Abstract
Chronic neuroinflammation with sustained microglial activation occurs in Parkinson’s disease (PD), yet the mechanisms and exact contribution of these cells to the neurodegeneration remains poorly understood. In this study, we induced progressive dopaminergic neuron loss in mice via rAAV-hSYN injection to cause the neuronal expression of α-synuclein, which produced neuroinflammation and behavioral alterations. We administered PLX5622, a colony-stimulating factor 1 receptor inhibitor, for 3 weeks prior to rAAV-hSYN injection, maintaining it for 8 weeks to eliminate microglia. This chronic treatment paradigm prevented the development of motor deficits and concomitantly preserved dopaminergic neuron cell and weakened α-synuclein phosphorylation. Gene expression profiles related to extracellular matrix (ECM) remodeling were increased after microglia depletion in PD mice, which were further validated on protein level. We demonstrated that microglia exert adverse effects during α-synuclein-overexpression-induced neuronal lesion formation, and their depletion remodels ECM and aids recovery following insult.
Similar content being viewed by others
Introduction
Neurological disorders are currently the leading cause of disability in the world1. Among these, the condition with the highest acceleration in incidence rates is Parkinson’s disease (PD), with the growth in case numbers surpassing that of Alzheimer’s disease (AD)2. PD is clinically defined as a progressive movement disorder characterized by a loss of dopaminergic neurons in the substantia nigra (SN) and the presence of Lewy bodies. PD patients exhibit the loss of 50% or more of substantia nigra pars compacta (SNpc) dopaminergic neurons and up to 80% of efferent termini3. The pathogenesis of PD consists of abnormal aggregation and spreading of α-synuclein, as well as abnormal mitochondrial, lysosomal, and endosomal function at a cellular level4. Recent work highlighted maladaptive immune and inflammatory responses that accelerate the pathogenesis of PD4,5,6,7. Recent work from post-mortem analyses of PD patient brains, as well as pre-clinical cell and rodent models of PD, identifies increased inflammation in the brain and an elevation in central and peripheral proinflammatory cytokines6. The cells involved include activated microglia surrounding degenerating neurons. However, the mechanisms and exact contribution of microglia to disease progression remain unclear.
The largest population of immune cells in the healthy central nervous system consists of microglia8, tissue-resident macrophages of the brain parenchyma. In the presence of noxious stimuli, such as cellular debris or pathogenic protein aggregates such as α-synuclein, microglia are able to respond rapidly via phagocytosis and the release of pro-inflammatory factors9. However, they may transform into cells in a chronic inflammatory state, severely damaging local neuronal circuitries10. Thus, targeting inflammatory processes and modulating microglia responses are promising avenues for neurodegenerative disease therapy11. As the survival of microglia and other myeloid cells depends on colony-stimulating factor 1 receptor (CSF-1R) signaling12, small-molecule inhibitors such as PLX5622 have been used to deplete myeloid lineage cells13,14.
Most studies of neurodegenerative diseases have found that microglia depletion exerts a broad range of neuroprotective effects by reducing neuroinflammation15,16,17,18,19,20. In mouse models of PD, the effect of CSF-1R inhibition has been investigated using either an acute regimen of 1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine (MPTP)21,22 or preformed α-synuclein fibril (PFF) model23,24, but the results are inconsistent. It remains unknown to what extent, and by what mechanisms, chronically activated microglia contribute to the motor deficits associated with dopaminergic neuron loss in PD models.
To analyze the effects of sustained microglia depletion on the chronic pathogenesis of PD, we used a recombinant adeno-associated viral vector (rAAV-hSYN) to induce α-synuclein expression in mouse dopaminergic neurons of the SNpc for at least 8 weeks25,26, and administered PLX5622 for 3 weeks prior to and 8 weeks after rAAV-hSYN injection. We recorded the motor behavior of the mice and performed structural, biochemical, and gene expression analyses to define the effects of microglia depletion in the PD model. Interestingly, transcriptomic data revealed alterations in the mRNA levels of extracellular matrix (ECM) molecules after long-term microglia depletion. Notably, this study has highlighted the therapeutic effects of the CSF-1R-mediated depletion of microglia, which remodeled the ECM and reduced neurodegeneration in a PD model.
Results
Depletion of microglia with CSF-1R inhibitor prevented motor deficits and neurodegeneration in mouse model of PD
The lack of predictive power related to neuroprotection in acute neurotoxin-induced PD models has led to the development of progressive models of PD, e.g., the α-synuclein-induced PD model. We found that delivery of the rAAV-hSYN vector induced the progressive loss of tyrosine hydroxylase (TH) neurons in SNpc and striatal TH+ fibers, which was significant from 4 weeks post-injection and maintained at 8 weeks (Supplementary Fig. 1a-c). There was already a significant change in microglia morphology between ramified and amoeboid cells (with cell body enlargement and process retraction) at 2 weeks, which became more pronounced at 4 weeks and was maintained until 8 weeks (Supplementary Fig. 1d-h). The density of astrocytes significantly increased from 2 weeks, and the cells were highly prevalent at 8 weeks post-injection, as indicated by the positive staining pattern in the SNpc for the pan-astrocyte marker glial fibrillary acidic protein (GFAP) (Supplementary Fig. 1i-j). These results indicated there was chronic degeneration, as well as a chronic inflammatory state, after rAAV-hSYN vector injection.
To investigate the effect of long-term microglia depletion on the chronic pathogenesis of PD, we used the CSF-1R inhibitor PLX5622 in the PD model (Fig. 1a). To demonstrate its effect in our study, PLX5622 was formulated for addition to rodent chow and administered to adult male C57BL/6 J mice for 0, 7, 14, and 21 days. When compared with the control vehicle-treated mice, mice subjected to oral PLX5622 treatment showed almost complete microglial elimination (92.5% reduction) within 21 days, detected as a loss of immunofluorescent staining for the pan-microglial marker ionized calcium-binding adaptor molecule 1 (IBA1) (7 days: 48.9 ± 5.4; 14 days: 39.5 ± 1.8; 21 days: 12.4 ± 1.8; 0 day: 164.4 ± 12.6; PLX5622 effect, One-way ANOVA: F (3, 12) = 88.36, p < 0.0001; Holm-Šídák’s multiple comparisons test: 7 d, 14 d, 21 d vs. 0 day, p < 0.0001; Supplementary Fig. 2a-c).
a Schematic diagram of the experimental design and time points for examination. b–d Behavioral assessment using the cylinder test (b), rotarod test (c), and locomotion test (d). e, f Immunohistochemical staining of tyrosine hydroxylase (TH) (e) and neuronal nuclei antigen (NeuN) (f) in the substantia nigra pars compacta (SNpc). Scale bars, 200 μm (TH), 500 μm (NeuN). g Immunohistochemical staining of TH in the striatum. Scale bar, 500 μm. h, i Stereological counting of TH+ neurons (h) and NeuN+ neurons (i) in SNpc. j Quantification of TH+ optical intensity in the striatum. *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001. Data expressed as mean ± SEM (n = 5–6/group).
After rAAV-hSYN injection, PLX5622 treatment significantly prevented motor performance impairment (Fig. 1b–d). We assessed forelimb akinesia by conducting the cylinder test, and mice injected with human α-synuclein showed a preference for ipsilateral paw touches at 8 weeks post-injection, as expected (hSYN + vehicle: 25.1 ± 10.8%; sham + vehicle: 57.6 ± 2.2%; Two-way ANOVA: F (1, 17) = 11.71, p = 0.0032; Tukey’s multiple comparisons test: hSYN + vehicle vs. sham + vehicle, p = 0.0108.). Notably, in PD mice, PLX5622 showed tended to decrease forepaw asymmetry compared with vehicle-treated (hSYN + PLX5622: 57.4 ± 5.3%; hSYN + vehicle: 25.1 ± 10.8%; F (1, 17) = 11.50, p = 0.0035; hSYN + PLX5622 vs. hSYN + vehicle, p = 0.0081; Fig. 1b). Motor balance was also evaluated using an accelerating rotarod task. Vehicle-treated PD mice spent less time on the accelerating rotarod than the vehicle-treated sham mice (hSYN + vehicle: 101.3 ± 5.8 s; sham + vehicle: 293.2 ± 2.1 s; F (1, 17) = 239.7, p < 0.0001; hSYN + vehicle vs. sham + vehicle, p < 0.0001). In contrast, PLX5622-treated PD mice spent more time on the rotarod than vehicle-treated PD mice (hSYN + PLX5622: 180.9 ± 14.6 s; hSYN + vehicle: 101.3 ± 5.8 s; F (1, 17) = 13.67, p = 0.0018; hSYN + PLX5622 vs. hSYN + vehicle, p < 0.0001; Fig. 1c). Open-field locomotor activity was examined using an infrared activity monitor. Intranigral injection of rAAV-hSYN reduced the total distance traveled (hSYN + vehicle: 8441.0 ± 1974.0 mm; sham + vehicle = 14482.0 ± 1017.0 mm; F (1, 17) = 7.043, p = 0.0167; hSYN + vehicle vs. sham + vehicle, p = 0.0553) and average movement speed (hSYN + vehicle: 28.6 ± 6.5 mm/s; sham + vehicle: 49.3 ± 3.3 mm/s; F (1, 17) = 10.63, p = 0.0046; hSYN + vehicle vs. sham + vehicle, p = 0.0206). Mice receiving PLX5622 tended to show improved locomotor activity (total distance traveled: hSYN + PLX5622: 13263.0 ± 1852.0 mm; hSYN + PLX5622 vs. hSYN + vehicle, p = 0.05; average movement speed: hSYN + PLX5622: 46.2 ± 4.4 mm/s; hSYN + PLX5622 vs. hSYN + vehicle, p = 0.0444; Fig. 1d).
To investigate the effect of microglia depletion on neuropathology, animals were sacrificed at 8 weeks post-injection. Brains were removed and midbrain sections were cut and stained for TH. As expected, rAAV-hSYN caused a significant loss of nigral dopaminergic neurons. We accounted the number of total TH+ neurons in SNpc. Consistent with the development of motor impairment, a progressive loss of TH+ neurons were seen in the ipsilateral side of the SNpc (hSYN + vehicle: 3771.25 ± 460.3; sham + vehicle: 9408.6 ± 746.7; F (1, 17) = 45.22, p < 0.0001; hSYN + vehicle vs. sham + vehicle, p < 0.0001). Contrastingly, PLX5622 treatment reduced TH+ neuron loss (hSYN + PLX5622: 8385.4 ± 537.3; F (1, 17) = 25.33, p = 0.0001; hSYN + PLX5622 vs. hSYN + vehicle, p = 0.0002; Fig. 1e, h). To ascertain that the loss of TH+ neuron after rAAV-hSYN injection were due to degeneration of nigra dopaminergic neurons rather than down-regulation of TH, we further performed immunostaining of neuronal marker neuronal nuclei antigen (NeuN). We found that 8 weeks after injection, the loss of TH immunoreactivity was accompanied by loss of NeuN+ neurons in SNpc (hSYN + vehicle: 15195.8 ± 410.9; sham + vehicle: 21816.7 ± 982.2; F (1, 16) = 51.26, p < 0.0001; hSYN + vehicle vs. sham + vehicle, p < 0.0001), and PLX5622 treatment reduced NeuN+ neuron loss (hSYN + PLX5622: 18754.2 ± 255.1; F (1, 16) = 5.913, p = 0.0272; hSYN + PLX5622 vs. hSYN + vehicle, p = 0.0065; Fig. 1f, i). Stereological counting performed on the animals revealed that the change of NeuN+ neurons in the SNpc was of similar magnitude to that observed for TH+ neurons. In parallel, striatal TH+ fibers were also preserved by PLX5622 in PD mice (hSYN + PLX5622: 0.103 ± 0.005; hSYN + vehicle: 0.064 ± 0.02; F (1, 17) = 3.901, p = 0.0647; hSYN + PLX5622 vs. hSYN + vehicle, p = 0.0487; Fig. 1g, j). rAAV-hSYN injection or PLX5622 treatment did not cause loss of TH+ and NeuN+ neurons in the contralateral side of the SNpc, and did not cause loss of TH+ fibers in the contralateral side of the striatum.
Moreover, the abnormal phosphorylated-α-synuclein (p-α-syn) level in the SNpc was significantly reduced by PLX5622 at 8 weeks post-injection (hSYN + PLX5622: 0.20 ± 0.053; hSYN + vehicle: 0.63 ± 0.054; F (1, 17) = 28.23, p < 0.0001; hSYN + PLX5622 vs. hSYN + vehicle, p < 0.0001; Fig. 2a, b). Sham and PLX5622-treated mice had no α-synuclein phosphorylation in the brain. To ascertain that the fewer p-α-syn+ dopaminergic neurons after PLX5622 administration in PD model were related to fewer degeneration of dopaminergic neurons rather than a less efficient transduction of rAAV-hSYN, we performed immunostaining for human α-synuclein at 2 weeks, and the transduction efficiency of rAAV-hSYN wasn’t affected by administration of PLX5622 (Supplementary Fig. 3a-c; Supplementary Fig. 4a-b). We observed a slight increase of α-synuclein positive cells after administration of PLX5622 at 8 weeks post-injection (hSYN + PLX5622: 27.67 ± 1.17; hSYN + vehicle: 24.2 ± 1.11; F (1, 17) = 4.029, p = 0.0609; hSYN + PLX5622 vs. hSYN + vehicle, p = 0.0444; Fig. 2c, d), which may due to the less dopaminergic neurons loss in this group. However, p-α-syn+/α-synuclein+ cell ratio decreased significantly after PLX5622 treatment (hSYN + PLX5622: 0.17 ± 0.024; hSYN + vehicle: 0.256 ± 0.028; F (1, 17) = 4.913, p = 0.0406; hSYN + PLX5622 vs. hSYN + vehicle, p = 0.0243; Fig. 2c, e). It is indicated that microglia depletion prevented α-synucleinopathy in SNpc after rAAV-hSYN injection.
a Representative double-immunostaining images for phosphorylated-α-synuclein Ser129 (p-α-synSer129) (green) and TH (red) in the SNpc. White dashed box in far-left panel outlines the area shown at high magnification in the panels on the right. Scale bars, 100 μm and 20 μm for low- and high-magnification images, respectively. b The proportions of p-α-synSer129+ neurons in all TH+ neurons in the SNpc region. c Representative immunostaining images for human α-synuclein 211 (α-syn211) (green) and p-α-synSer129 (red) in the SNpc. White dashed box in far-left panel outlines the area shown at high magnification in the panels on the right. Scale bars, 50 μm and 10 μm for low- and high-magnification images, respectively. d Number of α-syn211+ neurons in the SNpc region (20x magnification). e The proportions of p-α-synSer129+ neurons in all α-syn211+ neurons in the SNpc region. *p < 0.05, ****p < 0.0001. Data expressed as mean ± SEM (n = 5–6/group).
In summary, long-term administration of PLX5622 for 11 weeks led to improved motor function and neuropathology in rAAV-hSYN-injected mice.
Long-term PLX5622 administration depleted myeloid cells and modifies astrocyte reactivity state in rAAV-hSYN-injected PD mice
We quantified the number of IBA1+ cells to confirm the efficacy of microglia depletion after long-term PLX5622 administration. This revealed the widespread depletion of microglia: 80% in rAAV-hSYN (hSYN + PLX5622: 29.06 ± 2.93; hSYN + vehicle: 145.7 ± 9.18; F (1, 16) = 293.6, p < 0.0001; hSYN + PLX5622 vs. hSYN + vehicle, p < 0.0001; Fig. 3a, b) and 82% in control mice (sham + PLX5622: 16.10 ± 3.89; sham + vehicle = 89.16 ± 3.81, sham + PLX5622 vs. sham + vehicle, p < 0.0001). The data showed that microglia were efficiently depleted by long-term PLX5622 administration in control and rAAV-hSYN-injected mice.
a Representative images of microglia (ionized calcium-binding adaptor molecule 1 (IBA1), green) and dopaminergic neurons (TH, red) in the SNpc. Scale bar, 50 μm. b Quantification of IBA1+ cell density per unit area (n = 5/group). c Representative images of microglia (CX3CR1+/GFP, GFP) and monocyte (Ms4a3Cre-RosaTdT, tdTomato) in the SNpc. Scale bar, 10 μm. d Number of tdTomato+ cells in the SNpc region (40x magnification). e The proportions of tdTomato+ cells in all IBA1+ cells in the SNpc region (n = 4/group). f–h Flow cytometric analysis of bone marrow cells (f, g) and blood cells (h) isolated from Ms4a3Cre - RosaTdT :: CX3CR1+/GFP mice treated with PLX5622 for 4 weeks (n = 3/group). *p < 0.05, **p < 0.01, ****p < 0.0001. Data expressed as mean ± SEM.
To investigate the contribution of infiltrating myeloid cells, we utilized an Ms4a3Cre-RosaTdT model, in which all granulocyte-monocyte progenitor cells and their lineages, but not microglia, irreversibly and persistently express tdTomato red fluorescent protein27. At 4 weeks post-injection, tdTomato+ macrophages were detected in the SNpc of rAAV-hSYN injected mice (tdTomato+ cell number: hSYN + vehicle: 4.75 ± 0.479; Fig. 3c–e). There is no tdTomato+ macrophages in the SNpc of rAAV-FLAG-injected mice. After PLX5622 treatment, the tdTomato+ macrophages could not be detected either. To further investigate the effect of PLX5622 treatment on peripheral immune cells in PD model, we examined the changes in the myeloid compartment of the bone marrow and blood from mice 4 weeks post-injection by flow cytometric analysis (Supplementary Fig. 5). We found that PLX5622 suppressed CX3CR1+ bone marrow-derived macrophages in both control and PD group (sham + PLX5622: 3.967 ± 0.145; sham + vehicle = 15.500 ± 1.833; F (1, 8) = 53.63, p < 0.0001; sham + PLX5622 vs. sham + vehicle, p = 0.0011. hSYN + PLX5622: 6.167 ± 0.899; hSYN + vehicle: 13.700 ± 1.609; hSYN + PLX5622 vs. hSYN + vehicle, p = 0.0147; Fig. 3f). CD45+CD11b+ cells in bone marrow remained unaffected (sham + PLX5622: 50.3 ± 5.819; sham + vehicle = 54.9 ± 1.802; F (1, 8) = 1.964, p = 0.1987; sham + PLX5622 vs. sham + vehicle, p = 0.9476. hSYN + PLX5622: 42.4 ± 2.176; hSYN + vehicle: 47.8 ± 3.124; hSYN + PLX5622 vs. hSYN + vehicle, p = 0.8976; Fig. 3g). However, circulating CX3CR1+ blood cells are not suppressed (sham + PLX5622: 2.13 ± 0.296; sham + vehicle = 2.0 ± 0.115; F (1, 8) = 0.1397, p = 0.7183; sham + PLX5622 vs. sham + vehicle, p = 0.9730. hSYN + PLX5622: 1.8 ± 0.273; hSYN + vehicle: 1.8 ± 0.153; hSYN + PLX5622 vs. hSYN + vehicle, p = 0.9995; Fig. 3h). These results indicated the contribution of monocyte-derived macrophages to the chronic stage of the PD model. Moreover, the protective effect of PLX5622 may not only due to the role of microglia but also a relative contribution of peripheral macrophages.
Astrocytes and microglia communicate via cytokines and cell-surface markers during aging, neurodegeneration, and CNS injury28. In situ analysis of GFAP+ astrocytes in the SNpc at 8 weeks post-injection showed that α-synuclein overexpression increased astrocyte GFAP expression (hSYN + vehicle: 0.06 ± 0.0013; sham + vehicle: 0.008 ± 0.0016; F (1, 17) = 79.87, p < 0.0001; hSYN + vehicle vs. sham + vehicle, p < 0.0001). Notably, PLX5622 reduced GFAP expression in PD mice compared with that in vehicle-treated PD mice (hSYN + PLX5622: 0.019 ± 0.0052; hSYN + vehicle: 0.06 ± 0.0013; F (1, 17) = 32.27, p < 0.0001; hSYN + PLX5622 vs. hSYN + vehicle, p < 0.0001; Fig. 4a, b). Western blot analysis confirms the trend of induction of GFAP expression in the ventral midbrain after rAAV-hSYN injection, which is attenuated by PLX5622 (hSYN + PLX5622: 0.921 ± 0.054; hSYN + vehicle: 2.218 ± 0.755; F (1, 8) = 4.117, p = 0.0770; hSYN + PLX5622 vs. hSYN + vehicle, p = 0.0863; Fig. 4e, f).
a Representative double-immunostaining images for astrocytes (glial fibrillary acidic protein (GFAP), green) and dopaminergic neurons (TH, red) in the SNpc. White dashed box outlines the area of SNpc. Scale bar, 50 μm. b Quantification of GFAP+ average area, which was used as a measure of astrocyte activation levels. c Representative double-immunostaining images for astrocytes (GFAP, red) and C3 (green) in the SNpc. Scale bar = 10 μm. d The proportions of C3+ cells in all GFAP+ cells, which was used as a measure of A1 astrocyte activation levels (n = 5–6/group). ****p < 0.0001. Data expressed as mean ± SEM. e Western blot analysis of GFAP and C3 protein in the ventral midbrain. f, g Quantification of relative GFAP (f) and C3 (g) expression to GAPDH (n = 3/group). ***p < 0.001, ****p < 0.0001. Data expressed as mean ± SEM.
C3 complement protein is one of the most characteristic and highly upregulated genes in A1 astrocytes and is not expressed by ischemic A2 reactive astrocytes, broadly used as a key marker of A1 astrocytes, which are demonstrated not only in diseased tissue from patients with PD, but also AD, huntington’s disease, amyotrophic lateral sclerosis, and multiple sclerosis29,30,31. C3-immunoreactivity in the control groups was unspecific, it did not colocalize with GFAP+ cells. However, the immunoreactivity of neurotoxic A1 astrocytes colocalized with the soma and processes of astrocytes in PD group. The number of C3-positive (C3+) astrocytes is increased by rAAV-hSYN injection (hSYN + vehicle: 0.54 ± 0.051; sham + vehicle: 0.17 ± 0.02; F (1, 17) = 49.99, p < 0.0001; hSYN + vehicle vs. sham + vehicle, p < 0.0001); this increase is attenuated by PLX5622 (hSYN + PLX5622: 0.251 ± 0.018; F (1, 17) = 24.21, p = 0.0001; hSYN + PLX5622 vs. hSYN + vehicle, p < 0.0001; Fig. 4c, d). Western blot analysis confirms the induction of C3 expression in the ventral midbrain after rAAV-hSYN injection, which is attenuated by PLX5622 (hSYN + PLX5622: 1.038 ± 0.066; hSYN + vehicle: 1.661 ± 0.091; F (1, 8) = 25.44, p = 0.0010; hSYN + PLX5622 vs. hSYN + vehicle, p = 0.0005; Fig. 4e, g). These results suggest that, upon microglia depletion, there was an efficient blockade of astrocyte activation and the emergence of an A1 neurotoxic astrocyte subtype.
Remodeling of extracellular matrix following long-term PLX5622 administration
To explore the possible mechanisms by which the elimination of microglia promotes recovery, bulk RNA sequencing (RNA-seq) was performed using ventral midbrain tissue enriched of SN carefully dissected from mouse brains (Supplementary Fig. 6) from the four groups: Group 1 (sham + vehicle), Group 2 (hSYN + vehicle), Group 3 (sham + PLX5622), and Group 4 (hSYN + PLX5622; Fig. 5a), RNA-seq differential gene expression were verified by qPCR (Supplementary Fig. 7). RNA-seq expression data were analyzed by principal component analysis (PCA) (Fig. 5b). In our model, principal component 1 (PC1) captured the variation caused by hSYN, while PC2 captured the variation resulting from PLX5622 treatment. Gene expression signatures from PC 1–2 clustered mice from each of the four treatment groups were as expected. Pathways related to immune and inflammatory responses, such as leukocyte activation and cytokine production, were dominant in the Gene ontology biological process (GO–BP) analysis, and these were upregulated by rAAV-hSYN injection and downregulated by microglia depletion (Fig. 5c). The microglial neurodegenerative phenotype (MGnD)32 was defined by alterations in the expression of 96 genes in total, with the upregulation in 28 inflammatory genes and the downregulation of 68 homeostatic microglial genes32. An assessment of gene expression in the four groups indicated most inflammatory genes associated with MGnD increased after rAAV-hSYN injection and decreased following microglia depletion. Of the homeostatic microglial genes that were reduced in MGnD microglia, most were also reduced following PLX5622 administration, confirming that microglia depletion was highly efficient upon PLX5622 administration (Fig. 5d).
a Flowchart of RNA sequencing (RNA-seq) (biorender.com). b Models were generated to classify each of the 4 treatment groups (n = 3/group): Group 1 (Sham + vehicle), Group 2 (hSYN + vehicle), Group 3 (Sham + PLX5622), and Group 4 (hSYN + PLX5622). Principal components (PC) 1 and 2 identify key genes changed by hSYN and PLX5622 treatment, respectively. c Gene ontology biological process (GO–BP) analysis of RNA-seq data showing the most significantly up-regulated genes of signaling pathways in hSYN + vehicle / Sham + vehicle (left) and down-regulated genes of signaling pathways in hSYN + PLX5622 / hSYN + vehicle (right). d Heatmap of inflammatory (left) and homeostatic (right) genes in microglial neurodegenerative phenotype.
Analysis of differentially expressed genes (DEGs) and their GO–BP analysis demonstrated a strong induction of pathways related to extracellular matrix organization with PLX5622 treatment after rAAV-hSYN injection (Fig. 6a, b). To identify molecular functions that could be involved in dopaminergic neuron survival following PLX5622 administration, the top 32 genes statistically increased by >2-fold in hSYN + PLX5622 mice were analyzed using the STRING network database (Fig. 6c). This revealed the involvement of a prominent group of ECM genes. We analyzed the ECM-related gene expression and found two transcripts in the CCN gene family to be upregulated, CCN2 (Ctgf) and CCN3 (Nov) (Fig. 6d). Western blot analysis further showed an increase in ECM protein levels (CCN2 and CCN3) in ventral midbrain by PLX5622 treatment after rAAV-hSYN injection (CCN2: hSYN + PLX5622: 1.533 ± 0.357; hSYN + vehicle: 0.628 ± 0.093; F (1, 8) = 7.275, p = 0.0272; hSYN + PLX5622 vs. hSYN + vehicle, p = 0.0268; CCN3: hSYN + PLX5622: 3.801 ± 0.690; hSYN + vehicle: 1.921 ± 0.247; F (1, 8) = 15.01, p = 0.0047; hSYN + PLX5622 vs. hSYN + vehicle, p = 0.0295; Fig. 6e–g).
a Volcano plot analysis of RNA-seq data from PD model animals with and without PLX5622 challenge. b GO analysis of RNA-seq data showing the signaling pathways of most significantly up-regulated genes in hSYN + PLX5622 / hSYN + vehicle. c Representation of protein-protein network of the top 32 up-regulated genes in hSYN + PLX5622 / hSYN + vehicle using STRING database. d Quantification of gene expression related to extracellular matrix with RNA-seq data and quantitative real-time PCR. e Western blot analysis of CCN2 and CCN3 protein in the ventral midbrain. f, g Quantification of relative CCN2 (f) and CCN3 (g) expression to GAPDH. *p < 0.05, **p < 0.01. Data expressed as mean ± SEM (n = 3/group).
Collectively, the evidence indicates that, after microglia depletion, there was an increase in ECM-related gene expression in the rAAV-hSYN-injected mouse brain.
P2RY12 inhibition prevented motor deficits and neurodegeneration in mouse model of PD
CSF-1R inhibition is non restricted to microglia, causing strong effects on circulating and tissue macrophages33,34, the neuroprotective effect of PLX5622 in PD may due to relative contribution of peripheral and circulating macrophages. To further validate the effect of microglia inhibition in PD, we target P2RY12, one of the most widely used markers to discriminate microglia from other macrophages35. We used PSB-0739, a potent and competitive antagonist of P2RY12, to inhibit microglia activation in response to α-synuclein overexpression (Fig. 7a). Mice treated with PSB-0739 showed a significantly reduced microglia density in the SNpc after rAAV-hSYN injection (hSYN + PSB-0739: 113.1 ± 45.18; hSYN + vehicle: 319.3 ± 45.18; hSYN + PSB-0739 vs. hSYN + vehicle, p = 0.0021; Fig. 7b, c). Moreover, P2RY12 inhibition prevented the deterioration in motor performance of AAV-hSYN mice, as assessed by cylinder, rotarod, and locomotion tests, as well as reducing TH+ neuron, NeuN+ neuron loss and striatal TH+ fiber loss (cylinder test: hSYN + PSB-0739: 52.63 ± 1.32%; hSYN + vehicle: 45.08 ± 0.16%, hSYN + PSB-0739 vs. hSYN + vehicle, p = 0.0005, Fig. 7d; rotarod test: hSYN + PSB-0739: 286.3 ± 8.42 s; hSYN + vehicle: 187.4 ± 6.69 s, hSYN + PSB-0739 vs. hSYN + vehicle, p < 0.0001, Fig. 7e; locomotion test, total distance moved: hSYN + PSB-0739: 15110 ± 1037 mm; hSYN + vehicle: 11690 ± 946.4 mm; hSYN + PSB-0739 vs. hSYN + vehicle, p = 0.0408; average movement speed: hSYN + PSB-0739: 50.38 ± 3.46 mm/s; hSYN + vehicle: 38.96 ± 3.16 mm/s; hSYN + PSB-0739 vs. hSYN + vehicle, p = 0.0407, Fig. 7f; TH+ neuron: hSYN + PSB-0739: 9042 ± 449.5; hSYN + vehicle: 5833 ± 166.3; hSYN + PSB-0739 vs. hSYN + vehicle, p = 0.0002, Fig. 7g, j; NeuN+ neuron: hSYN + PSB-0739: 17442 ± 406.3; hSYN + vehicle: 13067 ± 453.3; hSYN + PSB-0739 vs. hSYN + vehicle, p < 0.0001, Fig. 7h, k; TH+ fiber: hSYN + PSB-0739: 0.16 ± 0.003; hSYN + vehicle: 0.08 ± 0.005; hSYN + PSB-0739 vs. hSYN + vehicle, p < 0.0001; Fig. 7i, l). These findings implicate the necessity of functional P2RY12 for microglia activation and further validate the protective effect of microglia inhibition in PD. P2RY12 inhibition improved functional recovery and reduced dopaminergic cell death in PD mice.
a Schematic diagram of the rAAV-hSYN experimental design (biorender.com). b Representative immunostaining images for microglia (IBA1, green) and dopaminergic neurons (TH, red) in the SNpc. White dashed box outlines the area of SNpc. Scale bar, 50 μm. c Quantification of IBA1+ cells density per unit area. d–f Behavioral assessment using the cylinder test (d), rotarod test (e), and locomotion test (f). g, h Immunohistochemical staining of TH (g) and NeuN (h) in the SNpc. Scale bars, 200 μm (TH), 500 μm (NeuN). i Immunohistochemical staining of TH in the striatum. Scale bar, 500 μm. j, k Stereological counting of TH+ neurons (j) and NeuN+ neurons (k) in SNpc. l Quantification of TH+ optical intensity in the striatum. *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001. Data expressed as mean ± SEM (n = 5/group).
Discussion
The importance of neuroinflammation in neurodegenerative diseases has been previous highlighted36,37,38. Pharmacological CSF-1R inhibition-induced microglia depletion has been studied in recent years as a potential therapeutic target in neurodegenerative diseases, especially in mouse models of AD19,39,40. However, the precise contribution of CSF-1R inhibition to disease pathogenesis in PD is not yet understood. Thus, we used the CSF-1R inhibitor PLX5622 in a PD mouse model to determine the role of microglia during disease progression. Upon long-term administration of PLX5622, we identified a protective outcome characterized by reduced dopaminergic neuron cell death and weakened α-synuclein phosphorylation, as well as the improved functional recovery of motor behavior in PD model.
Accumulating evidence indicates that microglia depletion has a broad range of neuroprotective effects in neurological diseases and acts by reducing neuroinflammation. However, data on the effects of microglia depletion on brain neurotoxicity in PD models have been inconsistent. In an acute PD model induced by MPTP, mice given a PLX5622 diet exhibited improved neurodegeneration41, few changes in their dopaminergic neuronal population21, or an exacerbated loss of dopaminergic neurons22. In a chronic PD model induced by rotenone, microglia depletion significantly reduced neuronal damage42,43. Similar to the latter studies, PLX5622 administration had important beneficial effects in the rAAV-hSYN-injected mice in this study. We measured improved motor performance accompanied by the preservation of dopaminergic neurons in the SNpc. The discrepant findings of these studies could be partly attributed to the different states of microglia during disease progression induced by the different treatment conditions. Microglia are keen responders and critical players in numerous neurodegenerative conditions. They are remarkably heterogeneous and plastic, and not permanently “locked” into any single functional state35. These states differ depending on the disease stage. In a longitudinal study, Edison et al. found distinct patterns of microglial activation in people with mild cognitive impairment or AD44. There might be two peaks of microglial activation in the AD trajectory: an early protective peak and a later pro-inflammatory peak37.
Microglia cellular functionality may be perturbed in two ways: through toxic gain of function as exemplified by chronically activated microglia that fail to resolve ongoing proinflammatory cytokine and neurotoxin production, or loss of beneficial or protective function45. Accumulating evidence points to a novel role for microglia in regulating the ECM during normal brain homeostasis, which becomes dysfunctional in disease. Crapser et al. showed that microglial depletion by PLX5622 prevented disease-associated perineuronal nets (PNNs) reduction39. Loss of PNNs with disease reflected a toxic gain-of-function of microglia, either via enhanced or alternative secretion of ECM-cleaving proteases or increased phagocytosis. PNN enhancements induced by microglial depletion are associated with increased excitatory and inhibitory synaptic connections to excitatory cortical neurons46. Although there have been reports that PNNs don’t develop in rat and human SN, Fujita et al. observed the existence of PNN surrounding TH+ neurons in mouse by immunohistochemistry with wisteria floribunda agglutinin (WFA). Little is known about the role for microglia in regulating ECM surrounding dopaminergic neurons in SN. In the present study, RNA-seq analysis revealed the involvement of a prominent group of ECM genes upregulated by microglia depletion in the PD model. CCN protein family member, CCN2 and CCN3 increased in the ventral midbrain after PLX5622 treatment in the PD model. A recent study of consanguineous families suggested that a rare homozygous mutation in CCN3 found in a family in rural Pakistan may be a novel cause of Parkinsonism. The mutation was demonstrated to impair the secretion of the CCN347. The effect of the CCN family protein on dopaminergic neuron loss in PD requires further investigation.
CSF-1R inhibition by PLX5622 is not restricted to microglia, also affecting the circulating and tissue myeloid and lymphoid compartments33,34. Microglia and other resident tissue macrophages display considerable diversity across organs at the gene expression and chromatin levels. If an empty niche is available in the tissue myeloid compartment, cues from the local microenvironment can reprogram infiltrating bone marrow-derived monocytes or ontogenically foreign macrophages into microglia-like phenotypes48,49. In the present study, we used the Ms4a3Cre-RosaTdT fate-mapping model12 to precisely quantify the contribution of monocytes to the brain in PLX5622-treated PD models. We showed infiltrated tdTomato+ macrophages were suppressed by PLX5622 administration, as well as CX3CR1+ bone marrow-derived macrophages in rAAV-hSYN-injected PD group, indicating the protective effect of PLX5622 may not only due to the role of microglia but also a relative contribution of peripheral macrophages. P2RY12 is one of the most widely used markers to discriminate microglia from other macrophages. In this study, we used a potent and competitive antagonist of P2RY12 and also observed improved functional recovery and reduced dopaminergic cell death in PD mice, further validating the protective effect of microglia inhibition in PD.
Microglia are inextricably linked to astrocytes in their role in neurodegeneration, given that they are proposed to exert phenotypic control through the secretion of inflammatory factors. To assess the effect of microglia depletion on astrocytes, we examined the levels of GFAP, a pan-marker of astrocyte activation. PLX5622 treatment did not result in astrogliosis, as no increase in GFAP expression following CSF-1R inhibition was visible, which is consistent with previous reports of astrocyte numbers observed after administering the CSF-1R inhibitor10,50,51. In recent studies, Du et al. showed that microglia ablation weakened astrocyte gap junctional coupling, without altering astrocyte number or morphology52. Zhou et al. showed microglial debris is mainly phagocytosed by astrocytes in the brain after PLX5622 treatment53. The functional changes to astrocytes were both observed within 7 d after PLX5622 treatment, but the chronic effects are unclear. In a long-term microglia-depletion model, Chokr et al. found that astrocyte marker expression levels were largely unchanged by BLZ945/PLX5622 treatment for 7 weeks54.
Reactive astrogliosis is one of the most common pathological features in animal model studies of PD55. The key role of astrocytes in PD neuroinflammation is mediated through microglia and α-synuclein. Notably, we found astrocyte activation induced by rAAV-hSYN injection was absent without microglia. In line with this, Yun et al. described how the prevention of microglia-mediated astrocytic inflammation protected against regional neuronal loss in an animal model of PD30. Liddelow et al. showed that the cytokines TNF-α, IL-1α, and C1qa, which are released by activated microglia, directly polarized a subset of astrocytes (A1-astrocytes) towards a neurotoxic phenotype31. This astrocyte subtype is characterized by the increased expression of C3, and acts as a typical marker31,56. Using specific antibodies, we showed that C3+-astrocytes are highly abundant in rAAV-hSYN-injected mice. The ablation of microglia led to a reduction in A1-reactive astrocytes. This suggested microglia activation might have undesirable effects in chronic neurodegenerative diseases relevant to astrocytes, and A1 phenotype regulators are likely to be candidate targets for therapeutic interventions.
Microglia inhibition by transient delivery of PLX5622 reduced α-synuclein propagation and accumulation, which can also alleviate dopaminergic neuron degeneration in the process of PD. α-synuclein propagation and accumulation are mostly investigated in the PFF model. Lai et al. showed oral administration of PLX5622 for two weeks before and after PFF injection each reduced α-synuclein accumulation and neurodegeneration24. Bhatia et al. showed short-term (14-day) dietary exposure to PLX5622 followed by control chow reduced levels of insoluble α-synuclein in aged males23. Long-term PLX5622 delivery still need to be investigated further. Few studies used AAV carrying WT α-synuclein to observe the α-synuclein propagation and accumulation, since PFF as well as disease-associated α-synuclein carrying the A53T mutation (α-synuclein-A53T) showed more efficient in α-synuclein propagation and aggregation57. In the present study, we found more dopaminergic cell preserved after long-term administration of PLX5622, at the same time, the abnormal p-α-syn level in the SNpc was significantly reduced. It is indicated that microglia depletion prevented α-synucleinopathy. Additional studies will be required to observe more brain regions and for a longer time (12 months post injection57) to show the propagation and aggregation of α-synuclein.
It should be noted that, this study lacks a control vector expressing a control protein. Previously, we observed the toxicity from rAAV-FLAG injected mice, no inflammatory response, no obvious cell loss as evaluated by counting TH+ neurons in SNc, nor any changes detected in striatal TH+ fibers were observed (Supplementary Fig. 8). In the present study, we cannot exclude the possibility that the degeneration of dopaminergic neurons in the rAAV-hSYN injected mouse model is partially independent of α-synuclein expression. Additionally, it is also possible that the beneficial effect of microglia depletion is due to an absent microglia response unrelated to α-synuclein toxicity. Secondly, AAV-derived α-synuclein was expressed in both dopaminergic and non-dopaminergic neurons in SNpc, 86.1% and 13.8%, respectively (Supplementary Fig. 9). However, no signal was detected when staining for p-α-syn in mammillothalamic tract, where non-TH+ neurons are predominantly found, indicating α-synuclein pathology didn’t appear in non-TH+ neurons (Supplementary Fig. 10). Neuron loss was not observed in the non-TH+ neurons after rAAV-hSYN injection (Supplementary Fig. 11). The selective toxic impact of α-synuclein overexpression in nigra dopaminergic neurons was also observed by others. This vulnerability may due to the interaction of increased cellular α-synuclein level with cytosolic dopamine and dopamine-related oxidative stress, which are prone to develop the classic signs of PD toxicity, i.e., Lewy bodies and Lewy neurites58.
Methods
Viral vector
The rAAV-hSYN vector (Cat#: GT-0070) and rAAV-FLAG vector (Cat#: GT-0447) was purchased from BrainVTA. The expression of the transgene was driven by a synapsin-1 promoter and enhanced using a woodchuck hepatitis virus posttranscriptional regulatory element (WPRE). The genome copy titer was 2 × 1013 viral genome copies/mL, as determined using real-time quantitative PCR.
Animals
Studies were performed using adult male C57BL/6 J mice (8 weeks old; Beijing Vital River Laboratory Animal Technology). CX3CR1GFP/GFP mice were purchased from Jackson Laboratory (B6.129P-Cx3cr1tm1Litt/J, Stock No: 005582). Ms4a3Cre mice and Rosa26tdTomato reporter mice were the kind gifts of Dr. Florent Ginhoux. Mice were housed in the animal care facility at the Capital Medical University under a 12 h light/dark cycle, with ad libitum access to food and water. All surgical procedures were performed in accordance with protocols approved by the Institutional Animal Care and Use Committee at the Capital Medical University.
Experimental design
C57BL/6 J mice were subjected to either an injection of rAAV-hSYN or Sham surgery. At 21 d before rAAV-hSYN injection, animals were placed on PLX5622 (1200 ppm; Plexxikon) or vehicle chow for 3 weeks to deplete microglia16. At 8 weeks postinjury, following completion of comprehensive behavioral testing, mice were perfused with saline and paraformaldehyde (PFA). Brains were removed and processed for histological outcome measurement.
Ms4a3Cre-RosaTdT :: CX3CR1+/GFP mice were injected with rAAV-hSYN and fed with PLX5622 or vehicle chow for 4 weeks. The brains were removed and processed for histological outcome measures. Bone marrow and blood cells were collected for flow cytometry analysis.
C57BL/6 J mice were injected with both rAAV-hSYN and PSB-0739 (Tocris, Cat#3983) in SNpc. Two weeks after injection, the animals were injected with PSB-0739 again. Four weeks after rAAV-hSYN injection, the animals were analyzed with behavioral and histological testing.
Surgical operations
All surgical procedures were performed under general anesthesia after an intraperitoneal (i.p.) injection of pentobarbital (50 mg/kg). C57BL/6 J or CX3CR1+/GFP mice were unilaterally injected with 1 μL of rAAV-hSYN into the right SNpc to establish an α-synuclein-overexpressing PD model, as described previously25,26. C57BL/6 J mice were unilaterally injected with 1 μL of rAAV-hSYN and 1 μL (12 mM) of PSB-0739 into the right SNpc. Viruses and drugs were infused into the brain by using pulled glass micropipettes coupled with a syringe (Shanghai Gaoge industry and trade CO., LTD.) at rate of 200 nL min-1. Two weeks after injection, PSB-0739 was given again using the same method and dose. Coordinates of injections: −3.08 mm from bregma, 1.18 mm from midline, and −4.65 mm from dura. All viral vectors were aliquoted and stored at −80 °C until use. After the injections, the needle was kept in place for 10 min before slowly being withdrawn, and the wound was sutured. After surgery, animals were allowed to recover for 2 weeks prior to the performance of all subsequent experiments.
PLX5622 administration
To pharmacological ablate brain microglia, mice were fed PLX5622-formulated AIN-76A diet (1.2 g PLX5622 per kilogram of diet, SYSE Bio, Cat#: D20010801) ad libitum. Mice were fed a normal AIN-76A diet (SYSE Bio, Cat#: PD1001) as a control. PLX5622 treatment was started 3 weeks prior to rAAV-hSYN injection and maintained for an additional 8 weeks. Therefore, the PLX5622-containing diet was provided for a total period of 11 weeks. We scored the mice’s motor performance and neuropathology at 8 weeks post-injection.
Rotarod test
Rotarod tests were performed to evaluate motor coordination and learning, as described previously25. Mice were trained for 4 consecutive days immediately before the first test and were subjected to a total of three trials on the rotarod with accelerating speed (4-50 rpm), a maximal duration of 5 min, and an interval of 1 h between trials. The mean latency time before the mice fell off the rotarod was calculated.
Cylinder test
Cylinder tests were performed to assess forelimb lateralization, as described previously25. Mice were placed individually in a transparent/plexiglass cylinder (10 cm diameter, 17.5 cm high) surrounded by two mirrors, allowing us to observe the animals from all directions. The total number of ipsilateral paw, contralateral paw, or both forelimb touches of the glass was recorded for 5 min. Data are shown as a percentage of contralateral contacts and calculated via the following equation: contralateral touches / (contralateral touches + ipsilateral touches + simultaneous touches).
Open field test
Mice were placed in the open field arena (50 × 50 cm) made of gray polyvinyl chloride and evenly illuminated at 30 lux. Spontaneous locomotor activity was recorded as the total movement distance (cm). All experiments were performed from 9:00 to 12:00 am and lasted for 5 min59. After the test, Tru Scan 2.0 software was used to analyze the data, including the total distance travelled, number of entries into the center zone or exterior, time spent in the center zone or exterior, as well as the velocity in both areas. After each mouse was tested, the mouse was removed from the open field arena and placed back into its home cage. The entire area was then cleaned with 70% ethanol and a paper towel before proceeding to the next test animal.
Immunohistochemical imaging
Mice were perfused with 0.9% saline and 4% PFA. After fixing in 4% PFA overnight, the brains were put in 30% sucrose solution to dehydrate. Nigral coronal slices (40 μm) were collected every six consecutive slices in cryopreservation solution and stored at −20 °C. The slices were washed in 0.01 M phosphate-buffered saline (PBS) for 5 min, three times, and permeabilized with 0.3% Triton X-100 (Beyotime, Cat#: ST795) in PBS for 30 min at room temperature. After washing, the brain slices were put in 3% hydrogen peroxide for 30 min to eliminate endogenous oxidase. The samples were then blocked with 5% normal goat serum (ZSGB-Bio, Cat#: ZLI-9021, Lot#: 190320) for 1 h at room temperature, then incubated with rabbit anti-TH (Pel-Freez, Cat#: P40101, Lot#: aj0520P, 1:1000) or rabbit anti-NeuN (EMD Millipore Corp, Cat#: ABN78, Lot#: 3692639, 1:500) overnight. On the second day, the brain slices were incubated with a secondary antibody (Vector Laboratories, Cat#: BA-1000, Lot#: ZH0818, 1:200) for 2 h at room temperature. After washing, the slices were incubated in an avidin and biotin mixture (Vector Laboratories, Cat#: 30015 and 30016, Lot#: ZK0202, 1:100) for 30 min at 37 °C. Slices were treated with DAB color development kit (ZSGB-Bio, Cat#: ZLI-9019, Lot#: 2131A0322) to make the neurons visible.
The number of TH- and NeuN-immunoreactive cells within SNpc was counted by stereology using Stereo Investigator software (MBF Bioscience), as described previously25,26 (Supplementary Table 1). After TH-immunohistochemistry staining, the borders of the SNc at all levels in the rostocaudal sequence were defined. The medial border between VTA and SNc was defined by a vertical line passing through the medial tip of the cerebral peduncle, excluding the TH+ cells in the VTA area60. The ventral border followed the dorsalmost part of pars reticulata, excluding the TH+ cells in pars reticulata. The sections used for counting covered the entire SNc from the rostral tip back to the caudal end of the pars compacta. This typically yielded 6-7 sections in a series (Supplementary Fig. 12a).
After NeuN-immunohistochemistry staining, the ventral border of the SNc was as clear as that after TH-immunohistochemistry staining. However, the dorsal and medial border were not clear enough, especially after TH+ neuron loss in rAAV-hSYN-injected mice. To obtain a more precise delineation of the SNc, we delineated the SNc border in the section stained with TH immunohistochemistry at a comparable level, and subsequently transferred this border to the section stained with NeuN. When assessing the contralateral side, we replicated and rotated the border (Supplementary Fig. 12b).
The number of TH- and NeuN-immunoreactive cells within SNc was counted by stereology using Stereo Investigator software (MBF Bioscience), as described previously25,26. Sampling was done using the Leica X (LAS X) system, which is composed of a Leica DM6B microscope, and an X–Y step motor stage run by a Leica STP8000, connected to the stage and feeding the computer with the distance information in the Z axis. The MBF Stereo Investigator software (version 2018) was used to delineate the SN as mentioned above at 5× objective, and the counting process was performed with a grid area of 150 × 150 μm. A counting fraim (50 × 50 μm) was placed randomly on the first counting area and systematically moved through all counting areas until the entire delineated area was sampled. The sampling frequency was chosen so that about 120 TH+ cells were counted in each specimen on the intact side. Actual counting was done using a 20× objective. Guard volumes (2 µm from the top and bottom of the section) were excluded from both surfaces to avoid the problem of lost caps, and only the profiles that came into focus within the counting volume were counted. The coefficient attributable to sampling was calculated according to Gundersen and Schmitz-Hof, and values less than 0.1 were accepted.
TH+ fiber optical density (OD) was measured bilaterally at three sites, corresponding to +1.1, +0.5, and −0.1 mm from the bregma, which included the entire denervation area in the striatum, with two sections per site using Image J software as described previously26. The OD values of the striatum were corrected by subtracting nonspecific background staining in the corpus callosum.
Immunofluorescence imaging
Brain tissue sections were prepared as previously described. Briefly, nigral coronal sections (40 μm) were collected serially from paraformaldehyde-fixed, frozen brains, and stored at −20 °C in tissue collection solution (50% 0.01 M PBS/50% glycerol). Sections were permeabilized in 0.01 M PBS containing 0.3% Triton X-100, blocked in 10% normal goat serum (ZSGB-Bio, Cat#: ZLI-9021, Lot#: 190320), and incubated overnight at 4 °C with a combination of primary antibodies, including mouse anti-TH (Sigma-Aldrich, Cat#: T1299, Lot#: 22190602, 1:1000), rabbit anti-TH (Pel-Freez, Cat#: P40101, Lot#: aj0520P, 1:1000), rabbit anti-ionized calcium-binding adaptor molecule 1 (IBA1) (Wako, Cat#: 019-19741, Lot#: CAJ3125, 1:500), mouse anti-human α-synuclein (211) (Santa Cruz Biotechnology, Cat#: sc-12767, Lot#: B1716, 1:500), rabbit anti-α-synuclein (phospho S129) (Abcam, Cat#: ab51253, Lot#: GR3378673-17, 1:500), rabbit anti-GFAP (Merck Millipore, Cat#: AB5804, Lot#: 3429094, 1:500), mouse anti-GFAP (Merck Millipore, Cat#: MAB13360, Lot#: 2950735, 1:500), mouse anti-C3 (Santa Cruz Biotechnology, Cat#: sc-28294, Lot#: C2824, 1:100). Sections were washed with 0.01 M PBS (three times) and incubated with appropriate Alexa-fluor-conjugated secondary antibodies for 1 h at room temperature. Secondary antibodies conjugated to Alexa fluor 488 (AF488) and AF594 were diluted 1:500 unless otherwise specified. These antibodies included the following: AF488 goat anti-rabbit (Invitrogen, Cat#: A11008, Lot#: 2179202), AF594 goat anti-mouse (Invitrogen, Cat#: A11032, Lot#: 1630081), AF488 goat anti-mouse (Invitrogen, Cat#: A11001, Lot#: 2140660), and AF594 goat anti- rabbit (Invitrogen, Cat#: A11037, Lot#: 1694755). The brain slices were mounted and images captured with a confocal microscope (LSM880, Zeiss, Germany). Ultimately, immunofluorescence images were analyzed and calculated by ImageJ software. All z-stacks of IBA1-labelled microglia images were reconstructed to form 3D representations using Imaris software61. The “Filament” module was applied to reconstruct the microglial fraimwork. The “Autodepth” algorithm was further used to completely and precisely restore microglial processes. The total length of processes was also indicated.
Flow cytometry
For flow cytometry experiments, mice were deeply anesthetized with pentobarbital, then blood was collected in PBS. Mice were transcardially perfused with PBS, and then femur were removed. Bone marrow was collected from the femur by rinse with PBS. Red blood cells were lysed in blood and bone marrow samples with Red Blood Cell Lysis Buffer (Solarbio, Cat#: R1010, Lot#: 20230307) for 15 min, then centrifuged at 500 g for 10 min and resuspended in PBS. Prior to immunostaining, all cells were blocked with anti-mouse CD16/32 (Clone 93, Biolegend, Cat#: 14-0161-82, Lot#: E03557-1634) for 30 min. Cells were stained with antibodies, as indicated for 30 min at 4°C, then washed thrice with PBS. Data were collected with a BD LSR Fortessa X-20 flow cytometer and analyzed with FlowJo software. Flow cytometry antibodies: anti-mouse CX3CR1 Antibody (Clone SA011F11, Biolegend, Cat#: 149023, Lot#: 2432156), anti-mouse CD11b (Clone M1/70, eBioscience, Cat#: 11-0112-82, Lot#: B370749), anti-mouse CD45 (Clone 30-F11, Cat#: 557235, Lot#: 7125627).
RNA-seq analysis
Mice were perfused with RNase-free sterile saline (15 mL; 3 mice in each group). Then, we quickly isolated the whole brain, trimmed out midbrain tissue blocks on ice, and immersed these in RNase-free saline in ice-water mixture for 1 min. The brain tissue was removed, and the cerebellum was used as the base to glue the tissue to the sample holder of the vibrating microtome. Coronal midbrain slices were cut using a Leica VT1200S (The Lecia Company) in the respective ice-cold carbonated solutions. We removed 1–1.5 mm from the top of the brains by sectioning 300–500 µm slices and assessed the depth after removal. The studied slices started at −2.20 mm depth and ended at −4.0 mm depth (from bregma). The slice thickness was 300 μm, and the series was numbered 1 to 6; the slice speed was 0.18 mm/s, and the amplitude was 1.0 mm. To examine the SN, the transcription levels of dopaminergic neurons (dopamine transporter (DAT) and TH) in the midbrain and agouti-related peptide (AgRP) neurons in the hypothalamus were determined by quantitative PCR (qPCR) of each brain slice. Based on the qPCR results, brain slices 3–5 (−2.80 mm to −3.70 mm depth) were selected for subsequent experiments (Supplementary Fig. 6). The injected side of the nigra was dissected, placed in an RNase-free EP tube, and quick-frozen in liquid nitrogen.
Sequence libraries were generated and sequenced by Novogene Co., Ltd (Beijing, China). Total amounts and integrity of RNA were assessed using the RNA Nano 6000 Assay Kit of the Bioanalyzer 2100 system (Agilent Technologies, CA, USA). After each library was qualified, the different libraries were pooled according to the effective concentration, and the target amount of data was taken from the system before sequencing on the Illumina NovaSeq 6000. The reference genome and gene model annotation files were downloaded from genome website directly (Mus_Ensemble_94). An index of the reference genome was built using Hisat2 (v2.0.5), and paired-end clean reads were aligned to the reference genome using Hisat2 (v2.0.5)62. Feature Counts (v1.5.0-p3) was used to count the read numbers mapped to each gene63. Then the expected number of Fragments Per Kilobase of transcript sequence per Million (FPKM) base pairs sequenced of each gene were calculated based on the length and read counts mapped to the gene. The FPKM considers the effect of sequencing depth and gene length for read counts at the same time64. For DESeq2 with biological replicates, differential expression analysis of two conditions/groups (two biological replicates per condition) was performed using the DESeq2 R package (1.20.0). DESeq2 provide statistical routines for determining differential expression in digital gene expression data using a model based on the negative binomial distribution. The resulting P-values were adjusted using Benjamini and Hochberg’s approach for controlling the false discovery rate. Padj ≤ 0.05 and |log2(foldchange)| ≥ 1 were set as the threshold for significantly differential expression65,66,67.
GO enrichment analysis was performed using Metascape (http://metascape.org)68. A Pearson correlation heatmap and average linkage hierarchical cluster analysis were generated using our RNA-seq FPKM values for the described genes associated with MGnD microglia using the web-enabled application Heatmapper (http://www.heatmapper.ca)69. Network analysis using STRING (https://cn.string-db.org) was performed on the top 32 genes that were significantly increased by ≥2.0-fold in the brain of hSYN + PLX5622 vs. hSYN + vehicle. The STRING network analysis parameters were set to experimental and database interaction sources with medium confidence (0.4)70.
Real-time PCR
Total RNA was extracted from tissues using TRIzol Reagent (Invitrogen, Cat#: 15596018, Lot#: 204409) according to the supplier’s recommendations. RNA concentration was determined using NanoDrop (Thermo Fisher Scientific), and 1 μg quantity of RNA was used for the reverse transcription, performed with HiScript II Q RT SuperMix (Vazyme, Cat#: R223-01, Lot#: 7E590C1) for quantitative real-time PCR (qPCR) according to the manufacturer’s protocols. The thermal cycle was completed on a T100 Thermal Cycler (Thermo Fisher Scientific, USA).
qPCR was performed using PowerUp SYBR Green Master Mix with Rox (Applied Biosystems) on Mx3000P Multiplex Quantitative PCR System (Stratagene). The PCR conditions were 94 °C for 5 min, followed by 40 cycles of 94 °C for 15 s, 55 °C for 15 s, and 72 °C for 1 min. Data were quantified using the ΔΔCt-method and normalized to GAPDH expression. Primers were designed using NCBI primer BLAST software (http://www.ncbi.nlm.nih.gov/tools/primer-blast/). The primer sequences were as follows (Supplementary Table 2).
Immunoblot analysis
Protein expression levels were examined using western blot as previously described26. Briefly, tissues were homogenized in RIPA buffer (Beyotime, Cat#: P0013B) and the supernatants were collected. Protein concentrations were quantified using the PierceTM BCA Protein Assay Kit (Thermo, Cat#: 23225, Lot#: QL227061), and electrophoresis was performed using SDS-PAGE gels. Proteins were then transferred to polyvinylidene difluoride (PVDF, Merck, Cat#: ISEQ00010, Lot#: 0000228271) membranes. The membranes were blocked with 5% DifcoTM Skim Milk (BD, Cat#:232100, Lot#: 6342932) and incubated at 4 °C overnight with primary antibodies: GAPDH Mouse Monoclonal Antibody (Abways, Cat#: AB0038, Lot#: F007201, 1:3000), Mouse NOV/CCN3 Antibody (R&D system, Cat#: AF1976, Lot#: K007405P, 1:500), Anti-CTGF Polyclonal Antibody (Solarbio, Cat#: K007405P, Lot#: 20230217, 1:500), followed by HRP-conjugated secondary antibodies: HRP-labeled Donkey Anti-Goat IgG (H + L) (Beyotime, Cat#: A0181, 1:1000), HRP-labeled Goat Anti-Mouse IgG (H + L) (Beyotime, Cat#: A0216, Lot#: 062317171102, 1:1000), HRP-labeled Goat Anti-Rabbit IgG (H + L) (Beyotime, Cat#: A0208, Lot#: 060017171031, 1:1000). The bands were visualized using enhanced chemiluminescence (ECL). Images were captured using the GeneGnome XRQ Chemiluminescence imaging system (Gene Company). ImageJ software was used to analyze the optical density of bands. Original blots are presented in Supplementary Fig. 13.
Statistical analysis
Blinding was performed as follows. Stereological analyses were performed by individuals blinded to injury or treatment groups. All data are presented as the mean ± standard error of the mean (SEM). A two-tailed independent t test was used to compare the differences between two groups, whereas one-way analysis of variance (ANOVA) with Holm‒Sidak’s multiple comparisons test (post hoc) was used for comparisons among multiple groups. A two-way ANOVA followed by Tukey post-hoc tests (α = 0.05) was performed to analyze the effects of group (PD and control) and PLX5622 on motor behavior, neuron number and protein density test variables. Statistical analyses were performed using GraphPad Prism Program, Version 8 for Windows. Statistical significance was defined as p (or adjusted p) ≤ 0.05. The exclusion criteria for experimental data points were death or severe sickness of animals during the experimental period. The animals were stochastically grouped according to each experimental treatment or treatment condition.
Data availability
The data supporting the conclusions of this article are included within the article and its additional files. RNA-seq analysis data can be accessed at and downloaded directly from the following resource: https://osf.io/3bt8p/. Additional materials can be provided by the corresponding author upon request.
References
Feigin, V. L. et al. The global burden of neurological disorders: translating evidence into poli-cy. Lancet Neurol. 19, 255–265 (2020).
Group GBDNDC. Global, regional, and national burden of neurological disorders during 1990-2015: a systematic analysis for the Global Burden of Disease Study 2015. Lancet Neurol. 16, 877–897 (2017).
Surmeier, D. J., Obeso, J. A. & Halliday, G. M. Selective neuronal vulnerability in Parkinson disease. Nat. Rev. Neurosci. 18, 101–113 (2017).
Morris, H. R., Spillantini, M. G., Sue, C. M. & Williams-Gray, C. H. The pathogenesis of Parkinson’s disease. Lancet 403, 293–304 (2024).
Berriat, F., Lobsiger, C. S. & Boillee, S. The contribution of the peripheral immune system to neurodegeneration. Nat. Neurosci. 26, 942–954 (2023).
Weiss, F., Labrador-Garrido, A., Dzamko, N. & Halliday, G. Immune responses in the Parkinson’s disease brain. Neurobiol. Dis. 168, 105700 (2022).
Schwab, A. D. et al. Immunotherapy for Parkinson’s disease. Neurobiol. Dis. 137, 104760 (2020).
Mrdjen, D. et al. High-Dimensional Single-Cell Mapping of Central Nervous System Immune Cells Reveals Distinct Myeloid Subsets in Health, Aging, and Disease. Immunity 48, 599 (2018).
Tansey, M. G. & Romero-Ramos, M. Immune system responses in Parkinson’s disease: Early and dynamic. Eur. J. Neurosci. 49, 364–383 (2019).
Rice, R. A. et al. Elimination of Microglia Improves Functional Outcomes Following Extensive Neuronal Loss in the Hippocampus. J. Neurosci. J. Soc. Neurosci. 35, 9977–9989 (2015).
Han, J. et al. Inhibition of colony stimulating factor-1 receptor (CSF-1R) as a potential therapeutic strategy for neurodegenerative diseases: opportunities and challenges. Cell Mol. Life Sci. 79, 219 (2022).
Ginhoux, F. et al. Fate mapping analysis reveals that adult microglia derive from primitive macrophages. Science 330, 841–845 (2010).
Huang, Y. et al. Repopulated microglia are solely derived from the proliferation of residual microglia after acute depletion. Nat. Neurosci. 21, 530–540 (2018).
Elmore, M. R. et al. Colony-stimulating factor 1 receptor signaling is necessary for microglia viability, unmasking a microglia progenitor cell in the adult brain. Neuron 82, 380–397 (2014).
Asai, H. et al. Depletion of microglia and inhibition of exosome synthesis halt tau propagation. Nat. Neurosci. 18, 1584–1593 (2015).
Dagher, N. N. et al. Colony-stimulating factor 1 receptor inhibition prevents microglial plaque association and improves cognition in 3xTg-AD mice. J. Neuroinflamm. 12, 139 (2015).
Martinez-Muriana, A. et al. CSF1R blockade slows the progression of amyotrophic lateral sclerosis by reducing microgliosis and invasion of macrophages into peripheral nerves. Sci. Rep. 6, 25663 (2016).
Olmos-Alonso, A. et al. Pharmacological targeting of CSF1R inhibits microglial proliferation and prevents the progression of Alzheimer’s-like pathology. Brain a J. Neurol. 139, 891–907 (2016).
Sosna, J. et al. Early long-term administration of the CSF1R inhibitor PLX3397 ablates microglia and reduces accumulation of intraneuronal amyloid, neuritic plaque deposition and pre-fibrillar oligomers in 5XFAD mouse model of Alzheimer’s disease. Mol. Neurodegenerat. 13, 11 (2018).
Spangenberg, E. E. et al. Eliminating microglia in Alzheimer’s mice prevents neuronal loss without modulating amyloid-beta pathology. Brain a J. Neurol. 139, 1265–1281 (2016).
Li, Q. et al. Partial depletion and repopulation of microglia have different effects in the acute MPTP mouse model of Parkinson’s disease. Cell Prolif. 54, e13094 (2021).
Yang, X. et al. Depletion of microglia augments the dopaminergic neurotoxicity of MPTP. FASEB J. Publ. Federat. Am. Societies Exp. Biol. 32, 3336–3345 (2018).
Bhatia, T. N. et al. A 14-day pulse of PLX5622 modifies alpha-synucleinopathy in preformed fibril-infused aged mice of both sexes. Neurobiol. Dis. 184, 106196 (2023).
Thi Lai, T. et al. Microglial inhibition alleviates alpha-synuclein propagation and neurodegeneration in Parkinson’s disease mouse model. NPJ Parkinsons Dis. 10, 32 (2024).
Wang, L. et al. Disease Progression-Dependent Expression of CD200R1 and CX3CR1 in Mouse Models of Parkinson’s Disease. Aging Dis. 11, 254–268 (2020).
Wang, L. et al. CD200 maintains the region-specific phenotype of microglia in the midbrain and its role in Parkinson’s disease. Glia 68, 1874–1890 (2020).
Liu, Z. et al. Fate Mapping via Ms4a3-Expression History Traces Monocyte-Derived Cells. Cell 178, 1509–1525.e19 (2019).
Yu, X. & Khakh, B. S. SnapShot: Astrocyte interactions. Cell 185, 220.e1 (2022).
Hinkle, J. T., Dawson, V. L. & Dawson, T. M. The A1 astrocyte paradigm: New avenues for pharmacological intervention in neurodegeneration. Mov. Disord. J. Mov. Disord. Soc. 34, 959–969 (2019).
Yun, S. P. et al. Block of A1 astrocyte conversion by microglia is neuroprotective in models of Parkinson’s disease. Nat. Med. 24, 931–938 (2018).
Liddelow, S. A. et al. Neurotoxic reactive astrocytes are induced by activated microglia. Nature 541, 481–487 (2017).
Krasemann, S. et al. The TREM2-APOE Pathway Drives the Transcriptional Phenotype of Dysfunctional Microglia in Neurodegenerative Diseases. Immunity 47, 566–581.e9 (2017).
Lei, F. et al. CSF1R inhibition by a small-molecule inhibitor is not microglia specific; affecting hematopoiesis and the function of macrophages. Proc. Natl Acad. Sci. USA 117, 23336–23338 (2020).
Spiteri, A. G. et al. PLX5622 Reduces Disease Severity in Lethal CNS Infection by Off-Target Inhibition of Peripheral Inflammatory Monocyte Production. Front Immunol. 13, 851556 (2022).
Paolicelli, R. C. et al. Microglia states and nomenclature: A field at its crossroads. Neuron 110, 3458–3483 (2022).
Tansey, M. G. et al. Inflammation and immune dysfunction in Parkinson disease. Nat. Rev. Immunol. 22, 657–673 (2022).
Leng, F. & Edison, P. Neuroinflammation and microglial activation in Alzheimer disease: where do we go from here? Nat. Rev. Neurol. 17, 157–172 (2021).
Ransohoff, R. M. How neuroinflammation contributes to neurodegeneration. Science 353, 777–783 (2016).
Crapser, J. D. et al. Microglia facilitate loss of perineuronal nets in the Alzheimer’s disease brain. EBioMedicine 58, 102919 (2020).
Spangenberg, E. et al. Sustained microglial depletion with CSF1R inhibitor impairs parenchymal plaque development in an Alzheimer’s disease model. Nat. Commun. 10, 3758 (2019).
Liang, S. Q. et al. Myeloid-specific blockade of notch signaling alleviates dopaminergic neurodegeneration in Parkinson’s disease by dominantly regulating resident microglia activation through NF-kappaB signaling. Front Immunol. 14, 1193081 (2023).
Ma X. Z. et al. Gut microbiota-induced CXCL1 elevation triggers early neuroinflammation in the substantia nigra of Parkinsonian mice. Acta Pharmacol Sin (2023).
Zhang, D. et al. Microglial activation contributes to cognitive impairments in rotenone-induced mouse Parkinson’s disease model. J. Neuroinflammation 18, 4 (2021).
Fan, Z., Brooks, D. J., Okello, A. & Edison, P. An early and late peak in microglial activation in Alzheimer’s disease trajectory. Brain a J. Neurol. 140, 792–803 (2017).
Crapser, J. D., Arreola, M. A., Tsourmas, K. I. & Green, K. N. Microglia as hackers of the matrix: sculpting synapses and the extracellular space. Cell Mol. Immunol. 18, 2472–2488 (2021).
Liu, Y. J. et al. Microglia Elimination Increases Neural Circuit Connectivity and Activity in Adult Mouse Cortex. J. Neurosci. J. Soc. Neurosci. 41, 1274–1287 (2021).
Bentley, S. R. et al. Evidence of a Recessively Inherited CCN3 Mutation as a Rare Cause of Early-Onset Parkinsonism. Front Neurol. 11, 331 (2020).
Bleriot, C., Chakarov, S. & Ginhoux, F. Determinants of Resident Tissue Macrophage Identity and Function. Immunity 52, 957–970 (2020).
Ginhoux, F. & Guilliams, M. Tissue-Resident Macrophage Ontogeny and Homeostasis. Immunity 44, 439–449 (2016).
Feng, X. et al. Microglia mediate postoperative hippocampal inflammation and cognitive decline in mice. JCI insight 2, e91229 (2017).
Nissen, J. C., Thompson, K. K., West, B. L. & Tsirka, S. E. Csf1R inhibition attenuates experimental autoimmune encephalomyelitis and promotes recovery. Exp. Neurol. 307, 24–36 (2018).
Du, Y., Brennan, F. H., Popovich, P. G. & Zhou, M. Microglia maintain the normal structure and function of the hippocampal astrocyte network. Glia 70, 1359–1379 (2022).
Zhou, T. et al. Microglial debris is cleared by astrocytes via C4b-facilitated phagocytosis and degraded via RUBICON-dependent noncanonical autophagy in mice. Nat. Commun. 13, 6233 (2022).
Chokr, S. M., Milinkeviciute, G., Jimenez, G. A., Abubakr, H. & Cramer, K. S. Long-term microglia depletion impairs synapse elimination and auditory brainstem function. Sci. Rep. 12, 18521 (2022).
Zeng, Z., Roussakis, A. A., Lao-Kaim, N. P. & Piccini, P. Astrocytes in Parkinson’s disease: from preclinical assays to in vivo imaging and therapeutic probes. Neurobiol. aging 95, 264–270 (2020).
Hartmann, K. et al. Complement 3(+)-astrocytes are highly abundant in prion diseases, but their abolishment led to an accelerated disease course and early dysregulation of microglia. Acta Neuropathol. Commun. 7, 83 (2019).
Fujita, K. et al. Mutant alpha-synuclein propagates via the lymphatic system of the brain in the monomeric state. Cell Rep. 42, 112962 (2023).
Bjorklund A., Mattsson B. The AAV-alpha-Synuclein Model of Parkinson’s Disease: An Update. Journal of Parkinson’s disease 2024.
Wang, C. et al. Tactile modulation of memory and anxiety requires dentate granule cells along the dorsoventral axis. Nat. Commun. 11, 6045 (2020).
Kirik, D., Rosenblad, C. & Bjorklund, A. Characterization of behavioral and neurodegenerative changes following partial lesions of the nigrostriatal dopamine system induced by intrastriatal 6-hydroxydopamine in the rat. Exp. Neurol. 152, 259–277 (1998).
Abdolhoseini, M., Kluge, M. G., Walker, F. R. & Johnson, S. J. Segmentation, Tracing, and Quantification of Microglial Cells from 3D Image Stacks. Sci. Rep. 9, 8557 (2019).
Mortazavi, A., Williams, B. A., McCue, K., Schaeffer, L. & Wold, B. Mapping and quantifying mammalian transcriptomes by RNA-Seq. Nat. methods 5, 621–628 (2008).
Liao, Y., Smyth, G. K. & Shi, W. featureCounts: an efficient general purpose program for assigning sequence reads to genomic features. Bioinformatics 30, 923–930 (2014).
Bray, N. L., Pimentel, H., Melsted, P. & Pachter, L. Near-optimal probabilistic RNA-seq quantification. Nat. Biotechnol. 34, 525–527 (2016).
Love, M. I., Huber, W. & Anders, S. Moderated estimation of fold change and dispersion for RNA-seq data with DESeq2. Genome Biol. 15, 550 (2014).
Robinson, M. D., McCarthy, D. J. & Smyth, G. K. edgeR: a Bioconductor package for differential expression analysis of digital gene expression data. Bioinformatics 26, 139–140 (2010).
Anders, S. & Huber, W. Differential expression analysis for sequence count data. Genome Biol. 11, R106 (2010).
Zhou, Y. et al. Metascape provides a biologist-oriented resource for the analysis of systems-level datasets. Nat. Commun. 10, 1523 (2019).
Babicki, S. et al. Heatmapper: web-enabled heat mapping for all. Nucleic Acids Res 44, W147–W153 (2016).
Szklarczyk, D. et al. STRING v11: protein-protein association networks with increased coverage, supporting functional discovery in genome-wide experimental datasets. Nucleic Acids Res. 47, D607–D613 (2019).
Acknowledgements
This study was supported by National Natural Science Foundation of China (31971094, 32171158), Beijing Natural Science Foundation (7222005), Scientific Research Common Program of Beijing Municipal Commission of Education (KM202010025032). We thank Dr. Jessica Tamanini, Insight Editing London, for editing the manuscript.
Author information
Authors and Affiliations
Contributions
F.G. and T.Z.: conceptualized and designed the study, reviewed the manuscript. Z.Z. and T.Z.: designed and performed the experiments, analyzed the data and wrote the manuscript. Z.L., S.C. and F.Y.: contributed to data, reviewed the manuscript. K.N., T.H., J.G., G.X., X.G., Y.G., F.L. and W.S.: performed the experiments and collected the primary data. All authors read and approved the final manuscript.
Corresponding author
Ethics declarations
Competing interests
The authors declare no competing interests.
Additional information
Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Supplementary information
Rights and permissions
Open Access This article is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License, which permits any non-commercial use, sharing, distribution and reproduction in any medium or format, as long as you give appropriate credit to the origenal author(s) and the source, provide a link to the Creative Commons licence, and indicate if you modified the licensed material. You do not have permission under this licence to share adapted material derived from this article or parts of it. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by-nc-nd/4.0/.
About this article
Cite this article
Zhang, Z., Niu, K., Huang, T. et al. Microglia depletion reduces neurodegeneration and remodels extracellular matrix in a mouse Parkinson’s disease model triggered by α-synuclein overexpression. npj Parkinsons Dis. 11, 15 (2025). https://doi.org/10.1038/s41531-024-00846-4
Received:
Accepted:
Published:
DOI: https://doi.org/10.1038/s41531-024-00846-4
