Activators of neurotoxic microglia in neurodegeneration: is the answer in blood?

Microglia are the resident immune cells of the central nervous system. Making up approximately 12% of cells, microglia are a key player in a diverse range of important functions including regulation of brain development and homeostasis, synaptic pruning and maintaining neuronal networks and clearing cell debris through phagocytosis. When activated, microglia migrate to the site of damage and participate in cellular responses to insult and injury. Microglia may be neuroprotective in pathologies associated with neurodegenerative diseases, such as dementias and Parkinson's disease. For example, loss-of-function mutations in triggering receptor expressed on myeloid cells 2 (TREM2), which prevent adequate microglial activation in response to insult, significantly increase the risk of Alzheimer's disease (AD).1 Further, microglia can modulate destructive inflammation by releasing anti-inflammatory mediators, including interleukin-10 and transforming growth factor-β, and promote neurogenesis after injury by facilitating increased stem cell proliferation and differentiation.2 Importantly, however, microglia are also implicated in chronic neuroinflammation and can drive neurodegeneration. Microglia can promote inflammation via release of inflammatory mediators such as inducible nitric oxide synthase, cathepsins, matrix metalloproteases and tumor necrosis factor.3 Persistently activated microglia become neurotoxic, participating in cellular necrosis, excitotoxicity, neuronal injury, apoptosis and neurodegeneration.3 The dual roles of microglia, as both neuroprotective and neurodegenerative agents, likely stem from different microglial states. In contrast to neuroprotective microglia, which release anti-inflammatory mediators and help promote tissue homeostasis and repair, disease-associated microglia (DAM) contribute to the release of proinflammatory mediators, neuronal loss and neurodegenerative processes.4, 5 Understanding the activation processes and promoting beneficial microglia as opposed to DAM development are critical for developing effective therapeutic interventions for neurodegenerative diseases. A recent study by Mendiola et al.6 took an essential step toward defining the activators of pathogenic microglia. Through a combination of unbiased transcriptomic and phosphoproteomic profiling of microglia and macrophages, the authors demonstrate that blood proteins can elicit widespread proinflammatory transcriptional and signaling changes in microglia and suggest that fibrin, a blood protein involved in coagulation, and fibrin–CD11b signaling are key mediators of neurotoxic microglial function in neurodegenerative disease (Figure 1). Using the plasma of wild-type mice delivered directly into the brain, Mendiola et al. show that blood proteins indeed induce pathogenic microglial changes, including upregulation of genes involved in oxidative stress and DAM. While the use of plasma from C3- or albumin-deficient mice similarly upregulated genes associated with neurotoxic microglia, fibrinogen-deficient mouse plasma had no effect on microglial gene expression. Further, the microglial response was similarly reduced in Fggγ390–396A mice that lack the fibrinogen CD11b–CD18 binding motif, suggesting that fibrin regulates the central nervous system's innate immune system through a distinct fibrinogen CD11b pathway. Mendiola et al. offer further evidence for a fibrin–CD11b signaling pathway through their use of three in vitro mouse cell lines, including primary microglia, bone marrow–derived macrophages and the immortal RAW264.7 macrophages. Single-cell transcriptomic analyses indicated that fibrin stimulation led to increased expression of genes associated with oxidative stress, lipid metabolism and type I interferon response. Phosphoproteomics indicated that fibrin regulated several phosphorylation signaling pathways, including vascular endothelial growth factor (VEGF), T-cell receptor and cytoskeleton organization, via casein kinase II (CSNK2A1), S6 kinase (RPS5KA1), protein kinases C and D and pyruvate dehydrogenase kinases. In addition, the pathways implicated in fibrin response were distinct from those induced by other immunogenic ligands. Unlike fibrin, complement iC3b and lipopolysaccharide were found to recruit host defense and classical inflammatory responses, respectively, further highlighting the specificity of fibrin–CD11b signaling in mediating neurotoxic microglial responses. Importantly, Mendiola et al. validated their findings in two mouse models of neurodegenerative disease, the experimental autoimmune encephalomyelitis (EAE) model of multiple sclerosis and the 5XFAD model of familial AD (FAD). Single-cell RNA-sequencing of oxidative stress producing cells (Tox-seq) indicated that in EAE, fibrin was a potent activator of oxidative stress and reactive oxygen species (ROS) expression in microglia and infiltrating macrophages, both of which are found in the neuroinflammatory lesions of EAE mice. In FAD, by contrast, fibrin facilitated the polarization of microglia toward a DAM and neurotoxic phenotype, an effect again shown to be dependent on the fibrinogen CD11b–CD18-binding motif. Further, in 5XFAD primary microglia, fibrin-regulated phosphorylation was associated with focal adhesion of platelets, proinflammatory responses and ROS generation. Mendiola et al.'s findings are consistent with many studies demonstrating that blood extravasation into the brain through a leaky blood–brain barrier (BBB) contributes to neuroinflammation and neurodegeneration.7 Although Mendiola et al. extend these findings to implicate fibrin–CD11b signaling as a central mechanism underlying this effect, there are questions that remain. First, given that blood is extravasating into the brain, it is critical to differentiate the microglial effects of fibrin from other blood-borne proteins. For example, thrombin has also been demonstrated to activate DAM, causing subsequent ROS and neurodegenerative processes,8 and it is not clear how its resultant microglial transcriptomic and phosphoproteomic profiles differ from fibrin. Another remaining question is whether fibrin, or other blood proteins, drive neurodegeneration or do they simply exacerbate it? The evidence presented by Mendiola et al. largely suggests the latter. In EAE and 5XFAD mouse models, neurodegenerative, disease-associated processes are already triggered by either active immunization, such as with myelin oligodendrocyte glycoprotein peptide MOG35–55, or rare genetic mutations, respectively, suggesting that fibrin further exacerbates disease. It may be the case, however, that fibrin drives neurodegenerative processes. For example, fibrin may interact with microvasculature and cells of the BBB and neurovascular unit, such as endothelial cells, astrocytes and, importantly, microglia, to promote a leaky BBB whereby it can trigger blood extravasation and DAM-mediated neurodegeneration. It would have been interesting if Mendiola et al. tested this by examining the long-term consequences of wild-type plasma injected into the brain, including whether it causes neurodegeneration. Another way to test this is through the use of next-generation BBB in vitro models that recapitulate key aspects of the spatial organization, cellular subtypes (including endothelial cells and astrocytes) and 3D microenvironment of the brain.9 Preventing neurotoxic microglia and DAM remains an important line of enquiry in the pursuit of developing effective therapeutic strategies for neurodegenerative diseases and it is therefore critical to answer these questions to better understand if fibrin may be a suitable therapeutic target. Overall, Mendiola et al. have clearly identified a fibrin–CD11b signaling pathway as one of the key mechanisms underlying blood protein-mediated activation of neurotoxic microglia. A remaining question is whether fibrin itself drives microglial-dependent neurodegeneration or whether it simply exacerbates ongoing neurodegenerative processes. Further questions are also raised about interactions between microglia and other cells of the BBB, including astrocytes and endothelial cells, and the complex milieu that they find themselves in. Preventing neurotoxic microglia development and harnessing their neuroprotective actions are important in the context of preventing and treating neurodegenerative diseases. It is therefore important to consider fibrin's role and to better understand if it might represent a therapeutic target for neurodegenerative diseases. The authors have no conflicts of interest to declare. Caitlin A Finney: Conceptualization; writing – original draft; writing – review and editing. Heather M Wilkins: Conceptualization; writing – review and editing. Russell H Swerdlow: Conceptualization; writing – review and editing. David A Brown: Conceptualization; writing – review and editing.