Unexpected role of complement component 8 gamma chain in the inflamed brain

Neuroinflammation, an intricate inflammatory process occurring in the central nervous system (CNS), plays an important role in host defense. Glial cells, including astrocytes and microglia, along with cytokines, chemokines, and the complement system are important components of neuroinflammation. A low level of neuroinflammation is associated with and contributes to various homeostatic and neuroprotective processes, such as removing pathogens or cellular debris and promoting tissue repair after brain injury. However, prolonged or maladaptive neuroinflammation has been implicated in neurodegenerative diseases, such as Alzheimer’s disease (AD), Parkinson’s disease, multiple sclerosis, and major depression. Increasing evidence indicates that targeting neuroinflammation may be a potential therapeutic intervention against neurodegeneration for delaying disease onset or progression. However, our current understanding of neuroinflammation remains limited.

The complement cascade, a part of the complex innate immune surveillance system, is deeply involved in inflammatory response. The complement components circulate as inactive precursors under normal conditions and can be cleaved into smaller active fragments through various proteolytic cascades. These fragments play an important role in defending the body against pathogens through phagocytosis or activating the adaptive immune system. Complement activation is achieved by three major processes: the lectin, classical, and alternative pathways. While the lectin pathway is initiated by binding of the ficolin-mannose-associated serine protease-2 complex or mannose-binding lectin-mannose-associated serine protease-2 complex to sugar monomers (mannose) on microbial surfaces, the classical pathway is initiated by binding of the first complement component, which consists of C1q, C1r, and C1s, to antigen–antibody complexes. Both pathways are similar in structure. In the classical complement cascade, the complement 3 (C3) convertase (C2aC4b complex), generated from the C1 complex, cleaves C3 into C3b and C3a. C3b then produces C5 convertase (C4bC2aC3b), which cleaves the C5 component. Cleavage of C3 and C5, the key proteins in the complement cascade, generates three types of effectors: anaphylatoxins (proinflammatory molecules C3a and C5a), opsonin (C3b and iC3b), and the membrane attack complex (MAC; C5bC6C7C8C9 and C5b–9). Finally, the alternative pathway is initiated by endotoxins and the spontaneous hydrolysis of C3 into active components. Subsequently, interactions among C3, properdin, factor B, and factor D generate the alternative pathway C3 convertase (C3bBbP) on microbial surfaces.

In the CNS, the complement components of the classical pathway are involved not only in neuroprotection from pathogens or harmful stimuli but also play homeostatic roles such as sculpting neural circuits (Kanmogne and Klein, 2021). For example, the anaphylatoxins C3a and C5a are strong chemoattractants for immune cells, thereby facilitating pathogen elimination via phagocytosis while enhancing the humoral response. During CNS development, C1q acts as an initiator for classical complement activation, and activated C3 provides phagocytic signals to microglia for synaptic pruning. C1q deficiency increases synaptic connectivity and induces atypical seizure behaviors, demonstrating its essential role in maintaining normal synapse numbers during CNS development. C1q and C3 are found in developing synapses of the visual system, whereas CR3 is localized to microglia during visual system development. C1q/C3–CR3 signaling is involved in engulfing synapses. Astrocytic transforming growth factor-β upregulates C1q in retinal ganglion cells.

Aberrant activation of complement cascades has been found to exacerbate certain neuropathologies and neurodegeneration. For example, a dramatic increase in C1q levels has been found in the normal aging brain, whereas C1q-deficient aged mice exhibit enhanced synaptic plasticity and less cognitive and memory decline (Wang et al., 2020). The C1q-dependent complement pathway is actively involved in microglial synapse elimination, which results in normal forgetting and early synapse loss in an AD mouse model (Wang et al., 2020). Additionally, excessive C4-mediated synapse elimination is strongly correlated with schizophrenia (Dalakas et al., 2020). These reports suggest that the upregulation of complement protein expression may disrupt synaptic networks by increasing the number of phagocytic microglia. Other studies suggested additional roles of complement proteins in age-related macular degeneration (factor H/heparin sulfate interaction) and in regulating phagocytic inflammatory responses (C1q/surface calreticulin interaction) (Herwald and Egesten, 2014). In the CNS, C5b–9 complexes play a crucial role in activating the cell cycle and survival of oligodendrocytes in the brain, although the activation of these complexes is also involved in the pathogenesis of various CNS diseases via pore formation in the cell membrane causing cell death (Soane et al., 2001). Additionally, C3aR and C5aR are closely involved in CNS development (Dalakas et al., 2020) .

We recently reported the novel role of complement component 8 gamma chain (C8G) in neuroinflammation (Kim et al., 2021b). In that study, we found that C8G has a distinct expression pattern and unexpected function in the brain (Figure 1). C8G belongs to a subset (alpha, beta, and gamma) of the eighth complement component (C8). C8 is part of the MAC, alternatively termed the terminal complement complex, leading to cell lysis and death by forming pores that disrupt the target cell membrane. The roles of C8A and C8B are directly associated with MAC formation, but C8G is not essential for the assembly or activity of the MAC (Parker and Sodetz, 2002). We observed that C8G expression in reactive astrocytes was upregulated by proinflammatory cytokines (mainly interleukin-1β and interleukin-6). These cytokines are major reactants of acute-phase proteins, which contribute to acute responses against pathogens, facilitating complement activation. Interestingly, C8A and C8B were not induced by these cytokines; C8G has an independent induction pattern from that of the other C8 subunits. While human C8A and C8B are located in proximity on chromosome 1p, human C8G is located on the chromosome 9q34.3 harboring a cluster of lipocalin genes, which encode a protein superfamily with immunomodulatory properties, such as lipocalin-2 and orosomucoid 2 (Suk, 2016; Jo et al., 2017). This implies that C8G may be a genetically linked subfamily of lipocalins involved in brain defense as part of an extended neuroinflammation network. Indeed, we observed that the C8G released from astrocytes inhibits microglial activation and modulates neuroinflammation in both lipopolysaccharide (LPS)-induced neuroinflammation and AD mice models. Moreover, an intracerebroventricular injection of a recombinant C8G protein attenuated neuroinflammation and cognitive deficits in AD mice model. Conversely, C8G knockdown led to microglial activation and neuroinflammation. Mechanistically, astrocyte-derived C8G antagonizes the microglial proinflammatory sphingosine 1-phosphate (S1P)/sphingosine 1-phosphate receptor 2 (S1PR2) signal pathway (Figure 1). S1PR2 is a G protein-coupled receptor that binds to the bioactive lipid mediator S1P. It is widely expressed in various tissues, including the immune system, cardiovascular system, and CNS. C8G binds to the N-terminus region of S1PR2 to obstruct or weaken the interaction between S1P and its binding pocket in S1PR2, subsequently inhibiting the RhoA/NF-κB activation pathway in microglia. S1P is generated by the catabolism of sphingomyelin, an abundant component of plasma membranes. Sphingosine kinase 1, which catalyzes the conversion of sphingosine to S1P, is upregulated under inflammatory conditions in the brain (Kim et al., 2021b). Additionally, S1P signaling is closely associated with neuroinflammatory diseases such as AD (Kim et al., 2021b). Therefore, this study led to the discovery of a novel role of C8G as an immunomodulatory factor inhibiting microglial activation in the inflamed brain.

Figure 1.

Novel roles of complement component 8 gamma in the inflamed brain.

Complement component 8 gamma (C8G) has been recognized as a constituent of the classical membrane attack complex in the peripheral organs. However, in the brain, C8G expression is upregulated in reactive astrocytes by proinflammatory acute response proteins, such as IL-1β and IL-6; the C8G induction is independent of other C8 subunits. Astrocyte-derived C8G, also known as immunocalin, suppresses microglial activation and attenuates blood-brain barrier (BBB) disruption by antagonizing the sphingosine 1-phosphate (S1P)/sphingosine 1-phosphate receptor 2 (S1PR2) signal pathway. Therefore, C8G is not only a potential biomarker of neuroinflammation, but also a novel therapeutic target. IL: Interleukin. Created with BioRender.com.

S1PR2 is involved in regulating various physiological processes, such as vascular permeability, immune cell trafficking, and smooth muscle contraction (Cartier and Hla, 2019). In the immune system, S1PR2 regulates immune cell migration and function, while in the vascular system, it regulates vascular tone, blood pressure, and endothelial cell function. The balance between S1PR1 and S1PR2 signaling is crucial for maintaining homeostasis in these processes. Endothelial S1PR1 activation enhances the formation of adherens junctions and tight junctions between endothelial cells, strengthening the endothelial barrier, while S1PR2 activation can have the opposite effect, increasing vascular permeability and contributing to the development of various vascular diseases. S1PR2 activation requires higher S1P concentrations compared to S1PR1 activation; this difference between S1PR1 and S1PR2 may be related to their distinct roles in regulating angiogenesis. Further understanding of the roles and signaling pathways of S1PR1 and S1PR2 is necessary for optimal therapeutic applications.

Because S1PR2 is crucial for both proadhesive and proinflammatory phenotypes of the endothelium and are involved in the modulation of blood-brain barrier (BBB) permeability, we previously investigated whether astrocytic C8G suppresses acute endothelial inflammation by inhibiting the activation of S1PR2 in the brain. We found that the expression of C8G was upregulated in perivascular astrocytes following intraperitoneal injection of LPS and confirmed the localization of S1PR2 in endothelial cells (Kim et al., 2021a). Using an in vitro model of the BBB, we confirmed that recombinant C8G treatment decreases LPS-induced endothelial cell activation and permeability; this was conversely increased by knockdown of C8G. In an LPS-induced neuroinflammation mouse model, intraperitoneal administration of the recombinant C8G protein reduced neutrophil infiltration into the brain. Thus, we found an additional novel function of C8G: protecting the BBB in neuroinflammatory conditions (Figure 1). From these observations, we propose that astrocytic C8G protects the brain by antagonizing S1PR2 in microglia and endothelial cells, indicating that C8G confers neuroprotection through a novel crosstalk between astrocytes and microglia or endothelial cells in the inflamed brain.

Accumulating evidence supports that the activation of the complement system may indicate the status of neurological disorders. For example, C3 and C4 levels highly correlate with aging. C1q, C3d, and C4 have been found in areas surrounding Aβ plaques and increased cerebrospinal fluid levels of these proteins have been detected in the early stages of AD. Upregulation of MAC proteins (C9 or C5b-9) in the early and advanced stages of AD has also been reported (Dalakas et al., 2020) . In our previous study, we found that C8G is also upregulated in the reactive astrocytes of the human AD brain and higher C8G levels were detected in the cerebrospinal fluid and plasma of patients with AD (Kim et al., 2021b). These results indicate that C8G is a potential biomarker of the severity of neuroinflammation in the AD brain. Given the independent induction pattern of astrocytic C8G, it could be a useful biomarker in the CNS. Generally, low levels of anti-inflammatory gene expression are often associated with an increased neuroinflammation and a faster disease progression. However, AD patients with elevated blood levels of anti-inflammatory cytokines, such as IL-4 and IL-10, have been shown to experience more rapid cognitive decline (Park et al., 2020). Possibly, this is because the anti-inflammatory cytokines or proteins, including C8G, may act as compensatory mechanisms during pathogenesis. A strong compensatory response could indicate the severe pathological states. A neuroinflammatory disease progression might be dependent on a balance between pro-inflammatory and anti-inflammatory mechanisms.

The critical roles of innate immune responses and neuroinflammation in neurodegenerative diseases offer potential therapeutic strategies for their treatment. However, despite significant efforts to develop disease-modifying therapeutic interventions against neuroinflammation, most failed to delay or ameliorate neurodegenerative progression. This high failure rate may be due to our limited understanding of neuroinflammation and a lack of appropriate diagnostic/prognostic biomarkers. Accumulating evidence has shown that complement components play a key role in CNS pathology; thus, a therapeutic strategy for neuroinflammatory diseases based on regulating their activation may be useful. However, a deeper and more extensive understanding of this cascade and the roles of complement components is required for successful clinical application. Despite the many unknown aspects surrounding the role of the complement system in the CNS, it can be a valuable target for the development of useful biomarkers and therapeutic interventions.

This work was supported by the National Research Foundation of Korea (NRF) grant funded by the Korean government (Nos. NRF-2017R1A5A2015391, 2020M3E5D9079764, to KS).

Additional file: Open peer review report 1 ^{(81.9KB, pdf)} .