Detection and Assessment of α -Synuclein Oligomers in the Urine of Parkinson’s Disease Patients

Western blot analysis

We loaded samples on to 4–15% gradient gels (#456-1086; Bio-Rad Laboratories, Inc.) and transferred them to nitrocellulose membranes. The primary antibodies used in this study are listed in Table 1. The secondary antibodies used for the detection of primary antibodieswere peroxidase-conjugated AffiniPure Goat Anti-Mouse IgG (H + L) (#115-035-003; Jackson Immunoresearch Laboratories Inc.) and peroxidase-conjugated AffiniPure Goat Anti-Rabbit IgG (H + L) (#111-035-144; Jackson Immunoresearch Laboratories Inc.). To visualize interesting protein bands, we used Luminata Crescendo Western HRP (#WBLUR0500; Merck) and images were acquired using MicroChemi 4.2 (Shimadzu).

Generation of standards for the ELISA

Recombinant human α-syn was purchased from Boston Bio (SP-480). To generate fibril standards, α-syn solution (70 μM, 1 mg/ml) was concentrated up to 3 mg/ml and dialyzed with phosphate-buffered saline (PBS).

As a pre-saturation step, the α-syn solution was incubated for one week at 37°C,whilst shaking at 100 rpm. After brief sonication, α-syn was re-incubated for 2 weeks under conditions identical to the pre-saturation conditions. Fibrillar α-syn was isolated following ultra-centrifugation at 100,000g for 1 h at 4°C. The precipitated α-syn fibril was washed with PBS twice, and the pellet was resuspended with PBS at a concentration of 1 mg/ml.

For another oligomeric ELISA standard, α-syn monomer solution (1 mg/ml) was incubated at 4°C for 5 days. The sample was then passed through a 0.2 μm filter to remove fibrillar conformers. The flow-through (FT) was subjected to 100 K filtration using a 100 K filter (100 K Omega Nanosep Centrifugal Devices, OD100C34, Pall), and we collected the retentate (100 K-Cut). The FT of the 100 K underwent 30 K filtration using a 30 K filter (3 K Omega Nanosep Centrifugal Devices, OD030C34, Pall), and the 30 K-FT was collected. The 100 K-Cut fraction was used for the standard of another oligomeric ELISA, and the 30 K-FT was used to verify its capacity as a standard.

Aggregates-trap assay and dot-blot assay

The nitrocellulose membrane was equilibrated with Tris-buffered saline with 0.1% TWEEN-20 (TBST) for 30 min at room temperature. Equivalent samples of each standard for the ELISA were then loaded onto the nitrocellulose membranes using a Dot-Blotter (Bio-Rad Laboratories, Inc.). For the detection of oligomeric standards, we used 2 μl of the standard mixtures in equal concentration for each ELISA. After blocking the membranes with 5% skimmed milk, the indicated primary antibodies were used at the same dilution ratio as the respective capture antibodies in the ELISAs. The rest of the steps followed the western blot assay protocol.

Sandwich ELISA

We coated 96-well immunoplates (Corning) with three different capture antibodies at the indicated dilution ratio (Table 1) in 60 μl of sodium bicarbonate buffer (pH 5.0), overnight at 4°C. Super Block T20 (PBS) Blocking Buffer (#37516; Thermo) was used for blocking and it was mixed with PBST (PBS with 0.1% TWEEN-20) to make the dilution buffer used for the standards and samples. Standards or samples of 100 μl were loaded to wells and incubated for 2 h at RT, with shaking at 100 rpm. For detection, we diluted the HRP-conjugated α-synuclein antibody (Table 1) in 100 μl of dilution buffer for 1.5 h at RT, shaking at 100 rpm, and 3, 3^′, 5, 5^′-tetramethylbenzidine ([TMB], T0440; Sigma) 100 μl was added and allowed to reactfor 5 to 15 min at RT. Then, 2 M HCl 10 μl was used to halt the TMB reaction. Absorbance at 450 nm was measured and used for estimations based on an interpolated standard curve.

Cell culture and rotenone treatment

The human neuroblastoma cell line, SH-SY5Y, was cultured with growth media (DMEM, 10% fetal bovine serum [FBS], and 1% Penicillin/Streptomycin). To differentiate SH-SY5Y to dopaminergic neuronal cells, we cultured SH-SY5Y cells with growth media containing 10 μM all-trans retinoic acid (RA) for 7 days.

Human iDP (induced human dopamine progenitor)-neurons (iDP-neuron) were differentiated from dopaminergic neural progenitors reprogrammed from human skin fibroblasts. The neuronal differentiation protocol has been previously described [17]. We treated both differentiated SH-SY5Y (dSH) and iDP-neurons with 3 μM rotenone (dissolved in DMSO) in growth media for 48 h. We collected the supernatant of the culture media, which was centrifuged to remove cell debris, and the supernatant was concentrated using a 3 K filter (3 K Omega Nanosep Centrifugal Devices, OD003C34, Pall). We also lyzed the cells with PBS containing 1% Triton X-100 (TX-100), Xpert Protease inhibitor cocktail ([PIC], P3100; GenDEPOT) and 5 mM EDTA under ice-cold conditions. Diluted crude lysates or concentrated culture media were loaded onto the ELISA. Estimated α-syn levels obtained from their respective ELISAs were normalized based on the concentration of total protein in the cell lysates. The α-syn levels in the concentrated culture media or diluted lysate were divided by the fold ratio of the concentration.

Urine collection and concentration

This study was approved by the Institutional Review Board of Boramae Hospital, Seoul National University (IRB No. 30-2018-20 and 30-2019-96). All experimental processes were performed in accordance with the relevant guidelines and regulations of the IRB. Urine samples were collected from PD patients as well as non-PD controls at Boramae Hospital (Seoul, Korea) on visits. All PD patients were able to volunteer because our expert neurologist did not recruit PD patients with cognitive decline. Hence, every patient signed the informed consent for themselves. Urine samples were stored at –80°C until use. All PD patients were diagnosed by a neurologist, and their Hoehn and Yahr (H & Y) stage was measured (Table 2).

We processed urine samples that were mixed with 1% TX-100 and 1X PIC. These were centrifuged (4000g for 30 min at 4°C) to remove debris, then concentrated using a10 K filter (AmiconUltra-4 Centrifugal filters Utracel-10 K, UFC801024, Merck). Each urine sample was subjected to distinct centrifugation protocols at 4°C and was concentrated to 10-fold of the original volume. Ten-fold concentrated urine samples were used in the ELISA, and this was accounted for in the subsequent analysis of α-syn. The serum was used as obtained from the patient.

To verify the presence of α-syn using western blot analysis, urine samples from PD patients were mixed with 1% Triton X-100 and 1X PIC, which was concentrated using a 100 K filter (VIVASPIN TURBO 4, VS04T41, Sartorius). Then, the flow-through underwent further concentration by 10 K filtration. Both 10-fold concentrated samples were mixed with 5X sample buffer (250 mM Tris-HCl, pH 6.8, 10% sodium dodecyl sulfate, 50% glycerol, 5% β-mercaptoethanol, and 0.1% bromophenol blue) and subjected to western blot assay. Lysates from rotenone-treated dSH and iDP-neurons were compared with urine samples.

Set-up and detection of monomeric or fibrillar α-syn standard for ELISA

To establish the sandwich ELISA for α-syn, we composed antibody-pairs for the sandwich ELISA. The three different antibodies that we used as capture antibodies in our ELISA are described in Table 1. Monomeric α-syn was used as the standard for the total α-syn ELISA; we then measured serial dilutions of monomer standards using a capture antibody (Clone 42) and detection antibody (Fig. 1A, B). We validated the combination of the total α-syn ELISA (Total-αS ELISA) antibody-pair using a monomer standard (Fig. 1C).

OPEN IN VIEWER

Fig. 1

Establishment of sandwich ELISA detecting total α-syn and β-sheet rich conformations of α-syn. A,B) The serial dilution of monomer standards was confirmed following western blot assay using the capture (A) or detection (B) antibody for Total-αS ELISA. C) Monomeric standards showed a linear increase in 450 nm absorbances with the increase in concentrations of monomeric standards. D, E) A dose-dependent increase in fibril standards was observed following western blot assay using the Fila-αS capture (D) and detection (E) antibodies. F) The capture antibody for the Total-αS ELISA validated the serial increases in fibril standards shown in E-F. G) The association between the concentration of fibril standards and absorbance showed good linearity. H) The highest concentration of monomeric standards was not detected by the capture antibody of Fila-αS ELISA using an aggregates-trap assay but positively reacted with the capture antibody of Total-αS ELISA. I) The monomeric standards did not show a dose-dependent increase in the 450 nm wavelength absorbances in Fila-αS ELISA. Most standards were measured below the standard of zero-background. J) Fibril standards showed a linear increase with their concentrations and it appears that saturation occurs between 80 ng/ml and 160 ng/ml of the fibril standard, thereby reorganizing the range of fibril standards at lower levels.

The fibrillar form of α-syn was used as the standard for filament-conformation specific (Fila) α-syn ELISA (Fila-αS ELISA) because the Fila antibody could detect the aggregation-prone formation of α-syn oligomers, which harbors a β-sheet-rich structure [18]. Fibril standards were confirmed with a capture and detection antibody using an aggregates-trap assay (Fig. 1D, E). Fibril standards were tested with a Total-αS ELISA capture antibody (Fig. 1F), and the fibril standards and antibody pair showed a linear correlation (Fig. 1G). The highest concentration of monomer α-syn was detected using the capture antibody against Total-αS ELISA, and not by the filament-conformation specific α-syn capture antibody (Fig. 1H). Moreover, monomeric standards did not fit with Fila-αS ELISA, which was contrary to the linear curve of the fibril standards (Fig. 1I, J). These results show that monomeric or fibril standards with antibody pairs were suitable for their respective ELISAs.

Detection of neuronal cell-derived α-syn using ELISA

Previous studies have demonstrated that rotenone-treatment results in the aggregation of α-syn and its release [19, 20]. Therefore, we treated dSH, a human dopaminergic neuronal cell line, and iDP-neurons which were derived from fibroblasts of a healthy donor, with rotenone. Rotenone treatment promoted a slight increase in cellular Total-αS levels in dSH and a significant increase in iDP-neurons (Fig. 2A, E). The intracellular accumulation of β-sheet-rich oligomeric α-syn following treatment with rotenone was significantly higher in dSH, but there was no significant change in iDP-neurons (Fig. 2B, F). However, secreted levels of Total-αS and Fila-αS showed significant increases following rotenone treatment for both cell types (Fig. 2C, D, G, H). As rotenone-induced cytotoxicity can promote the release of α-syn as well as the aggregation of α-syn, the results from a cell-based system may validate our Total- and Fila-αS ELISA.

OPEN IN VIEWER

Fig. 2

The validation of established ELISAs. We measured crude cell lysate from dSH using Total-αS ELISA (A) or Fila-αS ELISA (B). Culture media from dSH was also analyzed using the ELISA for Total-αS (C) or Fila-αS (D). Crude cell lysate from iDP-neurons was also subjected to Total-αS (E) or Fila-αS (F) ELISA. The measurements obtained using Total-αS ELISA (G) or Fila-αS ELISA (H) of the culture media from iDP-neurons are shown. A Student’s t-test was used for statistical analysis. *** p < 0.001; **** p < 0.0001.

To clarify the results of Total- or Fila-αS ELISA using the cell-based system, immune precipitation (IP) was used for the removal of α-syn in samples. We immune precipitated recombinant α-syn using the capture antibody against Total-αS ELISA and confirmed that IP pull-down α-syn was appropriate (Supplementary Figure 1A). The α-syn in naïve dSH lysate (Pre) was captured by the α-syn antibody-beads complex (IPed), and the α-syn levels following IP (Post) were decreased to undetectable levels (Supplementary Figure 1B). We observed that the removal of endogenous α-syn followed by IP decreased the levels of α-syn in Total- or Fila-αS ELISA (Supplementary Figure 1C, D). These results lead us to conclude that our developed ELISA is appropriate for the measurement of Total- or Fila-αS levels in biofluids.

Detection and analysis of α-syn in urine samples

The presence of α-syn in human urine has not been demonstrated with biochemical analyses to date. To confirm the validity of measuring α-syn levels in human urine, we attempted to detect monomeric and oligomeric α-syn in a urine sample that was concentrated 10-fold using sequential 100 K and 10 K filtration. Western blot analysis clearly showed the presence of monomeric α-syn in the 10 K filtration fraction (Fig. 3A). The high molecular weight bands of α-syn in the 10 K and 100 K fraction (arrowhead in Fig. 3A) might represent oligomeric α-syn in the urine. Various non-specific α-syn oligomers, which appeared as a ladder of numerous α-syn bands, were also observed in the lysates from dSH and iDP-neurons that were treated with 2 μM of rotenone. We could not observe non-specific reactions of the anti-mouse secondary antibody (Fig. 3B), and the total protein levels in urine showed relatively lower levels than those of the rotenone-treated dSH or iDP-neurons (Fig. 3C). Therefore, we concluded that the concentrated urine sample was amenable for use in our established ELISA.

OPEN IN VIEWER

Fig. 3

Detection of α-syn in human urine. A) Monomeric α-syn (arrow) in 10K-concentrated urine samples appeared at an identical size of approximately 18 kDa using the capture antibody of Total-αS ELISA. Both urine samples demonstrate the presence of oligomeric a-syn (arrowhead). B) The secondary antibody to detect Total-αS capture antibody did not show any detectable protein bands. C) MemCode staining validated the total amount of loaded protein.

Next, we tested 11 non-PD samples from non-related patients and 21 samples from PD patients. There was no difference in Total-αS levels between the two groups (Fig. 4A), but Fila-αS was significantly higher in PD patients’ urine (Fig. 4B). The correlation between Total-αS and Fila-αS in non-PD and PD patients was not significant following non-zero slope validation, but the slope in PD patients showed a weak association (p = 0.0569) (Fig. 4C, D). Interestingly, Total-αS levels increased with the H & Y stage of PD patients and the non-zero slope analysis was significant (p = 0.0321) (Fig. 4E). However, there was no significant association between Fila-αS levels in PD patients’ urine and H & Y stage (Fig. 4F). In addition, when αSyn levels in the blood were measured, the levels of Total- or Fila-αS in the serum were not related to those levels in the urine or the H & Y stage of the respective PD patients (Supplementary Figure 2). Taken together, only the levels of Fila-αS were significantly higher in PD patients’ urine than those in non-PD urine samples. Meanwhile, PD severity may be associated with Total-αS urine levels.

OPEN IN VIEWER

Fig. 4

Analyses of urinary α-syn levels among non-PD and PD samples. Ten-fold concentrated urine samples were subjected to Total-αS ELISA (A) and Fila-αS ELISA (B). We analyzed the association between Total-αS levels and Fila-αS levels in non-PD (C) and PD (D) groups. The relationship between the H&Y stage and Total-αS levels (E) or Fila-αS levels (F) was also analyzed. A Student’s t-test was used for comparisons between the non-PD and PD groups. Non-zero slope significance was analyzed in XY curve graphs. * p < 0.05.

Levels of various oligomeric formations of α-syn in urine

The heterogeneity of oligomeric α-syn in biofluids is not yet fully understood. However, we cannot not exclude the possibility of the presence of different α-syn oligomers in the urine. Therefore, we analyzed the levels of different α-syn oligomers in urine using the ASyO5 antibody against distinct conformations of α-syn oligomers [21]. The ASyO5 antibody attached to the specific oligomer α-syn which showed a spherical ultra structure as seen with Atomic Force microscopy, but the fibrillar α-syn was not detected using the ASyO5 antibody. To avoid the detection of fibrillar α-syn, we choose ASyO5 as the capture antibody in our ELISA which is distinct α-syn. To prepare a standard for the oligomeric α-syn, we collected each cut and FT fraction using sequential filtration following low-temperature incubation of monomeric α-syn for 5 d (Fig. 5A) and used the 100 K-cut α-syn as an appropriate standard for oligomeric α-syn (Oligo5-αS) ELISA (Fig. 5B). Serial dilution of the 100 K-cut α-syn showed a gradual decrease in oligomeric (arrow) and monomeric α-syn (Fig. 5C) using the detection antibody in western blot assays and with the capture and detection antibody pairs for Oligo5-αS ELISA in dotblot assays (Fig. 5D). Linearity of the oligomeric standards using the 100 K-cut α-syn fractions was observed, but serial dilutions of 30 K-FT α-syn did not fit a linear curve (Fig. 5E), suggesting an absence of Oligo5-αS in the 30 K-FT. As the capture antibody, Oligo5-αS, exhibited a weak binding to the 30 K-FT (Fig. 5B), and a serial dilution of 30 K-FT showed extremely low absorbances in the Oligo5-αS ELISA (Fig. 5E). The detection of monomeric α-syn by Oligo5-αS ELISA was concluded to be negligible. The level of Oligo5-aS was lower in PD patients than that in non-PD patients (Fig. 5F), and there was no correlation between Oligo5-αS levels and H&Y stage in PD patients (Fig. 5G). The increase in Oligo5-αS levels was not related to Total-αS or Fila-αS levels (Fig. 5H–K). ELISA for Oligo5-αS showed absorbance below that of the standard at zero, so we could not analyze these data (data not shown). Moreover, we applied another oligomer α-syn specific antibody, a clone named Syn33, to our established ELISA and observed a similar result to the Oligo5-αS ELISA. However, the sensitivity using the Syn33 antibody was too low for urinary α-syn analysis (data not shown).

OPEN IN VIEWER

Fig. 5

Analysis of the distinct conformations of α-syn oligomers. A) A scheme for the generation and isolation of oligomeric standards. B) Isolated α-syn oligomer in the 100 K-cut fraction was validated using three different capture antibodies. C) A serial dilution of oligomeric standards showed a dose-dependent presence of monomeric and oligomeric α-syn. The arrow indicates oligomeric α-syn. D) Dot-blot assay of the oligomeric standards for Oligo5-αS ELISA was consistent with the results of the western blot assay. E) Oligomeric standards made from the 100 K-cut fraction showed linearity, whereas the serial dilution of α-syn in the 30 K-FT fraction was not appropriate for oligomeric standards. F) Urine samples subjected to Oligo5-αS ELISA showed significantly lower levels in the PD group than in the non-PD group. G) Analysis of the association between Oligo5-αS levels and H & Y stage. Analysis of the correlation of Oligo5-αS levels with Total- αS levels in the non-PD (H) and PD (I) groups and Fila-αS levels in the non-PD (J) and PD (K) groups was also performed. A Student’s t-test and non-zero slope significance were used for the analyses. *p < 0.05.