Figure legends
Fig.1 HMGB1 A Box reduces macrophage infiltration in SN parenchyma. (A and B) Brains of animals receiving saline, MPTP, MPTP + HMGB1 A Box injection were separated and made into serial frozen sections of 30μm thickness were prepared at the end of the 7-day treatment, and sections of the SN site were subjected to CD45, F4/80, CD11c, CD31 immunofluorescence staining.
Fig.2 HMGB1 A Box reduces macrophages and Th17 infiltration in SN and protecting TH+ neurons. (A and B) Brains of animals receiving saline, MPTP, MPTP + HMGB1 A Box, MPTP + clodronate liposomes were separated after 7 days of treatment and serial frozen sections were made and then the sections of SN site were subjected to TH immunization group chemical staining, extract the protein from the SN for western blot detection of TH, and finally, the gray value of the results is counted (n=3; unpaired t-test; mean ± s.d.). (C) Digest mouse SN cells to obtain single cell suspension, and perform flow cytometry detection of CD45, CD11b, CD11c, MHC II, CD3, CD4, IFN-γ, IL-4 and IL-17A. (D-F) Proportion of CD45intCD11b+ microglia, CD45hiCD11b+ macrophages and CD45hiCD11c+ DCs in mouse SN (n=5; unpaired t-test; mean ± s.d.). (G-H) MHC II mean fluorescence intensity of CD45intCD11b+ microglia, CD45hiCD11b+ macrophages and CD45hiCD11c+ DCs in SN (n=5; unpaired t-test; mean ± s.d.). (J) Numbers of CD4+IFN-γ+ cells, CD4+IL-4+ cells and CD4+IL-17A+ cells in SN (n=3; unpaired t-test; mean ± s.d.). *P < 0.05, **P < 0.01.
Fig.3 Peripheral monocytes/macrophages promote MPTP-induced Th17 differentiation in mouse SN and are inhibited by HMGB1 A Box. (A) Mouse MPTP induction and midbrain cell isolation and culture timeline. (B) Mouse primary cells from the midbrain and CD4+ naïve T cells co-culture groups. (C) Co-cultured cells were collected for CD45, CD11b, CD4 and IL-17A flow cytometry assays. (D and G) Ratio of IL-17A+ cells to CD4+ cells in each group of cells (n=4; unpaired t-test; mean ± s.d.). (E) Number of CD45intCD11b+ microglial in each group (n=4; unpaired t-test; mean ± s.d.). (F) Number of CD45hiCD11b+ macrophages in each group (n=4; unpaired t-test; mean ± s.d.). *P < 0.05, **P < 0.01.
Fig.4 HMGB1 complexes with CXCL12 mediate macrophage migration.(A and B) Peritoneal macrophages were co-cultured with MPP+-treated primary midbrain cells in transwell system (n=4 random high-magnification fields of view; unpaired t-test; mean ± s.d.). (C) Immunofluorescence staining of 6×His and CXCR4 after 6 hour-HMGB1 A Box treatment of peritoneal macrophages. (D) After the culture supernatant of MPP+-treated midbrain cells was concentrated, the binding of HMGB1/CXCL12 was detected by immunoprecipitation assay. (E) Mouse peritoneal macrophages were cultured with MPP+ -treated midbrain cell supernatant for 6 hours, and the binding of CXCR4 to HMGB1 and CXCL12 was detected by immunoprecipitation assay. (F) Peritoneal macrophages were treated with HMGB1 A Box for 6 hours, and the binding of HMGB1 A Box-6×His and CXCR4 were detected by immunoprecipitation assay. (G and H) Midbrain microglia (mic), astrocytes (ast) and neurons (neu) HMGB1 were knocked down separately and co-cultured with peritoneal macrophage in transwell system in different combinations, and cells in the lower chamber were treated with MPP+ (n=4 random high-magnification fields of view; unpaired t-test; mean ± s.d.). ***P < 0.005, ****P < 0.001
Fig.5 CXCR4 of CD3+ T cell mediates their infiltration in the SN. (A) Immunofluorescence staining of 6×His and CXCR4 after HMGB1 A Box-6×His treatment of magnetic bead-sorted CD3+ T cells for 6h. (B) CD3+ T cells were treated with HMGB1 A Box for 6 hours, and the binding of HMGB1 A Box-6×His and CXCR4 were detected by immunoprecipitation assay. (C) FCM detection in peripheral blood of MPTP mice that were transfused with CD3+ T cells from GFP mice (n=3; unpaired t-test; mean ± s.d.). (D) GFP and CD3 tissue immunofluorescence staining of MPTP mouse SN which were transfused with CD3+ T cells from GFP mice. *P < 0.05, **P < 0.01,
Fig. 6 HMGB1/CXCL12-CXCR4 exists in the serum of PD patients.(A and B) Serum HMGB1 and CXCL12 levels of PD patients and healthy people were detected by ELISA. (C) Correlation analysis between serum HMGB1 levels and CXCL12 levels of PD patients. (D) Co-immunoprecipitation experiments of HMGB1 and CXCL12 in serum of PD patients. (E) Correlation analysis of serum HMGB1 level and age in PD patients. (F) Correlation analysis between serum CXCL12 level and age in PD patients. (G) Binding of HMGB1 and CXCL12 to CXCR4 in PD patient CD14+ monocytes by co-immunoprecipitation. (H) Immunofluorescence detection of CXCR4, HMGB1 and CXCL12 on CD3+ T cells and CD14+ monocytes in PD patients. (n=20; unpaired t-test; mean ± s.d.) *P<0.05, **P<0.01.
Fig. 7 HMGB1 A Box suppresses immune cell responses in the SN of PD mice and its mechanism. After neuronal damage in SN, frHMGB1 passively released from neurons binds to CXCL12 in the surrounding environment to form frHMGB1/CXCL12 complex. This complex can bind to CXCR4 on the surface of nearby microglia, macrophages in blood or lymphatic capillaries and T cells to induce their migration and infiltration into the SN parenchyma. The macrophages and microglia infiltrated in the SN jointly promote T cells to differentiate into Th17 and continue to damage neurons. Intravenous injection of HMGB1 A Box competitively inhibited this process by binding CXCR4 of these peripheral immune cells.