amyloid precursor protein
Appl Drosophial Amyloid precursor protein-like gene or protein
BACE site APP cleaving enzyme
BAK Bcl-2 homologous antagonist/killer
BAX Bcl-2–associated X protein
BSA Bovine serum albumin
bZIP Basic Leucine Zipper
C/EBP CCAAT-enhancer-binding proteins
CaKII Calmodulin kinase II
CALHM1 calcium homeostasis modulator 1
Cam Calmodulin
CBP Calcium binding protein
CHOP CCAAT/enhancer binding protein
CNS Central nervous system
Creb cAMP response element binding
DNA Deoxyribose nucleic acid
DSM Diagnostic and Statistical Manual of Mental Disorders
DTT Dithiothreitol
xiii
EDTA ethylenediaminetetraacetic acid
EGTA ethylene glycol tetraacetic acid
EMG Electromyography
ER Endoplasmic reticulum
ERAD ER-associated
fAD familial Alzheimer’s Disease
fALS familial Amyotrophic Lateral Sclerosis
FBS Fetal bovine serum
GADD153 Growth Arrest and DNA Damage
GFP Green fluorescent protein
HBSS Hank's buffered salt solution
HEK 293 Human Embryonic Kidney cell line 293
HRP Horse radish peroxidase
HS Human SOD1
HSP Mammalian heat shock protein
Hsp Drosophila heat shock protein
IHC Immunohistochemistry
IP3R nositol triphosphate receptor
LTP Long term potentiation
MND Motor neuron disease
NDS Normal donkey serum
NEP2 Mammalina neprilysin 2 gene or protein
Nep2 Drosophila neprilysin 2 gene or protein
NGS Normal goat serum
PAG proliferation associated gene
PBS Phosphate buffer saline
xiv
PI3K Phosphoinositide 3-kinases
PS Mouse presenilin gene or protein
PSEN Human presenilin gene
Psn Drosophila presenilin gene or protein
PVDF Polyvinylidene Fluoride
RPM Rotations per minute
RyR Ryanodine receptor
sAD sporadic Alzheimer’s Disease
sALS sporadic Amyotrophic Lateral Sclerosis
SDS Sodium Dodecyl Sulfate
SERCA Sarco/Endoplasmic Reticulum Ca2+
-ATPase
Sod Drosophila superoxide dismutase gene or protein
SOD1 Human superoxide dismutase gene or protein
TACE Tumor necrosis factor-α-converting enzyme
TBST Tris-Buffered Saline Tween-20
TUNEL Terminal deoxynucleotidyl transferase dUTP nick end labeling
UAS Upstream activating sequence
UPR Unfolded protein response
WT Wild type
1
CHAPTER 1
1.1 Introduction to human neurodegenerative diseases
Today, the aging Canadian population is faced with the grim reality that there are no effective
treatments let alone a cure for many of the diseases that target the elderly brain. Affected
individuals inevitably lose the ability to care for themselves causing significant emotional stress
to both patients and their family. The high costs of long-term palliative care and therapeutics,
which only partially alleviate some symptoms, poses a considerable economic burden on the
elderly, their caregivers as well as on our social health care system. It is estimated that
Canadians spend 5.5 billion dollars annually treating and caring for persons with dementia alone
(Ostbye and Crosse. 1994). Though staggering, the current expenditures pale in comparison with
what’s projected to come. By 2041 nearly a quarter of the Canadian population will be 65 years
of age or older, relative to the current proportion of 12% (Lindsay. 1999). Given that age is the
single most easily identifiable risk factor for neurodegenerative diseases, as the proportion of
seniors continues to increase, so will the prevalence of neurodegenerative diseases in our society
(Nelson. 1995;Anonymous1994).
Research has turned to animal models to explore the cellular and molecular mechanisms
involved in human disease pathogenesis in hopes of one day finding a cure. Modelling human
neurodegenerative diseases in animal models has been instrumental to our current understanding
about the biology of neurodegenerative disease. In addition, work performed in animal models is
indeed guiding the development of new therapeutics. In the proceeding sections two human
neurodegenerative diseases, Alzheimer’s Disease (AD) and Amyotrophic Lateral Sclerosis
(ALS), will be described and explored based on work performed in both vertebrate as well as
invertebrate animal models. In the final section the advantages of using Drosophila
melanogaster as a model for human neurodegenerative diseases will be specifically addressed.
2
1.2 Alzheimer’s Disease
Alzhiemer’s Disease (AD) is the most prevalent form of dementia and affects an estimated 5-8%
of the Canadian population (Anonymous1994;Lindsay et al. 2004). At onset, AD symptoms
include mild cognitive impairment that gradually progresses to severe, incapacitating memory
loss. According to the Diagnostic and Statistical Manual of Mental Disorders (DSM IV) AD
may be suspected if a patient presents with an impaired ability to learn new information or recall
previously learned information concomitant with other cognitive disturbances affecting
language, recognition or motor ability. The duration of disease can range from 1-25 years and
the cause of death is usually associated with malnutrition or pneumonia (Bird. 2008). A positive
AD diagnosis relies both on clinical as well as postmortem neuropathological assessments by
autopsy examination of the brain. AD brains exhibit gross cerebral cortex atrophy as well as
microscopic extracellular amyloid-β (Aβ) neuritic plaques and intraneuronal neurofibrillary
tangles (NFT) (Bird. 2008). NFTs are composed of the hyperphosphorylated form of the
microtubule-associated protein tau. Neuritic plaques, on the other hand, are composed primarily
of a 42- and to a lesser extent a 40-residue β-amyloid peptide (Aβ1-42 and Aβ1-40), amyloidogenic
cleavage products of the amyloid precursor protein (APP).
Synaptic loss in the frontal cortex has been documented in AD brains by electron microscopy
studies and correlates well with cognitive impairment highlighting the importance of preserved
synaptic connectivity for normal brain function (DeKosky and Scheff. 1990).
Immunohistochemical analysis (IHC) has revealed the specific loss of cholinergic neurons which
utilize acetylcholine as a neurotransmitter in AD brains (Geula and Mesulam. 1989). IHC has
also demonstrated intracellular NFT accumulation in AD affected brains (Giannakopoulos et al.
2003) and the deposition of NFTs also correlates with the degree of clinical impairment
(Giannakopoulos et al. 2003). Conversely, although some studies suggest a positive correlation
between increasing numbers of neuritic plaques and cognitive impairment (Hyman. 1997) others
studies show poor correlation between both the distribution and quantity of plaques and clinical
symptoms (Giannakopoulos et al. 2003) as well as neuronal cell loss (Terry et al. 1991). This is
not meant to imply that neuritic plaques do not play a role in AD pathogenesis but rather to
3
suggest that plaque deposition is not a reliable predictor of the degree of cognitive impairment in
AD.
1.2.1 Genetics of Alzheimer’s Disease
Although the majority of AD cases occur sporadically (sAD) approximately 10% are inherited
and are specifically referred to as familial AD (fAD). fAD is clinically and pathologically
indistinguishable from sAD with the exception that the onset of symptoms typically occurs
earlier in fAD (Bird. 2008). To date three genes have been identified that segregate with fAD
(Goate et al. 1991b).
1.2.1.1 The Amyloid Precursor Protein gene
The first fAD–linked gene mutation to be identified was a missense mutation in the amyloid
precursor protein (APP) gene (Goate et al. 1991b). Today over 20 fAD-mutations in the APP
gene have been identified (Gotz and Ittner. 2008). The APP gene encodes a type 1
transmembrane protein that is synthesized in the endoplasmic reticulum (ER), trafficked along
the secretory pathway to the plasma membrane where it undergoes proteolysis. APP proteolysis
is particularly relevant to AD because aberrant APP proteolysis promotes the deposition of β-
amyloid plaques (Figure 1). The extracellular N-terminal of APP can be cleaved by one of two
proteolytic complexes, α-secretase or β-secretase. α-secretase cleavage is catalyzed by the
tumour necrosis factor-α converting enzyme (TACE), which cleaves APP within the β-amyloid
domain precluding Aβ generation. For this reason α-secretase cleavage of APP is referred to as
the non-amyloidogenic pathway. β-secretase activity is attributed to the β-site-APP cleaving
enzyme (BACE). Subsequent to the N-terminal cleavage of APP by either β- or α-secretase,
APP undergoes intramembraneous proteolysis mediated by the γ-secretase complex. The γ-
secretase complex is composed of four proteins, presenilin (Psn) , presenilin enhancer-2 (Pen-2),
nicastrin (Nic) and anterior pharynx defective-1 (Aph-1). γ-secretase cleavage is heterogeneous
4
Figure 1. Schematic of intramembraneous APP proteolysis.
The amyloidogenic pathway consists of the sequential cleavage of APP by β-secretase and γ-
secretase liberating the AICD as well as the Aβ peptide. In the non-amyloidogenic pathway α-
secretase cuts within the Aβ domain thereby precluding the formation of Aβ peptides.
5
and produces various Aβ peptides ranging from 39-43 amino acids in length (Wolfe. 2007).
Importantly, fAD-mutations in APP cluster around proteolytic sites and favour the generation of
Aβ42, a more hydrophobic and aggregation prone Aβ species. The sequential cleavage of APP
by β-secretase followed by γ-secretase is referred to as the amyloidogenic pathway.
In addition to Aβ generation, γ-secretase cleavage also liberates an APP intracellular domain
fragment (AICD) which has been implicated in transcriptional activation (Cao and Sudhof. 2001)
as well intracellular calcium signalling (Leissring et al. 2002). The precise function of the APP
holoprotein is unclear and complete ablation of the APP gene does not give rise to any
substantial phenotypes in mice (Zheng et al. 1996). The lack of deleterious phenotypes in APP
knock-out animals has been attributed to genetic redundancy since the mammalian genome
encodes two APP-like genes which are highly homologous to APP and hence likely to
compensate functionally for the loss of APP.
1.2.1.2 The Presenilin genes
Shortly after mutations in the APP gene were linked to fAD, mutations in two homolgous genes,
presenilin-1 (PSEN1) (Sherrington et al. 1995) and presenilin-2 (PSEN2) (Rogaev et al. 1995)
were also found to segregate with fAD. The vast majority of these mutations are missense
mutations in the PSEN1 gene, which we now know account for over 50% of all fAD cases.
PSEN1 and PSEN2 protein topology includes 8 transmembrane domains with the N- and C-
termini as well as a large cytoplasmic loop residing in the cytoplasm (Figure 2). Presenilins are
broadly expressed in vertebrates including the brain (Lee et al. 1996). At a cellular level
presenilins primarily localize to the ER and Golgi apparatus and to a lesser extent to the plasma
membrane (Berezovska et al. 2003). Along the secretory pathway the presenilin holoprotein
undergoes proteolysis by an unknown protease generating a stable amino (N) and carboxy (C)
terminal heterodimer (Thinakaran et al. 1996). The proteolysis and stability of presenilin is
dependent on its association with the three other molecules composing the γ-secretase complex,
nicastrin, Aph-1 and Pen-2 (Takasugi et al. 2003). Apart from APP, the γ-secretase complex
6
Figure 2. The γ-secretase complex
The γ-secretase complex consists of four proteins, presenilin, nicastrin, anterior pharynx
defective-1 (Aph-1) and presenilin enhancer-2 (Pen-2). Presenilin is believed to be the catalytic
core of the γ-secretase complex since mutating two key aspartate residues (black asterisk) within
the presenilin protein eliminates γ-secretase activity. Together, this complex has proteolytic
activity and is known to cleave a variety of type 1 transmembrane proteins including APP and
Notch.
7
cleaves a variety of other type-1 transmembrane proteins including Notch, a highly conserved
protein that regulates numerous developmental events in most multicellular organisms
(Herreman et al. 2000;Zhang et al. 2000). The proteolytic activity of the γ-secretase complex is
attributed to presenilin since ablation of two aspartate residues in presenilin 1 has been shown to
abolish γ-secretase activity (Wolfe et al. 1999). Importantly, fAD-mutations in PSEN1 and
PSEN 2 promote the generation of the aggregation prone Aβ42 and hence facilitate neuritic
plaque formation.
Mice lacking presenilin 1 (PS1) die shortly after birth and exhibit significant neuronal loss,
impaired neurogenesis as well as hemorrhaging of the CNS (Shen et al. 1997). Mice deficient in
presenilin 2 (PS2), on the other hand, are viable and fertile (Herreman et al. 1999). PS1 and PS2
compound mutants die earlier than PS1 knock-out mice with phenotypes closely resembling
those of Notch-1 deficient mice (Herreman et al. 1999), a key γ-secretase substrate during
development. Hence, while PS2 activity does appear dispensable for normal Notch signalling
and embryonic development, it is not completely redundant to PS1.
Given that elimination of both PS1 and PS2 from the mouse genome is lethal several groups
have taken a conditional knock-out approach to eliminate both PS1 and PS2 specifically in the
postnatal mouse forebrain (Saura et al. 2004;Chen et al. 2008). Conditional inactivation of both
PS1 and PS2 faithfully recapitulates AD phenotypes including progressive neuronal
degeneration, memory deficits as well tau hyperphosphorylation. Although these results appear
to be at odds with the apparent gain-of-function mechanisms responsible for Aβ generation, as
well as the dominant mode of inheritance of fAD, several observations have provided compelling
evidence to account for this apparent paradox. First, mice expressing fAD-PS1 in a PS1-null
heterozygous background exhibit accelerated Aβ plaque deposition suggesting that reduced
presenilin function exacerbates the amyloid cascade (Wang et al. 2006). Second, successive
cleavage of Aβ by γ-secretase from longer to shorter Aβ peptides has been documented and
suggests that accumulation of the longer Aβ42 peptide may actually be indicative of incomplete
Aβ proteolysis, which is consistent with a partial loss of γ-secretase activity (Qi-Takahara et al.
2005). Third, unlike mutations in APP, which cluster around APP cleavage sites, mutations in
PS1 and PS2 are scattered throughout the protein, which is also most consistent with a partial
loss-of-function mechanism, possibly attributed to alterations in protein structure or stability.
8
Taken together, these results are consistent with an autosomal dominant mode of inheritance yet
also support a model where several of the characteristics of AD, including plaque deposition,
memory deficits, tau phosphorylation and neurodegeneration are due to partial loss of presenilin
function.
1.2.2 Alzheimer’s Disease aetiology
For most AD cases the aetiology is unknown however, a combination of genetic and
environmental factors is believed to be involved. Although no single monocausal event is likely
to account for all AD cases, several aberrant cellular processes have been implicated in both
sporadic as well as familial AD aetiology. In this section evidence supporting and opposing two
prevalent theories of AD, the amyloid cascade and intracellular calcium deregulation, will be
summarized and discussed.
1.2.2.1 The amyloid cascade hypothesis
The amyloid cascade hypothesis maintains that aberrant APP proteolysis followed by
progressive Aβ deposition and neuritic plaque formation is the primary cause of AD. For many
years this theory was the dominant model for AD pathogenesis. In order for the amyloid
hypothesis to be valid, the deposition of Aβ must mechanistically be linked to the genesis of all
phenotypic traits associated with AD and it must precede other AD-linked toxic cellular events
such as NFT deposition and neuronal dysfunction. The amyloid hypothesis was first proposed in
1992 and it was supported by several compelling observations (Hardy and Higgins. 1992). First,
Aβ42 and to a lesser extend Aβ40 peptides are the primary components of neuritic plaques, which
are cellular hallmarks of AD. Second, the majority of mutations in known fAD genes alter APP
proteolysis and augment the production of Aβ42 (Tanzi and Bertram. 2005). Third, fAD-
mutations in the APP gene are fully penetrant and hence can induce the onset of AD concomitant
with elevated plaque deposition. Fourth, although the nature of its toxicity is uncertain,
9
microinjection of Aβ into aged rhesus monkey brains results in neuronal loss and tau
phosphorylation (Geula et al. 1998). In addition, individuals with Down’s syndrome (trisomy of
chromosome 21) have three copies of the APP gene and invariably develop early onset AD,
including NFT deposition. Finally, Aβ immunotherapy can decrease Aβ and NFT levels in a
mouse fAD model expressing fAD-APP, fAD-PS1 and a frontal-temporal dementia associated
tau mutation (Oddo et al. 2004).
Equally compelling evidence exists that opposes the validity of the amyloid cascade hypothesis.
For example, while neuritic plaque formation remains a histopathological marker that defines
AD, spatial and temporal plaque deposition does not always correlate well with clinical AD
symptoms (Giannakopoulos et al. 2003). Furthermore, amyloid deposition does not always
precede NFT accumulation (Schonheit, Zarski, and Ohm. 2004) and expression of fAD-APP and
fAD-PS1 in the mouse model fails to recapitulate all AD phenotypic traits including NFT
deposition and neurodegeneration (Wong et al. 2002;Irizarry et al. 1997). Ultimately, while APP
proteolysis is unquestionably linked to AD aetiology it is unlikely that Aβ generation is the
primary and monocausal toxic mechanism responsible for triggering AD pathogenesis.
1.2.2.2 The calcium hypothesis
The calcium hypothesis maintains that defects in intracellular calcium homeostasis are the
primary cause of AD pathogenesis. In order for the calcium hypothesis to be accepted it must be
mechanistically linked to all AD phenotypic traits. The calcium hypothesis was initially
proposed in 1992 based on observations that calcium-dependent neuronal processes, which
regulate synaptic plasticity, learning and memory as well as cell survival, deteriorate with age
(Thibault, Gant, and Landfield. 2007). With the discovery that fAD-mutations in PS1 and PS2
cause profound disruptions in intracellular calcium storage and release, the calcium hypothesis
gained a great deal of credibility and quickly became an intensely studied field.
10
In healthy neurons intracellular calcium homeostasis is tightly regulated (Figure 3). Calcium can
enter the cytosol from the extracellular space through calcium channels present at the plasma
membrane or it can be released from internal stores, the largest of which is the ER. The
concentration of intracellular free calcium in most neurons is about 100 nM, relative to the mM
and μM calcium concentration in the extracellular space and ER, respectively. The ER has two
primary calcium channels the Ryanodine receptor (RyR) and the inositol triphosphate receptor
(IP3R). Calcium entry across plasma membrane calcium channels can stimulate ER calcium
release through RyRs. Alternatively, activation of phospholipase C leads to production of IP3,
which can activate calcium release from the ER through IP3Rs. The activity of both the RyR and
IP3R can be further modulated by calcium binding proteins such as calmodulin (Cam). Calcium
binding proteins, such as Cam, buffer incoming calcium and upon binding calcium, transduce the
calcium signal by impacting the activity of various enzymes and channels. For example, Cam
can directly bind to and impact the activity of both the RyR as well as the IP3R in a calcium
dependent fashion (Balshaw, Yamaguchi, and Meissner. 2002). Excess calcium is cleared from
the cytosol by actively pumping the calcium across the plasma membrane by calcium pumps or
exchangers or into internal stores via the sarco-endoplasmic reticulum ATPase (SERCA) pump.
Changes in cytosolic calcium concentration normally function as a second messenger system,
mediating a wide range of cellular processes relevant to AD including synaptic plasticity as well
as apoptosis. Long term potentiation (LTP) is an example of synaptic plasticity that is
particularly relevant to AD since LTP is believed to be the cellular correlate of learning and
memory. Specifically, LTP describes the long-lasting facilitation of chemical synaptic
transmission after repeated stimulation. LTP is dependent on calcium ions entering the cystosol
since loading of the postsynaptic neuron with a calcium chelator blocks LTP (Nicoll, Malenka,
and Kauer. 1989). Thus, LTP is dependent on transient increases in cytosolic calcium, which in
normal neurons is quickly cleared by mechanisms described in the previous section. However,
during aging and in disease states the ability of a cell to recover from a transient rise in cytosolic
calcium may become compromised and this deficiency can lead to cell death. Apoptosis is a
calcium-dependent programmed cell death mechanism. Apoptosis can be induced by various
intracellular or extracellular signals that result in sustained elevation of intracellular calcium
(Mattson and Chan. 2003). Persistent high levels of cytosolic calcium lead to changes in
11
Figure 3. Intracellular calcium signalling
Intracellular calcium is a tightly regulated process invols the sequestration of calcium ions into
the ER throught the SERCA pump. The reseting calcium concentration in most neurons is in an
nM range relative to the µM and mM range found in the cytosol and extracellular space,
respectively. Release of calcium from internal stores through the RyR and IP3R calcium
channels triggers signalling pathways necessary for coordinating apoptosis, synaptic plasticity as
well as gene transcription. The activity of calcium chanels is modulated by calcium binding
proteins such as calmodulin (Cam) as well as presenilin, which also resides in the ER.
12
mitochondrial membrane permeability, which results in the release of cytochrome C. Once
released, cytochrome C binds to and stimulates more calcium release from the IP3R, which
subsequently activates caspases and nucleases to initiate a proteolytic cascade required for cell
death (Boehning et al. 2003).
Intracellular calcium deregulation has been observed in fibroblasts from fAD patients
(Etcheberrigaray et al. 1998;Zatti et al. 2006) as well as in primary neuronal cultures from fAD-
PS1 transgenic animals (Zatti et al. 2006;Guo et al. 1996;Leissring et al. 2000) and in cells
transiently transfected with fAD-PS1 (Zatti et al. 2006;Chan et al. 2000;Cheung et al. 2008) and
PS2 (Zatti et al. 2006;Cheung et al. 2008;Zatti et al. 2004). The majority of studies suggest that
mutations in PS1 or PS2 potentiate calcium release from both RyR and IP3R sensitive stores.
How they do this remains uncertain. Furthermore, while some groups report overloaded calcium
stores in fAD-presenilin expressing cells, others observe the exact opposite. Those that
subscribe to the calcium-overload hypothesis believe that elevated calcium release in fAD-
presenilin expressing cells is driven by overloaded calcium stores. Conversely, others believe
that deficits in internal calcium stores is a consequence of exaggerated calcium release.
Presenilins have also been implicated in affecting the activity or expression of RyRs (Stutzmann
et al. 2007;Hayrapetyan et al. 2008;Rybalchenko et al. 2008) and IP3R (Cheung et al. 2008). In
addition, a separate but not mutually exclusive hypothesis suggests that presenilins may function
as passive calcium channels and that fAD-mutations in PS1 disable the passive calcium channel
activity leading to overloaded stores (Tu et al. 2006). However, others have been unable to
detect intrinsic calcium channel activity in presenilin (Cheung et al. 2008). Ultimately, although
the specific mechanism remain uncertain, what we do know is that presenilins are involved in
intracellular calcium homeostasis.
Deregulation of intracellular calcium has also been linked to Aβ generation and NFT formation.
Knocking down the activity of IP3R has been shown to reduce Aβ generation (Cheung et al.
2008) and stimulating release of calcium from the RyR has been implicated in increasing Aβ
production (Querfurth et al. 1997). Furthermore, tau is known to be phosphorytlated by a
calmodulin-dependent kinase (CaKII) (Steiner et al. 1990). Hence, intracellular calcium
signalling is linked to the generation of two histological hallmarks of AD. Finally, recently a
polymorphism in a novel calcium channel, calcium homeostasis modulator 1 (CALHM1), has
13
been identified as a risk factor in sporadic AD (Dreses-Werringloer et al. 2008). CALHM1
localizes primarily to the ER but can also be found at the plasma membrane and knocking down
CALHM1 appears to increase Aβ generation. Hence, calcium dysfunction has now also been
implicated in sporadic AD.
1.2.3 Therapeutic treatments
The observation that concomitant with cholinergic cell loss, AD brains also exhibit deficits in
acetylcholine levels prompted the cholinergic deficit hypothesis, which dominated much of the
early 1980’s (van Marum. 2008). Consequently, several drugs were developed that increase
acetylcholine levels in the CNS. These drugs have been shown to have modest clinical benefits
in AD by temporarily slowing down the rate of cognitive decline by 6-12 months (van Marum.
2008). Although measurable, the beneficial effects of these drugs are not large.
Memantine targets the plasma membrane calcium channel and glutamate receptor NMDA (N-
methyl-D-aspartate) by inhibiting calcium influx into the cytosol during aberrant synaptic
transmission. The clinical efficacy of mementine, however, is lower than that of drugs targeting
acetylcholine (van Marum. 2008).
Finally, much effort has been made to develop anti-amyloid therapies. One much anticipated
clinical trial involved an Aβ1-42 vaccination. Unfortunately, this trial did not reveal any benefits
with immunization and ended in disaster with the occurrence of meningeoencephalitis in a subset
of patients receiving vaccination (Gilman et al. 2005). Given that the γ-secretase complex has
many proteolytic substrates general γ-secretase inhibition is not considered to be a good
therapeutic option. Ultimately, despite intense research there are currently no effective
treatments that halt the progression of AD. As researchers unravel the finer mechanistic details
involved in AD aetiology more specific drugs will certainly be developed that target precise AD-
pathogenic processes.
14
1.3 Amyotrophic Lateral Sclerosis
Amyotrophic Lateral Sclerosis (ALS) is a fatal neurodegenerative disease defined by rapid and
progressive muscle weakness leading to paralysis. ALS, also known as Motor Neuron Disease
(MND) or Lou Gehrig’s Disease, is attributed to degeneration of motor neurons in the motor
cortex, brain stem and spinal cord. Today, approximately 3000 Canadians are living with ALS
and 50% of these case are expected to live only 3-4 years after the onset of symptoms
(Mitsumoto. 1997). Death is typically attributed to denervation of respiratory muscles and
diaphragm leading to failure of the respiratory system. ALS involves the progressive spread of
degeneration in lower and upper motor neurons (LMN and UMN, respectively) in at least one of
the four anatomical regions of the body (bulbar, cervical, thoracic or lumbosacral) (Brooks.
1994). Upper motor neurons are neurons that project to motor neurons while lower motor
neurons are neurons that innervate muscle. Clinical signs of ALS include loss of dexterity,
progressive muscle weakness and atrophy, spasticity, as well as the onset of pathological reflexes
each of which reflects progressive muscle denervation and death of motor neurons in any of the
four aforementioned anatomical regions (Mitsumoto. 1997;Brooks. 1994). No specific test exists
to unequivocally diagnose ALS and since ALS symptoms are common to several other
neuromuscular and skeletal disorders, an integral part of diagnosing ALS includes excluding
other possible causes using electromyographs (EMGs), X-rays, neuroimaing as well as muscle
and nerve biopsies. Hence, an ALS diagnosis is supported by the absence of evidence of
neuromuscular abnormalities or other injuries as well as the absence of sensory or cognitive
dysfunction. A positive ALS diagnosis is also supported by the presence of microscopic
ubiquitin-positive cytoplasmic inclusion bodies. Inclusion bodies being cytoplasmic depositions
of misfolded proteins aggregated into ubiquitin-positive proteinaceous bodies.
Hyperphosphorylated neurofilaments concomitant with motor neuron cell loss in the brainstem,
motor cortex and spine are also diagnostic signs of ALS (Mitsumoto. 1997;Brooks. 1994).
15
1.3.1 Genetics of Amyotrophic Lateral Sclerosis
Familial ALS (fALS) occurs in 5-10% of ALS cases and is clinically and neuropathologically
indistinguishable from sporadic ALS (sALS). Approximately 20-25% of fALS cases have been
attributed to missense mutations in the superoxide dismutase 1 gene (SOD1) and the vast
majority of these SOD1-fALS cases are inherited in an autosomal-dominant fashion. Several
other genes have been linked to fALS but validation of their involvement in ALS aetiology is
still pending. In the following section the validated and most intensely studied fALS-associated
gene, SOD1, will be discussed.
1.3.1.1 The superoxide dismutase 1 gene
In 1993 missense mutations in the SOD1 gene were found to segregate with fALS (Rosen et al.
1993). SOD1 encodes a ubiquitously expressed metalloenzyme that catalyzes the dismutation
(conversion) of superoxide radicals (O2˙¯ ) into oxygen (O2) and hydrogen peroxide (H2O2).
Hydrogen peroxide is subsequently converted into water by glutathione peroxidase or catalase.
SOD1 is a critical component of a cell’s anti-oxidant defense system against reactive oxygen
species (ROS), which are a by-product of oxidative metabolism. SOD1 functions as a
homodimer with each subunit containing one copper and zinc atom. The superoxide anion is
guided into the SOD1 catalytic site by the positively charged copper atom (Cu2+
), while the zinc
atom (Zn2+
) stabilizes the dismutation reaction and ensures the rapid purging of hydrogen
peroxide from the active site (Hand and Rouleau. 2002). Within the cell, SOD1 resides primarily
in the cytoplasm.
Today over 100 fALS-SOD1 mutations have been identified and these mutations are distributed
across all five exons of the SOD1 gene (Andersen et al. 2003). The clinical characteristics, age
of onset, and biochemical characteristics of the various fALS-SOD1 mutant proteins vary
greatly. Furthermore, the level of residual dismutase activity does not correlate with disease
severity suggesting that loss of SOD1 function is unlikely to be the primary cause of SOD1-
16
associated pathogenesis (Majoor-Krakauer, Willems, and Hofman. 2003;Boillee, Vande Velde,
and Cleveland. 2006). This theory was corroborated in SOD1 null mice, which do not develop
motor neuron disease and in fact are viable (Reaume et al. 1996;Lino, Schneider, and Caroni.
2002). Ubiquitous overexpression of fALS-SOD1 mutants, on the other hand, does give rise to
many of the characteristic features of ALS including progressive motor deficits, motor neuron
degeneration, formation of microscopic inclusion bodies and premature death (Gurney et al.
1994;Bruijn et al. 1997;Dal Canto and Gurney. 1997;Jaarsma et al. 2000). Importantly, SOD1
toxicity is dependent on gene-dosage hence, higher mutant protein expression levels positively
correlate with earlier disease onset and more severe progression of symptoms (Gurney et al.
1994;Dal Canto and Gurney. 1997).
The selective toxicity of fALS-SOD1 mutations to motor neurons is believed to involve both
cell-autonomous motor neuron defects as well as non-cell autonomous interactions between
motor neurons and glia. In accordance, expression of fALS-SOD1 in motor neurons (Lino,
Schneider, and Caroni. 2002;Lino, Schneider, and Caroni. 2002) or glia (Gong et al. 2000) alone
does not give rise to motor neuron disease. However, in an in vitro model of ALS, co-culturing
wild type motor neurons with fALS-SOD1 glia has been shown to decrease the survival of wild
type motor neurons (Di Giorgio et al. 2007). Furthermore, while diminishing fALS-SOD1
expression exclusively in glia does not have an impact on disease onset, it does slow down
disease progression and extends the lifespan of chimeric fALS-SOD1 transgenic mice (Boillee et
al. 2006). Finally, new evidence has surfaced which suggests that fALS-SOD1 expressing
astrocytes produce a substance that is selectively toxic to motor neurons (Nagai et al. 2007). In
fact, several aberrant processes involving glial cells including defective removal of
neurotransmitters from the synaptic cleft, as well as the production of toxins and inflammatory
factors during disease state, have been implicated in motor neuron degeneration (Bruijn et al.
1997;Di Giorgio et al. 2007;Nagai et al. 2007;Clement et al. 2003). Ultimately, the specific
mechanisms and the cellular origin of SOD1 induced toxicity remain uncertain and are active
areas of research.
17
1.3.2 Amyotrophic lateral sclerosis aetiology
The aetiology of most ALS cases is unknown but is believed to involve both genetic as well as
environmental factors. Multiple theories exist that attempt to account for motor neuron
degeneration in ALS including oxidative stress, glutamate excitoxicity, mitochondrial
dysfunction as well as protein aggregation. In the proceeding section two prevalent theories,
protein aggregation and oxidative stress, will be discussed specifically in the context of SOD1-
associated ALS.
1.3.2.1 The protein aggregation hypothesis
The protein aggregation hypothesis maintains that fALS-SOD1 mutant proteins misfold and
oligomerize into high-molecular-weight species that aggregate into proteinaceous inclusions,
which are selectively toxic to motor neurons. In support of this theory intracellular inclusions
containing SOD1, ubiquitin, heat shock chaperone proteins as well as the proteasome have been
found in tissues taken from fALS patients and ALS mouse models (Watanabe et al. 2001).
Although the precise mechanism by which inclusion bodies selectively damage motor neurons
remains unknown, sequestration of vital chaperones may be involved. Indeed, inclusion bodies
often sequester heat shock protein 70 (Hsp70) and chaperone activity is generally decreased in
fALS-SOD1 transgenic animals (Tummala et al. 2005). In addition, overexpression of Hsp70
has been shown to reduce the formation of SOD1-positive aggregates and to preserve the
viability of cultured primary motor neurons expressing fALS-SOD1 (Bruening et al. 1999).
Inclusion bodies may also impair protein degradation by sequestering the proteasome thus
impairing its ability to degrade misfolded proteins. SOD1 degradation is mediated by the
ubiquitin-proteasome pathway and inhibiting this proteolytic pathway increases fALS-SOD1
aggregate formation in spinal cord slices taken from fALS-SOD1 transgenic animals (Puttaparthi
et al. 2003). fALS-SOD1 aggregate formation, however, can be reversed by restoring
proteasome activity (Puttaparthi et al. 2003).
18
Although aggregates are a cellular hallmark of ALS, whether or not aggregates are actually
cytotoxic is a long standing debate. In support of the aggregation hypothesis, SOD1 aggregates
have been documented to coincide with the onset of neurodegeneration in mouse ALS models
(Bruijn et al. 1997;Jaarsma et al. 2000;Jonsson et al. 2004;Wong et al. 1995). In addition, fALS-
SOD1 proteins aggregate more readily relative to wild type proteins (Wang, Xu, and Borchelt.
2002). Finally, time-lapse studies in live cells have revealed a strong correlation between the
appearance of fALS-SOD1 aggregates and subsequent cell death (Matsumoto et al. 2005).
Conversely, given that aggregate formation is a late-stage event it may actually be a secondary
effect or potentially even represent a cellular defense mechanism, whereby misfolded, toxic
proteins are actively sequestered into inclusion bodies.
1.3.2.2 The oxidative stress hypothesis
The oxidative stress hypothesis maintains that fALS-associated mutant SOD1 proteins have
altered substrate affinity and/or engage in aberrant activity that results in the production of toxic
free-radicals and that these radicals are the primary trigger of the pathogenic process that gives
rise to ALS. In addition to its conventional dismutase activity, SOD1 can also act as a
peroxidase and thus can convert hydrogen peroxide into toxic hydroxyl radicals (Peled-Kamar et
al. 1997;Yim et al. 1996). Normally, access to the wild type SOD1 active site is limited by size
and charge, favouring superoxide as a substrate and excluding larger molecules such as hydrogen
peroxide. However, fALS-SOD1 has a higher affinity for hydrogen peroxide relative to wild
type SOD1 enabling fALS-SOD1 peroxidase function and consequently hydroxyl radical
production (Yim et al. 1996). The hydroxyl radical is highly reactive and can cause oxidative
damage to DNA, lipids and proteins including to SOD1 itself (Andrus et al. 1998). In support of
this theory there is evidence of oxidative pathology in human sporadic and familial ALS patients
including increased oxidative-stress associated damage to DNA, phospholipids and proteins
(Ferrante et al. 1997;Shaw et al. 1995).
An extension of the oxidative stress hypothesis suggests that oxidative damage to SOD1 itself
promotes misfolding, which leads to the oligomerization of misfolded SOD1 into toxic inclusion
19
bodies thus linking oxidative stress to protein aggregation. In support of this theory, SOD1 has
been shown to be one of the most heavily oxidized proteins identified in fALS-SOD1 transgenic
mice as well as in fALS patients (Andrus et al. 1998). Several factors make SOD1 selectively
vulnerable to oxidative stress specifically in motor neurons. First, SOD1 is highly abundant in
motor neurons (Pardo et al. 1995). In addition it has a particularly long half-life in motor
neurons (Borchelt et al. 1998). Finally, motor neurons are very large cells with high energy
demands. This energy is supplied by high metabolic rates and consequently, elevated oxidative
stress. Both fALS-SOD1 and to a lesser extent wild type SOD1 have been shown to be
susceptible to hydrogen peroxide induced oxidative damage, which leads to misfolding and
monomerization of the SOD1 homodimer prior to aggregation (Rakhit et al. 2004). Studies
using an antibody specifically engineered to only detect misfolded-monomeric SOD1 have
revealed the accumulation of misfolded-monomeric fALS-SOD1 in fALS patients as well as in
fALS-SOD1 rodent models specifically in motor neurons prior to neurodegeneration (Rakhit et
al. 2007). However, the degree to which monomer-misfolded SOD1 correlates with ALS disease
progression remains to be determined.
1.3.3 Therapeutic treatments
Currently, ALS treatments focus primarily on relieving symptoms and maintaining an optimum
quality of life. Riluzole is one of the few drugs that appears to have a modest effect on
prolonging the life of ALS patients (Lacomblez et al. 1996). The mechanism of action of
riluzole is uncertain but it may involve alleviating excitotoxic stress by inhibiting glutamate
release (Hand and Rouleau. 2002). A recent study, however, has suggested that riluzole can also
increase the expression of heat shock chaperone proteins and hence the beneficial impact of
riluzole may also, at least in part, include amelioration of protein misfolding and hence
aggregation (Yang et al. 2008).
20
1.4 Drosophila as a model of human neurodegenerative diseases
Numerous features of the fruit fly, Drosophila melanogaster, make it an ideal model organism to
study human neurodegenerative diseases. First, flies have a very short generation time. The life
cycle of a fly is between 10-12 days at 25 °C. Newly laid eggs take 24 hours to undergo
embryogenesis before hatching into first instar larvae, which continue to develop for another 48
hours into second, and then third instar larvae. Two days later, larvae transform into immobile
pupa, undergo metamorphosis and eclose in adult form 5-7 days later. Importantly, adult
females are fertile 12 hours post-eclosion and a single fly can produce hundreds of offspring
within days making it relatively easy to perform analysis on hundreds of flies within a matter of
days.
Many of the genes, cellular processes and basic building blocks of the nervous system are
conserved in flies (Yoshihara, Ensminger, and Littleton. 2001). Cross-genomic comparisons of
the Drosophila and human genomes have revealed a remarkably high degree of conservation
(Adams et al. 2000;Rubin et al. 2000). Over 60% of known human disease causing genes have a
fly orthologue (Rubin et al. 2000). What’s more, there is considerably less genetic redundancy
in Drosophila relative to vertebrate models hence, characterization of disease-causing gene
function is often less complicated. Many of the cellular processes known to be involved in
neurodegeneration including the signalling cascades that orchestrate apoptosis, intracellular
calcium homeostasis as well as oxidative stress are also conserved in flies, as are the proteins
that mediate these processes. The Drosophila nervous system is composed of some 200,000
neurons and supporting glia relative to the millions of neurons found in the mammalian brain.
Although simpler, fly neurophysiology is very similar to it’s mammalian counterpart. For
example, fly neurons exhibit synaptic plasticity and neurotransmission mediated by many of the
same neurotransmitters, synaptic proteins, receptors and ion channels found in the mammalian
brain (Yoshihara, Ensminger, and Littleton. 2001). Flies also exhibit complex behaviours and
like humans, many of these behaviours deteriorate with age including learning, memory and
motor ability (Mockett et al. 2003;Simon, Liang, and Krantz. 2006).
Finally, and perhaps most importantly, the Drosophila genome can be used to perform
sophisticated genetic analyses designed to further elucidate the pathological processes associated
21
with normal as well as aberrant disease gene activity. Genome manipulation in flies includes
exploiting naturally occurring transposable elements referred to as P-elements. Thousands of P-
element lines, representing independent transposon insertions spanning the majority of the fly
genome are available to the fly research community. These insertions often, but not always,
impact the expression or activity of the gene(s) proximal to the insertion site. Mobilization of P-
elements can also give rise to precise as well as imprecise excisions of the transposon, the latter
removing not only the sequence corresponding to the P-element but also the flanking sequence of
the proximal gene(s). Imprecise excision of P-elements is a common method for generating
genetic nulls in flies. P-elements can also be used as vectors to shuttle transgenes into the fly
genome to generate transgenics. The method of choice for targeting exogenous gene expression
in Drosophila is the bi-partite GAL4-UAS (upstream activating sequence) system (Figure 4)
(Brand and Perrimon. 1993). The insertion of the yeast GAL4 transcriptional activator
downstream of an endogenous fly promotor results in the expression of GAL4 in the same
spatial-temporal pattern characteristic of that promoter. Thus, when flies expressing GAL4 are
crossed to flies bearing a transgene downstream of the GAL4 binding UAS-sequence, the
progeny of this cross express the transgene in a spatial-temporal pattern determined by the
promoter driving GAL4 expression. The GAL4/UAS system can thus be used to determine
whether expression of wild type or mutant disease causing genes in various anatomical regions
of the fly results in observable phenotypes. With respect to generating a fly model of a human
neurodegenerative disesase, phenotypes that recapitulate human disease symptoms are in many
ways most desired but, not always necessary. Often studying a phenotype in a structure
unrelated to the human disease or even to human anatomy, for example the fly wing or retina,
can reveal valuable insights into the function of disease causing genes. So long as the relevant
signalling molecules are conserved, such studies may be very relevant to the human disease
process. Additionally, any phenotypic trait associated with the expression of a disease causing
gene can be subject to a genetic screen designed to isolate dominant mutations that suppress or
enhance the original disease gene induced phenotype. In such a way, the power of genetic
analysis in Drosophila can be used to identify novel molecules involved in human disease
processes, which can subsequently be validated in mammalian models. These topics will be
further explored in the proceeding sections with specific focus on how the fruit fly model has
been used to further our understanding of the normal and aberrant function of AD and ALS-
associated genes.
22
Figure 4. The GAL4/UAS system
The GAL4/UAS system is composed of two parts. The first part is the yeast GAL4
transcriptional activator which is under the control of a native promoter of interest. The second
is a short upstream activiating sequence (UAS) to which GAL4 binds and initiates transcription.
Thus the GAL4/UAS system enables tissue specifc transgene expression when flies expessing
GAL4 are crossed to flies bearing an UAS sequence upstream of a transgene. Adapted from:
Brand and Perrimon. 1993.
23
1.4.1 Drosophila models of Alzheimer’s Disease
Drosophila models of AD have included expressing wild type and fAD-mutant forms of APP
and presenilin. Drosophila has a single APP orthologue (Appl) which encodes a β-amyloid
precursor-like protein that notably lacks the Aβ domain. Although γ-secretase activity is
conserved in Drosophila, there is no known β-secretase orthologue hence, to date there is no
clear evidence that Aβ peptides are generated in flies. Overexpression of human Aβ42 in the
Drosophila CNS gives rise to, amyloid deposition, progressive learning defects, extensive
neurodegeneration and ultimately a shortened lifespan (Iijima et al. 2004). Another group took a
different approach and expressed Aβ42 in the Drosophila eye. Here too, overexpression of Aβ42
gave rise to progressive retinal degeneration concomitant with plaque formation (Finelli et al.
2004). The advantage of working with a retinal phenotype is that the eye is a non-essential organ
that is easy to score as a phenotypic trait in a genetic modifier screen, which is precisely what
was done. A random collection of P-element misexpression mutant strains (EP-strains) was used
to screen for suppressors or enhancers of the Aβ42-induced eye phenotype. Overexpression of the
Neprilysin 2 gene (Nep2) was identified as a suppressor of both retinal degeneration as well as
Aβ42 accumulation (Finelli et al. 2004). In addition, overexpression of Nep2 was shown to also
rescue the shortened lifespan observed in flies expressing Aβ42 in the fly CNS raising the
exciting possibility that increasing Nep2 activity could have therapeutic potential for AD (Finelli
et al. 2004). Nep2 encodes a metalloprotease that is known to be involved in Aβ42 degradation.
Importantly, Nep2 is conserved in mammals and transgenic mice overexpressing NEP2 and fAD-
APP exhibited significantly reduced amyloid plaque formation. Moreover, NEP2 expression
also rescued the premature lethality observed in mice expressing fAD-APP alone (Leissring et
al. 2003) thus validating that Drosophila can be used to identify and study the molecular
pathways involved in human neurodegenerative disease process.
The Drosophila genome encodes a single presenilin gene (Psn) (Boulianne et al. 1997). The
structure, proteolysis, trafficking and subcellular localization of Psn is conserved in flies as are
all the components and activity of the γ-secretase complex (Adams et al. 2000;Boulianne et al.
1997;Ye and Fortini. 1998;Fossgreen et al. 1998;Struhl and Greenwald. 1999). In accordance,
like mice, loss of Psn in flies gives rise to Notch-like phenotypes. Specifically, Psn-null flies
24
survive to early pupal stages due to maternal contribution and exhibit defective eye and wing
development as well as incomplete differentiation of central neurons (Ye and Fortini.
1998;Struhl and Greenwald. 1999;Guo et al. 1999). Elevated apoptosis is observed in the
developing eye of Psn-null larvae suggesting that like mice normal Psn activity plays a role in
cell survival (Ye and Fortini. 1999). Lack of maternal or zygotic Psn activity, on the other hand,
results in hyperplasia of the embryonic nervous system hence, loss of Psn can have pleiotropic
effects at different developmental stages (Ye, Lukinova, and Fortini. 1999). Overexpression of
wild type Psn in the fly eye also gives rise to apoptosis and co-expression of an fAD-form of Psn
exacerbates this phenotype (Ye and Fortini. 1999). Hence, both loss of Psn function as well as
its overxpression in the eye give rise to apoptosis. Likewise, both loss of Psn function (Struhl
and Greenwald. 1999), as well as its overexpression (Ye and Fortini. 1999) gives rise to wing
margin scalloping, a classic Notch loss-of-function phenotype. Given that Psn overexpression
can phenocopy Psn loss-of-function, it has been proposed that overexpression of Psn actually
gives rise to a dominant negative effect (Ye and Fortini. 1999). The nature of this dominant
negative effect is unclear but it may involve accumulated Psn holoprotein impairing ER function
or disrupting the assembly of active γ-complexes (Ye and Fortini. 1999).
Work performed in our lab has demonstrated that loss of Psn function in third instar larvae
results in learning deficits concomitant with impaired synaptic plasticity (Knight et al. 2007).
More specifically, Psn-null larvae were tested using olfactory and visual associative-learning
paradigms and in both assays mutant larvae performed significantly worse relative to wild type
controls. In another study, flies were generated that express either wild type or an fAD-mutant
form of Drosophila Psn under the control of the endogenous Psn promoter in a Psn-null genetic
background (Lu et al. 2007). These flies were subjected to heat shock stress and assayed for
long term memory using an olfactory associative learning paradigm. Flies expressing fAD-Psn
were shown to have significant long term memory deficits relative to flies expressing wild type
Psn, but only during the post-stress recovery period. Synaptic activity at the larval
neuromuscular junction (NMJ) was also measured pre- and post heat shock. Both Psn-null and
fAD-Psn but not wild type flies demonstrated prolonged calcium influx through voltage-gated
calcium channels on the plasma membrane but again, only post heat-shock, pointing to a role for
Psn in regulating post-stress calcium-channel activity. Finally, expression of 12/14 fAD-Psn
mutations under the control of the endogenous Psn-promoter only partially rescues Psn-null
25
lethality hence in flies, as in mice, fAD-mutations in Psn appear to be partial loss-of-function
mutations (Seidner et al. 2006). Lifespan analysis was not reported. The effects of wild type or
fAD-mutant Psn expression in the Drosophila CNS has also not been explored.
1.4.2 Drosophila models of Amyotrophic Lateral Sclerosis
The Drosophila genome encodes a single copper-zinc superoxide dismutase gene (Sod). Unlike
mice, flies are very sensitive to reductions in dismutase activity. Sod null flies die soon after
they eclose and exhibit significantly higher levels of oxidative stress relative to wild type
controls (Mockett et al. 2003;Phillips et al. 1989). The lethality of Sod null flies can be rescued
by introducing wild type human SOD1 into the fly genome, demonstrating that the fly and
human gene are functionally homologous (Parkes et al. 1998). Furthermore, there is evidence of
neuropathology in flies bearing a loss-of-function point mutation in the endogenous Sod gene
(Sod n108
) (Phillips et al. 1995). Degeneration of photoreceptor neurons in the retinas of Sod n108
flies was documented as early as 7 days post-eclosion (Phillips et al. 1995). In addition, Sod
loss-of-function mutants have also been shown to have reduced locomotor activity (Mockett et
al. 2003). Hence, unlike mice, loss of Sod function in flies can recapitulate some ALS-like
phenotypes. However, these phenotypes are all recessive, which is unlike the dominant, gain-of-
function mechanisms believed to underlie SOD1-associated ALS pathogenesis in mice and
humans.
Several studies in flies have taken a transgenic approach to express human fALS-SOD1 in
various parts of the fly. The first published study expressed human fALS-SOD1 specifically in
Drosophila motor neurons. The results were quite unexpected since expression of fALS-SOD1
actually extended lifespan and increased resistance to oxidative stress in flies (Elia et al. 1999).
While the precise mechanisms responsible are unclear, the ability of fALS-SOD1 to extend the
lifespan of flies is likely due to the positive outcome of elevating dismutase activity specifically
in motor neurons. In addition, though motor neurons are selectively killed in ALS, we know
from studies in mice that expression of fALS-SOD1 exclusively in motor neurons does not result
26
in motor neuron disease. Hence, perhaps in flies, like in mice, non-cell autonomous fALS-SOD1
toxicity is required to induce ALS-like phenotypes.
Another study used fly transgenics that expressed several different human fALS-SOD1
mutations under the control of the endogenous Drosophila Sod promoter (Mockett et al. 2003).
The longevity and climbing ability of these flies was measured in the context of a Sod null or Sod
wild type genetic background. Expression of fALS-SOD1 had no effect on longevity in a Sod
wild type background. However, in a Sod null background all fALS-SOD1 mutants partially
rescued the short lifespan of Sod null flies. fALS-SOD1 mutants also exhibited progressive
climbing impairments in a Sod null background. Specifically, while the walking speed of young
fALS-SOD1 mutant flies was initially no different to that of wild type flies, within a couple of
weeks the climbing ability of flies expressing fALS-SOD1 had dramatically declined relative to
controls. The climbing ability of fALS-SOD1 transgenics in a Sod wild type background was
not reported. At least two reasons may account for why these studies failed to produce gain-of-
function phenotypes. Both are based on the knowledge that high levels of fALS-SOD1 mutant
protein are often required to induce ALS-like phenotypes in other model systems. In the present
study each fALS-SOD1 transgene was present in only a single copy number. In addition, the
endogenous Sod promoter is not a strong driver of transgene expression. Hence, it is possible
that using a stronger, more robust promoter to drive fALS-SOD1 expression may be necessary to
induce gain-of-function phenotypes in flies.
Recently, a study was published that describes gain-of-function ALS-like phenotypes associated
with the expression of human SOD1 in flies (Watson et al. 2008). In this study, two fALS-SOD1
alleles as well as wild type SOD1 were used to generate independent fly transgenics which were
crossed to a motor neural GAL4 driver. As reported in earlier studies, motor neuron specific
expression of fALS-SOD1 did not impact fly lifespan. However, this study also measured
climbing ability. Flies expressing either wild type or fALS-mutant SOD1 were reported to have
progressive climbing deficits relative to flies expressing Drosophila Sod, although it should be
noted that some important controls were not included in this analysis precluding absolute
certainty that the observed climbing defects are indeed attributed to transgene expression. That
being said, the reported climbing defects were not accompanied by motor neuron loss but
electrophysiological examination did reveal progressive decrements in synaptic transmission.
Both wild type, as well as fALS-SOD1 mutant protein, were observed to accumulate in motor
27
neurons starting as early as day one, however, the SOD1 positive foci did not co-localize with
chaperone proteins indicating that SOD1 accumulation was not inducing a chaperone stress
response. It was noted, however, that a chaperone stress response was being induced in
surrounding glial cells. These studies have thus suggested that expression of fALS-SOD1 or
wild type human SOD1 in fly motor neurons can give rise to progressive climbing and synaptic
defects through cell-autonomous mechanisms. This is in contrast to work performed in the
mouse model, which argues that cell-autonomous mechanisms are necessary but not sufficient at
inducing ALS-like phenotypes (Boillee, Vande Velde, and Cleveland. 2006;Di Giorgio et al.
2007;Boillee et al. 2006;Nagai et al. 2007). What has not been explored, as of yet, is what
consequence fALS-SOD1 expression has on flies when driven by a strong, robust ubiquitous
driver.
1.5 Purpose of our studies
Using Drosophila as a model our ultimate goal is to further decipher the normal and aberrant
function of two genes linked to the onset of neurodegeneration in humans, presenilin and
superoxide dismutase. In the next chapter, we will present data that provides new insights into
the interaction of presenilin with the calcium release machinery. In the third chapter, we will
describe fALS-SOD1-induced locomotory deficits. In the final chapter we will discuss the
implication of our work in the context of the current understanding of the cellular, physiological
and behavioural processes involved in ALS and AD. We will also discuss how future studies
could pave the way towards novel effective treatments against neurodegeneration.
28
CHAPTER 2
2 Analysis of the interaction of presenilin with the intracellular
calcium stores machinery
Kinga Michno, Jorge M. Campussno, Diana van de Hoef , Diane O’Dowd and
Gabrielle L. Boulianne
A modified version of this chapter has been submitted for publication.
Statement of contribution:
A modified version of this chapter has been submitted for publication. Jorge M. Campussano
assisted with the generation of primary cultures and calcium imaging. Diana van de Hoef
constructed the wild type Drosophila presenilin expression construct used in the binding assay.
Commercially obtained reagents are indicated in the materials and methods. Kinga Michno
generated all other reagents and performed all other experiments and analysis as indicated in the
materials and methods. Diane O’Dowd pioneered the Drosophila primary culture protocol.
Calcium imaging analysis was performed under the co-supervision of Diane O’Dowd and
Gabrielle L. Boulianne. All other work was performed under the supervision of Gabrielle L.
Boulianne.
29
2.1 Abstract
Much of our current understanding about neurodegenerative diseases can be attributed to the
study of inherited forms of these disorders. For example, mutations in the presenilin 1 and 2
genes have been linked to early onset familial forms of Azheimer’s Disease (fAD). However, a
clear understanding of the sequence of presenilin-induced toxic events that trigger the onset of
dementia still remains largely unresolved. Using the Drosophila central nervous system as a
model we have investigated the role of presenilin in one of the earliest cellular defects associated
with Alzheimer’s Disease, intracellular calcium deregulation. We show that expression of either
wild type or fAD-mutant presenilin in Drosophila CNS neurons has no impact on resting
calcium levels but does give rise to deficits in internal calcium stores. Furthermore, we show
that a loss-of-function mutation in calmodulin, a key regulator of intracellular calcium, can
suppress the presenilin-induced deficits in intracellular calcium. Our data support a model
whereby presenilin plays a role in regulating intracellular calcium levels and demonstrates that
Drosophila can be used to study the link between presenilin and calcium deregulation.
30
2.2 Introduction
Alzheimer’s disease (AD) is a neurodegenerative disorder characterized clinically by progressive
dementia and, histopathologically, by the formation of senile neuritic plaques, neurofibrillary
tangles (NFT) and ultimately neuronal cell death. Despite being the most prevalent and intensely
studied form of dementia there is still no effective cure. Although the majority of AD cases are
sporadic, 10% are familial (fAD) and are inherited in an autosomal dominant fashion.
Approximately 50% of fAD cases have been attributed to mutations in three genes, amyloid
precursor protein (APP) (Goate et al. 1991a), presenilin-1 (PSEN1) (Sherrington et al. 1995) or
presenilin-2 (PSEN2) (Rogaev et al. 1995).
Presenilins are integral membrane proteins that are synthesized within the endoplasmic reticulum
(ER) as full-length holoproteins. In the ER, presenilins undergo proteolytic cleavage generating
N- and C-terminal fragments which remain associated. Along the secretory pathway, presenilins
associate with presenilin enhancer-2, nicastrin and anterior pharynx defective-1. Together these
proteins constitute the γ-secretase complex. This complex has proteolytic activity and is known
to cleave several type I transmembrane proteins including Notch and APP. APP proteolysis is
particularly important to AD because aberrant APP proteolysis results in the deposition of Aβ
fragments which are the primary components of senile plaques. However, while Aβ deposition
is a cellular hallmark of AD, it is remains unclear whether or not this process is the primary
cause of AD. In fact, neurodegeneration in the absence of senile plaque formation (Chui et al.
1999;Amtul et al. 2002;Dermaut et al. 2004;Raux et al. 2000) suggests that other toxic processes
in which presenilin could be involved may compromise neuronal function independent of Aβ
generation and ultimately set the stage for the onset of AD pathogenesis. In fact, AD aetiology is
believed to involve several aberrant cellular processes including protein aggregation, oxidative
stress as well as intracellular calcium deregulation.
Deregulation of intracellular calcium signalling is an early event in AD pathogenesis and
precedes any symptoms (Etcheberrigaray et al. 1998). More specifically, internal calcium stores
including the endoplasmic reticulum (ER) and Golgi apparatus have been reported to be either
under, or over-loaded, in cells expressing fAD-mutant forms of PS1 (Zatti et al. 2006;Leissring
31
et al. 2000;Smith et al. 2002;Leissring et al. 2001) or PS2 (Zatti et al. 2006;Zatti et al. 2004).
The apparent discrepancies in published results may be attributed to the use of different fAD-
presenilin mutations, different cell types (often non-neuronal) and different experimental
approaches. Thus further studies focused on understanding the role of presenilins in calcium
dysfunction are needed to resolve these inconsistencies.
Changes in cytosolic calcium concentration normally function as a second messenger system
mediating a wide range of cellular processes, many of which are relevant to AD aetiology
including learning and memory as well as cell death. Internal calcium stores play an important
role in facilitating intracellular calcium homeostasis by regulating calcium release and storage.
The ER contains two main types of calcium release channels, the ryanodine receptor (RyR) and
the inositol 1,4,5-triphosphate receptors (IP3R). Presenilins have been shown to physically
interact with both of these channels and to influence their activity (Cheung et al. 2008;Stutzmann
et al. 2007;Hayrapetyan et al. 2008;Rybalchenko et al. 2008). Presenilins have also been shown
to physically interact with a number of known transducers of calcium signalling including
calmyrin (O'Day and Myre. 2004;Morohashi et al. 2002), sorcin (Zhu et al. 2004) and calsenilin
(Leissring et al. 2001;Ahn et al. 2004). Finally, one study has suggested that presenilins
themselves may function as passive ER calcium channels (Tu et al. 2006). Despite all the
evidence linking presenilin function to intracellular calcium homeostasis, the precise
mechanisms by which presenilins regulate calcium dynamics remain unresolved.
In this study we have investigated the impact of wild type and mutant presenilin expression on
intracellular calcium dynamics in primary Drosophila cholinergic neurons. Importantly, unlike
most presenilin studies performed in Drosophila, our work focuses on presenilin function
specifically in the fly central nervous system (CNS). The genetic tractability of Drosophila
melanogaster makes this organism an ideal model to study the function of presenilin. The
Drosophila genome encodes a single presenilin gene (Psn) (Boulianne et al. 1997)
circumventing genetic redundancy. All the components of the γ-secretase complex are
conserved in flies (Adams et al. 2000) as is the proteolytic specificity and function of this
complex (Struhl and Greenwald. 1999;Ye, Lukinova, and Fortini. 1999). Since flies do not
generate Aβ peptides, Psn function can also be studied without the confounding impact of Aβ
deposition. In addition, studies performed in our laboratory as well as others has implicated Psn
function in synaptic plasticity (Knight et al. 2007), learning and memory (Knight et al. 2007;Lu
32
et al. 2007) as well as ER-stress induced calcium tail currents (Lu et al. 2007) further
demonstrating that Drosophila and mammalian Psn function is highly conserved and that
Drosophila Psn is required for processes that are affected in AD.
Here we demonstrate that Psn expression in primary Drosophila cholinergic neurons causes
deficits in internal calcium stores. Importantly, these deficits occur independent of Aβ
generation. We also describe a novel genetic, physiological and physical interaction between
Psn and calmodulin (Cam), a key regulator of intracellular calcium homeostasis. Specifically,
we show that Psn-induced deregulation of internal calcium stores can be suppressed by the loss
of a single copy of Cam. We also present evidence that Psn and Cam physically interact.
Finally, we also examined whether presenilin expression affects neuronal cell death or fly
survival. Taken together our data support a model whereby Psn plays a role in regulating
intracellular calcium levels that may be mediated by its interaction with Cam and demonstrate
that Drosophila can be used to study the link between presenilin and calcium deregulation.
2.3 Materials and Methods
Fly stocks
Flies bearing both a UAS-wild type Drosophila presenilin (UAS-PsnWT
) transgene as well as the
cut-GAL4 wing margin driver were recombined onto the same third chromosome (cut-Psn). cut-
Psn flies were then crossed at 29°C to flies bearing either a P-element insertion in the Cam gene
(Cam hypomorph) (Wang, Sullivan, and Beckingham. 2003) or a Cam null line (Camn339
)
(Heiman et al. 1996a). The genetic interaction of Cam and Psn at the wing margin was
confirmed by the chi-squared (2) 2 X 2 table method using Statistica software. For the calcium
analysis, full-length wild type UAS-PsnWT
(Guo et al. 1999) or fAD-M146V mutant transgene
(UAS-PsnfAD
) (Ye and Fortini. 1999) Drosophila Psn transgenes both on the third chromosome
were crossed at room temperature to flies bearing both a Cha-GAL4 and UAS-GFP transgene
(Salvaterra and Kitamoto. 2001). Lines bearing both the Camn339
allele as well as the UAS-
33
PsnM146V
(UAS-PsnFAD
) were generated and crossed to the Cha-GAL4 line described above to
assess the physiological interaction between Cam and Psn.
Western Analysis of Cha-GAL4 driven presenilin expression
10 male fly heads were collected and lysed in 50µL of RIPA buffer. Lysates were incubated for
15 minutes on ice, spun at 10,000 x G, supernatant collected and loaded onto a 10%
polyacrylimide gel. Gels were then transferred over-night onto PVDF membranes followed by
blocking in 2.5% each of milk, BSA, FBS, NGS and NDS in 1% Tris-buffer saline tween-20
buffer (TBST – 100 mM Tris-HCl pH 7.5, 150 mM NaCl, 0.1% tween-20) for 1 hour at room
temperature. PVDF membranes were then incubated with a rabbit polyclonal antibody raised
against the N-terminal fragment of Drosophila Psn (Guo et al. 1999) at a 1:1000 dilution in 1%
Tris-buffer saline tween-20 buffer (TBST – 100 mM Tris-HCl pH 7.5, 150 mM NaCl, 0.1%
tween-20) overnight at 4 °C. The next day membranes were washed three times followed by
incubation with an HRP-conjugated goat ant-rabbit antibody at 1:10,000 in 1% blocking solution
for one hour at room temperature followed by three washes prior to visualization using
chemiluminescence. The blots were then stripped, washed, incubated in 2.5% block as described
above and re-probed using a mouse-anti-actin antibody (Hybridoma bank, JL20) used at 1:1000
in 1% block. Membranes were then washed, incubated with an HRP conjugated goat-anti-mouse
antibody, washed again before exposing using chemiluminescence.
Fly imaging
Whole mount images of the pupal CNS were captured using Zeiss LSM 5 Pascal laser-scanning
confocal microscope using a 20X objective. Whole mount images of fly wings were generated
using the Zeiss Mirax Scan digital imaging platform.
Cell culture, calcium imaging and analysis
Cell culture, calcium imaging and analysis
34
Primary pupal CNS culture, calcium imaging and subsequent analysis was performed according
to previously published methods (Campusano et al. 2007;Sicaeros, Campusano, and O'Dowd.
2007) with the following amendments. After baseline recordings were established in four day
old primary cultures, the cultures were washed three times over a 200 second interval in HBSS
media (Campusano et al. 2007) containing zero calcium, 2 mM EGTA and 4 mM MgCl2. HBSS
media containing 5 µM ionomycin (Sigma) was then added to the cultures and the area under the
curve was calculated using Mini-Analysis (Synaptosoft) to estimate intracellular calcium
content. At least six independent cultures generated on at least three different culturing days
were analyzed for each experimental variation. From each neuronal culture, approximately
fifteen cells were selected based on positive GFP expression (indicating expression of the
cholinergic Cha-GAL4 driver). Each genotype was coded during analysis and not decoded until
all analysis was completed. Statistica software was used for all statistical analysis. Kolmogrov-
Smirnov test was used to analyze raw data distribution. Since the raw data of both the resting
calcium and ionomycin response measurements was not normally distributed the non-parametric
Kruskal-Wallace ANOVA of ranks followed by Mann-Whitney pair-wise comparisons was used
to analyze both the calcium baseline and ionomycin responses.
Binding Assay and Western Analysis
Drosophila S2 cells were maintained at room temperature in Schneider’s media supplemented
with 10% FBS. A construct containing full-length, wild type Drosophila Psn under the control
of the actin promoter was used to transfect a total of approximately 1 x 107 S2 cells using the
Cellfectin reagent (Invitrogen, 10362-010). 48 hours post transfection, microsomal cell fractions
were generated as follows: cells were washed in cold PBS followed by re-suspension in 1.35 mL
of 20 mM Tris pH 7.4 with protease inhibitors and then sheared through a 25 gauge needle with
subsequent sonication (3x30 seconds on ice). Lysates were then incubated on ice for 15 minutes
and spun at 1230 x G for 25 min at 4 °C. The supernatant was collected and spun further at
100,000 x G for 45 min at 4 °C. The microsomal pellet was re-suspended in 1 mL of 50 mM
Tris pH 7.5 plus protease inhibitors and pre-cleared with 100 µL of agarose beads (Sigma, 4B-
200) (pre-washed in 50 mM Tris pH 7.5 buffer) by rotating for 1 hr at 4 °C. The sample was
then split in half and incubated with either 25 µL of Cam-beads (Sigma, P-4385) or 25 µL of
35
beads alone (Sigma, 4B-200) for four hours at 4 °C. Beads were then washed three times with
50 mM Tris pH 7.5 buffer, allowing beads to settle by the force of gravity in between each 10
minute wash. Finally, bound proteins were eluted using 50 µL of 2 x loading buffer with DTT.
Western analysis was performed using a rabbit polyclonal antibody raised against the N-terminal
portion of the wild type Drosophila presenilin protein (Guo et al. 1999) used at 1:1000 and
incubated over-night at 4°C. The anti-calmodulin antibody (Zymed,61-8500) was used at 1:1000
over-night at 4 °C.
Cell Death Analysis
Four day old primary neuronal cultures were fixed using 4% paraformaldehyde at room
temperature in the dark. Cells were subsequently washed three times with PBS, 10 minutes each
wash. Cells were then permeabilized with PBT (PBS/0.2% triton X) for 15 minutes at room
temperature after which the cells were washed three times with PBS, five minutes in between
each wash. TMR red TUNEL reactions were performed according to manufacturers instructions
(Roche, 2 156 792) with the following amendments. The TUNEL reaction was incubated at 37
°C for 45 minutes. Cultures were then washed three times with PBS at room temperature, each
wash lasting five minutes. Nuclei were then stained with the DNA-binding Hoescht dye at 1:500
in PBS for ten minutes and washed three times with PBS, each wash lasting five minutes.
Cultures were imaged using a Zeiss Plan-Apochromat 40X/0.95 NA objective and Zeiss
Axiovert 200 inverted fluorescence microscope equipped with a Sony 3 chip CCD camera
(RGB) and a Hamamatsu Orca ER CCD camera. A total of three brains were imaged for each
genotype tested. Four different fields of view were taken from each culture and used to
determine an average for TUNEL positive and condensed Hoescht stained nuclei for each plate.
Percent of TUNEL positive cells was then quantified for either the total cell population, the
cholinergic GFP positive population or the non-GFP population. A one-way analysis of variance
(ANOVA) was used to test for statistical differences between the groups followed by Tukey
post-hoc analysis.
36
Lifespan Analysis
Male flies were segregated into ten vials housing at least ten individual flies at the beginning of
the experiment and kept at 25 °C. Vials were plugged with a sponge plug and placed
horizontally in racks so as to prevent older flies from getting stuck in the food. Death was
recorded every 3-4 days until the end of the experiment. Statistica software was used for all
statistical analysis. Kaplan-Meier was used to plot survival of each genotype. The Gehan’s
Wilcoxon test was used for two-sample comparisons of life span.
2.4 Results
2.4.1 Intracellular calcium dynamics in primary cholinergic Drosophila neurons
We wanted to investigate the effect of Psn expression on intracellular calcium dynamics in a cell
type relevant to AD and therefore chose to focus on central neurons. The Drosophila pupal CNS
is particularly amenable to culturing and was used for all calcium imaging experiments. The
cholinergic Cha-GAL4 driver was chosen to drive expression of either wild type (PsnWT
) or fAD-
Psn (PsnfAD
). The specific fAD-Psn mutant used is a methionine to valine substitution at amino
acid 146, one of the most intensely studies fAD-Psn mutations. We chose Cha-GAL4 in part
because the Drosophila CNS is primarily cholinergic (Fig. 5A) but also because cholinergic cell
loss is a prominent feature in AD brains (Geula and Mesulam. 1989;Gaburjakova et al. 2001). In
addition, the Cha-GAL4 line used in our studies also contains a UAS-GFP transgene (Salvaterra
and Kitamoto. 2001) enabling us to specifically select cells expressing Psn for calcium analysis
(Fig 5A-C). Calcium dynamics were measured using the calcium binding Fura-2AM fluorescent
dye (Fig. 5D). Plotted over time Fura-2 measurements reveal a calcium trace that can be used to
determine resting cytosolic calcium levels as well as calcium movement from internal stores into
the cytoplasm (Fig. 5D). Since Psn has been shown to impact the calcium content of more than
one internal store (Zatti et al. 2006) we chose to measure the release of calcium from all internal
37
Figure 5. Evaluation of calcium content in internal stores in Drosophila primary neuronal
cultures
A) Whole mount Drosophila CNS 56-72 hour post pupariation expressing GFP in cholinergic
neurons driven by the Cha-GAL4 driver (100 µm scale bar). B) Field of dissociated primary
culture showing cholinergic pupal neurons expressing GFP (50 m scale bar). C) DIC image of
primary pupal cultures overlayed with an inverted image showing GFP fluorescent signal in
cholinergic neurons (50 m scale bar). D) Pseudo-colored 340/380 ratio representation of
intracellular calcium concentration in Fura-2 loaded neurons. Cells showing a red hue indicate
high intracellular calcium levels while cell bodies in blue represent low calcium levels (50 m
scale bar). E) Over time Fura-2 measurements can be translated into estimates of real calcium
levels. Trace here illustrates results obtained in a typical experiment, where after recording in
basal resting conditions, cells are exposed to ionomycin in absence of external calcium.
38
stores using the calcium ionophore ionomycin in zero extracellular calcium conditions (Fig 5E).
Previous studies have shown that ionomycin treatment depletes intracellular calcium stores in
Drosophila cell (Yeromin et al. 2004). Ionomycin treatment causes a rapid increase in cytosolic
calcium concentration during the initial release of calcium from internal stores as can be seen in
Figure 5E. This increased cytosolic calcium concentration gradually returns to baseline as
internal stores are emptied and the calcium is extruded from the cell.
2.4.2 Psn-induces deficits in intracellular calcium stores content
To investigate the role of Psn in intracellular calcium dynamics we expressed wild type or fAD-
mutant Psn in Drosophila cholinergic neurons. Transgene expression was confirmed using
western analysis (Fig. 6). Analysis of basal calcium recordings revealed no significant
differences between neurons expressing wild type (Cha;PsnWT
, median = 90 nM Ca2+
) or mutant
Psn (Cha;PsnFAD
, median = 80 nM Ca2+
) relative to controls (Cha, median = 80 nM Ca2+
)
suggesting that Psn expression is not overtly toxic to these cells (Fig. 7A). Next, we wanted to
determine whether Psn expression could impact internal calcium stores. We measured
intracellular calcium store content using the calcium ionophore ionomycin in a recording
solution that does not contain calcium (zero extracellular calcium) (Fig. 7B). This analysis
revealed that expression of both wild type ( Cha;PsnWT
P < 0.01, median = 3597.780 nM•s) as
well as mutant (Cha;PsnFAD
P < 0.01, median = 3926.490 nM•s) Psn caused a significant
decrease in internal calcium store content relative to controls (Cha, median = 5438.02 nM•s).
There was no significant difference in internal calcium store between neurons expressing wild
type or fAD-Psn.
39
Figure 6. Expression of wild type and FAD-mutant Psn.
Western analysis of lysates generated from adult fly heads reveal lower expression levels of wild
type Psn holoprotein (single asterisk in Cha;PsnWT
lane) relative to that of the FAD-mutant Psn
(single asterisk in Cha;PsnFAD
lane). Loss of a single Cam allele does not appear to alter the
level of Psn holoprotein in flies expressing FAD-mutant Psn (single asterisk in
Cha/Camnull
;PsnFAD
lane). The N-terminal Psn fragment levels appears normal in all lanes
relative to the Cha-GAL4 control (double asterisk in Cha relative to all other lanes). Actin
protein levels serve as loading control (solid black arrow head in all lanes).
40
Figure 7. Calcium content in internal calcium stores is affected in cholinergic neurons
expressing Psn.
A) Expression of wild type or FAD-mutant presenilin protein in cholinergic neurons (Cha;PsnWT
and Cha;PsnFAD
, respectively), does not affect basal calcium levels compared to control strain
(Cha). B) Expression of Cha;PsnWT
or Cha;PsnFAD
results in decrements in calcium content
detected in intracellular reservoirs in these neurons. Data are represented as modified box-
whisker plots with the median indicated by the smaller white box and the 25 and 75 percent
quartiles indicated by the lower and upper margins of the large grey boxes respectively. Each
box represents recordings from cultures generated from at least six independent brains, cultured
on at least three independent culturing days. The area under the response curve was calculated
from baseline to the point of 50 percent return to baseline in Neurons treated with 5µM of
ionomycin
41
2.4.3 Loss-of-function mutations in Cam suppress Psn-induced wing scalloping
Recently, we reported that several known regulators of calcium homeostasis suppressed Psn-
induced phenotypes (Van de Hoef et al, Genesis, in press). Briefly, loss-of-function alleles
generated by P-element insertions in the genes encoding the Ryanodine receptor, cyclic-AMP
response element binding protein A, calcium binding protein as well as calmodulin (Cam)
suppressed the penetrance of either a wing scalloping or thoracic bristle phenotype induced by
Psn expression. Psn has previously been shown to physically interact with, and impact the
activity of, the RyR in vertebrates. Cam is a calcium signal transducer that activates various
enzymes (reviewed in (Stull. 2001)) and modulates the activity of various ion channels,
including the RyR (Balshaw, Yamaguchi, and Meissner. 2002) and IP3R (Cheung et al. 2008).
To date, an interaction between Psn and Cam has not been described however, it could represent
an important mechanism for regulating intracellular calcium stores.
To confirm that Psn and Cam genetically interact we generated a recombinant transgenic line
that carried both a wing margin-GAL4 driver (cut-GAL4) as well as a UAS-wild type Psn
transgene. Overexpression of Psn at the wing margin gave rise to a wing scalloping phenotype
with 58% penetrance (cut-Psn Fig. 8A & C). Of note, others have shown that loss of Psn
function also results in wing scalloping (Struhl and Greenwald. 1999). Overexpression of Psn in
Drosophila is believed to give rise to dominant negative effects since overexpression
phenocopies Psn loss-of-function (Ye and Fortini. 1999). Flies bearing either a P-element
insertion in Cam (characterized elsewhere as a hypomorphic Cam allele) (Wang, Sullivan, and
Beckingham. 2003) or an independent imprecise excision in Cam (Camnull
) (Heiman et al.
1996b), which was not used in the original screen, were crossed to cut-Psn recombinant flies.
Both the Cam hypomorph (33% penetrance, 2: P < 0.05) as well as the Cam null (24%
penetrance, 2: P < 0.05) significantly suppressed the penetrance of the Psn-induced wing
scalloping phenotype (cut-Psn penetrance = 58%, Fig. 8B & C) thereby confirming that Psn and
Cam genetically interact.
42
Figure 8. Psn-induced wing notching is suppressed by loss-of-function mutations in Cam.
A) Overexpression of wild type Psn under the control of cut-GAL4 induces a wing notching
phenotype in flies (500 µm scale bar). B) The loss of a single Cam allele suppresses the Psn-
induced wing phenotype (500 µm scale bar). C) Quantification of the penetrance of Psn-induced
wing notching phenotype and the suppression of this phenotype by two loss-of-function
mutations in Cam (Cam null) and a Cam hypomorph). Penetrance was scored based on the
presence of at least one wing margin notch. Asterisks denote significant differences in expected
penetrance relative to the original cut-Psn recombinant as determined by the 2 test.
43
2.4.4 Cam suppresses Psn-induced deficits in intracellular calcium stores
content
Given that Cam is known to play an important role in the regulation of intracellular calcium
levels, we wanted to examine whether Psn and Cam physiologically interact to regulate internal
calcium stores. We decided to focus on mutant fAD-Psn rather than wild type Psn since
expression of both transgenes gave rise to similar deficits in internal calcium stores and because
the interaction of Cam with an fAD-mutant phenotype would be more relevant to AD aetiology.
The resting calcium levels in Cha/Camnull
trans-heterozygotes (Cha/Camnull
median = 90 nM
Ca2+
) were not significantly different to resting levels in Cha-GAL4 alone (Cha median = 80 nM
Ca2+
) (Fig. 8A). Likewise, resting calcium levels in neuronal cultures generated from flies
expressing fAD-Psn in cholinergic neurons with only a single functional Cam allele
(Cha/Camnull
;PsnfAD
median = 80 nM Ca2+
), also appeared normal relative to Cha-GAL4 controls
(Fig. 9A). We then treated these cells with ionomycin to measure internal calcium stores.
Importantly, loss of a single Cam allele alone did not alter intracellular calcium stores relative to
controls (Fig. 9B). However, as can be seen in Figure 9B, the loss of a single Cam allele
suppressed (Cha/Camnull
;PsnfAD
, median = 5322 nM•s, P=0.01) the Psn-induced calcium store
decrements otherwise observed in neurons expressing fAD-Psn with two functional copies of
Cam (Cha;PsnfAD
, median = 3926 nM•s). There were no significant differences in ionomycin-
induced calcium release between Cha/Camnull
;PsnfAD
and Cha-GAL4 alone.
44
Figure 9. Psn-induced effects on intracellular calcium stores is suppressed by a loss-of-
function mutation in Cam mutations.
A) Resting calcium levels are not affected by the loss of a single Cam allele either in a wildtype
(Cha/Camnull
) or in a Psn mutant background (Cha/Camnull
;PsnfAD
), compared to their respective
controls (Cha and Cha;PsnFAD
, respectively). B) Loss of a single allele in Cam suppresses Psn-
induced deficits in calcium stores content (Cha/Camnull
;PsnfAD
compared to Cha;PsnfAD
, P=0.01).
Neurons were treated with ionomycin and the area under the curve was calculated from baseline
to the point of 50 percent return to baseline. Data are represented by modified box-whisker plots
as described in Figure 2. Light grey boxes identify groups that are statistically different from the
ChaGAL4 control. Each box represents recordings from cultures generated from at least six
independent brains, cultured on at least three independent culturing days.
45
2.4.5 Psn and Cam physically interact
We then sought to determine if the ability of Cam to suppress Psn-induced deficits in
intracellular calcium stores was due to a direct versus indirect interaction between the two
proteins. Presenilins are known to physically interact with other calcium sensing proteins. For
example, mammalian presenilin 2 has been shown to bind to the EF-hand motif of sorcin (Pack-
Chung et al. 2000). Since Cam contains four EF-hand motifs we reasoned that Psn may
physically interact with Cam as well. Cam is highly conserved among species (Fig. 10A-C).
This conservation has enabled us to take advantage of the commercial availability of agarose
beads covalently bound to bovine Cam to perform binding assays. Lysates were generated from
Drosophila S2 cells transfected with full-length wild type Psn. Equal amounts of protein were
incubated with either beads alone or beads covalently bound to Cam. Normally, full-length Psn
is rapidly processed into N- and C-terminal fragments hence full-length Psn is rarely observed.
However, since Psn processing is dependent on limiting factors, when Psn is overexpressed the
full-length holoprotein accumulates (Fig. 10D black arrow head) while the N- and C-terminal
fragment levels remain unaltered. The two N-terminal Psn bands (Fig. 10D asterisk) correspond
to two different isoforms resulting from alternative splicing. Western analysis revealed that
indeed, full-length Psn, and to a lesser extent, the cleaved N-terminal fragment bound to Cam-
beads but not to beads alone (Fig. 10D)
46
Figure 10. Cam binds to full-length as well as the N-terminal fragment of Psn.
A) Sequence alignment of Bovine as well as Drosophila Cam demonstrates the high degree of
conservation of Cam between the two species. Identical sequence homology (asterisk), similar
amino acids (light grey box) and the one different amino acid (black box) are highlighted. B)
Antibodies raised against bovine Cam recognize mammalian Cam (arrow head) in COS cells as
well as Drosophila Cam in S2 cells, highlighting the high degree of conservation of this protein.
C) The specificity of the Cam antibody was confirmed using lysates generated from Camn339
homozygous null larvae (Camnull
) where the band corresponding to Cam is absent, as opposed
lysates generated from Cam heterozygotes (Camnull
/+) and mammalian control COS cells (COS).
D) Equal amounts of protein were incubated with either Cam-beads or beads alone. Full length
(black arrow head) and the N-terminal fragment of Psn in a short (asterisk) or long (pound)
exposure were pulled down by Cam-beads but not beads alone. Two N-terminal Psn bands
indicated in the gels correspond to two isoforms generated by alternative splicing of Psn.
47
2.4.6 Primary neurons expressing wild type Psn have lower incidence of
apoptosis
To determine if the Psn-induced changes in intracellular calcium stores promote the onset of
other cellular defects associated with AD we looked at the incidence of programmed cell death, a
cellular process that is dependent on intracellular calcium signalling. DNA fragmentation in
apoptotic nuclei was assayed using TUNEL staining and nuclear condensationwas confirmed
using Hoescht staining (Fig. 11A). First, we looked at apoptosis in the total neuronal population.
The incidence of TUNEL positive nuclei in neurons expressing wild type Psn was 20.1% which
was significantly lower relative to the 37.9% recorded in control Cha-GAL4 neurons (P=0.02)
(Fig. 11B). There was no significant differences in the incidence of TUNEL positive nuclei in
neurons expressing fAD-Psn relative to Cha-GAL4 or neurons expressing wild type Psn.
Quantification of condensed nuclei revealed similar trends however, statistical analysis did not
reveal significant differences between the three genotypes (Fig 11 C). We then quantified the
incidence of TUNEL positive nuclei in GFP positive, cholinergic neurons (Fig 12 A). Again,
expression of wild type Psn (Cha;PsnWT
= 0.5% TUNEL positive nuclei) resulted in a
significantly lower incidence of TUNEL positive nuclei relative to Cha-GAL4 controls (Cha =
2.3% TUNEL positive nuclei, P=0.04). The incidence of TUNEL positive nuclei in neurons
expressing fAD-Psn (Cha;PsnFAD
= 1.9% TUNEL positive nuclei) was not different to the
incidence observed in Cha-GAL4 or wild type Psn expressing neurons. We reasoned that if there
was less apoptosis occurring in neurons expressing wild type Psn there should also be more GFP
positive neurons surviving in these cultures. Indeed, this is what we observed (Fig 12 B). We
found that 47.7% of the neuronal population was GFP positive in neuronal cultures expressing
wild type Psn, which was significantly higher than the 32.7% observed in Cha-GAL4 control
cultures (P=0.01). Of the neuronal population expressing fAD-Psn, 38.9% was GFP positive
which was not significantly different to either Cha-GAL4 or wild type Psn expressing neuronal
cultures. There were no significant differences in TUNEL reactivity between any of the
genotypes in the non-GFP cell population (data not shown).
48
Figure 11. Total incidence of apoptosis in cultures expressing wild type or FAD-mutant
Psn.
(A) Cells expressing wild type or FAD-mutant Psn driven by the cholinergic Cha-GAL4 driver
are marked by GFP. Red TUNEL positive nuclei depict DNA-nicking. Condensed apoptotic
nuclei are visualized using Hoescht staining. (B) Expression of wild type Psn (Cha;PsnWT
)
significantly decreases the incidence of apoptosis relative to Cha-GAL4 controls (P = 0.02) but
not relative to cells expressing FAD-mutant Psn (Cha;PsnFAD
). (C) Although the same general
trend is observed using Hoescht staining to quantify apoptotic condensed nuclei there were no
significant differences between the three genotypes.
49
Figure 12. Expression of wild type Psn in cholinergic neurons facilitates cell-autonomous
cell survival.
(A) Expression of wild type Psn in cholinergic neurons (Cha;PsnWT
) results in a lower incidence
of TUNEL positive nuclei in cholinergic cells (P = 0.04). The incidence of TUNEL positive
nuclei in GFP positive FAD-mutant Psn (Cha;PsnFAD
) cholinergic neurons was not different to
that of Cha-GAL4 (Cha) or neurons expressing wild type Psn (Cha;PsnWT
) (B) A higher
percentage of GFP positive cholinergic neurons was observed in cultures expressing wild type
Psn relative to Cha-GAL4 cultures (Cha). No differences in the percentage of GFP positive cells
were observed between cultures expressing mutant Psn relative to control Cha-GAL4 (Cha)
neurons or those expressing wild type Psn (Cha;PsnWT
).
50
2.4.7 Cholinergic expression of fAD-mutant Psn results in shortened lifespan.
To determine whether expression of Psn in cholinergic neurons over the entire life time of a fly
could influence lifespan we performed survival analysis (Table 1 & Apendix 1). Flies
expressing wild type Psn had a mean life span of 81.1 days relative to 82.6 days for Cha-GAL4
(Fig. 13). There was no significant difference between the lifespan of flies expressing wild type
Psn and Cha-GAL4 flies. However, flies expressing wild type Psn did live longer than flies
bearing UAS-PsnWT
alone (PsnWT
alone, mean lifespan = 76.8 days, P<0.01) as well as wild type
flies (w1118
, mean lifespan = 73.2 days, P<0.01) (Fig. 13).
The mean lifespan of flies expressing fAD-Psn was 67.9 days which was significantly shorter
than the lifespan of control w1118
flies (w1118
, mean lifespan = 73.2 days, P<0.01), Cha-GAL4
flies (Cha, mean lifespan = 82.6 days, P<0.01) and flies bearing the UAS-PsnFAD
transgene alone
(PsnWT, mean lifespan = 76.8 days, P<0.01) (Fig. 14, Table 1 & Appendix 1). Next we
examined whether one loss-of-function allele in Cam could suppress the short lifespan of flies
expressing fAD-Psn. Indeed, loss of a single Cam allele significantly extended the lifespan of
flies expressing fAD-mutant Psn from 67.9 days to 83.2 days (P<0.001). However, we also
observed that loss of a single Cam allele significantly extended the lifespan of Cha-GAL4 flies
from 82.6 to 97.8 days (P<0.001) (Fig. 15, Table 1 & Apendix 1). Hence, the extension of
lifespan attributed to the loss of a single Cam allele is not specific to Psn expressing flies.
51
Table 1. Survival Analysis
The mean and median lifespan in days for each genotype analyzed plus/minus the standard error
of the mean (SEM) and the standard deviation (SD). N represents the total number of flies
analyzed.
52
Figure 13. Survival analysis of flies expressing wild type Psn in cholinergic neurons.
Flies expressing wild type Psn (Cha;PsnWT
) live longer then wild type flies (w1118
) as well as
flies bearing only the UAS-PsnWT
transgene but not longer then Cha-GAL4 (Cha) control flies.
53
Figure 14. Survival analysis of flies expressing FAD-Psn in cholinergic neurons.
Flies expressing mutant Psn (Cha;PsnfAD
) have a shorter lifespan relative to wild type flies
(w1118
), Cha-GAL4 (Cha) control flies, as well as flies bearing only the UAS-PsnWT
transgene.
54
Figure 15. Survival analysis for flies expressing FAD-Psn in cholinergic neurons with or
without the loss of a single Cam allele.
Flies expressing FAD-mutant Psn (Cha;PsnfAD
) have a shorter lifespan relative to wild type flies
(w1118
), Cha-GAL4 (Cha) control flies, as well as flies bearing only the UAS-PsnWT
transgene.
Loss of a single Cam allele extends the lifespan of both of flies expressing FAD-mutant Psn
(Cam/Cha;PsnfAD
) as well as Cha-GAL4 (Cha) flies.
55
2.5 Discusion
Although the specific cellular mechanisms remain uncertain, increasing evidence suggests that
presenilin plays an important role in regulating intracellular calcium dynamics. We have
investigated Psn function in the context of intracellular calcium homeostasis using the
Drosophila CNS as a model system. Our data demonstrates that expression of wild type or fAD-
mutant Psn in Drosophila cholinergic neurons results in cell-autonomous deficits in calcium
stores. Decrements in calcium stores attributed to wild type or mutant Psn expression have been
documented in human as well as mouse models (Zatti et al. 2006;Cheung et al. 2008;Zatti et al.
2004;Cai et al. 2006). To date, most studies in Drosophila have focused on the role of Psn in
Notch signalling during development. Our data clearly demonstrate that Drosophila can be used
as a model to study additional functions of Psn and specifically, its role in the regulation of
intracellular calcium dynamics. Importantly, unlike mammals Drosophila does not generate Aβ.
Hence, in our model system, any effects on internal calcium stores can be attributed entirely to
presenilin and is not confounded by the production of Aβ peptides.
Interestingly, we found that expression of either wild type or fAD-Psn gave rise to deficits in
intracellular calcium stores. We believe that this is due to a loss of Psn function since previous
studies have shown expression of Psn in Drosophila gives rise to dominant negative effects (Ye
and Fortini. 1999). The mechanistic nature of the dominant negative effect is not known but it
may involve negative repercussions of holoprotein accumulation within the ER or competition
for limiting factors required to generate functional γ-secretase complexes. Moreover, there is
mounting evidence suggesting that loss of Psn function is responsible for some aspects of fAD
pathogenesis. For example, conditional knock-out of both presenilin 1/2 phenocopies AD-like
symptoms including, learning and memory impairments as well as progressive
neurodegeneration (Saura et al. 2004). Notably, these Psn-induced phenotypes were observed in
the absence of Aβ deposition. Another group has also demonstrated memory loss and
degeneration associated with loss of both presenilins in the mouse brain, once again, in the
absence of A generation (Chen et al. 2008). In accordance to these findings, we have
previously shown that loss of Psn function in Drosophila results in defects in synaptic plasticity
56
and learning (Knight et al. 2007). In addition, in both Drosophila (Seidner et al. 2006) and C.
elegans (Baumeister et al. 1997;Levitan et al. 1996), fAD-mutations in Psn fail to completely
rescue loss of wild type Psn function. Since decrements in internal calcium stores have been
documented in mammalian PS1/PS2 null cells (Kasri et al. 2006) our data are consistent with a
dominant negative effect arising from Psn overexpression.
A great deal of effort has been made in several model systems to resolve how presenilin function
impacts intracellular calcium dynamics. The results of one study suggested that wild type, but
not fAD-presenilin, exhibits passive calcium channel activity (Tu et al. 2006). Since our results
indicate that both wild type and fAD-mutant Psn cause decrements in intracellular calcium stores
our data do not support a putative passive calcium channel function for wild type Psn (Tu et al.
2006). Presenilins are not known to bind calcium directly hence their influence on intracellular
calcium signalling may be mediated by interactions with calcium binding proteins and indeed,
presenilins have been shown to bind several calcium binding proteins (Pack-Chung et al.
2000;Stabler et al. 1999;Buxbaum et al. 1998). Using two independent loss-of-function alleles in
Cam we have confirmed that loss-of-function mutations in Cam suppress a Psn-induced wing
phenotype. The mechanism for this suppression may also involve intracellular calcium stores.
Wing scalloping is a classic Notch loss-of-function phenotype. Although we have no direct
evidence to suggest that Psn-induced wing notching is attributed to disruptions of intracellular
calcium at the wing margin, it is known that Notch proteolysis and activity is impacted by
changes in internal calcium stores. For example, loss-of-function mutations in the Drosophila
calcium-ATPase gene Ca-P60A have been shown to cause aberrant Notch trafficking and
secretion due to alterations in internal calcium stores (Periz and Fortini. 1999). Hence, it is
conceivable that Psn-induced deregulation of internal stores is responsible for the observed
Notch phenotypes.
Since Psn has been linked to calcium deregulation and Cam is an important player in
intracellular calcium homeostasis we further investigated the genetic interaction between Psn
and Cam in a cellular context relevant to AD. Using primary Drosophila cholinergic neurons we
found that loss of a single functional Cam allele suppressed calcium store deficits otherwise
observed with the overexpression of fAD-Psn. Furthermore, we showed that Cam physically
interacts with both full-length as well as the N-terminal fragment of Psn, albeit to a lesser extent
relative to the holoprotein. A physical interaction between Cam and Psn has previously been
57
postulated using a Calmodulin Target Database, which identified putative Cam binding sites in
presenilin 1 and 2 (O'Day and Myre. 2004). In fact, this database identified putative Cam
binding sites in all of the components of the γ-secretase complex (O'Day and Myre. 2004). Since
Cam binds both full-length as well as the N-terminal fragments of Psn, Cam may play a role in
regulating Psn proteolysis or protein stability. Cam activity has previously been implicated in
regulating the stability and proteolysis of other integral membrane proteins (Li, Joyal, and Sacks.
2001). The ability of Cam to regulate Psn protein levels could explain how partial loss of Cam
activity can suppress both Psn-induced wing notching and calcium-deficits.
AD-associated neurodegeneration ultimately involves neuronal cell death including the induction
of programmed cell death. Under-loaded calcium stores have been implicated as both pro-
apoptosis (Nguyen, Wang, and Perry. 2002) as well as anti-apoptotic (Scorrano et al. 2003)
triggers. Under-loaded calcium stores are associated with ER stress, which can trigger several
signalling pathways that lead to apoptosis. One of these ER stress signalling cascades involves
the induction of a member of the C/EBP family of bZIP transcription factors referred to as
CHOP (GADD153). CHOP has been shown to promote apoptosis by inhibiting the pro-survival
activity of the Bcl-2 protein (McCullough et al. 2001). Conversely, mouse embryonic fibroblasts
deficient for the pro-apoptotic BAX and BAK proteins exhibit depleted ER stores concomitant
with a decreased incidence of apoptosis (Scorrano et al. 2003). How depletion of internal stores
can coincide with both pro- or anti-apoptotic events is uncertain. Our data demonstrate that
overexpression of Psn in the Drosophila CNS results in enhanced neuronal survival.
Furthermore, since the enhanced survival is only observed in cultures where wild type Psn is
expressed we cannot attribute the enhanced survival to deficits in intracellular calcium stores.
Evidence from both vertebrate and invertebrate model systems suggests that Psn function
involves both pro-apoptotic as well as anti-apoptotic activity. For example, in mammals wild
type but not fAD-presenilin 1 has also been shown to inhibit apoptosis by activating PI3K/Akt
signalling (Baki et al. 2004). Conversely, expression of wild type and to a lesser extent fAD-Psn
in the Drosophila eye results in the induction of apoptosis (Wu, Lu, and Xu. 2001). Elevated
levels of apoptosis are also a feature of Psn loss-of-function in both the fly (Ye and Fortini.
1999) and mammalian (Saura et al. 2004;Chen et al. 2008) CNS arguing that presenilin can act
as a pro-survival factor. Ultimately, Psn function with respect to cell survival and cell death is
likely pleiotropic and context dependent. Our analysis has revealed that expression of wild type
58
but not mutant Psn enhanced cell survival relative to control Cha-GAL4 central neurons.
Furthermore, our data suggests that the enhanced survival of neurons expressing wild type Psn is
cell autonomous since lower levels of apoptosis were observed in GFP positive neurons
expressing wild type Psn concomitant with a larger pool of surviving GFP positive cholinergic
neurons relative to Cha-GAL4. Hence, our data support a pro-survival role for Psn activity in the
fly CNS.
Several reasons may explain why no significant differences in cell survival were observed
between neurons expressing wild type Psn and fAD-mutant Psn. First, fAD-mutant Psn alleles
are believed to be partial loss-of-function alleles hence fAD-Psn may retain sufficient levels of
wild type function to exert some pro-survival activity, at least during the specific developmental
stage from which the neuronal cultures are generated. Second, the expression level of wild type
Psn is lower than that of fAD-Psn. Hence, if both proteins were expressed at equal levels a
difference in survival might be observed between the two genotypes. Third, in the mammalian
system fAD-Psn knock-in mice are asymptomatic until exposed to toxic insult (Guo et al. 1999).
Perhaps exposure of fAD-mutant Psn expressing cholinergic fly neurons to toxic stress is also
necessary to induce apoptosis. Finally, fAD-mutant Psn toxicity may only become observable
with age.
To address what impact expression of wild type or fAD-Psn has on fly lifespan we performed
survival analysis. The mean lifespan of flies expressing wild type Psn was in between the mean
lifespan of Cha-GAL4 flies and flies bearing the UAS-PsnWT
transgene alone. Hence, the pro-
survival capacity of wild type Psn is insufficient to extend lifespan. The lifespan of flies
expressing fAD-mutant Psn in cholinergic neurons, on the other hand, was shortened relative to
all controls. Thus, like AD patients, flies expressing fAD-mutant Psn die prematurely. Future
studies will have to address whether these fly die due to progressive neuronal degeneration and
whether these changes coincide with alterations in intracellular calcium in the aging fly CNS.
Although loss of a single Cam allele did indeed suppress the shortened lifespan of flies
expressing fAD-Psn it did so non-specifically since the lifespan of Cha-GAL4 flies was also
extended by the loss of a single Cam allele. Cam activity is known to change during aging in
both mammals (Hoskins and Ho. 1986) as well as flies (Massie et al. 1989) however, there is no
clear evidence that could explain how loss of Cam function could extend lifespan. Ultimately,
59
these studies should be repeated once all the various transgenic and mutants fly stocks are
backcrossed onto a common genetic background in order to reduce the genetic variability.
Since Cam activity is known to play a role during learning and memory, apoptosis as well as tau
phosphorylation the interaction between Psn and Cam may be very relevant to AD pathogenesis.
Only by characterizing how normal and aberrant Psn activity impact calcium homeostasis can
we begin to resolve how this cellular process contributes to AD pathogenesis. Future studies will
need to address whether presenilins’ effect on intracellular calcium channel activity is mediated
through Cam and what if any impact this interaction has on progressive neuronal cell death in
adult flies. Deciphering the significance of the molecular and neurophysiological interactions
between Psn, Cam, IP3R and Ryr in flies would be greatly facilitated by the fact that unlike
mammals each of these proteins is encode by a single gene in Drosophila. Ultimately, only by
characterizing how normal and aberrant Psn activity impacts calcium homeostasis can we begin
to resolve how this cellular process contributes to AD pathogenesis.
60
3 CHAPTER 3
Expression of human fALS–associated SOD1 in Drosophila induces progressive activity
deficits
Kinga Michno, Shirley Liu, Joel Levine and Gabrielle L. Boulianne
Statement of contributions:
Shirley Liu generated the mouse spinal lysates and provided the antibodies used in the
immunoprecipitation analysis. The adult fly activity analysis was performed in Joel Levine’s
laboratory. Commercially obtained reagents are indicated in the materials and methods. Kinga
Michno generated all other reagents and performed all other experiments and analysis as
indicated in the materials and methods. The activity analysis was performed under the co-
supervision of Joel Levine and Gabrielle L. Boulianne. All other work was performed under the
supervision of Gabrielle L. Boulianne.
61
3.1 Abstract
Amyotrophic lateral Sclerosis (ALS) is a fatal neurodegenerative disease that specifically targets
motor neurons and causes progressive paralysis. Currently, there are no effective treatments for
ALS and as a result most patients die 3-5 years after the onset of symptoms. Most ALS cases are
sporadic (sALS) however, approximately 10% of cases are familial (fALS). Since, sporadic and
familial cases are clinically indistinguishable much effort has been made to understand the
normal and aberrant activity of genes associated with fALS. For example, dominant gain-of-
function mutations in the superoxide dismutase 1 gene have been linked to fALS. fALS-SOD1
associated toxicity has been shown to involve oxidative stress as well as protein aggregation
concomitant with progressive decline in motor ability. In this study, we demonstrate that
ubiquitous expression of fALS but not wild type human SOD1 in Drosophila gives rise to
progressive locomotory deficits in adult flies. This Drosophila model of fALS-SOD1 associated
locomotory deficits can now be used to further characterize the cellular and molecular basis of
the locomotory defects and to search for novel therapeutic interventions.
62
3.2 Introduction
Amyotrophic lateral sclerosis (ALS) is a progressive and ultimately fatal neurodegenerative
disease that selectively targets motor neurons. Most ALS cases occur sporadically (sALS) but
10% are inherited and are specifically referred to as familial ALS (fALS) (Majoor-Krakauer,
Willems, and Hofman. 2003). Histopathologically, ALS is characterized by the presence of
ubiqutinated aggregates of misfolded proteins, referred to as inclusion bodies. These inclusion
bodies form primarily in motor neurons but can also be found in glial cells (Jonsson et al. 2004).
ALS patients succumb to progressive paralysis due to motor neuron death but are spared from
any apparent cognitive decline. Since there is currently no cure for ALS, identifying the unique
molecular events that result in selective motor neuron death has been the subject of intense
research. Such studies have implicated several gene mutations and consequently, several
cytotoxic processes in promoting motor neuron degeneration, including protein aggregation
(Watanabe et al. 2001;Rakhit et al. 2007;Johnston et al. 2000a) as well as oxidative stress (Peled-
Kamar et al. 1997;Yim et al. 1996). Understanding the pathogenic role of fALS-associated
genes is unquestionably valuable, since sALS and fALS are clinically indistinguishable. Thus,
any insights into the pathological processes involved in fALS may also shed light into sALS
pathogenesis.
Insight into the mechanisms underlying ALS has come from studying genes implicated in fALS.
The most intensely studied fALS-associated gene is superoxide dismutase 1 (SOD1). Mutations
in SOD1 account for approximately 20% of all fALS cases and 5% of sporadic ALS (Majoor-
Krakauer, Willems, and Hofman. 2003). To date, over 100 different fALS-associated mutations
have been identified in SOD1 yet how these different mutations give rise to the same motor
neuron pathology remains unresolved. The function of SOD1 provides critical insight into one
possible mechanism involved in motor neuron degeneration. SOD1 catalyzes the dismutation of
superoxide free radicals. Thus, SOD1 functions as a critical cellular antioxidant. Furthermore,
inclusion bodies in fALS patients stain positively for SOD1, suggesting that SOD1 aggregation
may also be toxic to motor neurons (Watanabe et al. 2001). Importantly, given that the level of
SOD1 activity does not correlate with disease severity, it is unlikely that loss of SOD1 function
is the primary cause of motor neuron degeneration (Ratovitski et al. 1999) .
63
To further understand fALS pathogenesis, several groups have explored normal and aberrant
SOD1 activity in mouse models. SOD1 activity appears to be dispensable in mice since SOD1-
null animals are viable and do not succumb to motor neuron disease (Reaume et al. 1996). This
corroborates the theory that loss of SOD1 enzymatic function is unlikely to be a major
contributor to the onset of motor neuron degeneration in mammals (Reaume et al. 1996).
Ubiquitous expression of human fALS-SOD1, on the other hand, does give rise to progressive
paralysis, premature death and other characteristic features of ALS (Gurney et al. 1994;Bruijn et
al. 1997;Wong et al. 1995). Importantly, expression of fALS-SOD1 exclusively in motor
neurons does not give rise to motor neuron disease suggesting that SOD1 pathogenesis involves
non-cell autonomous mechanisms (Lino, Schneider, and Caroni. 2002;Boillee et al. 2006).
Collectively, it is believed that toxic gain-of-function, rather than loss-of-function of SOD1 is
responsible for the vast majority of SOD1-associated fALS pathogenesis.
The function of SOD1 has also been explored in invertebrate model systems, including
Drosophila. Flies encode a single superoxide dismutase gene, Sod. Unlike mice, Sod is required
for viability in flies (Phillips et al. 1989). Interestingly, expression of human SOD1 exclusively
in fly motor neurons results in a significant increase in the otherwise very short lifespan of Sod
null flies (Parkes et al. 1998). This outcome is likely attributed to the positive effects of re-
introducing dismutase activity into Sod null flies (Mockett et al. 2003). Perhaps even more
interesting, in a wild type background, motor neuron expression of human SOD1 results in a
significant increase in the lifespan of flies concomitant with an increased resistance to oxidative
stress (Parkes et al. 1998). These studies suggest that although SOD1 is a ubiquitously expressed
enzyme its function is particularly critical in motor neurons, which are the precise targets of
ALS. Using the endogenous fly Sod promoter, expression of human fALS-SOD1 in flies has
been shown to induce progressive locomotory deficits, but unlike the human disease these
SOD1-induced defects were determined to be recessive (Mockett et al. 2003). More recently, a
study has described dominant, cell-autonomous phenotypes in flies expressing human wild type
as well as fALS-mutant SOD1 in motorneurons (Watson et al. 2008). Specifically, flies
expressing both wild type and fALS-SOD1 were shown to have progressive decrements in
climbing activity, concomitant with defects in synaptic transmission, but notably in the absence
of neurodegeneration (Watson et al. 2008). A limitation to this study is that several important
controls are missing from the analysis. In addition, work performed in the mouse model has
64
suggested that cell-autonomous fALS-SOD1 expression is not sufficient to induce the onset of
motor neuron disease. Ultimately, why expression of fALS-SOD1 in flies has thus far failed to
induce neurodegeneration remains unknown. One reason may be that flies simply do not live
long enough. Alternatively, perhaps using a strong ubiquitous promoter to drive the expression
of fALS-SOD1 is required to induce dominant ALS-like phentoypes.
Modeling motor neuron degeneration in invertebrate models has not been fully exploited and
warrants further investigation. An ideal invertebrate ALS model would exhibit dominant,
progressive locomotor deficits, decreased lifespan, concomitant with ALS-like neuropathology
including protein aggregation, elevated oxidative stress and progressive motor neuron loss. In
this study we show that ubiquitous expression of human fALS-SOD1, but not wild type SOD1 in
Drosophila leads to progressive decrements in adult fly locomotor activity. Importantly, the
observed activity deficits are not accompanied by fALS-SOD1 mutant protein aggregation
suggesting that other toxic mechanisms may be responsible for the progressive decline
locomotory deficits.
3.3 Methods and materials
Fly stocks
The generation of the UAS-wild type human SOD1 transgenes (HS1 and HS2) is described
elsewhere (Elia et al. 1999). Briefly, both the HS1 and HS2 transgenes are inserted on the
second chromosome. UAS-human fALS-SOD1 transgenes bearing a glycine to alanine
substitution at position 93 were inserted onto either the X chromosome (fALS1) or the third
chromosome (fALS2). Ubiquitous expression was accomplished using daughterless-GAL4 (da-
GAL4) to drive UAS-transgene expression. Wild type (w1118
) flies were used as controls. Each
of these stocks were backcrossed eight generations onto a common Canton S genetic background
to alleviate the confounding effects of genetic variability.
65
Adult fly locomotor behaviour
Fly locomotor behaviour was measured in TriKinetics locomotor activity monitors in the
laboratory of Joel Levine at the University of Toronto at Mississagua. A single 42-45 day old
male fly was placed within a narrow glass vial (5 mm in diameter and 65 mm in length)
containing food at one end and a cotton plug at the other. Vials were loaded onto a horizontally
positioned TriKinetics activity monitor housed at 25 °C and set to a 12 hour light, 12 hour dark
cycle. A minimum of 14 individual flies were loaded for each genotype at the beginning of the
experiment. As a fly walks from one end of the vial to the other its activity is monitored by a
beam of infrared light that bisects the tube perpendicular to its axis. The number of times a fly
breaks the beam of light within a 5 minute time frame is then digitally recorded by a computer.
The average activity counts for each genotype aged 42-45, 52-55 and 56-59 days was
determined. An activity count is defined as an event that is triggered by breaking a beam of light
and then recorded digitally.
Life span analysis
Male flies were segregated into ten vials housing at least twenty individual flies at the beginning
of the experiment and kept at 25 °C. Vials were plugged with a sponge plug and placed
horizontally in racks so as to prevent older flies from getting stuck in the food. Death was
recorded every 3-4 days until the end of the experiment. Statistica software was used for all
statistical analysis. Kaplan-Meier was used to plot survival of each genotype. The Gehan’s
Wilcoxon test was used for two-sample comparisons of life span.
66
Immunoprecipitation and western analysis
SOD1G93A mice (B6.Cg-Tg[SOD1-G93A]1Gur/J) were purchased from The Jackson
Laboratory. Mouse spinal lysates were generated from 4 month old mice by adding 10X volume
of RIPA buffer (50 mM Tris-HCl pH 7.5,150 mM NaCl, 1% NP-40, 0.25% Na-deoxycholate,
1mM EDTA, 0.1% SDS) + protease inhibitors (PI) to spinal tissue and homogenized for 30
strokes. Lysates were then incubated on ice for 20 min. Samples were spun at 15,000 X G at 4
°C for 15 min. Supernatant was transferred to a clean tube, protein concentration was
determined using BCA assay and frozen at -80 °C until ready to use. Fly lysates were generated
in the same manner as mouse lysates with the following amendments: 30 flies were
homogenized in 600 µL of RIPA + PI using a dounce homogenizer and protein lysates were used
immediately, not frozen. 1 mg of mouse spinal lysate and 1 mg of 30 day old fly lysate or 2 mg
of 56-59 day old fly lysates were used for immunoprecipitation (IP) and protein-A beads alone
incubation. 0.1% (1 µg) of the sample was set aside for the input lanes with the exception of 30
day old fly lysates for which 0.5% (50 µg) of total protein was used for input lanes. Protein A-
agarose beads were washed in RIPA buffer 3X, spun down at 6,000 RPM for 5 min at 4 °C in
between washes. Samples were pre-cleared with 20 µL of pre-washed protein A-agarose for 1hr
at 4 °C. Lysates were then spun down for 10 min at 10,000 X G for 10 min at 4 °C.
Supernatants were transferred to a clean eppendorf tube and 10 µL of rabbit anti-human
superoxide dismutase1 exposed dimer interface antibody (SEDI) (Rakhit et al. 2007) was added
and incubated overnight at 4 °C. The next day, 20 µL of pre-washed protein-A-agarose beads
was added and incubated for 4 hr at 4 °C . The IP was then spun down at 6,000 RPM for 5 min
at 4 °C and the supernatant discarded or frozen. Beads were washed with 1 mL of RIPA 3x
spinning down at 6,000 RPM for 5 min at 4 °C in between washes. Supernatants were
completely removed using a gel-loading tip after the last wash. 12 µL of 2 X SDS sample buffer
plus 5% of β-mercaptoethanol was added and the sample was boiled for 5min. Samples were
spun at 10,000 X G for 10 min at room temperature, transferred to a clean tube with a gel-
loading tip and loaded onto a 10% SDS-page electrophoresis gel. Gels were then transferred
over-night onto PVDF membranes. The membranes were blocked in 2.5% each of milk, BSA,
FBS, NGS and NDS in 1% Tris-buffer saline Tween-20 buffer (TBST – 100 mM Tris-HCl pH
7.5, 150 mM NaCl, 0.1% tween-20) for 1 hour at room temperature. PVDF membranes were
67
then incubated with a sheep antibody raised against human SOD1 (574597,Calbiochem ) diluted
at 1:1000 dilution in TBST overnight at 4 °C. The next day membranes were washed three times
in TBST followed by incubation with an HRP-conjugated rabbit anti-sheep antibody at 1:5,000
in 1% blocking solution for one hour at room temperature followed by three washes prior to
visualization using chemiluminescence.
3.4 Results
3.4.1 Analysis of transgenic flies expressing human wild type and fALS-SOD1
To determine if expression of fALS-SOD1 in flies could give rise to dominant ALS-like
phenotypes we expressed UAS-human SOD1 in transgenic flies. Two independent UAS-human
fALS-SOD1 (fALS1 and fALS2) transgenics both bearing a glycine to alanine substitution at
position 93 and two independent UAS-human wild type SOD1 (HS1 and HS2) transgenics were
crossed to the ubiquitous daughterless-GAL4 driver (da-GAL4). We chose to ubiquitously
express SOD1 since ubiquitous expression of fALS-SOD1 in the mouse model has proven to be
successful at recapitulating ALS-like phenotypes (Gurney et al. 1994;Bruijn et al. 1997;Wong et
al. 1995). Importantly, we backcrossed all UAS-trangenic as well as the da-GAL4 line onto a
common genetic background to minimize the confounding effect of genetic variability. To
ensure that both the wild type and mutant SOD1 transgenes were expressed, western analysis
was performed. As can be seen in Figure 16, wild type SOD1 protein was more abundant
relative to the fALS-SOD1. This is consistent with what has previously been shown in the
mouse model and is attributed to the fact that fALS-SOD1 is less stable than wild type SOD1
(Gurney et al. 1994).
68
Figure 16. Expression of wild type and fALS-mutant human SOD1.
The dark arrow head points to the band corresponding to human SOD1 protein expression driven
by the da-GAL4 driver (da). Human SOD1 protein levels are lower in lysates generated from
flies expressing fALS-mutant SOD1 protein (fALS2/da and fALS1;;da) relative to wild type
SOD1 (HS2;da and HS1;da) protein levels. Relative to the single insert transgenic lines protein
levels of human fALS-SOD1 are higher in flies expressing both fALS1 and fALS2 (fALS1;;
fALS2/da). Absence of a SOD1 band in lysates generated from control w1118
flies or da-GAL4
flies confirms the specificity of the antibody for human SOD1 protein. β-tublin protein levels
serve as loading control (Hollow black arrow head for all lanes).
69
3.4.2 Ubiquitous expression of fALS-SOD1 gives rise to progressive deficits in
adult fly locomotory activity
Progressive decline in locomotor activity is a prominent feature in ALS patients as well as fALS-
SOD1-associated animal models (Gurney et al. 1994;Watson et al. 2008). To determine if
expression of fALS-SOD1 in flies could also induce progressive motor deficits we ubiquitously
expressed wild type or fALS-SOD1 using da-GAL4 and measured adult fly locomotion. Day
time locomotory activity of adult flies was recorded using TriKinetics Drosophila activity
monitors. We focused on day time activity because during the day flies are more active relative
to night and hence more activity measurements can be recorded. As a fly walked from one end
of a narrow vial to the other its activity was monitored by a beam of infrared light that bisected
the tube perpendicular to its axis. The number of times a fly broke the beam of lights within a
five minute time frame was defined as an activity count. The average activity count for each
genotype within a 12 hour light cycle was determined. Since ALS affects older adults we chose
to begin our analysis with 42-45 day old flies. Flies ubiquitously expressing HS1 had higher
locomotor activity relative to flies bearing UAS-HS1 alone, but not higher relative to da-GAL4
flies (HS1;da: mean activity counts = 509.3; HS1: mean activity counts = 363.2, P = 0.04)( da:
mean activity counts = 459.3) (Fig. 17A) (Table 2). There were no differences in the locomotor
activity in flies expressing HS2 relative to their respective controls (Fig. 17 B) (Table 2). Flies
expressing fALS-SOD1 also had normal locomotor activity relative to da-GAL4 flies as well as
flies bearing the UAS-fALS SOD1 transgene alone (fALS1;da: mean activity counts = 347.6 and
fALS2/da: activity counts = 459.3; da: activity counts = 423.8; fALS1: mean activity counts =
361.9; fALS2: mean activity counts = 449.3) (Fig. 17C & D) (Table 2). Hence, at 42-45 days of
age ubiquitous expression of wild type or fALS –SOD1 did not result in locomotory deficits.
Ten days later we measured the locomotory activity of the same flies. At 52-55 days of age the
only differences in activity detected in flies expressing wild type SOD1 was between da-GAL4
flies and the UAS-HS1 line (da: mean activity counts = 264.8; HS1: mean activity counts =
253.4, P = 0.04). Flies expressing fALS1, however, did exhibit significantly lower activity
relative to both the UAS-line alone as well as da-GAL4 flies (fALS1;;da: mean activity counts =
162.8; fALS1: mean activity counts = 239.9, P = 0.04) (fALS1;;da: mean activity counts =
162.8; da: mean activity counts = 264.8, P < 0.01 ) (Fig. 18 D) (Table 3). Although flies
70
Figure 17. Day activity levels of 42-45 day old flies expressing wild type or fALS-SOD1.
A-D) Activity counts were quantified and averaged during a 12 hour day light cycle for each
genotype. A) Flies expressing wild type SOD1 from the HS1 insertion (HS1;da) had higher
activity relative to flies bearing UAS-HS1 alone (HS1) but not relative to the da-GAL4 control
(da). B-D) There were no significant differences between flies expressing HS2, fALS1 and
fALS2 relative to each respective control. Error bars represent the SEM.
71
Table 2. Average day time activity of 42-45 day old flies ubiquitously expressing wild type
or fALS-mutant human SOD1.
Activity counts were quantified and averaged during a 12 hour day light cycle for each genotype.
(Avg. activity). The standard error (SEM), standard deviation (SD) and sample size (N) was
recorded for each genotype. da-GAL4 is the daughterless-GAL4 driver alone, uncrossed. HS1
and HS2 represent independent UAS-transgenic stocks bearing human wild type SOD1 transgene
insertion, uncrossed . HS1;da and HS2;da represent the same UAS-transgenes crossed to the da-
GAL4 driver. fALS1 and fALS2 represent two independent UAS-transgenic stocks bearing a
glycine to alanine substitution at position 93 in human fALS-SOD1, uncrossed. fALS1;;da and
fALS2/da represent the same UAS-transgenes crossed to the da-GAL4 driver
72
Figure 18. Day activity levels of 52-55 day old flies expressing wild type or fALS-SOD1.
A-D) Activity counts were quantified and averaged during a 12 hour day light cycle for each
genotype. Significant differences are denoted using an asterisk (determined using two-tailed t-
tests). A double asterisks signifies that the activity levels induced by transgene expression was
significantly different to both the da-GAL4 as well as the respective UAS-transgene alone. A)
Flies bearing the UAS-HS1 transgene control alone had lower activity levels relative to da-GAL4
but not to flies expressing the HS1 transgene (HS1;da) B) There were no significant differences
between flies expressing HS2 and its respective controls. C) Flies expressing the fALS2-SOD1
transgene (fALS2/da) had significantly lower activity relative to da-GAL4 (da) flies but not to
flies bearing the fALS2-SOD1 transgene alone (fALS2). D) Flies expressing the fALS1-SOD1
transgene (fAL12;;da) had significantly lower activity relative to both da-GAL4 (da) flies as
well as flies bearing the fALS1-SOD1 transgene alone (fALS1). Error bars represent the SEM
73
Table 3. Average day time activity of 52-55 day old flies ubiquitously expressing wild type
or fALS-mutant human SOD1.
Activity counts were quantified and averaged during a 12 hour day light cycle for each genotype.
(Avg. activity). The standard error (SEM), standard deviation (SD) and sample size (N) was
recorded for each genotype. da-GAL4 is the daughterless-GAL4 driver alone, uncrossed. HS1
and HS2 represent independent UAS-transgenic stocks bearing human wild type SOD1 transgene
insertion, uncrossed . HS1;da and HS2;da represent the same UAS-transgenes crossed to the da-
GAL4 driver. fALS1 and fALS2 represent two independent UAS-transgenic stocks bearing a
glycine to alanine substitution at position 93 in human fALS-SOD1, uncrossed. fALS1;;da and
fALS2/da represent the same UAS-transgenes crossed to the da-GAL4 driver
74
expressing fALS2 had lower activity relative to da-GAL4 flies they were not different to UAS-
fALS2 control flies (fALS2/da: mean activity counts = 187.1; da: mean activity counts = 264.8,
P = 0.02) (fALS2: mean activity counts = 219.8) (Fig. 18 C) (Table 3). Hene, at 56-59 days of
age only flies expressing fALS1 has clear locomotory deficits relative to both respective
controls.
Four days later, we again measured the locomotory activity of the same flies. By 56-59 days of
age the activity of all the flies had considerably decreased relative to the activity levels of the
same flies at day 42-55 (Fig 17 & 19) (Table 2 & 4). da-GAL4 flies also had significantly lower
locomotory activity relative to flies expressing HS1 and flies bearing UAS-HS1 alone (HS1;da:
mean activity counts = 163.1, P < 0.01) (HS1: mean activity counts = 159.1, P = 0.04).
However, there were no significant differences in activity levels between flies expression HS1
and those bearing only the UAS-HS1 transgene (HS1: mean activity counts = 159.1; HS1;da:
mean activity counts = 163.1) (Fig. 18 A). The average activity of flies expressing HS2, was not
different to flies bearing UAS-HS2 alone or da-GAL4 flies (Fig. 19 B) (Table 4). Flies
expressing fALS2, however, did exhibit significantly lower locomotor activity relative to both
da-GAL4 flies as well as to flies bearing UAS-fALS2 alone (fALS2/da: mean activity counts =
134.2; da: mean activity counts = 248.4, P < 0.01) (fALS2/da: mean activity counts = 134.2;
fALS2: mean activity counts = 2236.4, P = 0.03) (Fig. 19 C) (Table 4). Although flies
expressing fALS1 were significantly less active relative to da-GAL4 flies they were no longer
less active relative to flies bearing UAS-fALS1 alone (fALS1;;da; mean activity counts = 111.9;
da: mean activity counts = 248.4, P < 0.01) (fALS1: mean activity counts = 167.1) (Fig. 19 D)
(Table 4). Hence, only flies ubiquitously expressing fALS2 exhibited locomotory deficits as 56-
59 days of age.
75
Figure 19. Day activity levels of 56-59 day old flies expressing wild type or fALS-SOD1.
A-D) Activity counts were quantified and averaged during a 12 hour day light cycle for each
genotype. Significant differences are denoted using an asterisk (determined using two-tailed t-
tests). A double asterisks signifies that the activity levels induced by transgene expression was
significantly different to both the da-GAL4 as well as the respective UAS-transgene alone. A)
da-GAL4 flies had higher activity levels compared to both flies expressing the HS1 transgene
(HS1;da) as well as flies bearing the UAS-HS1 transgene alone (HS1). B) There were no
significant differences between flies expressing HS2 and respective controls. C) Flies expressing
the fALS2-SOD1 transgene (fALS2/da) had significantly lower activity relative to both da-GAL4
(da) flies as well as flies bearing the fALS2-SOD1 transgene alone (fALS2). D) da-GAL4 flies
had higher activity levels compared to both flies expressing the fALS1-SOD1 transgene
(fALS2/da) as well as flies bearing the fALS2-SOD1 transgene alone (fALS21). Error bars
represent the SEM.
76
Table 4. Average day time activity of 56-59 day old flies ubiquitously expressing wild type
or fALS-mutant human SOD1.
Activity counts were quantified and averaged during a 12 hour day light cycle for each genotype.
(Avg. activity). The standard error (SEM), standard deviation (SD) and sample size (N) was
recorded for each genotype. da-GAL4 is the daughterless-GAL4 driver alone, uncrossed. HS1
and HS2 represent independent UAS-transgenic stocks bearing human wild type SOD1
transgene insertion, uncrossed . HS1;da and HS2;da represent the same UAS-transgenes crossed
to the da-GAL4 driver. fALS1 and fALS2 represent two independent UAS-transgenic stocks
bearing a glycine to alanine substitution at position 93 in human fALS-SOD1, uncrossed.
fALS1;;da and fALS2/da represent the same UAS-transgenes crossed to the da-GAL4 driver
77
3.4.3 Survival analysis of flies expressing wild type or fALS-SOD1
To determine what effects expression of fALS-SOD1 had on fly lifespan we performed survival
analysis. We expressed wild type SOD or fALS-mutant SOD1 using the da-GAL4 driver. Flies
expressing fALS1 did indeed exhibit a significantly reduced lifespan relative to da-GAL4 flies,
wild type flies, as well as flies bearing the UAS-fALS1 transgene alone (fALS1;;da: mean
lifespan = 64.7 days; da: mean lifespan = 69.9 days; w1118
: mean lifespan = 73.4 days; fALS1:
mean lifespan = 74.3) (Fig 20 B) (Table 5 & Apendix 2). Expression of fALS2, on the other
hand, did not give rise to a shorter life span relative to controls (Fig. 20 A) (Table 5 & Apendix
2).
Surprisingly, expression of HS1 also resulted in a shorter lifespan relative to wild type, da-
GAL4, as well as the to flies bearing UAS-HS1 alone (HS1;da: mean lifespan = 67.5 days; da:
mean lifespan = 69.9 days; w1118
: mean lifespan = 73.4 days; HS1: mean lifespan = 76.9 days)
(Fig. 21 A) (Table 5 & Apendix 2). Expression of HS2 did not alter the lifespan of flies relative
to controls (Fig. 21 B) (Table 5 & Apendix 2).
78
Figure 20. Survival analysis of flies ubiquitously expressing fALS-mutant SOD1.
A-B) Survival plots were generated using the Kaplan-Meier method. A) Flies ubiquitously
expressing fALS2 (fALS/da) did not have a shorter life span relative to control w1118
flies, da-
GAL4 flies or flies bearing the UAS-fALS2 transgene alone. B) Flies expressing fALS1-SOD1
had a shorter lifespan relative to control w1118
flies, flies bearing only the UAS-fALS1 transgene
(fALS1) as well as to da-GAL4 flies (da).
79
Table 5. Survival Analysis
The mean and median life span in days for each genotype analyzed plus/minus the standard error
of the mean (SEM) and the standard deviation (SD) and sample size (N). da-GAL4 is the
daughterless-GAL4 driver alone, uncrossed. HS1 and HS2 represent independent UAS-
transgenic stocks bearing human wild type SOD1 transgene insertion, uncrossed . HS1;da and
HS2;da represent the same UAS-transgenes crossed to the da-GAL4 driver. fALS1 and fALS2
represent two independent UAS-transgenic stocks bearing a glycine to alanine substitution at
position 93 in human fALS-SOD1, uncrossed. fALS1;;da and fALS2/da represent the same
UAS-transgenes crossed to the da-GAL4 driver
80
Figure 21. Survival analysis of flies ubiquitously expressing wild type SOD1.
A-B) Survival plots were generated using the Kaplan-Meier method. A) Flies expressing HS1
(HS1;da) had a shorter lifespan relative to control w1118
flies, flies bearing only the UAS-HS1
transgene (HS1) as well as to da-GAL4 flies (da). B) Flies ubiquitously expressing HS2
(HS2;da) had a shorter lifespan relative to control w1118
flies (w1118
), as well to flies bearing only
the UAS-HS1 transgene (HS1) but not shorter relative to da-GAL4 flies (da).
81
3.4.4 fALS-SOD1 does not appear to form aggregates in adult flies
To determine whether expression of fALS-SOD1 in flies could give rise to aggregate formation
as it does in fALS mouse models and fALS patients we evaluated SOD1 misfolding in flies
expressing wild type or fALS-SOD1. We used an antibody specifically engineered to only
recognize misfolded SOD1 (SEDI) to determine whether misfolded fALS-SOD1 protein was
present in Drosophila lysates. First, we looked at lysates generated from 30 day old adult flies
expressing fALS-SOD1. We initially focused on flies expressing fALS1 because this line had
the earliest deficits in locomotor activity as well as the shortest lifespan relative to controls. As a
positive control for the presence for misfolded fALS-SOD1 we used lystates generated from
symptomatic transgenic mouse spinal cords expressing the same fALS-SOD1 mutant as our flies
(a glycine to alanine substitution at position 93) (hSOD1fALS
). Non-transgenic (non-Tg) mice
were used as a negative control. As can be seen in Figure 22, misfolded monomeric fALS-SOD1
was pulled down from lysates generated from fALS-SOD1 but not from non-transgenic mice
(non-Tg) spinal lysates (Fig. 24). We were unable to detect misfolded fALS-SOD1 in lysates
generated from 30 day old flies expressing fALS-SOD1. We reasoned that 30 days of age may
be too early to detect misfolded SOD1 and decided to evaluate fALS-SOD1 misfolding in older
flies. In addition, we doubled the amount of protein used to 2 mg in an effort to facilitate the
detection of potentially very small amounts of aggregates. Again, misfolded fALS-SOD1 was
successfully immunoprecipated with the SEDI antibody from mouse fALS-SOD1 transgenic
lysates (hSOD1fALS
) but not from non-transgenic lystates (Fig. 23). High molecular weight
human SOD1 complexes were visible in the mouse fALS-SOD1 input lane and these were also
pulled down by the SEDI antibody (data not shown). We were unable to detect misfolded SOD1
in lysates generated from 56-59 day old flies expressing either fALS1 or fALS2 despite the fact
that the murine and Drosophila SOD1 protein levels appear similar in respective input (Fig. 23).
However, high molecular weight bands were also observed in lysates generated from transgenic
flies expressing fALS-SOD1 (fALS1;;da and fALS2/da) (Fig. 23).
82
Figure 22. Misfolded human SOD1 is not detected in lysates generated from 30 day old
flies expressing human fALS-SOD1.
Lysates totaling 1 mg of protein generated from either non-transgenic (non-Tg) mice expressing
fALS- mutant SOD1(hSOD1fALS
) or 30 day old whole flies ubiquitously expressing the same
human fALS-SOD1mutant (a glycine to alanine substitution at position 93) (fALS1;;da) were
incubated with an antibody that recognizes misofolded human SOD1 (SEDI) or beads alone. A
total of 10 µg of protein was loaded in both the mouse input lanes. A total of 50 µg of protein
was loaded as the input for the fly lysate. An antibody that recognizes both mouse (double
asterisk) as well as human (single asterisk) SOD1 was used to probe for immunoprecipitated,
misfolded SOD1. Misfolded human SOD1 was immunoprecipitated only from spinal lysates
generated from transgenic mice expressing fALS-mutant SOD1 (hSOD1fALS
). A total of 1 µg of
purified wild type human SOD1 protein was loaded as a positive control for SOD1detection by
western analysis
83
Figure 23. Misfolded human SOD1 is not detected in lysates generated from 56-59 day old
flies expressing human fALS-SOD1.
Lysates totaling 1 mg of protein generated from either non-transgenic (non-Tg) mice expressing
fALS- mutant SOD1(hSOD1fALS
) or 2 mg of protein generated from 56-59 day old whole flies
ubiquitously expressing the same human fALS-SOD1 mutant (a glycine to alanine substitution
at position 93) (fALS2/da and fALS1;;da) were incubated with an antibody that recognizes
misofolded human SOD1 (SEDI) or beads alone. A total of 10 µg of protein was loaded for both
mouse and fly input lanes. An antibody that recognizes both mouse (double asterisk) as well as
human (single asterisk) SOD1 was used to probe for immunoprecipitated, misfolded SOD1.
Misfolded human SOD1 was immunoprecipitated only from spinal lysates generated from
transgenic mice expressing fALS-mutant SOD1 (hSOD1fALS
). High molecular weight, SOD1
positive complex were visible in fly lysates (black arrowhead). The hollow arrow head points to
a non-specific band which is also present in fly lysates not expressing human SOD1 (data not
shown).
84
Finally, we generated a multiple-insert transgenic line that expressed both fALS1 and fALS2 in
an effort to increase the abundance of mutant protein and thus facilitate our ability to detect
misfolded mutant SOD1. Western analysis confirmed that fALS-SOD1 protein is more abundant
in lysates generated from the double-insert fALS-SOD1 line (Fig. 16). Again, we used 2 mg of
total protein to immunoprecipite misfolded human SOD1 using the SEDI antibody, however we
were unable to detect misfolded fALS-SOD1 in lysates generated from flies 56-59 day old flies
expressing both fALS1 and fALS2 (Fig. 24).
3.5 Discussion
4
We have used Drosophila as a model system to study locomotory behaviour, lifespan and
aggregate formation in flies expressing either wild type or fALS-SOD1. We show that
expression of human fALS-SOD1in Drosophila gives rise to locomotory deficits, reduced
lifespan but notably, these phentoypes occur in the absence of fALS-SOD1 misfolding.
Our findings demonstrate that ubiquitous expression of fALS-SOD1 (fALS1 and fALS2) but not
wild type SOD1 (HS1 and HS2) results in progressive deficits in locomotor activity.
Decrements in fly locomotor activity can be reflective of muscle or nervous system defects,
which are the precise targets of ALS. Importantly, locomotor deficits were not detected in 42-45
day old flies indicating that like humans, flies expressing fALS-SOD1 mutant alleles initially
have normal motor ability. By 52-55 days of age, however, flies expressing fALS1-SOD1
exhibit a significant reduction in activity relative to controls. Three days later, at 56-59 days of
age, flies expressing fALS2-SOD1 also exhibit significant deficits in locomotor activity. The
fact that fALS1-SOD1 expressing flies no longer exhibit activity deficits at 56-59 days of age
may be due to the fact that as the activity of flies decreases with age, it becomes more difficult to
detect small differences in activity.
85
Figure 24. Misfolded human SOD1 is not detected in lysates generated from 56-59 day old
flies expressing two copies of human fALS-SOD1.
A total of 2 mg of lysate generated from 56-59 day old whole flies ubiquitously expressing the
human fALS-SOD1mutant from two independent insertion (fALS1;;fALS2/da) was incubated
with an antibody that recognizes misofolded human SOD1 (SEDI) or beads alone. A total of 10
µg of protein was loaded for input lanes. An antibody that recognizes human SOD1 (single
asterisk) was used to probe for immunoprecipitated, misfolded SOD1. Misfolded human SOD1
was not detected in lysates generated from flies expressing the fALS-mutant SOD1
(fALS1;;fALS2/da).
86
In our study, expression of wild type SOD1 did not have an affect on locomotor activity
indicating that the deficits we observe in flies expressing fALS-SOD1 are specific to the
expression of the mutant allele. Previous studies have described climbing deficits in flies
expressing wild type or fALS-SOD1 specifically in motor neurons (Watson et al. 2008). Since,
fALS-SOD1 induced toxicity has been shown to involve non-cell autonomous mechanisms in
promoting motor neuron cell death, ubiquitous expression of SOD1 may more faithfully
recapitulate fALS-SOD1 specific toxicity rather than motor neuron specific expression (Clement
et al. 2003). Further analysis using nervous system or muscle specific GAL4 drivers will need to
be performed to determine if the primary defect caused by fALS-SOD1 expression originates in
neurons, glia, muscle cells or a combination thereof.
Lifespan analysis of flies expressing fALS-SOD1 revealed that ubiquitous expression of the
fALS1 allele but not fALS2 resulted in a shorter lifespan. Although not detectable by western
analysis this discrepancy may be due to slightly different levels of transgene expression. More
sensitive techniques such as real-time RTPCR could determine whether this is indeed the case.
We were unable to test the lifespan of flies expressing two copies of fALS-SOD1 on account of
the fact that this line, unlike all the other fly stocks used in this study, was not backcrossed onto a
common genetic background. Surprisingly, expression of one of the wild type SOD1 alleles
(HS1) also gave rise to a shorter lifespan. Neurodegenerative toxicity as a result of wild type
SOD1 expression has been observed in a SOD1 mouse model where moderate motor neuron
loss, mild motor impairments and SOD1 aggregate formation but no premature death was
reported (Jaarsma et al. 2000). In addition, work performed by others has shown that
overexpression of wild type SOD1 can catalyze surrogate reactions that produce highly toxic
hydroxyl radicals (Peled-Kamar et al. 1997). Hydroxyl free radicals can then go on to devastate
other biological molecules including SOD1 itself. Furthermore, oxidative damage to SOD1 has
been shown to promote SOD1 aggregate formation (Rakhit et al. 2004). It is thus possible that
sustained oxidative damage to SOD1 can lead to SOD1 misfolding and aggregate formation
which ultimately may lead to toxicity and premature death in flies. In such a way, perhaps high
levels of wild type SOD1 in certain cell types, other than motor neurons, actually results in
toxicity.
87
Work performed in the mouse model has shown that very small quantities of misfolded mutant
SOD1 coincide with aggregate formation (Rakhit et al. 2004) and progressive fALS-SOD1-
induced neurodegeneration in vivo (Rakhit et al. 2007). We used an antibody specifically
engineered to recognize misfolded human SOD1 (SEDI) to determine if misfolded SOD1 could
also be detected in lysates generated from flies ubiquitously expressing human fALS-SOD1.
Using, this approach we did not detect misfolded SOD1 in fly lysates expressing fALS-SOD1
from single or multiple-insert transgenic lines. There are several possible reasons that could
explain why we do not see misfolding of human SOD1 in flies. First, perhaps the fALS-SOD1-
induced toxic mechanisms responsible for the activity deficits in flies do not involve protein
misfolding. Second, perhaps minute amounts of misfolded SOD1 are present but are below the
threshold of detection. Although we observed similar fALS-SOD1 protein levels in lysates
generated from flies and mice, the fly lysates were generated from whole flies while the mouse
lysate was exclusively generated from symptomatic mouse spines where misfolded, aggregated
SOD is expected to be most abundant. Third, perhaps the specific human fALS-SOD1 epitope
used to generate the SEDI antibody is not exposed in Drosophila. Although we cannot
mechanistically account for how this could happen we suspect that that presence of high
molecular bands in lysates generated from flies expressing human SOD1 may represent
aggregated SOD1. High molecular weight, SOD1 positive bands were also detected in the
mouse spinal lysates generated in this study as well as by others (Wang, Xu, and Borchelt. 2002)
and importantly these bands too were pulled down by the SEDI antibody, indicating they also
contain misfolded SOD1. In order to confirm whether aggregates are present but perhaps
spatially localized in flies an independent method such as immunohistochemistry (IHC) should
be used. Work performed by others has shown that expression of fALS-SOD1 in the Drosophila
eye gives rise to SOD1 positive aggregates upon cell-autonomous fALS-SOD1 expression
(Watson et al. 2008). Using an antibody specific for human SOD1 to probe for in situ aggregates
in older adult brain and ventral nerve cord would confirm whether SOD1 positive aggregates
accumulate in flies and whether aggregate formation is tissue specific, as it is in mice. If
aggregates are found by IHC it would indicate that either the amount of aggregated SOD1 is
diluted in whole fly lysate to a point that it is undetectable, even by immunoprecipitation, or that
the SEDI antibody does not recognize misfolded SOD1 in flies. If aggregates are not found it
would indicate that aggregate formation does not contribute to the locomotor activity deficits
88
observed with fALS-SOD1 expression and that alternative fALS-SOD1-induced toxicity such as
oxidative stress may be the underlying cause.
We have shown that expression of human fALS-SOD1 in Drosophila results in dominant,
progressive locomotory deficits. We can now express fALS-SOD1 in different cell types
including glial cells, various neuronal populations as well as muscle cells in various
combinations to determine the precise cellular origin of the fALS-SOD1-induced locomotor
defect. Such studies coupled with in situ analysis of aggregate formation will further our
understanding about what role aggregate formation plays in promoting ALS-like phenotypes.
Defining the specific interactions of glia, neurons and muscles during fALS-SOD1-induced
progressive motor deficits in Drosophila will help guide the course of ALS research aimed at
identifying novel treatments and ultimately a cure for ALS.
89
CHAPTER 4
4.1 Discussion
Many important insights into the molecular mechanisms underlying neurodegenerative diseases
have come from studying genes associated with inherited forms of these disorders. However,
several questions remain unanswered. For example, although genetic mutations that cause
neurodegenerative diseases are expressed throughout the lifetime of an organism, their
deleterious activities take decades to clinically manifest. In addition, the factors that determine
the progressive nature of neurodegeneration also remain largely unknown. In the proceeding
section, age-related forms of cumulative cytotoxic stress will be discussed as potential
determining factors that trigger the onset and progressive decline of neuronal function. Cellular
coping strategies in place to counteract these stressors will also be discussed with particular
emphasis on how the Drosophila model can be exploited to further our understanding of these
toxic mechanisms and how to treat them.
4.1.1 Are aggregates toxic?
Aggregate formation is a classical hallmark of most neurodegenerative diseases including AD
and ALS, yet a causative relationship between aggregate formation and the onset of pathogenesis
is still not certain. Are aggregates a cause of disease or a consequence? Are they toxic or
protective? Perhaps cells actively sequester mutant PSEN, APP, SOD1 or other cytotoxic
proteins into inactive aggregates as a means to control the aberrant activity of the soluble mutant
protein? In fact, this mechanism has been documented in several cell types including neurons.
When a cells’ capacity to degrade misfolded or aggregated proteins by conventional ubiquitin-
proteasome mechanisms has been exceeded or compromised it will expend energy to sequester
and actively transport aggregates into large, highly structured proteinacious bodies (Johnston,
90
Ward, and Kopito. 1998;Johnston et al. 2000b). What happens to these inclusion bodies is
unclear, but degradative autophagy may be involved (Keller et al. 2004). Evidence exists linking
Presenilins, SOD1 and Aβ to aggresome formation (Johnston, Ward, and Kopito. 1998;Johnston
et al. 2000b). So long as aggregates are efficiently sequestered or cleared neuronal function
likely continues uninterrupted. However, every cellular coping mechanism has a threshold and it
is likely that at a certain point cells run out of space to safely sequester aggregates, especially if
back-up degradation mechanisms themselves become overwhelmed. Since proteosomal decline
is observed during normal aging, the process of aging itself may be the single most important
determinant of how long a cell can hold-off the negative impact of aggregate accumulation
(Keller et al. 2004). Long-term accumulation of protein aggregates, in turn, may compromise
cell viability by sequestering vital proteins such as chaperones or by physically impairing
intracellular trafficking, which happens to be particularly critical to neuronal physiology.
Intriguingly, within a single day AD patients can exhibit significant fluctuations in functional
abilities, especially memory (Palop, Chin, and Mucke. 2006). Why this happens is uncertain but
it is unlikely to involve a sudden loss or gain of neurons. Instead these fluctuations may actually
represent temporary amelioration of chronic intoxication, such as the clearance of a large
aggregate clogging axonal transport. Whether or not aggregates are the primary cause of the
disease process, however, remains uncertain but is arguably unlikely.
4.1.2 What role does stress play?
Oxidative stress is an unavoidable consequence of oxidative metabolism and as we age, our
cellular defense mechanisms against oxidative stress decline. Oxidative stress is considered one
of the principle causes of progressive decline in cellular function during normal aging (Keller et
al. 2004). What’s more, oxidative damage to proteins can be a primary trigger of protein
aggregation (Rakhit et al. 2007;Rakhit et al. 2004). For example, oxidative damage to SOD1 has
been shown to promote SOD1 misfolding and aggregation (Rakhit et al. 2007;Rakhit et al.
2004). Since, fALS-SOD1 mutants are known to promote oxidative stress, this would exacerbate
oxidative damage and consequently further promote protein aggregation (Peled-Kamar et al.
1997;Yim et al. 1996). In addition, oxidative damage to proteins involved in protein
91
degradation, such as the proteasome, can lead to further accumulation of unfolded, aggregated
proteins. Hence, cumulative oxidative damage over the entire lifespan of a post-mitotic neuron
may be a primary toxic trigger for subsequent aggregate formation.
In our model system, overexpression of fALS-SOD1 was shown to give rise to dominant
locomotory deficits in the absence of detectable aggregates, suggesting that aggregate formation
may not be a primary cause of locomotory decline. However, others have successfully detected
fALS-SOD1 aggregates in flies (Watson et al. 2008). The unanswered question is, whether our
failure to detect aggregates is due to the fact that the SED1 antibody cannot detect misfolded
SOD1 aggregates in flies, or that the aggregates are below the threshold of detection.
Alternatively, it may be that an earlier primary toxic insult, such as oxidative stress, is
responsible for the observed locomotory deficits rather than aggregate formation. This
hypothesis could be tested by challenging flies with oxidative stress and assaying whether flies
expressing fALS-SOD1 are more sensitive to this stress. These experiments would specifically
address whether oxidative stress causes flies expressing fALS-SOD1 to die even earlier or
exhibit even earlier locomotory defects relative to controls. In addition, oxidative insult may
also promote SOD1 aggregation to a level where it becomes detectable. This would argue that
indeed, oxidative stress is the primary trigger of motor deficits in flies and that protein
aggregation is a secondary effect. This model could then be further subjected to several
screening methods. For example, it may be worthwhile to examine whether protein aggregation
could also be induced at earlier larval stages. If so, fALS-SOD1-GFP constructs could be
generated and aggregation visualized through the translucent body walls of larvae. These
transgenenics could then be subjected to a drug screen designed to identify chemicals that slow
down or reduce the accumulation of aggregates. Promising therapeutic compounds could then be
screened further by assaying whether they are also able to ameliorate locomotory deficits or
premature death of fALS-SOD1 expressing flies. The small size, short lifespan and relative low
costs associated with fly husbandry facilitates such large scale analyses, which can be conducted
in a relatively short time frame. Ultimately, however, the efficacy of any findings in flies would
have to be rigorously confirmed in mice before moving into human clinical trial.
Another form of stress related to protein aggregation and relevant to both AD and ALS is ER
stress. The ER is a multifunctional organelle, acting as a calcium storage sink but also as an
important site for protein synthesis, modification and folding. Not surprisingly, disruptions in
92
intracellular calcium homeostasis, and specifically depletion of ER calcium stores, have been
shown to cause ER stress (Nguyen, Wang, and Perry. 2002;Kudo et al. 2002;Verkhratsky and
Petersen. 2002). ER stress disrupts normal protein folding and leads to accumulation of unfolded
proteins. This triggers an evolutionarily conserved cellular response referred to as the unfolded
protein response (UPR). The UPR upregulates the transcription of chaperones and purges
misfolded proteins into the cytoplasm by a processes referred to as ER-associated degradation
(ERAD). Once eradicated, misfolded proteins are ubiquitnated and targeted for protesosomal
degradation. If the UPR fails to clear the back-up, ER stress can lead to apoptosis. What’s more,
fAD-mutations in presenilin have been shown to attenuate the UPR (Kudo et al. 2002).
Collectively, it is possible that in fAD brains aberrant presenilin activity facilitates amyloid
deposition by first inducing intracellular calcium deficits, which sets the stage for ER stress.
Misfolded proteins accumulate, degradation systems start to back-up to a point that the cell has
no choice but to initiate apoptosis. Since all the components of the UPR are conserved in flies it
would be interesting to determine whether the UPR is indeed being induced in our model system.
Our data did not reveal elevated apoptosis associated with fAD-Psn expression, however, pupal
neurons are relatively young and likely efficiently cope with stress, ER stress included. Our data
also suggests that Psn has a pro-survival function in the CNS. This pro-survival function may be
sufficient to counteract any pro-apoptotic signals arising from calcium-deficit induced ER stress.
However, if neurons expressing fAD-Psn were challenged with further stress, elevated levels of
apoptosis may be revealed. This would be consistent with work performed in the mouse model.
Specifically, neurons cultured from fAD-presenilin1 knock-in mice do not exhibit overt cell
death unless challenged with cellular stress (Guo et al. 1999). In future studies, pupal neuronal
cultures expressing fAD-Psn could be challenged with various forms of stress, including
oxidative stress, and apoptosis re-quantified. I hypothesize that any residual pro-survival activity
that fAD-Psn will be insufficient to deal with the additional external cellular stress. In fact,
accumulation of oxidative damage, which we know is a normal function of aging, may be the
reason why fAD patients and flies expressing fAD-Psn die prematurely. It would also be very
informative to determine the status of internal calcium stores and UPR in older flies.
Unfortunately, the adult fly CNS is not amenable to culturing. However, several labs have
pioneered protocols involving whole mount adult CNS preparations which may in the future be
adapted to measure intracellular calcium (Olsen and Wilson. 2008).
93
Mutations in SOD1 are also linked to ER stress. Recent evidence, has suggested that fALS-
SOD1 impairs ERAD, disabling the degradation of misfolded proteins and activating proteins
involved in triggering apoptosis (Nishitoh et al. 2008). This evidence links SOD1 activity with a
primary histopathalogical hallmark of ALS, ubiquitinated aggregates of misfolded proteins.
There is reason to believe that ERAD is conserved in flies since genes known to function in
ERAD coordination are conserved in flies, but await characterization (Ryoo and Steller. 2007).
Drosophila would be an ideal platform to characterize these genes and to investigate whether
mutations in the ERAD machinery can modify fALS-SOD1 or Psn-induced phenotypes. In fact,
since protein aggregation is common to so many neurodegenerative diseases any insights into
how disease-causing genes interact with the components of ER-stress machinery may pave the
way towards novel therapeutic avenues.
4.1.3 Behavioural Genetics in Drosophila
Our work has demonstrated that expression of fAD-presenilin in cholinergic neurons results in
premature death. What we don’t know is whether expression of fAD-mutant or wild type Psn
has any effect on learning and memory. Previous work performed in our lab has determined that
Psn-null larvae exhibit defects in olfactory-associative learning suggesting that indeed Psn does
play a role in learning and memory, at least during larval development (Knight et al. 2007). In
addition, there is some evidence suggesting that expression of fAD-Psn driven by the
endogenous Psn promoter may give rise to learning defects in adults. However, this study did
not include some important controls, including a wild type non-transgenic control. Hence, the
role of Psn in learning and memory in the adult fly brain warrants further investigation.
Specifically, it would be interesting to test the learning and memory abilities of adult flies
expressing wild type or fAD-mutant Psn in cholinergic neurons, especially since these neurons
are affected in AD. Using an olfactory learning paradigm where flies are trained to associate a
specific odour with an electric shock future studies could test learning and memory at various
stages of adult life. In addition, given that wild type Psn has a pro-survival effect in the
Drosophila CNS perhaps flies expressing wild type Psn will exhibit enhanced learning or
memory.
94
4.1.4 Concluding thoughts
Drosophila is an extremely versatile model system and in many ways is ideal for studying the
genetic basis of human disease. The high degree of genetic conservation coupled with low
genetic redundancy make this model particularly well suited for studying the function of disease
causing genes. Importantly, Drosophila is amenable to cellular, physiological as well as
behavioural analysis and unlike mammalian models these studies can be performed in a
relatively short time frame at minimal costs. Pioneering work in Drosophila can then be
validated in mammalian model systems and hopefully one day contribute to the development of
effective treatments against neurodegeneration.
95
References
"Canadian Study of Health and Aging: Study Methods and Prevalence of Dementia." CMAJ :
Canadian Medical Association Journal = Journal De l'Association Medicale Canadienne
150, no. 6 (Mar 15 1994): 899-913.
Adams, M. D., S. E. Celniker, R. A. Holt, C. A. Evans, J. D. Gocayne, P. G. Amanatides, S. E.
(TRUNCATED). "The Genome Sequence of Drosophila Melanogaster." Science 287, no.
5461 (Mar 24 2000): 2185-2195.
Ahn, E. Y., S. T. Lim, W. J. Cook, and J. M. McDonald. "Calmodulin Binding to the Fas Death
Domain. Regulation by Fas Activation." Journal of Biological Chemistry 279, no. 7 (Feb 13
2004): 5661-5666.
Amtul, Z., P. A. Lewis, S. Piper, R. Crook, M. Baker, K. Findlay, A. Singleton, M. Hogg, L.
Younkin, S. G. Younkin, J. Hardy, M. Hutton, B. F. Boeve, D. Tang-Wai, and T. E. Golde.
"A Presenilin 1 Mutation Associated with Familial Frontotemporal Dementia Inhibits
Gamma-Secretase Cleavage of APP and Notch." Neurobiology of Disease 9, no. 2 (Mar
2002): 269-273.
Andersen, P. M., K. B. Sims, W. W. Xin, R. Kiely, G. O'Neill, J. Ravits, E. Pioro, Y. Harati, R.
D. Brower, J. S. Levine, H. U. Heinicke, W. Seltzer, M. Boss, and R. H. Brown Jr. "Sixteen
Novel Mutations in the Cu/Zn Superoxide Dismutase Gene in Amyotrophic Lateral
Sclerosis: A Decade of Discoveries, Defects and Disputes." Amyotrophic Lateral Sclerosis
and Other Motor Neuron Disorders : Official Publication of the World Federation of
Neurology, Research Group on Motor Neuron Diseases 4, no. 2 (Jun 2003): 62-73.
Andrus, P. K., T. J. Fleck, M. E. Gurney, and E. D. Hall. "Protein Oxidative Damage in a
Transgenic Mouse Model of Familial Amyotrophic Lateral Sclerosis." Journal of
Neurochemistry 71, no. 5 (Nov 1998): 2041-2048.
Baki, L., J. Shioi, P. Wen, Z. Shao, A. Schwarzman, M. Gama-Sosa, R. Neve, and N. K.
Robakis. "PS1 Activates PI3K thus Inhibiting GSK-3 Activity and Tau
Overphosphorylation: Effects of FAD Mutations." The EMBO Journal 23, no. 13 (Jul 7
2004): 2586-2596.
Balshaw, D. M., N. Yamaguchi, and G. Meissner. "Modulation of Intracellular Calcium-Release
Channels by Calmodulin." The Journal of Membrane Biology 185, no. 1 (Jan 1 2002): 1-8.
96
Baumeister, R., U. Leimer, I. Zweckbronner, C. Jakubek, J. Grunberg, and C. Haass. "Human
Presenilin-1, but Not Familial Alzheimer's Disease (FAD) Mutants, Facilitate
Caenorhabditis Elegans Notch Signalling Independently of Proteolytic Processing." Genes
and Function 1, no. 2 (Apr 1997): 149-159.
Berezovska, O., P. Ramdya, J. Skoch, M. S. Wolfe, B. J. Bacskai, and B. T. Hyman. "Amyloid
Precursor Protein Associates with a Nicastrin-Dependent Docking Site on the Presenilin 1-
Gamma-Secretase Complex in Cells Demonstrated by Fluorescence Lifetime Imaging." The
Journal of Neuroscience : The Official Journal of the Society for Neuroscience 23, no. 11
(Jun 1 2003): 4560-4566.
Bird, T. D. "Genetic Aspects of Alzheimer Disease." Genetics in Medicine : Official Journal of
the American College of Medical Genetics 10, no. 4 (Apr 2008): 231-239.
Boehning, D., R. L. Patterson, L. Sedaghat, N. O. Glebova, T. Kurosaki, and S. H. Snyder.
"Cytochrome c Binds to Inositol (1,4,5) Trisphosphate Receptors, Amplifying Calcium-
Dependent Apoptosis." Nature Cell Biology 5, no. 12 (Dec 2003): 1051-1061.
Boillee, S., C. Vande Velde, and D. W. Cleveland. "ALS: A Disease of Motor Neurons and their
Nonneuronal Neighbors." Neuron 52, no. 1 (Oct 5 2006): 39-59.
Boillee, S., K. Yamanaka, C. S. Lobsiger, N. G. Copeland, N. A. Jenkins, G. Kassiotis, G.
Kollias, and D. W. Cleveland. "Onset and Progression in Inherited ALS Determined by
Motor Neurons and Microglia." Science (New York, N.Y.) 312, no. 5778 (Jun 2 2006): 1389-
1392.
Borchelt, D. R., P. C. Wong, M. W. Becher, C. A. Pardo, M. K. Lee, Z. S. Xu, G. Thinakaran, N.
A. Jenkins, N. G. Copeland, S. S. Sisodia, D. W. Cleveland, D. L. Price, and P. N. Hoffman.
"Axonal Transport of Mutant Superoxide Dismutase 1 and Focal Axonal Abnormalities in
the Proximal Axons of Transgenic Mice." Neurobiology of Disease 5, no. 1 (Jul 1998): 27-
35.
Boulianne, G. L., I. Livne-Bar, J. M. Humphreys, Y. Liang, C. Lin, E. Rogaev, and P. St George-
Hyslop. "Cloning and Characterization of the Drosophila Presenilin Homologue."
Neuroreport 8, no. 4 (Mar 3 1997): 1025-1029.
Brand, A. H., and N. Perrimon. "Targeted Gene Expression as a Means of Altering Cell Fates
and Generating Dominant Phenotypes." Development (Cambridge, England) 118, no. 2 (Jun
1993): 401-415.
Brooks, B. R. "El Escorial World Federation of Neurology Criteria for the Diagnosis of
Amyotrophic Lateral Sclerosis. Subcommittee on Motor Neuron Diseases/Amyotrophic
97
Lateral Sclerosis of the World Federation of Neurology Research Group on Neuromuscular
Diseases and the El Escorial "Clinical Limits of Amyotrophic Lateral Sclerosis" Workshop
Contributors." Journal of the Neurological Sciences 124 Suppl (Jul 1994): 96-107.
Bruening, W., J. Roy, B. Giasson, D. A. Figlewicz, W. E. Mushynski, and H. D. Durham. "Up-
Regulation of Protein Chaperones Preserves Viability of Cells Expressing Toxic Cu/Zn-
Superoxide Dismutase Mutants Associated with Amyotrophic Lateral Sclerosis." Journal of
Neurochemistry 72, no. 2 (Feb 1999): 693-699.
Bruijn, L. I., M. W. Becher, M. K. Lee, K. L. Anderson, N. A. Jenkins, N. G. Copeland, S. S.
Sisodia, J. D. Rothstein, D. R. Borchelt, D. L. Price, and D. W. Cleveland. "ALS-Linked
SOD1 Mutant G85R Mediates Damage to Astrocytes and Promotes Rapidly Progressive
Disease with SOD1-Containing Inclusions." Neuron 18, no. 2 (Feb 1997): 327-338.
Buxbaum, J. D., E. K. Choi, Y. Luo, C. Lilliehook, A. C. Crowley, D. E. Merriam, and W.
Wasco. "Calsenilin: A Calcium-Binding Protein that Interacts with the Presenilins and
Regulates the Levels of a Presenilin Fragment." Nature Medicine 4, no. 10 (Oct 1998):
1177-1181.
Cai, C., P. Lin, K. H. Cheung, N. Li, C. Levchook, Z. Pan, C. Ferrante, G. L. Boulianne, J. K.
Foskett, D. Danielpour, and J. Ma. "The Presenilin-2 Loop Peptide Perturbs Intracellular
Ca2+ Homeostasis and Accelerates Apoptosis." The Journal of Biological Chemistry 281,
no. 24 (Jun 16 2006): 16649-16655.
Campusano, J. M., H. Su, S. A. Jiang, B. Sicaeros, and D. K. O'Dowd. "NAChR-Mediated
Calcium Responses and Plasticity in Drosophila Kenyon Cells." Developmental
Neurobiology 67, no. 11 (Sep 15 2007): 1520-1532.
Cao, X., and T. C. Sudhof. "A Transcriptionally [Correction of Transcriptively] Active Complex
of APP with Fe65 and Histone Acetyltransferase Tip60." Science (New York, N.Y.) 293, no.
5527 (Jul 6 2001): 115-120.
Chan, S. L., M. Mayne, C. P. Holden, J. D. Geiger, and M. P. Mattson. "Presenilin-1 Mutations
Increase Levels of Ryanodine Receptors and Calcium Release in PC12 Cells and Cortical
Neurons." Journal of Biological Chemistry 275, no. 24 (Jun 16 2000): 18195-18200.
Chen, Q., A. Nakajima, S. H. Choi, X. Xiong, and Y. P. Tang. "Loss of Presenilin Function
Causes Alzheimer's Disease-Like Neurodegeneration in the Mouse." Journal of
Neuroscience Research 86, no. 7 (May 15 2008): 1615-1625.
Cheung, K. H., D. Shineman, M. Muller, C. Cardenas, L. Mei, J. Yang, T. Tomita, T. Iwatsubo,
V. M. Lee, and J. K. Foskett. "Mechanism of Ca2+ Disruption in Alzheimer's Disease by
98
Presenilin Regulation of InsP3 Receptor Channel Gating." Neuron 58, no. 6 (Jun 26 2008):
871-883.
Chui, D. H., H. Tanahashi, K. Ozawa, S. Ikeda, F. Checler, O. Ueda, H. Suzuki, W. Araki, H.
Inoue, K. Shirotani, K. Takahashi, F. Gallyas, and T. Tabira. "Transgenic Mice with
Alzheimer Presenilin 1 Mutations show Accelerated Neurodegeneration without Amyloid
Plaque Formation." Nature Medicine 5, no. 5 (May 1999): 560-564.
Clement, A. M., M. D. Nguyen, E. A. Roberts, M. L. Garcia, S. Boillee, M. Rule, A. P.
McMahon, W. Doucette, D. Siwek, R. J. Ferrante, R. H. Brown Jr, J. P. Julien, L. S.
Goldstein, and D. W. Cleveland. "Wild-Type Nonneuronal Cells Extend Survival of SOD1
Mutant Motor Neurons in ALS Mice." Science (New York, N.Y.) 302, no. 5642 (Oct 3
2003): 113-117.
Dal Canto, M. C., and M. E. Gurney. "A Low Expressor Line of Transgenic Mice Carrying a
Mutant Human Cu,Zn Superoxide Dismutase (SOD1) Gene Develops Pathological Changes
that most Closely Resemble those in Human Amyotrophic Lateral Sclerosis." Acta
Neuropathologica 93, no. 6 (Jul 1997): 537-550.
DeKosky, S. T., and S. W. Scheff. "Synapse Loss in Frontal Cortex Biopsies in Alzheimer's
Disease: Correlation with Cognitive Severity." Annals of Neurology 27, no. 5 (May 1990):
457-464.
Dermaut, B., S. Kumar-Singh, S. Engelborghs, J. Theuns, R. Rademakers, J. Saerens, B. A.
Pickut, K. Peeters, M. van den Broeck, K. Vennekens, S. Claes, M. Cruts, P. Cras, J. J.
Martin, C. Van Broeckhoven, and P. P. De Deyn. "A Novel Presenilin 1 Mutation
Associated with Pick's Disease but Not Beta-Amyloid Plaques." Annals of Neurology 55,
no. 5 (May 2004): 617-626.
Di Giorgio, F. P., M. A. Carrasco, M. C. Siao, T. Maniatis, and K. Eggan. "Non-Cell
Autonomous Effect of Glia on Motor Neurons in an Embryonic Stem Cell-Based ALS
Model." Nature Neuroscience 10, no. 5 (May 2007): 608-614.
Dreses-Werringloer, U., J. C. Lambert, V. Vingtdeux, H. Zhao, H. Vais, A. Siebert, A. Jain, J.
Koppel, A. Rovelet-Lecrux, D. Hannequin, F. Pasquier, D. Galimberti, E. Scarpini, D.
Mann, C. Lendon, D. Campion, P. Amouyel, P. Davies, J. K. Foskett, F. Campagne, and P.
Marambaud. "A Polymorphism in CALHM1 Influences Ca2+ Homeostasis, Abeta Levels,
and Alzheimer's Disease Risk." Cell 133, no. 7 (Jun 27 2008): 1149-1161.
Elia, A. J., T. L. Parkes, K. Kirby, P. St George-Hyslop, G. L. Boulianne, J. P. Phillips, and A. J.
Hilliker. "Expression of Human FALS SOD in Motorneurons of Drosophila." Free Radical
Biology & Medicine 26, no. 9-10 (Jun 1999): 1332-1338.
99
Etcheberrigaray, R., N. Hirashima, L. Nee, J. Prince, S. Govoni, M. Racchi, R. E. Tanzi, and D.
L. Alkon. "Calcium Responses in Fibroblasts from Asymptomatic Members of Alzheimer's
Disease Families." Neurobiology of Disease 5, no. 1 (Jul 1998): 37-45.
Ferrante, R. J., S. E. Browne, L. A. Shinobu, A. C. Bowling, M. J. Baik, U. MacGarvey, N. W.
Kowall, R. H. Brown Jr, and M. F. Beal. "Evidence of Increased Oxidative Damage in both
Sporadic and Familial Amyotrophic Lateral Sclerosis." Journal of Neurochemistry 69, no. 5
(Nov 1997): 2064-2074.
Finelli, A., A. Kelkar, H. J. Song, H. Yang, and M. Konsolaki. "A Model for Studying
Alzheimer's Abeta42-Induced Toxicity in Drosophila Melanogaster." Molecular and
Cellular Neurosciences 26, no. 3 (Jul 2004): 365-375.
Fossgreen, A., B. Bruckner, C. Czech, C. L. Masters, K. Beyreuther, and R. Paro. "Transgenic
Drosophila Expressing Human Amyloid Precursor Protein show Gamma-Secretase Activity
and a Blistered-Wing Phenotype." Proceedings of the National Academy of Sciences of the
United States of America 95, no. 23 (Nov 10 1998): 13703-13708.
Gaburjakova, M., J. Gaburjakova, S. Reiken, F. Huang, S. O. Marx, N. Rosemblit, and A. R.
Marks. "FKBP12 Binding Modulates Ryanodine Receptor Channel Gating." Journal of
Biological Chemistry 276, no. 20 (May 18 2001): 16931-16935.
Geula, C., and M. M. Mesulam. "Cortical Cholinergic Fibers in Aging and Alzheimer's Disease:
A Morphometric Study." Neuroscience 33, no. 3 (1989): 469-481.
Geula, C., C. K. Wu, D. Saroff, A. Lorenzo, M. Yuan, and B. A. Yankner. "Aging Renders the
Brain Vulnerable to Amyloid Beta-Protein Neurotoxicity." Nature Medicine 4, no. 7 (Jul
1998): 827-831.
Giannakopoulos, P., F. R. Herrmann, T. Bussiere, C. Bouras, E. Kovari, D. P. Perl, J. H.
Morrison, G. Gold, and P. R. Hof. "Tangle and Neuron Numbers, but Not Amyloid Load,
Predict Cognitive Status in Alzheimer's Disease." Neurology 60, no. 9 (May 13 2003):
1495-1500.
Gilman, S., M. Koller, R. S. Black, L. Jenkins, S. G. Griffith, N. C. Fox, L. Eisner, L. Kirby, M.
B. Rovira, F. Forette, J. M. Orgogozo, and AN1792(QS-21)-201 Study Team. "Clinical
Effects of Abeta Immunization (AN1792) in Patients with AD in an Interrupted Trial."
Neurology 64, no. 9 (May 10 2005): 1553-1562.
Goate, A., M. C. Chartier-Harlin, M. Mullan, J. Brown, F. Crawford, L. Fidani, L. Giuffra, A.
Haynes, N. Irving, and L. James. "Segregation of a Missense Mutation in the Amyloid
100
Precursor Protein Gene with Familial Alzheimer's Disease." Nature 349, no. 6311 (Feb 21
1991a): 704-706.
"Segregation of a Missense Mutation in the Amyloid Precursor Protein Gene with Familial
Alzheimer's Disease." Nature 349, no. 6311 (Feb 21 1991b): 704-706.
Gong, Y. H., A. S. Parsadanian, A. Andreeva, W. D. Snider, and J. L. Elliott. "Restricted
Expression of G86R Cu/Zn Superoxide Dismutase in Astrocytes Results in Astrocytosis but
does Not Cause Motoneuron Degeneration." The Journal of Neuroscience : The Official
Journal of the Society for Neuroscience 20, no. 2 (Jan 15 2000): 660-665.
Gotz, J., and L. M. Ittner. "Animal Models of Alzheimer's Disease and Frontotemporal
Dementia." Nature Reviews.Neuroscience 9, no. 7 (Jul 2008): 532-544.
Guo, Q., W. Fu, B. L. Sopher, M. W. Miller, C. B. Ware, G. M. Martin, and M. P. Mattson.
"Increased Vulnerability of Hippocampal Neurons to Excitotoxic Necrosis in Presenilin-1
Mutant Knock-in Mice." Nature Medicine 5, no. 1 (Jan 1999): 101-106.
Guo, Q., K. Furukawa, B. L. Sopher, D. G. Pham, J. Xie, N. Robinson, G. M. Martin, and M. P.
Mattson. "Alzheimer's PS-1 Mutation Perturbs Calcium Homeostasis and Sensitizes PC12
Cells to Death Induced by Amyloid Beta-Peptide." Neuroreport 8, no. 1 (Dec 20 1996):
379-383.
Guo, Y., I. Livne-Bar, L. Zhou, and G. L. Boulianne. "Drosophila Presenilin is Required for
Neuronal Differentiation and Affects Notch Subcellular Localization and Signaling." The
Journal of Neuroscience : The Official Journal of the Society for Neuroscience 19, no. 19
(Oct 1 1999): 8435-8442.
Gurney, M. E., H. Pu, A. Y. Chiu, M. C. Dal Canto, C. Y. Polchow, D. D. Alexander, J.
Caliendo, A. Hentati, Y. W. Kwon, and H. X. Deng. "Motor Neuron Degeneration in Mice
that Express a Human Cu,Zn Superoxide Dismutase Mutation." Science 264, no. 5166 (Jun
17 1994): 1772-1775.
Hand, C. K., and G. A. Rouleau. "Familial Amyotrophic Lateral Sclerosis." Muscle & Nerve 25,
no. 2 (Feb 2002): 135-159.
Hardy, J. A., and G. A. Higgins. "Alzheimer's Disease: The Amyloid Cascade Hypothesis."
Science (New York, N.Y.) 256, no. 5054 (Apr 10 1992): 184-185.
Hayrapetyan, V., V. Rybalchenko, N. Rybalchenko, and P. Koulen. "The N-Terminus of
Presenilin-2 Increases Single Channel Activity of Brain Ryanodine Receptors through
Direct Protein-Protein Interaction." Cell Calcium 44, no. 5 (Nov 2008): 507-518.
101
Heiman, R. G., R. C. Atkinson, B. F. Andruss, C. Bolduc, G. E. Kovalick, and K. Beckingham.
"Spontaneous Avoidance Behavior in Drosophila Null for Calmodulin Expression."
Proceedings of the National Academy of Sciences of the United States of America 93, no. 6
(Mar 19 1996a): 2420-2425.
"Spontaneous Avoidance Behavior in Drosophila Null for Calmodulin Expression." Proceedings
of the National Academy of Sciences of the United States of America 93, no. 6 (Mar 19
1996b): 2420-2425.
Herreman, A., D. Hartmann, W. Annaert, P. Saftig, K. Craessaerts, L. Serneels, L. Umans, V.
Schrijvers, F. Checler, H. Vanderstichele, V. Baekelandt, R. Dressel, P. Cupers, D.
Huylebroeck, A. Zwijsen, F. Van Leuven, and B. De Strooper. "Presenilin 2 Deficiency
Causes a Mild Pulmonary Phenotype and no Changes in Amyloid Precursor Protein
Processing but Enhances the Embryonic Lethal Phenotype of Presenilin 1 Deficiency."
Proceedings of the National Academy of Sciences of the United States of America 96, no. 21
(Oct 12 1999): 11872-11877.
Herreman, A., L. Serneels, W. Annaert, D. Collen, L. Schoonjans, and B. De Strooper. "Total
Inactivation of Gamma-Secretase Activity in Presenilin-Deficient Embryonic Stem Cells."
Nature Cell Biology 2, no. 7 (Jul 2000): 461-462.
Hoskins, B., and I. K. Ho. "Effects of Maturation and Aging on Calmodulin and Calmodulin-
Regulated Enzymes in various Regions of Mouse Brain." Mechanisms of Ageing and
Development 36, no. 2 (Oct 1986): 173-186.
Hyman, B. T. "The Neuropathological Diagnosis of Alzheimer's Disease: Clinical-Pathological
Studies." Neurobiology of Aging 18, no. 4 Suppl (Jul-Aug 1997): S27-32.
Iijima, K., H. P. Liu, A. S. Chiang, S. A. Hearn, M. Konsolaki, and Y. Zhong. "Dissecting the
Pathological Effects of Human Abeta40 and Abeta42 in Drosophila: A Potential Model for
Alzheimer's Disease." Proceedings of the National Academy of Sciences of the United States
of America 101, no. 17 (Apr 27 2004): 6623-6628.
Irizarry, M. C., M. McNamara, K. Fedorchak, K. Hsiao, and B. T. Hyman. "APPSw Transgenic
Mice Develop Age-Related A Beta Deposits and Neuropil Abnormalities, but no Neuronal
Loss in CA1." Journal of Neuropathology and Experimental Neurology 56, no. 9 (Sep
1997): 965-973.
Jaarsma, D., E. D. Haasdijk, J. A. Grashorn, R. Hawkins, W. van Duijn, H. W. Verspaget, J.
London, and J. C. Holstege. "Human Cu/Zn Superoxide Dismutase (SOD1) Overexpression
in Mice Causes Mitochondrial Vacuolization, Axonal Degeneration, and Premature
Motoneuron Death and Accelerates Motoneuron Disease in Mice Expressing a Familial
102
Amyotrophic Lateral Sclerosis Mutant SOD1." Neurobiology of Disease 7, no. 6 Pt B (Dec
2000): 623-643.
Johnston, J. A., M. J. Dalton, M. E. Gurney, and R. R. Kopito. "Formation of High Molecular
Weight Complexes of Mutant Cu, Zn-Superoxide Dismutase in a Mouse Model for Familial
Amyotrophic Lateral Sclerosis." Proceedings of the National Academy of Sciences of the
United States of America 97, no. 23 (Nov 7 2000a): 12571-12576.
"Formation of High Molecular Weight Complexes of Mutant Cu, Zn-Superoxide Dismutase in a
Mouse Model for Familial Amyotrophic Lateral Sclerosis." Proceedings of the National
Academy of Sciences of the United States of America 97, no. 23 (Nov 7 2000b): 12571-
12576.
Johnston, J. A., C. L. Ward, and R. R. Kopito. "Aggresomes: A Cellular Response to Misfolded
Proteins." The Journal of Cell Biology 143, no. 7 (Dec 28 1998): 1883-1898.
Jonsson, P. A., K. Ernhill, P. M. Andersen, D. Bergemalm, T. Brannstrom, O. Gredal, P. Nilsson,
and S. L. Marklund. "Minute Quantities of Misfolded Mutant Superoxide Dismutase-1
Cause Amyotrophic Lateral Sclerosis." Brain : A Journal of Neurology 127, no. Pt 1 (Jan
2004): 73-88.
Kasri, N. N., S. L. Kocks, L. Verbert, S. S. Hebert, G. Callewaert, J. B. Parys, L. Missiaen, and
H. De Smedt. "Up-Regulation of Inositol 1,4,5-Trisphosphate Receptor Type 1 is
Responsible for a Decreased Endoplasmic-Reticulum Ca2+ Content in Presenilin Double
Knock-Out Cells." Cell Calcium 40, no. 1 (Jul 2006): 41-51.
Keller, J. N., E. Dimayuga, Q. Chen, J. Thorpe, J. Gee, and Q. Ding. "Autophagy, Proteasomes,
Lipofuscin, and Oxidative Stress in the Aging Brain." The International Journal of
Biochemistry & Cell Biology 36, no. 12 (Dec 2004): 2376-2391.
Knight, D., K. Iliadi, M. P. Charlton, H. L. Atwood, and G. L. Boulianne. "Presynaptic Plasticity
and Associative Learning are Impaired in a Drosophila Presenilin Null Mutant."
Developmental Neurobiology 67, no. 12 (Oct 2007): 1598-1613.
Kudo, T., T. Katayama, K. Imaizumi, Y. Yasuda, M. Yatera, M. Okochi, M. Tohyama, and M.
Takeda. "The Unfolded Protein Response is Involved in the Pathology of Alzheimer's
Disease." Annals of the New York Academy of Sciences 977 (Nov 2002): 349-355.
Lacomblez, L., G. Bensimon, P. N. Leigh, P. Guillet, and V. Meininger. "Dose-Ranging Study of
Riluzole in Amyotrophic Lateral Sclerosis. Amyotrophic Lateral Sclerosis/Riluzole Study
Group II." Lancet 347, no. 9013 (May 25 1996): 1425-1431.
103
Lee, M. K., H. H. Slunt, L. J. Martin, G. Thinakaran, G. Kim, S. E. Gandy, M. Seeger, E. Koo,
D. L. Price, and S. S. Sisodia. "Expression of Presenilin 1 and 2 (PS1 and PS2) in Human
and Murine Tissues." The Journal of Neuroscience : The Official Journal of the Society for
Neuroscience 16, no. 23 (Dec 1 1996): 7513-7525.
Leissring, M. A., Y. Akbari, C. M. Fanger, M. D. Cahalan, M. P. Mattson, and F. M. LaFerla.
"Capacitative Calcium Entry Deficits and Elevated Luminal Calcium Content in Mutant
Presenilin-1 Knockin Mice." The Journal of Cell Biology 149, no. 4 (May 15 2000): 793-
798.
Leissring, M. A., W. Farris, A. Y. Chang, D. M. Walsh, X. Wu, X. Sun, M. P. Frosch, and D. J.
Selkoe. "Enhanced Proteolysis of Beta-Amyloid in APP Transgenic Mice Prevents Plaque
Formation, Secondary Pathology, and Premature Death." Neuron 40, no. 6 (Dec 18 2003):
1087-1093.
Leissring, M. A., F. M. LaFerla, N. Callamaras, and I. Parker. "Subcellular Mechanisms of
Presenilin-Mediated Enhancement of Calcium Signaling." Neurobiology of Disease 8, no. 3
(Jun 2001): 469-478.
Leissring, M. A., M. P. Murphy, T. R. Mead, Y. Akbari, M. C. Sugarman, M. Jannatipour, B.
Anliker, U. Muller, P. Saftig, B. De Strooper, M. S. Wolfe, T. E. Golde, and F. M. LaFerla.
"A Physiologic Signaling Role for the Gamma -Secretase-Derived Intracellular Fragment of
APP." Proceedings of the National Academy of Sciences of the United States of America 99,
no. 7 (Apr 2 2002): 4697-4702.
Levitan, D., T. G. Doyle, D. Brousseau, M. K. Lee, G. Thinakaran, H. H. Slunt, S. S. Sisodia,
and I. Greenwald. "Assessment of Normal and Mutant Human Presenilin Function in
Caenorhabditis Elegans." Proceedings of the National Academy of Sciences of the United
States of America 93, no. 25 (Dec 10 1996): 14940-14944.
Li, Z., J. L. Joyal, and D. B. Sacks. "Calmodulin Enhances the Stability of the Estrogen
Receptor." The Journal of Biological Chemistry 276, no. 20 (May 18 2001): 17354-17360.
Lindsay, Colin. A Portrait of Seniors in Canada. 3rd Ed. Ottawa, Canada: Ottawa: Statistics
Canada, 1999.
Lindsay, J., E. Sykes, I. McDowell, R. Verreault, and D. Laurin. "More than the Epidemiology
of Alzheimer's Disease: Contributions of the Canadian Study of Health and Aging."
Canadian Journal of Psychiatry.Revue Canadienne De Psychiatrie 49, no. 2 (Feb 2004):
83-91.
104
Lino, M. M., C. Schneider, and P. Caroni. "Accumulation of SOD1 Mutants in Postnatal
Motoneurons does Not Cause Motoneuron Pathology Or Motoneuron Disease." The Journal
of Neuroscience : The Official Journal of the Society for Neuroscience 22, no. 12 (Jun 15
2002): 4825-4832.
Lu, Y., Y. Lv, Y. Ye, Y. Wang, Y. Hong, M. E. Fortini, Y. Zhong, and Z. Xie. "A Role for
Presenilin in Post-Stress Regulation: Effects of Presenilin Mutations on Ca2+ Currents in
Drosophila." The FASEB Journal : Official Publication of the Federation of American
Societies for Experimental Biology 21, no. 10 (Aug 2007): 2368-2378.
Majoor-Krakauer, D., P. J. Willems, and A. Hofman. "Genetic Epidemiology of Amyotrophic
Lateral Sclerosis." Clinical Genetics 63, no. 2 (Feb 2003): 83-101.
Massie, H. R., Valerie Aeillo R, Trevor R. Williams, and L. K. DeWolfe. "Calcium and
Calmodulin Changes with Aging in Drosophila." Age 12 (1989): 7-11.
Matsumoto, G., A. Stojanovic, C. I. Holmberg, S. Kim, and R. I. Morimoto. "Structural
Properties and Neuronal Toxicity of Amyotrophic Lateral Sclerosis-Associated Cu/Zn
Superoxide Dismutase 1 Aggregates." The Journal of Cell Biology 171, no. 1 (Oct 10 2005):
75-85.
Mattson, M. P., and S. L. Chan. "Calcium Orchestrates Apoptosis." Nature Cell Biology 5, no.
12 (Dec 2003): 1041-1043.
McCullough, K. D., J. L. Martindale, L. O. Klotz, T. Y. Aw, and N. J. Holbrook. "Gadd153
Sensitizes Cells to Endoplasmic Reticulum Stress by Down-Regulating Bcl2 and Perturbing
the Cellular Redox State." Molecular and Cellular Biology 21, no. 4 (Feb 2001): 1249-1259.
Mitsumoto, H. "Diagnosis and Progression of ALS." 48, no. Suppl 4 (1997): S2.
Mockett, R. J., S. N. Radyuk, J. J. Benes, W. C. Orr, and R. S. Sohal. "Phenotypic Effects of
Familial Amyotrophic Lateral Sclerosis Mutant Sod Alleles in Transgenic Drosophila."
Proceedings of the National Academy of Sciences of the United States of America 100, no. 1
(Feb 7 2003): 301-306.
Morohashi, Y., N. Hatano, S. Ohya, R. Takikawa, T. Watabiki, N. Takasugi, Y. Imaizumi, T.
Tomita, and T. Iwatsubo. "Molecular Cloning and Characterization of CALP/KChIP4, a
Novel EF-Hand Protein Interacting with Presenilin 2 and Voltage-Gated Potassium Channel
Subunit Kv4." Journal of Biological Chemistry 277, no. 17 (Apr 26 2002): 14965-14975.
105
Nagai, M., D. B. Re, T. Nagata, A. Chalazonitis, T. M. Jessell, H. Wichterle, and S. Przedborski.
"Astrocytes Expressing ALS-Linked Mutated SOD1 Release Factors Selectively Toxic to
Motor Neurons." Nature Neuroscience 10, no. 5 (May 2007): 615-622.
Nelson, L. M. "Epidemiology of ALS." Clinical Neuroscience (New York, N.Y.) 3, no. 6 (-1996
1995): 327-331.
Nguyen, H. N., C. Wang, and D. C. Perry. "Depletion of Intracellular Calcium Stores is Toxic to
SH-SY5Y Neuronal Cells." Brain Research 924, no. 2 (Jan 11 2002): 159-166.
Nicoll, R. A., C. Malenka, and J. A. Kauer. "The Role of Calcium in Long-Term Potentiation."
Annals of the New York Academy of Sciences 568 (1989): 166.
Nishitoh, H., H. Kadowaki, A. Nagai, T. Maruyama, T. Yokota, H. Fukutomi, T. Noguchi, A.
Matsuzawa, K. Takeda, and H. Ichijo. "ALS-Linked Mutant SOD1 Induces ER Stress- and
ASK1-Dependent Motor Neuron Death by Targeting Derlin-1." Genes & Development 22,
no. 11 (Jun 1 2008): 1451-1464.
O'Day, D. H., and M. A. Myre. "Calmodulin-Binding Domains in Alzheimer's Disease Proteins:
Extending the Calcium Hypothesis." Biochemical and Biophysical Research
Communications 320, no. 4 (Aug 6 2004): 1051-1054.
Oddo, S., L. Billings, J. P. Kesslak, D. H. Cribbs, and F. M. LaFerla. "Abeta Immunotherapy
Leads to Clearance of Early, but Not Late, Hyperphosphorylated Tau Aggregates Via the
Proteasome." Neuron 43, no. 3 (Aug 5 2004): 321-332.
Olsen, S. R., and R. I. Wilson. "Cracking Neural Circuits in a Tiny Brain: New Approaches for
Understanding the Neural Circuitry of Drosophila." Trends in Neurosciences 31, no. 10 (Oct
2008): 512-520.
Ostbye, T., and E. Crosse. "Net Economic Costs of Dementia in Canada." CMAJ : Canadian
Medical Association Journal = Journal De l'Association Medicale Canadienne 151, no. 10
(Nov 15 1994): 1457-1464.
Pack-Chung, E., M. B. Meyers, W. P. Pettingell, R. D. Moir, A. M. Brownawell, I. Cheng, R. E.
Tanzi, and T. W. Kim. "Presenilin 2 Interacts with Sorcin, a Modulator of the Ryanodine
Receptor." The Journal of Biological Chemistry 275, no. 19 (May 12 2000): 14440-14445.
Palop, J. J., J. Chin, and L. Mucke. "A Network Dysfunction Perspective on Neurodegenerative
Diseases." Nature 443, no. 7113 (Oct 19 2006): 768-773.
Pardo, C. A., Z. Xu, D. R. Borchelt, D. L. Price, S. S. Sisodia, and D. W. Cleveland. "Superoxide
Dismutase is an Abundant Component in Cell Bodies, Dendrites, and Axons of Motor
106
Neurons and in a Subset of Other Neurons." Proceedings of the National Academy of
Sciences of the United States of America 92, no. 4 (Feb 14 1995): 954-958.
Parkes, T. L., A. J. Elia, D. Dickinson, A. J. Hilliker, J. P. Phillips, and G. L. Boulianne.
"Extension of Drosophila Lifespan by Overexpression of Human SOD1 in Motorneurons."
Nature Genetics 19, no. 2 (Jun 1998): 171-174.
Peled-Kamar, M., J. Lotem, I. Wirguin, L. Weiner, A. Hermalin, and Y. Groner. "Oxidative
Stress Mediates Impairment of Muscle Function in Transgenic Mice with Elevated Level of
Wild-Type Cu/Zn Superoxide Dismutase." Proceedings of the National Academy of
Sciences of the United States of America 94, no. 8 (Apr 15 1997): 3883-3887.
Periz, G., and M. E. Fortini. "Ca(2+)-ATPase Function is Required for Intracellular Trafficking
of the Notch Receptor in Drosophila." The EMBO Journal 18, no. 21 (Nov 1 1999): 5983-
5993.
Phillips, J. P., S. D. Campbell, D. Michaud, M. Charbonneau, and A. J. Hilliker. "Null Mutation
of copper/zinc Superoxide Dismutase in Drosophila Confers Hypersensitivity to Paraquat
and Reduced Longevity." Proceedings of the National Academy of Sciences of the United
States of America 86, no. 8 (Apr 1989): 2761-2765.
Phillips, J. P., J. A. Tainer, E. D. Getzoff, G. L. Boulianne, K. Kirby, and A. J. Hilliker.
"Subunit-Destabilizing Mutations in Drosophila copper/zinc Superoxide Dismutase:
Neuropathology and a Model of Dimer Dysequilibrium." Proceedings of the National
Academy of Sciences of the United States of America 92, no. 19 (Oct 12 1995): 8574-8578.
Puttaparthi, K., C. Wojcik, B. Rajendran, G. N. DeMartino, and J. L. Elliott. "Aggregate
Formation in the Spinal Cord of Mutant SOD1 Transgenic Mice is Reversible and Mediated
by Proteasomes." Journal of Neurochemistry 87, no. 4 (Nov 2003): 851-860.
Qi-Takahara, Y., M. Morishima-Kawashima, Y. Tanimura, G. Dolios, N. Hirotani, Y. Horikoshi,
F. Kametani, M. Maeda, T. C. Saido, R. Wang, and Y. Ihara. "Longer Forms of Amyloid
Beta Protein: Implications for the Mechanism of Intramembrane Cleavage by Gamma-
Secretase." The Journal of Neuroscience : The Official Journal of the Society for
Neuroscience 25, no. 2 (Jan 12 2005): 436-445.
Querfurth, H. W., J. Jiang, J. D. Geiger, and D. J. Selkoe. "Caffeine Stimulates Amyloid Beta-
Peptide Release from Beta-Amyloid Precursor Protein-Transfected HEK293 Cells." Journal
of Neurochemistry 69, no. 4 (Oct 1997): 1580-1591.
Rakhit, R., J. P. Crow, J. R. Lepock, L. H. Kondejewski, N. R. Cashman, and A. Chakrabartty.
"Monomeric Cu,Zn-Superoxide Dismutase is a Common Misfolding Intermediate in the
107
Oxidation Models of Sporadic and Familial Amyotrophic Lateral Sclerosis." The Journal of
Biological Chemistry 279, no. 15 (Apr 9 2004): 15499-15504.
Rakhit, R., J. Robertson, C. Vande Velde, P. Horne, D. M. Ruth, J. Griffin, D. W. Cleveland, N.
R. Cashman, and A. Chakrabartty. "An Immunological Epitope Selective for Pathological
Monomer-Misfolded SOD1 in ALS." Nature Medicine 13, no. 6 (Jun 2007): 754-759.
Ratovitski, T., L. B. Corson, J. Strain, P. Wong, D. W. Cleveland, V. C. Culotta, and D. R.
Borchelt. "Variation in the biochemical/biophysical Properties of Mutant Superoxide
Dismutase 1 Enzymes and the Rate of Disease Progression in Familial Amyotrophic Lateral
Sclerosis Kindreds." Human Molecular Genetics 8, no. 8 (Aug 1999): 1451-1460.
Raux, G., R. Gantier, C. Thomas-Anterion, J. Boulliat, P. Verpillat, D. Hannequin, A. Brice, T.
Frebourg, and D. Campion. "Dementia with Prominent Frontotemporal Features Associated
with L113P Presenilin 1 Mutation." Neurology 55, no. 10 (Nov 28 2000): 1577-1578.
Reaume, A. G., J. L. Elliott, E. K. Hoffman, N. W. Kowall, R. J. Ferrante, D. F. Siwek, H. M.
Wilcox, D. G. Flood, M. F. Beal, R. H. Brown Jr, R. W. Scott, and W. D. Snider. "Motor
Neurons in Cu/Zn Superoxide Dismutase-Deficient Mice Develop Normally but Exhibit
Enhanced Cell Death After Axonal Injury." Nature Genetics 13, no. 1 (May 1996): 43-47.
Rogaev, E. I., R. Sherrington, E. A. Rogaeva, G. Levesque, M. Ikeda, Y. Liang, H. Chi, C. Lin,
K. Holman, and T. Tsuda. "Familial Alzheimer's Disease in Kindreds with Missense
Mutations in a Gene on Chromosome 1 Related to the Alzheimer's Disease Type 3 Gene."
Nature 376, no. 6543 (Aug 31 1995): 775-778.
Rosen, D. R., T. Siddique, D. Patterson, D. A. Figlewicz, P. Sapp, A. Hentati, D. Donaldson, J.
Goto, J. P. O'Regan, and H. X. Deng. "Mutations in Cu/Zn Superoxide Dismutase Gene are
Associated with Familial Amyotrophic Lateral Sclerosis." Nature 362, no. 6415 (Apr 4
1993): 59-62.
Rubin, G. M., M. D. Yandell, J. R. Wortman, G. L. Gabor Miklos, C. R. Nelson, I. K. Hariharan,
M. E. Fortini, P. W. Li, R. Apweiler, W. Fleischmann, J. M. Cherry, S. Henikoff, M. P.
Skupski, S. Misra, M. Ashburner, E. Birney, M. S. Boguski, T. Brody, P. Brokstein, S. E.
Celniker, S. A. Chervitz, D. Coates, A. Cravchik, A. Gabrielian, R. F. Galle, W. M. Gelbart,
R. A. George, L. S. Goldstein, F. Gong, P. Guan, N. L. Harris, B. A. Hay, R. A. Hoskins, J.
Li, Z. Li, R. O. Hynes, S. J. Jones, P. M. Kuehl, B. Lemaitre, J. T. Littleton, D. K. Morrison,
C. Mungall, P. H. O'Farrell, O. K. Pickeral, C. Shue, L. B. Vosshall, J. Zhang, Q. Zhao, X.
H. Zheng, and S. Lewis. "Comparative Genomics of the Eukaryotes." Science 287, no. 5461
(Mar 24 2000): 2204-2215.
108
Rybalchenko, V., S. Y. Hwang, N. Rybalchenko, and P. Koulen. "The Cytosolic N-Terminus of
Presenilin-1 Potentiates Mouse Ryanodine Receptor Single Channel Activity." The
International Journal of Biochemistry & Cell Biology 40, no. 1 (2008): 84-97.
Ryoo, H. D., and H. Steller. "Unfolded Protein Response in Drosophila: Why another Model can
make it Fly." Cell Cycle (Georgetown, Tex.) 6, no. 7 (Apr 1 2007): 830-835.
Salvaterra, P. M., and T. Kitamoto. "Drosophila Cholinergic Neurons and Processes Visualized
with Gal4/UAS-GFP." Brain Research.Gene Expression Patterns 1, no. 1 (Aug 2001): 73-
82.
Saura, C. A., S. Y. Choi, V. Beglopoulos, S. Malkani, D. Zhang, B. S. Shankaranarayana Rao, S.
Chattarji, R. J. Kelleher 3rd, E. R. Kandel, K. Duff, A. Kirkwood, and J. Shen. "Loss of
Presenilin Function Causes Impairments of Memory and Synaptic Plasticity Followed by
Age-Dependent Neurodegeneration." Neuron 42, no. 1 (Apr 8 2004): 23-36.
Schonheit, B., R. Zarski, and T. G. Ohm. "Spatial and Temporal Relationships between Plaques
and Tangles in Alzheimer-Pathology." Neurobiology of Aging 25, no. 6 (Jul 2004): 697-711.
Scorrano, L., S. A. Oakes, J. T. Opferman, E. H. Cheng, M. D. Sorcinelli, T. Pozzan, and S. J.
Korsmeyer. "BAX and BAK Regulation of Endoplasmic Reticulum Ca2+: A Control Point
for Apoptosis." Science (New York, N.Y.) 300, no. 5616 (Apr 4 2003): 135-139.
Seidner, G. A., Y. Ye, M. M. Faraday, W. G. Alvord, and M. E. Fortini. "Modeling Clinically
Heterogeneous Presenilin Mutations with Transgenic Drosophila." Current Biology : CB 16,
no. 10 (May 23 2006): 1026-1033.
Shaw, P. J., P. G. Ince, G. Falkous, and D. Mantle. "Oxidative Damage to Protein in Sporadic
Motor Neuron Disease Spinal Cord." Annals of Neurology 38, no. 4 (Oct 1995): 691-695.
Shen, J., R. T. Bronson, D. F. Chen, W. Xia, D. J. Selkoe, and S. Tonegawa. "Skeletal and CNS
Defects in Presenilin-1-Deficient Mice." Cell 89, no. 4 (May 16 1997): 629-639.
Sherrington, R., E. I. Rogaev, Y. Liang, E. A. Rogaeva, G. Levesque, M. Ikeda, H. Chi, C. Lin,
G. Li, and K. Holman. "Cloning of a Gene Bearing Missense Mutations in Early-Onset
Familial Alzheimer's Disease." Nature 375, no. 6534 (Jun 29 1995): 754-760.
Sicaeros, B., J. M. Campusano, and D. K. O'Dowd. "Primary Neuronal Cultures from the Brains
of Late Stage Drosophila Pupae." Journal of Visualized Experiments : JoVE (4), no. 4
(2007): 200.
109
Simon, A. F., D. T. Liang, and D. E. Krantz. "Differential Decline in Behavioral Performance of
Drosophila Melanogaster with Age." Mechanisms of Ageing and Development 127, no. 7
(Jul 2006): 647-651.
Smith, I. F., J. P. Boyle, P. F. Vaughan, H. A. Pearson, R. F. Cowburn, and C. S. Peers. "Ca(2+)
Stores and Capacitative Ca(2+) Entry in Human Neuroblastoma (SH-SY5Y) Cells
Expressing a Familial Alzheimer's Disease Presenilin-1 Mutation." Brain Research 949, no.
1-2 (Sep 13 2002): 105-111.
Stabler, S. M., L. L. Ostrowski, S. M. Janicki, and M. J. Monteiro. "A Myristoylated Calcium-
Binding Protein that Preferentially Interacts with the Alzheimer's Disease Presenilin 2
Protein." The Journal of Cell Biology 145, no. 6 (Jun 14 1999): 1277-1292.
Steiner, B., E. M. Mandelkow, J. Biernat, N. Gustke, H. E. Meyer, B. Schmidt, G. Mieskes, H.
D. Soling, D. Drechsel, and M. W. Kirschner. "Phosphorylation of Microtubule-Associated
Protein Tau: Identification of the Site for Ca2(+)-Calmodulin Dependent Kinase and
Relationship with Tau Phosphorylation in Alzheimer Tangles." The EMBO Journal 9, no.
11 (Nov 1990): 3539-3544.
Struhl, G., and I. Greenwald. "Presenilin is Required for Activity and Nuclear Access of Notch
in Drosophila." Nature 398, no. 6727 (Apr 8 1999): 522-525.
Stull, J. T. "Ca2+-Dependent Cell Signaling through Calmodulin-Activated Protein Phosphatase
and Protein Kinases Minireview Series." Journal of Biological Chemistry 276, no. 4 (Jan 26
2001): 2311-2312.
Stutzmann, G. E., I. Smith, A. Caccamo, S. Oddo, I. Parker, and F. Laferla. "Enhanced
Ryanodine-Mediated Calcium Release in Mutant PS1-Expressing Alzheimer's Mouse
Models." Annals of the New York Academy of Sciences 1097 (Feb 2007): 265-277.
Takasugi, N., T. Tomita, I. Hayashi, M. Tsuruoka, M. Niimura, Y. Takahashi, G. Thinakaran,
and T. Iwatsubo. "The Role of Presenilin Cofactors in the Gamma-Secretase Complex."
Nature 422, no. 6930 (Mar 27 2003): 438-441.
Tanzi, R. E., and L. Bertram. "Twenty Years of the Alzheimer's Disease Amyloid Hypothesis: A
Genetic Perspective." Cell 120, no. 4 (Feb 25 2005): 545-555.
Terry, R. D., E. Masliah, D. P. Salmon, N. Butters, R. DeTeresa, R. Hill, L. A. Hansen, and R.
Katzman. "Physical Basis of Cognitive Alterations in Alzheimer's Disease: Synapse Loss is
the Major Correlate of Cognitive Impairment." Annals of Neurology 30, no. 4 (Oct 1991):
572-580.
110
Thibault, O., J. C. Gant, and P. W. Landfield. "Expansion of the Calcium Hypothesis of Brain
Aging and Alzheimer's Disease: Minding the Store." Aging Cell 6, no. 3 (Jun 2007): 307-
317.
Thinakaran, G., D. R. Borchelt, M. K. Lee, H. H. Slunt, L. Spitzer, G. Kim, T. Ratovitsky, F.
Davenport, C. Nordstedt, M. Seeger, J. Hardy, A. I. Levey, S. E. Gandy, N. A. Jenkins, N.
G. Copeland, D. L. Price, and S. S. Sisodia. "Endoproteolysis of Presenilin 1 and
Accumulation of Processed Derivatives in Vivo." Neuron 17, no. 1 (Jul 1996): 181-190.
Tu, H., O. Nelson, A. Bezprozvanny, Z. Wang, S. F. Lee, Y. H. Hao, L. Serneels, B. De
Strooper, G. Yu, and I. Bezprozvanny. "Presenilins Form ER Ca2+ Leak Channels, a
Function Disrupted by Familial Alzheimer's Disease-Linked Mutations." Cell 126, no. 5
(Sep 8 2006): 981-993.
Tummala, H., C. Jung, A. Tiwari, C. M. Higgins, L. J. Hayward, and Z. Xu. "Inhibition of
Chaperone Activity is a Shared Property of several Cu,Zn-Superoxide Dismutase Mutants
that Cause Amyotrophic Lateral Sclerosis." The Journal of Biological Chemistry 280, no. 18
(May 6 2005): 17725-17731.
van Marum, R. J. "Current and Future Therapy in Alzheimer's Disease." Fundamental & Clinical
Pharmacology 22, no. 3 (Jun 2008): 265-274.
Verkhratsky, A., and O. H. Petersen. "The Endoplasmic Reticulum as an Integrating Signalling
Organelle: From Neuronal Signalling to Neuronal Death." European Journal of
Pharmacology 447, no. 2-3 (Jul 5 2002): 141-154.
Wang, B., K. M. Sullivan, and K. Beckingham. "Drosophila Calmodulin Mutants with Specific
Defects in the Musculature Or in the Nervous System." Genetics 165, no. 3 (Nov 2003):
1255-1268.
Wang, J., G. Xu, and D. R. Borchelt. "High Molecular Weight Complexes of Mutant Superoxide
Dismutase 1: Age-Dependent and Tissue-Specific Accumulation." Neurobiology of Disease
9, no. 2 (Mar 2002): 139-148.
Wang, R., B. Wang, W. He, and H. Zheng. "Wild-Type Presenilin 1 Protects Against Alzheimer
Disease Mutation-Induced Amyloid Pathology." The Journal of Biological Chemistry 281,
no. 22 (Jun 2 2006): 15330-15336.
Watanabe, M., M. Dykes-Hoberg, V. C. Culotta, D. L. Price, P. C. Wong, and J. D. Rothstein.
"Histological Evidence of Protein Aggregation in Mutant SOD1 Transgenic Mice and in
Amyotrophic Lateral Sclerosis Neural Tissues." Neurobiology of Disease 8, no. 6 (Dec
2001): 933-941.
111
Watson, M. R., R. D. Lagow, K. Xu, B. Zhang, and N. M. Bonini. "A Drosophila Model for
Amyotrophic Lateral Sclerosis Reveals Motor Neuron Damage by Human SOD1." The
Journal of Biological Chemistry 283, no. 36 (Sep 5 2008): 24972-24981.
Wolfe, M. S. "When Loss is Gain: Reduced Presenilin Proteolytic Function Leads to Increased
Abeta42/Abeta40. Talking Point on the Role of Presenilin Mutations in Alzheimer Disease."
EMBO Reports 8, no. 2 (Feb 2007): 136-140.
Wolfe, M. S., W. Xia, B. L. Ostaszewski, T. S. Diehl, W. T. Kimberly, and D. J. Selkoe. "Two
Transmembrane Aspartates in Presenilin-1 Required for Presenilin Endoproteolysis and
Gamma-Secretase Activity." Nature 398, no. 6727 (Apr 8 1999): 513-517.
Wong, P. C., H. Cai, D. R. Borchelt, and D. L. Price. "Genetically Engineered Mouse Models of
Neurodegenerative Diseases." Nature Neuroscience 5, no. 7 (Jul 2002): 633-639.
Wong, P. C., C. A. Pardo, D. R. Borchelt, M. K. Lee, N. G. Copeland, N. A. Jenkins, S. S.
Sisodia, D. W. Cleveland, and D. L. Price. "An Adverse Property of a Familial ALS-Linked
SOD1 Mutation Causes Motor Neuron Disease Characterized by Vacuolar Degeneration of
Mitochondria." Neuron 14, no. 6 (Jun 1995): 1105-1116.
Wu, L. J., Y. Lu, and T. L. Xu. "A Novel Mechanical Dissociation Technique for Studying
Acutely Isolated Maturing Drosophila Central Neurons." Journal of Neuroscience Methods
108, no. 2 (Jul 30 2001): 199-206.
Yang, J., K. Bridges, K. Y. Chen, and A. Y. Liu. "Riluzole Increases the Amount of Latent HSF1
for an Amplified Heat Shock Response and Cytoprotection." PLoS ONE 3, no. 8 (Aug 6
2008): e2864.
Ye, Y., and M. E. Fortini. "Apoptotic Activities of Wild-Type and Alzheimer's Disease-Related
Mutant Presenilins in Drosophila Melanogaster." The Journal of Cell Biology 146, no. 6
(Sep 20 1999): 1351-1364.
"Characterization of Drosophila Presenilin and its Colocalization with Notch during
Development." Mechanisms of Development 79, no. 1-2 (Dec 1998): 199-211.
Ye, Y., N. Lukinova, and M. E. Fortini. "Neurogenic Phenotypes and Altered Notch Processing
in Drosophila Presenilin Mutants." Nature 398, no. 6727 (Apr 8 1999): 525-529.
Yeromin, A. V., J. Roos, K. A. Stauderman, and M. D. Cahalan. "A Store-Operated Calcium
Channel in Drosophila S2 Cells." The Journal of General Physiology 123, no. 2 (Feb 2004):
167-182.
112
Yim, M. B., J. H. Kang, H. S. Yim, H. S. Kwak, P. B. Chock, and E. R. Stadtman. "A Gain-of-
Function of an Amyotrophic Lateral Sclerosis-Associated Cu,Zn-Superoxide Dismutase
Mutant: An Enhancement of Free Radical Formation due to a Decrease in km for Hydrogen
Peroxide." Proceedings of the National Academy of Sciences of the United States of America
93, no. 12 (Jun 11 1996): 5709-5714.
Yoshihara, M., A. W. Ensminger, and J. T. Littleton. "Neurobiology and the Drosophila
Genome." Functional & Integrative Genomics 1, no. 4 (Mar 2001): 235-240.
Zatti, G., A. Burgo, M. Giacomello, L. Barbiero, R. Ghidoni, G. Sinigaglia, C. Florean, S.
Bagnoli, G. Binetti, S. Sorbi, P. Pizzo, and C. Fasolato. "Presenilin Mutations Linked to
Familial Alzheimer's Disease Reduce Endoplasmic Reticulum and Golgi Apparatus Calcium
Levels." Cell Calcium 39, no. 6 (Jun 2006): 539-550.
Zatti, G., R. Ghidoni, L. Barbiero, G. Binetti, T. Pozzan, C. Fasolato, and P. Pizzo. "The
Presenilin 2 M239I Mutation Associated with Familial Alzheimer's Disease Reduces Ca2+
Release from Intracellular Stores." Neurobiology of Disease 15, no. 2 (Mar 2004): 269-278.
Zhang, Z., P. Nadeau, W. Song, D. Donoviel, M. Yuan, A. Bernstein, and B. A. Yankner.
"Presenilins are Required for Gamma-Secretase Cleavage of Beta-APP and Transmembrane
Cleavage of Notch-1." Nature Cell Biology 2, no. 7 (Jul 2000): 463-465.
Zheng, H., M. Jiang, M. E. Trumbauer, R. Hopkins, D. J. Sirinathsinghji, K. A. Stevens, M. W.
Conner, H. H. Slunt, S. S. Sisodia, H. Y. Chen, and L. H. Van der Ploeg. "Mice Deficient
for the Amyloid Precursor Protein Gene." Annals of the New York Academy of Sciences 777
(Jan 17 1996): 421-426.
Zhu, J., S. M. Stabler, J. B. Ames, I. Baskakov, and M. J. Monteiro. "Calcium Binding
Sequences in Calmyrin Regulates Interaction with Presenilin-2." Experimental Cell Research
300, no. 2 (Nov 1 2004): 440-454.
113
Appendices
Apendix 1. Gehan’s Wilcoxon Survival Analysis
5
UAS-PsnWT
represents a UAS-transgenic stock bearing a full length copy of the Drosophila
presenilin gene. UAS-PsnFAD
represents a UAS-transgenic stock bearing a methionine to valine
substigution at position 146 in the full length Drosophila presenilin protein. Camnull
is a loss of
function calmodulin allele. w1118
flies are wild type, non-trangenic flies.
114
Apendix 2. Gehan’s Wilcoxon Survival Analysis.
HS1 and HS2 represent independent UAS-transgenic stocks bearing human wild type SOD1
transgene insertion, uncrossed . HS1;da and HS2;da represent the same UAS-transgenes crossed
to the da-GAL4 driver. fALS1 and fALS2 represent two independent UAS-transgenic stocks
bearing a glycine to alanine substitution at position 93 in human fALS-SOD1, uncrossed.
fALS1;;da and fALS2/da represent the same UAS-transgenes crossed to the da-GAL4 driver
115