Neuron-released oligomeric α-synuclein is an endogenous agonist of TLR2 for paracrine activation of microglia

Abstract

Abnormal aggregation of α-synuclein and sustained microglial activation are important contributors to the pathogenic processes of Parkinson's disease. However, the relationship between disease-associated protein aggregation and microglia-mediated neuroinflammation remains unknown. Here, using a combination of in silico, in vitro and in vivo approaches, we show that extracellular α-synuclein released from neuronal cells is an endogenous agonist for Toll-like receptor 2 (TLR2), which activates inflammatory responses in microglia. The TLR2 ligand activity of α-synuclein is conformation-sensitive; only specific types of oligomer can interact with and activate TLR2. This paracrine interaction between neuron-released oligomeric α-synuclein and TLR2 in microglia suggests that both of these proteins are novel therapeutic targets for modification of neuroinflammation in Parkinson's disease and related neurological diseases.

Figure 1

dSY5Y-released α-synuclein activates microglia. Rat primary microglia was treated with LZCM, αSCM, or LPS (1 μg/ml, a positive control) for the indicated hours. LPS is an endotoxin that activates microglial responses that we tested, and therefore, was used as a positive control. (a) Percentage of microglial cells with amoeboid morphology (n = 6). (b) NO produced from microglia (n = 5). (c) Microglial proliferation (n = 6). (d) Relative expression of IL-1μ mRNA. Real-time PCR data were normalized with the average value of LZCM (n = 3). (e) Quantification of intracellular ROS levels using flow cytometry. This is the representative result of three independent experiments. (f) Quantification of cytokines using ELISA in the microglial culture media (n = 3). (g) Depletion of α-synuclein and microglia activation activity from αSCM. Three successive rounds of depletion were performed using an affinity resin. For cytokine induction (n = 3), the amount of α-synuclein in αSCM used was 0.1 μg/ml. (h) Cytokine induction by his-tagged α-synuclein pulled down from his-αSCM. Western blot shows different amounts of pulled-down α-synuclein used in microglia activation (n = 3). Morphology analysis (a), NO production analysis (b), proliferation assay (c), iROS production (e), and cytokine ELISA quantification (f) were performed at 24 hours post treatment. Relative mRNA expression (d,g,h) was determined at 2 hours post treatment. Morphology analysis (a), NO production analysis (b), relative mRNA expression (g,h) data were compared by one-way ANOVA. Proliferation assay (c), cytokine gene expression (d), and cytokine ELISA (f) data were analyzed using unpaired t-test. Error bars represent ± s.e.m. *P < 0.05; **P < 0.01; ***P < 0.001. “n” represents the number of independent experiments and each experiment was performed at least in triplicate.

Figure 2

Hypothetical signaling network of microglia activated by cell-released α-synuclein. (a) Hypothetical signaling network was constructed by using DEGs from αSCM-exposed microglia. Alterations of gene expression were detected at two different time points (Inner circle; 6 hours and Outer circle; 24 hours). (b) RT-PCR analysis of genes identified in the network. (c, d) IκB degradation (c) and phosphorylation of p38 MAP kinase (d) in microglia exposed to αSCM for 15 minutes (n = 4). All data were analyzed using unpaired t-test. Error bars represent ± s.e.m. **P < 0.01. “n” represents the number of independent experiments.

Figure 3

Microglia activation by cell-released α-synuclein is mediated by TLR2. (a) Expression of cytokines and chemokines upon treatment of αSCM in wild type (WT), Tlr2^-/-, Tlr3^-/-, and Tlr4^-/- mouse microglia (n = 3). (b) αSCM-induced cytokine production and release in WT and Tlr2^-/- microglia (n = 3). (c) Effects of TLR2 blocking antibody (T2.5) on induction of IL-1β mRNA. Microglia were pre-incubated with either T2.5 or control IgG for 30 min before addition of αSCM or LZCM (n = 3). (d) TLR2 activity was determined in the HEK-Blue-TLR2 reporter cells. Pam3CSK4 (10 ng/ml) is a known TLR2 agonist and used as a positive control (n = 3). (e) Induction of IL-1β mRNA by different amounts of α-synuclein purified from αSCM (n = 3). (f) TLR2 activation by the endogenous α-synuclein released from mouse primary cortical neurons. Culture media from wild type and α-synuclein null mice primary neurons were treated to the HEK-Blue-TLR2 reporter cells (n = 3). (g, h) TLR2-dependent IκB degradation (g) and phosphorylation of p38 MAP kinase (h) in microglia exposed to αSCM (n = 3). Relative mRNA expressions (a,c,e) were determined at 2 hours post treatment. Cytokine ELISA (b) was performed at 6 hours post treatment. IκB degradation (g) and p38 phosphorylation (h) analyses were performed at 15 minutes post treatment. All data were analyzed using unpaired t-test. Error bars represent ± s.e.m. *P < 0.05; **P < 0.01; ***P < 0.001. “n” represents the number of independent experiments, and each experiment was performed at least in triplicate.

Figure 4

Properties of TLR2 as the receptor for cell-released α-synuclein. (a) Competition between TLR2 antibody (T2.5) and dSY5Y-released α-synuclein for binding to the surface of BV-2 microglia (n = 3). Scale bars: 10 μm. (b) Uptake of α-synuclein of αSCM by wild type and Tlr2^-/- mouse microglia. Cells were treated with αSCM for 30 minutes, and then analyzed. (c) Effect of a TLR2 blocking antibody on the internalization of α-synuclein of αSCM into microglia. Primary mouse microglia was pre-incubated with either TLR2 blocking antibody (T2.5) or control (IgG) for 30 minutes before addition of αSCM to the media for 2 hours. (d) Ectopic TLR2 expression enhanced uptake of α-synuclein of αSCM in COS7 cells. (e) Colocalization of dSY5Y-released α-synuclein with TLR2 in mouse microglia after internalization (top panels). Tlr2^-/- mouse microglia show markedly reduced uptake (bottom panels). Scale bars: 10 μm. (f) Fluorescence intensities of internalized α-synuclein in microglia (n = 3). All data were analyzed using one-way ANOVA. Error bars represent ± s.e.m. **P < 0.01; ***P < 0.001. “n” represents the number of independent experiments (see the methods section).

Figure 5

TLR2-dependent microglial activation by α-synuclein overexpression in nigral dopaminergic neurons. (a) Experimental scheme. (b) Immunofluorescence images of the SN (dotted lines). Scale bars: 100 μm (c) MHC II fluorescence in the SN of wild type mice that received striatal injections of viral vectors expressing GFP and α-synuclein (n = 5). (d) MHC II fluorescence in the SN of wild type and Tlr2^-/- mice after striatal injections of α-synuclein vectors (n = 5). (e) Correlation between MHC II and GFP/α-synuclein fluorescence in SN (WT; n = 9, Tlr2^-/-; n = 7, GFP; n = 12). MHC II fluorescence data (c,d) were analyzed using the Mann-Whitney U test. Error bars represent ± s.e.m. *P < 0.05. “n” represents the number of animals.

Figure 6

Conformation-sensitive agonist activity of α-synuclein for TLR2. (a) Development of TLR2 agonist activity with endotoxin-free recombinant α-synuclein by in vitro incubation (n = 3). (b) Separation of “aged” recombinant α-synuclein by SEC. The line trace and the bars indicate TLR2 activity and the α-synuclein levels, respectively. (c) Dot blot analysis of fractions from b. (d) CD spectra of fractions 8 and 14 from b. (e) Binding of pure α-synuclein oligomers (fraction 8) on the surface of HEK293 cells expressing human TLR2/CD14 (n = 3). (f) Competition between TLR2 antibody (T2.5) and pure preparation of oligomeric α-synuclein for the binding to the surface of BV-2 microglia (n = 3). Binding data of α-synuclein oligomers (e) were analyzed by unpaired t-test. TLR2 fluorescence data (f) were analyzed using one-way ANOVA. Error bars represent ± s.e.m. *P < 0.05; **P < 0.01. “n” represents the number of independent experiments (see the methods section).

Figure 7

Conformation-sensitive agonist activity of cell-released oligomeric α-synuclein for TLR2. (a) Separation of αSCM by SEC. The line trace and the bars indicate TLR2 activity and the α-synuclein levels, respectively. (b) Dot blot analysis of fractions from a. (c) MHC II fluorescence in the cerebral cortices of wild type and Tlr2^-/- mice after injections of fraction 8 and fraction 14 from a (n = 5). Data were analyzed using unpaired t-test. Error bars represent ± s.e.m. *P < 0.05; **P < 0.01. “n” represents the number of animals.

Figure 8

TLR2 expression in DLB and α-synuclein transgenic mice. (a) Immunohistochemical analysis of human (top panels, n = 6) and mouse samples (bottom panels, n = 6). The TLR2 antibody immunolabeled neuronal cells (N) as well as microglial cells (arrow heads). (b) Double immunofluorescence studies with antibodies against TLR2 and Iba-1 showing colocalization of the two markers in microglial cells (top panels). These cells usually were close to capillaries (c) that also displayed TLR2 labeling. Double labeling for TLR2 and α-synuclein in the α-synuclein transgenic mice (bottom panels). (c) Western analysis of brain homogenates. All data were analyzed using unpaired t-test. Error bars represent ± s.e.m. *P < 0.05; **P < 0.01. “n” represents the number of patients and animals.