Partial depletion and repopulation of microglia have different effects in the acute MPTP mouse model of Parkinson’s disease

Abstract Objectives Parkinson's disease (PD) is a common neurodegenerative disorder characterized by the progressive and selective degeneration of dopaminergic neurons. Microglial activation and neuroinflammation are associated with the pathogenesis of PD. However, the relationship between microglial activation and PD pathology remains to be explored. Materials and Methods An acute regimen of MPTP was administered to adult C57BL/6J mice with normal, much reduced or repopulated microglial population. Damages of the dopaminergic system were comprehensively assessed. Inflammation‐related factors were assessed by quantitative PCR and Multiplex immunoassay. Behavioural tests were carried out to evaluate the motor deficits in MPTP‐challenged mice. Results The receptor for colony‐stimulating factor 1 inhibitor PLX3397 could effectively deplete microglia in the nigrostriatal pathway of mice via feeding a PLX3397‐formulated diet for 21 days. Microglial depletion downregulated both pro‐inflammatory and anti‐inflammatory molecule expression at baseline and after MPTP administration. At 1d post‐MPTP injection, dopaminergic neurons showed a significant reduction in PLX3397‐fed mice, but not in control diet (CD)‐fed mice. However, partial microglial depletion in mice exerted little effect on MPTP‐induced dopaminergic injuries compared with CD mice at later time points. Interestingly, microglial repopulation brought about apparent resistance to MPTP intoxication. Conclusions Microglia can inhibit PD development at a very early stage; partial microglial depletion has little effect in terms of the whole process of the disease; and microglial replenishment elicits neuroprotection in PD mice.


| INTRODUC TI ON
Parkinson's disease (PD) is the second most common neurodegenerative disease, threatening human health. PD is clinically characterized by symptoms such as resting tremor, bradykinesia, rigidity and postural instability, accompanying non-motor symptoms such as olfactory hypotension, sleep disturbance, depression and anxiety. 1 and CD86 or TGFβ, IL4, IL10, CD206, CCL22 and TREM2 are upregulated, respectively. 7 In the conditions of brain injury and neurodegenerative diseases, both neuroprotective and detrimental effects of microglia have been revealed. The development of pharmacological inhibition and genetic targeting (eg, DTR expressing exclusively in microglia), which can deplete microglial cells or suppress microglial activation in vivo by specific inhibitors or DT toxin, has much expanded our understandings of microglia. 8 Microglial activation and neuroinflammation are common features in the early stage of Parkinson's disease, suggesting that neuroinflammation contributes to the onset and progression of PD. 9,10 The progressive death of dopaminergic neurons in PD causes the release of endogenous damage-associated molecular patterns (DAMP), which leads to excessive activation of microglia, and subsequently, activated microglia elicit an inflammatory response. [11][12][13] Maintenance of the microglial population depends on the continuing activation of the receptor for colony-stimulating factor 1 (CSF1R), which is also essential for microglial development. [14][15][16] The pharmacological blockade of CSF1R by PLX3397 and many other similar inhibitors lead to a rapid depletion of microglia, and microglia reach full repopulation soon after removal of the inhibitors. 14,[17][18][19][20][21] With the strategy of microglial elimination, Yang et al showed that microglial ablation led to the augmented dopaminergic neurotoxicity in MPTP-treated mice, 18 while Oh et al found that in 6-OHDA-induced PD rats, depletion of microglia elicited beneficial effects on motor and non-motor symptoms of PD. 22 Moreover, in rotenone-induced mouse PD models, microglial activation contributed to neurodegeneration in the locus coeruleus and cognitive impairments. 23,24 Thus, whether microglia are beneficial or harmful remains controversial in PD. And the effects of repopulated microglia in PD are also warranted to illustrate.
In this study, to clarify the function of microglia in Parkinson's disease, an acute regimen of MPTP was administered to adult C57BL/6J mice with a normal, much reduced or repopulated microglial population. We comprehensively analysed the impact of microglia at different stages in PD animal models, including the damages of the nigrostriatal pathway, the levels of inflammation, the changes in glial cells and the mouse behaviours. Our results demonstrate that microglial depletion has little effect on dopaminergic injuries; however, repopulated microglia elicits neuroprotection, in PD mice.

| Animal research
Male C57BL/6J mice, 10 weeks old and weighing 26-29 g, were  Mice were fed with a PLX3397-formulated diet (PLX3397) for 7, 10, 14 and 21 days to determine the optimal days for microglial elimination and repopulation in the nigrostriatal pathway (n = 3 per group). After 21 days, a subset of mice with the PLX3397 diet were switched to a control diet (CD) for 4 and 7 days (n = 3 per group) to repopulate microglia.

| Compounds
Mice were fed with a PLX3397 diet or a control diet for 21 days and followed by a diet according to the requirements of different experiments to clarify the function of microglia in PD (n = 4-5 per group).
The design of each experiment was shown as a diagram, respectively.

| Mouse treatment
Mice were intraperitoneally injected with MPTP•HCl at a dose of 10 mg/kg (Sigma, USA) or normal saline (NS) for four times at 2 h of intervals. Animals were sacrificed at different time points after the last injection.

| Immunofluorescence
Animals were anaesthetized and perfused transcardially with normal cold saline, followed by 4% paraformaldehyde. Brains were Conclusions: Microglia can inhibit PD development at a very early stage; partial microglial depletion has little effect in terms of the whole process of the disease; and microglial replenishment elicits neuroprotection in PD mice. collected and post-fixed with 4% paraformaldehyde overnight and subsequently immersed in 15% and 30% sucrose at 4°C overnight.
Dissected brains were embedded in the OCT compound then cut into coronal sections (30 μm) by a freezing microtome and stored in a cryoprotectant solution at −20°C for further analysis. The sections were rinsed and permeabilized with PBS containing .5% Triton X-100 for 20 min, blocked in 10% normal donkey serum containing 0.05% Tween-20 for .5 h and then incubated with primary antibodies at 4°C overnight. The following antibodies were used as follows: mouse anti-TH (1:1000, ImmunoStar, USA), chicken anti-GFAP (1:500, EnCor Biotechnology, USA) and rabbit anti-Iba1 (1:500, Abcam, USA). Appropriate secondary antibodies conjugated with Alexa fluorophore 488, 594 or 647 were used for visualization.

| Western blot
Protein samples were lysed in RIPA lysed buffer containing protease inhibitor cocktail (1:100, BioMake, China). The lysates were centrifuged at 16 000g for 15 min at 4°C, and then, the supernatants were collected. The protein concentration of samples was determined using a BCA protein assay kit (Beyotime, China). Equal amounts of 30 µg total proteins were loaded onto SDS-PAGE. Following electrophoresis, proteins were transferred to Immobilon-PSQ membranes

| Quantitative real-time PCR assays
The total RNA was extracted using TRIzol reagent according to the manufacturer's protocols. Quantitative analysis of RNA was performed by using NanoDrop spectrophotometer. 2 µg total RNA was used for complementary DNA (cDNA) synthesis using the One-Step gDNA Removal and cDNA Synthesis SuperMix (Transen, China).
Real-time PCR was performed in duplicates with a quantitative thermal cycler (Thermo Fisher Scientific, USA). The following PCR con-

| Multiplex immunoassay
Protein samples were lysed in a T-PER lysed buffer containing pro-

| High-performance liquid chromatography
The dissected striatum was homogenized in .4 M HClO 4 for 30 s.
The lysates were centrifuged at 12,000g for 10 min at 4°C. Then, the supernatants were collected to determine the concentrations of DA and its metabolites homovanillic acid (HVA) and 3, 4-dihydroxyphenylacetic acid (DOPAC), as well as serotonin  and hydroxyindole acetic acid (5-HIAA) by using the chromatograph (ESA, USA) with a 5014B electrochemical detector.

| Rotarod test
One day before the experiment, mice were pre-trained on a rotarod instrument (MED Associates, USA) three times, separated by onehour intervals. At testing days, mice were placed on the rod rotating at a speed of 20 rpm, and the latency to fall and times of drop were measured. Total time spent on the rotarod was measured up to a maximum of 300 s.

| Pole test
The pole test was carried out 3 and 7 days after MPTP injection to evaluate the motor function impairment. A wood pole with a rough surface, 75cm in length and 1cm in diameter, was placed vertically in the cage of mice to be tested. Briefly, mice were placed with their head facing upside on top of the pole; the time required for the mice to turn around and climb down was recorded to assess the movement initiation and coordination.

| Cell counting
Stereological cell counting was conducted to measure the density of TH-positive cells in the substantia nigra par compacta (SNpc) using a Stereo Investigator system (Micro Brightfield, USA) attached to a Leica microscope as previously described 25

| Statistical analysis
All data were expressed as means ± SEM. Data were assessed for normal distribution by the Shapiro-Wilk test. Unpaired two-tailed Student's t tests were used to compare two groups. One-way or two-way analysis of variance (ANOVA) followed by Holm-Sidak's or Dunn's multiple comparisons test was used to compare multiple groups as appropriate. Statistical analysis was performed using the Prism 7 software (GraphPad Software Inc, USA). P < .05 was considered statistically significant. The interaction between the treatments was tested, and corresponding F-, dfn-, dfd-and P-values of significant interaction were showed.

| PLX3397 dramatically reduces the microglial population in the substantia nigra, but has no obvious effect on the basal status of nigrostriatal pathway in mice
To evaluate whether microglial depletion affects the nigrostriatal pathway in mice, we first established an appropriate treatment paradigm for the microglial depletion and repopulation models with CSF1R inhibitors PLX3397. C57BL/6 mice were fed with a PLX3397-formulated diet for up to 21 days ( Figure 1A). Mice showed a dramatic reduction in body weight 7 days after PLX3397 treatment, which returned straightly back to normal body weight afterwards ( Figure S1A

| Partial depletion of microglia has distinct effects on dopaminergic neurons degeneration in the processes of MPTP-induced injuries
An acute regimen of MPTP (10 mg/kg body weight) was adopted in this study. Mice were treated with a control diet (CD) or CD-and PLX3397-treated mice ( Figure S2A, B).
In the striatum, immunohistochemistry staining showed that the density of TH-positive nerve fibres decreased dramatically after MPTP administration, whereas it did not differ between two groups of mice at all experimental time points ( Figure 2D, E). Similarly, striatal protein levels of TH were determined by Western blot assays, which were dramatically reduced from 1d up to 7d after MPTP intoxication, and the reduction was close between the two groups of mice ( Figure 2F, G). Furthermore, the striatal levels of dopamine (DA), DOPAC, HVA, 5-HT and 5-HIAA were analysed by HPLC assay.
In striatum of mice challenged with MPTP, the levels of DA and its metabolites DOPAC and HVA decreased, which were comparable between the CD and PLX3397 groups ( Figure 2H a-c); and the levels of neurotransmitter 5-HT and its metabolite 5-HIAA did not differ at all experimental time points between the two groups ( Figure 2H f, g).
In addition, there were no differences in the ratios of DOPAC to DA, HVA to DA and 5-HIAA to 5-HT between CD-and PLX3397-treated In the striatum, the numbers of microglia increased rapidly in CDtreated mice and restored at 5 days after MPTP administration; on the contrary, microglial cells did not proliferate in PLX3397-treated mice during the whole experimental process [F(4, 60) = 7.388, P < .0001; Figure 3B, F]. In terms of astrocytes, we found that MPTP exposure elicited increases in the astrocyte densities at 1 day after MPTP administration and maintained activated afterwards, whereas there was no difference between the two groups ( Figure 3C, G).
Besides, Western blot assays showed that the striatal GFAP protein levels were upregulated in the striatum at 3 days after MPTP injection, whereas there was no difference between the CD and PLX3397 groups (Figure S3 A, B). whereas GFAP, TNFα, IL1β, COX2 and iNOS transcripts did not differ between the CD and PLX3397 groups ( Figure 4C).

| Partial depletion of microglia mitigates the production of inflammatory mediators in the nigrostriatal pathway
To evaluate whether microglial depletion affects the production of inflammation-related mediators in the nigrostriatal pathway after MPTP exposure, we detected both the pro-inflammatory and the anti-inflammatory cytokines were detected by Multiplex  Figure 5A, B). Anti-inflammatory cytokines IL4 and CCL22 were significantly reduced in the SN, but not in the striatum of PLX3397-treated mice compared with CD-treated mice in both NS and MPTP groups, and IL13 was dramatically reduced in the SN, but not in the striatum of PLX3397-NS mice, while IL10 levels did not change in the nigrostriatal pathway ( Figure 5C, D).

| Fully repopulated microglia lead to neuroprotection in MPTP-induced PD mice
Microglia can restore rapidly after the withdrawal of PLX3397. Here, the effects of repopulated microglia in the mouse model of PD were investigated. Two paradigms of repopulation were adopted ( Figure   S5A, Figure 6A). First, repopulation of microglia by the cessation of PLX3397 at the time of MPTP administration yielded no difference in the nigrostriatal degeneration compared to the mice fed with a control diet and PLX3397-formulated diet during the whole experimental process, at 7 days post-MPTP injection ( Figure S5B, C). In the second paradigm, after the full recovery of microglia in the mouse brain, MPTP was injected to induce the PD model. We found that at 7 days post-MPTP injection, fully repopulated microglia brought about the mitigated reduction in dopaminergic neurons in mice compared with CD-treated mice ( Figure 6B, C). However, the numbers of microglia did not differ ( Figure 6D, F did not change ( Figure 7B).
Finally, since glial cells can phagocyte apoptotic cells, therefore avoiding further inflammation, 41,42 we measured the transcripts of phagocytosis-related molecules, including triggering receptor expressed on myeloid cells 2 (TREM2) and complement receptor 3 (CD11b).
The transcript level of TREM2, but not CD11b increased dramatically in microglial repopulation groups ( Figure 7C). Microglial depletion is a powerful approach to decipher the roles of microglia in various brain injuries. Beneficial, detrimental, dual roles and even null impacts of microglia in differential disease processes have been reported. Microglial ablation exacerbates post-ischaemic inflammation and brain injury, 20 and aggravates the severity of acute and chronic seizures in mice, 53 suggesting the neuroprotective effects of microglia in such brain injury models. However, the elimination of microglia mitigates brain injury after intracerebral haemorrhage, 54  Microglia rely on the colony-stimulating factor 1 receptor (CSF1R)

| D ISCUSS I ON
signalling to proliferate and survive. 14 CSF1R inhibition, using a selective pharmacological ATP competitive inhibitor GW2580, does not deplete microglia but attenuates disease-stimulated microglial proliferation. GW2580 manifests protective effects in multiple brain disorders, including AD, 63 amyotrophic lateral sclerosis, 64 spinal cord injury, 65 and PD as well. 66 Notably, in rodent models of PD, the controversial effects of microglial depletion were brought up. Previous studies and our data showed that the microglia rapidly repopulated the brain after PLX3397 withdrawal for 4-7 days, which was reported to be stemmed from the residual microglia. 68 Repopulated microglia appear to be less reactive, which could repopulated microglia promote brain repair in a mouse model of traumatic brain injury. 69 Replacement of microglia in the aged brain reverses cognitive deficits in mice. 70 Rice et al have reported that removing the reactive microglia and repopulating with new microglial cells offer a strategy to resolve neuroinflammation and promote recovery. 71 Microglial repopulation promotes an anti-inflammatory, trophic microenvironment and normalizes pro-inflammatory gene expression in organotypic hippocampal slice cultures. 72 Additionally, Willis et al have reported that repopulated microglia support functional neurogenesis during a critical time window after traumatic brain injury in an IL6-dependent manner. 69 Microglia in the mammalian brain can be manipulated into a neuroprotective and regenerative phenotype to help repair and alleviate cognitive deficits caused by brain injury. 69 Therefore, all of the above mentioned support the possibility of microglial depletion/repopulation as a strategy to reverse chronic neuroimmune activation. We found

ACK N OWLED G EM ENTS
This work was supported by grants from the National Natural

DATA AVA I L A B I L I T Y S TAT E M E N T
The data that support the findings of this study are available from the corresponding author upon reasonable request.