Inflammation in the pathogenesis of depression: a disorder of neuroimmune origin

Abstract

There are several hypotheses concerning the underlying pathophysiological mechanisms of major depression, which centre largely around adaptive changes in neuronal transmission and plasticity, neurogenesis, and circuit and regional connectivity. The immune and endocrine systems are commonly implicated in driving these changes. An intricate interaction of stress hormones, innate immune cells and the actions of soluble mediators of immunity within the nervous system is described as being associated with the symptoms of depression. Bridging endocrine and immune processes to neurotransmission and signalling within key cortical and limbic brain circuits are critical to understanding depression as a disorder of neuroimmune origins. Emergent areas of research include a growing recognition of the adaptive immune system, advances in neuroimaging techniques and mechanistic insights gained from transgenic animals. Elucidation of glial-neuronal interactions is providing additional avenues into promising areas of research, the development of clinically relevant disease models and the discovery of novel therapies. This narrative review focuses on molecular and cellular mechanisms that are influenced by inflammation and stress. The aim of this review is to provide an overview of our current understanding of depression as a disorder of neuroimmune origin, focusing on neuroendocrine and neuroimmune dysregulation in depression pathophysiology. Advances in current understanding lie in pursuit of relevant biomarkers, as the potential of biomarker signatures to improve clinical outcomes is yet to be fully realised. Further investigations to expand biomarker panels including integration with neuroimaging, utilising individual symptoms to stratify patients into more homogenous subpopulations and targeting the immune system for new treatment approaches will help to address current unmet clinical need.

Keywords: Depression; Inflammation; Neuroimmunology.

Figure 1. Differentiation of CD4 effector T-cell subsets

Upon engaging an infectious agent, dendritic cells (DC) may become activated via their pathogen recognition receptor and phagocytose the pathogen, presenting antigens via the MHC class II pathway. Naïve T-cell activation is initiated through engagement of this antigen: MHC II complex with the T-cell receptor (signal 1), along with co-stimulation through CD28 (T cell) and CD80/86 (antigen presenting cell; APC) binding (signal 2). Upon activation, Th cells undergo clonal expansion and differentiate into one of seven effector subsets (Th1, Th2, Th17, Tfh, Th22, Th9 and pTreg cells) depending on the cytokines present in their microenvironment (signal 3). Signalling induced by these polarising cytokines activates specific STATs and transcription factors, which direct the function of the Th cell type, including production of their signature cytokines. Peripheral Treg cells have a similar phenotype and function to thymic-derived Treg cells, but thymic derived cells do not require APC engagement. Created with BioRender.com.

Figure 2. The relationship between stress and immune function

Stress stimulates the HPA axis causing the hypothalamus to secrete corticotrophin-releasing hormone (CRH). This in turn stimulates the release of adrenocorticotropic hormone (ACTH) from the pituitary gland into the bloodstream. ACTH stimulates the release of glucocorticoid hormones from the adrenal cortex and adrenaline from the adrenal medulla. Stress stimulates the release of noradrenaline from sympathetic nerve endings. Noradrenaline and adrenaline released via SAM axis activation influence immune function. (1) Stimulation of the β-adrenoceptor leads to increased release of anti-inflammatory IL-10 [122]. IL-10 has an anti-inflammatory effect by suppressing synthesis of IL-1α, IL-1β, IL-6, IL-8, TNF-α, GM-CSF and G-CSF [123]. (2) Acetylcholine has been shown to have anti-inflammatory properties, attenuating the release of pro-inflammatory TNF, IL-1β, IL-6 and IL-18 in LPS-stimulated human macrophage cultures, while SAM activation is known to suppress the parasympathetic nervous system, leading to reduced acetylcholine release [124,125]. (3) Activation of α-adrenoceptors in macrophages augments the immune response by increasing production of proinflammatory TNF-α [126]. Furthermore, activation of the β₂-adrenoceptor with adrenaline increases proinflammatory IL-6 production in macrophages [127]. (4) Chronic stimulation of the β-adrenoceptor via SAM activation leads to down-regulation of the receptor leading to adaptive changes [128]. Glucocorticoids also have the capacity to modulate immune function. (5) Glucocorticoids inhibit phospholipase A₂ which limits the conversion of cell membrane phospholipids to arachidonic acid and then into proinflammatory eicosanoids [118]. (6) Activation of the glucocorticoid receptor causes modulation of transcription of glucocorticoid response elements which have an immunosuppressive effect [115]. (7) Glucocorticoid receptor activation disrupts NFκB coactivator activity and prevents NFκB-mediated activation of inflammatory genes [120]. (8) Chronic HPA axis activation leads to glucocorticoid resistance and adaptive changes in immune function [120]. Chronic stress can also lead to adaptive immune changes through (9) ageing of the immune system [129] and (10) epigenetic modification of transcription of immune- and stress-related genes [130]. Stress also leads to neuronal activation in fear and threat appraisal centres in mouse stress models (marked by increased cFos and ∆FosB) which coincides with microglial activation. This neuronal and immune activation also recruits peripheral immune cells into the CNS, leading to exacerbation of existing neuroinflammation [50]. Created with BioRender.com.

Figure 3. Reactions affected by tetrahydrobiopterin (BH4) availability

BH4 acts as a cofactor for nitric oxide synthase, phenylalanine hydroxylase, tyrosine hydroxylase and tryptophan hydroxylase, which catalyse the synthesis of nitric oxide, tyrosine, L-DOPA and 5-hydroxytryptophan (a precursor to serotonin) respectively. Created with BioRender.com.

Figure 4. Action of KP metabolites at the NMDA receptor

(1) Binding of NMDA receptor antagonist KYNA prevents activation of the NMDA receptor and subsequent calcium influx. (2) Binding of NMDA receptor agonists glutamate or QUIN activate the receptor, causing the ion channel to open (if the membrane is also depolarised) and facilitating the flow of Ca²⁺ into the cell. (3) Increase in intracellular Ca²⁺concentration activates Ca²⁺/calmodulin-dependent protein kinases (CaMK) which in turn (4) regulate the activity of neuronal NOS (nNOS) which produces NO [204]. (5) Downstream nNOS-dependent activation of protein kinase G (PKG) increases activity of NADPH oxidase in the cell membrane [205]. (6) NADPH oxidase activity in turn mediates superoxide production [205]. (7) Superoxide reacts with water to form hydrogen peroxide which passes through the membrane and accumulates intracellularly [206]. (8) Oxidative KP metabolites also generate ROS which further enhance the production of hydrogen peroxide. (9) Accumulated hydrogen peroxide in turn has the capacity to activate cytosolic phospholipase A₂, increasing the production of arachidonic acid [207]. (10) Mitochondrial metabolism of arachidonic acid results in the formation of additional superoxide, hydrogen peroxide and cytochrome c. (11) Hydrogen peroxide causes oxidative damage to cellular molecules including lipids, proteins and DNA, leading to cell death. (12) Cytochrome c activates cellular caspases resulting in apoptosis. (13) Hydrogen peroxide facilitates the calmodulin-mediated dephosphorylation of calcineurin and protein phosphatase 1 (PP1) leading to decreased AMPA receptor activity [206]. (14) Calmodulin also dephosphorylates CaMKII, leading to downregulation of the NMDA receptor. Abbreviations: 3-HK, 3-hydroxykynurenine; 3-HAA, 3-hydroxyanthranillic acid; QUIN, quinolinic acid. Created with BioRender.com.