Neuroinflammation in addiction: A review of neuroimaging studies and potential immunotherapies

Highlights

• Glial activation contributes to neural adaptations in substance use disorders.

• PET is a useful technique to quantify TSPO levels following chronic drug exposure.

• Neuroinflammation promotes addiction-related brain and behavioral deficits.

• Interventions that reduce inflammation may be efficacious for substance use disorders.

Abstract

Addiction is a worldwide public health problem and this article reviews scientific advances in identifying the role of neuroinflammation in the genesis, maintenance, and treatment of substance use disorders. With an emphasis on neuroimaging techniques, this review examines human studies of addiction using positron emission tomography to identify binding of translocator protein (TSPO), which is upregulated in reactive glial cells and activated microglia during pathological states. High TSPO levels have been shown in methamphetamine use but exhibits variable patterns in cocaine use. Alcohol and nicotine use, however, are associated with lower TSPO levels. We discuss how mechanistic differences at the neurotransmitter and circuit level in the neural effects of these agents and subsequent immune response may explain these observations. Finally, we review the potential of anti-inflammatory drugs, including ibudilast, minocycline, and pioglitazone, to ameliorate the behavioral and cognitive consequences of addiction.

Introduction

Neuroinflammation has been attributed to the pathogenesis of a number of central nervous system (CNS) diseases (Block and Hong, 2005; Chen et al., 2016; Tansey et al., 2007), and although classically defined as the accumulation of mobile innate and/or adaptive immune cells in the tissue, there is diversity in what is considered to be inflammation in the brain, including gliosis, microglia activation, and the release of cytokines, chemokines, and pro-inflammatory factors (see “Neuroinflammation in psychiatric disorders: an introductory primer” in this Special Issue for additional background information). Broadly, neuroinflammation is thought to contribute to the neural adaptations following chronic exposure to drugs of abuse (Lacagnina et al., 2017; Liu et al., 2016; Pocock and Kettenmann, 2007), as many drugs render the brain more vulnerable to inflammation and resultant neuropathology. There is considerable interest in the mechanism by which drug use interacts with inflammatory processes, contributing to brain dysfunction, impairing cognitive control, and consequently promoting drug-use behavior. Preclinical studies show that drug exposure increases the release of pro-inflammatory cytokines, and glial cells (microglia and astrocytes) with chemokine and cytokine receptors respond quickly to CNS injury (Pocock and Kettenmann, 2007). Drug-induced dysregulation of neuroimmune signaling may compromise neuronal function, exacerbate neurodegeneration, and increase neurotoxicity, which may contribute to drug-related behavior through the activation of microglia and other glia-mediated synaptic remodeling (Lacagnina et al., 2017; Liu et al., 2016; Pocock and Kettenmann, 2007). Although the neural circuits relevant to substance use disorders may be impaired before inflammation or drug use, drug-induced inflammation may further compromise brain function in individuals with substance use disorders. It is, therefore, important to examine the combination of insults and interactive effects of substance use and neuroinflammation as new therapeutic strategies are considered.

The neuroimmune response to drugs of abuse is characterized, in part, by proliferation and morphological and functional changes of microglia and astrocytes (Ransohoff and Brown, 2012). Microglia are distributed throughout the brain with greatest concentrations found in substantia nigra, basal ganglia, and hippocampus (Lawson et al., 1990). Microglia respond directly to drug-induced CNS injury and are activated by stimulation of chemokine and cytokine receptors or by peripheral signals, potentially resulting from drug-induced damage to the blood brain barrier (Lacagnina et al., 2017; Loftis and Huckans, 2013). Activation of microglia results in a number of downstream processes including cell migration to the site of injury and phagocytosis (Hanisch, 2002; Otten et al., 2000), the production of pro-inflammatory factors, such as interleukin (IL)-1β, IL-6, and tumor necrosis factor-α (TNF-α), and the generation of reactive oxygen and nitrogen species that cause neuronal damage (Beardsley and Hauser, 2014). Astrocytes play a critical role in the uptake of synaptically-released glutamate (Cui et al., 2014), are affected by the activity level of dopamine (DA) neurons (Imaizumi et al., 2008), and can shape DA neuron activity and plasticity (Jucaite et al., 2012). Like microglia, astrocytes produce and secrete pro-inflammatory cytokines in response to tissue injury or other insults (Ransohoff and Brown, 2012), including exposure to substances of abuse (Lawson et al., 1990). Thus, excess neurotransmitters (e.g., DA and glutamate) released by drug use may bind to receptors expressed on glial cells and further amplify inflammatory signaling via additional release of cytokines and chemokines, potentially contributing to positive feedback that promotes inflammation.

A number of animal studies have established a link between neuroinflammation and drug exposure (Lacagnina et al., 2017; Loftis and Huckans, 2013), and it is important that work in humans expand on preclinical work to extend the clinical relevance and address the greater complexity of human drug use. The following review will, therefore, focus on findings from human neuroimaging studies of addiction, with an emphasis on positron emission tomography (PET). Although astrocytes are the most abundant type of glial cell in the brain and are affected by substances of abuse (Bull et al., 2015; Cao et al., 2016), there are no techniques to directly quantify astrocyte activation in humans in vivo. Currently, in vivo quantification of glial activation is only available to examine ligand binding of the 18-kDa translocator protein (TSPO), a protein formerly known as the peripheral benzodiazepine receptor and primarily located on the outer membrane of mitochondria. Within the CNS, a variety of cells are capable of expressing TSPO; however, the pattern of expression appears to differ between normal and injured CNS (Cosenza-Nashat et al., 2009). Second generation radiotracers such as [¹⁸F]FEDAA1106, [¹¹C]PBR28, [¹¹C]DPA-713, and [¹⁸F]DPA-714 now provide better specific binding ratios (Imaizumi et al., 2008) and higher test-retest, intra-individual reproducibility (Jucaite et al., 2012) compared with first generation [¹¹C](R)-PK11195. A number of clinical studies show changes in TSPO binding with [¹¹C]PBR28 (Hannestad, 2012), and more studies have begun investigating TSPO levels in addictions, specifically alcohol, nicotine, methamphetamine, and cocaine use disorders. As clinical and preclinical studies have demonstrated a link between immunological cells in blood and activated microglia (Kanegawa et al., 2016), this review will highlight the work conducted using PET as an index for neuroinflammation and also examine relevant work with magnetic resonance imaging linking brain function to peripheral markers of inflammation in substance use disorders. The review includes the limitations of PET imaging as an index of neuroinflammation and concludes with a brief summary of therapeutic strategies that may help target and treat the combination of insults and interactive effects of substance abuse and neuroinflammation.