Brain homeostasis is maintained by “danger” signals stimulating a supportive immune response within the brain’s borders

Abstract

An organism’s behavior is determined by the way it senses and perceives the surrounding environment, and by its responses to these stimuli. The major factors known to affect the behavioral response to an event are genetic background, environmental factors, and past experiences, and their imprinting on the relevant brain circuits. Recently, circulating immune cells were introduced as novel players into this system. It was proposed that the brain and circulating immune cells engage in a continuous dialogue that takes place within the brain’s territory, though outside the parenchyma (occurring within the brain’s borders – the choroid plexi, the brain meninges and the cerebrospinal fluid (CSF)). The cytokines secreted by activated leukocytes residing at the borders were shown to affect neurotrophic factors production within the parenchyma. Here, we suggest that such a dialogue is stimulated at the brain’s borders, upon need, by a “danger” signal that originates in the parenchyma in response to any destabilizing event, and discuss the potential role of reactive oxygen species (ROS) in transmitting this signal. Accordingly, a failure to restore balance is likely to lead to aberrant responses to subsequent events. This view thus supports the contention that circulating immune cells are required to maintain the brain’s balanced activity and suggests a novel mechanism whereby the surveying immune cells are sensing the brain’s status and needs.

Introduction

When challenged by an external stimulus, an individual’s behavior is directed towards evaluating the destabilizing potential of the stimulus. The interface between the incoming information and the evaluation process is formed within limbic brain structures, which include the hippocampus, amygdala and prefrontal cortex. These structures integrate the physiological, emotional, and memory components of the individual’s reaction to the stimulus (Sullivan et al., 2006). The two major factors known to determine the perception of an event and the consequential response are genetic background (Binder and Nemeroff, 2010, Jovanovic and Ressler, 2010) and past experiences (McCauley et al., 1997, Sullivan et al., 2006), and their imprinting on the limbic and stress-response systems.

Experiences that are most likely to leave their mark on the wiring patterns of the limbic synaptic systems are those encountered during infancy and early childhood, times which are considered critical for the fine tuning of neuronal wiring (Sullivan et al., 2006). For example, filial imprinting, the process by which an emotional bond to the mother or caregiver is formed, causes changes in synaptic connectivity in prefrontal forebrain regions (Sullivan et al., 2006). While experience-dependent fine-tuning provides an optimal adaptation of the brain to a given environment, when synaptic reorganization is driven by an adverse environment, it could result in “defective” synaptic wiring that will cause aberrant behavior throughout life (Andersen and Teicher, 2004). In addition, traumatic experiences at any stage were shown to cause an over sensitization of the central stress response system (McGuire et al., 2010, Coplan et al., 1996, Bhatnagar and Dallman, 1998, Ulrich-Lai et al., 2007, Heim et al., 2008, Zoladz et al., 2008). Normally, the adaptive stress response is coordinated by secretion of two neuropeptides: corticotropin-releasing hormone (CRH) and vasopressin (AVP), which are secreted by the hypothalamus and activate the HPA axis, resulting in secretion of corticosteroid hormones. The spread of corticosteroids through the circulation allows the coordination of brain and somatic functions that are geared towards coping with stress, recovery and adaptation (reviewed in: (de Kloet et al., 2005)). Hypersensitivity of the HPA axis due to traumatic childhood experience increases the risk of developing depression later in life in response to stress (Heim et al., 2008). Similarly, patients with post traumatic stress disorder (PTSD) show elevated CRH levels in their cerebro-spinal fluids (CSF) (Bremner et al., 1997), and enhanced secretion of cortisol following a traumatic event (Elzinga et al., 2003).

Regardless of past mental experiences or episodes, evidence suggests that immune related dysfunction, either congenital or as a result of a postnatal event, could, by itself, lead to behavioral, mental or cognitive malfunctions at adulthood. For example, strong activation of an immune response during pregnancy was shown to cause a persistent immune abnormality in the offspring (Mandal et al., 2010, Yamaguchi et al., 1983, Fujii and Yamaguchi, 1992, Cardon et al., 2010). These offspring show increased susceptibility to various mental disorders, such as schizophrenia, and autism (Ciaranello and Ciaranello, 1995, Shi et al., 2003, Brown, 2006). Similarly, infection during early childhood is correlated with subsequent development of Tourette syndrome (Church et al., 2003). A recent study by our group demonstrated that congenital immune deficiency in mice causes abnormal sensorimotor gating (Cardon et al., 2010), an activity that is impaired in schizophrenia (Swerdlow et al., 2006), as well as in other neuropsychiatric disorders (Swerdlow et al., 1993, Swerdlow et al., 1995; Castellanos et al., 1996). Another example connecting immune profile to mental health is the case of post traumatic stress disorder (PTSD), in which a specific gene expression pattern in peripheral blood mononuclear cells of patients hospitalized immediately following the experience of a stressful event predicted emergence of PTSD (Segman et al., 2005). One possible explanation for these findings is that such immune abnormalities, similar to the defective neuroendocrine stress response, impair the individual’s ability to cope with stressful life events, thus increasing the susceptibility to develop behavioral abnormalities.

Several studies done in our laboratory over the last few years provided the basis for the hypothesis that links adaptation to stress with the presence/activity of circulating immune cells by showing that the presence of a functional adaptive immune system at the time of exposure to mental stress reduces susceptibility to post-traumatic behavioral abnormalities (Cohen et al., 2006, Lewitus et al., 2008, Lewitus et al., 2009). In this perspective article, we suggest that protective immune-derived factors are produced by the circulating immune cells within the brain’s borders in response to an alarm signal that is emitted by the stimulated brain (Fig. 2).

The first line of immune defense within the CNS is mediated by microglia, which are the resident macrophages of the CNS parenchyma (McKercher et al., 1996, Ransohoff and Cardona, 2010). Although previously referred to as “resting” cells, it is now becoming clear that microglia continuously sample their environment to monitor changes in CNS homeostasis (Nimmerjahn et al., 2005), and rapidly respond to threats (Davalos et al., 2005). It has been suggested that microglia are able to polarize their activation state to achieve the appropriate responses to varying challenges (Ransohoff and Perry, 2009), in a manner similar to peripheral macrophages (Geissmann et al., 2010, Martinez et al., 2009).

In addition to the monitoring performed by resident microglia, systemic immune cells are also engage in a constant immune surveillance of the CNS that takes place primarily within the CSF. Such immune surveillance is carried out primarily by memory T cells, which were shown to migrate from the blood to the CSF through the choroid plexus and meninges, and comprise 80% of the cells in the CSF of healthy individuals (Ransohoff et al., 2003). Furthermore, the continuous circulation of T cells between the periphery and the CNS was demonstrated. T cells re-enter the bloodstream from the CSF and are replaced by new lymphocytes approximately every 12 h (Kivisakk et al., 2003, Ransohoff et al., 2003, Engelhardt and Ransohoff, 2005, Reboldi et al., 2009). The initial encounter of naïve T cells with neuroantigens occurs mainly in the peripheral lymphatic organs. Despite the lack of classical lymphatic vessels in the CNS, there are indications for drainage of CNS antigens to the CSF circulatory pathway, finally reaching the deep cervical lymph nodes, the nasal lymphatics and the spleen, via the circulation (Ransohoff et al., 2003, Engelhardt and Ransohoff, 2005). While circulating within the CSF, the T cells can be reactivated by encountering their cognate antigen presented by APCs that populate the choroid plexus, the CNS meninges, and the perivascular and sub-arachnoid spaces (McMenamin et al., 2003, Ransohoff et al., 2003, Kawakami et al., 2004).

The entrance of T cells through the choroid plexus into the territory of the CNS was primarily investigated to explain the initiation of neuroinflammation. It was demonstrated that the first wave of encephalitogenic T cells enter the CNS through the choroid plexus (Reboldi et al., 2009). Their subsequent reactivation in the CSF induces expression of adhesion molecules on the cerebral blood vessel endothelium, enabling the penetration of a second wave of T cells to the parenchyma (Bartholomaus et al., 2009, Reboldi et al., 2009). Several molecules that are expressed on the choroid plexus epithelium were found to mediate the penetration of T cells, including CCL20, which binds to CCR6 on the T cells (Reboldi et al., 2009), CD73 (Mills et al., 2008), and P-selectin (Kivisakk et al., 2003). In the following section we will discuss recent lines of evidence suggesting a fundamental role for these surveying T cells in supporting normal brain function and plasticity.

The need for systemic immune cells to support brain function and plasticity was first demonstrated using immune deficient mice, and mice lacking specific immune-cell populations. It was found that adult neurogenesis, neurotrophic factor production, and hippocampus-dependent functions such as spatial memory and sensorimotor gating are all dependent on immune cell availability (Kipnis et al., 2004, Ziv et al., 2006, Brynskikh et al., 2008, Ron-Harel et al., 2008, Wolf et al., 2009a, Wolf et al., 2009b, Cardon et al., 2010). Apparently, the immune-brain dialogue needed for normal brain function under physiological conditions occurs within the brain’s territory and is mediated by the CNS-surveying T cells: A recently published study demonstrated that hippocampus-dependent cognitive ability is specifically supported by the T cells that reside within the brain meninges (Derecki et al., 2010). The “long-distance” communication between the relevant brain structures (i.e. hippocampus) and the T cells might be mediated by cytokines that are secreted in the meninges and arrive at the parenchyma through the CSF. Specifically, it was shown that IL-4, which is secreted by meningeal resident T cells, supports spatial memory performance by induction of BDNF production in the hippocampus (Derecki et al., 2010). The need for immune-cell activity to ensure normal brain function is ongoing: Sudden T cell depletion in young healthy adult mice causes cognitive impairment, whereas immune reconstitution restores spatial memory abilities in immune deficient mice (Brynskikh et al., 2008, Ron-Harel et al., 2008). The specific cells that support normal cognitive performance and neurogenesis are CNS-specific autoreactive CD4⁺ T cells. This observation was based on the comparison made between: Tova-transgenic mice, in which the majority of the T cell population is specific to the non-self antigen (ovalbumin), and Tmbp-transgenic mice, in which the majority of the T cell population is specific to the CNS antigen myelin basic protein (MBP). While Tmbp-transgenic mice showed normal cognitive ability and increased hippocampal neurogenesis, the Tova-transgenic mice resembled immune deficient mice in their impaired cognitive ability, and their reduced hippocampal neurogenesis (Ziv et al., 2006).

Increased T cell numbers are found in the borders of the CNS following increased brain activity (i.e. cognitive testing (Derecki et al., 2010) and acute mental stress (Lewitus et al., 2008)). These findings, taken together with the contribution of systemic immune cells to successful coping with mental stress (Cohen et al., 2006), raise the question as to whether local cytokine secretion by T cells within the brain’s territory (in the choroid plexus/meninges/CSF) can restore balance following any mental activity, thereby ensuring an adequate behavioral response to subsequent stimuli (Fig. 2).