Stress can have profound effects on physiological, molecular, and neuronal mechanisms that lead to long-lasting changes in behavior. Physiological and psychological stressors activate multiple systems simultaneously, with one of the most immediate responses being increased noradrenergic signaling. In the periphery, norepinephrine is released from postganglionic neurons that regulate the sympathetic nervous system to increase heart rate and prime the organism for “fight or flight” (42). The primary source of norepinephrine in the brain is the locus coeruleus (6). Here, norepinephrine contributes to regulating a number of cognitive functions relevant to ensuring immediate survival including arousal, attention, and cognitive flexibility (43, 44). Sympathetic tone, assessed by increased heart rate variability, is elevated in individuals suffering from depression (45), anxiety (46), and post-traumatic stress disorder (PTSD) (47). Increased sympathetic tone is a key feature of the hyper-arousal symptom cluster in PTSD (47). Locus coeruleus activity correlates with stress and stress vulnerability as it predicts anxiety and depression scores (48). In rodents, the locus coeruleus is activated by a wide range of acute or chronic stressors (49, 50). Notably, locus coeruleus activity can be increased by corticotropin releasing factor (CRF) (51, 52), an initiator of the neuroendocrine branch of the stress response (33), which is activated by stress in parallel with noradrenergic signaling.

The hypothalamic–pituitary–adrenal (HPA) axis represents the neuroendocrine branch of the stress response. Here, a stressor increases the activation of parvocellular neurons in the paraventricular nucleus of the hypothalamus (PVN), which release CRF (53), thereby initiating HPA axis activation. PVN neurons project to blood vessels in the median eminence, which deliver CRF to the anterior pituitary. CRF binding to its receptors on corticotropic cells in the anterior pituitary triggers the calcium-dependent cleavage of adrenocorticotropic hormone (ACTH) from pro-opiomelanocortin (54). ACTH is released into blood circulation where it binds to MC2 and 3 receptors in the zona fasciculata of the adrenal gland (55). These G protein coupled receptors increase intracellular cyclic adenosine monophosphate (cAMP) concentrations, leading to the activation of protein kinase A and increased expression of enzymes responsible for converting cholesterol to glucocorticoids like corticosterone (rodents) and cortisol (humans) (56). Corticosterone binds to mineralocorticoid (MRs) and glucocorticoid receptors (GRs) (57, 58), which are expressed in a wide range of cell types including neurons and immune cells in the brain and periphery (56). Once bound and activated, GRs bind to glucocorticoid response elements throughout the genome and recruit co-activators or co-repressors to modulate gene expression (39). MRs have a higher affinity for corticosterone than glucocorticoid receptors do (56, 59). Because of this, mineralocorticoid receptors are saturated at baseline glucocorticoid levels, so the effects of stress-induced glucocorticoids are primarily mediated by GRs (57). GRs regulate a number of functions including, but not limited to, altering the structure and function of neurons (60–62), inducing metabolic changes (63), and reducing inflammation (38).

Thus, in humans and rodents, chronic stress activates both noradrenergic signaling and the HPA axis. Since a major result of noradrenergic signaling is an increase in inflammatory markers, an important role for GR signaling is the compensatory reduction of inflammatory processes (39). These systems are generally balanced in response to mild or acute stress. In response to chronic stress, the anti-inflammatory effects of glucocorticoids can become less effective. This, along with excess noradrenergic signaling, is believed to contribute to the increased inflammatory cytokine levels observed in stress-related mood disorders.

Acute stressful experiences result in increased inflammatory marker expression and other neurophysiological changes. The Trier Social Stress Test (TSST) is commonly used as an acute psychological stressor in humans (64). In this public speaking task, participants are required to provide an interview-style presentation, followed by a surprise mental arithmetic test in front of a panel that does not provide feedback or encouragement. Indeed, the TSST increases HPA axis activation (65) and noradrenergic signaling as the sympathetic nervous system is activated (66). The TSST is associated with increased transcriptional activity of nuclear factor kappa B (NFkB) (35, 67). NFkB is a master transcriptional regulator of inflammatory processes. Following its activation via translocation to the nucleus, NFkB increases the expression of a wide range of inflammatory transcripts while suppressing anti-inflammatory gene expression through multiple mechanisms, including inhibiting GR-mediated transcription (68). Increased inflammation during the TSST is likely due to noradrenergic signaling regulated by the sympathetic nervous system as norepinephrine increases NFkB-mediated transcription (35). NFkB activity inversely correlates with cortisol levels during the TSST (67), which may be attributed to the direct and indirect mechanisms by which GRs inhibit NFkB (39, 69). GRs suppress the activity of NFkB (68–70) via direct binding that prevents NFkB interactions with DNA, by recruiting co-repressors to pro-inflammatory transcriptional sites, through competition for co-activators of NFkB, and by increasing the expression of proteins that inhibit NFkB function (41, 71). NFkB, in turn, inhibits GRs by similar mechanisms (69).

A meta-analysis of 34 studies measuring circulating inflammatory markers indicate that in individuals exposed to acute stress under laboratory conditions, levels of interleukin (IL)-1β, IL-6, and tumor necrosis factor α (TNFα) are increased (72). Acute stressors in this meta-analysis included recall tasks, mock interviews, and Stroop tasks along with being put under social evaluative stressors like the TSST. Peripheral blood samples taken before and after stressors revealed an increase in inflammatory markers (72). Acute stress also increases inflammatory processes in other peripheral fluids such as gingival crevicular fluid, in which IL-1 β is increased (73).

Single and multiple stressors also increase inflammatory markers in rodents. In established rodent stress models, such as social defeat stress, inflammatory processes are activated in the brain and periphery. A single social defeat session in the resident-intruder paradigm increases IL-6 mRNA expression in the medial prefrontal cortex (mPFC), while increases in IL-1β and TNFα are only observed in the mPFC after at least 7 days of stress (28). The mPFC is an important stress-related brain region as it provides top-down negative control of the HPA axis (74) and the amygdala (75), a brain region important for fear and anxiety. Therefore, the mPFC is important for suppressing neuroendocrine and behavioral responses to stress. The mPFC is also important for cognitive flexibility (76) and resilience to stress (77). We demonstrated that IBA1+ cell densities and TNFα are increased in the mPFC of vulnerable rats, which display increased anxiety- and depression-like behavior following 7 days of social defeat in the resident-intruder paradigm. Levels of these inflammatory markers in the mPFC of resilient rats, which behave similarly to non-defeated controls, are lower and similar to those of control rats (29). Indeed, chronic administration of infliximab, a TNFα antagonist, reduces anxiety- and depression-like behavior following chronic stress in rodents (78).

Mood disorders associated with chronic stress are also associated with increases in inflammatory markers in the blood (18, 20, 27, 79, 80) and brain (27). Veterans with PTSD from combat exposure have higher pro-inflammatory cytokine levels than veterans that did not develop PTSD following combat (80). The inflammatory markers TNF-α and IL-6 are also higher in the blood of individuals with major depression (79), and generalized anxiety disorder (GAD) (19, 20). In suicide victims, mRNA and protein levels of the pro-inflammatory cytokines IL-6, IL-1β, TNF-α and lymphotoxin A were increased in the prefrontal cortex (PFC) (27). Individuals who have had autoimmune diseases that can dysregulate the immune response are 45% more likely to develop depression (81). Therefore, chronic stress disorders are associated with increased inflammation and auto-immune disorders can lead to stress-related mood disorders in humans. These increases in inflammatory cytokines are not only symptomatic of immune dysfunction and systemic inflammation, but contribute to symptoms of depression.

Inflammatory cytokines are not only increased by stress but contribute to the pathology of stress-related mood disorders. A broad range of social (8), anatomical (15), and molecular (82, 83) factors contribute to the onset and development of stress-related mood disorders. It is well-established that inflammatory markers are increased in the blood (79) and brain (27) of individuals afflicted with stress-related mood disorders. Moreover, blocking inflammation can reduce phenotypes associated with stress-related mood disorders (24, 25), whereas increasing inflammatory processes can contribute to the onset and development of stress-related mood disorders (84, 85).

For example, a meta-analysis of 14 trials with 6,262 participants revealed that pharmacological cytokine inhibitors or non-steroidal anti-inflammatory drugs (NSAIDS) can reduce depression scores in adults with depressive symptoms compared to placebo controls (24). A subsequent meta-analysis showed similar anti-depressant effects of NSAIDs in trials that included both men and women (86). In a randomized trial of participants with treatment-resistant depression, the TNFα inhibitor infliximab reduces depression scores, but only in the subpopulation of patients displaying elevation of the inflammatory marker C-reactive protein (25). Similar findings have been observed in preclinical studies as weekly intraperitoneal injections of infliximab reduce depression- and anxiety-like behavior in rats following an 8-week chronic mild stress paradigm (78). Therefore, anti-inflammatory treatments reduce symptoms of depression in certain human populations and rodents.

Conversely, treatments using exogenous cytokines increase symptoms of depression. Interferon-α (IFNα) exacerbates symptoms of depression when used to treat hepatitis C (84). IFNα also increases symptoms of depression when used to treat melanoma (87, 88). SSRIs can reduce symptoms of depression in hepatitis C and cancer patients receiving IFNα treatments (89, 90), suggesting that IFNα might increase depression severity by reducing serotonergic neurotransmission. In rodents, induction of inflammatory processes using intraperitoneal lipopolysaccharide (LPS) administration increases depressive-like behavior in mice as characterized by immobility in the forced swim test and tail suspension test (91). Therefore, exogenous cytokines are sufficient to induce symptoms of depression in humans and mice.

Studies in the absence of any treatments also support a relationship between stress-related mood disorders and immune dysfunction as depression and autoimmune disorders share comorbidity, common symptoms, and inflammatory processes (92). Diseases characterized by increased peripheral inflammation, like rheumatoid arthritis (93) and sepsis (81) have been linked to an increased risk for depression. Together, these findings indicate that inflammatory processes contribute to symptoms of depression in certain individuals and are sufficient to induce or exacerbate depressive symptoms.

Stress-induced noradrenergic signaling can promote inflammatory processes (35, 37, 94). Noradrenergic signaling can be pharmacologically inhibited using the α1 adrenergic receptor antagonist prazosin or the β adrenergic receptor antagonist propranolol. The α1 adrenergic receptors are G_q/11 protein coupled receptors (GPCRs) that activate phospholipase C β, which increases concentrations of diacylglycerol and inositol trisphosphate, thus activating protein kinase C signaling and increasing intracellular calcium concentrations. The β adrenergic receptors are G_sPCRs that activate adenylyl cyclase, converting adenosine triphosphate to 3′,5′-cyclic adenosine monophosphate (cAMP), which activates protein kinase A signaling (95). Norepinephrine, but not epinephrine, increases peripheral inflammation as it induces NFkB-mediated transcription in mononuclear cells (35). Indeed, antagonism of α1 adrenergic receptors using prazosin can reduce inflammatory markers in the blood of PTSD patients (96). In mice, the β adrenergic receptor antagonist propranolol mitigates social defeat-induced microglial reactivity in the prefrontal cortex, the hippocampus, and the amygdala (37) and foot shock-induced increases in IL-1β in the hypothalamus (36). Each of these brain regions receive noradrenergic input from the locus coeruleus (97). Propranolol also prevents stress-induced IL-6 elevation in healthy young adult humans (40). PTSD symptom severity is associated with both increased sympathetic tone and inflammation (98), which is consistent with systemic inflammation being driven by noradrenergic neurotransmission in humans. Together, these findings indicate that noradrenergic neurotransmission, which is increased by stress in the brain and periphery, increases peripheral and central inflammatory processes.

The context of psychological stress might be an important factor contributing to the pro-inflammatory effects of norepinephrine. In response to certain inflammatory stimuli, such as exposure to lipopolysaccharides, norepinephrine can provide anti-inflammatory effects (99). Norepinephrine can also inhibit microglia reactivity, at least in certain mouse models of Parkinson’s disease (100). Generally, G_q-coupled α1 and G_i-coupled α2 adrenergic receptors are thought to mediate pro-inflammatory effects (101). G_s-coupled β adrenergic receptors may mediate anti-inflammatory effects by inhibiting nuclear translocation of NFkB through activation of protein kinase A induced by cAMP (102, 103). However, β adrenergic receptors have been proposed to adopt non-canonical mitogen-activated protein kinase (MAPK) pathways that can increase inflammatory processes (104). These findings underscore the complex nature by which norepinephrine affects inflammation. Additional work is needed to better understand the contexts, molecular states, and immune cell types in which noradrenaline primarily exerts anti- vs. pro-inflammatory effects.

Norepinephrine is the most extensively studied signaling molecule to be implicated in stress-induced inflammation, although other singling molecules increased by stress also contribute to inflammatory processes. Stress-induced CRF release has been studied most extensively in PVN neurons. However, stress can also increase CRF release in the prefrontal cortex (105), amygdala (106), and periphery (107, 108). In the periphery, CRF increases inflammation through direct interactions with macrophages in cell lines derived from mouse myeloma cells (107). CRF potentiates the pro-inflammatory effects of LPS as TNF-α, IL-1β, and IL-6 concentrations are higher in the presence of CRF and LPS compared to LPS alone (107). The neuropeptide substance P is increased by certain stressful stimuli, like loud sounds (109) and cold water (110). Substance P stimulates cytokine secretion from peripheral immune cells during stress (110) and in the absence of stress (111). Substance P also exerts pro-inflammatory effects in the central nervous system as it increases nuclear translocation of NFkB in cultured microglia (112). Pro-inflammatory effects of substance P in the periphery and central nervous system have been previously reviewed (113). Therefore, like noradrenergic signaling, stress-induced CRF and Substance P signaling can promote inflammatory processes in the periphery and brain.

Stress increases the expression and release of inflammatory molecules, which modulate neuronal activity to affect mood and behavior. Inflammatory processes can induce sickness behavior, characterized by anhedonia, reduced motivation, lethargy, reduced social interaction, and altered sleep architecture. These behaviors share similarity with symptoms of MDD (22). In addition to central inflammatory processes induced by locus coeruleus-derived noradrenergic signaling, peripheral inflammation resulting from sympathetic nervous system activity can also affect the brain. While peripherally and centrally derived cytokines likely affect neuronal activity similarly, peripheral cytokines must first cross the blood–brain barrier (BBB). Peripheral cytokines can access the brain through circumventricular organs, specific cytokine transporters, recruitment of peripheral immune cells by reactive microglia, a leaky BBB caused by stress and/or prior inflammation, and other mechanisms. These mechanisms have been reviewed extensively by others (22, 23, 26, 114).

Cytokines have direct effects on neuronal activity. In response to inflammation, interferons and interleukins are released and levels of indoleamine 2,3 deoxygenase (IDO) are increased. IDO is responsible for catabolizing tryptophan into kynurenine, which further catabolizes into metabolites like quinolinic acid in microglia and kynurenic acid in astrocytes (30, 31, 115). Because tryptophan is a key precursor for serotonin, this reduces the amount of tryptophan available for serotonin synthesis, thus reducing levels of serotonin in the brain (31). Quinolinic acid can have excitotoxic effects via activation of N-methyl-D-aspartate (NMDA) receptors whereas kynurenic acid can antagonize NDMA receptors and α7 nicotinic acetyl choline receptors (30). Inflammatory cytokines can also reduce serotonergic neurotransmission by promoting serotonin reuptake as IL-1 β increases serotonin transporter (SERT) activity (116). These mechanisms, especially those reducing serotonergic neurotransmission, can promote depression- and anxiety-like behavior as the role of serotonin in regulating mood is well-documented (114).

Increased serotonergic neurotransmission can facilitate neurogenesis in the subgranular zone of the dentate gyrus (117), which is important for the antidepressant effects of selective serotonin reuptake inhibitors (SRRIs) (118). Indeed, stress-induced IL-1 β release contributes to depression-like behavior and reduced neurogenesis as pharmaceutical inhibition or genetic knockout of the IL1 receptor prevents anhedonia-like behavior and anti-neurogenic effects in a mouse model of chronic stress (119). The inflammatory cytokine interferon-α can also reduce dopamine levels by reducing levels of one its precursors, tyrosine, and metabolizing dopamine directly (120). Several other neurotransmitters can also be impacted via similar mechanisms to control mood and brain function (114).

The neural substrates underlying pain represent an important, and often overlooked, system that can cause a positive feedback loop that increases inflammation during stress. A heightened sense of pain represents one of the most prominent symptoms caused by inflammation. Pain is an unpleasant sensory and emotional experience caused by actual or potential tissue damage. While the sensation originates with activation of nociceptors throughout the periphery, the signals are sent to the brain via the spinothalamic tract and further shaped by supraspinal brain regions that encode it with a negative emotional valence (121). Interestingly, many of the stress-related brain areas such as the locus coeruleus, PVN, hippocampus, amygdala, and mPFC are also involved in pain processing and play a role in the transition to chronic pain (122–125). The locus coeruleus, which provides analgesic effects during acute pain, has been proposed to contribute to allodynia and hyperalgesia in the context of chronic pain (125). Despite projections of nociceptive pathways to the PVN, chronic pain models do not elicit changes in HPA axis activation. However, chronic pain causes changes in limbic regions associated with stress and mood (126). Indeed, reduced hippocampal volume is a biomarker for both chronic pain (127, 128) and depression (129). It is unclear whether reduced hippocampal volume associated with chronic pain is attributed to nociceptive spinothalamic pathway projections or secondary effects caused by increases in inflammatory cytokines, which are known to reduce neurogenesis and dendrite growth in the hippocampus (130). Regardless of the mechanism(s) responsible, reduced hippocampal volume caused by pain likely contributes to aspects of depression due to the importance of hippocampal volume in promoting stress resilience (15). Conversely, chronic pain increases the activity of the amygdala, which regulates fear and anxiety-like behavior. This is at least partially due to dysfunction of GABAergic neurons within the amygdala, allowing for enhanced plasticity and facilitated excitatory inputs from the parabrachial nucleus, which transmits nociceptive information from the periphery. Heightened activity within the amygdala also contributes to inhibition of the mPFC (131). The mPFC is involved in a wide range of functions including attention (132), cognitive flexibility (76), stress resilience (29, 77) and negative regulation of the HPA axis (74). Therefore, pain modulates the activity of multiple stress-related brain regions that might contribute to depression- and anxiety-like behavior.

While nociceptors can be activated by noxious chemical, thermal, and mechanical stimuli at baseline, local inflammation from injury or infection can sensitize them. Mast cells, macrophages, and keratinocytes release inflammatory mediators such as TNF-α, IL-1β, and IL-6 that increase infiltration of immune cells that further promote inflammation (133). These mediators increase hyperactivity of nociceptors by activating intracellular signaling pathways that increase sensitivity and expression of sodium channels and nociceptor-specific ion channels such as the Transient Receptor Potential Vanilloid Type 1 (TRPV1) receptor and Transient Receptor Potential Ankyrin Type 1 (TRPA1) receptor (134). Indeed, initiating these inflammatory processes with the injection of formalin, complete Freund’s adjuvant (CFA), carrageenan, or an inflammatory soup (i.e., an experimental mixture of serotonin, bradykinin, prostaglandin E2, and histamine) results in acute and long-term pain-related behavior in rodents (135, 136). Interestingly, this inflammatory sensitization at the periphery is bidirectional as substance P and calcitonin gene-related peptide (CGRP), which are important for neurotransmission in nociceptive neurons, further increase inflammation (137). Indeed, substance P increases inflammatory markers in the periphery (110, 111) as well as cultured brain cells (112, 113). Even in the absence of tissue injury, aberrant neuronal signaling can induce inflammation and swelling of tissue (138). Therefore, cross-talk between the immune system and nociceptive neurons is reciprocal, which might lead to positive feedback cycles that exacerbate inflammatory processes in chronically stressed individuals.

Repeated activation of these nociceptors under chronic pain conditions can also cause microglia reactivity, leading to central sensitization of nociceptive pathways. Here, microglia reactivity is thought to occur due to the release of the chemokine CCL2 from the chronically active nociceptor (134). Once reactivate, microglia enter a state of microgliosis and release pro-inflammatory mediators such as IL-1β and TNF-α. These sensitize the pre-synaptic nociceptor and post-synaptic second-order neuron while also activating astrocytes that further release pro-inflammatory cytokines. Given the involvement of microglia and astrocytes in the maintenance of the BBB, chronic inflammatory pain leads to an increase in BBB permeability (139), allowing peripheral cytokines to access the brain parenchyma more easily. Interestingly, these neuroinflammatory processes do not appear to occur under acute conditions and thus may be important for the maintenance of pain under chronic conditions (140).

Understandably, treating inflammatory pain with non-selective COX inhibitors and other non-steroidal anti-inflammatory drugs (NSAIDs) is considered first-line therapy for clinicians (141). Other treatments such as infliximab, glucocorticoids, and minocycline are also effective for pain, presumably through modulating peripheral and central inflammatory processes (142–145). Surprisingly, however, several other therapeutics that target stress-related systems also show efficacy in treating inflammatory pain. Topical prazosin reduced allodynia and hyperalgesia in patients with complex regional pain syndrome (146). Propranolol reduced pain in a larger percentage of patients with temporomandibular disorder (TMD) in a randomized placebo-controlled clinical trial and reduced temporomandibular joint pain in a female rats via anti-inflammatory mechanisms (147, 148). Antidepressants including monoamine oxidase inhibitors, tricyclic antidepressants, monoamine reuptake inhibitors, and glutamatergic antidepressants have all shown efficacy in treating multiple forms of chronic pain (149). Indeed, it is difficult to separate the psychological components of stress, depression, and pain when seeking to understand the mechanisms behind therapeutic strategies that have efficacy in all three disorders. Given that these all these antidepressants affect inflammatory mechanisms, however, it suggests stress-induced inflammation could be a key contributor to their shared efficacy (150).

Impairments in the anti-inflammatory mechanisms regulated by GRs can lead to systemic inflammation in chronically stressed individuals. Negative feedback of the HPA axis prevents continuous activity of itself during periods of prolonged stress. Here, negative feedback is mediated by glucocorticoids, which bind to glucocorticoid receptors in multiple tissues including the anterior pituitary (151), PVN (152), and PFC (74). Negative feedback of the HPA axis is overactive in individuals with PTSD, resulting in lower levels of circulating glucocorticoids in the absence of stress (153). In some depressed individuals, glucocorticoid resistance is observed as dexamethasone-induced anti-inflammatory effects are less efficacious (154, 155). Glucocorticoid resistance is widely believed to be an important factor contributing to systemic inflammation observed in stress-related psychiatric disorders. This impaired ability of glucocorticoid receptors to exert anti-inflammatory processes can leave pro-inflammatory mechanisms, such as increased noradrenergic signaling, to go unchecked. Notably, cytokines can also contribute to glucocorticoid resistance (156), causing a positive feedback loop that exacerbates inflammatory processes during chronic stress.

A number of factors can cause glucocorticoid resistance. Glucocorticoid signaling can be impaired by DNA methylation of the gene encoding GRs, thereby reducing GR expression (157). Additionally, previous exposure to stress or glucocorticoids reduces the affinity of GRs for glucocorticoids during subsequent exposure to glucocorticoids (158). Glucocorticoid resistance can also be caused by impaired nuclear translocation of GRs. GRs that are not bound to cortisol or corticosterone are retained in the cytoplasm as part of a complex that includes multiple proteins including, but not limited to, heat shock protein 70 (HSP70), p23, and FK506-binding protein 51 (FKBP5) (159). Upon binding of GRs to cortisol or corticosterone, FKBP5 is replaced with FKBP4, which interacts with the motor protein dynein. This allows for the translocation of the GR-bound complex toward the nucleus along microtubules (160, 161). Once in the nucleus, GRs regulate the expression of thousands of genes, including FKBP5. Induction of FKBP5 by GRs allows for a short negative feedback loop whereby GR-mediated transcription impairs GR nuclear translocation. Stress increases FKBP5 expression in the rodent prefrontal cortex and hippocampus, which impairs nuclear translocation of GRs and the expression of genes regulated by GRs (162). Thus, FKBP5 is considered to be an important factor contributing to glucocorticoid resistance (163). Interestingly, these molecular mechanisms involving FKBP5-mediated glucocorticoid resistance are returned to normal expression and function following SSRI treatment (162). Therefore, this mechanism might contribute to the antidepressant effects of SSRIs.

Glucocorticoid resistance can also be caused by NFkB, which impairs the ability of GRs to bind DNA and regulate gene expression (38, 164). NFkB can suppress GR function via direct binding that prevents GR binding to DNA, by outcompeting GRs for co-activators, and by recruiting co-repressors to GR-binding sites (69, 70). GRs inhibit NFkB via similar mechanisms (41, 71, 165, 166). Therefore, pro- and anti-inflammatory mechanisms that are activated by stress are normally in balance. In chronically stressed individuals, this reciprocal inhibition can be disturbed. This can lead to glucocorticoid resistance and positive feedback cycles that increase inflammatory processes (Figure 1).

Aberrant glucocorticoid signaling is a hallmark of stress-related psychiatric disorders. MDD patients display increased salivary cortisol in the first hour of wakefulness, but salivary cortisol concentrations are similar to controls within 8 h of wakefulness (167). Because the glucocorticoid resistance experienced by individuals with MDD is likely continuous, overall glucocorticoid signaling might be reduced in depressed individuals, despite increased cortisol early in the day. In a randomized trial, dexamethasone was shown to improve Hamilton Depression Rating Scale scores in depressed patients compared to placebo over the course of 4 days (168), suggesting that impaired GR activation might contribute to symptoms of depression in some MDD patients. PTSD is characterized by reduced baseline cortisol levels, (169) which can be attributed to hypersensitivity to glucocorticoid-mediated negative feedback of the HPA axis (153). Low-dose dexamethasone treatment can reduce inflammatory markers in PTSD patients and reduce the severity of certain PTSD symptoms (19).

Recently, we found that GR expression is increased in the mPFC of rats resilient to the adverse behavioral effects of social defeat. We found that a specific target of GRs, sphingosine-1-phosphate receptor 3 (S1PR3), prevented anxiety- and depression-like behavior by mitigating stress-induced inflammation. S1PR3 mRNA was also reduced in the blood of veterans with PTSD (29), which is consistent with increased inflammatory markers in PTSD. Conditional knockout of GRs in the forebrain causes depression-like behavior in rodents (170). Of course, GRs reduce inflammatory processes in the periphery and brain through regulating many transcriptional targets. Inflammatory cytokines can affect neuronal activity, and likely mental health, by altering neurotransmission. Some of the adverse effects on mood caused by inflammatory cytokines may also involve pain, which has both physiological and psychological components. While glucocorticoids reduce pain caused by inflammation (145) and can improve measures of mental health at certain doses in some individuals (19, 168), whether glucocorticoids improve mental health by alleviating pain has not been directly investigated. Impaired GR function and circulating glucocorticoid levels outside the optimal range can contribute to symptoms of depression and PTSD. Of course, chronic increases in circulating glucocorticoid concentrations can contribute to a wide range of adverse biological and psychological effects, but glucocorticoid signaling also has positive effects on mental health in certain contexts. This includes, but is not limited to, mitigating stress-induced inflammatory processes prior to the development of glucocorticoid resistance.

Decades of research has identified many key molecular mechanisms underlying the interplay among stress, inflammation, pain, and behavior. The nervous system is central to this interplay, but we have yet to identify many of the circuits responsible for this interplay. For example, we know that noradrenergic neurotransmission contributes to inflammatory processes in immune cells of the blood and brain using pharmacology. These noradrenergic neurons are likely postganglionic sympathetic neurons and locus coeruleus neurons. However, it is possible that certain subregions of these noradrenergic centers are more important for stress-induced inflammatory processes than others. For example, thoracic, but not lumbar, sympathetic neurons might underlie stress-induced inflammation. We know that norepinephrine induces microglia reactivity and cytokine production and assume this norepinephrine is derived from the locus coeruleus, but this has yet to be directly demonstrated. We also do not know if the pro-inflammatory effects of norepinephrine in microglia are due to NFkB-mediated transcription as they are in peripheral immune cells.

The activity of noradrenergic neurons and specific neural circuits can be manipulated using chemogenetic or optogenetic methods. In particular, viral vectors expressing hyperpolarizing hM4D and depolarizing hM3D Designer Receptors Exclusively Activated by Designer Drugs (DREADDs) under the control of the PRSx8 promoter have been developed (171). These constructs allow for decreasing or increasing the activity of noradrenergic neurons within heterogenous neuronal populations. For the manipulation of neural circuits, Cre-dependent DREADDs can be used in combination with CAV2-Cre, which retrogradely expresses Cre recombinase. Here, Cre-dependent DREADDs are injected into a particular brain region and CAV2-Cre is injected in a target region of choice (172, 173). This allows DREADDs to be expressed in a subpopulation of neurons that project to a specific brain region. This technology is also an excellent tool for specifically targeting pain-related circuits, especially when used in combination with nociceptive neuron-specific promoters, such as the substance P promoter. Of course, optogenetic approaches can also be used, although DREADDs are typically better suited for stress paradigms as they avoid the use of cumbersome and sensitive lasers. When using Cre-dependent DREADDs, it is important to consider collateral projections, as these will also be inhibited. For example, using this system to chemogenetically inhibit mPFC-projecting locus coeruleus neurons will inhibit the entire locus coeruleus neuron, which may also innervate other brain regions via collateral projections. This can be avoided using optogenetic approaches, although this might not be conducive to stress paradigms involving physical contact between rodents.