Introduction

Neurodegenerative Diseases
At only three pounds, the human brain is by far the most important organ in the human body.  The brain is our central processing unit; it is involved in every action we carry out, sensation we perceive, or thought we contemplate.  From the feel of the sun’s warmth on our skin to the complex muscle coordination necessary to walk up a flight of stairs, the brain plays a pivotal role in every action taken throughout life.  This universal involvement in our lives requires a highly organized and overwhelmingly complex connection of neurons, the signaling cells of the nervous system, in the brain.  In fact, the neuronal circuits of the brain are so complex that disruption of these specific connections results in a plethora of distinct neurological disorders, including schizophrenia, depression, autism, and epilepsy.  In addition, a family of tragic, incurable disorders termed neurodegenerative diseases can afflict the brain as well.  Neurodegenerative diseases increase in prevalence with age and have, unfortunately, become increasingly common in society as medical advances lead to increased longevity of the general population.

The unique environment of our brain contributes to its vulnerability to neurodegenerative disease in two important ways.  Firstly, the adult brain is characterized by very select neurogenesis (birth of new neurons) (Purves et al., 2008).  As such, neurons are not replaced in the same way that the skin or intestines replace lost cells.  Secondly, cellular regeneration in the central nervous system (CNS) is constrained by a number of mechanisms (Grandpre et al., 2000).  Neural circuits are precisely wired pathways of communication between cells that form when we are children.  These regenerative barriers were favored by evolution to prevent incorrect wiring of neurons after damage occurs, as incorrect rewiring could lead to further harm (Purves et al., 2008).  Once damage occurs in the CNS, it cannot be fixed, a property that makes studying neurodegeneration all the more important.

Each neurodegenerative disease is characterized by progressive death of neurons in a specific area of the brain or spinal cord.  Several of the most well known neurodegenerative diseases are Alzheimer’s disease (AD), multiple sclerosis, Huntington’s disease (HD), prion diseases such as Creutzfeldt-Jakob disease and bovine spongiform encephalopathy (mad cow disease), amyotrophic lateral sclerosis (ALS), and Parkinson’s disease (PD).  The specificity of degeneration in each disease is quite astounding.  For example, neurons in the cortex and hippocampus (involved in memory) degenerate in AD, resulting in the characteristic memory impairment seen in patients (Waldemar et al., 2007).  In ALS, upper and lower motor neurons (cells essential for muscle movements) deteriorate, leading to the inability of sufferers to initiate voluntary movement (Shaw et al., 2001).  Although neurodegenerative diseases are distinct in that they affect different regions of the brain, they are unified by the presence of protein deposits (often called aggregates) in or around the affected neurons (Taylor et al., 2002).  These deposits form when the involved proteins come out of solution, similar to how cheese or yogurt comes out milk.  In each neurodegenerative disease, one or two specific culprit proteins thought to be intimately involved in pathogenesis (disease development) compose a large portion of the aggregates.  However, the role that these protein aggregates play in cell death in neurodegenerative diseases is not fully understood.
Some neurodegenerative diseases can be grouped together.  One group of neurodegenerative diseases is termed the synucleinopathies, and this family includes dementia with lewy bodies (DLB), multiple system atrophy (MSA), lewy body dysphagia (LBD), and PD (Figure 1A and 1B).  DLB primarily afflicts the cerebral cortex, and it results in dementia reminiscent of AD.  MSA stems from loss of neurons in the putamen, globus pallidus, and caudate nucleus, and this disease leads to degeneration of movement and the body’s autonomic functions.  Lewy body dysphagia results from the degeneration of the vagus nerve, a nerve essential for muscle movements in the esophagus, and a hallmark symptom is difficulty swallowing.  While these diseases afflict several regions of the brain, each one is characterized by the accumulation of misfolded and aggregated α-synuclein in the dying neurons (Spillantini et al., 1998; Wakabayashi, 1999; Burn et al., 2001; Heidebrink et al., 2001).  Interestingly, DLB, PD, and MSA all involve the substantia nigra.  My thesis research centers on increasing understanding of the molecular basis behind one specific synucleiopathy, PD, but my findings will help understand all synucleinopathies.

Introduction to Parkinson’s Disease
PD is the most common movement-based disorder in the elderly and the second most common neurodegenerative disease after AD (Serulle et al., 2006; de Silviera et al., 2002).  Over 4 million people suffer from the disorder worldwide, and this number is expected to double by 2030 (Dorsey et al., 2007; Jain et al., 2005).  In PD patients, selective death of midbrain dopaminergic neurons in the substantia nigra pars compacta (SNpc) occurs (Barbosa et al., 1997).

Figure 1: Synucleinopathies Impact Numerous Brain Regions

(A) Schematic of the brain & the structures involved in four synucleinopathies.  Distinct regions of the brain are affected by individual synucleionpathic diseases.  Blue: Lewy Body Dimentia. Orange: Multiple System Atrophy.  Green: Parkinson’s Disease.  Red: Lewy Body Dysphagia (B) List of the regions afflicted by synucleionpathies.  While each individual synucleinopathy afflicts a different region of the brain, they are all united in the presence of misfolded, aggregated  α-synuclein within the dying neurons.
Brain image from: http://www.paulnussbaum.com/

These neurons, localized to a 1-2mm wide strip and darkened by the pigment neuromelanin, are part of the basal ganglia, a neural circuit situated at the base of the forebrain responsible for the initiation of voluntary movement (Figure 2A; Olanow and Tatton, 1999).  SNpc neurons project to the striatum, a structure that influences timing and coordination of muscle movements.  Additionally, SNpc neurons synthesize the neurotransmitter dopamine, an important signaling molecule in this area of the brain.  Diminished dopamine levels resulting from cell death further disrupt neuronal signaling.  The loss of SNpc neurons in PD patients results in gaps in the neural circuit and the onset of several movement based symptoms, such as rigidity, resting tremors, and slow movement (Goedert et al., 2001).  The exact cause of cell death in the SNpc is unknown, but many researchers believe the protein α-synuclein is the culprit (Outeiro and Lindquist, 2003; Cooper et al., 2006).

Upon examination under a light microscope, large protein accumulations are visible inside the dying neurons of PD patients (Muchowski, 2002; Eriksen et al., 2005; Robinson, 2008). These protein aggregates, as with all neurodegenerative diseases, are the hallmark symptom of PD.  They were first identified by Friederich Lewy in 1912 and have since been termed Lewy Bodies (LBs; Forster and Lewy, 1912).  Although LBs contain several different proteins, they are chiefly comprised of misfolded α-synuclein, a protein expressed throughout the brain (Spillantini et al., 1998).  Interestingly, LBs are present in both of the major forms of PD: sporadic and familial.

Variety of Parkinson’s Disease: Sporadic and Familial
Many diseases have both environmental or genetic causes.  Diseases that arise without a known genetic cause are defined as sporadic.  Conversely, diseases caused by heritable genetic mutations are termed familial.  Over 90% of PD cases arise sporadically, and possible initiators are pesticide exposure (Ascherio et al., 2006), infection (Altschuler, 2007) and contact with heavy metals (Calne et al., 1994).  Mitochondrial dysfunction (Langston et al., 1983; Langston et al., 1984) and free radical damage (Jenner and Olanow, 1996; Maguire-Zeiss et al., 2005) have also been implicated in disease onset.  Additionally, sporadic PD is closely linked to the misfolding and aggregation of α-synuclein, but the exact cause remains unidentified (Dawson and Dawson, 2003; Greenamyre and Hastings, 2004). 

A smaller percentage (5-10%) of PD is genetic in origin, and disease onset is linked to genetic mutations. Familial PD arises from autosomal recessive mutations in Parkin, PINK1, and DJ-1 or autosomal dominant mutations in UCHL1, LRRK2, and SNCA (α-synuclein) (Leroy et al., 1998; Zimprich et al., 2004; Polymeropoulos et al., 1997; Kitada et al., 1998; Valente et al., 2004; Bonifati et al., 2003).  These genetic mutations, specifically the mutants in the α-synuclein gene, have provided the field with unique insight into the mechanism behind PD because mutations known to cause disease allow scientists to directly study a direct link to disease.

To date, three point mutations in the α-synuclein gene are known to cause familial PD.  Point mutations affect amino acids, the building blocks of proteins.  Amino acids are analogous to beads on a piece of string.  The sequence of beads determines whether or not the protein will acquire a correct shape (Alberts et al., 2009).  Each bead has unique properties, and some of these beads are more important to the shape of a protein than others.  In 1997, the first familial mutation in α-synuclein to cause PD was discovered in a family of Greek origin.  The mutation results in an alanine to threonine (A53T) substitution in α-synuclein at the 53rd amino acid (Polymeropoulous et al., 1997).  A year later, a second α-synuclein missense mutation was traced to a family in Germany.  In this instance, proline replaces an alanine at the 30th amino acid (A30P; Krueger et al., 1998).  The most recent α-synuclein mutant, discovered in 2004, results from a glutamic acid to lysine swap at the 46th amino acid (E46K; Zarranz et al, 2004).  All three of the familial mutations lead to an increase in α-synuclein misfolding, providing support for the link between misfolded protein and cellular toxicity (Figure 2B; Conway et al., 1998; Giasson et al., 1999; Conway., et al 2000).  In addition, familial PD is also caused by duplication or triplication of the α-synuclein gene, suggesting that one mechanism for the disease is over-expression of α-synuclein (Singleton et al., 2003; Chartier-Harlin et al., 2004).  These familial mutations and the presence of α-synuclein in LBs of sporadic PD patients strongly implicate α-synuclein in the development of PD.  My thesis sought to further understand α-synuclein’s pathotoxic properties in both familial and sporadic PD.

A Closer Look at the Culprit: The α-Synuclein Protein
α-Synuclein belongs to the synuclein family of proteins.  This small group of proteins contains two other members: β-synuclein, which is present in AD neurofibrillary lesions, and γ-synuclein, a protein closely linked to breast carcinoma progression (George, 2002; Bruening et al., 2000).  Both β-synuclein and γ-synuclein differ from α-synuclein in that they do not cause PD (Uverskey et al., 2008).  Interestingly, expression of β-synuclein in a mouse model actually protects from α-Synuclein induced toxicity (Fan et al., 2006).  The exact functions of the synuclein proteins are not currently known (Uverskey et al., 2006).

α-Synuclein is a short, highly flexible protein found throughout the brain (Uéda et al., 1993; Jakes et al., 1994 ;Weinreb et al., 1996). The protein was initially described in relation to AD, as the non amyloid component (NAC) peptide of α-synuclein consistently localized to AD plaques.  Interestingly, α-synuclein is also found associated with tubulin, a similarity it shares with tau protein (Alim et al., 2004).

Figure 2: α-Synuclein and Cell Death (A) Diagram demonstrating where cell death occurs in Parkinson’s disease (PD).  The substantia nigra is located in the midbrain.  A horizontal cut through this structure reveals the pigmented cells of the substantia nigra.  Parkinson’s diseased brains are characterized by a loss of pigmented neurons in the substantia nigra.  In addition, accumulations of α-synuclein, termed Lewy Bodies, are visible in the dying cells when examined under a light microscope. (B) Flow chart demonstrating the similarities and differences of sporadic and familial forms of PD.  In sporadic PD, wild-type α-synuclein misfolds and aggregates.  Genetic mutations are responsible for the onset of familial PD.  In either case, it is unclear whether the misfolded α-synuclein or a toxic intermediate is responsible for disease onset.
Brain image from http://www.paulnussbaum.com/. Substantia nigra image from http://www.urmc.rochester.edu/neuroslides/slides/slide199.jpg. Lewy body image from http://www.saigata-n.go.jp/saigata/rinken/neuropat/library/SN295LEWYSYNUCLX100.JPGLewy body image from http://neuropathology.neoucom.edu/chapter9/images9/9-lb2.jpg

α-Synuclein primarily localizes to pre-synaptic terminals of dopaminergic neurons (neurons that release dopamine).  Dopamine is released from these neuron in vesicles, structures that can be thought of as transport containers, and α-synuclein is thought to be involved in vesicle trafficking and dopamine release (Maroteaux and Scheller, 1991; Cabin et al., 2002).  α-Synuclein also appears to be involved in endocytosis, a mechanism that moves vesicles throughout cells, as expression of either WT or A53T disrupts endoplasmic reticulum (ER) to Golgi transport and enhances ER stress in a yeast model (Cooper et al., 2006).

While α-synuclein is normally a cytoplasmic protein, it also binds to phospholipid membranes (Clayton and George, 1998).  This interaction is mediated by several KTKEGV motifs in α-synuclein amphipathic N-terminal domain (Soper et al., 2008).  Additionally, α-synuclein has a natural tendency to aggregate due to its flexible structure.  α-Synuclein aggregation occurs in a stepwise manner (Caughey et al., 2003).  α-Synuclein monomers (single molecules) misfold and begin to coalesce into spherical protofibrils (short chains of α-synuclein).  These protofibrils link together into longer chains, forming α-synuclein oligomers (Giasson et al., 1999).  Subsequently, rapid fibril (or aggregate) formation occurs after the oligomers appear, resulting in the LBs visible in dying SNpc cells (Caughey et al., 2003).  Both the domains of α-synuclein and key amino acids have been implicated in the protein’s propensity to misfold and aggregate, but their exact contribution is still being evaluated (Baba et al., 1998; Soper et al., 2008).  Exactly how α-synuclein contributes to toxicity and whether LBs are the toxic mechanism in PD is unclear. 

Several hypotheses exist in the field as to how these LBs relate to neurotoxicity (Caughey et al., 2003).  Over the past decade, support has increased for the idea that intermediate protofibrils, which precede LB formation, are responsible for toxicity in PD.  While all three α-synuclein familial mutations enhance fibril formation, the A30P familial mutation specifically increases formation of protofibrils (Conway et al., 2000; Li et al., 2001).  Electron microscopy revealed that α-synuclein protofibrils form pore-like structures in cell membranes, and these protofibrils can permeabilize synthetic membranes (Lashuel et al., 2000; Volles and Lansbury, 2002). Furthermore, α-synuclein’s affinity for phospholipids is enhanced by protofibrils but not by fibrils (Volles and Lansbury, 2002; Ding et al., 2002). Thus, LB formation might be a protective response by the cells since fibril formation reduces the concentration of protofibril intermediates (Caughey et al., 2003). The properties of A53T and A30P familial mutants are well established, but the relationship between E46K and the impact it has on α-synuclein’s aggregation is still being examined. α-Synuclein is also heavily modified by covalent bonds on specific amino acids in LBs.  These modifications include ubiquitination (Shimura et al., 2001), glycosylation (Shimura et al., 2001), nitration (Hodara et al., 2004), and phosphorylation (Okochi et al., 2000; Fujiwara et al., 2002), but the role these modifications play in cellular death and aggregation is not fully understood and are all key research areas in PD (Sharon et al., 2001).

Yeast PD Models: A Research Alternative
Numerous models exist to study PD, including cell culture systems, C. elegans, drosophila, rat, and mice (Nass and Prezdborski, 2008).  Neuronal culture systems provide direct evidence from living neurons, but they are fragile systems and cell culture reprograms their apoptotic and senescence pathways (Nass and Prezdborski, 2008).  Both C. elegans and Drosophila models provide easily manipulable genetic systems, but genetic screens in these organisms are impractical due to the wide array of assays necessary (Nass and Prezdborski, 2008).  While mammalian models provide terrific insight into the disease process in a species closely related to humans, genetic experiments are costly and difficult to perform (Nass and Prezdborski, 2008).  Although they do not model the exact environmental conditions of a neuron, yeast provides an attractive alternative for PD research (Outiero and Lindquist, 2003; Brandis et al., 2006; Sharma et al., 2006). 

For my thesis, I utilized two yeast PD models: budding yeast (Saccharomyces cerevisiae) and fission yeast (Schizosaccharomyces pombe).  Both budding and fission yeast genomes share high homology with many human genes.  In addition, these eukaryotic fungi synthesize, fold, and degrade proteins similar to humans, providing human relevance for discoveries using yeast (Nass and Prezdborski, 2008).  Practically, they are cheap, reproduce rapidly, and their entire genome is sequenced.  Genetic knockouts of any non-essential genes are available in budding yeast, and a similar database currently in development for fission yeast (Nass and Prezdborski, 2008).

Budding yeast has a rich history of modeling human diseases, including cancer, mitochondrial disorders, and neurological diseases (Nass and Prezdborski, 2008).  More relevant to my research, budding yeast serve as effective models for several neurodegenerative diseases, including prion diseases (Ma and Lindquist, 2002), AD (Komano et al., 1998), HD (Meriin et al., 2002), and ALS (Corson et al., 1998).  In the PD field, a number of successful budding yeast model systems have been established since 2004 (Outeiro and Lindquist, 2003; Willingham et al., 2003; Dixon et al., 2005; Zabrocki et al., 2005; Sharma et al., 2006).  Yeast do not normally make α-synuclein,  so researchers use standard recombinant DNA technology to synthesize the protein in yeast.  Susan Lindquist’s lab developed the first budding yeast model in 2003.  Their landmark paper clearly demonstrated a dose-dependent toxicity when yeast expressed α-synuclein.  A single copy of the α-synuclein gene resulted in no toxicity and α-synuclein plasma membrane localization while two copies imparted significant toxicity and caused aggregation (Outiero and Lindquist, 2003).  While the aggregates appeared reminiscent of classical LBs, future studies that visualized these aggregates in yeast suggested that at least some of the LBs are actually accumulations of α-synuclein associated with vesicles (Soper et al., 2008).  α-Synuclein also disrupts ER-to-Golgi trafficking in yeast, a defect that is rescued by Rab1, providing additional evidence that the aggregates are vesicular in nature (Cooper et al., 2006).  Thus, aggregates in yeast might be vesicles rather than true aggregates.

Our lab induces human α-synuclein in yeast using using methods similar to the previously mentioned studies.  In budding yeast, WT and A53T α-synuclein localize primarily to the plasma membrane (Figure 3; Sharma et al., 2006).  In contrast, A30P α-synuclein is typically diffuse throughout the cytoplasm, correlating well with previous research suggesting the proline substitution alters the interaction of the N-terminus with phospholipids (Sharma et al., 2006; Jensen et al., 1998).  However, unlike the Lindquist lab, neither WT α-synuclein nor the familial mutants induce toxicity in budding yeast, which is most likely a result of a moderate expression levels (Sharma et al., 2006; Outiero and Lindquist, 2003).  Interestingly, yeast with a genetically compromised proteasome (the cell’s protein recycling center), exhibit significant toxicity when they express α-synuclein (Sharma et al., 2006).  However, deletion of the numerous E1, E2, and E3 ubiquitin ligase genes (genes that mark proteins for degradation) does not affect toxicity (Herrera Thesis, 2005).  However, deletion of manganese superoxide dismutase (SOD2; responsible for destroying damaging free radicals) and expression of α-synuclein coupled with a hydrogen peroxide challenge also proves lethal to yeast.  Interestingly, when several other oxidative stress pathway genes are deleted, no toxicity is seen (Brandis Thesis, 2005; Sharma et al., 2006).  Because of their ability to recapitulate several important PD related properties, our lab has also used budding yeast to research the role of endocytosis (Ayala Thesis, 2009; Perez Thesis, 2010), autophagy (Choi Thesis, 2009) and lipid synthesis (Kukreja Thesis, 2007), in α-synuclein biology.

In 2006, our lab also pioneered a fission yeast model to study α-synuclein to determine if α-synuclein behaves similar in another yeast species (Figure 3).  Using different strength promoters, our lab demonstrated that α-synuclein aggregates in a concentration-dependent manner in fission yeast (Brandis et al., 2006).

Figure 3: The DebBurman Lab Yeast Models (A) Diagram explaining budding and fission yeast models used by the DebBurman lab.  Our expression system centers on a plasmid vector containing a human α-synuclein gene fused to green fluorescent protein (GFP).  In budding yeast, expression is controlled by a galactose inducible promoter, and selection takes place using media lacking uracil.  In fission yeast, expression is controlled by a thiamine repressible promoter, and selection takes place using media lacking leucine.  WT α-synuclein associates with the plasma membrane in budding yeast.  In fission yeast, WT α-synuclein aggregates in the cytoplasm.

The familial mutant A53T aggregates more aggressively than WT, but neither WT nor A53T localize to the plasma membrane (Brandis et al., 2006).  In addition, despite extensive aggregation, no toxicity was seen with any α-synuclein variants (Brandis et al., 2006).  This finding might suggest a protective role of aggregation, but it is not yet clear if the aggregates observed are true aggregates or accumulations of vesicular structures associated with α-synuclein.

My Focus
Several key questions in the PD field remain unsolved.  1)  How is α-synuclein involved in the oxidative damage present in afflicted neurons?  2)  By what route is α-synuclein degraded?  3)  What role does α-synuclein have in neuronal death outside of the SNpc?  4)  What are the properties of the E46K familial mutant in living systems?  5)  What role does covalent modification of α-synuclein have in disease pathogenesis?  6)  Which structural aspects of α-synuclein contribute to its aggregation and membrane binding properties?  My thesis will focus on the last three questions, and I will now provide more background for each question I address.

The Lonely Mutant: E46K
Despite its discovery nearly six years ago, surprisingly little is known about the E46K familial α-synuclein mutant.  Discovered in a Spanish family in 2004, E46K is an autosomal dominant mutation (one mutated copy will cause PD) in the α-synuclein gene (Zarranz et al., 2004).  Following its discovery, a group from Cambridge examined the E46K mutant in vitro and compared its properties to the A30P and A53T familial mutants.  Two important observations were made.  First, E46K enhanced α-synuclein binding to liposomes in a lipid binding assay by a factor of two compared to WT, A30P, and A53T.  Second, E46K increased the formation of fibrils at a rate similar to A53T in a concentration-dependent manner.  Interestingly, the E46K fibrils were more tightly twisted than A53T (Choi et al., 2004; Greenbaum et al., 2005). 

Figure 4: Properties of α-Synuclein (A) Diagram of α-synuclein demonstrating the location of serine-129 and serine-87, two primary sites of phosphorylation in Lewy Bodies.  In vitro experiments demonstrated that serine phosphorylation enhances aggregation of α-synuclein.  However, there is conflicting evidence as to the toxic nature of serine phosphorylation.  (B)  Diagram of α-synuclein demonstrating location of the E46K familial mutation.  The mutation occurs when a glutamic acid (E) is mutated to lysine (K) due to a genetic mutation in the α-synuclein gene.  The E46K mutant was shown to enhance membrane interaction and aggregation of α-synuclein in vitro.  (C)  Diagram of α-synuclein demonstrating the location of alanine-76 within the NAC domain of the protein.  Alanine-76 was predicted to govern the hydrophobic properties of the protein, and the A76E and A76R mutations decreased aggregation of α-synuclein in vitro.

Most recently, researchers investigated how the E46K mutation alters the conformation (shape) of α-synuclein.  α-Synuclein is composed of three domains (regions):  the N (aa1-57), NAC (aa61-95), and C (aa96-140) domains (Soper et al., 2008; Giasson et al., 2001).  Previous studies demonstrate that A53T and A30P decrease the interaction between the N- and C-terminal regions of the protein (Bertonici et al., 2005).  Disrupting the interaction between those two domains was suggested to contribute to the aggregation of these mutants in vitro.  In contrast to A30P and A53T, the E46K mutant enhanced contact between the N- and C-terminus, suggesting that the interaction with the C-terminus does not play a role in the propensity of the protein to aggregate since A53T aggregates similarly to WT without enhancing this interaction (Rospigliosi et al., 2009).  The properties of the E46K mutant still require further elucidation (Figure 4A).  The first goal of my thesis was to describe the properties of the E46K mutation in our budding and fission yeast models.

The Mystery of Serine Phosphorylation
Phosphorylation involves the covalent addition of a phosphate group to a protein. This modification typically alters a protein’s conformation, which is useful in regulating the on or off state of an enzyme (Alberts et al., 2008).  The α-synuclein located in LBs is heavily phosphorylated at two serine residues: serine-129 and, to a lesser extent, serine-87 (Fujiwara et al., 2002; Palelogou et al., 2010).  Several enzymes that phosphorylate proteins (called kinases) demonstrate the ability to phosphorylate α-synuclein in cell culture, including Lrrk2 (Chen and Feany, 2005), Gprk2 (Okochi et al., 2000), casein kinase 1 and 2 (Kim et al., 2005), and Dyrk1A (Sakamoto et al., 2009), but the kinase responsible for phosphorylation of α-synuclein in the LBs of PD patients has yet to be identified.  In vitro fibrillization assays revealed that phosphorylation of α-synuclein enhances fibril formation (Fujiwara et al., 2002). α-Synuclein in LBs is also phosphorylated at several tyrosine residues, and research suggests tyrosine phosphorylation appears influential in preventing aggregation (Nakamura et al., 2001; Ellis et al., 2001; Chen et al., 2009).

Conflicting evidence exists in the field as to what role serine phosphorylation has in dopaminergic cell death.  In 2005, the Feany lab demonstrated that an α-synuclein mutant that blocks phosphorylation at serine-129 through mutation to an alanine (S129A) attenuates toxicity in a Drosophila model (Chen and Feany, 2005).  An α-synuclein phosphorylation-mimic mutant (S129D) enhanced toxicity in their fly model.  The S129A mutation also increased inclusion formation, suggesting that LBs might be neuroprotective (Chen and Feany, 2005).  However, a lab using a rat model to study phosphorylation found S129A to be extremely toxic as compared to wild-type (WT) and S129D synuclein.  The S129D mutant increased inclusion formation in SNpc neurons (Gorbatyuk et al., 2008).  A second paper published in 2009 also saw increased toxicity caused by S129A expression in a rat model (da Silveira et al., 2009).  Further complicating the issue is a recent study by the same authors of the Drosophilastudy.  They found no difference in toxicity between rats expressing WT, S129A, and S129D α-synuclein (McFarland et al., 2009).  Thus, the role that phosphorylation plays in toxicity is still unclear and may be organism and cell-dependent (Figure 4B).  Describing the properties in yeast provides an additional organism to compare results in the field with.  The second goal of my thesis is to provide insight further into the role that α-synuclein serine phosphorylation has in PD pathogenesis by describing the properties of several phosphorylation mutants in our budding and fission yeast models.

The Role of α-Synuclein’s Hydrophobic Amino Acids
The N-terminus has been implicated in α-synuclein’s ability to bind membranes due to its flexible nature and the presence of several imperfect KTKEGV amino acid repeats (Davidson et al., 1998; Perrin et al., 2000; Kim et al., 2006).  Upon binding to membranes, the N-terminus shifts from an α-helical to a β-sheet conformation (Kim et al., 2006.)  The A30P familial mutation, which occurs in repeat two, was shown to disrupt membrane binding in vitro,demonstrating how essential these repeats in the N-domain are for lipid interaction (Jensen et al., 1998). More recently, a 2008 study evaluated an α-synuclein N-terminal truncation mutant in a yeast model and found that the mutant failed to bind to plasma membranes, a well documented property of α-synuclein in budding yeast  (Soper et al., 2008; Sharma et al., 2006; Outeiro and Lindquist, 2003; Dixon et al., 2005; Zabrocki et al., 2005).  These findings suggest that the proteins domains contribute significantly to α-synuclein’s properties.

The NAC domain, also called the non-β-amyloid component, is a hydrophobic domain essential for aggregation and fibrillization both in vitro and in vivo (Giasson et al., 2001; Periquet et al., 2007).  Within the NAC domain is a region of hydrophobic (water hating) residues from aa71-82 that was shown to be crucial for α-synuclein to aggregate (Giasson et al., 2001).  On its own, the NAC domain is highly amyloidogenic and induces toxicity when expressed in PC12 (rat adrenal) and SHSY-5Y cells (neuroblastoma cancer cells) (Han et al., 1995; El-Agnaf et al., 1998a; El-Agnaf et al., 1998b; Bodles et al., 2001).  In addition, β-synuclein, which does not aggregate as α-synuclein does, lacks the NAC domain (Biere et al., 2000).  Addition of the NAC domain to β-synuclein is sufficient to induce aggregation of the protein (Biere et al., 2000).  Soper et al. 2008) also demonstrated that, in yeast expressing a NAC truncation mutant, cytoplasmic aggregates failed to form.

In light of the controversy regarding the potentially toxic role of protein aggregation in PD, understanding how α-synuclein acquires its shape is an important question.  In 2003, researchers modeled α-synuclein aggregation in relation to hydrophobicity, charge, and tendency to convert from α-helix to β-sheet (Chiti et al., 2003).  They predicted that alanine-76 within the NAC domain would be a key contributor to α-synuclein’s hydrophobic properties because the A76R and A76E mutants would significantly alter the hydrophobicity of the polypeptide chain (Chiti et al., 2003).  In addition, in vitro analysis of A76E and A76R indicated that these two mutants aggregate more slowly than WT α-synuclein (Giasson et al., 2001).  Thus, alanine-76 appears potentially relevant because it affects a pathology-linked property: protein aggregation (Figure 4C).  However, no PD patients actually have an A76E or A76R mutation.  A third goal of my thesis is to assess the relevance of alanine-76 using our budding and fission yeast models.