Introduction

Mutations Drive Evolution and Disease Alike

Each and every one of the trillions of cells in our bodies is equipped with the instructions necessary to regulate and maintain life. These instructions manifest themselves in the form of our DNA. Although our DNA plays such a pivotal role, it is still susceptible to change. These changes to the DNA are called mutations. While the word mutation may have a negative connotation, mutations are not intrinsically bad. Mutations actually drive evolution: they have allowed us to survive and reproduce.

Benefits resulting from mutations include resistance or a lower risk of developing diseases such as HIV and AIDS (Cohen et al., 2008), the ability to see a wider range of colors (Neitz and Neitz, 2011), and an increase in bone density (Boyden et al., 2002). Changes in the DNA are also linked to many diseases. Cystic fibrosis, a disease of both the mucus and sweat glands, results from mutations in the CFTR gene (Straniero et al., 2016). Breast cancer is linked to mutations in the BRCA1 or BRCA2 genes (Pearson et al., 2017). The list of diseases with their corresponding mutations is a long one that extends even further.

When the Loss of the Original Amino Acid is Key

A single amino acid mutation at the 6th position of hemoglobin from glutamine to valine causes sickle cell anemia, a blood disorder in which red blood cells are abnormally shaped and more likely to rupture (Bender and Douthitt, 2003). Two events happen in that change: the orig­inal amino acid coded for by the DNA is being lost and a new amino acid is being gained. Glutamine is a polar amino acid, which can participate in hydrogen bonding. Valine, however, cannot participate in hydrogen bonding because it has no partially negatively charged atom that can in­teract with hydrogens. The loss of the original amino acid and its ability to hydrogen bond is ultimately responsible for the disastrous, conformational change in hemoglobin’s structure, and the disease itself (Figure 1). A sin­gle amino acid mutation at the 173rd position of a different protein, ApoA- 1 Milano, is critical in cardiovascular disease (Eriksson et al., 2009). The positively charged amino acid arginine is changed to the neutral cysteine. Because the positive charge of arginine contributes to ApoA-1 Milano’s ability to bind lipids, its loss transforms the secondary structure of the pro­tein. This loss increases the transportation of cholesterol from the tissues to the liver, granting protection against cardiovascular disease, as seen in Figure 1.

When the Gain of the New Amino Acid is Key

The loss of the original amino acid is not always the root of the positive or negative effects. A single amino acid mutation in the B-Raf pro­tein at the 600th position, which changes the nonpolar valine to the acidic, negatively charged glutamic acid, is the culprit in many cases of papillary thyroid carcinoma, colorectal cancer, melanoma, and non-small-cell lung cancer (Jiang et al., 2016). The gain of the glutamic acid is key because once it replaces valine, the kinase activity of B-Raf increases. This is seen in Figure 1. Blood disorders, heart conditions, and cancers are just a few examples of how a single change in an amino acid can produce drastic effects. Neurodegenerative diseases are no exception.

Neurodegenerative Diseases

As suggested by the name, neurodegenerative diseases involve the progressive death of neurons in the central nervous system (CNS). Different diseases are characterized by different neurons that die. In Alzheimer’s disease (AD), neurons in the region of the brain known as the hippocampus, which functions in memory, and the cortex die. In Hun­tington’s disease (HD), neurons in the striatum, which functions in control of motor movement, are the ones that die off.

Although the death of neurons can occur in many different parts of the brain, the things that are common among all of the neurodegenerative disorders are protein misfolding, aggregation, and accumulation. For my thesis, I will focus on Parkinson’s disease (PD) and the implicated protein, α-synuclein.

Parkinson’s Disease

PD is the second most common neurodegenerative disease after AD, and it affects more than 10 million people worldwide (Dorsey et al., 2007; de Silva et al., 2002). There are two types of symptoms of PD: those that are motor and those that are non-motor. The motor symptoms are more readily apparent and include bradykinesia (slowness of move­ment), resting tremors, and rigidity (Olanow and Tatton, 1999).

Although less apparent, non-motor symptoms still have a significant impact on PD patients and their everyday life. These include difficulty chewing and swallowing, reduced ability to smell and taste, constipation, bladder dysfunction, cognitive impairment, and sometimes dementia. While the appearance of both types of symptoms may seem indicative of the onset of PD, changes in the brain actually get underway long before symptoms. Specifically, in PD, the symptoms don’t appear until more than ninety percent of midbrain dopamine-producing neurons in the substantia nigra pars compacta (SNc) have died (Barbosa et al., 1997). The SNc is part of the brain known as the basal ganglia, which is the motor circuit that controls the initiation of voluntary movement. The SNc neurons nor­mally release the neurotransmitter dopamine (DA), which both promotes voluntary movement and inhibits unnecessary involuntary movement (van Raaij et al., 2006; Squire et al., 2012). The cell death leads to reduced dopamine release overall, which inhibits the output of the basal ganglia to control the timing of voluntary movement and the motor symptoms mentioned above begin to appear, as seen in Figure 2.

Two Causes: Genetic and Sporadic

PD can be caused by heritable genetic mutations or envi­ronmental factors. These represent the two types of PD, familial and sporadic, respectively. About ten percent of cases are familial, while the other ninety percent are sporadic. Both, however, are characterized by brains that are ridden with Lewy bodies: thick, dark deposits of misfolded, aggregated, and accumulated protein (Pollanen et al., 1993; Caughey and Lansbury, 2003). α-Synuclein is a major component of these Lewy bodies (Da Costa et al., 2000).

Sporadic PD is linked to exposure to pesticides (Ashcerio et al. 2006), exposure to heavy metals (Calne et al., 1994), mitochondrial dysfunction (Langston et al. 1983; Langston et al., 1984), and oxidative stress (Jenner and Olanow, 1996; Maguire-Zeiss et al., 2005). Sporadic PD can be stud­ied in the lab, however the study of familial PD is more well-established.

Familial PD is linked to mutations in at least eight genes. The mutations fall into one of two inheritance patterns: autosomal dominant or autosomal recessive. The genes that give rise to PD through autosomal dominant mutations are UCHL1 (Leroy et al., 1998), SNCA (Polymero­poulos et al., 1997), and LRRK2 (Funayama et al., 2002). These genes lead to disease because the mutations result in a gain of function of toxic properties. The genes that give rise to PD through autosomal recessive mutations are Parkin (Mizuno et al., 2001), PINK1 (Valente et al., 2004), ATP13A2 (Najim al-Din et al., 1994), and DJ-1 (Bonifati et al., 2003). These genes cause disease because they result in a loss of normal func­tion. LRRK2 codes for a protein that is involved in intracellular signaling and kinase activity, and a mutation in LRRK2 increases kinase activity. LRRK2 also regulates the mitochondria, as do PINK1 and Parkin, and mutations in all three result in mitochondrial and oxidative stress (Norris et al., 2004). PINK1 also normally plays a role protecting the mitochon­dria during cellular stress, as does DJ-1, and mutations in both put the mitochondria at higher risk for damage and disease (Deng et al., 2008; Yang et al., 2005). The SNCA gene codes for a protein called α-synuclein and point mutations in α-synuclein are associated with its misfolding, ag­gregation, and accumulation. In addition to genetic mutations that cause PD, genetic risk factors pose a risk of developing PD. Almost 30 risk factors have been identified to date, but there are likely many more that have not been discovered yet and this illustrates just how little we know about the genetics underlying PD. A few examples of these risk factors include mutations in GBA, RAB7L1, BST1, HLA-DRB5, and VPS35 genes (Brás et al., 2015). My thesis focuses on the α- synuclein gene, SNCA, in particular and how the point mutations affect various properties of the α- synuclein protein and on the VPS35 risk factor.

α-Synuclein

α-Synuclein is a relatively small protein, which is only 140 amino acids in length. It has three domains: the N-, M-, and C- domains, each linked to different α- synuclein properties. The N- domain contains amino acids 1-60 and promotes lipid binding (Chandra et al., 2003). The M- domain contains amino acids 61-95 and is required for α-synuclein to form fibrils. The C-domain contains amino acids 96-140 and is related to the solubility and flexibility of α-synuclein (Bodles et al., 2001; Giasson et al., 2001; Lücking and Brice, 2000). α-Synuclein can also undergo many post-translational modifications: SUMOylation which appears protective (Krumova et al., 2011), nitration which appears toxic (Sevcsik et al., 2011), and phosphorylation which also appears to be toxic (Oueslati et al., 2012).

The role of α-synuclein in a healthy cell is still being elucidated. Researchers believe it plays an important role in neurotransmission by aiding in vesicle release (Davidson et al., 1998; Liu et al., 2004). It is also implicated in the process of endocytosis, which brings nutrients into the cell and recycles proteins (Kuwahara et al., 2008; Ben-Gedalya et al., 2009) and the transport of proteins from the endoplasmic reticulum (ER) to the Gogi body (Cooper et al., 2006; Chua and Tang, 2006; Gitler et al., 2008). α-Synuclein interacts with other proteins. It binds microtubules (Alim et al., 2004) and actin fibers, which both function in cell movement (Sousa et al., 2009).

Therefore, although we might not have a comprehensive understanding of α-synuclein’s normal function, what is clear is that it is involved in a variety of processes required for cell functioning. When α-synuclein becomes misfolded, all of these basic processes thus become disrupted, ultimately leading to cell death. However, the mechanism by which these cells die is not clearly understood. The genetic mutations in α-synucle­in have been and continue to be studied in attempt to understand this mechanism.

Six Familial Mutations of α-Synuclein

There are six known familial mutations on α-synuclein that are linked to early onset, aggressive PD and they are all curiously located on the membrane binding- involved N-domain (Figure 3). Three of these mutations (A30P, E46K, A53T) were discovered earlier and thus are well studied in multiple model organisms and are well characterized as a whole. I will refer to these as the old familial mutants. The other three mutations (H50Q, G51D, A53E) were discovered more recently – within the last four years - so comparatively much less is known about them. I will refer to these as the new familial mutants. Our lab studies all familial mutants, old and new, in relation to the wildtype (WT) form of α-synuclein that most people have. I focus on the new familial mutants in my thesis because we currently know less about them. Early research that has been done suggests that each new familial mutant has its own distinct phenotype, which is the case with the old familial mutants as explained below.

Well-Understood Older Familial Mutants

A53T

The A53T mutation was the first α-synuclein mutation dis­covered in 1997 and thus is the most well studied of all of the familial mutants, new and old. It was originally discovered in a Greek family. Thereafter, A53T was also discovered in three unrelated Greek fami­lies (Polymeropoulos et al., 1997). The mutation replaces the nonpolar alanine with the uncharged polar threonine at the 53rd position (Alberts et al., 2009). Early in vitro studies demonstrated that A53T showed different properties depending on the concentration. At low concentrations, it was disordered like WT, but at high concentrations, it formed spherical assemblies very quickly (Conway et al., 1998). In vivo studies provided more insight. In a transgenic mouse model, A53T was shown to bind to membranes just like WT (Sharon et al., 2003). This membrane binding in A53T and WT was also shown in budding yeast; however, in fission yeast formation of aggregates was observed instead for A53T and WT and this was determined to be toxic (Sharma et al., 2006; Brandis et al., 2006). More recently, A53T is being studied in the context of oligomerization. This is because oligomers, the preliminary species to aggregates, are believed to be key players in PD pathology.

A53T appears to form oligomers that are structurally different than WT (Tosatto et al., 2015), which suggests oligomer structure may be more important than concentration in leading to pathology. Recent work also suggests heat shock protein responses and autophagy may be able to alleviate A53T toxicity (Xiong et al., 2015).

A30P

After the A53T mutation, the A30P mutation was discovered in a British family in 1998. The mutation replaces the nonpolar alanine for the nonpolar proline at the 30th site. Thus, the overall charge of the amino acid is maintained (Krüger et al., 1998) While A53T showed mem­brane binding, A30P did not (Jensen et al., 1998). Structural studies pro­vided further insight into this. They revealed that A30P ultimately changed the structure of α-synuclein by preventing it from forming helices, which are involved in membrane binding ability (McLean et al., 2000; Jo et al., 2002). In vitro experiments revealed that at low concentrations A30P was disordered like WT, but at high concentrations it formed spherical particles (Conway et al., 1998). Work with A30P in budding and fission yeast also showed lack of membrane binding, consistent with previous research, and instead showed cytoplasmic diffusion (Sharma et al., 2006; Brandis et al., 2006; Fares et al., 2014). In more recent work on oligomerization, A30P, like A53T, forms oligomers that are structurally different than WT and this may be related to its toxicity (Tosatto et al., 2015). A30P also formed the most dimers in vitro compared to the other old mutants, which suggests that it aggregates differently than the rest (Lv et al., 2015).

E46K

The E46K mutant was discovered the last of all the old familial mutants in 2004 in a Spanish family and the mutant replaces the nega­tively charged glutamic acid for the positively charged lysine at the 46th site (Zarranz et al., 2004). Like the other old mutants, E46K has been studied both in vitro and in vivo. Collectively, those studies found that E46K distinctly bound liposomes twice as much as WT, A53T, and A30P, had the ability to form fibrils as fast as A53T, and also formed aggregates (Choi et al., 2004; Greenbaum et al., 2005). This unique phenotype is attributed to increased N- and C- domain interactions that are not present in either of the other old familial mutants (Bertoncini et al., 2005; van Raaij et al., 2006). In budding yeast, E46K was shown to bind lipids, while in fission yeast it was shown to form aggregates and cause toxicity (Fiske et al., 2011b). Recent work points to E46K as the only old mutant that impairs macroautophagy (Yan et al., 2014) and the only mutant that sig­nificantly enhances α-synuclein phosphorylation at the S-129 site (Mbefo et al., 2015). Thus, collectively from the older mutants we know that although they are all linked to early onset, aggressive PD, each one has a distinctive phenotype and is toxic in its own way. Now we look to the more recently discovered new familial mutants.

Less Characterized New Familial Mutants

H50Q

The H50Q mutation was discovered in an English patient, fol­lowed by a Canadian patient in 2013 (Proukakis et al., 2013a; Proukakis et al., 2013b; Appel- Cresswell et al., 2013). The mutation replaces the positive, basic amino acid histidine with the polar amino acid glutamine at the 50th site. So far, the H50Q mutation has been found to increase α-synuclein aggregation and toxicity and to be secreted extracellularly (Khalaf et al., 2014). The way that H50Q oligomerizes has been inter­preted to be quite similar to the way WT α-synuclein does, although the exact mechanism is still being studied (Ghosh et al., 2013). Work in yeast has provided some more insight, although different properties emerged in different types of yeast. In budding yeast, H50Q was found to form aggre­gates, but then go to the membrane over time like WT, and it was found to be more toxic than WT at low levels of expression. In fission yeast, H50Q binds the endomembrane more than WT and is comparatively less toxic (Tembo et al., manuscript in prep).

G51D

The G51D mutation was discovered in a British patient also in 2013. The mutation replaces the nonpolar amino acid glycine for the acid­ic aspartic acid at the 51st position (Lesage et al., 2013). G51D actually has one of the earliest onsets and most rapid progressions of PD: pa­tients with this mutation usually die within five to seven years of disease onset (Tokutake et al., 2013, Lesage et al., 2013). Furthermore, G51D is also associated with multiple system atrophy (MSA), as some British G51D PD patients had inclusions indicative of MSA (Kiely et al., 2013). Recent work has suggested that G51D promotes cell toxicity, but actually attenuates α-synuclein membrane binding and aggregation in mammalian cells and primary neurons (Fares et al., 2014). This observed attenuation is curious given what we see in patients and needs to be investigated further. Work in budding yeast has suggested that G51D binds the mem­brane less than WT in favor of a cytoplasmically diffuse phenotype and is actually less toxic than WT, and work in fission yeast has also suggested less toxicity compared to WT as well as less aggregation (Tembo et al., manuscript in prep).

A53E

The A53E mutation was discovered most recently in 2014 in a Dutch patient and her family and it is the least well-studied of all of the familial mutants. The patient’s brain pathology extended beyond the SNc. It contained cytoplasmic inclusions as well as Lewy body inclusions char­acteristic of MSA and dementia with Lewy bodies (Pasenen et al., 2014). The mutation replaces the hydrophobic amino acid alanine with

the acidic glutamic acid at the 53rd position. Work thus far suggests that A53E decreases α-synuclein aggregation and membrane binding in vitro (Ghosh et al., 2014). In budding yeast, A53E was found to bind the mem­brane less than WT and grow at the same rate comparatively while in fission yeast, A53E was found to form less aggregates than WT (Tembo et al., manuscript in prep).

Other Cellular Disruptions in PD

In addition to the α-synuclein misfolding and aggregation that I have focused on thus far, there are also many other disruptions that occur inside the cell in PD. Oxidative and nitrative stress are elevated- this is manifested in an excess production of reactive oxygen and nitrogen species (Giasson et al., 2000; Giasson et al., 2002; Danielson and An­dersen, 2008). The mitochondria, which is involved in energy production, is dysfunctional and mitochondrial dysfunction itself also contributes to the generation of reactive oxygen species, which suggests one cellular disruption can fuel another (Lin and Beal, 2006; Morais et al., 2014). Neu­rons show calcium dysfunction (Sheehan et al., 1997; Chan et al., 2007) and disruptions in the trafficking of proteins between the ER and the Golgi, which is followed by defective endocytosis or recycling of proteins (Cooper et al., 2006; Lashuel and Hirling, 2006; Su et al., 2010). The process of autophagy, which degrades unwanted cellular debris, is also dysfunctional (Anglade et al., 1997; Ventruti and Cuervo, 2007). My thesis focuses on two of the PD- related cellular disruptions, which are easy to look at in yeast: nitrative stress and defective endocytosis. In doing so, I hope to gain a better understanding of the relationship between α-synu­clein pathology and disruptions in the larger context of the cell.

Nitrative Stress

Nitrative stress is observed in the dying neurons (Giasson et al., 2002; Danielson and Andersen, 2008; Hartmann, 2004) and cerebro­spinal fluid (CSF) of PD patients (Fernández et al., 2013). In vitro work and in vivo work with many model systems from yeast to mice have yield­ed more insight into the effect of this nitrative stress. One in vitro study showed that exposing α-synuclein to nitrogen species induces the for­mation of nitrated α-synuclein oligomers, which suggests nitrative stress is involved in the pathogenesis of Lewy bodies (Souza et al., 2000). A few years later, the same lab showed that exposure to nitrogen species is also associated with the cross- linking within α-synuclein, suggesting nitrative stress stabilizes aggregated species (Norris et al., 2003). Work in yeast has shown that increasing nitrative stress by knocking out the cytochrome c oxidase subunit 5a (COX5A) protein, which normally de­creases nitrative stress, increases toxicity, and decreasing nitrative stress by knocking out the cytochrome c oxidase subunit 5b (COX5B) protein, which normally increases nitrative stress, decreases toxicity (Chung et al., 2013). Additionally, the A53T familial mutant worsens nitrative stress and is more toxic under conditions of nitrative stress compared to WT α-sy­nuclein in vitro (Paxinou et al., 2001). The A30P familial mutant has also been implicated in nitrative stress. Studies have reported A30P shows increased aggregation under conditions of nitrative stress in mouse models (Unal-Cevik et al., 2011) and in yeast (Kleinknecht et al., 2016). It is not known how the toxicity of the new familial mutants are altered when subjected to different levels of nitrative stress and this is why I want to explore this as part of my thesis.

Endocytosis

Endocytosis was first implicated in PD in yeast models (Outerio and Lindquist, 2003). Since then, more research has been done in vitro and in other models (Shin et al., 2008; Nemani et al., 2010) as we know that PD patients show less endocytosis (Alieva et al., 2014; Dijkstra et al., 2015). Mouse models show defective endocytosis just like in humans (Burré et al., 2010). Additionally, although there are many endocytosis-as­sociated proteins, previous work in yeast shows that mimicking defective endocytosis by knocking out vacuolar protein sorting-associated protein 28 homolog (VPS28) in particular results in more aggregation in terms of localization (Price and Shrestha, 2005) and more toxicity (Willingham et al., 2003; Price and Shrestha, 2005). Knocking out another endocytosis protein, vacuolar protein sorting-associated protein 35 homolog (VPS35); however, does not appear to have the same toxic effects (Dhungel et al., 2015). This suggests certain endocytosis proteins may be more vital to functional endocytosis than others and this warrants more research. In re­gard to the old familial mutants, it appears that A53T inhibits endocytosis in mammalian cell culture (Xu et al., 2016). However, it is not known how defective endocytosis affects the new familial mutants and this is why I want to investigate this in my thesis.

Gap in Knowledge

All of the familial mutants have been characterized to some de­gree. However, it is not known for any of the mutants what is responsible for toxicity: loss of the original amino acid or the gain of the new amino acid. Nature provides one clue. We already know that two of the six familial PD mutants, A53T and A53E, both involve the loss of the original amino acid alanine. This suggests that the loss of the original amino acid is important in causing disease. Additionally, work with H50Q has sug­gested that the loss of the original, positively charged histidine is key, but this work was done in vitro (Chi et al., 2014). We do not know if the same results would appear in living model organisms nor do we know anything about the other new familial mutants since they have not been studied in this regard as of yet. The new mutants have also not been studied in strains altered for nitrative stress or endocytosis, which are disrupted in PD. Since not much is known about the new mutants, my thesis will focus 1) on determining if it is the loss of the original or gain of the new amino acid that is responsible for observed toxicity in the new mutants and 2) on determining how the new mutants are affected by altered nitrative stress and defective endocytosis.

Although I am focusing on the new familial mutants, my work will shed light on all of the familial mutants of α-synuclein as they all in­volve the loss of an original amino acid and the gain of a new amino acid.

Yeast: A Powerful Model Organism

There are several model organisms that are used to study PD. These include primates, rats, mice, Drosophila, and C. elegans. Cell culture is also used. Each model has its own benefits and disadvantages which researchers weigh to pick which model is best for them. Mammali­an models seem like an obvious choice because they are the most close­ly related to humans. They also show the observable motor PD behaviors well; however, it is costly, difficult, and time consuming to manipulate them genetically (Nass and Przedborski, 2008). Although not as closely related to humans as mammalian models, Drosophila and C. elegans seem feasible to use because they are easy to genetically manipulate. Unfortunately, they also require a wide range of assays and so are impractical for our use (Feaney and Bender, 2000; Nass and Przedborski, 2008). Cell culture is another option. Although no living organism is used here, the neurons studied in cell culture are living, isolated systems. The difficulty here is that cell culture is extremely fragile (Nass and Przedbor­ski, 2008).

Our lab along with many others has opted for another unusual, simple, but powerful model organism in the study of PD: yeast. Although many may think yeast are a peculiar choice, yeast have proven to be useful in many PD studies (Outeiro and Lindquist, 2003; Brandis et al., 2006; Sharma et al., 2006). The choice of yeast is rooted in the fact that many genes and processes have been evolutionarily conserved from yeast to humans. Yeast and humans actually share the same genes for protein synthesis, folding, and degradation. Since PD is a protein-misfold­ing disease, the conservation of these genes makes yeast a fitting model to study. Furthermore, yeasts are cheap, they have a short lifespan, and reproduce rapidly (Nass and Przedborski, 2008). Taking all of these factors into consideration, yeast are more than a sensible choice of model organism to study PD.

One type of yeast, budding yeast, has been well-established as a model organism in the study of human diseases, specifically neurode­generative diseases which include Alzheimer’s disease (Komano et al., 1998), Huntington disease (Meriin et al., 2002), and ALS (Corson et al., 1998). Several labs have created budding yeast models, the first being Susan Lindquist’s lab in 2003 (Outeiro and Lindquist, 2003) followed by others (Willingham et al., 2003; Zabrocki et al., 2005; Sharma et al., 2006). Yeast does not naturally contain the α-synuclein gene, so researchers must introduce the α- synuclein gene into yeast. This can be done by inserting the α-synuclein gene into the yeast genome directly or into a plasmid vector. Once incorporated, yeast may then synthesize the α-synuclein protein. In terms of characterization, the Lindquist lab revealed that the toxicity caused by α-synuclein expression was dose-de­pendent - a single copy of the α-synuclein gene was not toxic whereas two copies were toxic and caused aggregation (Outeiro and Lindquist, 2003). Later it was shown that these aggregates were similar to Lewy bodies, as well as vesicles which are characteristic of human PD brains (Soper et al., 2008). The Lindquist lab and several other labs have also already shown yeast are well- established to study both nitrative stress and endocytosis disruptions (Outeiro and Lindquist, 2003; Chung et al., 2013; Dhungel et al., 2015).

I used two yeast models in my thesis: budding yeast, Sac­charomyces cerevisiae, and also fission yeast, Schizosaccharomyces pombe, to study the substitutions of the new familial mutants. Although they sound deceptively similar, budding yeast and fission yeast are quite distinct from one another and use of both is warranted. Not only do they reproduce by different means, but budding yeast also shows α-synuclein’s ability to bind membranes well while fission yeast highlights α-synuclein’s ability to form aggregates. Our lab has a rich history in using both of these yeast models. Since yeast do not contain the α-synuclein gene, our lab has used a high copy 2-micron vector to express α-synuclein in budding and fission yeast (Sharma et al., 2006; Brandis et al., 2006). Brandis et al. (2006), Fiske et al. (2011a), Fiske et al., (2011b), Natalie Kukulka (Thesis, 2013), and Katrina Campbell (Thesis, 2014) have all used both yeast models inserted with the α-synuclein gene to study the old familial mutants, A30P, E46K, and A53T. Likewise, Maiwase Tembo (Thesis, 2015) used the same models in her study of the new familial mutants, H50Q, G51D, and A53E, finding each new mutant to have a distinctive phenotype. Thus, the studies done with yeast models both in and outside of our lab have shown yeast to be a practical and powerful model for PD research. I chose to do my thesis with yeast models and my hypothesis and aims regarding the substitution mutants are explained below.

Hypothesis and Aims

Hypothesis

Firstly, the loss of the original amino acid characteristics is responsible for the distinctive phenotypes of each of the new familial mu­tants and secondly, the new familial mutants’ properties will be worsened by nitrative stress and defective endocytosis.

Aim 1

My first aim was to create four substitution mutations represent­ing the four functional classes of amino acids (polar, hydrophobic, acidic, basic) for H50Q, G51D, and A53E in both budding and fission yeast. Note that for amino acids, N/Q are polar; R is basic; D/E are acidic; A is hydro­phobic. An outline of all of the mutations can be seen in Figure 4. For my project, I was able to successfully create all substitution mutants in both budding and fission yeast, as described in chapter 1 of results.

Aim 2

My second aim was to examine localization, expression, and toxicity properties of substitution mutants compared to WT and the original familial mutant in budding and fission yeast. In order to acquire data on localization, expression, and toxicity, I used three assays: green fluorescent protein (GFP) microscopy, Western blot analysis, and serial dilution spotting. GFP microscopy allowed me to visualize where α- synu­ clein was located in each cell. I could compare the dominant phenotype for each substitution mutant to the dominant phenotype of WT α-synuclein and the original new familial mutant. Western analysis allowed me to see the level of α-synuclein expression across the substitution mutants, WT, and the original new familial mutant. Finally, serial dilution spotting allowed me to examine the amount of cell growth of the substitution mu­tants compared to WT and the original new familial mutant.

Western blot analysis was omitted for fission yeast substitution mutants due to technical issues and it was also omitted for new mutant PD-related disruptions, but in the latter case it was only in the interest of time.

One set of my predictions for localization, expression, and toxicity properties of the substitution mutants based on my hypothesis can be seen in Figure 5. The figure shows only my predictions for the H50Q substitutions, but I used the same logic to make predictions for all the other substitution mutants. I predicted that the substitution mutant that conserved the property of the original amino acid would show WT-like properties and the rest of the substitution mutants which lost the property of the original amino acid would have properties like the original new familial mutant. The results are described in chapter 3 of results. I found that each substitution mutant has its own unique characterization and the substitution mutants that exhibit diffuse localization phenotypes tend to be the least toxic, suggesting that the loss of the original and the gain of the new amino acid are both important.

Aim 3

My third aim was to examine the impact of the new familial mutants on the PD- related cellular disruptions of nitrative stress and endocytosis in budding yeast. vps28Δ and vps35Δ knockout strains were used to study properties when endosomal sorting proteins VPS28 and VPS35 are inactive. cox5aΔ and cox5bΔ strains were used to study increased and decreased nitrative stress respectively. My predictions for the properties in these strains can be seen in Figure 6. I predicted the properties would be worsened: more aggregation and toxicity. The results are described in chapter 4. I found that increasing nitrative stress increased toxicity for all new mutants, but only increased aggregation for A53E. With regard to endocytosis, I found that both vps28Δ and vps35Δ strains increased aggregation for all new mutants, but only the vps28Δ strain was toxic.

To view the whole text please visit www.lakeforest.edu/eukaryon.

Appendices

Figure 1: Amino Acids: Loss and Gain are Both Important in Nature

There are cases in which the loss of the original amino acid is key and in which the gain of the original amino acid is key in a given mutation. Two examples where the loss of the orginal amino acid is key in hemoglobin, where the loss of glutamine’s polar quality results in no hydrogen bonding and this changes the shape of hemo­ globin to a sickle shape, leading to Sickle Cell Anemia. Another example is the loss of the positively charged arginine in the ApoA-1 Milano protein, which decreases lipid binding, leading to an increase in transportation of cholesterol from the tissues to the liver. One example where the gain of the new amino acid is key is when the negatively charged glutamic acid replaces the nonpolar valine in the B-raf protein, a kinase protein, increasing activity in a signaling pathway which regulates cell proliferation. This is implicated in many types of cancer.

 

Figure 2: Death of Dopaminergic Neurons in the SNe leads to PD Motor Symp­toms Dopamine-producing neurons in the substantia nigra are a part of the motor circuit in the basal ganglia: dopamine release normally promotes voluntary move­ment and inhibits unnecessary involuntary movement. When over ninety percent of those dopamine-producing neurons die in PD, this overall decrease in dopamine release breaks the motor circuit and motor symptoms such as bradykinesia, rigidity, and temors start to show themselves.

 

Figure 3: Old and New Familial Mutantions of Alpha Synuclein

The three old familiar mutants are A30P, E46K, and A53T in green. They have been characterized and well-studied in multiple model organism systems. The three new familial mutants are H50Q, G51D, and A53E in purple. They are just beginning to be characterized. Note that all familial mutations are located on the N-terminus ad that two of the familial mutations are on the 53rd sire (A53E and A53T).

 

Figure 4: Four Substiution Mutations for H50Q, G51D, and A53E

The four substitions represeting the four functional classes of amino acids made for each new familial mutant are shown above. When the new amino acid is the same color as the original amino acid, the original amino acid properties are conserved. In all other cases, the original amino acid properties are lost.

 

Figure 5: Expected Phenotypes for H50Q Subsitution Mutants

The epected properties for the H50Q substitution mutants are shown above.H50R is expected to have properties similar to WT since it conserves the original basic property of the histidine. In terms of localization, it is expected to show foci initially and then go to the membrane. In terms of toxicity, it is expected to grow as well as WT. Finally, it is expected to show a similar level of expression compared to WT. Finally, it is expected to show a similar level of expression compared to WT. H50D, H50A, and H50N are expected to show similar properties to the basic property of the histidine. In terms of localization, they are expected to go to the membrane like WT, but at a faster rate. In terms of toxicity, they ae expected to grow as well as H50Q. Finally, they are expected to show a similar level of expression compared to H50Q

 

Figure 6: Expected Phenotypes for Altered for Nitrative Stress and Endocy­tosis The expected properties for the new familial mutants in strains altered for nitrative sgress and defective endocytosis are shown above. Compared to the new mutants in the control BY4741 strain, those in strains altered for increased nitrative stess (cox5astrain) and defective endocytosis (vps28 strain and v[ps35 strain) are expected to show more disease related characteristics. In terms of localization, more aggregation is expected. In terms of toxicitiy, the new mutants are expected to be more toxic.

Related Links:

• 188-192_Ong_SeniorThesis (PDF)