Recombinant α-synuclein was expressed in E. coli BL21 (DE3) (Novagen) grown at 37˚C in M9 medium using the expression vector pET15b (Novagen) and purified as described previously [13]. The ¹⁵N-labled α-synuclein was expressed using ¹⁵NH₄Cl as a nitrogen source. The solutions of 0.2 mM α-synuclein containing 1.0 mM DA were assembled into oligomers in the absence and presence of 1.0 mM GSH by incubating in10 mM ammonium acetate (pH 7.5) at 30˚C with continuous stirring at 450 rpm. These solutions were used in analysis of ESI-MS and size exclusion chromatography.

The 2D ¹H-¹⁵N HSQC spectra were recorded using sample solutions of, 1) 0.2 mM¹⁵N-labeled α-synuclein and 1.0 mM DA; and 2) 0.2 mM¹⁵N-labeled α-synuclein, 1.0 mM DA and 1.0 mM GSH at the incubation time of 0, 24 and 72 h on a Bruker DRX-500 spectrometer. All HSQC spectra were acquired at 4˚C using the identical receiver gain. The experimental parameters used for the 2D ¹H-¹⁵N HSQC were: data size in t₁ = 128 complex points; data size in t₂ = 1024; spectral width in f₁ = 2048 Hz; spectral width in f₂ = 6000 Hz; number of transients per increment = 16. The WATERGATE W5 LOGSY [16] spectra were recorded using a solution of 0.1 mM α- synuclein in the presence of 8.0 mM DA and/or 8.0 mM GSH at 20˚C on a Varian Inova 500 spectrometer. The experimental parameters used for the WATERGATE W5 LOGSY were: data size = 8192 points; spectral width = 10,000 Hz; number of scans = 64.

ESI-MS measurements were carried out using a time-of-flight mass spectrometer (JEOL JMS-T100). The experimental conditions were as follows: acceleration voltage, 2.0 kV; needle voltage, 2600 V; orifice voltage, 65 V; desolvation temperature, 100˚C; sample flow rate, 5.0 μL/min. All spectra were obtained via infusion of the sample solutions in the positive ionization mode.

The incubated sample solutions were filtrated through a membrane filter Dismic-03 (pore size, 0.45 μm; Toyo Roshi), and then applied onto a TSKgel G3000SW_XL gel filtration column (7.8 mm I.D. × 300 mm; Tosoh Bioscience) with 50 mM sodium phosphate buffer, pH 7.5, containing 100 mM NaCl. The flow rate was 0.4 mL/ min.

In observation of interactions between α-synuclein and DA at the atomic level, the 2D ¹H-¹⁵N HSQC spectra were acquired using the ¹⁵N-labeled α-synuclein in the presence of DA (Figure 1). The expanded spectra, indicating the chemical shift changes of L8, S9 and S129, are shown in Figure 2. The chemical shift changes were identified for the residues of V3, L8, S9, A27, H50, N113, D119, E126, M127 and S129, which are considered to interact with DA. In the presence of GSH, the chemical shift changes were identified for the residues of V3, L8, A27, N113, D119, E126, M127 and S129 (data not shown). Any chemical shift changes of α-synuclein were not identified only in the presence of GSH (data not shown). These results indicate that N-terminal and Cterminal regions of α-synucleinare involved in the inter-

actions with DA, and that GSH is not involved. Because the oligomer formation of α-synuclein was distinctly induced upon binding with DA [10,15], both Nand C-termini are considered to play a key role in oligomerization. An importance of C-terminal in aggregation of α-synuclein was demonstrated by Mazzulli, et al., using the protein lacking the polypeptide segment ¹²⁵YEMPS¹²⁹ [17]. It was shown that the intracellular inhibition of α-synuclein aggregation required the specific noncovalent interaction of the oxidized catechols with the residues ¹²⁵YEMPS¹²⁹. On the other hand, another study demonstrated that a truncated α-synuclein mutant terminating at residue 124 could still form soluble oligomers in the presence of DA, and that mutagenesis of all four Met residues to Ala significantly reduced the propensity to form oligomers [18,19]. In the present chemical shift perturbation study using the ¹H-¹⁵N HSQC spectra, three residues, E126, M127 and S129, were involved in the polypeptide segment ¹²⁵YEMPS¹²⁹ in the C-terminal region. Since two of four Met residues are positioned in the N-terminal region, the present results suggested that both Nand C-terminal regions are required in oligomerization.

Interactions of α-synuclein with DA in the presence and absence of GSH were also observed using the WATERGATE W5 LOGSY as shown in Figure 3. In this NMR method, the small molecules which bind to protein can be selectively detected, and the unbound small molecules are observed in the opposite phase. The ¹H signals of DA were selectively observed as expected (Figure 3(a)), and interactions of GSH with α-synuclein could not be identified as shown in Figure 3(b). It was shown that DA was able to bind with α-synuclein in the presence of GSH (Figure 3(c)), indicating that the unoxidized α-synuclein had an ability to bind with DA. The above interactions were also observed by ESI-MS. Envelopes of ions between m/z 852 and 2412, corresponding to +17 through +6 charged states of monomers, were observed (data not shown). The expanded spectra between m/z 1420 and 1570 are shown in Figure 4. The envelopes of ion peaks between m/z 1447 and 1450 corresponded to the monomeric α-synuclein (mw = 14460) with +10 charged state. In this envelope, the ion peaks consisted of the free α-synuclein and its sodium adducts. In the presence of DA (mw =153), the ion peaks corresponding to the complexes of α-synuclein and DA were observed (Figure 4(a)). The maximum of two DA molecules were detected to bind with one molecule of α-synuclein even in the presence of GSH (Figure 4(b)). The ion peaks of the complexes comprising α-synuclein and DA were observed by ESI-MS, and ¹H signals of DA bound to α-synuclein were selectively observed by NMR spectroscopy. These results indicate a distinct interaction of DA with theunoxidized α-synuclein.

The effects of GSH in oligomerization of α-synuclein in the presence of DA were monitored using size exclusion chromatography as shown in Figure 5. In the chromatograms of α-synuclein, the monomerwas eluted at 24 minutes. Several high molecular weight oligomers with retention times from ca. 17 min to 23 min were observed in the presence of 1 mM DA at 24, 72 and 120 hour incubations (Figure 5(a)). Distribution of the molecular weights was estimated to be in the range of ca. 65 - 150 kDa based on the calibration curve provided by Tosoh Bioscience. Observation of the high molecular weight species indicates formation of the DA-induced oligomers. Interestingly, any formation of oligomers could not be observed in the presence of 1 mM GSH (Figure 5(b)). Although the results of NMR and ESI-MS indicated that DA had an affinity with α-synuclein in the presence of GSH, the result of size exclusion chromatography demonstrated the distinct suppression of oligomerization under the anti-oxidant condition. These results suggest that an association with DA could be a trigger of oligomeri-

zation, and some oxidations of α-synuclein are required to proceed with oligomerization. The observations of oligomers comprising α-synuclein with four oxidized Met [9-11] coincide with the present results of size exclusion chromatography. In γ-synuclein, which is another member of the synuclein family, M38 and Y39 can be easily oxidized and γ-synuclein can form annular oligomers that accumulate in cells [20]. Its oxidation plays a key role in the ability of γ-synuclein to aggregate and seed the aggregation of α-synuclein. Oxidation of the synuclein family could induce a cascade of events leading to synucleinopathies.