Innate immunity in the central nervous system

Abstract

Immune responses in the CNS are common, despite its perception as a site of immune privilege. These responses can be mediated by resident microglia and astrocytes, which are innate immune cells without direct counterparts in the periphery. Furthermore, CNS immune reactions often take place in virtual isolation from the innate/adaptive immune interplay that characterizes peripheral immunity. However, microglia and astrocytes also engage in significant cross-talk with CNS-infiltrating T cells and other components of the innate immune system. Here we review the cellular and molecular basis of innate immunity in the CNS and discuss what is known about how outcomes of these interactions can lead to resolution of infection, neurodegeneration, or neural repair depending on the context.

Immune privilege: CNS innate immune cells do not phone home

An essential function of innate immunity is to provide the informational input for adaptive immunity. In peripheral organs, innate DCs detect the presence and nature of pathogens (viral, bacterial, or protozoal; intracellular or extracellular) and, through the release of selective mediators, educate T cells about the specifics of pathogen threat. Once the T cell has been informed (primed and polarized), it is directed to the site that harbors the pathogen (1–4). Here other resident or infiltrating innate cells decode the expressed array of T cell cytokines and, in a perfect immunological world, carry out the appropriate host attack on pathogen (Figure 1).

Figure 1. Innate immunity in the periphery and CNS.

(A) In the face of a peripheral infection, innate immune cells prime and instruct T lymphocytes. Tissue DCs internalize microbial protein antigens, process them into peptides, and display them on their surfaces with MHC class II molecules; migrate to draining lymph nodes; and present antigens to naive CD4⁺ T cells. DCs direct the quality of the subsequent inflammatory response by decoding distinctive pathogen-associated signals and transmitting this information to T cells in the form of regulatory cytokines such as IL-12 (for Th1), IL-4 (Th2), or IL-6/TGF-β (Th17). In addition, lymph node environmental cues can provide information about the site of infection (gut, skin, or other). Armed with this information, effector T cells migrate to infected tissues. Upon reactivation, Th1, Th2, and Th17 cells express phenotype-defining cytokines that act on resident and recruited innate cells, which operate collectively with factors such as complement to clear the infection. (B) Resident microglia and astrocytes exert multiple functions in the CNS, including protective and restorative responses to CNS infection or injury. Cytokines and chemokines expressed by resident CNS cells also promote the recruitment of circulating lymphocytes and myeloid cells from the periphery to assist in pathogen clearance. Innate responses in the CNS cannot directly initiate adaptive immunity. Innate CNS reactions also occur during neuroinflammatory disorders and utilize many of the same components as do host defense responses.

Inflammation in the CNS: the role for DCs

DCs play a critical role in initiating T cell responses by taking up protein antigens in tissues, processing them into small peptides and then displaying them on their surface physically associated with MHC class II molecules. DCs migrate through afferent lymphatics to draining lymph nodes and present antigen to naive or memory T cells. Importantly, there is no evidence that DCs with such capacities reside in the healthy CNS parenchyma, nor do CNS resident immune cells prime naive T cells (reviewed in ref. 5). Cells carrying DC surface markers (e.g., CD11b, CD11c) are readily detected in the meningeal coverings of the CNS and in the choroid plexus, the site of cerebrospinal fluid synthesis (6). Although cells with DC markers are abundant in the inflamed CNS parenchyma, they are primarily observed after blood-brain barrier (BBB) disruption, suggesting that many are peripherally derived while others represent resident microglia induced to express such markers by locally expressed cytokines (7–10).

The lack of parenchymal DCs and the fact that no other parenchymal CNS cells fit the operational definition of a DC (antigen uptake, migration to draining lymph nodes, and presentation to naive T cells) constitute the cellular basis of CNS immune privilege. Immune privilege of the CNS, a hoary concept whose definition has become swollen and imprecise over more than six decades can be reduced to two observations: (a) immunogens such as xenografts, viruses, or bacterial lysates fail to elicit adaptive immune responses following non-traumatic micro-injection into the CNS parenchyma and (b) peripheral immunization with the same immunogen leads to a brisk immune response to the CNS depot of antigen.

Why is CNS tissue immune privileged? Two possibilities are salient: (a) robust intrathecal inflammatory reactions can damage delicate, non-regenerating post-mitotic cells such as neurons and oligodendrocytes, suggesting that the lack of adaptive immune responses might confer a survival advantage; and (b) pathogen ingress into the CNS always involves transit from a peripheral site of entry that will first elicit a response in the draining lymph nodes or spleen. Therefore, it would be redundant to endow the CNS with the ability to generate adaptive immune responses de novo.

The BBB has its phylogenetic origin in invertebrates and evolved to provide a precisely calibrated chemical and ionic environment to optimize neuronal function. Yet the BBB is also well suited to restrain CNS inflammation by excluding plasma proteins as well as peripherally derived innate and adaptive immune cells and their associated inflammatory molecules (11, 12). Additionally, the parenchymal CNS environment is anti-inflammatory, featuring high local concentrations of inflammation-suppressive cytokines such as TGF-β and IL-10 and is replete with gangliosides, which can be toxic to T cells (13–17).

Cumulatively, the lack of resident DCs and the relative anti-inflammatory environment of neural tissue lead to innate immune processes that are muted and secluded within the CNS. There is no efficient outward migration of CNS innate immune cells to sound the alarm in lymphoid organs, requiring that resident innate immune cells deal directly with pathogens and tissue damage. Under many circumstances resident cells recruit inflammatory cells from the circulation and interact with these cells to facilitate vigorous inflammatory responses.

Recognizing and responding to microbial pathogens is the cardinal function of innate immune cells. Basic host defense mechanisms are operational in microglia and astrocytes, despite their sequestration within the CNS. Host defense begins with recognition of structural signatures characteristic of pathogens (reviewed in refs. 18–20). Microbial warnings are mediated by pathogen-associated molecular patterns (PAMPs) and include bacterial, viral, and protozoal products (protein, lipid, nucleic acid, carbohydrate). PAMPs are recognized by TLRs, which reside on the plasma membrane or in endosomal compartments (21). In a prototypical scenario, the engagement of TLRs evokes NF-κB activation, resulting in increased transcription of genes encoding IL-1 family cytokines (Figure 2). Pro-forms of resulting peptides, for example pro–IL-1β, remain cytoplasmic until cleaved enzymatically by activated caspase-1, releasing active IL-1β (22).

Figure 2. Innate recognition of infection or tissue injury.

Endogenous DAMPs such as HSP or exogenous bacterial PAMPs activate the innate immune cells of the CNS by engaging cell surface receptors such as TLRs. Ligation of TLRs initiates an intracellular signaling cascade that involves activation of NF-κB and MAPK activity, which leads to formation of transcription factor AP1. Together these components drive transcription of IL-1 family cytokine precursor proteins pro–IL-1β and pro–IL-18. NLRs are activated by cytosolic microbial products, changes in cytosolic pH, or potassium levels often associated with stimulation of the ligand-gated ion channel P2X7R by extracellular ATP. NLRs such as NALP3 are central constituents of inflammasomes, multiprotein complexes that mediate activation of caspase-1, which catalyzes cleavage and maturation of IL-1 family cytokines. Functional domains of inflammasome components include caspase activation and recruitment domains (CARDs), which are present in both caspase-1 and the adapter apoptosis-associated Speck-like protein containing CARD (ASC). PYD, pyrin domain.

Activation of caspase-1 is initiated by signaling from a second set of innate immune receptors, termed “nucleotide-binding domain leucine-rich repeat–containing (LRR-containing) receptors” (NLRs), whose function is dependent on the assembly of large (~700-kDa) complexes termed “inflammasomes” (23, 24). NLRs have been studied extensively in hematopoietic cells including myeloid lineage cells in the CNS such as microglia. Our understanding of specific NLR functions is encumbered by an unwieldy and ever-changing terminology. The largest NLR subfamily (with 14 members), and the one most pertinent for neuroinflammation, is designated the NACHT domain–, LRR domain–, and pyrin domain–containing protein (NALP) family (25). Inflammasomes, defined by their core NALPs, are activated by the cytoplasmic presence of specific microbial components, tissue-injury products, or inflammation-associated metabolic alterations including low cytosolic potassium (26). For NALP3 inflammasomes, effective stimuli include bacterial muramyl dipeptide, bacterial RNA, ATP, and uric acid. Inflammasomes recruit and activate caspase-1, thereby complementing TLR signaling to generate mature IL-1β and IL-18. Another family member, IL-33, is sequestered within cell nuclei, released by cell injury, and inactivated by caspase-1 cleavage (27–29). Along with IL-1α, IL-33 is considered an alarmin (indicator of cell damage) (30).

Dissection of TLR and NLR signals involved the convergence of two distinct lines of research. Toll and spätzle, the index TLR family receptor/ligand pair, were discovered as regulators of Drosophila dorsal-ventral patterning and, later, antifungal immunity (18, 21, 31). NALP3, also known as cryopyrin, was characterized as the mutated gene in autosomal-dominant autoinflammatory disorders such as Muckle-Wells syndrome and familial cold autoinflammatory syndrome, typified by excess IL-1β activity and effectively treated by IL-1β sequestration (22, 25). These pathways were linked by the discovery that cooperative signaling through TLRs and NLRs culminated in secretion of IL-1 family cytokines (32).

The interplay of TLR and NLR signaling effectively protects against pathogens, and both receptor families are expressed in resident CNS cells that participate in innate immunity. Microglia, myeloid cells of the CNS, express all TLRs (33). A more restricted array is expressed in astrocytes (34, 35). Upon pathogen exposure, activated microglia secrete biologically active IL-1β and IL-18 through expression of NLR-mediated inflammasome activity, which in turn elicits production of a secondary inflammatory cytokine cascade by both microglia and astrocytes (23, 36, 37). For example, IL-1β can induce expression of TNF-α and IL-6, while IL-18 stimulates production of IL-17. Inflammatory cytokines also diminish BBB barrier function and enhance recruitment of hematogenous leukocytes (38).

Innate recognition of tissue injury: variation on a theme

TLRs and NLRs are also highly effective at sensing and responding to non-infectious sterile tissue injury, as observed in stroke or trauma (Figure 2). Just as pathogens are detected by virtue of releasing “stranger” signals, so do damaged cells release “danger” signals, designated damage-associated molecular patterns (DAMPs). TLRs and NLRs sense DAMPs: TLR3, TLR7, and TLR9 detect microbial nucleic acids and also those released from necrotic cells (39). TLR2 and TLR4 respond to cellular hsp such as Hsp60, Hsp70, and αB crystallin. NLRs can be activated by endogenous cellular products such as uric acid crystals (as in gouty arthritis) and aggregated peptides (20, 40). ATP from damaged cells activates purinergic receptor-regulated channels to cause cytosolic ion fluxes that are detected by NLRs (41, 42).

Cellular soldiers of CNS innate immunity

Microglia.

Microglia, the archetypal cells of CNS innate immunity (43), are a unique myeloid cell population, derived from the yolk sac during a narrow time window before vascularization or definitive hematopoiesis in the embryo (44). Once established in the CNS parenchyma, microglia are sustained by proliferation of resident progenitors, independent of blood cells (45). In vitro, microglial activation by diverse stimuli (46) induces varied programs of gene expression, yet these gene-expression patterns have not been validated in vivo (47). Activation of microglia is accompanied by morphological changes (Figure 3 and ref. 48). Despite their dissimilar embryonic origins, microglia are related to resident tissue macrophages. Monocyte-derived macrophages are classified as M1, M2a, M2b, and M2c subsets (49, 50). It is plausible that microglia also transcribe context-dependent, activation-related genes that confer unique phenotypes, however the M1/M2 paradigm has not been extended to any tissue-resident macrophages, let alone a population as unusual as microglia. Repurposing techniques including parabiosis (51) might help in accurately defining subsets of microglia (reviewed in ref. 52).

Figure 3. Microglial cells, as described by Pio del Rio-Hortega (161).

(A) Pio del Rio-Hortega (1882–1945), who characterized and named microglial cells. (B) Images of ramified microglial cells drawn by Hortega. (C) Morphological transformation of microglia to phagocytic macrophage. Panels as lettered in C: A, Microglial cell with modestly thickened processes as compared with ramified microglia; B, microglial cell with short, thick processes and enlarged cell body; C, microglial cell with pseudopodia; D, amoeboid microglial cell; E, amoeboid microglial cell with pseudopodia; F, microglial cell with phagocytosed leukocyte; G, microglial cell with numerous phagocytosed erythrocytes; H, microglial cell with lipid inclusions, also termed “foam cell”; I, microglial cell in mitotic division. Reproduced with permission from Physiological Reviews (161).

Systemic inflammation also activates microglia (53–57). Paradoxically, microglial responses to innate stimuli such as systemic LPS show interesting neuroprotective properties in experimental systems. In this paradigm, (stress preconditioning), systemic challenges elicit cytokine responses, which activate microglia and ameliorate injury after subsequent CNS insults including stroke or physical trauma (58–61). The molecular bases and clinical relevance of stress preconditioning remain uncertain.

Chronic neurodegeneration also leads to microglial activation, although the outcome of the activation may be beneficial, deleterious, or neutral. Neurons constitutively express cell-surface and secreted microglial inhibitors; it is conceivable that neuronal cell death or injury removes this suppression (46). If so, the microglial response to neurodegeneration represents a specialized danger signal. Genetic models have unraveled certain microglial contributions to neurodegeneration. In a genetic mouse model of motor neuron disease, targeted deletion of the causative mutant superoxide dismutase gene in microglia remarkably prolonged the lifespan of the mice even though the mutant transgene was still expressed by neurons and astrocytes (62). Targeted ablation of the CX3CR1 chemokine receptor gene (expressed in the CNS only by microglia) modulates microglial reactivity, in most cases increasing cytokine production and effector functions (63). CX3CR1-deficient mice show enhanced amyloid clearance in Alzheimer’s disease (AD) amyloid deposition models (64), consistent with beneficial activation of microglia (52, 63, 65, 66). By contrast, CX3CR1 deficiency worsens toxicity in t hyperphosphorylation models (67) and in models incorporating both pathologies, and complex effects are seen in models incorporating both pathologies (68). Amyloid β, the principal component of AD-associated amyloid, directly activates microglia through the NALP3 inflammasome pathway (36), and subsequent production of IL-1β may worsen neurodegeneration (69).

Astrocytes.

With the exception of microglia and mast cells, CNS resident cells descend from neuroepithelial stem cells and are categorized as neurons and glia, with glia further subdivided into astrocytes, oligodendrocytes, and polydendrocytes. A traditional view holds that glia exist to serve and protect neurons. However, neurons and glia function in intimate interconnections to support every aspect of brain development and function (as reviewed in refs. 70–74). Astrocytes are the best-characterized innate immune neuroglia. The main functions of astrocytes include buffering CNS potassium, removing and recycling potentially toxic glutamate, adjusting water balance, and modulating synaptic activity and blood flow. Astrocytes also produce neurotrophins and anti-inflammatory cytokines such as IL-10 (75).

Upon activation by TLR and NLR signals, astrocytes participate in innate immune reactions and are the principal CNS sources of innate inflammatory mediators, including several complement components, IL-1β, IL-6, and chemokines such as CCL2, CXCL1, CXCL10, and CXCL12 (76–89). Essential homeostatic functions of astrocytes are compromised during inflammatory reactions, potentially worsening outcomes. For example, CXCL12 signaling to astrocytes promotes physiological release of glutamate during synaptic transmission, and also induces release of small amounts of TNF-α. In inflammatory conditions, CXCL12 plus TNF-α signal to microglia to produce large quantities of TNF-α. This cytokine, at high concentrations, impairs the capacity of astrocytes to detoxify glutamate, resulting in neuronal loss through a mechanism termed “excitotoxicity” (79, 90, 91). Microglial-astrocyte interactions are also critical in CNS innate immunity. The deciphering of microglial-astrocyte communication at the molecular level is still in its infancy but already shows promise for identifying interesting therapeutic targets (92, 93).

In a mouse model, the inflammatory transcriptional regulator NF-κB was silenced in astrocytes by transgenic overexpression of a naturally occurring NF-κB inhibitor (94). The blocking of NF-κB signaling in astrocytes showed benefit in disease and injury models — reduced retinal ganglion cell death after ischemic injury; improved recovery from spinal cord trauma, along with increased axonal sparing and regeneration; and lessened inflammation in EAE, a rodent model of the human inflammatory demyelinating disease MS. These findings highlighted the contributions of astrocyte-specific inflammatory signaling for a multitude of CNS pathologies (94–98).

Interactions between innate immune cells and T cells in the CNS

CNS innate immune cells respond to primed T cells and their cytokine directives. Under T cell–mediated inflammatory conditions, the CNS admits large numbers of peripheral innate immune cells. Indeed, CNS infiltration by peripheral cells is critical for protective host defense against infection and for repair after stroke or physical trauma (99–104). However, restraint is required because hematogenous inflammation causes profound damage if the reaction is excessive or inappropriate. The interaction of the CNS innate immune system with infiltrating T cells is typified by MS and EAE (reviewed in refs. 105, 106). EAE can be induced by actively immunizing rodents with myelin protein peptides, which are emulsified in immune-stimulating adjuvants. IFN-γ–producing Th1 cells and IL-17–producing Th17 cells subsequently accumulate in the CNS and initiate demyelination. This immunization protocol also generates T cells that cause disease upon adoptive transfer to naive recipients, a process termed “passive immunization” (107). Myelin-specific CD4⁺ T cells are found in peripheral blood of healthy individuals and in MS patients (108, 109). Clonally expanded and potentially autoreactive CD4⁺ and CD8⁺ T cells have been detected at autopsy in CNS tissues from individuals with MS but not in relevant controls (110, 111). Thus it is likely that, in MS as in EAE, disease-causing T cells are initially activated in peripheral lymphoid organs, where they undergo differentiation and expansion. When autoreactive T cells are reactivated in the CNS by cognate antigen, release of immune mediators facilitates an extensive local inflammatory reaction, which abrogates the trafficking restraints and barrier functions of the BBB. Large-scale CNS infiltration by inflammatory cells and entry of plasma proteins culminates in demyelination, edema, compromise of neural cell function, and neurobehavioral impairment.

How do T cells communicate with resident and infiltrating innate cells?

T cell adoptive transfer experiments demonstrate that encephalitogenicity of Th cells is exquisitely dependent on their production of GM-CSF — even though GM-CSF–deficient Th cells enter CNS and produce other cytokines, disease does not develop. GM-CSF is required to recruit CD11b⁺ myeloid cells, which are thought to sustain local CNS inflammation, mediate direct myelin damage, and support the survival of lymphocytes (112–114).

Microglia express IL-23 and IL-1β, both of which promote GM-CSF expression by T cells (115), implicating microglial–T cell interactions in intrathecal T cell survival and effector function. In transgenic mice, EAE was less severe if microglial responses were impaired (116). Antibodies to GM-CSF are being utilized in clinical trials of inflammatory diseases so that translational applications of this line of research are feasible (117).

Neutrophils

That neutrophils are considered the first line of defense against extracellular and intracellular bacteria is illustrated by life-threatening conditions that result from neutrophil deficiency (118, 119). Neutrophils are rapidly mobilized from the bone marrow in response to signals from CXC family chemokines to mediate pleiotropic functions in immune-inflammatory responses (reviewed in ref. 120). Neutrophils respond to PAMPs and DAMPs through TLRs and NLRs and are also activated by cytokines such as TNF-α and IFN-γ. Once activated, neutrophils upregulate CD15 and CD11b, adhesion molecules that enhance their association with endothelium and migration into tissues (120). Activated neutrophils also produce reactive intermediates through their vigorous respiratory burst and release a plethora of pre-formed mediators: cytokines, chemokines, colony-stimulating and angiogenic factors, lytic enzymes, and antimicrobial peptides. Neutrophils influence lymphocyte migration as well; TNF-α–induced production of CXCL9 and CXCL10 or CCL20 by neutrophils recruits Th1 or Th17 cells, respectively (121–123). Neutrophil-lymphocyte interactions induce survival factors that prolong the lifespan of the short-lived neutrophils. Adding to the inflammatory cascade, T cells recruit neutrophils by secreting IL-17 (124).

Neutrophils are implicated in inflammatory conditions of the CNS. Bacterial meningitis elicits neutrophil infiltration, which is often associated with unfavorable outcomes, potentially because of the severity of the infection (125). Roles of neutrophils in chronic sterile neuroinflammation (as in MS) are under investigation. G-CSF, a growth factor that supports neutrophil activation, worsens MS disease activity (126). Neutrophils are not detected in postmortem MS tissues, nor are there increased neutrophils in the blood or CSF of MS patients (127). By contrast, lesions of neuromyelitis optica (NMO), an autoimmune CNS disease caused by aquaporin 4 antibodies, show abundant neutrophils, which may also be found in CSF during active disease (128). Variable acuity of NMO and MS may contribute to these different findings. NMO lesions are much more destructive and more likely to cause death during acute disease, whereas fatal outcomes of MS occur through complications of immobility after decades of disease. Therefore, the absence of neutrophils in lesions of MS (studied at autopsy) may not be proof of their absence during lesion formation.

Animal models also implicate neutrophil involvement in MS. In EAE, neutrophils are among the earliest CNS-infiltrating cells (129, 130), and neutrophil depletion reduces EAE severity dramatically (131). Furthermore, CXCR2^–/– mice are resistant to EAE induction (131, 132).

Neutrophil influx into the CNS during EAE results from TNF-α production by meningeal mast cells (133). Because neutrophils also promote B cell survival and proliferation (120), innate neutrophils and mast cells might contribute to the B cell follicle–like structures that are found at autopsy in the meninges of MS tissues (134, 135).

Mast cells

Mast cells are myeloid cells defined by c-kit^hi FcεRI⁺ expression and are well known for roles in allergic disease and host defense (136, 137). Mast cells are particularly numerous within tissues exposed to the external environment, such as skin, gut, and respiratory tract, but are also found in brain, spinal cord, and meninges. Classic antimicrobial mast cell responses involve the release of TNF-α and IL-1β (136, 138–140).

Collectively, mast cells comprise a large population of CNS cells, yet they are fixed and widely dispersed, which poses hurdles for direct study. Nevertheless, provocative correlative findings have been reported that implicate these cells in CNS inflammation. Mast cells are present in active MS plaques (141, 142), and mast cell–specific transcripts encoding tryptase and FcεRI are detected in lesions of chronic MS (143). Tryptase and histamine are present in the CSF of MS patients but not healthy individuals (144, 145). Mast cells in the CNS parenchyma likely contribute to local inflammatory responses, and CNS mast cells appear to exert both neuroprotective and damaging effects following concussion injury or stroke (146).

There are limitations to the commonly used experimental models that utilize c-kit–mutant, mast cell–deficient mice (147) for the study of mast cell function, as mast cell development is exquisitely dependent on SCF signaling through c-kit. Mice with reduced SCF signaling due to mutations in the c-kit receptor (W/Wv or Wsh mice) exhibit a loss of mast cells. Mast cells can be reconstituted by systemic or local transfer of bone marrow–derived mast cell precursors in mice harboring c-kit mutations. The c-kit–mutant mice have additional hematologic and developmental abnormalities, and it is therefore essential to use mast cell reconstitution to confirm that the observed phenotypic differences between wild type and Kit mutant mice are mast cell dependent (148, 149). Unfortunately, transferred mast cells fail to reconstitute the brains and spinal cords of c-kit mutant mice, making it challenging to use this model to address the functions of CNS-resident mast cells in health or disease (147).

Initial EAE studies using c-kit^W/Wv or c-kit^Wsh mice employed diverse disease-induction protocols and subjective neurobehavioral scoring, yielding inconsistent and conflicting conclusions about EAE severity in mast cell–deficient mice (150–153).

Objective, quantitative disease severity measures revealed that mast cells were critical for fulminant disease in both chronic (C57BL/6-MOG_35–55–induced) and relapsing remitting (SJL-PLP_139–151–induced) EAE (133, 154). BBB integrity was enhanced and decreased inflammatory cell infiltrates were decreased in c-kit^W/Wv mice with EAE, as compared with wild-type animals (133). Interestingly, TNF-α production by dura mater and pia mater mast cells regulates BBB function as well as T cell and myeloid cell infiltration into the CNS (133), consistent with the idea that these mast cells are protective first responders to microbial CNS challenge, as they are in the periphery.

During the preclinical phase of EAE, T cells interact with APCs and proliferate in the leptomeninges around the spinal cord, suggesting that the leptomeninges is a site of T cell reactivation (155–158). Given that mast cells direct immune cell traffic to peripheral sites of infection (159), it is tempting to speculate that meningeal mast cells may also promote cellular cross-talk between T cells and myeloid cells in EAE. Determination of how CNS parenchymal and meningeal mast cells contribute both to protective CNS immunity and to harmful neuroinflammation now appears critical for identifying effective and comprehensive therapies for CNS diseases. The ongoing development of novel genetic murine models with selective mast cell defects will help toward this goal (160).

Summary

The immune-privileged status of the CNS has evolved to maintain homeostasis required for neural function and host defense. The inability to generate robust and potentially harmful adaptive immune responses therefore requires a primary reliance for host defense on the sequestered and moderate innate responses of microglia, astrocytes, and other resident innate cells. Nonetheless, pathologic neuroinflammation is inherent in all diseases, which disrupt CNS tissue elements, including MS, AD, Parkinson’s disease, stroke, and traumatic brain injury. Our understanding of the interactions between resident and peripheral immune cells, neurons, and glial cells and their implications for host defense, tissue repair, and neurodegeneration is still in its infancy. However, the delineation of the molecular interactions between the immune and nervous systems is proceeding rapidly and will yield translational application in the years to come.