Microglia and other brain resident macrophages

Microglia, resident hematopoietic cells in the CNS parenchyma, maintain homeostasis and surveil the environment in the healthy brain[89]. These cells originate from yolk sac-derived erythromyeloid precursors, which in a process called microgliogenesis, populate the brain during early stages of fetal development[90–

92]. According to these findings, yolk sac macrophages enter the murine brain rudiment via the bloodstream and migrate to the neuroepithelium around E9.5.

These cells resemble macrophages in terms of their morphology and F4/80 and CD11b expression. It has been shown that microgliogenesis is dependent on Irf8/PU.1[93]. Under healthy conditions, microglial cells are highly ramified, express low levels of CD45, MHCII, and Fc surface markers and their replenishment rate is relatively low[94]. Ischemic stroke leads to microglial cell activation, resulting in morphological changes to an ameboid shape, which facilitates an effective response to the environmental changes. Subsequently, microglia start to proliferate and migrate toward lesions, release various inflammatory factors and phagocyte cell debris.

There are several types of non-parenchymal macrophages, e.g. perivascular, subdural meningeal and choroid plexus macrophages, which have been shown to

participate in brain immune responses[95]. These cells originate from hematopoietic precursors, and with the exception of choroid plexus macrophages, represent a stable population with a low turnover with blood-derived monocytes[95]. As these cells are residing in strategical locations within CNS and interact with the vasculature, one of their main responsibilities during neuroinflammation includes antigen presentation to circulating lymphocytes[96]. It has been recently shown in a rat model of tMCAo, that CD163⁺ cells defined as perivascular and meningeal macrophages are key players in control of leukocyte chemotactic infiltration, as well as prominently impact BBB integrity in the acute phase of ischemic stroke[97].

2.1.4.2.3.1 Microglial activation in ischemic stroke

Activated microglia express typical myeloid markers and acquire round amoeboid shape, which makes them barely distinguishable from blood-borne macrophages[98], a major challenge in neuroinflammation research. Nevertheless, recent studies of distinct molecular and functional signatures in microglia have pinpointed several microglia-specific genes, like surface markers P2ry12, Gpr34, Mertk, secreted C1qa, Pros1 and Gas6 in humans, and Fcrls, Tmem119, Olfml3, Hexb, Tgfbr1, and Sall1 in mice[99]. In addition, CD11b and Iba1 remain typical markers of activated microglia, together with molecules associated with antigen presentation, such as MHC-II[100]. Under brain ischemia microglial activation is regulated by a number of receptors, including purinergic ATP receptors, such as P2RX7 and P2Y12, TLR4, CX3CR1, PPAR-γ, CB2, TREM2 and CD200R[101]. Microglia become activated very rapidly after the ischemic insult, their reactivity peaks a few days after the stroke and may persist for several weeks. This is evident both in animal stroke models[102,103] and in patients[104,105]. Spatiotemporal characteristics of immune cell activation in cerebral ischemia are summarized in Figure 3.

Figure 3. Spatiotemporal profile of microglia and myeloid immune cell polarization in transient brain ischemia. Adapted from Benakis et al., 2015.

Polarization of activated microglial cells results in acquiring M1 or M2-like phenotype, mediating different cellular responses. Classically activated M1-type microglia/macrophages produce proinflammatory cytokines like IL-1β, IL-6, IL-23, while M2-type, alternatively activated cells are able to induce Th2 cell responses and release anti-inflammatory cytokines like IL-4[106]. Of note, neutrophils are also capable of acquiring N1 pro- or N2 anti-inflammatory phenotype, and their activation profiles resemble those of microglia and macrophages (Fig. 3). The types of microglial and macrophage activation are summarized in Figure 4. M1 phenotype, modulated primarily through TLRs, is generally considered detrimental, since persistent release of proinflammatory cytokines, like TNF-α[107], MCP1, and MIP-1 α[108], triggers increased ROS production, representing a potential source of cytotoxicity[109]. M1-type microglial cells also release other factors, such as reactive oxygen and nitrogen species[110], MMPs and other proteases. MMPs activity, apart from the phagocytosis of endothelial cells by perivascular microglia, have been highlighted as one of the substantial factors responsible for loss of BBB integrity leading to leukocyte influx into the ischemic penumbra[111]. Harmful mediators derived from M1-type microglial cells play a proven role in ischemic injury and inflammation, as pharmacological or miRNA-induced regulation of microglial activity ameliorates ischemic brain injury[112–114]. Rational immunomodulation in the early stages after ischemic stroke has been highlighted as a promising treatment strategy in experimental stroke models.

M2-type microglia, also called alternatively activated microglia, are characterized by release of anti-inflammatory factors, such as IL-4, IL-10, IL-13, and TGF-β, and are considered beneficial for ischemic stroke outcome[115]. M2-type microglia are mostly found in the early stages after ischemic injury, whereas in later stages the M1-like phenotype becomes increasingly predominant[108]. M2-type microglia polarize into different subpopulations (M2a, M2b, M2c and M2d), which have unique features, distinct gene expression profiles and specific biological functions[116,117].

The M2a (alternative) subtype is induced by IL-4 and IL-13, M2b (immunoregulatory) elicited by IL-1 receptor ligands, immune complexes and LPS, while M2c (immunosuppressing) is mediated by IL-10, TGF-β and glucocorticoids[118]. IL-6, TLR agonists and adenosine are responsible for macrophage/microglial polarization into M2d subtype[119]. Interestingly, some studies have shown that in the tMCAo model, selective ablation of proliferating resident microglia resulted in larger brain infarction, associated with an increased amount of apoptotic neurons[120].

Phagocytosis of damaged cells is a crucial process exerted mainly by microglia and macrophages, supporting structural and functional reorganization of the injured brain[121]. Complement-mediated phagocytosis shifts the cytokine profile from proinflammatory to anti-inflammatory[122], however transient complement

inhibition in the acute stroke phase can also alleviate neuroinflammation and improve neuronal survival[123]. CD68, a common marker used to visualize active phagocytic cells, is upregulated in the penumbra early in pMCAo, and then gradually increases in the core to peak at 7 dpi[124]. Scavenger receptors MARCO, CD36, mannose receptor CD206, Ym1, and Trem2 are other markers linked to phagocytic activity or prevention of extracellular matrix degradation[125–129].

Neuronal death and functional impairment can be prevented by blocking specific phagocytic pathways. For example, abolishing MerTK and MFG-E8 action effectively suppresses phagocytosis of viable neurons and in turn ameliorates neurological deficits after focal brain ischemia[129].

Figure 4. Scheme representing the types of microglial and macrophage activation states. The division for M1 and M2 polarization is considered rather simplistic, yet practical approach to discriminate between the microglia and macrophage multiple subtypes. Adapted from Rőszer, 2015.

2.1.4.2.4 Blood-derived monocytes/macrophages in cerebral ischemia