Central Nervous System Co-signaling Pathways
Introduction
The central nervous system (CNS) constitutes a highly specialized environment where immune activation can be detrimental. Traditionally, it was believed that the CNS is immunologically privileged and shielded from the peripheral immune system by the blood-brain barrier (BBB), but it is clear that peripheral immune cells can enter the CNS as part of physiological immune surveillance, recognize cognate antigens, and elicit an adaptive inflammatory response (Korn and Kallies, 2017). Resident CNS cells, in particular activated microglia, play an important role in CNS inflammation as professional antigen presenting cells (APCs) that secrete a vast array of inflammatory mediators, which modulate both innate and adaptive immune responses (Almolda et al., 2015). Many neurological diseases are associated with CNS inflammation, not only the classical inflammatory demyelinating diseases exemplified by multiple sclerosis (MS), but also neurodegenerative disorders such as Alzheimer’s disease (AD), amyotrophic lateral sclerosis (ALS), Parkinson’s disease (PD), and ischemic stroke (Cebrian et al., 2015; Famakin, 2014; Gendelman and Mosley, 2015; Puentes et al., 2016). As neurons are unable to regenerate and often are integrated into highly complex networks, immune responses in the CNS need to be tightly controlled, balancing the need to protect the organism from foreign threats while minimizing immune-mediated bystander damage.
The system of co-signaling pathways provides the immune system with the means to fine tune immune responses by turning on and off immune cell activation. It has long been known that activation of T cells requires two signals: the first consists of the interaction of the T cell receptor (TCR) with its cognate antigen presented in the context of major histocompatibility complex (MHC) molecules on the surface of APCs; and the second consists of the interaction of co-signaling receptors on the T cell with ligands expressed by the APCs (Bretscher and Cohn, 1970). As co-signaling receptors can be both stimulatory, promoting T cell activation, differentiation, and survival, or inhibitory, dampening T cell activation, the net outcome of TCR stimulation is decided by the combination of the different stimulatory and inhibitory signals (Zhang and Vignali, 2016). Similarly, B cells and other immune cells also require two signals for their activation, maturation, and function (Bretscher and Cohn, 1970). In addition to microglia, other resident cells of the CNS including astrocytes, neurons, and neural stem and progenitor cells may also express co-signaling molecules under various conditions providing a mechanism by which CNS cells can interact with the immune system and modulate its effects. In this chapter, we will review the literature on some important co-signaling pathways and discuss their role in mediating and controlling neuroinflammation.
Resident CNS cells
Microglia are the main immunocompetent cells in the CNS and function as the resident macrophages of the CNS parenchyma. They are constantly surveying the CNS parenchyma and are ready to rapidly respond to disturbances in the microenvironment, serving as a first line of defense against infection or injury (Ransohoff and El Khoury, 2015). Under normal conditions, microglia display a quiescent phenotype characterized by the lack or low expression of many molecules normally expressed by other tissue macrophages to prevent the CNS from unwanted immune-mediated inflammation (Perry, 2016). Consistent with this quiescent phenotype, unstimulated microglia express only low or undetectable levels of costimulatory molecules (Almolda et al., 2015; Aloisi et al., 1998). An exception is Tim-3, an inhibitory co-signaling receptor regulating peripheral tolerance and the expansion of effector Th1 cells, preventing uncontrolled inflammation, which is constitutively expressed by microglia isolated from autopsy tissue from subjects with no evident inflammatory disease (Anderson et al., 2007).
Neurons in the local microenvironment control microglial activation and contribute to their quiescent state by secreting soluble factors and through cell-cell interactions (Chavarria and Cardenas, 2013). The expression of the costimulatory molecules CD40 and CD86 is down-regulated in cultured microglia by nerve growth factor (NGF) released by neurons (Wei and Jonakait, 1999). Neurons also express CD200 (Koning et al., 2009), a non-signaling molecule that triggers anti-inflammatory signaling in CD200R expressing cells including microglia (Walker and Lue, 2013). Microglia from CD200 deficient mice exhibited an activated phenotype, while IL-4-mediated neuronal CD200 expression protected against lipopolysaccharide (LPS)-induced microglial activation (Hoek et al., 2000; Lyons et al., 2009).
Consistent with their role as sentinels of the CNS, microglia rapidly become highly activated upon insults to the brain. Depending on the specific stimuli and the condition of the microenvironment, microglia can exert neuroprotective functions or upregulate factors involved in phagocytosis, antigen presentation, and secretion of neurotoxic factors, showing a high degree of plasticity (Ransohoff and Perry, 2009). In vitro experiments have consistently shown that activated microglia can be induced to express costimulatory molecules. The expression of CD40 is upregulated on microglia by proinflammatory cytokines such as IFN- and GM-CSF, while anti-inflammatory cytokines such as TGF- and IL-4 downregulate its expression (Almolda et al., 2015). Activated microglia isolated from the CNS during demyelinating diseases such as MS and experimental autoimmune encephalomyelitis (EAE), its animal model, have been shown to express CD40 (Becher et al., 2001; Ponomarev et al., 2006; Vogel et al., 2013), CD80/CD86 (Issazadeh et al., 1998), and PD-L1 (Ortler et al., 2008; Pittet et al., 2011; Schreiner et al., 2008). CD40 positive microglia can also be detected in the CNS during neurodegenerative disorders such as AD and the SOD1 mouse model of ALS (Okuno et al., 2004; Togo et al., 2000), while CD40 and PD-1 are upregulated in microglia during reperfusion after middle cerebral artery occlusion (MCAO), an animal model of ischemic stroke in mice (Klohs et al., 2008; Ren et al., 2011).
The role of astrocytes in regulating the immune system is less well understood. Astrocytes have a range of functions most importantly maintaining neuronal health (Sofroniew and Vinters, 2010). Reactive astrogliosis, a defensive reaction of astrocytes aimed at limiting tissue damage and restoring homeostasis, is observed in many neurological disorders. Reactive astroglia signal to surrounding cells both by cell-cell interactions and by secreting numerous growth factors, neurotransmitters, cytokines, and chemokines (Pekny et al., 2016). Astrocytes protect the CNS from unwanted adaptive immune responses by the expression of inhibitory co-signaling receptors. In vitro experiments have shown that astrocytes can suppress T cell proliferation and effector functions by upregulation of the inhibitory receptor CTLA-4 (Gimsa et al., 2004). Astrocytes can also be induced in vitro to express PD-L1 and galectin-9 (Gal-9) that bind to the coinhibitory receptors PD-1 and Tim-3, respectively (Magnus et al., 2005; Pittet et al., 2011; Yoshida et al., 2001). Furthermore, PD-L1 and Gal-9 immunoreactivity have been reported in astrocytes in a mouse model of axonal injury, and in MS brain lesions (Anderson et al., 2007; Lipp et al., 2007; Pittet et al., 2011).
Astrocytes may also express CD40L, a ligand for the stimulatory receptor CD40, both during physiologic aging and in AD (Calingasan et al., 2002) showing the complexity of astrocyte responses. In vitro studies have shown inconsistent results in regards to the expression of CD80 and CD86 on astrocytes (Chastain et al., 2011), but it should be noted that both astrocytes and microglia are known to rapidly change their phenotype and gene expression profile when removed from the complex support structure in the brain rendering them difficult to study in vitro (Butovsky et al., 2014). CD80 and CD86 expression was reported in astrocytes during EAE in some models but not others (Cross and Ku, 2000; Issazadeh et al., 1998), and could be observed on astrocytes in chronic active MS lesions (Togo et al., 2000; Zeinstra et al., 2003).
Neurons can directly interact with autoreactive encephalitogenic T cells and convert them to Tregs, either of the traditional Foxp3+ phenotype or a recently described FoxA1+ phenotype (Liu et al., 2014; Liu et al., 2006). FoxA1+ Tregs were originally identified in IFN-β deficient mice and can be induced by IFN-β both in vivo and in vitro (Liu et al., 2014). These cells are characterized by high expression of PD-L1 and can suppress autoreactive T cells in the CNS during EAE in a PD-L1 dependent manner (Liu et al., 2014). PD-L1 expression on neurons is essential for their ability to interact with encephalitogenic T cells and convert them to FoxA1+ Tregs (Liu et al., 2017b). Mechanistic studies showed that endogenous neuronal IFN-β triggers the PI3K/Akt pathway through autocrine signaling resulting in translocation of the transcription factor FoxA1 to the nucleus, inducing PD-L1 expression (Liu et al., 2017b).
Multiple Sclerosis and Experimental Autoimmune Encephalomyelitis
MS is an immune-mediated disease characterized by immune cell infiltration in the CNS and associated inflammation, demyelination, and neuronal degeneration (Dendrou et al., 2015). The initiating factor is still unknown, but evidence suggests that MS, especially in the earlier stages of the disease, is a T cell driven disease where autoreactive T cells are activated and migrate to the CNS, where they trigger an inflammatory response with accumulation of macrophages, T cells, B cells, and plasma cells (Dendrou et al., 2015; Ransohoff et al., 2015). This inflammatory response is accompanied by activation of microglia and astrocytes, as well as damage to myelin and neurons. As the disease progresses, the inflammation becomes more diffuse and sequestered within the CNS with fewer infiltrating cells and more pronounced neurodegeneration (Dendrou et al., 2015). Much insight into the pathogenesis of MS has been gained from its animal model, EAE that is typically initiated by peripheral immunization with myelin proteins/peptides or by adoptive transfer of autoreactive T cells. Peripherally generated pathogenic CD4+ T helper (Th) cells expressing Th1 and Th17 cytokine profiles, cross the BBB and trigger an inflammatory response in the CNS upon encountering their cognate antigens resulting in a model primarily resembling aspects of the early inflammatory phase of MS (Lassmann and Brad, Acta Neuropath, 2017).
All current FDA approved disease-modifying drugs in MS target the immune system, confirming that modulation of inflammation can reduce disease activity in MS. Several co-signaling pathways have been targeted in EAE with positive effects, but none of the molecules tested in clinical trials this far have shown efficacy in MS. Studies of co-signaling molecules in MS and EAE have, however, provided much insight into the role of the immune system in CNS inflammatory diseases and how these diseases are regulated.
