Biochemical Society Transactions

8th International Meeting on Yeast Apoptosis

Aggresome formation and segregation of inclusions influence toxicity of α-synuclein and synphilin-1 in yeast

Erwin Swinnen, Sabrina Büttner, Tiago F. Outeiro, Marie-Christine Galas, Frank Madeo, Joris Winderickx, Vanessa Franssens

Abstract

PD (Parkinson's disease) is a neurodegenerative disorder, caused by a selective loss of dopaminergic neurons in the substantia nigra, which affects an increasing number of the elderly population worldwide. One of the major hallmarks of PD is the occurrence of intracellular protein deposits in the dying neurons, termed Lewy bodies, which contain different proteins, including aggregated α-synuclein and its interacting protein synphilin-1. During the last decade, a number of groups developed yeast models that reproduced important features of PD and allowed the deciphering of pathways underlying the cytotoxicity triggered by α-synuclein. Here, we review the recent contributions obtained with yeast models designed to study the presumed pathobiology of synphilin-1. These models pointed towards a crucial role of the sirtuin Sir2 and the chaperonin complex TRiC (TCP-1 ring complex)/CCT (chaperonin containing TCP-1) in handling misfolded and aggregated proteins.

  • α-synuclein
  • polarisome
  • sir2
  • synphilin-1
  • TCP-1 ring complex/chaperonin containing TCP-1 (TRiC/CCT)

Introduction

In healthy cells, proteins are continuously in a delicate balance between conservation of the functional protein conformation, refolding of misfolded proteins and the degradation of proteins that are beyond repair. To maintain this proteostasis, cells rely on chaperone-based quality control mechanisms and proteolytic systems. Chaperones detect misfolded proteins and facilitate their ATP-mediated refolding and repair [1]. However, when these chaperones are unsuccessful in achieving natively folded proteins, the proteins are redirected to the ubiquitin–proteasome system or the autophagy-lysosomal pathway for degradation [2]. Furthermore, cells are also equipped with systems to deal with abnormal levels of ROS (reactive oxygen species), which promote protein damage and misfolding [3]. In most age-dependent neurodegenerative diseases, the protein quality control systems and anti-oxidative defence mechanisms fail or become overwhelmed, resulting in elevated levels of misfolded proteins and aggregates and eventually in cellular demise. In PD (Parkinson's disease), intracellular inclusions constitute an important pathological hallmark. These so-called Lewy bodies mainly contain aggregated α-synuclein, in addition to other proteins such as synphilin-1 [4,5]. A good understanding of the cellular mechanisms connected to aggregation of these proteins is essential for the development of therapeutic strategies that effectively intervene in the progression of the disease. The present mini-review focuses on the interaction and aggregation of α-synuclein and synphilin-1 and describes the added value of yeast as a tool to study the role of these proteins in PD. In addition, we provide and discuss additional data on the role of sirtuins and the TRiC (TCP-1 ring complex)/CCT (chaperonin containing TCP-1) chaperonin complex in preventing cell death in response to protein misfolding and aggregation.

Aggregation of α-synuclein as a hallmark of PD

Not long after the discovery that fibrillar α-synuclein was the major constituent of Lewy bodies in the brain of PD patients [4], mutations in the SNCA gene, encoding α-synuclein, as well as duplication or triplication of the SNCA allele were found to be associated with familial cases of PD. So far, three α-synuclein missense mutations have been identified: A30P, E46K and A53T, all of which render the protein more prone to form amyloid fibrils in vitro [6]. In patients with sporadic PD, intracellular protein aggregates contain wild-type α-synuclein. To date, the triggers inducing oligomerization, fibrillation and aggregation of α-synuclein are not fully understood, although a number of mechanisms have been proposed. Consistent with its putative role in vesicle dynamics, α-synuclein interacts with lipids and membranes and this appears to relate to its toxic activity. However, there are conflicting data regarding the role that membranes play in the α-synuclein aggregation process, since both enhancement [7] and suppression [8] of aggregation have been reported. In Lewy bodies, α-synuclein displays several post-translational modifications, including oxidation and nitration, ubiquitination and SUMOylation, C-terminal truncation and phosphorylation [9]. The latter involves the interplay between various kinases able to phosphorylate α-synuclein either at Ser87 and Ser129 or Tyr125, Tyr133 and Tyr136, thereby increasing or inhibiting the aggregation propensity of the protein [10].

PD pathology is also closely associated with mitochondrial dysfunction, which influences the formation and accumulation of α-synuclein aggregates by means of an increase in oxidative and nitrative radicals as well as by a reduction in ATP levels required for the proper removal of α-synuclein misconformers by molecular chaperones and the ubiquitin–proteasome system [11].

Synphilin-1 as an interacting partner and aggregation seeder of α-synuclein

The presynaptic protein synphilin-1 was originally identified as an interacting partner of α-synuclein using a yeast-two-hybrid screening [12]. Meanwhile, this interaction has been confirmed by several in vitro and in vivo studies [5,13]. As for α-synuclein, the physiological function of synphilin-1 is unknown, but given the presynaptic localization and its affinity to lipids and membranes [14], it is plausible that synphilin-1 also plays a role in synaptic vesicle dynamics [15]. Synphilin-1 contains a central coiled-coil domain allowing the protein to interact with the N-terminal domain of α-synuclein and to promote α-synuclein inclusion formation in mammalian cells [16]. Overexpression of synphilin-1 alone also induces inclusion formation in cellular models [14] and dopaminergic neurons of transgenic mice [17].

A number of studies provided insight into the mechanisms that regulate the interaction between synphilin-1 and α-synuclein and their tendency to form inclusions. Phosphorylation of synphilin-1 by protein kinase CK2 was shown to stimulate the interaction with α-synuclein and to enhance inclusion formation [18]. In addition, phosphorylation of synphilin-1 by GSK3β (glycogen synthase kinase 3β) was reported to decrease ubiquitination of the protein and its subsequent degradation by the ubiquitin–proteasome system. Consistently, pharmacological inhibition of GSK3β or mutation of its phosphorylation site on synphilin-1 strongly increased the formation of inclusions in HEK (human embryonic kidney)-293 cells when treated with proteasome inhibitors [19]. Synphilin-1 is ubiquitylated by different E3 ubiquitin-ligases, including parkin, dorfin and SIAH (seven in absentia homologue) [2022]. This ubiquitination appears to be crucial for inclusion formation, since a catalytically inactive SIAH-1 mutant is unable to trigger the formation of synphilin-1 inclusions in the presence of proteasome inhibitors [22]. Recent studies identified a synphilin-1 isoform, known as synphilin-1A, which has enhanced aggregation properties [23]. This isoform acts as regulator of SIAH as it decreases the auto-ubiquitination and degradation of SIAH. In addition, synphilin-1A prevents SIAH-dependent mono-ubiquitination of substrates such as α-synuclein, thereby reducing α-synuclein inclusion formation in a mammalian cell line model [24].

Are inclusions toxic or cytoprotective?

Although inclusion formation is considered to be a common pathological hallmark of many neurodegenerative disorders, opinions concerning the toxic nature of these inclusions are still diverse. While there are various studies linking the presence of α-synuclein inclusions to toxicity, other studies provided evidence for a protective role of these inclusions. A plausible explanation is that mature inclusions and aggregates represent an end stage of several events and that their formation may help to sequester soluble misconformers that are otherwise neurotoxic [25]. In line with this, there are studies demonstrating that inclusions of synphilin-1 are beneficial, rather than harmful, as they elicit a positive effect on viability of human cell lines [14]. However, when it comes to in vivo studies in brains of transgenic mice, the situation is far from clear. One study found no signs of neurodegeneration despite the presence of insoluble synphilin-1 deposits [26], while other studies demonstrated that the presence of ubiquitin-positive synphilin-1 inclusions coincided with cell loss in the cerebellum [27] and dopaminergic neurons [17]. Most recently, a double-transgenic model was generated with combined expression of synphilin-1 and the A53T α-synuclein mutant and in this case synphilin-1 inclusions appeared to attenuate α-synuclein-induced neuronal demise [28].

Yeast models to study PD

Over the years, several higher organisms, including rat, mouse, fly and worm, have been used as models to study the pathobiology of proteins associated with PD and other neurodegenerative disorders. More recently, the budding yeast, Saccharomyces cerevisiae, has also been validated as a powerful model system, thereby providing an unsurpassed tool to decipher disease-associated cellular processes and identify novel therapeutic targets using genome-wide screenings [2934]. The rationale for this is that although yeast is a unicellular organism, the basic molecular machineries necessary for cellular functioning and the execution of various cell death modalities are highly conserved between humans and yeast [35,36]. In fact, about 60% of the yeast genes show sequence homology with human genes and, conversely, approximately 25–30% of the positionally cloned human disease genes have a homologue encoded by the yeast genome [36].

As the yeast genome does not encode an orthologue of α-synuclein, humanized yeast models have been developed based on heterologous expression of human wild-type α-synuclein and clinical mutants. These models not only faithfully recapitulated several features of PD, but also allowed us to refine processes and identify novel players involved in mediating α-synuclein-instigated cytotoxicity [3739]. To briefly summarize, expression of α-synuclein in yeast inhibits growth and induces cell death in a concentration-dependent manner [32]. This coincides with the formation of α-synuclein inclusions at the plasma membrane [32,34,40], the obstruction of vesicular trafficking, endocytosis and vacuolar degradation [34,4043] as well as the induction of oxidative stress, mitochondrial dysfunction and apoptosis [29,31,44].

We recently reported on humanized yeast models to investigate the importance of synphilin-1 in PD [45]. In these models, synphilin-1 was expressed alone or together with α-synuclein in yeast cells. In the case of the latter, synphilin-1 moderately aggravated the cytotoxic effect of α-synuclein, which is in accordance with data obtained in mammalian cells [16] and coincided with enhanced α-synuclein aggregation. Also when synphilin-1 was expressed alone it induced only a moderate growth defect during the exponential phase (Figure 1A) and this despite the observation that it readily formed inclusions in about one-third of the cells (Figure 1B). As this value is more than ten times higher than the number of cells displaying inclusions upon cytotoxic expression of α-synuclein, the data suggested that synphilin-1 and α-synuclein are being processed via different cellular pathways. This was further evidenced when analysing the localization and inclusion formation of both proteins. Synphilin-1 localized mainly in the yeast cytoplasm, though it formed small foci that co-localized with lipid droplets and lipid rafts in endomembranes and that transformed into larger inclusions when cells approached the diauxic shift. In contrast, α-synuclein localized and formed foci at the plasma membrane and exhibited only limited interaction with lipid rafts [45]. Furthermore, the pattern of inheritance by the daughter cells differed for the two proteins, as emerging buds often received cytoplasmic synphilin-1 inclusions from the mother cells, while α-synuclein was transmitted almost exclusively as plasma membrane-associated protein (Figure 1B). Further analysis demonstrated that the growing synphilin-1 inclusions represented aggresomes, which are formed by convergence of smaller aggregates via microtubule-based transport and are generally believed to be cytoprotective because they can be easily cleared via autophagy [46]. Interestingly, the yeast synphilin-1 model provided evidence that aggresome formation also depends on actin-mediated transport (Figure 2A) and probably involves the sirtuin Sir2 as further discussed below [45]. This is reminiscent of the model recently proposed by Nyström and co-workers [47] where misfolded and damaged proteins are subject to retrograde transport from the daughter cells back to the mother cells along polarisome-anchored actin cables (Figure 2B). Moreover, our previously reported genome-wide screen to identify yeast deletion mutants displaying enhanced aggregation of α-synuclein [34] retrieved the major components of the polarisome as well as the GimC/prefolding complex (Figure 2C), which is required for efficient transfer and folding of newly synthesized actin and tubulin by the chaperonin TRiC/CCT [48]. This suggests that α-synuclein misconformers might also, at least to some extent, be subject to retrograde transport and clearance from the daughter cells.

Figure 1 Properties of yeast strains expressing α-synuclein or synphilin-1

(A) Growth of the BY4741 wild-type strains transformed with an empty plasmid (□) or constructs allowing for expression of α-synuclein (■), synphilin-1 (○) or a combination of both proteins (●). The data represent the mean of at least three independent transformants. (B) Images of exponential BY4741 cells expressing the dsRed-synphilin-1 (dsRed-SY) or the α-synuclein–enhanced green fluorescent protein (α-Syn–EGFP) fusion proteins showing that daughter cells inherit cytosolic synphilin-1 inclusions and plasma membrane-associated α-synuclein. The percentages refer to the number of cells with or without inclusions in an exponential culture. (C) Relative quantification of viable cells able to form colonies at 48 h after inoculation of the BY4741 wild-type strain or the sir2Δ mutant transformed with empty plasmids or constructs, allowing for expression of α-synuclein (α-Syn) or synphilin-1 (SY), either alone or in combination as indicated. The number of viable cells in samples of the two strains transformed with the empty plasmids was set at 100%. (D) Visualization and (E) quantification of cells after 36 h of growth expressing α-synuclein (α-Syn), synphilin-1 (SY) or the combinations, as indicated, and stained with DHE (dihydroethidium) (D, upper panels) to demonstrate accumulation of ROS or co-stained with annexin V and PI (propidium iodide) (D, lower panels) to identify cells that display phosphatidylserine externalization or loss of membrane integrity. See [45] for more detailed information. DIC, differential interference contrast.

Figure 2 Actin-mediated retrograde transport of α-synuclein or synphilin-1 inclusions

(A) Images of late exponential wild-type cells expressing the dsRed-synphilin-1 (dsRed-SY) fusion stained with Alexa Fluor® 488–phalloidin to visualize actin patches and actin fibres and with Calcofluor to visualize the cell wall. See [45] for more details. (B) Model proposed by Nyström and co-workers [47] for retrograde transport of inclusions along actin cables. (C) Images of inclusions formed by the synuclein–EGFP (enhanced green fluorescent protein) fusion protein in exponential BY4741 cells and mutant cells lacking components of the polarisome or the GimC/prefolding complex. The percentages refer to the number of cells with inclusions. See [34] for more detailed information.

As many neurodegenerative disorders are associated with aging of post-mitotic cells, the behaviour of synphilin-1 was also studied in stationary phase and aging yeast cultures [45]. In contrast with exponentially growing cells, synphilin-1 confers toxicity and dramatically reduces the survival of aging cells to an extent comparable to that triggered by α-synuclein (Figure 1C). Consistently, these aging cells display an increment of oxidative stress and show signs of apoptotic and necrotic cell death (Figures 1D and 1E). As several age-related phenomena in higher and lower eukaryotes are controlled by the activity of sirtuins, the involvement of the yeast sirtuin Sir2 was further examined. This analysis demonstrated that synphilin-1 largely failed to induce additional toxicity in aging sir2Δ cells (Figure 1C), indicating that synphilin-1 exerts its toxic effect via Sir2-dependent processes. Also the toxicity of α-synuclein was reduced in sir2Δ cells, albeit to a lesser extent than that of synphilin-1 (Figure 1D) [45]. Although it will require further investigation to decipher the exact role of Sir2 in conveying synphilin-1 and α-synuclein toxicity in aging yeast cells, it is likely that this relates to accelerated aging and failure to segregate dysfunctional proteins in cells lacking Sir2 [49], which is due to a requirement of Sir2 to maintain the activity of the TRiC/CCT complex [47]. Alternatively, or additionally, the role of Sir2 may also relate to its involvement in maintaining the autophagic flux [50].

Funding

This work was supported by a Marie Curie International Reintegration Grant, an EMBO Installation Grant and grants from the Fundação para a Ciência e Tecnologia (to T.F.O.), by the Austrian Science Fund FWF (to S.B. and F.M.) [grant numbers T414-B09 and LIPOTOX], by an EU-Marie Curie PhD Graduate School NEURAD grant and a Tournesol grant from Egide (Partenariat Hubert Curien), in collaboration with the Flemish Ministry of Education (to M.-C.G. and J.W.), by the Flemish society Alzheimer Research (SAO) (to V.F.) and by the IWT-Vlaanderen (SBO Neuro-Target and Baekeland), the KULeuven Research Fund and the Fund of Scientific Research of Flanders (FWO) (to J.W.).

Footnotes

  • 8th International Meeting on Yeast Apoptosis: An Independent Meeting held at Keynes College, University of Kent, Canterbury, U.K., 2–6 May 2011. Organized and Edited by Paula Ludovico (University of Minho, Braga, Portugal).

Abbreviations: CCT, chaperonin containing TCP-1; GSK3β, glycogen synthase kinase 3β; PD, Parkinson's disease, ROS, reactive oxygen species; SIAH, seven in absentia homologue; TRiC, TCP-1 ring complex

References

View Abstract