Journal of Neuroimmune Pharmacology

, Volume 1, Issue 2, pp 117–126

Microglia as a Pharmacological Target in Infectious and Inflammatory Diseases of the Brain

Authors

  • R. Bryan Rock
    • Center for Infectious Diseases and Microbiology, Translational Research and the Department of MedicineUniversity of Minnesota Medical School
    • Center for Infectious Diseases and Microbiology, Translational Research and the Department of MedicineUniversity of Minnesota Medical School
Invited Review

DOI: 10.1007/s11481-006-9012-8

Cite this article as:
Rock, R.B. & Peterson, P.K. Jrnl NeuroImmune Pharm (2006) 1: 117. doi:10.1007/s11481-006-9012-8

Abstract

Following an eclipse of scientific inquiry regarding the biology of microglia that lasted 50 years, recognition toward the end of the 20th century of their neuropathogenic role in HIV-associated dementia and in neuroinflammatory/neurodegenerative diseases fueled a renaissance of interest in these resident macrophages of the brain parenchyma. Results of a large number of in vitro studies, using isolated microglial cells or glial/neuronal cell cultures, and parallel findings emerging from animal models and clinical studies have demonstrated that activated microglia produce a myriad of inflammatory mediators that both serve important defense functions against invading neurotropic pathogens and have been implicated in brain damage in infectious as well as neuroinflammatory/neurodegenerative diseases, such as multiple sclerosis, Alzheimer's disease, Parkinson's disease, and amyotrophic lateral sclerosis. This review provides a brief background regarding the physiological and pathophysiological roles of microglia and highlights current pharmacological approaches that target activated microglia with the goal of ameliorating infectious and neuroinflammatory/neurodegenerative diseases of the brain. Although this aspect of the field of neuroimmunopharmacology is in its infancy, it holds great promise for developing new treatments and prevention of diseases that are, in many cases, epidemic throughout the world.

Keywords

MicrogliaCentral nervous system infectionsNeuroinflammationNeurodegenerative diseases

Introduction

Studies of the origin, distribution, and functions of microglia date back to the 19th century, and by the mid-1930s much was understood about the nature of these “resident macrophages of the nervous system” (del Rio-Hortega, 1932). This early period of exuberant discovery was followed by an eclipse of scientific interest that lasted half a century, only to be reawakened toward the end of the 20th century. The renaissance of research interest in microglia has been fueled by recognition of their pivotal role in the neuropathogenesis of HIV and their involvement in many, if not all, neuroinflammatory/neurodegenerative diseases of the brain.

The physiological and pathophysiological features of microglia have been reviewed elsewhere (Streit, 2002; Rock et al., 2004; Farber and Kettenmann, 2005). The concept of microglia as a pharmacological target in infections and inflammatory diseases of the brain, the subject of this review, is relatively new. This concept has emerged from a broader understanding of the mechanisms underlying the neuroprotective and neuropathogenic properties of microglia and an increased interest in translating these insights into new treatments and prevention of brain diseases that are, in many cases, epidemic. Although the main purpose of this article is to highlight pharmaceutical approaches to altering microglial cell activities that are germane to controlling infections and suppressing deleterious inflammatory processes within the central nervous system (CNS), a summary will also be provided of key functional properties of microglia and their role in CNS infections and inflammatory diseases of the brain.

Functional properties of microglia

In his treatise on microglia in 1932, the Spanish neuroanatomist Pio del Rio-Hortega articulated many enduring concepts regarding these brain macrophages, e.g., their derivation from mesoderm (most likely from bone marrow-derived blood monocytes populating the brain during early fetal development) and their plasticity (an amoeboid form found in the fetus), a ramified (resting) morphology found throughout the nervous system, and, in response to a variety of insults to the nervous system, a resumption of a motile, amoeboid form “necessary for active discharge of their macrophagic function” (del Rio-Hortega, 1932). Microglia are distinguished from other populations of brain macrophages (Guillemin and Brew, 2004) by their parenchymal location, where it is estimated they are present in numbers equal to neurons. Like the macroglia (astrocytes and oligodendrocytes), microglia are thought to provide a supportive function for neurons, but little is known about their physiological function in the healthy brain. However, recent studies indicate that these ramified microglia are not resting or quiescent but are rather continually surveying their microenvironment with extremely motile processes and protrusions (Nimmerjahn et al., 2005).

Although the nature of the communication between ramified microglia and neurons in the healthy brain is only beginning to be understood, much is known about the properties of microglia that have reassumed their amoeboid morphology following pathological insults to the nervous system. In this “reactive” form, microglia have the capacity to rapidly up-regulate a large number of receptor types (Table 1) and to release a myriad of secretory products (Table 2) that can contribute to defense, but also damage or kill neurons. As will be discussed below, several of the secretory products that play a direct or indirect antimicrobial role have also been implicated in the pathogenesis of neuroinflammatory/neurodegenerative diseases.
Table 1

Microglial cell membrane receptorsa

Cell adhesion molecules

 Immunoglobulin (Ig) superfamily

   Ig Fc receptors (FcγRI, RII, RIII)

   Major histocompatability (MHC) class I glycoproteins

   MHC class II glycoproteins

   CD4 receptors

   Intercellular adhesion molecule 1 (ICAM-1)

 Integrins

   Leukocyte function-associated antigen 1 (LFA-1; CD11a/CD18: CR1)

   Mac-1 (CD11b/CD18; CR3)

   p150, p95 (CD11c/CD18; CR4)

  Complement receptors: C1q, C5a

Cytokine/chemokine receptors

 Interferon (INF)-α, IFN-β, IFN-γ

 Interleukin (IL)-1, IL-6, IL-10, IL-12, IL-16, IL-23

 Tumor necrosis factor (TNF)-α

 Macrophage-colony stimulating factor (M-CSF), Granulocyte-macrophage (GM)-CSF

 CCR, CXCR, CX3CR

Toll-like receptors

CD14 receptors

Mannose receptors

Purinogenic receptors

Opioid receptors (μ, κ)

Cannabinoid receptors (CB1, CB2)

Benzodiazepine receptors (mitochondrial membrane)

aReceptors reported in the literature, whose expression may be influenced by the state of activation as well as by the anatomic location, age, and animal species from which the microglia are derived.

Table 2

Secretory products of microgliaa

Cytokines (IL-1α, IL-1β, IL-6, IL-10, IL-12, IL-16, IL-23, TNF-α, Transforming growth factor [TGF]-β)

Chemokines

 CC: CCL2/MCP-1, CCL3/MIP-1α, CCL4/M1P-1β, CCL5/RANTES

 CXC: CXCL8/IL-8, CXCL9/MIG, CXCL10/IP-10, CXCL12/SDF-1α

 CX3C: CX3CL1/fractaline

Matrix metalloproteinases (MMP-2, MMP-3, MMP-9)

Free radicals: superoxide, nitric oxide

Eicosanoids: PGD2, leukotriene C4

Growth factors: nerve growth factor, fibroblast growth factor

Proteases: elastase, plasminogen

Cathepsins B and L

Quinolinic acid, glutamate

Amyloid precursor protein

Complement factors: C1, C3, C4

aSecretory products reported in the literature, whose generation is influenced by the state of activation as well as by the anatomic location, age, and animal species from which the microglia are derived.

Historically considered an “immunologically privileged” site, the nervous system is more accurately regarded as immunologically specialized, with microglia and other brain macrophage populations serving as an innate immune system. Also, it has become appreciated only recently that the main communication signals used by microglia, i.e., cytokines and chemokines (“immunotransmitters”), are also received and, in some cases, produced by neurons. And conversely, certain neurotransmitters and neuropeptides can signal or are produced by microglia. Thus, cells within the nervous system use an elaborate and complex language to communicate among themselves, and when necessary, some of these same signals direct traffic of microglia within and recruit cells into the nervous system from the vascular space.

The blood–brain barrier

Prior to any discussion of the therapeutic manipulation of microglial cell function for the treatment of either infectious or neuroinflammatory/neurodegenerative diseases, the natural barriers to drug penetration into the CNS must be addressed. These natural barriers consist of both physical impediments, such as the blood–brain barrier (BBB) and blood–cerebral spinal fluid (CSF) barrier, as well as the presence of drug-specific transporters in the cells comprising these barriers. Classically, the physical impediments imposed by the BBB include endothelial cells with tight junctions, low permeability, and low rates of pinocytosis (Loscher and Potschka, 2005). Astrocytic end-foot processes create the brain side of the BBB. Ultimately, drugs intended for the CNS must either enter via passive diffusion or receptor-mediated transport. However, even if a drug has the lipid-soluble profile to successfully penetrate the physical barriers or is able to use a specific receptor-mediated transport system, numerous drug efflux transporters, such as P-glycoprotein and multidrug resistance proteins, can effectively prevent the accumulation of the drug within the CNS (Loscher and Potschka, 2005). Specific drugs targeted by these efflux transporters have been identified, including antiretroviral drugs, antibiotics, anticancer drugs, and analgesics. Given the formidable processes in place to prevent drug penetration into the CNS, pharmacologic agents targeting microglial cell function need to overcome these hurdles.

Microglia and CNS infections

As the major defenders of the brain parenchyma, microglia are called upon to perform a pivotal defense function against a broad array of pathogens that target the CNS (Table 3). Although microglia clearly provide direct innate immunity to foreign invaders, they also are involved in the recruitment of cells to the CNS from the periphery. This evolving concept of microglia as both an immune effector cell and a coordinator of the trafficking of other immune cells into the nervous system is critical to the notion that the neuropathogenesis of these pathogens may be manipulated by targeting microglia. In the following section, we examine the role of microglia in HIV-associated dementia, CNS tuberculosis, and cryptococcal meningitis as examples of CNS infections in which targeting microglia may be a novel method for intervening in these infections.
Table 3

Interacellular pathogensa

Viruses

Bacteria

 Retroviruses

Mycobacterium tuberculosis

   HIV

Treponema pallidum

   Human T lymphotropic virus type 1

Borrelia burgdorferi

Nocardia asteroides

 Herpes group

Leptospira

   HSV

Brucella

   CMV

Rickettsia

   Epstein–Barr virus

Mycoplasma

   Human herpesvirus 6

Ehrlichia

   B virus

Parasites

 Enteroviruses

Cysticercus

   Polioviruses

Toxoplasma gondii

   Coxsackieviruses

Fungi

   Echoviruses

Cryptococcus neoformans

 Arboviruses

Coccidioides immitis

 Rabies virus

Histoplasma capsulatum

 Mumps virus

Blastomyces dermatitidis

 Lymphocytic choriomeningitis virus

Candida

Zygomycetes

 Measles virus

Aspergillus

 Rubella virus

Sporothrix schenckii

 Nipah virus

Prions

 Hendra virus

 

 JC virus

 

aHuman pathogens that have the capacity to invade, multiply, and elicit a pathologic response within cells of the brain parenchyma.

HIV infection

HIV-associated dementia (HAD) is typically a late manifestation of HIV infection and is characterized by cognitive, behavioral, and motor deficits ranging from mild disease to profound dementia. A key element in the development of HAD appears to be infection of microglia, which are the only brain cell type that is productively infected by this virus, and—because of their ability to harbor viral particles intracellularly—are also potential reservoirs of the virus. Using histopathological techniques, the number of activated microglia and macrophages in the CNS is a better correlate with HAD than the presence and amount of HIV-1-infected cells in the brain (Glass et al., 1995), and microglial cell activation is a better correlate of neuronal damage than productive HIV-1 infection in the CNS (Adle-Biassette et al., 1999). These observations not only demonstrate the importance of microglia in viral production, but also underscore the importance of activated microglia/macrophages in HIV-1 neuropathogenesis.

It has also become increasingly clear that neurotoxic mediators released from microglia play a major role in HIV-1 neuropathogenesis. This is supported by the fact that the number and localization of HIV-1-infected microglia is quite limited compared to the diffuse CNS abnormalities that occur in HAD (Lipton and Gendelman, 1995), which suggests that diffusible factors released from microglia are contributing to neuronal loss. Indeed, it is now recognized that HIV-1-infected microglia and other brain macrophages actively secrete both neurotoxins such as tumor necrosis factor (TNF)-α, interleukin (IL)-1β, CXCL8, glutamate, quinolinic acid, platelet activating factor, eicosanoids, and nitric oxide (NO) as well as neurotoxic viral proteins such as Tat, gp120, and gp41. Ultimately, a complex interaction between activated microglia/macrophages, astrocytes, and neurons triggers the onset of neuronal dysfunction or apoptosis and progression of CNS damage.

Although antiretroviral therapy (ART) is extremely important in the control of HIV infection, its role in HAD is less clear. Specific ARTs, such as efavirenz, have been shown to inhibit HIV replication within human microglia at physiological concentrations (Albright et al., 2000). The drugs indinavir and zidovudine have been shown to inhibit lipopolysaccharide (LPS)-stimulated microglial production of matrix metalloproteinase (MMP)-9, thereby demonstrating an effect on microglia that extends beyond pure antiviral activity (Liuzzi et al., 1999).

Agents with activity against HIV may not be enough to impact HAD. As stated previously, microglia release neurotoxins, which may ultimately be more important in HAD than viral replication per se. Also, there is some evidence that ART itself may contribute to overall neuroinflammation. An example of this is found in an autopsy study that compared HIVdisease pre- and post-ART in regards to neuroinflammation. In this study, a surprising degree of inflammation was observed among the post-ART cases compared to pre-ART cases in the areas of the hippocampus and basal ganglia (Anthony et al., 1997).

For these reasons, ideal therapeutic agents for the treatment of HAD would attenuate HIV-1 neuropathogenesis through both direct inhibition of viral expression and suppression of microglia-produced neurotoxins. One agent that appears to combine each of these attributes is the antimicrobial agent minocycline. This second generation tetracycline, which easily penetrates into the CNS, has demonstrated anti-inflammatory effects on microglia and also inhibits HIV-1 production by microglia (Si et al., 2004). Another promising class of agents are the benzodiazepines, which cross the CNS, bind to microglia, inhibit LPS-induced TNF-α production, suppress HIV-1 Tat protein-induced chemotaxis, and also inhibit HIV-1 expression in microglia (Rock et al., 2004). These agents may yet prove to be useful in controlling HAD by targeting microglia as the central player in this process.

Tuberculosis

CNS tuberculosis accounts for 1–10% of all cases of tuberculosis, carries a high mortality, and is disproportionately more frequent and more virulent in children (Farer et al., 1979) and in HIV-infected individuals (Berenguer et al., 1992). Whereas Mycobacteriumtuberculosis gains access to the CNS parenchyma via hematogenous spread, entry into the cerebral spinal fluid occurs via rupture into the subarachnoid space from an adjacent parenchymal tubercle (Leonard and Des Prez, 1990).

The notion that microglial cells may be a potential therapeutic target for antituberculosis therapy comes from several key observations. First, human microglial cells are productively infected with M. tuberculosis and are the principal cell target in the CNS (Curto et al., 2004; Rock et al., 2004). In our laboratory, we have found that exposure of purified human microglia and astrocytes to M. tuberculosis is associated with a selective infection of microglia, and that ingestion of nonopsonized M.tuberculosis by human microglia is facilitated by the CD14 receptor (Rock et al., 2004).

Second, among various mycobacterial species, M. tuberculosis may have more successfully adopted an efficient method of gaining access to microglial cells. One study demonstrated that human microglia are more efficient at ingesting M. tuberculosis than virulent and avirulent strains of M. avium, and that following infection there is a lasting inhibition of both IL-1 and IL-10 production (Curto et al., 2004). The authors of this study suggested that mycobacterial infection induces immunosuppressive effects on microglial cells, which is more evident with more virulent strains.

From these observations, microglia, like other tissue macrophages, have emerged as being key to understanding the neuropathogenesis of tuberculosis. Although specific antimycobacterial therapy against M. tuberculosis-infected microglia in a CNS tuberculosis model has yet to be been evaluated, our laboratory has examined the effect of dexamethasone therapy on production of proinflammatory cytokines and chemokines by M. tuberculosis-infected microglial cell cultures (Rock et al., 2005). This corticosteroid is recommended as adjunctive therapy in the treatment of CNS tuberculosis. In these experiments, M. tuberculosis-infected microglia elicited robust amounts of several cytokines/chemokines, including TNF-α, IL-6, IL-1β, CCL2, CCL5, and CXCL10. Treatment with dexamethasone markedly suppressed production of these mediators. This study, in effect, supports the concept that microglia play an important role in neuropathogenesis of tuberculosis and that the benefits of dexamethasone therapy, which are seen clinically, could operate via modulation of the production of proinflammatory cytokines/chemokines by microglia.

Cryptococcosis

Cryptococcus neoformans is an important cause of meningitis and, in some geographic locations, is a major opportunist in AIDS patients. This fungus is a ubiquitous pathogen that has a pronounced predilection for the CNS. This “neurotropism” appears to be the result of the ability of cryptococci to grow unimpeded in the CNS. Defending the CNS against C. neoformans requires both innate and adaptive immunity.

Several studies have demonstrated the central role of microglia in this CNS infection. First, microglia can ingest and inhibit the growth of C. neoformans (Lee et al., 1992; Blasi et al., 1995). Nonopsonized cryptococci can be ingested by porcine (Lipovsky et al., 1997) and murine (Blasi et al., 1995) microglia, but opsonization is required for human microglia to ingest cryptococci (Lee et al., 1992; Lipovsky et al., 1998a). Also, in the presence of specific antibody, human microglia produce CCL3, CCL4, CCL2, CXCL8, and low levels of CCL5 via Fc-receptor activation (Goldman et al., 2001). Finally, enhanced phagocytosis and killing of cryptococci by IFN-γ- stimulated microglia occurs in murine microglia (Blasi et al., 1995), but only growth arrest is achieved in human microglia under similar conditions (Lipovsky et al., 1998b). This animal species difference appears to be associated with the relatively inefficient generation of NO by cytokine-activated human microglia compared to that by murine microglia.

In addition to antifungal therapy directed against C. neoformans, other agents have demonstrated effects on cryptococcalmicroglial cell interactions, which may ultimately impact the overall neuropathogenesis of this agent. In human microglia, morphine can enhance the uptake of cryptococci by both μ-opioid receptors and complement receptors (Lipovsky et al., 1998a). This is contrasted with evidence that morphine suppresses porcine microglial uptake of cryptococci (Sowa et al., 1997). Also, chloroquine enhances the anticryptococcal activity of the murine microglia-derived cell line BV2 in vitro (Mazzolla et al., 1997).

Although the focus on microglia as a pharmacological target in the treatment of these and other CNS infections is only in its infancy, increased attention is being given to the development of drugs that suppress or kill microbes replicating in these cells as well as to agents that inhibit production of neurotoxic mediators released from infected or activated microglia.

Microglia and neuroinflammatory diseases

Neuroinflammation/neurodegeneration

As techniques for identifying activated microglia have become available, such as imaging with [11c]PK 11195, a ligand for peripheral benzodiazepine receptors, their presence has been demonstrated in a variety of neuroinflammatory/neurodegenerative diseases, including multiple sclerosis (MS) (Chen et al., 2004), Alzheimer's disease (AD) (Cagnin et al., 2001), and amyotrophic lateral sclerosis (ALS) (Turner et al., 2004). Historically, such findings would have been interpreted as evidence of the supportive function of microglia in these diseases, e.g., serving as scavengers of dead cells and debris. However, as an appreciation of the potentially deleterious properties of activated microglia has emerged, the hypothesis has been entertained that these brain macrophages participate in the demise of neurons, i.e., the so-called “dark side of microglia.”

A neuropathological role of microglia has now been postulated in most, if not all, neuroinflammatory/neurodegenerative diseases. As shown schematically in Fig. 1, neurotoxic mediators, such as cytokines, free radicals, and excitatory neurotransmitters, released from activated microglia provide the proposed link between neuroinflammatory and neurodegenerative processes in MS, AD, Parkinson's disease (PD), and ALS. In most cases, the activating triggers are unknown but may be either exogenous factors, e.g., microbial products, environmental toxins or neurotoxic drugs, or endogenous proteins that have taken on pathological properties, e.g., amyloid β peptide (Aβ) in AD and α-synuclein in PD. As suggested in this schematic diagram, the factors activating microglia may, when present in sufficiently high concentrations, also be directly toxic to neurons or they may act synergistically with inflammatory mediators in the killing of neurons. Also, in diseases such as MS, recruitment of activated T cells is of crucial neuropathological significance.
https://static-content.springer.com/image/art%3A10.1007%2Fs11481-006-9012-8/MediaObjects/11481_2006_9012_Fig1_HTML.gif
Fig. 1

Simplified schematic diagram of the interactions between microglia and neurons in neuroinflammatory/neurodegenerative diseases. Lipopolysaccharide (LPS), tumor necrosis factor (TNF)-α, reactive oxygen intermediates (ROI), reactive nitrogen intermediates (RNI), quinolinic acid (QA), multiple sclerosis (MS), Alzheimer's disease (AD), Parkinson's disease (PD), amyotrophic lateral sclerosis (ALS).

Clearly, the pathogenesis of all neuroinflammatory/neurodegenerative diseases is complex and the schematic representation in Fig. 1 is overly simplistic. In addition to the neurotoxic potential of activated microglia, their neuroprotective properties and interactions with astrocytes and oligodendrocytes as well as neurons need to be taken into account. Moreover, risk factors such as genetic predisposition, age, gender, and the varied anatomic specificities of each of the neuroinflammatory/neurodegenerative diseases need to be considered. For example, in MS, an autoimmune disorder that predominates in young women, and in the animal models of this demyelinating disease, oligodendrocytes are particularly vulnerable to injury, and it has been postulated that inflammatory mediators (cytokines, chemokines, and free radicals) released from activated microglia contribute to damage of these glial cells (Jack et al., 2005; Li et al., 2005). Interestingly, although microglia are found in all areas of the nervous system, they are more heavily concentrated in white than gray matter.

By way of contrast to MS, AD is a neurodegenerative disease associated with aging, and neural damage begins in the cerebral cortex and progresses over time to the hippocampus and amygdala. AD was the first of the classic neurodegenerative diseases in which activated microglia were considered to be of neuropathogenic importance (McGeer and McGeer, 1995). Pathological diagnosis of AD requires identification of neurofibrillary tangles and Aβ plaques. Although early work demonstrated that Aβ peptide is directly toxic to neurons, subsequent studies provided evidence that Aβ peptide also stimulates release of neuroinflammatory/neurotoxic factors from activated microglia (Rogers and Lue, 2001).

Although many recent studies have shed light on the potentially “dark side of microglia,” a case can also be made for the neuroprotective role of microglia in AD (Streit, 2005). Evidence supporting this neuroprotective role includes the microglial production of transforming growth factor β1 in an acute neuronal injury model (Lehrmann et al., 1998), in vitro studies demonstrating microglial production of neurotrophic factors (Nakajima and Kohsaka, 2002), and evidence that transplanted microglia/macrophages into an injured brain model enhances neurite outgrowth (Rapalino et al., 1998). Furthermore, Butovsky et al. (2006) have advanced the concept of “protective autoimmunity” by demonstrating that whereas microglia activated by endotoxin block neurogenesis, microglia activated by interleukin-4 or interferon-γ associated with CD4 lymphocyte infiltration induces neurogenesis.

The delicate balance between the neurotoxic and neuroprotective properties of microglia is perhaps best underscored by a clinical trial in which patients who were immunized with Aβ peptide demonstrated clinical improvement, and autopsy revealed that Aβ plaques were cleared from the brain. However, the trial had to be prematurely stopped because of the development of severe meningoencephalitis in some patients related to vaccine-generated cytotoxic T cells. Recent analysis has demonstrated that prevaccine gene expression patterns indicative of proinflammatory pathways were associated with development of immunotherapy-related meningoencephalitis, whereas expression of genes related to protein synthesis, protein trafficking, and DNA repair were strongly associated with a beneficial immunoglobulin G response to immunization (O'Toole et al., 2005).

Mounting evidence also exists incriminating activated microglia in the pathogenesis of PD (Block and Hong, 2005). In PD, degeneration of dopaminergic neurons in the substantia nigra and the nerve terminals of the striatum is the critical site of brain damage. Cytoplasmic inclusions (Lewy bodies), which are localized to these neurons, are a histopathological hallmark of PD, and one of the components of Lewy bodies, α-synuclein, has been shown to activate microglia, much in the same way as Aβ peptide has been shown in studies relevant to AD. It has been proposed that generation of reactive oxygen intermediates (ROI) and reactive nitrogen intermediates (RNI) by activated microglia plays a pivotal role in neurodegeneration of dopaminergic neurons, which are highly susceptible to injury caused by oxidative stress (Block and Hong, 2005).

Interestingly, microglia are overrepresented in the substantia nigra, and in a unifying hypothesis, Block and Hong proposed that a number of triggers of microglial cell activation (microbes or microbial products, environmental toxins, and drugs) elicit free radicals as well as other inflammatory and neurotoxic mediators that induce neurodegeneration of dopaminergic neurons. Support for the notion that activated microglia continue to wreak havoc in the substantia nigra even after the offending or triggering agent is long gone has been provided by recent findings in a primate model (Barcia et al., 2004) and in an autopsy study of humans (Hald and Lotharius, 2005) who were exposed to the parkinsonism-inducing toxin, 1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine (MPTP). Preliminary studies suggest a parallel model of microglial cell activation and neurodegeneration of dopaminergic neurons by methamphetamine (Thomas et al., 2004). Given the current epidemic of addiction to this agent, it is likely that another neuroinflammatory/neurodegenerative disease will emerge as a major public health problem.

In addition to possessing D1 and D2 dopamine receptors, microglia express receptors for all other classes of neurotransmitters (acetylcholine, GABA, catecholamines, and glutamate). The role of these receptors in neuronal–microglial cell communication is not understood; however, the production of glutamate and the glutamate receptor ligand quinolinic acid by activated microglia has been implicated in excitotoxicity of neurons in several neurodegenerative diseases (Matute et al., 2002).

Cyclooxygenase-2 (COX-2) is an enzyme central to the production of prostaglandins, a family of powerful inflammatory mediators produced by activated microglia that can have both deleterious and neuroprotective effects in the inflamed brain (Tzeng et al., 2005). COX-2 and prostaglandin E2 are elevated in the CNS of patients with ALS, and recent studies have implicated activated microglia in this “neuron only” disease (Weydt and Moller, 2005). Prostaglandin production by activated microglia has been implicated in other neuroinflammatory/neurodegenerative diseases, and not surprisingly, COX-2 has been considered a major therapeutic target.

Therapeutic agents

Based upon the growing body of evidence summarized above that activated microglia play a pathogenic role in many neuroinflammatory/neurodegenerative diseases and that the same neurotoxic mediators (prostaglandins, free radicals, cytokines, and glutamate) are involved in one or more of these diseases, several related microglial cell targets have been considered in development of new treatments of AD, PD, MS, and ALS. Given the enormous need for new therapies for these devastating diseases, this line of clinical research is likely to accelerate. A brief description follows of the approaches that have been taken thus far.

Following early epidemiological evidence that chronic use of nonsteroidal anti-inflammatory drugs (NSAIDS) was associated with a decreased risk of developing AD (McGeer and McGeer, 1995), the use of anti-inflammatory agents in the treatment of AD has been intensely studied. Hypothetically, by inhibiting microglial cell cyclooxygenases (COX-1 or COX-2), the metabolism of arachadonic acid is curtailed and production of deleterious proinflammatory prostaglandins is suppressed (Hoozemans and O'Banion, 2005). In addition to the suppression of prostaglandin production, COX inhibitors may also decrease the formation of Aβ peptide (Hirohata et al., 2005) and thus indirectly abrogate microglial cell activation by this neurotoxic protein. Although preclinical and early clinical data suggested that COX-2 inhibitors may have a beneficial role in AD, results of subsequent studies and the development of unanticipated side effects of COX-2 inhibitors have dampened enthusiasm for the use of these agents in the management of AD.

Based upon the results of numerous in vitro experiments and animal models showing the importance of free radical (ROI and RNI) production by activated microglial cells in neuroinflammatory/neurodegenerative diseases, antioxidants have also been considered as candidate therapeutic agents in diseases such as PD and AD (Block and Hong, 2005). Results of a recent study of the potent antioxidant vitamin E, however, failed to show any benefit in the treatment of AD patients with mild cognitive impairment (Petersen et al., 2005).

In a rat model, the vitamin A derivative retinoic acid has been shown to inhibit the up-regulation of NO production as well as to suppress the production of TNF-α by LPS and Aβ peptide-stimulated microglia (Dheen et al., 2005); moreover, similar results have been reported with the antioxidant resveratol (Bi et al., 2005), the antibiotic minocycline (Fan et al., 2005), the cholesterol-lowering statins (Lindberg et al., 2005), and neuroleptics (Peng et al., 2005)—suggesting that these agents should be considered for clinical trials in AD and PD.

As mentioned earlier, activated microglial cells release the excitatory neurotransmitter glutamate, which—via activation of neuronal NMDA receptors—can induce neuronal death. Astrocytes are known to play a key role in the metabolism of glutamate and thereby provide an important neuroprotective function. Because microglia possess glutamate receptors, uptake of glutamate by microglia could also contribute to neuronal protection. Of relevance to ALS, a drug screening program recently revealed that ceftriaxone modulates glutamate transporters and protects neurons both in vitro and in an animal model of ALS (Rothstein et al., 2005). Given the safety profile of this β-lactam antibiotic, achieving good levels in the CNS, it is being considered for clinical trials in this neuroinflammatory/neurodegenerative disease.

Conclusion

As mentioned earlier, the development of therapeutic agents for infectious and inflammatory diseases of the brain faces a number of formidable hurdles, which are both pharmacological (e.g., transport across the BBB and administration of the right concentration of drug at the earliest time possible in the disease process), and biological [i.e., activated microglia have both neurotoxic properties (reviewed above) as well as neuroprotective potential). In addition to ongoing laboratory investigation of microglial cell pathobiology and clinical trials of candidate drugs that have come from this line of research, it is predicted that the field of neural stem cell research will yield additional therapeutic and preventive measures for CNS infections and neuroinflammatory/neurodegenerative diseases. Although early studies have shown that microglia-derived TNF-α induces apoptosis of hippocampal progenitor cells (Cacci et al., 2005), other studies indicate that microglia and neural stem cells themselves (Pluchino et al., 2005) have anti-inflammatory and neuroprotective effects. Learning how to harness the “bright side of microglia,” as well as to suppress their “dark side” represents a major scientific and clinical challenge, but one that will have extraordinary payoffs in terms of reducing suffering and mortality from a large and growing number of patients with epidemic CNS infections and neuroinflammatory diseases.

Acknowledgements

This work was supported, in part by U.S. Public Health Service grants DA04381, DA09783, and DA020398.

Copyright information

© Springer Science+Business Media 2006