Mitochondrial-membrane association of α-synuclein: Pros and cons in consequence of Parkinson's disease pathophysiology

Abstract

The mitochondrial association of α-synuclein has been known to play a crucial homeostatic role in the neuronal cells. The slightest of a tweak in this association cause for activating the neurodegenerative mechanism, ultimately leading to neuronal loss, the pathological feature of Parkinson's disease. Studies so far have highlighted several intrinsic and extrinsic factors that alter this association, promoting protein misfolding and mitochondrial dysfunction, both of which bring about cooperative cellular degeneracy. Despite the years of effort, the exact molecular association remains unspecified, forbidding the characterization of disease etiology. In this review, we summarize the accumulating evidence for this mitochondrial membrane association, sewing the scattered links to a clear understanding of this interaction.

Introduction

Parkinson's disease (PD) is the most well-spread movement disorder and the second most common neurodegenerative disorder (Meissner et al., 2011) after Alzheimer's disease (AD). The disease is identified by motor disturbances such as resting tremor, rigidity, bradykinesia, and postural instability, prompted by the slow and progressive death of dopaminergic (DA) neurons from the substantia nigra (Wood-Kaczmar et al., 2006; Gazewood et al., 2013; Rao et al., 2006; Rezak, 2007). The SNCA locus from the fourth chromosome, encoding the α-synuclein protein is among the first loci recognized and identified to be associated with the autosomal dominant inherited form of PD (Polymeropoulos et al., 1997; Klein and Westenberger, 2012; Bras and Singleton, 2009). Several point mutations in the SNCA gene have been known to underlie the early onset of PD (Corti et al., 2011; Gasser, 2009). The disease pathophysiology is almost always marked by the presence of dense cellular inclusions mainly composed of the amyloidogenic protein aggregates in the form of Lewy bodies in the DA neurons (commonly referred to as Lewy neurites). Ever since its discovery, (Nussbaum, 2017) and association with the Lewy neurites, the protein, and its familial mutants have surfaced as a much sought-after subject of interest for many scientific groups (Stefanis, 2012; Dehay et al., 2015) studying protein dynamics associated with PD. However, despite the decades of research, the exact disease etiology remains inexplicit. Especially, not much knowledge has been gathered over the natural physiological role of the wild-type (WT) protein even though more about the disease association of the protein is known.

α-Synuclein is localized particularly in the nerve terminals and has also been known to be distributed in the cytosol, associated with the nucleus, endoplasmic reticulum (ER) and mitochondria; more specifically the mitochondria-associated membranes (MAMs) with ambiguous functionality (George et al., 1995; Iwai et al., 1995; Li et al., 2007; Guardia-Laguarta et al., 2014). Several convergent studies have highlighted the membrane-associated folding of the protein that underlies the dynamics, affecting protein homeostasis in vivo. However, to understand the disease etiology, it is particularly important to understand the triggering factors that initiate the protein misfolding associated with the disease pathophysiology. Studies have shown that the α-synuclein aggregation kinetics depends on the dynamic equilibrium between the native structures, i.e., the monomers and tetramers (Bartels et al., 2011; Dettmer et al., 2015a). The protein is natively unfolded and can adopt several conformations upon interactions with the cellular factors. Alteration of a single factor can promote the transformation into the neurotoxic species of the protein (Rajagopalan and Andersen, 2001). Several internal and external factors control the aggregation process of α-synuclein. High protein concentration (due to gene mutations and constitutive expressions), low pH (acidosis in PD patients), high interaction with cellular metabolites, and altered native conformations (mutations) has been known to regulate the aggregation propensity of the molecule (Hashimoto et al., 1998). Thus disease pathophysiology is marked by either discrete conditions or overlapping situations that determine the rate of disease onset and progression. Physiological homeostasis hangs by a thin string, and the slightest of a tweak in it brings down the entire homeostatic machinery. Newer studies have changed their focus on the functioning of the WT protein with the intracellular counterparts (Skamris et al., 2019; Cabin et al., 2002; Murphy et al., 2000). In this context, several parallel studies have concentrated on the protein's functioning upon interaction with the membranes, particularly the synaptic vesicles. More and more studies have shown a functional role of the protein at this interface at the presynaptic compartment of the cell, associated with neurotransmission (Cabin et al., 2002; Murphy et al., 2000). Alternatively, these studies also revealed homeostatic functioning of the protein upon association with other cell organelles, including the ER and mitochondria (Calì et al., 2012). Evidence collected on mitochondrial bioenergetics dysfunction and α-synuclein's SNCA gene mutations have revealed it to be an entirely new pathogenic component in PD generation (Polymeropoulos et al., 1997; Lin and Beal, 2006). The central role played by α-synuclein and mitochondria, suggests a possible convergent mechanism in PD pathogenesis. Numerous studies revealed that α-synuclein modulated the mitochondrial dynamics to regulate mitochondrial fission-fusion, transport, mitophagy and mitochondrial calcium homeostasis (Calì et al., 2012; Nakamura et al., 2011; Choubey et al., 2011), all of which have been adding to the rising fame of this α-synuclein-mitochondrial association. The mitochondrial membrane association of the protein is of particular interest, given the existing evidence on a membrane-associated folding of the protein (Wiedemann et al., 2004). Despite the increasing number of studies, the specific mitochondria-associated pathway of PD progression remains largely unknown. Parallel studies have shown more than one pathway where the two interact and participate in the pathogenesis (Fujita et al., 2014; Nakamura, 2013). While some have focused on the membrane-directed interaction, others have delved into understanding the interaction of α-synuclein with the proteins in the electron transport chain (ETC) associated with mitochondrial functioning (Di Maio et al., 2016; Nakamura et al., 2008). Despite this multi-strata interaction in PD etiology, all the studies are convergent upon the initial mitochondrial membrane affinity of the protein and the downstream dynamics.

Interestingly, most of the collected evidence holds true for the mitochondrial membrane association of the protein- as seen in the diseased brain. However, direct association as a part of the natural functioning of either counterpart (mitochondria and α-synuclein) has not been extensively dealt with. Understanding this natural physiological association of the two might enable us to correlate and gain a complete picture as to how and what triggers the diseased condition. A comprehensive understanding of the actual biological functioning is essential to analyze the pathophysiological condition. In this review, thus, we have attempted to bring together the studies focusing mainly on this association with the mitochondrial membrane that has explained a possible mechanism of α-synuclein aggregation in compliance with mitochondrial dysfunction. Through these comprehensive studies, we put an effort to understand this alternative pathway of PD initiation at the mitochondrial membrane interface.

α-Synuclein is a small intrinsically disordered protein (IDP), known to form the amyloidogenic core for the Lewy body inclusions serving as the neuropathological hallmarks of PD (Polymeropoulos et al., 1997; Wakabayashi et al., 2013). The appearance of Lewy neurites is always accompanied by an excessive DA neuronal loss in the substantia nigra pars compacta (Polymeropoulos et al., 1997; Wakabayashi et al., 2013) in PD pathophysiology. The protein has been widely accepted to be monomeric in the cytosol, primarily in a closed conformation, which inhibits its intrinsic aggregation propensity (Binolfi et al., 2012; Burré et al., 2013; Fauvet et al., 2012b). Parallel studies identified α-helically folded tetramers as the endogenous state in living cells that are in dynamic equilibrium with the unstructured monomers, resisting fibrillation within healthy neurons (Bartels et al., 2011; Dettmer et al., 2015a; Wdeang et al., 2011). Over the years several intrinsic and extrinsic factors have been shown to perturb this WT conformational dynamics, thus initiating misfolding with a heterogeneous pool of intermediates that compromise homeostasis (Breydo et al., 2012; Villar-Piqué et al., 2016). These structurally and functionally distinct intermediates initiate a downstream cascade of chemical interactions underlying the pathological aggregation and subsequent propagation in PD etiology (Angot et al., 2012; Recasens and Dehay, 2014; Tran et al., 2014). The amyloidogenic oligomers or proto-fibrillar forms of the protein are the most toxic conformations when compared to the monomers and the matured fibrils (Lashuel et al., 2013). The resultant neuronal toxicity appears to involve many pathways and cellular functions, including endocytosis, Golgi homeostasis, ER-to-Golgi transport, presynaptic trafficking, the ubiquitin-proteasome system (UPS), autophagy, mitochondrial fragmentation, and oxidative and nitrosative stress (Wang and Hay, 2015; Norris et al., 2003; Yamin et al., 2003). This clearly demonstrates the involvement of the protein in several cellular homeostatic functioning. Despite the obtained evidence, not much has been unraveled about this multi-strata involvement. Nevertheless, the malice associated with α-synuclein amyloidogenesis has been widely studied. Duplications, triplications, and point mutations of the SNCA gene result in autosomal dominant PD with the overexpression of α-synuclein (Tan et al., 2005). All known familial mutations in α-synuclein are related to the dominantly inherited forms of PD (Polymeropoulos et al., 1997). Extensive studies have shown that these point mutations affect the WT conformations, initiating amyloidogenesis (Flagmeier et al., 2016; Bhattacharyya et al., 2018). Interestingly, all the familial point mutations are reported in the N-terminal region of α-synuclein (Fig. 1), suggesting the significance of this domain in the pathological dysfunction of α-synuclein (Dehay et al., 2015; Bhattacharyya et al., 2018). Intriguingly, the N-terminal plays a crucial role in binding with the lipid membranes of the several target cell organelles, including the neurotransmitter vesicles (the primary function of α-synuclein reported till date) and mitochondria (Bartels et al., 2010).

Analyses of the primary structure of the protein and its physiological traits suggest a membrane-mediated homeostatic functioning of the protein in vivo (Gitler et al., 2008; Snead and Eliezer, 2014). Various convergent studies show a discrete role of α-synuclein at the membrane interface of several intracellular components (Pfefferkorn et al., 2012). In fact, α-synuclein-membrane interactions have been shown to modulate both protein and membrane properties (Pineda and Burré, 2017; Shi et al., 2015). The protein harbors varying degrees of affinity for different membranes adopting alternate conformations that possibly underlies its homeostatic functioning at the lipid-peptide interface. Depending on the lipid composition, some studies have shown that the membrane induced α-synuclein conformations have the potential to disrupt the associated membranes (Winner et al., 2011). Still, others have shown a vesicle turnover functioning of the protein (Fusco et al., 2016), while some suggest a membrane-associated signaling role of the molecule (Lee et al., 2002a). All these studies indicated that the molecular interaction with the membranes initiates the adoption of higher order aggregates (Lee et al., 2002a; Hu et al., 2016). These higher order conformations are susceptible to misfolding and unchecked aggregation, often related to disease etiology (Dikiy and Eliezer, 2012; Bhattacharyya et al., 2019). The aggregation rate of α-synuclein in the presence of a particular membrane might depend on two factors: the chemical composition of lipids and the lipid-to-protein ratio (Galvagnion et al., 2016; Zhu et al., 2003). This molecular association results in membrane remodeling, alterations in curvature, membrane thinning, and membrane expansion (Westphal and Chandra, 2013; Pandey et al., 2011; Jiang et al., 2013; Ouberai et al., 2013). Cationic N-terminal region of α-synuclein preferentially bind to the membrane composed majorly of negatively charged lipids by electrostatic interactions (Bhattacharyya et al., 2019; Stöckl et al., 2008; Pirc and Ulrih, 2015). Studies have shown the favored binding of α-synuclein to membranes that mainly contain lipid-packing defects and are marked by a high membrane curvature (Garten et al., 2015). α-Synuclein has been suggested to be able to insert and accommodate more easily into lipid bilayers, inducing lateral expansion of the lipid layer, resulting in a reduction of the average bilayer thickness (Shi et al., 2015; Ouberai et al., 2013).

The 140 amino acid long α-synuclein protein, encoded by the SNCA gene is divided into three distinct domains: (i) a positively charged N-terminal region (amino acids 1–60); (ii) the central fibrillating core (amino acid 61–95) that has a high propensity to aggregate (Giasson et al., 2001); and (iii) a highly acidic C-terminal domain (amino acid 96–140) rich in proline and negatively charged residues (Guardia-Laguarta et al., 2014; Bayer et al., 1999) (Fig. 1). The N-terminal region contains six imperfect 11-amino acid (KTKEGVVAAAE) repeat sequences, allowing the formation of α-helical structures, reflecting the protein's propensity in binding to phospholipid vesicles (Vamvaca et al., 2011; Bodner et al., 2009). Several studies have shown that this N-terminal mediated anchorage of the protein to the membrane surface opens the fibrillating core region of the molecule, making it susceptible to fibrillation (Bhattacharyya et al., 2019; Vamvaca et al., 2009; Davidson et al., 1998) (Fig. 2). The fibrillating core region of the molecule is marked by the central stretch of hydrophobic residues that have previously been shown to be the Non-Amyloid-β Component (NAC), in patients with AD (Bisaglia et al., 2006). The NAC region has a high propensity for adopting β-sheet secondary structure conformation and is essential for aggregation (Giasson et al., 2001; Uéda et al., 1993). Membrane association provides hydrophobic protection mediated by the acyl chain regions of the lipid layers (Fig. 2), hence serving as a template for further protein-protein interaction associated with the higher ordered conformations (Zhu et al., 2003; Gaspar et al., 2018). The C-terminal is inherently unstructured with five prolines, making it more flexible (Ulmer et al., 2005). Formerly known to be responsible for maintaining the protein's solubility, the C-terminal remains free when associated with the membrane surfaces (Levitan et al., 2011; Wang et al., 2016). This allowed subsequent studies that showed a phosphorylation-prone sequence that further affects the protein's natural physiological functions in association with biologically significant membranes (Sato et al., 2013; Ma et al., 2016; Nakamura et al., 2001). Apart from phosphorylation, several other post-translational modifications such as acetylation (Anderson et al., 2006; Fauvet et al., 2012a; Iyer et al., 2016a; Iyer and Claessens, 2019) methionine oxidation (Anderson et al., 2006; Okochi et al., 2000; Ellis et al., 2001), nitration (Giasson et al., 2000) ubiquitination (Hasegawa et al., 2002), etc. have been reported for the molecule's functional attribute in association with membranes. This segment-specific modification has been considered to be responsible for affecting the protein's net charge and hence its membrane directed functionality (Vamvaca et al., 2009; Iyer and Claessens, 2019; Giasson et al., 2000; Maltsev et al., 2013). Several convergent studies have shown that while some of these post-translational modifications might function for the homeostatic regulation of the normal physiology, others have a disruptive effect, being prevalent only in the pathophysiological scenario (Iyer and Claessens, 2019). Thus membrane association allows segment-specific segregation of functionality that remains subdued in the closed cytosolic monomers.