Microglial Activation & Chronic Neurodegeneration

Abstract

Microglia, the resident innate immune cells in the brain, have long been implicated in the pathology of neurodegenerative diseases. Accumulating evidence points to activated microglia as a chronic source of multiple neurotoxic factors, including TNFα, NO, IL1-β, and reactive oxygen species (ROS), driving progressive neuron damage. Microglia can become chronically activated by either a single stimulus (ex. LPS or neuron damage) or multiple stimuli exposures to result in cumulative neuronal loss over time. While the mechanisms driving these phenomena are just beginning to be understood, reactive microgliosis (the microglial response to neuron damage) and ROS have been implicated as key mechanisms of chronic and neurotoxic microglial activation, particularly in the case of Parkinson’s Disease. Here, we review the mechanisms of neurotoxicity associated with chronic microglial activation and discuss the role of neuronal death and microglial ROS driving the chronic and toxic microglial phenotype.

Neurodegenerative diseases (ex. Alzheimer’s Disease, Parkinson’s Disease, Huntington’s Disease, ALS, etc.) share many common characteristics, such as changes in microglial number and morphlogy, elevated cytokine levels, oxidative stress, and progressive neuronal loss. Increasing evidence reports that microglia can become a chronic source of cytokines and reactive oxygen (ROS) to drive progressive neuron damage and are implicated in the chronic nature of neurodegenerative diseases.¹ Here, while predominantly focusing on Parkinson’s disease (PD), we discuss mechanisms through which microglia can become neurotoxic, explain current views on why the microglial response is chronic, and discuss the meaning of these findings in respect to the progressive nature of neurodegenerative disease.

Microglial Origin & Maintenance in the Central Nervous System (CNS)

Microglia reside in the CNS, comprise approximately 12% of the brain² (depending on brain region, health or pathology), and serve as the brain’s immune defense. Microglia are unique from neurons, oligodendrocytes, and astrocytes in that they are not derived from the neuroectoderm. Instead, it is generally accepted that the original microglial population in the CNS differentiates from cells of the myeloid lineage that originate in the bone marrow², which occurs early in embryonic development.³ These myeloid cells can be found within the CNS by embryonic day 8 in rodents⁴, and by the 12^th week of gestation in humans.⁵ At this stage, the cells are referred to as fetal macrophages, and are the earliest form of microglial precursor cells present in the embryonic CNS.

Once fetal macrophages come to reside in the developing CNS, they begin the differentiation process that will result in the formation of fully-matured microglia. Although the course of this differentiation is not fully understood, one of the early steps is the formation of rounded ‘ameboid’ microglia that cluster within distinct anatomical regions in the developing brain and may act as a source of microglial progenitors.⁶ Later in embryonic and early fetal development, these progenitors will follow a path of migration and differentiation leading to the mature microglia, a process that extends into neonatal development. Differentiation first involves the formation of partially ramified microglia followed by the development of fully ramified, or branched, microglia that express cell surface molecules characteristic of resting microglia (discussed below).⁷

While the origin of initial microglia populations within the CNS has been well supported, the replacement of microglial populations is matter of more debate. Due to the presence of the blood brain barrier, it was originally perceived that the circulating immune cells did not have immediate access to the CNS, keeping populations of microglia distinct from similar, circulating blood cells. However, there is increasing evidence that bone-marrow derived cells are capable of entering the CNS and differentiating into microglia in adults.⁸ For example, studies using bone marrow-chimeras have shown that circulating monocytes infiltrate the CNS under different conditions of injury, inflammation, or disease.^9–11 This has been shown to be possible even when the blood brain barrier remains intact, suggesting a mechanism for entry into the CNS.⁹ A possible mechanism for this could be replenishment from perivascular cells, which are bone marrow derived and have been shown to enter the CNS and differentiate into microglia.^{12, 13} The mechanisms through which circulating cells are recruited to the CNS, and whether they enter the CNS under normal, resting conditions, however, are poorly understood. At present, there is also ongoing debate regarding the function of infiltrating cells vs. resident microglia, where it has been suggested that infiltrating precursor cells may be predominantly beneficial cellular actors of wound healing¹⁴ and an under-utilized therapeutic resource.⁸

What also remains unclear is whether microglia are capable of a level of self-renewal that is sufficient to support the population of microglia in both resting and activated states. Microglia have a low mitotic rate when at rest, but are capable of high rates of proliferation when activated, suggesting that they have at least a partial ability to counteract cell-turnover.^{15, 16} In addition, populations of microglial progenitors have been proposed to persist in the adult brain that are capable of proliferating to replace microglial populations.^{4, 17} At present, we are just beginning to understand these basic principles regarding the origin and replacement of microglia and further research is needed to fully elucidate the mechanisms involved in the complicated life-cycle of the immune system of the CNS.

Microglial Activation & Function

Resting Microglia

Analogous to the role of macrophages and lymphocytes in the periphery, one role of microglia is to act as the brain’s immune defense against disease and injury. In addition to these duties, however, microglia are involved in number of processes in the normal, healthy CNS. In a normal brain, microglia are said to be resting, and can be distinguished by both their morphology and pattern of gene expression. In this state, microglia take on a ramified appearance, where long thin processes extend from the cell body and into the surrounding milieu. The immunological phenotype of this state is characterized by low expression of major histocompatability complex (MHC) proteins and other antigen presenting surface receptors.¹⁸ This phenotype is in stark contrast to that of other macrophages, which exist in a more readily activated state. One of the reasons for this may be the absence of serum proteins in the brain that have been shown to cause macrophage activation.¹⁹ In addition, the expression of certain receptors, such as CD200 and CX3CR1 on the microglia cell surface, may interact with ligands that keep microglia in a resting state.^{20, 21} Resting, ramified, microglia cell bodies are spaced throughout the CNS to avoid cell body overlap, but have been shown to be present with variable density in different brain regions.²² Notably, this ramified morphology occurs only in vivo and is relatively absent from microglia in cell cultures.

Despite the fact that these ramified microglia are referred to as resting, they are constantly surveying the surrounding environment by extending and retracting their processes.^{23, 24} In doing this, microglia are able to sample the microenvironment, maintain homeostasis, and identify signals that require a response. When reacting to extracellular signals, such as the presence of pathogens, foreign material and dead or dying cells, microglia may undergo a morphological change into an ameboid shape with short or non-existent processes.²⁵ This morphological change is also accompanied by changes in signaling and gene expression that can result in changes in surface receptor expression, the release of pro- or anti-inflammatory factors, recruitment molecules, and ROS, among others.^26–29 The cumulative effect of these changes in morphology and phenotype is a shift from resting to activated microglia.

Microglial Activation & Function in the Healthy CNS

Microglia are activated in response to brain injuries and immunological stimuli^{25, 30–32} to undergo dramatic alterations in morphology, changing from resting, ramified microglia into activated, amoeboid microglia²⁵, which is thought to favor phagocytosis and mobility. Unfortunately, changes in morphology are unlikely to differentiate between benign and toxic microglial activation.³³

In non-pathological states, microglia respond to extracellular stimuli in a number of ways. In the developing brain, and in areas of remodeling, microglia are responsible for the phagocytosis of cellular debris resulting from apoptosis and normal cell death.¹⁸ Microglia are also responsible for the phagocytosis of other debris present in the extracellular space, including damaged cells, plaques, and foreign matter. For microglia surrounding neurons, subtypes of microglia can provide trophic support to neurons through the release of nerve growth factors, neurotrophins, and other neurotrophic factors.³⁴ These microglia are also capable of assisting in synaptic plasticity, an observation that was first made in the mid-20^th century.³⁵ Along these lines, microglia have been implicated as the “brains electricians”,³⁶ where the release of neurotrophic factors and anti-inflammatory cytokines from microglia has been shown to promote synaptic plasticity.^37–39 Specifically in response to injury, activated microglia have been shown to surround damaged neurons and participate in synaptic stripping, a process of removing branches from damaged neurons to promote repair and regrowth,^{35, 40, 41} although recent evidence shows that microglia may play an indirect, anti-inflammatory role in this process.⁴² Notably, these beneficial microglial functions often involve changes resembling an activated morphology and protein expression, yet the function is distinct from a classic pro-inflammatory response.

In fact, the majority of microglial functions are beneficial and necessary for a healthy CNS, as activated microglia are critical for CNS wound healing.³⁶ For example, ablation of infiltrating bone marrow derived cells (that become microglia) in spinal cord injury at specific times has been shown to have disastrous neurotoxic consequences.¹⁴ In addition, microglia have also been shown to release anti-inflammatory and trophic molecules to enhance the survival of surrounding neurons.^{43, 44} Thus, evidence supports that when microglia become neurotoxic, this is due to both the loss of the beneficial functions³⁶ and/or a shift to a pro-inflammatory phenotype.^{1, 45, 46}

Pro-inflammatory Microglial Activation

Microglia detect and respond to pro-inflammatory triggers by changing to an activated phenotype, resulting in a shift of cellular function to releasing cytotoxic factors (ex. TNFα, NO, & ROS) aimed at destroying the invading pathogens. Surface molecules associated with the innate immune response, such as complement receptors and MHC molecules, are also upregulated upon microglial activation.^{47, 48} For example, upregulation of MHC proteins enable microglia to act as antigen-presenting cells to T-cells that will enter the brain during active infections.¹⁸

Microglia are readily activated by an extensive list of pro-inflammatory stimuli, such as lipopolysaccharide (LPS),^49–51 pesticides (ex. Paraquat,⁵² dieldrin,^{53, 54} lindane,⁵⁴ and rotenone⁵⁵), disease proteins (ex. Beta-amyloid (Aβ),⁵⁶ α synuclein, ⁵⁷ and HIV-Tat⁵⁸), air pollution,⁵⁹ and even neuron damage.⁴⁶ In fact, it has been proposed that many disease proteins and environmental toxicants trigger a toxic microglial response because they are misinterpreted as a pathogen.¹ In response to these triggers, microglia can produce of a large array of cytotoxic factors, such as superoxide (O₂^•−),²⁷ nitric oxide (NO),^{60, 61} tumor necrosis factor alpha (TNFα)^{62, 63} and inflammatory prostaglandins.⁶⁴ While the detailed components of the microglial response can be stimulus specific, there are common factors of microglial activation.⁴⁶

LPS is a cell wall component of gram negative bacteria and is a potent stimulus of the microglial innate immune response. The microglial response to LPS has been well characterized and provides excellent insight into the timing of the multiple factors produced in a pro-inflammatory response. Notably, extracellular superoxide (O₂^•−) is an immediate toxic factor released by microglia in response to LPS. Interestingly, unique to microglia, LPS-induced production of O₂^•− is not mediated through the traditional LPS receptor, the toll-like 4 receptor (TLR4).⁵⁰ In fact, recent work has indicated that the MAC-1 receptor is responsible for the LPS-induced activation of NADPH oxidase and the consequent production of O₂^•− in microglia.⁴⁹ Thus, the microglial (O₂^•−) response is typical (although less robust) when compared to other professional phagocytes, but the mechanisms mediating this response may be cell-specific.

While the microglial superoxide response is immediate, microglia also respond to LPS with a later pulse of pro-inflammatory factors, such as TNFα, IL1-β, and NO, both in vivo and in vitro.⁶⁵ Specifically, there is a delay in the production of TNFα, NO, prostaglandin E2 (PGE₂) and IL-1β, where TNF-α peaks at 6 hours, NO peaks at 12 hours and both PGE₂ and IL-1β peak at approximately 24 hours.⁶⁵ Thus, one mechanism through which microglia are thought to cause neuron damage is through the excessive and inappropriate release of toxic factors.

Microglia in Disease

Neuroinflammation is characteristic of many neurodegenerative diseases and microglia have been implicated to play both causative and exacerbating roles. In fact, microglia and inflammation-mediated neurodegeneration has been implicated in numerous other diseases and conditions, such as hypoxia,⁶⁶ stroke,⁶⁷ and neuropathic pain.⁶⁸ Neurodegenerative diseases are characterized by chronic and progressive neuronal loss, pathological levels of cytotoxic substances such as extracellular debris, elevated levels of pro-inflammatory factors, and production of reactive oxygen species, resulting in oxidative stress. These factors, in addition to the release of others that can activate and recruit microglia, support a role for microglia in diseases such as Alzheimer’s disease,⁶⁹ Parkinson’s disease,⁶⁹ multiple sclerosis,⁷⁰ amyotrophic lateral sclerosis,^{71, 72} and HIV-associated neurocognitive disorder (HAND).⁷³

Although microglia undergo changes as a result of normal aging, including increases in activation or decreases in function and proliferation, ⁷⁴ the activation of microglia in age-dependent neurodegenerative diseases, such as Alzheimer’s disease and Parkinson’s disease, seems to be a distinct process.⁷⁵ One of the hallmarks of Alzheimer’s disease pathology is the existence of β-amyloid plaques, an extracellular protein aggregate that is normally cleared by microglia. Activated microglia and their toxic effects have been associated with Alzheimer’s disease for decades.^76–78 This has lead to research showing that not only is β-amyloid directly toxic to neurons,⁷⁹ but it also causes microglia to cluster around plaques and become activated, which may perpetuate neuronal damage and death.^{80, 81} Similarly, activated microglia are associated with damaged neurons in patients with Parkinson’s disease,⁶⁹ which is discussed in more detail below.

In contrast to diseases specifically associated with aging, microglial activation can also play a role in diseases not related to age and may involve unique properties of microglial cells. A strong example of this is in multiple sclerosis, a disease associated with severe inflammation and demyelination of axons. Usually considered an autoimmune disease, multiple sclerosis is associated with lesions within the white and gray matter of the CNS that have increased levels of activated microglia.^{82, 83} In addition to increases in microglia-released ROS and pro-inflammatory cytokines,^84–86 microglia may play a large part in the initiation of disease by acting as antigen-presenting cells targeting myelin.⁸⁷ In HAND, microglia play a strong part in harboring the HIV virus and acting as a site of viral replication.^{88, 89} During this process, microglia become activated and release pro-inflammatory cytokines.⁹⁰ In addition, microglia are activated by HIV proteins, such as Tat,⁹¹ to produce ROS. Thus, the interaction between microglia, viral replication/proteins and the production of cytotoxic factors in HIV-associated neurocognitive disorder have strong implications for disease progression.

These neurodegenerative diseases, as well as others, such as amyotrophic lateral sclerosis, Huntington’s disease and prion diseases, highlight the role that activated microglia can play in cell damage and death. In cases of these pathologies, however, the exact role of microglia remains controversial. Ongoing research seeks to answer questions pertaining to microglial activation in both the development and progression of neurodegeneration.

Microglia Activation and Parkinson’s Disease

In contrast to the beneficial housekeeping duties of resting and moderately activated microglia, over-activation of microglia resulting in excess production of inflammatory mediators is in fact neurotoxic ^92–94 and microglial activation has been strongly linked to pathology in PD.^{1, 95} The term over-activation describes the state in which microglia continually produce inflammatory mediators which accumulate to levels that are harmful to neurons,^96–98 and often in combination, lead to neurodegeneration.^{94, 98, 99}

Pioneering work by McGeer and colleagues⁶⁹ discovered increased HLD-DR staining in the substantia nigra (SN) of post mortem PD patient brains, indicating the presence of activated microglia, and first implicating these cells may have an active pathological role in disease. Since then, many inflammatory mediators such as TNFα, IL-1β, and interleukin 6 (IL-6) have been identified in the striatum in human PD post-mortem brains^100–104 in addition to an upregulation of inducible nitric oxide synthase (iNOS) and cyclooxygenase 2 (COX2) in ameboid microglia located in the SN of PD patients,¹⁰⁵ further suggesting a link between activated microglial cells and neuronal damage in disease. Research has identified a critical role for neuroinflammation in dopaminergic (DA) neuron damage, as cytokines such as TNFα^106–108 have been shown to exert DA neuron damage. In fact, continuous expression of low levels of TNFα in the SN induced by an adenoviral vector will cause chronic microglia/macrophage activation, progressive neurodegeneration, and delayed motor symptoms.¹⁰⁹ Thus, it is not surprising that anti-inflammatory approaches,^{95, 110–112} including those focusing on TNFα,^{99, 113, 114} have been targeted as therapeutic strategies.

Subsequent studies provide a wealth of evidence confirming the presence of activated microglia in PD and PD-like brains. For instance, 1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine (MPTP) is an illicit drug contaminant linked to human Parkinsonism cases¹¹⁵ that is commonly used as an animal model of Parkinson’s disease.¹¹⁶ After MPTP administration, microglial activation in the SN of mouse brain has been confirmed by an increase in cell number, changed cell morphology, increased lectin staining, larger cell bodies, and thicker processes.^117,118 Activated microglia are also present in the brains of MPTP intoxicated monkeys.^{119, 120} Using PET imaging, a 6-hydroxydomapine (6-OHDA) model of DA neuron damage showed decreased binding of C-PK11195 in the striatum, indicating increased activation of microglia.¹²¹ Importantly, these findings parallel what is seen in the human PD post-mortem brain.^{100, 122, 123} More recently, PET imaging has shown that over-activated microglia are also present in the SN of living PD patients and are associated with DA neuron damage.^{124, 125} Thus, evidence strongly supports that microglia may play a role in the active pathology driving Parkinson’s disease.

Microglia-mediated Dopaminergic Neuron Damage

Several studies using animal models have demonstrated that the presence of microglia and the production of neurotoxic factors can initiate and amplify neuron damage. Interestingly, multiple factors and toxins are shown to selectively damage DA neurons through microglial activation, such as rotenone,¹²⁶ diesel exhaust particles (DEP),¹²⁷ paraquat (PQ),⁵² α-synuclein,⁵⁷ and the beta amyloid peptide (Aβ).¹²⁸ However, early studies using the immunogen LPS, demonstrate that a pro-inflammatory trigger such as LPS is toxic to DA neurons only in the presence of microglia,^{127, 129, 130} where LPS was one of the first microglia-mediated selective DA toxins identified. At present, it is unclear why DA neurons are more vulnerable to microglial activation when compared to other cells types, but inherent vulnerability to oxidative stress has been implicated.¹

Microglial NADPH Oxidase, ROS, and Neurotoxicity

Redox Signaling

Unregulated, excessive ROS can indiscriminately damage biomolecules (ex. protein carbonyls) to impair cellular function, a mechanism through which ROS is thought to contribute to neurodegenerative disease.¹³¹ However, in addition to permanent damage, ROS is capable of a more elegant and selective modification of proteins, where ROS is known to target thiol functional groups on cysteine amino acid residues.¹³¹ These reversible modifications can be likened to phosphorylation, and can regulate protein function, acting as an important process of signal transduction in multiple cell types, including microglia.

Microglial ROS Production & NADPH Oxidase

The production of ROS in phagocytes is derived from multiple sources, such as peroxidases inside the cell, NADPH oxidase on the membrane surface, or the oxidative processes of mitochondria.¹³² NADPH oxidase is a multi-subunit enzyme that catalyzes the production of O₂^•− from molecular oxygen and is the predominant source of extracellular ROS in phagocytes, such as microglia.^{129, 130} The enzyme complex is dormant in resting phagocytes and can be activated upon exposure to specific stimuli, such as bacteria and inflammatory peptides¹³³ In resting cells, the cytosolic subunits of NADPH oxidase (p47, p67, p40, and Rac2) are distributed between the cytosol and the membranes of intracellular vesicles and organelles.¹³³ However, activation induces the cytosolic subunits to translocate to the membranes, where they bind to the membrane-associated subunits (p22 and gp91), assembling the active oxidase that produces O₂^•−.¹³³

NADPH oxidase and intracellular ROS have been implicated in several cellular functions such as migration¹³⁴ and phagocytosis.¹³⁵ Specific to microglia, ROS produced from NADPH oxidase has been shown to mediate changes in microglia morphology, ¹³⁰ pro-inflammatory gene expression,¹³⁰ and upregulation of markers in response to immunological stimuli.¹³⁶

Interestingly, NADPH oxidase is upregulated in neurodegenerative diseases such as PD¹³⁷ and Alzheimer’s disease,¹³⁸ indicating a potential role for microglial NADPH oxidase activation in disease and neuron death. In fact, the critical role of NADPH oxidase in mediating inflammation-related neurotoxicity has been well documented,¹³⁰ where the LPS-induced loss of nigral DA neurons in vivo and in vitro was significantly less pronounced in NADPH oxidase-deficient mice, when compared to control mice. NADPH oxidase has also been linked to microglia-derived oxidative stress from a variety of neurotoxic insults, such as rotenone, ¹²⁶ DEP,¹²⁷ α-synuclein,⁵⁷ Aβ,¹²⁸ PQ,¹³⁹ dieldrin,⁵⁴ DA neuronal injury,^{137, 140} prothrombin kringle-2,¹⁴¹ β2 adrenergic receptor activation,¹⁴² angiotensin,¹⁴³ and cerebral ischemia-reperfusion injury,^{144, 145} indicating that microglial NADPH oxidase activation may also be a common denominator of microglial activation associated with neurotoxicity. Currently, the precise species of ROS responsible for NADPH oxidase-induced neurotoxicity is unknown. However, SOD and catalase mimetics, which remove O₂^•− and hydrogen peroxide (H₂O₂), respectively_, reduce LPS-induced DA toxicity¹⁴⁶ indicating the critical importance of H₂O₂ and O₂^•− in microglia-mediated neurotoxicity. Thus, microglial-derived ROS may be an essential and common factor of toxic microglial activation.

NADPH Oxidase & Redox Signaling

The phagocytic oxidase (PHOX)-ROS signaling pathway is the signaling mechanism induced by the increase in intracellular ROS in phagocytes as a response to NADPH oxidase activation (pro-inflammatory redox signaling in phaogocytes). The increase in intracellular ROS in phagocytes, such as microglia, includes a number of oxygen species, such as O₂^•−, hydroxyl radical (OH^•−), lipid hydroperoxides, and their by-products (e.g., H₂O₂).¹⁴⁷ While multiple cellular organelles and enzymes contribute to intracellular ROS, it is not surprising that the amount of intracellular ROS produced by NADPH oxidase is dependent upon the stimuli. For example, NADPH oxidase contributes to approximately 50% of the LPS-induced intracellular ROS increase¹³⁰ in microglia, but substance P-induced intracellular ROS is nearly completely dependent upon NADPH oxidase.¹⁴⁸ In the traditional phagocyte, the PHOX-ROS pathway has been shown to amplify pro-inflammatory gene expression through their function as second messengers to regulate several downstream signaling molecules, including protein kinase C, mitogen activated protein kinase, and nuclear factor-κB (NFκB),^{135, 149–151} through redox signaling.

Using neuron-glia cultures from NADPH oxidase-deficient mice, studies have shown that NADPH oxidase initiates an intracellular ROS signaling pathway¹⁵² that can activate microglia and amplify the production of pro-inflammatory cytokines, such as TNFα¹³⁰ or PGE₂.¹⁵³ Additionally, Min and colleagues¹⁵⁴ demonstrated that ganglioside induces the activation of microglia, where the production of IL-1β, TNF-α, and iNOS are attenuated by the addition of the NADPH oxidase inhibitor, diphenylene iodonium (DPI). Furthermore, NADPH oxidase inhibitors and catalase are shown to suppress LPS-induced expression of cytokines (IL-1, IL-6, and TNFα), iNOS, mitogen activated protein kinase (MAPK), and NFκB phosphorylation.¹⁵⁵ Thus, both microglial cellular function and signaling pathways are modified by NADPH oxidase ROS production.

Accumulating evidence supports that NADPH oxidase contributes to microglia-mediated neurotoxicity through two mechanisms. First, activation of NADPH oxidase results in the production of extracellular ROS that is toxic to neurons. Second, activation of NADPH oxidase causes an increase in the microglial intracellular ROS, which enhances the production of pro-inflammatory factors that are toxic to neurons. Given the dual impact of NADPH oxidase activation on neurotoxicity, the role of NADPH oxidase in generating toxic ROS, and the prevalence of NADPH oxidase activation upon microglial activation, this suggests that microglial NADPH oxidase is a critical mechanism of neuronal death across multiple neurodegenerative diseases. However, the specific mechanisms defining how ROS causes this toxic microglial phenotype through redox-signaling is unknown.

Chronic Microglial Activation - Propagation of Disease

Microglia are unique when compared to differentiated myeloid immune cells in the periphery for multiple reasons. In addition to the obvious morphological differences and unique resting profiles characterized in microglia, these cells are also more likely to establish chronic pro-inflammatory responses, rather than demonstrate resolution of the innate immune response, as is common in the peripheral immune system.¹⁵⁶ While the mechanisms are unclear, we believe that this microglial tendency for a chronic pro-inflammatory response is a key factor driving progressive neuron damage, contributing to the chronic nature of neurodegenerative diseases.

Reactive Microgliosis-Chronic Response to a Single Stimulus

While microglial activation was initially perceived as a transient event, ³¹ it is now thought to be chronic and culpable in the propagation of disease.^{1, 94, 157} Reports have shown that microglia can remain chronically activated^{140, 158, 159} in a process that has been termed reactive microgliosis. Reactive microgliosis can be defined as microglial activation that occurs in response to neuronal damage, which is then perpetuated by further microglial activation and neurotoxicity (Figure 1). Thus, a self-propelling and progressive cycle of microglial activation and neuron damage ensues.¹

Figure 1. Reactive Microgliosis Drives Chronic Neuron Damage.

Both stimulation of microglia with pro-inflammatory triggers (ex. LPS) and direct neuron damage (ex. Glutamate excitability) result in microglial activation causing the release of neurotoxic factors, such as IL-1β, NO, TNFα, ONOO⁻, O₂^•− and H₂O₂.. Subsequently, following damage with either a pro-inflammatory trigger or a direct neurotoxin, the neuron releases microglial activators (Soluble Neuron-Injury Signals) such as μ calpain, MMP3, α-synuclein, and neuromelanin which activate microglial cells and propagate the cycle. This self-perpetuating cycle of neurotoxicity is known as reactive microgliosis. Abbreviations: Hydrogen peroxide (H₂O₂); Interleukin 1 beta (IL-1β); Lipopolysaccharide (LPS); Matrix Metalloproteinase 3 (MMP3); 1-Methyl-4-Phenyl-1,2,3,6-Tetrahydropyridine (MPTP); 1-methyl-4-phenylpyridinium ion (MPP⁺); nitric oxide (NO); peroxynitrite (ONOO⁻); prostaglandin E2 (PGE₂); superoxide (O₂^•−); tumor necrosis factor alpha (TNFα). This figure was slightly modified from Block & Hong (2007)¹.

For example, to damage DA neurons, MPTP is metabolized to 1-methyl-4-phenylpyridinium (MPP⁺), which is then selectively taken up by the dopamine transporter (DAT), resulting in inhibition of the mitochondrial electron transport chain complex I.¹⁶⁰ In addition to this mechanism of direct neurotoxicity, MPTP-induced neurotoxicity is also clearly linked with microglial activation in vivo and in vitro.^{137, 140, 159}

It is particularly interesting that chronic microglial activation can continue years after MPTP exposure in humans¹²² and primates,¹⁵⁹ despite the fact that the exposure to MPTP was brief, indicating an incessant, active pathologic process. Importantly, several studies have shown that microglia have an active role in the process of neuronal death in MPTP/MPP⁺-induced neurotoxicity.^{137, 140, 159} Specifically, in vitro studies show that while MPTP directly damages DA neurons, both MPTP and MPP⁺ fail to directly activate microglia.¹⁴⁰ Rather, microglial activation in response to MPTP or MPP⁺ occurs only when neurons are present and this response takes time (days) to accumulate.¹⁴⁰ Further, the addition of microglia to enriched neuron cultures greatly enhances MPTP-induced DA toxicity,¹⁴⁰ demonstrating that microglia cause DA neuron damage in addition to the direct toxic effects of MPTP/MPP⁺ on the neuron.

In vivo studies also emphasize the important role of inflammation as a toxic component of MPTP/MPP⁺ neurotoxicity,¹⁶¹ where DA neuron damage in response to MPTP is significantly reduced in mutant mice with deficient production of pro-inflammatory factors, such as nitric oxide,¹⁶² superoxide,^{137, 163} prostaglandins,^{164, 165} and TNFα.¹⁰⁸ Thus, several lines of evidence suggest that microglial activation initiated by neuronal damage may be toxic and persistent, continuing long after the initiating damaging/toxic stimulus is gone.

Recently, we employed an in vitro approach to separate neuron injury factors from the cellular actors of reactive microgliosis in an attempt to begin to discover molecular signals (soluble neuron injury factors) responsible for chronic and toxic microglial activation.¹⁶⁶ We found that upon injury with the DA neurotoxin MPP⁺, DA neurons released soluble neuron injury factors that activated microglia and were selectively toxic to DA neurons in mixed mesencephalic neuron-glia cultures through NADPH oxidase.¹⁶⁶ This is consistent with other studies that have identified other soluble neuron injury factors that signal toxic microglial activation such as matrix metalloproteinase 3,¹⁶⁷ α synuclein,⁵⁷ and neuromelanin.¹⁶⁸ In addition, we identified μ calpain as a key soluble neuron-injury signal released from damaged neurons, causing selective DA neuron death through activation of microglial NADPH oxidase and superoxide production, converging on the common mechanism of toxic microglial activation through ROS production.¹⁶⁶ Together, these findings support that DA neurons may be inherently susceptible to reactive microgliosis, providing much needed insight into the chronic nature of Parkinson’s disease. Notably, while DA neurons may be more vulnerable to reactive microgliosis, it is likely that reactive microgliosis is a contributing factor to most neurodegenerative diseases.

LPS-Chronic Response to a Single Stimulus

LPSis reported to activate microglia both in vivo and in vitro causing the progressive and cumulative loss of DA neurons over time.^{129, 169, 170} While traditionally thought of as a model of infectious insult, a recently developed PD animal model uses a systemic LPS administration, and shows that a single pro-inflammatory stimulus can persistently activate microglia to cause neuron death.¹⁵⁶ Specifically, we have recently shown that systemic LPS administration activates cells in the liver to produce TNFα, which is distributed in the blood and transferred to the brain through TNFα receptors to induce the synthesis of additional TNFα and other pro-inflammatory factors, creating a persistent and self-propelling neuroinflammation that induces delayed and progressive loss of DA neurons in the SN of adult animals.¹⁵⁶ This work is the first to support that a single pro-inflammatory stimulus (whether a pathogen or environmental in origin) in the adult animal can cause progressive neuron damage later in life, suggesting a wide therapeutic window for the effective use of anti-inflammatory therapy in neurodegenerative disease.

However, earlier reports have already shown that exposure to LPS early in life can induce and enhance DA neuron damage later in life. Studies show that during critical periods of embryonic development (E 10.5), maternal exposure to low concentrations of LPS in mice impacts microglial activation and DA neuron survival in offspring that persists into adulthood.^{170, 171} Interestingly, LPS has been implicated in the potential etiology of sporadic PD through sepsis and early life exposure during pregnancy.^{169, 172, 173} Further, diagnosed parkinsonian symptoms, brain inflammation, and damage to the SN have been described in a patient who had received accidental systemic administration of LPS.¹⁷⁴ However, the most important implication of these findings is that not only can microglia induce neuron damage, but that microglia can become persistently activated to produce continuous and uncontrolled neurotoxicity that fails to resolve, long after the instigating stimulus has dissipated.

Priming: Lowering the Threshold to Initiate Neurotoxicity

The phenomenon of microglial priming offers valuable insight into why microglia continue to respond to additional stimuli in the chronic cycle of neuroinflammation (Figure 1). In the case of priming, microglia are not just exhibiting and enhanced toxic microglial response. Rather, in the case of priming, the microglial phenotype shifts, where a much lower stimulus is needed to exact a toxic microglial response, enhancing the probability that the chronic cycle of toxic reactive microgliosis will continue.

Systemic LPS has been widely explored for the ability to amplify ongoing neuropathology in adults. Systemic LPS administration has been shown to enhance prion-induced cognitive deficits, neuroinflammation, and neuropathology.¹⁷⁵ Further, systemic neonatal exposure to LPS is shown to significantly amplify neuronal death associated with ischemic insult.¹⁷⁶ Finally, immunological perturbation during critical periods of development¹⁷⁰ or aging¹⁷⁷ and sentience,¹⁷⁸ can prime microglia, where additional stimuli result in an exaggerated and prolonged pro-inflammatory response that enhances neuron damage.

By altering concentrations of intracellular ROS and consequent redox-signaling, NADPH oxidase is reported to prime the microglial response to further insult. Triggers of microglia activation, such as rotenone⁵⁵ and neuronal death,¹⁷⁹ are shown to prime microglia through NADPH oxidase and result in synergistic microglial activation that is associated with neurotoxicity upon additional insult with LPS. This has particular importance given that a multiple hit hypothesis¹⁸⁰ has been proposed as a potential mechanism through which environmental toxicants induce neurodegenerative disease over an in individual’s lifetime and may provide significant insight into the progressive nature of neuroinflammation.

Conclusions & Implications

Microglia can be continuously activated to produce toxic factors (cytokines and reactive oxygen species) by either single or chronic exposure to disease proteins, environmental toxins, cytokines, and neuronal damage (reactive microgliosis), resulting in the progressive loss of neurons over time, a fundamental component of neurodegenerative disease. Recent work suggests that DA neurons may be inherently vulnerable to reactive microgliosis, providing much needed insight into the progressive nature of Parkinson’s disease. Current research also suggests that redox signaling in microglia may be a critical mechanism of chronic neuroinflammation that propagates the microglial pro-inflammatory response through amplification of cytokine production and lowering of the stimulus threshold to cause a neurotoxic toxic response. Future research will need to focus on why the microglial response fails to resolve when compared to the peripheral immune system, the mechanisms through which ROS is signaling toxic microglial activation, and harness this information for the identification of novel therapeutic approaches.

Abbreviations

6-OHDA

6-hydroxy dopamine

Aβ

Beta amyloid peptide

ALS

amyotrophic lateral sclerosis

CNS

central nervous system

COX2

cyclooxygenase 2

DA

dopamine

DAT

dopamine trasporter

DEP

Diesel Exhaust Particles

DPI

diphenylene iodonium

HAND

HIV associated neurocognitive disorder

H₂O₂

hydrogen peroxide

IL-1β

interleukin 1-β

IL-6

interleukin 6

iNOS

inducible nitric oxide synthase

LPS

lipopolysaccharide

MHC

major histocompatability complex

MPP+

1-methyl-4-phenylpyridinium

MPTP

1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine

NADPH

nicotinamide adenine dinucleotide phosphate

NFκB

nuclear factor-κB

NO

nitric oxide

O₂^•−

superoxide

PD

Parkinson’s disease

PGE₂

prostaglandin E₂

PHOX

phagocytic oxidase

PQ

paraquat

ROS

reactive oxygen species

SN

substantia nigra

TLR4

toll-like receptor 4

TNFα

tumor necrosis factor-alpha