Clearance and Phosphorylation of Alpha-Synuclein Are Inhibited in Methionine Sulfoxide Reductase A Null Yeast Cells
- First Online:
- Cite this article as:
- Oien, D.B., Shinogle, H.E., Moore, D.S. et al. J Mol Neurosci (2009) 39: 323. doi:10.1007/s12031-009-9274-8
- 138 Views
Aggregated α-synuclein and the point mutations Ala30Pro and Ala53Thr of α-synuclein are associated with Parkinson’s disease. The physiological roles of α-synuclein and methionine oxidation of the α-synuclein protein structure and function are not fully understood. Methionine sulfoxide reductase A (MsrA) reduces methionine sulfoxide residues and functions as an antioxidant. To monitor the effect of methionine oxidation to α-synuclein on basic cellular processes, α-synucleins were expressed in msrA null mutant and wild-type yeast cells. Protein degradation was inhibited in the α-synuclein-expressing msrA null mutant cells compared to α-synuclein-expressing wild-type cells. Increased inhibition of degradation and elevated accumulations of fibrillated proteins were observed in SynA30P-expressing msrA null mutant cells. Additionally, methionine oxidation inhibited α-synuclein phosphorylation in yeast cells and in vitro by casein kinase 2. Thus, a compromised MsrA function combined with α-synuclein overexpression may promote processes leading to synucleinopathies.
KeywordsOxidative stressPosttranslation modificationNeurodegenerative diseasesParkinson’s diseaseAntioxidantsProtein aggregationYeastSynuclein
One of the hallmarks of Parkinson’s disease (PD) is the formation of Lewy bodies (Lucking and Brice 2000). These neuronal inclusion bodies are composed of aggregated proteins and consist mainly of α-synuclein fibrils (Baba et al. 1998). The α-synuclein protein is a presynaptic protein of unknown function and its involvement in the pathogenesis of PD is still not clear. However, the α-synuclein point mutations Ala30Pro and Ala53Thr are found in autosomal dominant forms of PD (Kruger et al. 1998; Polymeropoulos et al. 1997). Overexpression of α-synuclein is also associated with the risk of developing PD (Takeda et al. 1998), and α-synuclein locus triplication can cause PD (Singleton et al. 2003). When α-synuclein is abnormally expressed or modified, various studies indicate that it may cause alterations in mitochondrial and proteasomal function, protein aggregation, and accumulation of reactive oxygen species (ROS; Lindersson et al. 2004; Sawada et al. 2004; Snyder et al. 2005; Vekrellis et al. 2004).
Yeast cells have been used to study events related to α-synuclein biology and toxicity. This toxicity was shown to be mediated by increased cellular ROS and decreased proteasome function (Chen et al. 2004; Sharma et al. 2006; Willingham et al. 2003). The monomeric α-synuclein forms fibrils at a relatively slow rate. We hypothesize that modifications alter the physical properties and degradation rate of α-synuclein, overall promoting its oligomerization. This is consistent with the associated increase in oxidative stress among PD, aging, and other neurodegenerative diseases. Increased ROS and cellular α-synuclein levels promote its aggregation (Giasson et al. 2000; Hsu et al. 2000). The presence of ROS can cause posttranslational modifications, commonly to the sulfur-containing residues methionine (Met) and cysteine (Cys). The α-synuclein protein does not have Cys, but does contain four Met residues that can be readily oxidized to Met sulfoxide (MetO) in vitro. Under conditions of oxidative stress, MetO-α-synuclein can form fibrils when triggered by certain metals (e.g., Zn2+, Al3+, etc.) in vitro (Yamin et al. 2003). Furthermore, cellular conditions such as higher α-synuclein protein concentrations (Fink 2006) and a saline environment (Munishkina et al. 2004) could increase the rate of α-synuclein fibril formation. Posttranslational modifications can alter the solubility of α-synuclein, possibly decreasing the clearance of this short-lived protein. Cytoplasmic α-synuclein can be degraded by the ubiquitin–proteasome system, the autophagy system, and lysosomic pathway (Giorgi et al. 2006); however, degradation of α-synuclein is inhibited in PD. It is plausible that the latter inhibition decreases the clearance and promotes aggregation of α-synuclein. Modifications to α-synuclein may alter its solubility and interfere with its ability to be degraded, thereby resulting in a decreased clearance of α-synuclein.
The MetO modifications formed in the presence of ROS produce the enantiomers Met-S-O and Met-R-O. In proteins, MetO is reduced solely by Msr enzymes, which are MsrA (reduces Met-S-O) and MsrB (reduces Met-R-O; Moskovitz 2005; Oien and Moskovitz 2008). However, MsrA is thought to be the major Msr because it positively regulates MsrB expression (Moskovitz 2007; Moskovitz and Stadtman 2003). Lack of MsrA has been demonstrated in several organisms to cause a hypersensitivity to oxidative stress conditions (Oien et al. 2008; Pal et al. 2007) and premature death (Moskovitz et al. 2001; Oien and Moskovitz 2007). In contrast, overexpression of msrA may suppress dopaminergic cell death and protein aggregation (Liu et al. 2008). One hypothesis to explain the late onset of certain neurodegenerative diseases, such as PD, is that they occur in aging neurons when the capacity of the quality control system to cope with accumulating misfolded proteins is exceeded (Berke and Paulson 2003). This situation has been modeled in yeast by increasing α-synuclein expression (Outeiro and Lindquist 2003). Due to the lack of substantial information about MetO-α-synuclein in vivo, we further explored this by overexpressing α-synuclein and two mutants in msrA null (msrA-KO) yeast cells. The effects of the α-synucleins on protein fibril formation, degradation, and phosphorylation were monitored.
Materials and Methods
Overexpressing Human α-Synuclein in Yeast
The human α-synuclein protein and its mutants were overexpressed in msrA-KO and WT Saccharomyces cerevisiae strains with the pYES6 expression vector used previously (Moskovitz et al. 1997). This vector harbors the yeast GAL1 promoter for high-level inducible protein expression by galactose. It has a C-terminal peptide encoding a V5 epitope and a polyhistidine (6His) tag for detection and purification of the recombinant fusion α-synuclein. This vector contains the ampicillin and blasticidin resistance genes for selection of bacteria and yeast colonies harboring the plasmid, respectively. Three forms of the human α-synuclein cDNA (wild type (α-synuclein), Ala53Thr (SynA53T), and Ala30Pro (SynA30P)) were inserted into the vector keeping the sequence in frame with the 6His peptide. For that purpose, suitable 5′ forward and 3′ reverse complement primers (harboring Hind III and Xba I restriction sites, respectively) were used with each of the α-synuclein cDNAs (kindly provided by Dr. Mikiei Tanaka, NIH, NHLBI) in a routine polymerase chain reaction (PCR) procedure. The resulting PCR products were subcloned into the suitable complementary restriction sites on the pYES6 and transformed into bacterial cells (DH5α). Each plasmid was isolated from bacterial cells and transformed into WT and msrA-KO yeast. Vector-only transformed cells served as controls. Following growth selection on media containing blasticidin (Invivogen), all the transformed yeast strains (WT and msrA-KO containing either the vector only, vector containing α-synuclein, or vector containing either SynA53T or SynA30P) were grown in induction media containing 2% galactose and 1% raffinose for up to 24 h. At various time points, equal amounts of cells were taken from each culture (determined by optical density (OD600nm)) and extracted by glass beads and a bead-beater homogenizer apparatus (BioSpec Products, Bartlesville, OK, USA). The α-synuclein proteins were purified from the corresponding extracts by using of ProBond™ Nickel-Chelating Resin (Invitrogen). Equal volumes of purified α-synuclein proteins were subjected to sodium dodecyl sulfate (SDS)-gel electrophoresis and detected by Western blotting using anti-6His antibodies (Immunology Consultants Laboratories).
Total Protein Degradation and Synthesis in Yeast Expressing α-Synuclein
The procedure was performed as done previously (Chen et al. 2005) with modifications. Yeast cells (WT and msrA-KO) expressing the three α-synuclein variants (α-synuclein, SynA30P, SynA53T) were grown in yeast extract–peptone–dextrose (YPD; Fisher Scientific) medium to midlogarithmic phase while shaking at 30°C. The medium was removed by centrifugation and replaced with 2% galactose and 1% raffinose media lacking Met. Next, 50 μCi [35S]Met (Perkin-Elmer) was added and incubated for additional 30 min. The medium was removed and radioactive labeling was chased by media containing Met without radioactive isotopes. At different time intervals, equal aliquots of cells were removed from each culture and homogenized. The corresponding protein extracts were mixed with trichloroacetic acid (TCA) at a final concentration of 20% (v/v). After incubation at 4°C for 1 h, the samples were centrifuged, and the radioactivity (counts per minute) of the TCA-insoluble precipitates and soluble material was measured by liquid scintillation counting. For quantification of protein synthesis, the radioactivity (counts per minute) of the TCA-insoluble precipitate was measured by liquid scintillation counting after labeling with 50 μCi [35S]Met and immediate precipitation with TCA.
Phosphorylation of α-Synuclein in Yeast (Ex Vivo System)
The WT and msrA-KO yeast cells expressing α-synuclein, SynA30P, and SynA53T were grown in YPD medium to midlogarithmic phase shaking at 30°C. The medium was removed by centrifugation and substituted with induction media containing 2% galactose and 1% raffinose for 1 h. The cells were then incubated in media with only 10% of the nutrients for 1 h to reduce free phosphate. To each culture, 5 μCi/mL [32P]orthophosphoric acid (Perkin-Elmer) was added and incubated for 1 h. The cells were harvested and α-synuclein was isolated by resin affinity as described above with the addition of NaF and Na2VO4 as phosphatase inhibitors. The α-synuclein was eluted from the resin with 150 mM imidazole. Equal amounts of each eluted protein were subjected to an SDS-gel electrophoresis. At the end of electrophoretic separation, the gel was stained with GelCode Blue (Thermo Scientific, Rockford, IL, USA). Bands corresponding to the molecular weight of the α-synucleins were excised, added to vials with liquid scintillation cocktail, and their counts per minute measurements were determined by liquid scintillation counting.
Phosphorylation of α-Synuclein In Vitro
Recombinant α-synuclein, SynA30P, and SynA53T (rPeptide) were incubated overnight in 200 mM H2O2 at 37°C in phosphate buffered saline buffer. The H2O2 was removed by adding catalase and incubating 30 min (oxidant removal verified by adding α-synuclein after oxidants were removed, this served as the nonoxidized α-synuclein control). To reduce MetO, recombinant yeast MsrA and PilB were produced as previously described (Moskovitz et al. 1997; Olry et al. 2002). These Msrs were added at a concentration of 0.3 mg/mL with 20 mM dithiothreitol for 1 h. Each α-synuclein was phosphorylated by adding 6 mCi [γ-32P]-ATP (Perkin-Elmer), 20 µM cold ATP, 10 mM MgCl2, and 0.3 µg casein kinase 2 (Upstate) for 1 h. Reactions were stopped by adding an equal volume of protein gel sample buffer. Equal amounts of α-synuclein proteins were subjected to an SDS-gel electrophoresis. At the end of the electrophoretic separation, the gel was stained with GelCode Blue Stain reagent and exposed to X-ray film.
Protein Fibrillation in Yeast Strains Expressing α-Synuclein
In the presence of fibrils, Thioflavin T undergoes a “red shift” where it is excited at a wavelength of 442 nm and emits fluorescence at 482 nm. Yeast cells were grown in YPD and incubated in media with 2% galactose and 1% raffinose for 3 h. The cells were precipitated by centrifugation and resuspended in 1% Thioflavin T solution. Cells were washed in buffer and aliquots were diluted into a 96-well plate. Finally, Thioflavin T fluorescence was imaged and quantified using an IN Cell Analyzer 1000 (GE Healthcare) high content imaging system.
Comparison and calculation of P values was performed using Student t test analysis. Values of P < 0.05 were considered significant.
Results and Discussion
Overexpression of α-Synuclein in Yeast
The expression patterns of α-synuclein suggests that the three types of α-synuclein are degraded faster in the WT compared to the msrA-KO strain (Fig. 1). Moreover, the SynA30P variant of α-synuclein was the most resistant to degradation in the msrA-KO strain (Fig. 1). This result is supported by previous studies showing that WT yeast cells overexpressing SynA30P are inhibited in their general protein degradation by the proteasome, compared to nonmutated α-synuclein (Chen et al. 2005). We suggest that the degradation of α-synuclein is inhibited more in msrA-KO cells compared to WT cells, which may be an enhanced inhibitory effect of the MetO on cellular degradation. Using matrix-assisted laser desorption/ionization-mass spectrometry (MALDI-MS) techniques, we have verified that SynA30P and SynA53T expressed in the msrA-KO cells had elevated levels of Met oxidation relative to same α-synucleins expressed in WT cells (data not shown).
Protein Degradation and Synthesis in Yeast Strains Overexpressing α-Synuclein
It is noteworthy that the degradation rate over time was elevated in the msrA-KO versus WT yeast strain when sham vector was present. One possible explanation is that accumulation of MetO-containing proteins (the case in the msrA-KO strain) induces the action of the degradation machinery as an attempt to remove the faulty proteins (Stadtman and Berlett 1998). However, this response is inhibited by excess levels of specific α-synuclein types (Figs. 1 and 2a). The general inhibition of protein degradation was more pronounced in the msrA-KO cells expressing native and A30P α-synucleins (Fig. 2a). The overexpressed native α-synuclein effect on cellular degradation may be relevant to its accumulation that is thought to be a risk factor for PD, and specifically impaired α-synuclein degradation may be involved (Singleton et al. 2003).
The mechanism by which the nature of α-synuclein and MetO modifications to α-synucleins inhibits degradation is not known. The degradation pathways may be inhibited by the cellular localization of α-synuclein. In yeast, each α-synuclein mutant accumulated at the same level as nonmutated α-synuclein, but the cellular distributions differed profoundly. Similar to native α-synuclein, SynA53T is concentrated at the plasma membrane, while SynA30P dispersed throughout the cytoplasm (Brandis et al. 2006; Bussell and Eliezer 2003; Jo et al. 2002). Therefore, the cytoplasmic localization of SynA30P may cause the protein to be more prone to oxidation (compared to a membrane region) and accordingly less protected from the accumulation of MetO in the absence of cytoplasmic MsrA. This may also partly explain the special negative effect of MetO-α-synuclein accumulations on cytoplasmic degradation, as accumulations of oxidized proteins can inhibit protein degradation (Starke-Reed and Oliver 1989).
Inhibition of protein synthesis by native and Ala30Pro α-synuclein expressed in yeast cells was previously described (Chen et al. 2005). Therefore, we examined the contribution of msrA ablation to the α-synuclein inhibitory effect on protein synthesis. Expression of α-synuclein caused an overall inhibition of protein synthesis in both strains compared to control cells expressing the sham vector (Fig. 2b). The inhibition levels were similar in both yeast strains expressing either SynA30P or SynA53T. However, a combination of msrA ablation and native α-synuclein expression in msrA-KO cells caused a stronger inhibitory effect on protein synthesis compared to WT cells expressing native α-synuclein (Fig. 2b).
As discussed by other groups (Chen et al. 2005), the basis for decreased protein synthesis in cells expressing α-synuclein is unknown. They suggest that the impairments in this synthesis may either represent negative feedback (result of impaired protein degradation) or α-synuclein induction of endoplasmic reticulum stress impairing protein synthesis (Chen et al. 2005). It is possible that alterations to the structure and function of native α-synuclein expressed in msrA-KO cells are more effective in fostering inhibition of protein synthesis (Fig. 2b).
Accumulation of Fibrillated Proteins in Yeast Strains Overexpressing α-Synuclein
Several posttranslational modifications can alter the solubility of α-synuclein. Oxidation will cause a Met residue to be more hydrophilic, but the overall MetO-protein may become more hydrophobic by exposing hidden hydrophobic residues (Chao et al. 1997). Degradation of α-synuclein is inhibited in PD, and thus, it is a plausible theory that this inhibition (also shown in Figs. 1 and 2) decreases the clearance of α-synuclein and promotes its aggregation, especially when the protein undergoes modifications that reduce its solubility. All types of α-synuclein caused enhanced cellular fibrillation in the absence of MsrA compared to the vector, while SynA30P had the most effect on forming fibrils in the msrA-KO strain (Fig. 3). This result would be the expected outcome of enhanced inhibited degradation of the α-synuclein A30P (Figs. 1 and 2). Since the SynA53T aggregation is less affected by the presence of MsrA (Fig. 3), it may reflect the enhanced aggregation found in PD patients carrying this mutation (Lee et al. 2002; Ostrerova-Golts et al. 2000). However, a combination of a compromised Msr system (from mutations or lowered expression with age) and overexpression of SynA30P could cause even greater levels of α-synuclein fibrils in PD patients.
Phosphorylation of α-Synuclein and Its Dependency on Methionine Sulfoxide Reductase
To further investigate the possibility that methionine oxidation of α-synuclein may contribute to the phosphorylation inhibition, we used recombinant α-synuclein proteins in reduced and oxidized form, [32P]ATP, and casein kinase 2 (a major kinase for α-synuclein Ser129 phosphorylation; Ishii et al. 2007). To confirm that MetO formed in the oxidized α-synuclein is responsible for the observed effect on phosphorylation level, MsrA and PilB (naturally fused MsrA–MsrB protein from Neisseria gonorrhoeae) were added to the reaction mixture to reduce α-synuclein MetO residues. All samples were subjected to SDS-gel electrophoresis, gels were stained with GelCode Blue Stain (Fig. 4b), and phosphorylation was visualized by autoradiography. The oxidation of the three types of α-synuclein inhibited the phosphorylation by casein kinase 2 (Fig. 4c), while both MsrA and PilB were able to reverse this inhibition via MetO reduction. Both of the mutant α-synuclein proteins were more phosphorylated when reduced by the Msr enzymes (Fig. 4c). Thus, it is possible that conformational changes that may occur in these α-synuclein mutants may also facilitate MetO reduction by the Msr system due to better accessibility of the enzyme to MetO residues. It is important to note that casein kinase 2 phosphorylation activities on control samples were not altered when Msr enzymes were added. This excludes the possibility that the enhanced α-synuclein phosphorylation, shown by Msr enzymes, is due to their reducing activity on casein kinase 2. Also, using MALDI-MS, we have verified that the recombinant α-synucleins had elevated levels of oxidation relative to nonoxidized and Msr-reduced α-synucleins (data not shown).
The role of phosphorylation of α-synuclein (especially at Ser129) is controversial. The phosphorylation α-synuclein at Ser129 by casein kinase 2 forms fibrils faster than unmodified α-synuclein in vitro (Fujiwara et al. 2002) and increases the formation of cytoplasmic inclusion bodies in cell culture models of synucleinopathies (Smith et al. 2005). However, in vivo studies showed no obvious correlation between phosphorylated Ser129 and of α-synuclein fibril formation (Chen and Feany 2005). In contrast, another study demonstrated that phosphorylation at Ser129 inhibits α-synuclein fibril formation in vitro (Paleologou et al. 2008). With regard to the vicinity of Met127 to Ser129, we have examined the effect of MetO modifications on the rate of the in vivo phosphorylation of α-synuclein and casein kinase 2-mediated phosphorylation of α-synuclein. The observed decrease in α-synuclein phosphorylation due to MetO following MsrA ablation in cells (Fig. 4a) may be relevant to processes that regulate α-synuclein phosphorylation. As the provided yeast system is only a model to investigate the role of MsrA in maintaining α-synuclein function, further investigations in a parallel mammalian system will greatly contribute to clarify the physiological role of MsrA on α-synuclein phosphorylation.
A functional Msr system and Met oxidation may have a role in regulating α-synuclein and the phosphorylation of other proteins. Recent studies suggested that MetO formation on inhibitory factor of nuclear factor-kappaB (IkBα) prevents its subsequent degradation. Similar to α-synuclein, the IkBα Met45 residue is close to the phosphorylation sites Ser32 and Ser36. The MetO formation, particularly within helices, can cause major conformational rearrangements (Bigelow and Squier 2005), and the oxidation of Met45 may cause a structural disruption that makes the protein a less effective kinase target (Kanayama et al. 2002; Midwinter et al. 2006; Mohri et al. 2002). In another recent study, MsrA was shown to suppress dopaminergic protein aggregation induced by mutant α-synuclein via MetO repair. The MsrA was able to reduce several oxidized Met residues in recombinant α-synuclein (Liu et al. 2008). Overall, it is suggested that oxidation of Met in α-synuclein inhibits phosphorylation, which may be involved in processes relevant to its degradation or aggregation depending on cellular environment.
In summary, both the nature of α-synuclein and presence of MsrA determined the overall protein and α-synuclein degradation in yeast. Furthermore, Met oxidation of α-synuclein inhibits α-synuclein phosphorylation. This occurs parallel to increasing protein fibrillation, especially when SynA30P is expressed in msrA-KO cells. To establish the relationships between α-synuclein overexpression and MsrA function in mammals, similar studies will be conducted in transgenic α-synuclein mice on an MsrA−/− background.
This work was supported by the Kansas EPSCOR/National Science Foundation and the National Institute of Aging, AG027363.