Assessing the Subcellular Dynamics of Alpha-synuclein Using Photoactivation Microscopy

Alpha-synuclein (aSyn) is implicated in Parkinson’s disease and several other neurodegenerative disorders. To date, the function and intracellular dynamics of aSyn are still unclear. Here, we tracked the dynamics of aSyn using photoactivatable green fluorescent protein as a reporter. We found that the availability of the aSyn N terminus modulates its shuttling into the nucleus. Interestingly, familial aSyn mutations altered the dynamics at which the protein distributes throughout the cell. Both the A30P and A53T aSyn mutations increase the speed at which the protein moves between the nucleus and cytoplasm, respectively. We also found that specific kinases potentiate the shuttling of aSyn between nucleus and cytoplasm. A mutant aSyn form that blocks S129 phosphorylation, S129A, results in the formation of cytoplasmic inclusions, suggesting phosphorylation modulates aggregation in addition to modulating aSyn intracellular dynamics. Finally, we found that the molecular chaperone HSP70 accelerates the entry of aSyn into the nuclear compartment.


Introduction
Misfolded and aggregated alpha-synuclein (aSyn) is the major component of intraneuronal inclusions known as Lewy bodies (LBs), the pathological hallmark of Parkinson's disease (PD) and other synucleinopathies [1]. Despite the growing knowledge on aSyn, the normal function of the protein remains largely unclear. However, it is thought to play a role in synaptic function and plasticity, cell differentiation and vesicular trafficking [2,3]. The subcellular distribution of aSyn is also controversial and, although it is considered a pre-synaptic protein, it has also been found to be evenly distributed throughout the cells in different cellular models and in mice [4][5][6][7][8].
Although the majority of PD cases are idiopathic, three missense mutations in aSyn gene (A30P, E46K and A53T), restrained in the N-terminal domain, have been identified in rare, autosomal-dominant inherited forms of PD, as well as duplications and triplications of the aSyn-containing locus [9][10][11][12]. In vitro studies revealed that the A30P mutation blocks the membrane association and inhibits the synaptic localisation of aSyn by destabilising its first helical structure [5,13]. Conversely, A53T and E46K mutations enhance the binding to phospholipids [14].
In the normal brain, 4 % of aSyn is phosphorylated at serine 129 (S129), contrasting with 90 % of aSyn that is found to be phosphorylated in LBs. This suggests that S129 phosphorylation might interfere with the oligomerisation and aggregation process and contribute to the pathogenesis of PD [15,16]. Among others, G protein-coupled receptor kinases (GRKs) and Polo-like kinases (PLKs) were found to phosphorylate the S129 residue of aSyn [17,18]. Although it was proposed that S129 phosphorylation inhibits aSyn-induced regulation of tyrosine hydroxylase activity [19], the exact role of this posttranslational modification (PTM) in both physiological and pathological conditions remains unclear.
Other known modifiers of aSyn aggregation are molecular chaperones, such as HSP70, which modulates the misfolding, aggregation and toxicity of aSyn in different model systems [4,[20][21][22][23]. However, the mechanisms by which HSP70 suppresses aSyn toxicity are still unclear.
The new era of time-lapse bioimaging tools combined with GFP-derived fluorescent labels enables the characterisation of protein kinetics in real time, providing invaluable insights into the molecular processes in which they are involved. Photoactivation (PA) microscopy is an emerging technique in the field of neuroscience [24] in which a non-fluorescent molecule is converted into an activated and fluorescent state by the use of an intense and brief irradiation in a selected region of the cell. This process enables the direct tracking of a protein by photo-inducing fluorescence, instead of interfering with a steady state fluorescent signal, as photobleaching (PB) methods do. The newly activated pool, obtained through the use of an ultraviolet laser, contrasts with a background of nonactivated molecules and can be followed within the cells as they reach their kinetics equilibrium [25,26] (Fig. 1a).
In order to further understand the biology of aSyn, we investigated the dynamics of aSyn between the nucleus and cytoplasm in living cells using PA microscopy (Fig. 1b). We found that the N-terminal of aSyn-wild-type (WT) determines its entry into the nuclear compartment. Moreover, aSyn shuttles between the nucleus and cytoplasm at rates which depend on mutations, phosphorylation state and on the presence of HSP70. Altogether, our novel approach provides novel insights into the biology of aSyn in living cells and may enable the development of novel strategies for therapeutic intervention in synucleinopathies.

Blocking the N-terminal of aSyn Modulates Nuclear Localisation
To assess whether the subcellular localisation of aSyn-WT is affected by appending different tags to either the N or C terminus, we performed immunocytochemistry analysis in cells expressing either untagged aSyn-WT or GFP, Myc-or V5tagged versions. Both tagged and non-tagged aSyn-WT were widely distributed throughout the cell, including the nucleus (Fig. 2a-d). To further investigate the intracellular dynamics of aSyn-WT, we generated fusions with a photoactivatable green fluorescent protein (PAGFP) in order to follow the movement of a specific pool of aSyn over time at the N or C termini (aSyn-WT-PAGFP and PAGFP-aSyn-WT, respectively). H4 cells were transiently transfected with plasmids encoding aSyn-WT-PAGFP, PAGFP-aSyn-WT or PAGFP alone, as a control. As aSyn is widely distributed in the cell, we characterised the shuttling of aSyn between the nucleus and the cytoplasm. Reporter proteins were photoactivated in the nucleus or cytoplasm for 2 s using a 405-nm laser, their cellular trafficking was monitored and the fluorescence intensities quantified. After PA, the PAGFP control was quickly detected in the cytoplasm or in the nucleus after nuclear or cytoplasmic PA, respectively, reaching an equilibrium of PAGFP molecules between the two cellular compartments after 500 s. In contrast, we observed different trafficking behaviours for aSyn-WT-PAGFP and PAGFP-aSyn-WT (Figs. 2e and S1a; Videos S1, S2, S3, S4, S5 and S6). aSyn-WT-PAGFP displayed different dynamics depending on the region where PA was performed. Upon cytoplasmic PA, aSyn-WT-PAGFP molecules entered into the nucleus and after 1,000 s were evenly distributed between the two subcellular compartments. Conversely, upon nuclear PA, aSyn-WT-PAGFP was maintained in this compartment for the remainder of the time analysed (1,000 s). In contrast with the behaviour observed for photoactivated aSyn-WT-PAGFP, PAGFP-aSyn-WT was not detected in the contiguous compartment to where the PA was performed and remained in the same subcellular region (Figs. 2e and S1a). Immunoblotting analysis of nuclear and cytoplasmic extracts from cells expressing either protein confirmed the predominant localisation observed with microscopy ( Fig. S2a). To further validate the observations obtained using PA, we performed fluorescence PAGFP-labelled proteins become visible and the dynamics and fate of the activated molecules can be followed over time. b PAGFP constructs used in this study. We used WT, A30P, E46K, A53T and S129A aSyn variants recovery after PB (FRAP) experiments in H4 cells expressing aSyn-WT-GFP, GFP-aSyn-WT or GFP alone. Upon PB of aSyn-WT-GFP in the nucleus, we found that the recovery of fluorescence in this compartment occurred after 500 s while fluorescence was not significantly recovered in the cytoplasm after PB in this region. For GFP-aSyn-WT, PB in nucleus slightly recovered fluorescence of the reporter protein after 1,000 s (about 20 % of recovery) while no cytoplasmic recovery was observed upon PB in this region ( Fig. 2f; Videos S7, 8 and 9).
Altogether, our novel PA and FRAP experiments show, for the first time, that the movement of aSyn between the nucleus and the cytoplasm depends on the availability of the N terminus of the protein. Since we established a model to study aSyn intracellular dynamics, we next investigated whether PD familial mutations (A30P, E46K and A53T) in aSyn affected its trafficking in the cell (Fig. 3a). In comparison to the aSyn-WT-PAGFP, cytoplasmic aSyn-A30P-and aSyn-E46K-PAGFP are shuttled into the nucleus in half of the time (500 s) of the WT protein (at time points 500 and 1,000 s after PA, p values between 0.0044 and <0.0001, respectively, α=0.05; Table SIa). Because of this rapid movement into the nucleus, the fluorescence intensities in the cytoplasm and nucleus equalised earlier than in the control situation and remained in equilibrium for several minutes. In contrast to A30P and E46K mutants, aSyn-A53T-PAGFP remained in the compartment where the PA was performed, similarly to the WT protein.
In the fusions where PAGFP was on the N terminus of aSyn, the behaviour of A30P was similar to that of aSyn-WT. Additionally, upon cytoplasmic PA, there was a slight increase in molecules that remained on the nucleus (p value= Immunobloting analysis of total aSyn levels in cells expressing aSyn-WT and mutant-aSyn reporter proteins 0.0185 at 1,000 s after nuclear or cytoplasmic PA, α=0.05; Table SIa; Fig. 3a). In contrast, PAGFP-aSyn-E46K was translocated from the nucleus to the cytoplasm after nuclear PA (p value=0.001 at 1,000 s after nuclear PA, α=0.05, Table SIa). The same occurred with PAGFP-aSyn-A53T, although at a more pronounced rate, since the fluorescence intensity was higher in the cytoplasm than in the nucleus, starting at 700 s after PA (p value=0.0013 at 1,000 s after PA, α=0.05; Table  SIa; and Fig. 3a). In order to investigate if the dynamics results were influenced by differences in the levels of WT and mutant aSyn, we performed immunobloting analysis. We verified that the levels of total aSyn did not differ between the WT and mutant forms of the protein (Fig. 3b). As expected, subcellular fractionation followed by immunobloting analysis confirmed the presence of the protein in both cytoplasmic and nuclear compartments (Fig. S2b).
Altogether, these experiments show that A30P and E46K, but not the A53T mutation, promoted a faster shuttling of aSyn with a free N terminus into the nucleus when compared with aSyn-WT. On the other hand, aSyn-A53T and aSyn-E46K with a free C terminus, but not aSyn-A30P, were delayed in the cytoplasm.
aSyn Phosphorylation by GRK5 or PLK2 Modulates Its Subcellular Trafficking In order to assess the role of S129 aSyn phosphorylation on the intracellular dynamics of the protein, we co-expressed GRK2, GRK5, PLK2 or PLK3 kinases with aSyn-WT tagged with PAGFP on either its N-or C-terminal. The kinases tested did not significantly alter the dynamics of aSyn-WT-PAGFP. For aSyn-WT-PAGFP, co-expression with PLK2 did not alter the dynamics obtained in the absence of the kinase, but resulted in a more rapid progression towards the equilibrium fluorescence. The other kinases tested, GRK2, GRK5 and PLK3, neither did not induce significant differences on aSyn-WT-PAGFP (Table SIb; Fig. S3).
In contrast, stronger effects were observed for PAGFP-aSyn-WT. Overexpression of GRK2 did not affect the shuttling of aSyn-WT constructs between the nucleus and the cytoplasm as the values of fluorescence intensity were similar in the presence or absence of the kinase during the 1,000 s of imaging. Interestingly, GRK5 promoted the trafficking of PAGFP-aSyn-WT into the nucleus upon cytoplasmic PA, although the difference did not reach statistical significance. Nuclear PA in cells overexpressing GRK5 did not significantly alter the dynamics of PAGFP-aSyn-WT ( Fig. 4a; Table SIb).
Co-expression of PAGFP-aSyn-WT with PLK2 promoted its shuttling to the cytoplasm upon nuclear PA (similar nuclear and cytoplasmic fluorescence levels were reached 500 s after PA, p value=0.0141, α=0.05; Table SIb; Fig. 4a). Upon cytoplasmic PA, PLK2 slightly accelerated the movement into the nucleus 500 s after PA (Table SIb).
Co-expression of PAGFP-aSyn-WT with PLK3 promoted its shuttling to the cytoplasm upon nuclear PA but this movement was faster in the presence of PLK2. Although the difference was already significant 500 s after PA (p value= 0.0198, α=0.05; Table SIb), at 1,000 s the levels of fluorescence were still higher in the nucleus than in the cytoplasm. Upon cytoplasmic PA, the presence of PLK3 did not alter the dynamics of PAGFP-aSyn-WT ( Fig. 4a; Table SIb). We performed immunobloting analysis to investigate if the results on the dynamics of aSyn were influenced by differences in expression in the presence and absence of the kinases tested. We verified that the levels of total aSyn were not altered in the presence of the kinases (Fig. 4b). As expected, we also confirmed that the levels of aSyn phosphorylated at S129 were increased in the presence of both kinases (Fig. S2c).
An important difference between GRK5 and PLKs was that the former induced a tendency of aSyn to traffic from the cytoplasm to the nucleus and the latter had a strong effect in promoting the trafficking of aSyn from the nucleus to the cytoplasm.
Next, we tested the dynamics of S129A aSyn, a phosphorylation-incompetent mutant, in order to further assess the effect of S129 phosphorylation in the trafficking of aSyn. No differences were observed on the dynamics of aSyn-S129A-PAGFP except for the existence of more photoactivated protein in the nucleus when compared with aSyn-WT immediately after nuclear PA ( Fig. S3d; Table SIa). However, PAGFP-aSyn-S129A moved into the cytoplasm after 1,000 s of PA if PA was performed in the nucleus while the aSyn-WT did not. Moreover, PAGFP-aSyn-S129A remained in the cytoplasm if PA was performed in that compartment (p value= 0.0099, α=0.05; Table SIa; Fig. 4c). These findings were in agreement with the fact that both aSyn-S129A fusion proteins were only marginally detected in the nuclear protein fraction, in contrast to aSyn-WT (Figs. S2a, S2d and S3e), although the total protein levels of aSyn-WT and aSyn-S129A were comparable ( Fig. 4d; Fig. S3f).
Interestingly, we found that expression of either the N-or C-terminal S129A fusion proteins promoted the formation of cytosolic inclusions scattered around the nucleus (Fig. 4e;  Fig. S3g).
HSP70 Modifies the Trafficking of aSyn HSP70 modulates the accumulation of oligomeric and aggregated forms of aSyn in different model systems. Thus, we next asked whether HSP70 could interfere with the subcellular dynamics of aSyn. HSP70 did not change the intracellular dynamics of aSyn-WT-PAGFP (Table SIb; Fig. 5a). Conversely, upon cytoplasmic PA, PAGFP-aSyn-WT was shuttled into the nucleus in the presence of HSP70 within 100 s after PA (p value<0.0001, α=0.05; Table SIb; Fig. S1b; Fig. 5a). Although the total levels of the protein were not altered in the presence of the chaperone (Fig. 5b), the levels of aSyn were higher in the nucleus for both fusion proteins in this situation. Interestingly, HSP70 was present in the nuclear fraction only when aSyn was present (Fig. S2e). Upon nuclear PA, the tendency was for the protein to move into the cytoplasm, but not as quickly as in the former situation (p value at 100 s=0.7836, p value at 1,000 s=0.0014, α=0.05; Table SIb; Fig. S1b; Fig. 5a).
In summary, HSP70 increased the shuttling of PAGFP-aSyn-WT between the nucleus and cytoplasm.

Discussion
Here, we investigated the intracellular dynamics of aSyn in living cells using photoactivatable GFP as a reporter. To control for putative effects of tagging aSyn in particular domains, we engineered fusions with PAGFP on either the N-or C-terminal of aSyn. We found that, although both aSyn fusion proteins were evenly spread throughout the cell, aSyn required a free N terminus in order to move from the cytoplasm into the nucleus.
aSyn is evenly distributed throughout the cell in different in vitro models and in mice [4][5][6][7][8]. The N terminus of aSyn seems Fig. 4 Effect of S129 phosphorylation on the subcellular dynamics of aSyn. a Fluorescence intensities after PA in the nucleus (light grey line) and in the cytoplasm (dark grey line) of PAGFP-aSyn-WT fusion protein co-expressed with GRK2, GRK5, PLK2 and PLK3 (solid line) or an empty vector (dashed lines) over time. Fluorescence intensities of photoactivated control PAGFP-aSyn-WT construct after cotransfection with an empty vector are shown by a dashed line. Values represent mean+SD of up to15 cells analysed per condition. b Immunobloting analysis of total aSyn levels in cells co-expressing aSyn-WT reporter proteins with the four tested kinases. c Fluorescence intensities after PA in the nucleus (light grey line) and in the cytoplasm (dark grey line) of PAGFP-aSyn-S129A fusion protein over time. Fluorescence intensities of photoactivated control PAGFP-aSyn-WT are shown in dashed lines. Values represent mean+SD of up to15 cells analysed per condition. d Immunobloting analysis of total aSyn levels in cells expressing PAGFP-aSyn-WT or PAGFP-aSyn-S129A. e Cytosolic inclusions in cells expressing PAGFP-aSyn-S129A. Images were taken 500 s after PA in the nucleus in order to detect the cytosolic inclusions. Scale bar, 10 μm to be important for membrane binding in various model organisms, ranging from yeast to rats [14,[27][28][29]. Although the role of aSyn in the nucleus has not been defined, it is described to interact with histones, inhibiting acetylation and enhancing chromatin binding, and promoting neurotoxicity in cellular models, in mouse nigral neurons and in Drosophila [7,30,31].
We also investigated whether aSyn mutations associated with familial forms of PD altered the shuttling of aSyn between the nucleus and cytoplasm. We found that (a) the A30P mutant is more prone to be located in the nucleus than the aSyn-WT; (b) the E46K mutant loses the subcellular compartmentalisation characteristic of the WT form; and (c) the A53T mutation is more prone to be located in the cytoplasm than aSyn-WT.
Until recently, aSyn was thought to be an intrinsically unfolded protein [32][33][34]. Nevertheless, it acquires two αhelical structures upon interaction with vesicles, contained in the residues 1-42 and from 45 to 98 [35][36][37][38]. In vitro studies showed that A30P disrupts membrane binding [5,13], perhaps being more available to shuttle into the nucleus. Thus, it is likely that the differences in dynamics between the aSyn familial forms are related with the location and effect of the mutation on the secondary structure of aSyn protein. In addition, aSyn seems to regulate actin bundling inside the cell, and the A30P mutant affects the structure and dynamics of the actin cytoskeleton, potentiating the formation of actin foci [39]. Our results are also consistent with data showing that the A30P increases the nuclear localisation of the protein [30]. A53T is described to promote the formation of cytosolic aggregates [5,40], which is compatible with its tendency to be localised in the cytoplasm when compared with aSyn-WT.
PTMs are known to modulate the intracellular fate of proteins, including their sub-cellular distribution. aSyn is thought to have five residues prone to phosphorylation: S87, Y125, S129, Y133 and Y136. S129 phosphorylation is the most studied, and little information exists on the kinases phosphorylating the other residues [41,42]. In LBs, the majority of aSyn is thought to be phosphorylated on S129, contrasting with almost no phosphorylation of this residue in normal brain, However, the role of this PTM is still unclear and controversial. Phosphorylation of aSyn by GRKs inhibits its interaction with phospholipids [43]. In Drosophila, coexpression of GRK2 with aSyn leads to S129 phosphorylation and enhanced aSyn neurotoxicity [44]. Moreover, the levels of specific PLKs are increased in brains of patients with Alzheimer's or LB disease [45].
Here, we investigated whether a selected group of kinases, GRK2, GRK5, PLK2 and PLK3, modulated the dynamics of aSyn distribution in the cell. The kinases tested only affected the dynamics of aSyn with a free C terminus, although both fusion proteins were phosphorylated in S129. This can be due to the fact that when the C-terminal of aSyn is free the protein is more prone to phosphorylation at S129, resulting in a stronger effect in its intracellular dynamics. Overall, aSyn phosphorylation by GRKs or PLKs results in different dynamics of the protein. While GRK5 potentiates the nuclear localisation of aSyn, PLKs modulate the shuttling of the protein between the nucleus and cytoplasm. In particular, PLK2 modulates the intracellular dynamics of PAGFP-aSyn-WT by increasing the movement from the nucleus to the cytoplasm at a higher rate than PLK3. Since aSyn has more residues prone to phosphorylation, the different results obtained with GRK5 and PLKs might reflect different phosphorylation patterns in residues other than S129. Due to the limited availability of antibodies these studies are still not easy to perform but as novel tools become available one might be able to discriminate between the effects of phosphorylation in different residues.
Interestingly, we also observed that nuclear PAGFP-aSyn-S129A tends to move to the cytoplasm while cytoplasmic PAGFP-aSyn-S129A remains in this subcellular compartment. This tendency might at least partially explain the cytoplasmic inclusions detected in the cells expressing this mutant aSyn and suggest that the phosphorylation status on S129 is crucial for aggregation, in agreement with recent findings in yeast, in which S129A mutant potentiates the formation of aSyn foci [46]. GRKs and PLKs modulate the dynamics of aSyn in different ways and we did not find a consistent pattern that can explain the role of S129 phosphorylation on the distribution of aSyn. One possibility is that the effects of the kinases are also due to phosphorylation of other targets in addition to aSyn. Nevertheless, we verified that the phosphorylation status of aSyn on S129 was related with the aggregation state of the protein.
Molecular chaperones, such as HSP70, hold great potential as therapeutic targets due to their ability to reverse protein aggregation and to refold or promote degradation of misfolded proteins [47]. HSP70 was shown to inhibit formation of toxic pre-fibrillar forms of aSyn [23] and to reduce its aggregation in aSyn transgenic mice [22]. In flies, it was shown that coexpression of HSP70 with aSyn-WT suppresses the loss of dopaminergic neurons, and hence, the toxicity associated with aSyn-WT overexpression [20]. A similar effect was observed in yeast and in mammalian cell models, suggesting that HSP70 inhibits aSyn toxicity by binding to the exposed hydrophobic NAC domain (non-A beta component of AD plaques; residues 61-95 of aSyn) and sequestering the protein [21,[48][49][50].
Here, we found that HSP70 boosted the shuttling of PAGFP-aSyn-WT to the adjacent compartment, suggesting it may assist aSyn to adopt a conformation that is more likely to cross the nuclear envelope. The selective effect with PAGFP-aSyn-WT and not with aSyn-WT-PAGFP suggests the interaction might take place through the C terminus of aSyn, which is not blocked by PAGFP in this fusion protein.
In conclusion, we showed that PD-associated mutations in aSyn, S129 phosphorylation, and HSP70 exert different effects on aSyn trafficking within the cell (Table SII; Fig. 6). While additional studies will be important to clarify the relative contribution of each condition, our goal was to demonstrate the usefulness of PA microscopy for the study of aSyn dynamics in living cells, which is not possible to achieve with other types of approaches or with untagged protein.
Our data provide novel insights into the subcellular dynamics of aSyn by taking advantage of a powerful method for monitoring protein dynamics in living cells. A complete understanding of aSyn localisation, intracellular dynamics and proteinprotein interactions will be crucial for understanding the normal function of aSyn and may enable the development of novel strategies for intervention in PD and other synucleinopathies.
In order to obtain PAGFP-aSyn-WT construct, PAGFP in C1 vector was amplified by PCR with primers 5′TAAGC TAGCATGGTGAGCAAGGGCGAGG3′ (which contains a NheI restriction site) and 5′GGACTTAAGCTTGTA CAGCTCGTCCATGCC3′ (which contains a AflII restriction site and eliminates the stop codon from PAGFP). PAGFP PCR product and GN-link-aSyn were digested with NheI and AflII and ligated using T4 DNA ligase.
To obtain aSyn-WT-PAGFP, PAGFP in C1 vector was PCR amplified with the primers 5′GGGTCTAGACTATTACTTG TACAGCTCGTCCATGCC3′ (which contains a XhoI restriction site and eliminates ATG site from PAGFP) and 5′GTATC TAGACTATTACTTGTACAGCTCGTCCATGCC3′ (which contains a XbaI restriction site and the stop codon of PAGFP). PAGFP PCR product and aSyn-GC were digested with XhoI/XbaI, and ligated using T4 DNA ligase.
All constructs were generated in the pcDNA3.1 vector and verified by DNA sequencing.
Cell Culture, Transfections and Immunocytochemistry H4 human neuroglioma cells were maintained under standard conditions and passaged the day before transfection [51]. Transfections with aSyn, GFP and PAGFP constructs were performed using Fugene™ 6 reagent from Roche, according to the manufacturer's instructions. Immunocytochemistry experiments were performed as described previously for Myc-aSyn, aSyn-V5, Myc-aSyn-V5 and untagged aSyn constructs [51]. For PA experiments, cells were cotransfected with a mRFP in order to identify transfected cells. Cells were incubated for 48 h before imaging.

Live Cell Imaging
Cells were imaged using a Zeiss LSM510 META microscope with a ×63 1.4 NA oil immersion objective. aSyn-WT GFP-tagged was excited at 488 nm using an argon laser (5 % transmission) and a 505-to 550-nm band pass filter. For PA experiments, cells transfected with PAGFP constructs were first identified through a 561-to 10-nm DPSS laser (1 % transmission) to detect mRFP (561 nm) using a 575-nm long-pass filter and photoactivated using a diode laser line at 405 nm (100 % transmission) either in the cytoplasm or in the nucleus, using standard procedures for PA. PAGFP fluorescence emission was detected by excitation at 488 nm (5 % transmission) using a 505-to 550-nm band pass filter. About 500 images from each cell were taken with an interval of 2 s, and PA was performed after the second image; 15 cells per condition were analysed.
For FRAP experiments, cells transfected with GFP constructs were photobleached in the nucleus or in the cytoplasm using a diode laser line at 405 nm (100 % transmission), using standard procedures for FRAP. Recovery fluorescence of GFP constructs was detected by excitation at 488 nm (5 % transmission) using a 505-to 550-nm band pass filter. About 1,000 images from each cell were taken per second, and FRAP was performed after the second image; 15 cells per condition were analysed.

Image Analysis
The dynamics of aSyn diffusion (after PA or FRAP) was followed by analysing time-lapse series of the PAGFP or GFP reporter protein by measuring the fluorescence intensity over time, in the nucleus and in the cytoplasm, using ImageJ LSM toolbox plugin and LSM Image browser.

À Á
Where N and C refer to nucleus and cytoplasm, respectively. t=0 s refers to the time lapse immediately after PA or PB. These normalisations corrected for the loss of fluorescence caused by imaging both in PA and in FRAP procedures.

Statistical Analysis
The numerical results are given as mean of NF or CF±SD of up to 15 independent experiments.
The significance of the difference between the experimental and the control values of fluorescence was evaluated at three time points, 100, 500 and 1,000 s, in the nuclear compartment using 95 % confidence intervals (α=0.05) through single comparisons by the two-tailed unpaired Student's t test followed by a Fisher's exact test to compare variances between the control and experimental groups.