Neurological disorders are the predominant cause of disability and the second leading cause of death worldwide [1]. In the past 30 years mortality attributed to neurological disorders has risen by ~39% largely due to increased population and aging [1]. Mainly due to the inaccessibility of the human brain, insights into complex neurological disorders rely heavily on the development of model systems [2]. These model systems include human primary and immortalised cell lines, animal models (encompassing mammals—mouse and rat; nonmammalian vertebrates—zebrafish; invertebrates—nematodes and vinegar flies) as well as simpler eukaryotic models (e.g., yeast and Dictyostelium discoideum) ^{[3][4][5][6][7][8]}[3,4,5,6,7,8]. All of them have fully sequenced genomes, but the simple eukaryotes are unsurpassed in their short life cycles, low cost, ease of clonal growth and genetic manipulation. These benefits make them well suited for screening, testing and developing therapeutic agents. Despite having no central nervous system, their experimental tractability provides advantages for studying the function of disease-associated genes and underlying cytopathological pathways. They are amongst a small cohort of organisms recognized by the NIH for their usefulness as biomedical models [9].

D. discoideum has all the usual genetic and experimental tractability of an established, simple eukaryotic model, but in addition, it has a unique life cycle which, depending upon nutrient availability can be either solitary or social and provides diverse phenotypic “readouts” of underlying cytopathological pathways. When nutrients are plentiful the amoebae remain solitary, however, upon nutrient deprivation they differentiate and then release pulses of the chemoattractant cAMP which induces the aggregation of ~100,000 cells into mounds [10]. The cells then further differentiate, resulting in two distinct populations of cells—prestalk and prespore cells which participate in multicellular morphogenesis and further differentiation to ultimately form a fruiting body. This exceptional lifecycle distinguishes D. discoideum from other microbial models and has provided an exciting, simple, cost-effective, non-sentient model organism for studying complex biochemical pathways underpinning neurological disorders [11].

The genome of D. discoideum has been entirely sequenced [12] and many orthologs of human genes associated with neurological disorders have been identified. The genetic tractability of the organism allows genes related to such disorders to be easily manipulated and phenotypically analysed allowing insight into their conserved cellular functions. In addition to altering endogenous genes, several human genes have been expressed in D. discoideum and studied in the absence of endogenous copies, for example, human α-synuclein and Tau protein have been expressed singly and in combination in D. discoideum to elucidate the mechanisms of cellular toxicity in synucleinopathies and tauopathies ^[13][14][13,14]. Furthermore, many cellular processes are highly conserved in D. discoideum allowing for the investigation of underlying cytopathological mechanisms. For example, D. discoideum does not contain a homolog of human amyloid precursor protein (APP) but it has been expressed in D. discodeum and shown to be processed as in humans to produce amyloid beta (Aβ) peptides and other APP metabolites [15].

AD is the most common form of dementia, accounting for around 75% of cases ^[16][22], and the incidence is expected to triple by 2060 ^[17][23]. The central pathological characteristics of AD are the accumulation of neurofibrillary tangles comprised of hyperphosphorylated Tau proteins and of plaques comprised of amyloid-β proteins. These two key features have been hotly debated as the origin of the disease ^[18][24]. Support for AD pathology relating to plaque formation was initially advanced by the identification of mutations in patient populations with an early onset (familial) form of AD ^[19][25], arising from a range of loss-of-function mutations in components of the γ-secretase complex, especially in two presenilin proteins (Psen1 and Psen2), and in the amyloid precursor protein (APP) that is cleaved by the γ-secretase complex to form amyloid-β proteins. Research into AD pathology and treatment has therefore focused on the processes leading to the formation of tangles and plaques ^[20][26], but this has provided limited advances in treatments, suggesting that improved understanding of the role of the γ-secretase complex and presenilin proteins beyond tangle and plaque formation may provide alternative approaches for treatment.

D. discoideum provides a tractable model system, with a well-conserved γ-secretase complex including two presenilin proteins, where single and multiple genes encoding these proteins can be easily ablated and resulting isogenic mutants analysed. Furthermore, the model does not form tangles and plaques, allowing the effect of loss of function γ-secretase complex components to be investigated independent of tangle and plaque formation ^{[15][21][22][23]}[15,27,28,29]. The key complex components (Aph1, Ncst1, Pen2 and two presenilin proteins) in D. discoideum are of similar size, domain structure and catalytic amino acids to those found in mammals, suggesting conserved function. Analysis of the role of these individual components initially revealed that loss of both of the D. discoideum presenilin proteins (PsenA and PsenB), interrupts the process of multicellular development, to provide a clear and easily assessed phenotype for monitoring presenilin function ^[21][27]. The relevance of this model system and assay to the human condition is demonstrated by the rescue of this developmental block through the expression of the human Psen1 protein in the double mutant, confirming the functional homology of human and D. discoideum presenilin proteins. Interestingly, the presence of the key catalytic aspartic acid residues on either the D. discoideum PsenB (D348A and D394A) or the human Psen1 (D257A and D358A) is not required for developmental rescue ^[21][22][27,28], indicating a non-proteolytic function for both proteins in development. However, both human Psen1 and Psen2 retain auto-proteolytic activity in D. discoideum, through the cleavage of the human presenilin proteins to yield short C-terminal fragments ^[21][27], although the endogenous substrates remain to be identified. Development is not compromised by the loss of the other γ-secretase components, highlighting key roles for presenilin proteins rather than the γ-secretase in development in this model. Fluorescent tagging of the D. discoideum γ-secretase complex components (PsenB, Ncst, and Aph1) localised the complex to the endoplasmic reticulum, consistent with what is found in mammalian models ^[24][30]. Thus, presenilin proteins control development in D. discoideum, and this cellular function is conserved in the human homologues, evidencing D. discoideum as a suitable model to investigate human γ-secretase complex and presenilin protein function.

Further research into presenilin and the γ-secretase activity in D. discoideum identified a key role for the γ-secretase complex in regulating autophagy, in a proteolytic independent mechanism. Since the γ-secretase in mammalian models regulates endocytosis ^[25][26][27][31,32,33], and in D. discoideum regulates phagocytosis [15], a role for this complex was further examined in D. discoideum. Here, the dominant form of liquid uptake is via macropinocytosis, and this was shown to be reduced in a mutant lacking a component of the γ-secretase (Aph1) or both presenilin proteins ^[22][28], suggesting a key role for this complex is conserved in D. discoideum. Furthermore, reduced macropinocytosis following the loss of both presenilin proteins was restored through the expression of the human Psen1 protein, confirming evolutionary conservation of these proteins in this role. Surprisingly, the two key catalytic aspartic acids necessary for proteolytic activity are also not necessary for this endocytosis activity. Further analysis identified the role of the γ-secretase complex (including the presenilin proteins) in phagosomal proteolysis and autophagic flux, in regulating the time to acidification of the autophagosome, and thus reduced activity of the protein recycling complex in autophagy.

Thus, in D. discoideum, the function of presenilin proteins and the γ-secretase complex in regulating development and autophagy are independent of presenilin proteolytic activity. The activity of presenilin proteins in this proteolytic independent activity is also conserved in the human presenilin 1 protein. These data, therefore, suggest that presenilin proteins, and to a lesser extent the γ-secretase complex, function to regulate autophagy, phagosomal proteolysis and autophagic flux, in a manner that is independent of proteolytic cleavage activity. This role may be further investigated in relation to pathogenic changes in AD.

Parkinson’s disease is the leading cause of motor dysfunction and the second most prevalent neurodegenerative disorder after Alzheimer’s disease ^[28][34]. PD is characterised by motor symptoms including tremor, bradykinesia, rigidity, and gait abnormalities and non-motor clinical symptoms including sleep, behavioural disturbances and cognitive decline ^[28][29][34,35]. A defining hallmark of the disease is the loss of dopaminergic neurons from the substantia nigra region of the midbrain and the inclusion of Lewy bodies in surviving neurons ^[30][36].

Most cases of PD are sporadic or idiopathic with no known underlying genetic cause ^[31][37]. There are, however, rarer cases (about 10% of all PD) termed familial or genetic PD in which mutations to a single gene have been identified as the cause and many of these genes appear to regulate common cellular pathways such as mitochondrial signalling and lysosomal degradation. By genetically manipulating D. discoideum, various PD models have been created which alter the levels of one or more of the gene products known to be associated with PD. These models include manipulation of homologs of PD-associated genes and also the expression of human PD-associated genes which do not have homologs in D. discoideum. These models have been used to help unravel the complex pathophysiological mechanisms underlying PD and to better understand why certain mutations or repetitions give rise to the disease. Each of these PD-associated gene models will be discussed in further detail below.

Mutations in LRRK2 are the most common genetic cause of autosomal dominant, late-onset PD and genome-wide associated studies (GWAS) have subsequently linked the gene to sporadic cases of PD ^[32][47]. LRRK2 is a member of the Roco protein family, first discovered in D. discoideum in 2002 by Goldberg et al. ^[33][48]. The exact role LRRK2 plays in the pathogenesis of PD remains elusive, as does its normal role in healthy cells. There are 11 Roco family members in D. discoideum ^[34][38], more than is known in any other organism. These Roco proteins have diverse functions including roles in chemotaxis, development and growth. Roco1, also known as cyclic GMP-binding protein C (GbpC), plays a central role in chemotaxis. Upon starvation, and the subsequent release of the chemoattractant cAMP, Roco1 translocates to the cell membrane and interacts with myosin II ^[35][46]. Myosin II is regulated by phosphorylation and is essential for chemotaxis and cell polarity. Roco1 null mutants have abnormally phosphorylated myosin II and this cements a role for Roco1 in chemotaxis [15].

Like Roco1, Roco2/Pats1, also functions in chemotaxis wherein null mutants aggregate more slowly into larger mounds ^[36][39]. In addition, Pats1 has a role in cell division as null mutants display defects in cytokinesis ^[37][40]. Roco3/QkgA plays a role in cell proliferation with null mutants growing quicker in shaking culture and strains overexpressing QkgA growing slower ^[38][41]. Null mutants of Roco4 and Roco11 display aberrant fruiting body morphology revealing that both proteins are essential to normal multicellular development ^[39][43].

It is unclear which of the D. discoideum Roco kinases is the functional homolog of human LRKK2, but Roco4 has the highest combined sequence and structural similarity and in addition to the core LRR, Roc and kinase domains it also contains a WD40 domain like LRRK2 ^[40][42]. Specific human LRKK2 pathogenic variants are rare and often confined to a few families, however, the most studied mutation, a substitution at residue 2019 (G2019S) in the kinase domain, accounts for ~4% of familial and ~1% of sporadic cases worldwide ^[41][49]. This G residue is conserved in D. discoideum Roco4 at residue 1179 and the equivalent mutant form has been created (G1179S) and expressed in the Roco 4 null background ^[39][43]. As in human LRRK2, mutation of this residue in Roco 4 results in increased kinase activity in vitro ^[40][42]. Roco4 plays a key role in the D. discoideum developmental cycle. It is expressed throughout multicellular development with a peak in expression at the multicellular slug stage ^[34][40][38,42]. Its ablation results in normal aggregation but delayed tip formation and subsequent multicellular development with culmination occurring after 70 h compared to 12 h in the wild type (WT). The Roco4 null mutant also displayed a reduction in the number of fruiting bodies that formed and had a severe defect in the fruiting body morphology with very few stalk cells ^[39][43]. The reduction in stalk cells was determined to be due to a lack of cellulose rather than a defect in the stalk differentiation pathway ^[34][38]. The lack of cellulose produces unordered stalk cells which are unable to form a rigid structure and cannot support the sorus resulting in sori appearing to form on the agar surface ^[34][38]. The developmental defect was rescued by ectopic overexpression of the wild type Roco4 protein, or its kinase domain only, or a chimeric Roco4/Lrrk2 protein in which the Roco4 kinase domain was replaced with the human LRRK2 kinase domain. However, the mutant Roco4^G1179S protein only partially rescued the developmental defect in the null background ^[39][43]. Although the Roco4^G1179S strains had no defect in the number of fruiting bodies, their development was still significantly delayed, and their morphology was aberrant with smaller stalks. At the multicellular slug stage Roco4 null slugs did not migrate and Roco4^G1179S strains had impaired phototactic orientation towards the light source. In both strains, these defects were rescued by the expression of the Roco4 wild type protein ^[39][43].

Given that phototactic defects can often manifest due to impaired mitochondrial function ^[42][50] and that human LRKK2 is implicated in mitochondrial function, this was assayed directly using two measures of mitochondrial function—reactive O₂ species (ROS) levels (determined fluorometrically) and mitochondrial oxidative phosphorylation (using live cell respirometry in a Seahorse Extracellular Flux analyser) ^[43][51]. WT and Roco4⁻ strains displayed similar amounts of ROS, however after the addition of the protonophore CCCP, the Roco4⁻ mutant had elevated ROS compared to WT, indicating the Roco4⁻ mutant had a reduced ability to remove ROS following mitochondrial uncoupling ^[39][43]. Alternatively, the null mutant may have a greater capacity to generate ROS after CCCP uncoupling because of its elevated complex I activity under these conditions.

Respirometry experiments revealed basal oxygen consumption rates (OCR), OCR contributing to ATP synthesis and OCR by Complex I, were all elevated in the Roco4⁻ mutant. In contrast, basal OCR was significantly reduced in Roco4^G1179S strains. Interestingly, in the Roco4 wild type rescue and Roco4^G1179S strains maximum, OCR and spare capacity OCR were significantly reduced ^[39][43]. Electron microscopy revealed no defects in mitochondrial morphology or structure and no reduction in the number of mitochondria in the D. discoideum Roco4⁻ mutant. ^[39][43]. This suggests that the increased kinase activity present in the point mutant results in the impaired mitochondrial function which is in agreement with a gain of function property of the PD-associated mutation.

HTRA2, or Omi protease, is a nuclear-encoded protein that, in humans, primarily resides in the intermembrane space of the mitochondria ^[44][52]. It is a serine protease chaperone that is released to the cytosol prior to apoptosis and cleaves Inhibitor of Apoptosis Proteins (IAPs), initiating the apoptotic cascade ^[45][53]. Mutations in the gene are associated with late-onset autosomal dominant PD and it has been suggested that this is due to loss of HTRA2′s neuroprotective role ^[46][54]. Missense mutations in HTRA2 lead to reduced protease activity, resulting in an accumulation of unfolded proteins within the mitochondria, eventually leading to mitochondrial dysfunction ^[47][55]. HTRA2 is also regulated by PINK1, a putative mitochondrial protein kinase, wherein PD-associated mutations in PINK1 lead to reduced phosphorylation of HTRA2. Interestingly, when HTRA2 is overexpressed, this too can result in mitochondrial defects and unregulated apoptosis ^[45][53].

Mitochondrial dysfunction has been implicated as the cause of neuronal death in PD patients with reduced HTRA2 activity but the low penetrance of HTRA2 and the fact that previously reported pathogenic variants have been discovered in seemingly healthy asymptomatic individuals does not support this ^[48][44]. These discrepancies may be explained by the complex phenotype–genotype relationship that is often described in mitochondrial biology.

To understand the normal and pathogenic roles of HTRA2 D. discoideum has been employed. A single homologue of HTRA2 was identified in D. discoideum (htrA) and as in humans, the protein was shown to localise to the mitochondria ^[48][44]. Antisense inhibition of htrA revealed that reduction of HTRA2 as suggested by the reduced mRNA levels of HTRA2, partially phenocopied D. discoideum mitochondrial disease models with aberrant multicellular development and reduced growth rates with no corresponding defect in endocytosis ^[48][44]. As its protease activity has been implicated in its neuroprotective role the researchers created strains that overexpressed a protease-dead HTRA2 protein with an amino acid substitution at residue 300 in the kinase domain (S300A). HTRA2^S300A expressing strains displayed phenotypes resembling the HTRA2 antisense-inhibited strains, including deranged fruiting body morphology and reduced growth rates in the absence of defects in endocytosis. This suggested that HTRA2′s role in these phenotypes could be ascribed to loss of the protease activity. The researchers measured mitochondrial respiration directly via Seahorse respirometry and no defect in either the antisense-inhibited or S300A-expressing strains was detected. Overexpression of HTRA2 proved lethal to D. discoideum suggesting hyperactivity of the protein is also cytotoxic. The study revealed that HTRA2 in D. discoideum has a similar role to that in humans where tight regulation of HTRA2 is needed to avoid cytopathological dysfunction. Due to these similarities, D. discoideum may be useful in further elucidating HTRA2′s exact role in PD including its role in mitochondrial function. It has highlighted the importance of HTRA2′s protease activity and how the loss of this activity in the mitochondria can result in cytopathological outcomes, even in the absence of measurably impaired mitochondrial respiratory function ^[48][44]. It is possible that other mitochondrial functions are more sensitively impaired by loss of HTRA2 function than is respiratory oxidative phosphorylation.

Unlike HTRA2, mutations in DJ-1 (Park7) have extremely high penetrance despite being very rare ^[32][47]. Mutations in DJ-1 account for 1% of early-onset PD and are autosomal recessive ^[49][56]. DJ-1 has also been implicated in idiopathic late-onset PD, whereby post-mortem analysis of PD brains has shown that the protein is more abundant and oxidatively damaged ^[50][45]. Interestingly, mutations in DJ-1 appear to cause an increase in the clinical presentation of non-motor PD symptoms such as anxiety, psychosis, and cognitive decline ^[51][57]. DJ-1 appears to be particularly important to high energy-demanding tissues, such as the brain, where reactive oxygen species are generated more readily ^[52][58]. DJ-1 has multiple reported cellular roles, as a molecular chaperone, antioxidant, and protease, as well as ROS scavenger and transcriptional regulator. A conserved cysteine residue at position 106 is vital to DJ-1-mediated protection against oxidative stress ^[52][58]. DJ-1 has been shown to share common biological roles with PINK1, another protein that protects cells against mitochondrial dysfunction. In Drosophila, DJ-1 can rescue PINK1 knockout-associated phenotypes despite no substantial evidence for a direct interaction, indicating they share a common biological pathway ^[53][59].

DJ-1 also interacts with other PD-associated proteins such as α-synuclein, reducing its toxicity by sequestering α-synuclein monomers preventing its aggregation, however, DJ-1 can also act on α-synuclein fibrils and remodel these into toxic oligomers ^[54][60]. DJ-1 can directly regulate ROS production by increasing the expression of mitochondrial uncoupling proteins (UCP4 and 5) leading to reduced mitochondrial membrane potential and lower ROS production ^[52][58]. This has led to the conclusion that DJ-1 has potential roles in both neurodegeneration and neuroprotection ^[52][54][58,60]. DJ-1 acts as a sensor for oxidative stress and a molecular chaperone, but the mechanism by which DJ-1 achieves this is unclear ^[52][54][58,60].

DJ-1 has been reported to localise in the nucleus, mitochondria and cytoplasm ^[55][61]. Despite many studies, the exact function and localisation of DJ-1 remain uncertain ^[50][45]. D. discoideum has a single homologue of mammalian DJ-1, encoded by deeJ, and has been found to localise predominately to the cytoplasm. DJ-1 plays a vital role in multicellular development in D. discoideum. Antisense-inhibited strains exhibit aberrant fruiting body morphology with shorter, thicker stalks. This suggests that more cells are undergoing autophagic cell death and that DJ-1 may play a role in the regulation of autophagy. DJ-1 also has a role in growth and endocytosis with knockdown strains displaying reductions in growth in liquid media and on bacterial lawns. In both cases, the reduced growth was attributed to a reduced endocytic rate, reduced pinocytosis for liquid growth and reduced phagocytosis for growth on bacterial lawns. The converse was observed in DJ-1-overexpressing strains with increases in growth and endocytic rates ^[50][45].

Interestingly, studies performed in primary cortical neurons of DJ-1 KO mice have revealed that synaptic vesicle endocytosis is severely impaired ^[56][62]. Similar results were also achieved using DJ-1 KO mouse primary astrocytes, further validating DJ-1′s positive regulation of endocytosis ^[57][63]. D. discoideum is a well-established model for the study of endocytic pathways, with many endocytic proteins having already been characterised ^[58][64]. This provides an opportunity to properly characterise DJ-1′s role in regulating endocytosis and further elucidate its role in humans.

Despite the cytoplasmic localisation of DJ-1 in D. discoideum and its positive role in endocytosis, when overexpressed, DJ-1 has been shown to inhibit certain parameters of mitochondrial respiration ^[50][45]. Seahorse Extracellular Flux experiments revealed that, in DJ-1-overexpressing strains, basal and uncoupled respiration were reduced, as were ATP synthesis and Complex I activity. Conversely, the knockdown of DJ-1 slightly enhanced these parameters. Regression analysis revealed that both ATP synthase and Complex I are functionally normal in strains expressing altered levels of DJ-1, however, the total amount of respiratory activity was altered. These results are inconsistent with the proposed role of DJ-1 in protecting mitochondria, but are consistent with the elevated respiration rates also observed in other PD models including lymphoblasts and fibroblasts from idiopathic PD patients and in a neuroblastoma model exposed to α-synuclein fibrils ^[59][60][61][65,66,67]. Together this suggests that mitochondria may play a more complex role in the cytopathology of PD and warrants further investigation.

To investigate if DJ-1 played a role in protection against oxidative stress in D. discoideum, cells with altered expression of DJ-1 were exposed to H₂O₂ ^[35][46]. Like DJ-1 knockdown, exposure of wild type cells to H₂O₂ also caused impaired morphogenesis, growth and endocytosis, but the combination of DJ-1 knockdown and oxidative stress caused more severe defects than either on their own. Neither oxidative stress, nor DJ-1 knockdown caused a defect in slug phototaxis, but in combination, a dramatic defect in slug phototactic accuracy was produced. Taken together these results show that, as in other model organisms and in humans, DJ-1 provides cytoprotection against the effects of oxidative stress.

To understand how DJ-1 might exert these cytoprotective effects, it is important to understand the cytopathological mechanisms that are elicited by oxidative stress. Chen et al. ^[35][46] explored the possibility that AMP-activated protein kinase (AMPK) may act downstream of oxidative stress in the DJ-1 knockdown strains ^[35][62][46,68]. AMPK is known to play a role in the impairment of phototaxis and other phenotypes caused by mitochondrial dysfunction and may play a similar role after oxidative stress, especially when the protective role of DJ-1 has been compromised. If this was the case, then antisense inhibition of AMPK in DJ-1 knockdown strains should rescue the phenotypic defects observed after exposure to H₂O₂. This was found to be true, with knockdown of AMPK rescuing the phagocytosis, growth and fruiting body morphologies present after oxidative stress. Furthermore, the deranged phototaxis observed only in oxidatively stressed DJ-1 knockdown strains was rescued by antisense-inhibition of AMPK. This shows that these adverse phenotypic outcomes caused by oxidative stress are mediated by AMPK and exacerbated by the loss of DJ-1′s protective function.

To determine if DJ-1′s protective role was exerted at the level of preventing mitochondrial oxidative damage, respirometric analysis of mitochondrial function with and without oxidative stress was conducted in strains expressing altered levels of DJ-1 ^[35][46]. Basal OCR, maximum OCR, OCR consumption by ATP synthesis and OCR consumption by “proton leak” were all proportionately lowered by the exposure to H₂O₂ regardless of DJ-1 expression levels. Furthermore, elevated DJ-1 expression impaired mitochondrial respiration (rather than enhancing it) and DJ-1 knockdown enhanced mitochondrial function (rather than impairing it) in oxidatively stressed cells to the same extent as in unstressed cells ^[50][45]. Thus, although DJ-1 protects cells from the AMPK-dependent consequences of oxidative stress, it does not do so by protecting the mitochondria.

D. discoideum has provided insight into the complexities surrounding DJ-1′s cellular roles in endocytic pathways and protection of cells during oxidative stress. The results were consistent with a model in which DJ-1 and oxidative stress each exert independent inhibitory effects on mitochondrial respiratory function ^[35][46]. Under oxidative stress, AMPK becomes activated, and DJ-1 becomes oxidized. The chronic hyperactivity of AMPK has multiple cytopathological consequences, including impaired endocytosis and growth mediated via inhibition of oxidized DJ-1, as well as DJ-1-independent impairment of phototaxis, growth and morphogenesis. Genetic loss of DJ-1 function results in impaired endocytic pathways and consequently growth, these aberrant phenotypes being exacerbated by oxidative stress because of AMPK’s activation and the loss of DJ-1′s inhibition of AMPK under these conditions. The D. discoideum model has thus contributed to the understanding of DJ-1′s cellular roles and the mechanisms by which its loss can contribute to cytopathological outcomes, particularly in combination with elevated oxidative stress.

D. discoideum has also allowed researchers the opportunity to study human PD-associated genes and genetic variants in the absence of endogenous genes and α-synuclein is one such example. Alpha-synuclein is involved in vesicle trafficking and is concentrated at the pre-synaptic terminals of neurons ^[63][69]. Misfolding of the protein results in a group of diseases named synucleinopathies which include PD as well as other neurodegenerative diseases such as multiple system atrophy (MSA) and dementia with Lewy bodies ^[64][70]. Misfolded α-synuclein is the most abundant protein in the characteristic intraneuronal aggregates or Lewy Bodies present in PD and is crucial to the progression of PD. Particular polymorphisms in the SNCA gene encoding α-synuclein substantially increase the risk of developing sporadic PD. Furthermore, multiplication and consequential overexpression of the SNCA locus, as well as certain point mutations within the gene result in early-onset, autosomal dominant PD ^[32][47]. Truncation of the C-terminus of α-synuclein increases the tendency of the protein to form cytotoxic oligomers as well as to aggregate ^[65][71]. However, the exact cytotoxic mechanisms underlying α-synuclein pathology remain elusive [13].

To create D. discoideum models for α-synuclein cytotoxicity, the wild type human protein and two PD-associated mutant forms of it were expressed in D. discoideum [13]. The full length (WT) and A53T mutated α-synuclein were found to be enriched in the cellular cortex, but the C-terminally truncated α-synuclein was found throughout the cytoplasm, suggesting the C-terminus is essential for cortical localisation. In support of this, a fusion protein consisting of the C-terminal 20 amino acids of α-synuclein fused to GFP localised to the cell cortex, demonstrating that these residues are not only necessary but are also sufficient for cortical localisation [13]. This result contrasts with reports of mitochondrial localisation of α-synuclein in human dopaminergic neurons ^[66][67][72,73] but agrees with other reports that α-synuclein is enriched in the cell cortex and its binding to membranes involves the N- and C-termini ^[68][74].

D. discoideum α-synuclein expressing strains were analysed for hallmark mitochondrial dysfunction phenotypes including defective slug phototaxis, reduced growth with unaffected endocytosis and impaired fruiting body morphology [13]. Strains expressing wild type α-synuclein displayed very mild phototactic defects which were not statistically significant, the A53T α-synuclein-expressing strains had a moderate defect and the C-terminally truncated α-synuclein-expressing strains showed a larger phototactic impairment in addition to impaired thermotaxis. Fruiting body morphology was not altered by the expression of WT or A53T mutated α-synuclein, however, C-terminally truncated α-synuclein, resulted in a reduced number of fruiting bodies which had thicker and shorter stalks. All species of α-synuclein caused defects in plaque expansion on bacterial lawns which was partially attributed to a reduction in phagocytosis [13].

With the exception of the defective fruiting body morphology observed in the C-terminally truncated α-synuclein strains, antisense inhibition of expression of the AMPK α subunit rescued or partially rescued all of these defective phenotypes [13]. This suggested that cellular stress, perhaps a reduction in ATP production, may be occurring in these strains and activating the energy sensor AMPK. The researchers investigated mitochondrial function directly using Seahorse respirometry and showed that strains expressing α-synuclein did not have impaired but rather enhanced mitochondrial function [13]. Strains expressing the full-length wild type (WT) α-synuclein were most affected and displayed elevated basal and maximum respiration and the main components of these measures were also elevated. C-terminally truncated α-synuclein expressing strains also displayed these increases in respiration but only reached statistical significance for maximum respiration, not basal respiration. The smallest effect was seen in the A53T-expressing strains whose elevated measures did not reach statistical significance for any of the measured parameters. The increased respiration in the α-synuclein strains was not due to an impairment of any of the complexes, with the relative contributions of each complex to basal or maximum remaining unchanged. Therefore, the complexes were more active but functionally normal [13]. Whilst this elevated respiration is in contrast to some reports of mitochondrial dysfunction in PD, it accords with the elevated respiration rates in multiple cellular PD models ^{[50][59][60][61]}[45,65,66,67].

The aggregation and accumulation of α-synuclein is thought to involve impaired interactions with other proteins and studying these interactions in model systems can assist in characterising the cytopathological disease pathways ^[13][14][13,14]. D. discoideum has been used to investigate the interaction of α-synuclein with the human microtubule-associated Tau protein [14]. D. discoideum lacks endogenous Tau and α-synuclein which allowed investigation of the effects of human Tau alone and in combination with human α-synuclein. The longest isoform of Tau was used to create a D. discoideum model. It was found that Tau was present throughout the cytosol where it was also seen colocalized in close association with tubulin [14]. In human cells, Tau is phosphorylated on many sites and in the phosphorylated state dissociates from microtubules ^[69][75]. D. discoideum was shown to phosphorylate Tau on at least one of these regulatory sites (S404) and to colocalise with tubulin in the cytosol. Strains expressing both α-synuclein and Tau showed that these two human neurodegeneration-associated proteins colocalised at the cortex where α-synuclein is most prominent. The close proximity (within 10 nm) suggests that Tau and α-synuclein have a physical interaction, not just a colocalization [14].

The strains expressing Tau or α-synuclein were examined phenotypically and shown to cause different patterns of phenotypic abnormalities [14]. Whilst expression of WT α-synuclein resulted in growth defects, impaired phagocytosis and increased mitochondrial respiratory activity, expression of Tau resulted in impaired phototaxis, thermotaxis, growth in liquid and an isolated respiratory complex V defect in oxidative phosphorylation. Tau expression did not alter plaque expansion rates on bacterial lawns, while α-synuclein expression did. Tau-expressing strains were also more susceptible to Legionella sp. infection, whereas α-synuclein strains were not. Tau-expressing strains displayed aberrant fruiting body morphology with shorter and thicker stalks, indicative of mitochondrial dysfunction, whereas α-synuclein expressing strains were comparable to WT.

When both proteins were coexpressed, a distinct profile of altered phenotypes was evident, different from that caused by expressing either of these proteins alone [14]. Thus, the coexpression of tau and α-synuclein is exacerbated (phototaxis, fruiting body morphology), or reversed (phagocytosis, growth on plates, mitochondrial respiratory function, Legionella proliferation) the abnormalities caused by either Tau or α-synuclein alone [14]. These results, together with the colocalisation experiments, indicate Tau and α-synuclein interact and influence each other’s cytotoxic effects.

To further elucidate the interaction between Tau and α-synuclein, whole-cell proteomic analysis was performed. Like the phenotypic data, the expression profiles of proteins in the Tau-expressing and α-synuclein-expressing strains were distinct. In Tau-expressing strains, cytosolic proteins were more abundant whereas, in α-synuclein expressing strains, the levels of cortex-associated proteins were elevated, consonant with the subcellular locations of both proteins. Genes associated with protein turnover were upregulated in strains expressing Tau, whereas strains expressing α-synuclein had no upregulation of genes involved in protein synthesis or degradation. This difference may reflect homeostatic feedbacks in Tau-expressing strains that increase protein catabolism as a means to increase ATP synthesis. In strains coexpressing both Tau and α-synuclein a downregulation of cytoskeletal genes was observed in all strain groups, which may explain the phenotypes observed such as impaired phototaxis, thermotaxis, phagocytosis and multicellular development, as these genes play key roles in these biological processes. The downregulated expression in these strains of proteins involved in cell morphogenesis, polarity and motility may explain why phototaxis was more deranged in these strains and “slugs” migrated shorter distances [14].

The results of ectopically expressing Tau and α-synuclein individually and in combination in the D. discoideum model have shown compelling evidence that these proteins interact in the cortex of cells, probably physically and also functionally in the altering of phenotypes. The phenotypic consequences highlight a complex relationship between the two proteins which could be further explored and produces distinct dysfunctional phenotypes suggesting that the two proteins exert their cytotoxic effects through distinct pathways and mechanisms [14].

Huntington’s disease (HD) is a fatal neurodegenerative disease that is characterised by involuntary movements, cognitive decline and psychiatric symptoms. The disease presents clinically in the 4th and 5th decades of life and the average life expectancy is within 15–20 years of disease onset. HD is an autosomal dominant disorder caused by mutations in the huntingtin (htt) gene. The mutations result in an expansion of the CAG trinucleotide repeat in excess of 35 repeats. The normal and mutant Huntingtin protein is expressed ubiquitously but certain neuronal populations are more susceptible to damage and death, specifically striatal and cortical neurons ^[70][77]. In humans, numerous roles have been ascribed to the Huntingtin protein including roles in embryogenesis and development, apoptosis, autophagy, transcription, vesicle trafficking, axonal transport and mitochondrial impairment. It is widely accepted that the mutation results in a gain of function, but it may also interfere with Huntingtin’s normal role ^[71][78].

The Huntingtin protein has been well conserved in evolution and is present in all eukaryotes with the exception of plants and fungi. D. discoideum contains a single htt gene (DDB0238473) encoding the Huntingtin homologue, Htt. The protein contains a polyglutamine tract which is 19 residues in length, comparable to the normal range of the expansion in humans. The polyglutamine residues are encoded by the CAA codon interrupted by a single CAG codon and are not positioned in exon 1 as in mammals ^[72][79]. Santarriaga et al. ^[73][80] genetically manipulated the D. discoideum Htt protein by creating a polyglutamine expansion in exon 1 and showed that unlike all other organisms tested including yeast and mammalian cells this protein did not aggregate. D. discoideum encodes a large number of proteins with polyglutamine expansions and such proteins in other organisms including mammals are generally associated with protein aggregation and disease, yet in D. discoideum these proteins do not aggregate. The properties that make D. discoideum proteins with polyglutamine stretches resistant to aggregation are not clearly defined but it has been suggested that chaperone proteins specifically small heat shock proteins may be involved and D. discoideum encodes a large number of these proteins ^[73][80]. The Htt protein is expressed throughout development and is located in the cytosol ^[72][79]. D. discoideum htt gene knockouts have been created to study the function of the protein and, as in mammalian cells, D. discoideum Htt is multifunctional.

Htt in D. discoideum functions in growth and development, but unlike in higher eukaryotes where it is embryonically lethal, D. discoideum Htt⁻ mutants are viable. When Htt⁻ cells were grown in adherent culture their growth rate was unimpaired ^[72][79] but when grown in an axenic shaking culture they grew slower than the wild type parental strains ^[74][81]. This difference in ability to grow in different media was found to be due to a defect in the EDTA-resistant cell adhesion properties in shaking culture. The Htt null cells showed an altered expression of the key regulator of EDTA-resistant cell-cell adhesion contact site A (csA) glycoprotein. These altered-expression and EDTA-resistant cell adhesion properties were rescued in the null mutant by supplementation with calcium ^[72][75][79,82]. It is unknown how calcium can rescue these defects, but the results clearly show that Htt plays a role in cell adhesion in D. discoideum as it does in higher eukaryotes suggesting this function is well conserved through evolution. D. discoideum provides a useful system to further interrogate this pathway and determine how the repeat expansion interferes with this function.

HTT, the human Huntingtin gene, is essential and its deletion is embryonically lethal in mice ^[76][83]. In D. discoideum, the Htt protein also plays a role in development with Htt⁻ mutants showing delayed development, beginning with a delayed onset of aggregation and streaming and failure of many mounds to progress through further development. The fruiting bodies which did form did so at a delayed rate and the sori had a glassy appearance, due to the premature germination of spores in the sori with a decrease in spores with increasing age of the fruiting bodies. Spores were also less viable and less efficiently formed ^[72][79]. Using chimeras, Myre et al. ^[72][79] showed that htt null cells in the presence of wild type cells would not populate the prespore region and consequently did not end up in the spores. Furthermore, Bhadoriya et al. ^[74][81], using LacZ reporter fusion and neutral red staining demonstrated a misregulation of spatial expression of prestalk cells. Conversely, the expression of prespore cells was the same as the wild type until it reached the slug stage and its expression significantly increased in the htt mutant. The researchers then examined the expression of STAT transcription factors and noted a decrease in the expression of STATs involved in the decision to proceed with slug migration or culminate. Together this data shows that Htt is required for proper patterning and maintenance of stalk and spore cell boundaries ^[74][81].

The delay in aggregation that was observed in the Htt⁻ cells was due to a defect in cAMP signalling. The ability of starving cells to aggregate under conditions of nutrient stress is dependent on chemotaxis to cAMP. Thompson et al. ^[77][84] showed that the speed and accuracy of chemotaxis is reduced in the htt⁻ mutant ^[77][84]. Bhadoriya et al. ^[74][81] analysed cAMP signalling in htt⁻ cells by measuring cAMP protein levels and levels of transcripts of key genes. They noted a reduction in absolute cAMP levels during all stages of development with no characteristic increases during starvation or in the loose aggregate stage. cAMP is produced by three different adenylate cyclases and all three showed altered levels of transcripts in the htt⁻ cells. AcA (adenylyl cyclase of aggregation) which is normally expressed during aggregation was reduced at all developmental stages, AcrA (adenylyl cyclase with response regulator domain) normally expressed at mid development and is essential for spore production was reduced in later stages of development and AcgA (adenylyl cyclase of germination) which is important for maintaining spore dormancy was increased during development ^[74][81]. Furthermore, the catalytic subunit of protein kinase A (PKA) was increased in the vegetative cells, but was reduced in the slug and fruiting body whereas the regulatory subunit was decreased at all stages of development. This suggests that PKA in htt⁻ cells is regulated independently of cAMP and ensures that culmination can proceed despite the reduced cAMP signalling and delayed aggregation ^[74][81].

Myosin II forms filaments in chemotaxis and is required for the structural reorganisation of the actin network and contraction of the posterior of the cell. It normally accumulates in the cortex after cAMP stimulation and the amount of myosin II decreases in wild type cells after stimulation with cAMP ^[78][85]. This did not occur in the htt⁻ mutants which did not show a cAMP-stimulated accumulation in the cortex, displaying a reduced amount of myosin II in the triton-insoluble fraction that did not change with cAMP signalling. Myosin II is regulated by its phosphorylation status and the amount of phosphorylated myosin II was increased in the htt⁻ mutant ^[78][85].

Myosin II is also required for cytokinesis where it accumulates at the cleavage furrow. In agreement with their altered myosin II function, the htt⁻ mutant cells displayed a defect in the late stage of cytokinesis on non-adherent coverslips. Myosin II accumulated at the cleavage furrow early in cytokinesis but was subsequently lost from this location, likely due to a reduction in the activity of PP2A and an increase in phosphorylated myosin II ^[78][85].

Myosin interacts with the actin cytoskeleton whose complex rearrangements are necessary for normal motility and division. In addition to the cytokinesis and chemotaxis defects, Htt-vegetative cells displayed defects in response to changing environmental conditions. When Htt⁻ cells were grown in a non-nutrient buffer a reduction in pseudopods and a rounding of the cells was observed. This was accompanied by a relocation of F-actin from the cortex to the cytosol. The htt⁻ cells were also sensitive to osmotic shock and, unlike wild type cells, when placed in water did not form contractile vacuoles, lost adhesion and lysed. Thus, the htt⁻ cells have a defect in contractile vacuole activity under hypotonic stress ^[72][79].

As in other organisms Htt in D. discoideum is multifunctional and many of these functions seem to be conserved across evolution. Further characterising these roles in D. discoideum, where genetic manipulation of the pathways is possible and multiple phenotypes are measurable, will increase the understanding of the normal role of Htt and how mutation of the protein can lead to Huntington’s Disease.