Microglial cells participate in neurodegenerative pathologies development. Thus, the analysis of their specific role in these processes has lately emerged as an important focus of research, seeking for therapeutic approaches for diseases like MS 38. In this study, we used the CSF-1R antagonist, PLX5622, in order to assess the effect of microglia depletion in EAE development. However, we cannot exclude that the effect observed in EAE development after PLX5622 treatment could be partly due to the elimination of meningeal macrophages. We demonstrated that microglia and meningeal macrophages interfere with peripheral macrophage dynamics by controlling their entry and migration. The massive infiltration of macrophages in the absence of microglia and meningeal macrophages does not significantly affect neurological damage at EAE peak or recovery in EAE chronic phase. However, microglia and meningeal macrophages ablation delays EAE onset demonstrating a specific role of these cells in antigen presentation and T cell proliferation at early stages of EAE.
PLX5622 has largely been assumed to be microglia-specific but more recent studies described alterations in myeloid and lymphoid populations in peripheral tissues (Lei et al., 2020; Spiteri et al., 2022). In our study we did not find any alteration in the immune populations in spleen and blood nor in the expression of antigen presenting proteins after PLX5622 treatment, a fact suggesting that PLX5622 did not affect directly immune cells, including peripheral monocytes. While tissue macrophages and circulating monocytes express CSF-1R, their survival is not only depending on CSF1/CSF1R signaling but relies also on CCL2/CCR2 signaling40 which is not present in microglia. In contrast to infiltrating and peripheral macrophages, we found that CAMs, in particular meningeal macrophages, were also depleted after PLX5622 treatment. In accordance, perivascular macrophages survival is also dependent on CSF1/CSF1R signaling, as PLX5622 treatment induced a 60% reduction in their number 41.
We showed that microglial and meningeal macrophages depletion provoke a massive infiltration of CCR2+ peripheral macrophages during EAE progression in comparison to control mice, which could constitute a compensatory mechanism. Importantly, infiltrating macrophages in PLX5622-treated mice colonize CNS parenchyma including both white and grey matter of the spinal cord, and their location is not limited to demyelinated lesions as in control mice. These data suggest that microglia control the dispersion of macrophages throughout the CNS parenchyma. In accordance, microglia surround and confine infiltrating macrophages into the CNS parenchyma, limiting macrophage dispersion after LPC-induced demyelination 42. In addition, we detected a higher accumulation of macrophages in demyelinated lesions after PLX5622 treatment, suggesting that microglia and meningeal macrophages also control the infiltration, proliferation or survival of macrophages into the CNS parenchyma. In accordance, previous data using parabiosis in the EAE model demonstrated that peripheral macrophage infiltration is preceded by microglia cell death 9. Thus, microglia could potentially help to limit peripheral CNS inflammation and to maintain the “CNS immune-privileged” status.
PLX5622 treatment induced a consistent delay in the appearance of the first EAE symptoms. This suggests that microglia and meningeal macrophages play a role in the effector stage of the disease model. Indeed, inhibition of chemokine receptor-dependent recruitment of monocytes to the CNS blocked EAE progression, not EAE onset, suggesting that microglia is essential for EAE onset whereas macrophages contribute to EAE progression 9. Specifically, we have seen that microglia and meningeal macrophages ablation caused an alteration in antigen presentation to the first T cells arriving to the CNS, both in dendritic cells and infiltrated macrophages. Frequently, DCs have been identified as the main APCs during EAE 20,43,44. Interestingly, we showed that microglial and meningeal macrophages ablation reduced the expression of co-stimulatory molecules (CD80 and CD86) in all antigen presenting cells, including DC cells, in the spinal cord but not periphery, excluding the possibility of a direct effect of PLX5622 treatment. This suggests microglia and/or meningeal macrophages early activation after EAE induction potentially promote antigen presentation capacity in other cell types. This result is in accordance with those obtained in a virus model, in which PLX5622 also alters T cell local reactivation in CNS decreasing B7 co-stimulatory signals in CD11c+ cells 45. Altogether, these data highlight the relevance of microglia and meningeal macrophages in orchestrating the CNS immune response.
Numerous studies determine that microglia activation occurs during the onset and peak of EAE and its activation is necessary to EAE development. Microglia control T cell encephalogenicity through the release of IL23, specifically the P40 subunit and its deletion in microglia cells suppressed EAE by shifting T cell response towards a Th2 rather than Th1 46. Another important signal for microglia activation is the TGFβ-activated kinase (TAK1). Thus, microglia selective ablation of TAK1 blocks its activation, the release of pro-inflammatory mediators such as IL1β and CCL2, and the immune cell infiltration, suppressing completely EAE 10. Overall, our data, in accordance with previous studies, indicate that microglia and its persistent and overt activation has a detrimental role in CNS autoimmunity onset, and preventing or suppressing this process may be therapeutic.
Targeting CSF-1R has also been tried in different EAE models and at different time windows leading to diverse outcomes. Inhibiting CSF-1R with PLX5622 after EAE onset attenuated EAE pathology and promoted recovery 47. However, blocking CSF-1R with PLX3397 at later stages in EAE exacerbated neuroinflammation and neurological damage in a model of progressive MS 48. This model used non-obese diabetic mice, which have a genetic background of high innate immunity activation and aberrant activation of microglia that leads to an exacerbation of symptoms at the chronic phase 49. Indeed, microglia depletion in the chronic phase (not tested before) blocked EAE progression 48. Thus, the results of these studies are not comparable with ours because both the genetic background of the mice and the time window of the treatment are different.
More importantly, microglia depletion exacerbates demyelination and impairs remyelination in a neurotropic coronavirus infection model of MS 50. The authors showed a higher accumulation of damaged myelin deposits and debris in microglia depleted mice, that leads to impaired myelin repair and prolonged clinical disease. They propose that microglia functions could not be compensated by infiltrating macrophages. These results, although in a different MS model, are at odds with our data as we did not detect a deficit on myelin clearance, suggesting that in the EAE paradigm used in the current study macrophages compensate and efficiently phagocytose myelin. Accordingly, whereas microglia depletion increases monocyte infiltration in EAE lesions in the spinal cord, which could compensate for the microglia loss, it decreases the number of macrophages in the virus model of MS 50.
To sum up, we provide solid evidence showing that microglia and meningeal macrophages limit during EAE induction the infiltration and dispersion of peripheral macrophages throughout the CNS. Moreover, our results suggest that microglia are not essential for the development of chronic EAE nor for proper remyelination in this model. In addition, we described a microglia-related mechanism promoting early antigen presentation in the CNS, by modulating the expression of B7 co-stimulatory molecules in other APCs, such as DCs or other myeloid populations. This lack of antigen presentation to infiltrating T cells delays the reactivation of these lymphoid cells and therefore provokes a slowdown of the EAE early events.