P2X4 receptor controls microglia activation and favors remyelination in autoimmune encephalitis

Figure 2. P2X4R blockade increases autoimmune inflammation

• A. Clinical score of vehicle (n = 10)‐ and TNP‐ATP (10 mg/kg)‐treated mice (n = 10) after EAE induction. Right, scheme, raw data, and histogram showing axon conduction latency in the corticospinal tract of control (n = 4), EAE (n = 10), and TNP‐ATP‐treated EAE mice (n = 10). Symbols indicate significance versus control (*) or versus EAE (^#).

• B. Histology of spinal cord sections using Iba1 antibodies shows a significant increase in microglia/macrophage cell number in control mice (n = 3), in TNP‐ATP‐treated EAE mice (n = 4), and in non‐treated EAE mice (n = 5). Scale bar = 50 μm.

• C. Neurological score, axon conduction latency, and microglia/macrophage quantification in control WT (n = 3) and P2X4^−/− mice (n = 3) and after EAE induction in WT (n = 10) and P2X4^−/− mice (n = 10).

• D. Neurological score after EAE induction in P2X4^−/− mice treated with vehicle (n = 4) or TNP‐ATP (n = 5).

• E. Left, flow cytometry gating strategy for analysis of infiltrates in the spinal cord of EAE mice at peak (18–21 dpi). Right, graph showing the neurological score of the mice used for the analysis (top) (n = 8). Histogram showing flow cytometric quantification of CD3⁺, CD8⁺, CD4⁺, and γδ T cells, neutrophils (NP), macrophages (MC), and dendritic cells (DC) in the spinal cord at EAE peak (bottom).

• F. Immune response analysis in vitro. T‐cell proliferation assay and flow cytometric quantification of cytokine expression in T cells after PMA/ionomycin stimulation in the absence or presence of TNP‐ATP (10–30 μm, 3 days; n = 3 independent experiments).

Data information: Data are presented as mean ± s.e.m. Statistics were performed with Mann–Whitney U‐test (neurological score) and Student's t‐test. *^/#P < 0.05, **P < 0.01, ***P < 0.001.

Figure EV1. Immune cell analysis in EAE and TNP‐ATP‐treated EAE mice

• A. Correlation between the neurological score and the number of Iba1⁺ cells in gray matter and white matter of the spinal cord.

• B. Flow cytometric quantification of immune cell infiltrates (gating as in Fig 2E) in the brain of TNP‐ATP‐treated or non‐treated mice analyzed at the peak of symptoms after EAE induction (n = 8). Data are presented as mean ± s.e.m. and were analyzed by Student's t‐test. *P < 0.05.

Figure 3. P2X4R does not interfere with immune priming

• A. Neurological score of vehicle (n = 23 from three independent experiments)‐ and TNP‐ATP‐treated (n = 21 from three independent experiments) mice after EAE. Mice were treated daily with TNP‐ATP from 0 dpi to EAE peak.

• B. Flow cytometric quantification of CD8⁺, CD4⁺, and γδ T cells in lymph nodes (LN), spleen, and spinal cord at EAE peak of vehicle (n = 7)‐ and TNP‐ATP‐treated mice (n = 5) (gating strategy as described in Fig 2E).

• C. Relative mRNA expression of Foxp3, Ifng, and Ror in LN, spleen, and spinal cord at EAE peak of vehicle (n = 6)‐ and TNP‐ATP‐treated mice (n = 7).

• D. Representative images of ¹⁸F‐MMPi PET imaging in control mice (n = 6), EAE mice (n = 10), and EAE mice treated with TNP‐ATP (n = 9) as described in (A). ¹⁸F‐MMPi signal in the lumbar spinal cord was expressed as %ID/g.

Data information: Data are presented as mean ± s.e.m. and were analyzed by one‐way ANOVA. ***P < 0.001.

Figure 4. P2X4R blockade increases pro‐inflammatory gene expression after EAE

• A. Neurological score of EAE (n = 4) or TNP‐ATP‐treated EAE mice (n = 4) used for gene expression analysis. Mice were treated from the onset to the end of the experiment. Data are presented as mean ± s.e.m. and were analyzed by Mann–Whitney U‐test. *P < 0.05.

• B. Heatmap showing significant changes in pro‐inflammatory and anti‐inflammatory mRNA expression in the spinal cord at the EAE recovery phase in the presence or absence of TNP‐ATP (n = 3). The expression levels of genes are presented using fold‐change values transformed to Log2 format compared to control. The Log2 (fold‐change values) and the color scale are shown at the bottom of the heatmap. Tables indicate statistical significance between control and EAE (*) and between EAE and EAE + TNP‐ATP (^#). Data were analyzed by Student's t‐test.

• C. Heatmap showing changes in pro‐inflammatory and anti‐inflammatory markers in FACS‐isolated microglia from control, EAE WT, and EAE P2X4^−/− mice at the recovery phase (n = 2 in duplicate). Tables indicate statistical significance between control WT and EAE WT (*) and between EAE WT and EAE P2X4^−/− (^#). Data were analyzed by Student's t‐test.

• D, E iNOS (green) expression was increased in Iba1⁺ cells (red) in the spinal cord of P2X4^−/− mice versus WT mice (D) and in TNP‐ATP‐treated mice versus non‐treated mice after EAE induction (E) (n = 3). Analysis was performed at the recovery phase of the EAE. Scale bar = 50 (D, top; E) and 25 (D, bottom) μm.

Figure EV2. Heatmap of pro‐inflammatory and anti‐inflammatory genes at EAE peak

Expression of most pro‐inflammatory genes and some anti‐inflammatory genes was increased in the spinal cord of EAE mice (n = 3) at the peak phase, but TNP‐ATP did not induce significant changes (n = 3). Tables indicate statistical significance between control and EAE (*) and between EAE and EAE + TNP‐ATP (^#). Statistics were performed with Student's t‐test.

Figure 5. P2X4R modulates microglia polarization

• A. Schematic representation of microglia polarization protocol.

• B. Staining for iNOS (red) (n = 7) and mannose receptor (MRC1, green) (n = 4) in different activated microglia in the absence or presence of TNP‐ATP (10 μM). Scale bar = 50 μm.

• C. qPCR quantification of pro‐inflammatory genes (Ccl2 and Nos2) and anti‐inflammatory genes (Arg1 and Mrc1) in different activated microglia in the absence or presence of TNP‐ATP (10 μM) (n = 3).

Data information: Data are presented as mean ± s.e.m. and were analyzed by one‐way ANOVA. *^/#P < 0.05, **^/##P < 0.01, ***P < 0.001 versus control (*) or versus pro‐/anti‐inflammatory microglia (^#).

Figure EV3. Pro‐inflammatory and anti‐inflammatory polarization in wild‐type and P2X4^−/− microglia

Staining for iNOS (red) and mannose receptor (MRC1, green) in control, pro‐inflammatory, and anti‐inflammatory microglia from wild‐type and P2X4^−/− mice. Scale bar = 50 μm (n = 4 experiments performed in triplicate). Data are presented as mean ± s.e.m. and were analyzed by Student's t‐test. *^/#P < 0.05, **P < 0.01. Symbols indicate significance versus control (*) or versus pro‐/anti‐inflammatory microglia (^#).

Figure 6. Microglial P2X4R modulates oligodendrocyte differentiation

• A. Microglia–OPC coculture stained for P2X4R (left column) or P2X7R (right column) (red), Olig2 (white), isolectin B4 (IB4; green), and Hoechst (blue). Scale bar = 20 μm. Notice the presence of P2X4R on isolectin B4⁺ microglial cells and the absence in Olig2⁺ cells. Histograms (right) show quantification of P2X4R and P2X7R at different stages of oligodendrocyte development, 4 h in vitro (HIV), 1 day in vitro (DIV), and 3 DIV in cocultures (n = 3).

• B. Microglia (MG) express functional P2X7R (blocked with A438079) and P2X4R (blocked with TNP‐ATP), whereas oligodendrocytes (OL) only express functional P2X7 receptors. Representative traces showing ATP (100 μM)‐evoked inward, non‐desensitizing currents in microglia and oligodendrocytes in acute slices in the absence or presence of P2X7R antagonist A438079 (1 μM) or of P2X4R antagonist TNP‐ATP (10 μM) (n = 9 cells from three different mice). Symbols indicate significance versus ATP currents (*) or versus ATP+A438079 currents (^#).

• C. OPCs were treated with control, anti‐inflammatory, and pro‐inflammatory microglia‐conditioned media (MCM) or with fresh media ± polarizing factors and stained for NG2 (green) and MBP (red). Scale bar = 20 μm (n = 3). Symbols indicate significance versus polarizing factors (*) or versus MCM (^#).

• D. Representative immunoblots of BDNF levels in microglia in vitro and densitometry quantification (n = 10).

• E. Bdnf and Mbp mRNA expression (bottom) and correlation of the expression (top) at the EAE recovery phase in control mice, EAE mice, and TNP‐ATP‐treated EAE mice (n = 10 mice/group).

• F. Sagittal sections of cerebellum (300 μm) were treated with lysolecithin (LPC) to induce demyelination and allowed to remyelinate ± TNP‐ATP (10 μM) during 7 days. Representative immunoblots of MBP and GAPDH and densitometry quantification (n = 5).

Data information: Data are presented as mean ± s.e.m. and were analyzed by one‐way ANOVA (B, C, E) and Student's t‐test (D, F). *^/#P < 0.05, **^/##P < 0.01, ***^/###P < 0.001.Source data are available online for this figure.

Figure EV4. Functional role of P2X4R in oligodendrocytes in vitro

• A. Effect of ATPγS (10 μM) ± ivermectin or TNP‐ATP (10 μM) on oligodendrocyte differentiation quantified on the basis of the number of MBP⁺ cells/total cell number and the area of the cells. Area was calculated with ImageJ software (n = 3).

• B. Oligodendrocyte cell viability after 24‐h incubation with BzATP, a broad‐spectrum agonist of purinergic receptors, in the absence or presence of TNP‐ATP (10 μM) (left) or ATPγS (right) in the absence or presence of IVM (3 μM) (n = 3).

• C. Western blot analysis of BDNF in control and IVM‐treated wild‐type and P2X4^−/− microglia (n = 3).

• D. Bdnf mRNA levels in the spinal cord of WT and P2X4^−/− mice in control and after EAE induction (n = 4 mice/group).

Data information: Data are presented as mean ± s.e.m. and were analyzed by one‐way ANOVA (B) and Student's t‐test (C, D). *P < 0.05, ***P < 0.001.Source data are available online for this figure.

Figure 7. P2X4R modulates myelin phagocytosis

• A. Spinal cord sections of control mice, EAE mice, and EAE mice treated with TNP‐ATP at the recovery phase stained for MBP (red) and Iba1 (green). Scale bar = 50 μm. Representative images from n = 6 mice/group from two independent experiments.

• B. Alexa 488‐labeled myelin endocytosis (1 h, 37°C) in Iba1⁺ microglia polarized in the absence or presence of TNP‐ATP (10 μM). Scale bar = 50 μm.

• C. Degradation of Alexa 488‐labeled myelin by control, pro‐inflammatory, and anti‐inflammatory microglia at chase time 0, 3, and 6 days. Left, representative images at 0 and 6 days. Scale bar = 50 μm. Right, histogram shows the percentage of 488‐myelin retained in the cells after the 3‐ and 6‐day chase periods with respect to the 488‐myelin at 0‐h chase time.

• D. Histogram represents the effect of TNP‐ATP (10 μM), applied during differentiation and chase time (6 days), on myelin degradation. Myelin degradation at 6 days was expressed as fold change versus control microglia at the same chase time.

Data information: Data are presented as mean ± s.e.m. and were analyzed by Student's t‐test. *^/#P < 0.05, ^##P < 0.01. Symbols indicate significance versus control (*) or versus pro‐/anti‐inflammatory microglia (^#). In vitro data come from n = 4 independent experiments performed in duplicate.

Figure EV5. Role of P2X4R in myelin phagocytosis and lysosome function

• A. Spinal cord sections of wild‐type (n = 6) and P2X4^−/− mice (n = 7) 35 days after EAE induction stained for MBP. Scale bar = 50 and 10 (inset) μm. Inset images come from different fields at higher magnification. Right, demyelination score analyzed in MBP‐stained sections.

• B. Effect of IVM (3 μM) on myelin endocytosis (1 h) and degradation (3 days) in microglia from wild‐type and P2X4^−/− mice (n = 3).

• C. Treatment with IVM (3 μM; 16 h) increased late endosome–lysosome (LEL) size in microglia from wild‐type and P2X4^−/− mice. Representative images and histogram for LEL size distributions under the conditions indicated (n = 20 cells from two independent experiments). Scale bar = 10 μm.

• D. IVM‐treated microglia (3 μM) and anti‐inflammatory microglia showed an acidic shift in lysosomal pH, as analyzed on the basis of fluorescein/rhodamine–dextran ratio (n = 20 cells from two independent experiments). Scale bar = 10 μm.

Data information: Data are presented as mean ± s.e.m. and were analyzed by one‐way ANOVA. ^#P < 0.05, **P < 0.01, ***P < 0.001. Symbols indicate significance versus control (*) or versus pro‐inflammatory microglia (^#).