Monocytes Inadequately Fill In for Meningeal Macrophages

Abstract

In the meninges, a unique subset of tissue-resident macrophages serves as a sentinel population against invading microorganisms while suppressing harmful inflammation. A recent report shows that when murine meningeal macrophages are killed by viruses, circulating monocytes repopulate the open niche. However, they fail to mimic all aspects of tissue-resident macrophages.

Tissue-resident macrophages are universally found across mammalian organs, and each subset takes on a different role to manage homeostasis in its unique environment. Even in tissues that are thought to be devoid of other immune cells, such as the central nervous system (CNS), specialized macrophages known as microglia are present that aid in various functions ranging from the removal of cellular debris to the modulation of inflammatory processes. Traditionally, the CNS was considered to be immune-privileged owing to the lack of lymphatic drainage and the presence of the blood–brain barrier. However, recent discoveries have unveiled new immune networks within the meninges – the membranes that enclose the entire CNS [1,2]. Moreover, advances in technology have allowed investigators to probe the vast array of myeloid and lymphoid cell populations within the CNS [3], thus challenging the traditional dogma that the CNS is devoid of immune surveillance. Furthermore, this has fueled an initiative to revisit some precepts regarding immune regulation in the CNS.

A large breadth of work has previously focused on resident microglia and other leukocytes infiltrating the brain parenchyma to further our understanding of CNS immunity. However, less is known about the meninges, which are thought to act as a containment barrier for the cerebrospinal fluid. In mammals, the meninges consist of three layers of protective tissue – the dura, arachnoid, and pia. Recent discoveries have shown that this anatomical compartment functions as more than merely a barrier structure, as it harbors a large network of functional lymphatic vessels that connect the CNS to the immune system [4], and it is where a set of unique innate immune cells – similar to microglia – reside. McGavern and coworkers have led key studies to further our understanding of this distinct compartment, including a recent discovery outlining how distinct myeloid cell subsets can promote meningeal remodeling and repair after mild traumatic brain injury in rodents [5].

Many tissue-resident macrophages, including meningeal macrophages (MMs), originate from the yolk sac, are self-renewing [6], and are distinct from circulating monocytes. However, during inflammation the tissue-resident niche can be repopulated by circulating monocytes, as seen for Kupffer cells in the liver in mice [7]. To examine which cells repopulate the meningeal compartment after viral infection, McGavern and colleagues used an attenuated form of lymphocytic choriomeningitis virus (LCMV), r3LCMV, to cause sublethal meningitis in mice [8]. r3LCMV infection led to MM death after the first week of infection, but by day 30 post-infection the MM niche was completely replaced (Figure 1). To determine whether the compartment was being replenished by surviving noninfected MMs or by circulating monocytes, the authors first generated irradiation-induced chimeras reconstituted with the bone marrow of actin–teal fluorescent protein (TFP) mice, and observed that, following infection, the MM niche was repopulated with bone marrow-derived donor TFP⁺ cells. Chemokine Cx3cr1^CreER/+ × Stop^fl/fl TdTomato mice pulsed with tamoxifen were then used to label tissue-resident MMs for lineage-tracing analysis. This showed that TdTomato-negative cells – which correspond to circulating monocytes – repopulate the meningeal niche following r3LCMV infection. Together, these findings demonstrate that the MM niche can be repopulated by circulating monocytes (Figure 1).

Figure 1. Circulating Monocytes Can Repopulate an Empty Meningeal Macrophage (MM) Niche Following Lymphocytic Choriomeningitis Virus (r3LCMV) Infection in Mice.

Murine MMs (red) reside throughout the meninges where they regulate tissue homeostasis. Day 6 post-infection (p.i.): upon intracerebral infection with r3LCMV (yellow), MMs are infected and killed, leaving an open niche for immune cell repopulation. Days 7–30 p.i.: circulating monocytes (green) fill these niches and start differentiating into MMs in an interferon (IFN)-γ (pink)-dependent manner. Day 30 p.i.: circulating monocytes fully repopulate the empty MM niche and differentiate into MM-like cells, but exhibit increased expression of MHC II/MHC I proteins and reduced expression of immune-sensing and regulatory genes, such as Cd209b and Chnrb4, relative to the original MM population.

Rua et al. [8] then explored the characteristics of this newly engrafted MM population. Using flow cytometry, an increased frequency of a distinct subset of monocytes expressing MHC II and MHC I was noted among the newly engrafted monocyte-derived macrophages (Figure 1). Furthermore, gene microarray analysis showed enriched expression of interferon (IFN)-γ-related genes in these monocyte-derived macrophages. Using mice in which the IFN-γ receptor was deleted in MM as well as in monocyte populations, the repopulating cells had lower frequency of MHC II expressing cells and overall lower expression of MHC I proteins relative to control mice. This indicated that the differentiation of infiltrating monocytes into MM-like cells is mediated by IFN-γ signaling [8]. Because tissue damage during viral infections is often accompanied by induction of IFN-γ, and because this cytokine can also mediate the differentiation of repopulating cells into tissue-resident subsets [8], this may represent an interesting circuit that merits further exploration. For instance, a dedicated subset of T cells or natural killer (NK) cells constituting a source of IFN-γ might potentially promote MM repopulation following viral challenge.

Microarray analysis also revealed that several genes important for inflammatory functions are differentially expressed in these newly engrafted MMs [8]. For instance, the expression of the gene (Cd209b) encoding the microbial sensor SIGNR-1 – that is involved in the recognition of microbial carbohydrates (including lipopolysaccharide, LPS) – was considerably downregulated in MMs following repopulation by monocytes during r3LCMV-mediated meningitis, relative to MMs from uninfected mice. The authors hypothesized that loss of Cd209b might render the meningeal compartment more susceptible to bacterial infections. In line with this, mice previously infected with r3LCMV were challenged with LPS; these were repopulated with monocyte-derived MMs, but presented a defect in neutrophil recruitment into the meninges relative to control mice. This in turn supported the idea that these new MMs might not be as effective in microbial sensing as the original MMs.

Furthermore, the repopulating MMs displayed downregulated expression of a nicotinic cholinergic receptor (Chnrb4) compared with the original MMs. This finding is relevant in that, relative to tissue-resident MMs, these newly engrafted MMs (upon infection) were less able to suppress acetylcholine-induced expression of the interferon-stimulated gene 15 (Isg15) – a surrogate for inflammation – following LPS stimulation [8]. This finding is consistent with previous reports showing that acetylcholine can dampen inflammatory responses in macrophages [9]. Thus, the newly engrafted MMs were defective not only in immune response signaling (Cd209b/SIGNR-1) but, by responding inappropriately to parasympathetic input (acetylcholine), were presumably also unable to quench an active inflammatory response.

Although McGavern and colleagues explored the consequences of these defects by using model inflammatory stimulants, how these defects result in susceptibility to meningeal infections warrants further investigation, including additional in vivo models of bacterial and viral infections. The study also prompts the question of whether other forms of inflammation in the meningeal compartments, such as autoimmune meningitis, irradiation, myeloablative chemotherapy, or head injury, might contribute to shaping future responses to infections.

Collectively, the results from this study elegantly reveal the effect of one form of viral infection in the murine meningeal compartment, and demonstrate that tissue-resident MMs can be replaced by circulating monocytes after infection [8]. This repopulation results in a new tissue-resident macrophage population that has deficits in several immune-related gene expression pathways, including microbial sensing and nicotinic cholinergic receptor function. It will be interesting to determine whether the defects in these repopulating MMs persist for a defined period of time or whether monocyte-derived macrophages are eventually able to regain phenotypes that are similar to their original MMs. This is of particular interest because MM function, if not species-specific, might potentially impact the pathogenesis and/ or disease course of neurodegenerative diseases such as Alzheimer’s disease. Indeed, Alzheimer’s disease has been previously associated with meningeal lymphatic dysfunction in mice [4] as well as with viral infections in humans [10,11]. As a first barrier of defense against infiltrating microbes into the CNS, the meninges represent an under-appreciated compartment with unique immunological properties. The study by Rua et al. [8] thus reveals an important aspect of homeostasis restoration within the meningeal immune system, and gives important new insights into the meninges as a putative ‘immune organ’ that can participate in CNS immune surveillance.