There is growing evidence supporting a role of extracellular alpha-synuclein in the spreading of Parkinson's disease (PD) pathology. Recent pathological studies have raised the possibility that the enteric nervous system (ENS) is one of the initial sites of alpha-synuclein pathology in PD. We therefore undertook this survey to determine whether alpha-synuclein can be secreted by enteric neurons. Alpha-synuclein secretion was assessed by immunoblot analysis of the culture medium from primary culture of ENS. We show that alpha-synuclein is physiologically secreted by enteric neurons via a conventional, endoplasmic reticulum/Golgi-dependent exocytosis, in a neuronal activity-regulated manner. Our study is the first to evidence that enteric neurons are capable of secreting alpha-synuclein, thereby providing new insights into the role of the ENS in the pathophysiology of PD.
Alpha-synuclein, a 140-amino acid neuronal protein, attracted great interest since 1997 after a mutation in its gene was identified in autosomal dominant Parkinson's disease (PD) (Polymeropoulos et al. 1997), and its aggregates were found to be the primary components of Lewy bodies, the pathological hallmarks of PD (Spillantini et al. 1997). Although initially considered as an intracellular protein, recent reports have shown that alpha-synuclein also exerts its effects extracellularly (Borghi et al. 2000; El-Agnaf et al. 2003). Alpha-synuclein can be secreted into the culture medium of differentiated human neuroblastoma cells and primary cortical neurons (Lee et al. 2005) and detected in human cerebrospinal fluid and plasma (El-Agnaf et al. 2003). Recent reports have shown that alpha-synuclein can be directly transferred from nerve cells that over-express the protein to neighboring cells both in tissue culture and in transgenic animals (Desplats et al. 2009; Hansen et al. 2011). This suggests that alpha-synuclein has the potential to spread from one nerve cell to another, thereby contributing to the diffusion of neuropathology from one brain region to the next (Angot et al. 2010).
The distribution of alpha-synuclein pathology in PD is much greater than formerly appreciated. Lewy bodies distribution extend well beyond the substantia nigra and involves peripheral nervous networks, especially the enteric nervous system (ENS) (Beach et al. 2009). Braak and colleagues have determined that the appearance of alpha-synuclein aggregates occurs in the ENS at the earliest stage of the disease, leading to the assumption that PD pathology may in fact begin in the gastrointestinal tract, further spreading to the central nervous system (CNS) via the vagal pre-ganglionic innervation of the gut (Braak et al. 2006). Alpha-synuclein expression in the ENS is restricted to a subset of enteric neurons that are synaptically linked with alpha-synuclein-positive vagal neurons, thus providing a candidate alpha-synuclein-expressing pathway for the retrograde transport of PD pathology (Phillips et al. 2008). In this context, it is crucial to determine whether enteric neurons are capable of secreting alpha-synuclein. The aim of this survey was therefore to study the secretion of alpha-synuclein in primary culture of ENS.
Material and methods
Reagents and antibodies
Forskolin, veratridine, brefeldin A (BFA) were purchased from Sigma (Saint Quentin Fallavier, France). Recombinant alpha-synuclein was obtained from Millipore (Molsheim, France). The following commercially available antibodies were used for western blotting: rabbit polyclonal anti-alpha-synuclein C-20 (1 : 500; Santa Cruz Biotechnology, Heidelberg, Germany), mouse monoclonal anti-alpha-synuclein Syn-1 (1 : 500; BD Bioscience, Le Pont-De-Claix, France), mouse monoclonal anti-ubiquitin P4D1 (1 : 1000; Cell Signaling, Ozyme, Saint Quentin en Yvelines, France), rabbit polyclonal anti-bovine serum albumin (BSA) (1 : 1000; Millipore), and mouse monoclonal anti-Protein Gene Product (PGP) 9.5 (1 : 1000; Ultraclone limited, Isle of Wight, UK). Mouse monoclonal anti-alpha-synuclein Syn-1 and rabbit polyclonal anti-alpha-synuclein C-20 were used for ELISA.
Primary cultures of ENS
Primary cultures of ENS were performed as previously described using small intestine of E15 rat embryos (Paillusson et al. 2010).
Culture medium and cell lysates western blots
The culture medium (CM) was collected and centrifuged at 4000 g for 10 min at 4°C to remove cell debris. The CM was further concentrated at 1800 g for 28 min using 9-kDa-cutoff concentrators (Pierce, Brébières, France). Harvesting of primary cultures of ENS as well as western blotting analysis and quantification were performed as described previously (Paillusson et al. 2010). PGP9.5 and BSA were used as protein loading controls for cell lysates and CM samples, respectively.
Alpha-synuclein ELISA assay
For ELISA, immuno 96 MicroWell™ Solid Plates (Thermo Scientific, Brebières, France) were coated for 24 h at 20°C with 0.5 μg/mL of rabbit polyclonal anti-alpha-synuclein antibody C-20 in 100 mM NaHCO3 at pH 9.3. Plates were washed three times in wash buffer (50 mM Tris-HCl, 150 mM NaCl, and 0.04% v/v Tween-20) and blocked for 30 min with phosphate-buffered saline (PBS) containing 1% (w/v) BSA. Recombinant alpha-synuclein diluted in PBS (50 mM potassium phosphate; 150 mM NaCl; pH 7.2) containing 1% (w/v) BSA was used as standard. Concentrated CM or standards were added and plates were incubated at 37°C for 2.5 h. After washing three times with wash buffer, 50 μL of mouse monoclonal anti-alpha-synuclein Syn-1 (1 : 2500 in PBS with 1% w/v BSA) was added to each well and incubated for 1 h at 20°C. Then, wells were washed three times and filled with 50 μL of biotinylated goat anti-mouse antibody (1 : 2500; Invitrogen, Life Technologies, Saint Aubin, France) for 1 h followed by a washing step procedure. Finally, horseradish peroxidase-coupled streptavidin (1 : 1000; Thermo Scientific) was added for 30 min and after a three-times washing procedure, TMB substrate (BD Biosciences, Le Pont de Claix, France) was added to each well for 15 min at 20°C. The chemiluminescence was integrated for 1 s.
Cell death assays
Neuronal cell death was assessed by quantifying the release of neuron-specific enolase (NSE) into culture medium as described previously (Abdo et al. 2010).
All data are given as the mean ± standard error of the mean (SEM). Comparisons of means between groups were performed by Student's t-test for unpaired data or by analysis of variance followed by Dunnett's test. When data were not normally distributed, a Mann–Whitney test was performed. Differences were considered statistically significant if p < 0.05.
Alpha-synuclein is physiologically secreted by enteric neurons
Primary culture of ENS contains smooth muscle, enteric glial cells, and neurons. We have previously shown that only neurons express alpha-synuclein in this primary culture model (Paillusson et al. 2010). To determine whether enteric neurons secrete alpha-synuclein, culture medium (CM) was concentrated and immunoblotted with C-20 antibodies. Alpha-synuclein was detected in the CM as early as 3 h and accumulated over time (Fig. 1a and b), while the expression levels of intracellular alpha-synuclein remained constant (Fig. 1a). The identity of the band detected by alpha-synuclein antibodies was confirmed by comigration with recombinant alpha-synuclein (data not shown). The accumulation of alpha-synuclein in the CM was not explained by the membrane leakage caused by cell death as (i) the amount of NSE in the CM remained relatively stable throughout the 48-h period (8.7 ± 2.3 ng/mL at 6 h and 9.5 ± 1.6 ng/mL at 48 h; n = 5 p > 0.05), (ii) the cytosolic protein ubiquitin was not detected in the CM and did not accumulate over time (Fig. 1a). We quantified the levels of secreted alpha-synuclein in the CM after 48 h by ELISA. The concentration of alpha-synuclein was estimated to be 12.8 ± 3.7 pg/mL. Altogether, these results show that alpha-synuclein is physiologically secreted by enteric neurons.
Studies performed in SH-SY5Y cells over-expressing alpha-synuclein suggested that it is constitutively secreted via an unconventional secretory pathway because brefeldin A (BFA), a classical inhibitor of the endoplasmic reticulum (ER)/Golgi-dependent secretion pathway had no effect on alpha-synuclein secretion (Lee et al. 2005; Emmanouilidou et al. 2010). We have therefore assessed the effects of BFA in primary culture of ENS. Treatment of primary culture of ENS with 1 μg/mL BFA for 6 h significantly reduced alpha-synuclein release as compared with controls (Fig. 2a and b). By contrast, when the same treatment was applied to culture of primary cortical neurons, no changes in the levels of extracellular alpha-synuclein were observed (data not shown). Cell viability in primary culture of ENS as assessed by NSE release in the CM was not compromised following a 6-h treatment with BFA (8.9 ± 1.1 ng/mL for controls, 8.5 ± 3.7 ng/mL for BFA-treated cells, n = 3, p > 0.05). This suggests that alpha-synuclein secretion in enteric neurons occurs through a secretory mechanism likely to be dependent on ER/Golgi-related vesicular transport.
Another remarkable result obtained in SH-SY5Y cells was that the amount of released alpha-synuclein in the CM correlated with its intracellular expression levels (Lee et al. 2005; Emmanouilidou et al. 2010). To address this issue we used forskolin, as we have previously demonstrated that such a treatment was capable of increasing intracellular alpha-synuclein expression in enteric primary culture without any deleterious effect on cell survival (Paillusson et al. 2010). We show in this survey, that the 2.7-fold increase in intracellular alpha-synuclein expression levels evoked by 20 μM forskolin for 48 h was not associated with changes in the amount of extracellular alpha-synuclein (Fig. 2c and d). These results are in sharp contrast with that obtained in neuronal cell lines (Lee et al. 2005; Emmanouilidou et al. 2010) and suggest that the release of alpha-synuclein in the CM by enteric neurons is, at least partially, regulated.
Alpha-synuclein secretion is regulated by neuronal activity
Given that alpha-synuclein seems to be secreted through a regulated mechanism dependent on classical vesicular transport, we tested whether neuronal activity regulates its secretion. A 48-h treatment of primary culture of ENS with the sodium channel activator veratridine at 30 μM significantly increased the release of alpha-synuclein in the CM as compared with controls (Fig. 3a and b). This was associated with a statistically significant drop in the level of intracellular alpha-synuclein (Fig. 3a and b). Treatment of primary culture of ENS with the sodium channel blocker tetrodotoxin at 1 μM for 48 h induced a significant increase in intracellular alpha-synuclein expression as compared with controls (Fig. 3a and b). In parallel, a decrease in alpha-synuclein level in the CM was observed, but it did not reach significance. Under veratridine and tetrodotoxin treatments, cell viability as assessed by the release of NSE, was not compromised (11.1 ± 2.4 ng/mL for controls, 14.5 ± 1 ng/mL for veratridine-treated cells, n = 3, p = 1; 20.6 ± 17 ng/mL for controls, 35.2 ± 10.5 ng/mL for tetrodotoxin-treated cells, n = 3, p = 0.4).
To further investigate the role of neuronal activity, we studied whether intracellularly accumulated alpha-synuclein can be secreted. To this end, primary culture of ENS were treated with 20 μM forskolin then depolarized with veratridine. Treatment with veratridine induced the secretion of alpha-synuclein that accumulated following forskolin treatment (Fig. 3c and d).
Taken as a whole, these results demonstrate that alpha-synuclein secretion in enteric neurons is regulated by neuronal activity.
We have shown in this survey that alpha-synuclein is physiologically secreted by enteric neurons and that its secretion is regulated by neuronal activity.
Because of the lack of an ER signaling peptide from its sequence, alpha-synuclein was considered to be an exclusive intracellular protein. Using SH-SY5Y cells over-expressing alpha-synuclein, two different groups have convincingly shown that alpha-synculein can be secreted in the extracellular space, thereby affecting neighboring neurons (Lee et al. 2003; Emmanouilidou et al. 2010). We show in this report that alpha-synuclein can also be secreted from peripheral and more particularly from enteric neurons. The levels of secreted alpha-synuclein in primary culture of ENS were similar to those described in biological fluids and in tissue (Emmanouilidou et al. 2011). In accordance with the previous reports, secreted alpha-synuclein in primary culture of ENS accumulated over time and was not attributable to cell death as measured by the release of NSE in the extracellular space. This suggests that at least a portion of alpha-synuclein from enteric neurons is secreted in a constitutive manner or following spontaneous enteric neuronal activity.
This led us to investigate the mechanisms of alpha-synuclein secretion by enteric neurons and especially the role of neuronal activity. It has been suggested previously that alpha-synuclein exocytosis is mediated by a non-classical, BFA-independent, secretory mechanism (Lee et al. 2005; Emmanouilidou et al. 2010). In sharp contrast with these two previous reports, we show in this survey that treatment of primary culture of ENS with BFA resulted in a significant decrease in alpha-synuclein secretion, strongly suggesting that enteric alpha-synuclein follows a conventional secretory pathway. This is reinforced by our observations showing a tight regulation of alpha-synuclein release by neuronal activity, which is known to regulate conventional exocytosis. In support of this idea, a portion of pre-synaptic alpha-synuclein has been shown to be present within synaptic vesicles either in SH-SY5Y cells or in rat brain (Lee et al. 2005; Lee 2008), and neuronal activity controls the pre-synaptic accumulation of alpha-synuclein (Fortin et al. 2005). Our study is the first to directly address the regulation of alpha-synuclein secretion by depolarization. Although Lee and collaborators elegantly demonstrated that alpha-synuclein was localized within vesicle lumen, they did not mention the effects of depolarization, probably because alpha-synuclein secretion was BFA independent in their cell system (Lee et al. 2005). More recently, Emmanouilidou et al. showed that a portion alpha-synuclein is secreted via an exosomal calcium-dependent mechanism in SH-SY5Y cells (Emmanouilidou et al. 2010). Nevertheless, as stated in their discussion, it is not possible to rule out the possibility that alternative mechanisms for alpha-synuclein secretion, such as secretory vesicle-mediated exocytosis that is also calcium-sensitive, may also operate. Altogether, our results and the available data on alpha-synuclein secretion show that alpha-synuclein secretion from neurons is likely to occur through several pathways, either following conventional or unconventional exocytosis.
Our findings may be relevant to the pathogenesis of PD. Although precise etiology of the disease remains unknown, it is suggested that, besides genetic factors or in combination with, environmental factors could be critically involved. Some recent evidences suggest that the pathological process of PD affects the ENS at a very early stage of the disease (Braak et al. 2006). Remarkably, the enteric neurons are directly in contact with the environment, leading to the postulate, the so-called Braak's hypothesis, that they could represent a route of entry for an hitherto unknown environmental factor to initiate the pathological process further spreading to the CNS and more precisely to the dorsal motor nucleus of the vagus via the vagal pre-ganglionic innervation of the gut (Braak et al. 2002, 2006). If Braak's theory is true, two necessary conditions must be satisfied. First, an uninterrupted pathway that expresses alpha-synuclein throughout its trajectory should allow the retrograde transport of the pathological process from the gastrointestinal tract to the CNS. Second, enteric neurons should be able to secrete alpha-synuclein to transmit the pathological process from cell to cell as suggested for CNS neurons (Hansen et al. 2011). The first condition is fulfilled as Phillips and coworkers have elegantly shown that vagal efferent axons and terminals, which originate from the dorsal motor nucleus of the vagus, are positive for alpha-synuclein and that some of these pre-ganglionic efferent neurons synapse on alpha-synuclein-positive intrinsic neurons in the myenteric plexus of both the stomach and duodenum (Phillips et al. 2008). Regarding the second condition, our results allow us to consider that alpha-synuclein could behave in the ENS like in the CNS and thus transmit the pathology from neuron to neuron (Angot et al. 2010).
This work was supported by a grant from France Parkinson and CECAP (Comité d'Entente et de Coordination des Associations de Parkinsoniens). Work in Michel Neunlist's lab is supported by Michael J Fox Foundation for Parkinson's Research, Fondation de France, France Parkinson, ADPLA (Association des Parkinsoniens de Loire Atlantique), FFPG (Fédération française des groupements parkinsoniens), Parkinsoniens de Vendée. SP is a recipient of a ‘bourse MENRT’. The authors declare no conflicts of interest.