Analysis of monocyte infiltration in MPTP mice reveals that microglial CX3CR1 protects against neurotoxic over-induction of monocyte-attracting CCL2 by astrocytes

BACKGROUND: Evidence from mice suggests that brain infiltrating immune cells contribute to neurodegeneration, and we previously identified a deleterious lymphocyte infiltration in Parkinson’s disease mice. However, this remains controversial for monocytes, due to artifact-prone techniques used to distinguish them from microglia. Our aim was to reassess this open question, by taking advantage of the recent recognition that chemokine receptors CCR2 and CX3CR1 can differentiate between inflammatory monocytes and microglia, enabling to test whether CCR2(+) monocytes infiltrate the brain during dopaminergic (DA) neurodegeneration and whether they contribute to neuronal death. This revealed unexpected insights into possible regulation of monocyte-attracting CCL2 induction. METHODS: We used acute 1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine (MPTP) mice and assessed monocyte infiltration by combining laser microdissection-guided chemokine RNA profiling of the substantia nigra (SN) with immunohistochemistry and CCR2-GFP reporter mice. To determine contribution to neuronal loss, we used CCR2-deletion and CCL2-overexpression, to reduce and increase CCR2(+) monocyte infiltration, and CX3CR1-deletion to assess a potential implication in CCL2 regulation. RESULTS: Nigral chemokine profiling revealed early CCL2/7/12-CCR2 axis induction, suggesting monocyte infiltration in MPTP mice. CCL2 protein showed early peak induction in nigral astrocytes, while CCR2-GFP mice revealed early but limited nigral monocyte infiltration. However, blocking infiltration by CCR2 deletion did not influence DA neuronal loss. In contrast, transgenic astrocytic CCL2 over-induction increased CCR2(+) monocyte infiltration and DA neuronal loss in MPTP mice. Surprisingly, CCL2 over-induction was also detected in MPTP intoxicated CX3CR1-deleted mice, which are known to present increased DA neuronal loss. Importantly, CX3CR1/CCL2 double-deletion suggested that increased neurotoxicity was driven by astrocytic CCL2 over-induction. CONCLUSIONS: We show that CCR2(+) monocytes infiltrate the affected CNS, but at the level observed in acute MPTP mice, this does not contribute to DA neuronal loss. In contrast, the underlying astrocytic CCL2 induction seemed to be tightly controled, as already moderate CCL2 over-induction led to increased neurotoxicity in MPTP mice, likely due to the increased CCR2(+) monocyte infiltration. Importantly, we found evidence suggesting that during DA neurodegeneration, this control was mediated by microglial CX3CR1 signaling, which protects against such neurotoxic CCL2 over-induction by astrocytes, thus hinting at an endogenous mechanism to limit neurotoxic effects of the CCL2-CCR2 axis. ELECTRONIC SUPPLEMENTARY MATERIAL: The online version of this article (doi:10.1186/s12974-017-0830-9) contains supplementary material, which is available to authorized users.