Intertwined associations between oxytocin, immune system and major depressive disorder

Abstract

Major depressive disorder (MDD) is a prominent psychiatric disorder with a high prevalence rate. The recent COVID-19 pandemic has exacerbated the already high prevalence of MDD. Unfortunately, a significant proportion of patients are unresponsive to conventional treatments, necessitating the exploration of novel therapeutic strategies. Oxytocin, an endogenous neuropeptide, has emerged as a promising candidate with anxiolytic and antidepressant properties. Oxytocin has been shown to alleviate emotional disorders by modulating the hypothalamic-pituitary-adrenal (HPA) axis and the central immune system. The dysfunction of the immune system has been strongly linked to the onset and progression of depression. The central immune system is believed to be a key target of oxytocin in ameliorating emotional disorders. In this review, we examine the evidence regarding the interactions between oxytocin, the immune system, and depressive disorder. Moreover, we summarize and speculate on the potential roles of the intertwined association between oxytocin and the central immune system in treating emotional disorders.

Graphical Abstract

1. Introduction

Major depressive disorder (MDD), a stress-related disorder, is characterized by sadness, emptiness, anhedonia, disturbance of sleep, and appetite, and is accompanied by some physical and cognitive dysfunctions [1]. MDD is one of the most prevalent psychiatric disorders, with an annual prevalence of up to 6% worldwide [1]. Due to the ongoing COVID-19 pandemic, the annual prevalence of MDD has increased by approximately 25.6% worldwide [2].

For past decades, a lot of studies were dedicated to exploring the neural mechanisms underlying MDD and developing possible antidepressants. Historically, MDD was thought to be caused by the dysfunction of the central monoaminergic system [3], so some first-line antidepressants targeting the central monoaminergic system, including monoamine oxidase inhibitors (MAOIs), selective serotonin reuptake inhibitors (SSRIs) and Tricyclic antidepressants (TCAs), have been discovered and used in clinic [1]. However, current therapeutics only alleviate the symptoms of ∼1/3 of MDD patients and are partially useful for some symptoms in another 1/3 [4]. Antidepressant treatment outcomes that are unsatisfactory due to a lack of understanding of depression mechanisms. Emerging evidence suggests that immune system dysfunctions play a role in the development of depressive disorders [5], [6].

In humans, MDD is mainly induced by different stressors (e.g., psychosocial stress and biological stress), which cause different pathogeneses and result in diverse episodes of depression [1]. In the laboratory, animal models of depression are mainly built by three means: long-term chronic stress (e.g., chronic social defeat stress, early life stress, and chronic mild stress), corticosterone supplementation, and systemic inflammatory stimulation (e.g., administration of lipopolysaccharide or interferon-α) [7]. However, none of these models is able to fully reproduce the MDD symptoms observed in humans, which indicates that multiple mechanisms are involved in the disease.

Oxytocin (OXT), an endogenous neurohormone/neuropeptide, mainly binds to oxytocin receptors (OXTRs) to exert biological effects. OXTRs are found to be widely expressed in the hypothalamic-pituitary-adrenal (HPA) axis, neurons, astrocytes, and microglia in the central nervous system (CNS) [8]. OXT has been shown to modulate central immune activity [9]. OXT and OXTR is extensively altered in depressive disorders [10], [11], [12], [13], and exogenous OXT administration has been shown to alleviate depression by modulating the stress-HPA axis-immune crosstalk [14]. As such, intervening with OXT in the CNS may be one potential treatment for MDD in clinic.

In order to better understand the interactions between OXT, the immune system, and depressive disorders, we summarize current studies about the dysfunctions of the OXT system and the central immune system in depression disorders in both humans and animals, and discuss the potential anti-depressive roles of intervening central OXT and immune signals in this review.

2. Antidepressant actions of OXT in animal preclinical studies and clinical studies in humans

2.1. Animal-based studies

Since the first report by Meisenberg [15] in 1981, a number of animal studies have demonstrated the antidepressant effects of oxytocin (OXT). Intracerebroventricular (ICV) administration of OXT has been shown to decrease mouse immobility in the forced swim test [15]. Subsequent studies have confirmed that OXT administration, whether systemically or locally in the brain, can improve reduced immobility time in the forced swimming test (FST) in depressive animals [13], [16], [17], [18], [19], [20]. The antidepressant effect of OXT has been found to be comparable to that of imipramine [16]. In addition to OXT, other medications can also target OXT/OXTR to produce antidepressant effects. For example, Sildenafil, a drug used to treat erectile dysfunction in men, can have antidepressant effects by increasing the release of OXT[21]. Moreover, OXT's antidepressant effect is dependent on its binding to OXTR [22], [23], [24].

2.2. Human-based studies

Due to the consistent beneficial effects of OXT on depression in animal experiments, many researchers have attempted to use OXT for treating depression or depressive symptoms associated with diseases. However, the clinical application results are inconsistent.

Some studies show that OXT doesn't help with depression, but instead makes people feel worse. Early OXT administration had no effect on short-term postpartum depression in nulliparous women with primary labor dysfunction [25]. In major depression, OXT not only showed no significant differences in behavioral and vital signs, but also increased subjective anxiety [26]. Intranasal OXT impaired the ability of students with depressive symptoms to ignore task-irrelevant sad facial expressions, indicating that OXT administration may increase the risk of depression [27].

However, most researchers believe that OXT can alleviate depression directly or indirectly. Intranasal OXT twice daily for four weeks significantly improved clinical scores in depressive patients [28]. Intranasal OXT administration for ten days resulted in improvements in the symptoms of depression in schizophrenia [29]. More research supports the idea that OXT can be used as an adjunct to antidepressant treatment or to alleviate some symptoms of depression by decreasing responsiveness to stressful behaviors. In the postpartum depression, OXT did not improve the symptom of depression, but their perception of the relationship with their baby improved [30], [31], [32]. In depressive episodes and recurrent depressive disorder, OXT reduced the initial allocation of attention to angry faces and increased sustained attention to happy faces [33]. In chronic depression disorders, OXT increased attentional preference for the eye region during facial emotion recognition than controls, which can be a means of augmenting psychotherapy [34]. Individuals with high depression scores undervalued positive social evaluation, a common symptom of depression that was normalized by oxytocin [35]. In resistant depression, intranasal OXT twice daily for 4 weeks as an adjunct with escitalopram improves depressive symptoms [36]. fMRI revealed major depression is associated with increased paralimbic activity during emotional mental attribution of others, and this situation was improved after OXT administration [37]. OXT improved connectivity between the amygdala and the cerebellar and dorso-medial prefrontal regions involved in emotion regulation, which were associated with levels of subclinical depression [38].

Overall, the current body of clinical literature on the use of OXT as an antidepressant in humans is limited and does not present conclusive findings. While OXT has been employed in various clinical trials to alleviate certain cognitive and intimate behaviors commonly associated with depression, the heterogeneity of psychiatric diagnoses among the participating patient populations presents a challenge for evaluating and comparing OXT's efficacy as an antidepressant in humans. Additionally, the majority of studies examining OXT's therapeutic effect have utilized small sample sizes, which may not have sufficient power to detect existing effects [39].

3. The OXT/OXTR system in depressive disorder

3.1. OXT system

As the first nine amino acids neurohormone and neuropeptide to be biochemically described and synthesized, OXT is well recognized for its role in the induction of labor and lactation. The synthesis of endogenous OXT primarily occurs in OXT neurons located in the paraventricular nucleus (PVN) and supraoptic nucleus (SON). Upon release, OXT can bind to both OXTR and vasopressin V1a receptor (AVPR1A) to exert its biological effects, with a higher affinity for OXTR. Additionally, OXT can also act as an agonist on pain-sensing transient receptor potential vanilloid-1 (TRPV1) receptor and as a positive allosteric modulator at the mu-opioid receptor(MOR) [40]. In this review, we focus on discussing the role of OXT and OXTR in major depressive disorder (MDD).

The OXTR gene, located on chromosome 3p24–26, encodes a G-protein–coupled receptor (GPCR) with a seven-transmembrane domain. In addition to being packaged as secretory granules and transported to the posterior pituitary to be released into peripheral circulation, OXT neurons have broad projections in the brain [41], [42]. OXT can be released via volume transmission as well as axons to many brain regions associated with emotional behaviors, including the ventral tegmental area, prefrontal cortex, hippocampus, and amygdala [42], [43], [44], [45], [46]. OXTR is found throughout the central nervous system, including areas linked to psychiatric disorders (e.g., prefrontal cortex, hippocampus, amygdala), the HPA axis, and immune cells [40].

3.2. The alteration of OXT level in depressive disorder

Numerous studies have reported significant changes in the levels of OXT in the plasma and brain of individuals with depression. However, these changes are not consistently observed across studies. Specifically, a majority of studies have reported a decrease in plasma OXT levels in MDD [47], [48], [49]. Furthermore, the levels of plasma OXT have been found to be negatively associated with depression scores in MDD [50], which may have predictive value in the context of chronic depression and psychotherapy outcomes[51], [52]. Studies have also shown that OXT levels in plasma during pregnancy are associated with the development of postpartum depression [53], but are inversely correlated with postpartum depression scores [54]. However, van Londen and colleagues have reported that plasma OXT concentrations in MDD are comparable to controls, but exhibit a wider range [55]. A meta-analysis including 368 patients and 346 healthy controls also found no significant difference in endogenous OXT levels, including those in blood plasma, blood serum, saliva, urine, or cerebrospinal fluid (CSF), between individuals with depression and healthy controls [56]. Conversely, some studies showed that subclinical depression [57]，MDD and treatment-resistant adolescent major depression[58], [59] have been associated with higher OXT levels in plasma compared to healthy subjects.

In the brain, OXT level change was also inconsistent in depressive disorders in different studies. For example, PVN OXT mRNA and protein levels were decreased in gestational restraint stress induced depressive rats [13]. Prenatally stressed induced depressive rats had lower levels of OXT mRNA in the hippocampus [60]. In addition, long-term social isolated induced depressive mice exhibited depressive-like behaviors in FST and the sucrose preference tests, accompanied by a reduction of OXT mediated intra-central amygdala (CeA) inhibitory synaptic transmission, which indicates a decreased OXT level [11]. However, after 60 days of social isolation, depressed adult female prairie voles had a higher OXT level in PVN than normal objects [61]. In a human study, they discovered that OXT levels in CSF were elevated in depressed patients [55]. Another study also found a higher PVN OXT expression in MDD than controls [62].

These results mentioned above indicated that plasma and brain oxytocin levels were heterogeneously altered in depressive disorders in both animal and human studies. Several factors may contribute to the diverse changes in OXT levels observed in different studies. Firstly, the methods used to measure endogenous OXT are different. Bioassays, immunoassays (e.g., enzyme-linked immunosorbent assays, radioimmunoassay, and chemiluminescence immunoassays), and mass spectrometry can all be used to determine OXT levels [63], [64]. However, even the two most commonly used plasma oxytocin assays, immunoassay and mass spectrometry, showed quite different results when testing the same samples [63]. Different measurements used in different studies may result in disparate results. With the advancement of OXT sensors, it will be possible to measure endogenous OXT levels in real time, accurately tracking OXT changes [65], [66]. Second, OXT level is reported to differ between male and female, which may account for the mixed results observed in various studies. As estrogen [67] and androgen [62] can modulate OXT expression. Female estrogen levels are higher than male estrogen levels [68], and they also fluctuate with the menstrual cycle, resulting in a wide range of OXT baseline levels. Third, plasma and brain OXT levels do not reflect the "same OXT". It has been observed that OXT concentrations in the CSF are consistently higher than those in peripheral blood [69]. Additionally, the release of OXT in the blood and CNS differs in terms of half-life [69] and production by different neurons (parvOT vs. MagnOT) [70]. In the brain, OXT functions as a neuropeptide that regulates the activity of neurons, astrocytes [71] and microglia [72]. Conversely, when OXT is secreted into the systemic circulation, it may act as a hormone [73]. Therefore, any measurement of PVN mRNA and protein levels reflects a heterogeneous mixture of both the hormonal and neuropeptide forms of OXT [74]. However, neurobehavior is primarily regulated by the centrally released neuropeptides that are present in the extracellular fluid of specific brain areas, rather than in the CSF or plasma [73]. Furthermore,OXT is typically synthesized in the soma of neurons, but its release can occur through various mechanisms, including dendritic, axonal, synaptic or volumetric release [70], [75]. The diverse modes of OXT release imply that peripheral OXT levels may not serve as a reliable biomarker for MDD. Lastly, OXT level varies according to species, age, diseases, and disease progression. When we detect OXT, we frequently detect changes at a specific time point, which does not accurately reflect the overall changes of OXT in the disease. For example, young mice have higher plasma OXT level OXT than old mice [76]. Plasma OXT level was lower in obese and newly diagnosed type 2 diabetic patients [77]. Clarifying OXT changes throughout the MDD disease process may help us better understand the relationship between OXT and MDD.

3.3. The alteration of OXTR in depressive disorder

3.3.1. The OXTR expression

OXTR is widely expressed in neurons, astrocytes, and microglia in brain regions associated with depression, including the olfactory bulb, prefrontal cortex, hippocampus, amygdala, and hypothalamus [78], [79], [80]. There is little direct evidence for OXTR expression changes in MDD. A human postmortem study discovered that MDD was linked to an increase in OXTR mRNA levels in the dorsolateral prefrontal cortex [81]. However, some clues about OXTR changes in MDD can be found. MDD is primarily caused by various stressors [1], and stress, which is commonly used to induce depression, can alter OXTR expression. In young mice, postnatal chronic stress followed by a second single stressor, such as forced swimming, increased OXTR gene expression in the hippocampus at adulthood [82]. Prenatal restraint stress significantly increased OXTR mRNA levels in both the hippocampus and amygdala in rats [60]. However, this increase in OXTR expression in response to stress does not always last, and it appears that when the stimulus exceeds a certain threshold, it causes a decrease in OXTR. Han and colleagues, on the other hand, demonstrated that long-term isolation, a chronic stress intervention, may exhaust the hyperplastic response of OXTR and reduce OXTR expression in CeA [11]. All together, these evidences suggest that OXTR expression may also be changed in MDD models, which should be investigated in future study.

3.3.2. The epigenetic change of OXTR

Depression is a genetically susceptible disease. The dynamic epigenetic alterations of OXTR, including DNA methylation (DNAm) and single nucleotide polymorphism (SNP), have been reported in depression.

Animal studies revealed that OXTR DNAm regulates the brain region-specific OXTR gene expression [83]. Hypermethylation of OXTR in hippocampal tissues increased the early life stress-induced susceptibility to depression [84]. Furthermore, the effect of DNAm on OXTR expression regulated maternal care, social behavior, and synaptic responses to OXT stimulation, which are common abnormalities associated with depression [85].

In humans, the DNA methylation (DNAm) alterations of the OXTR have been found to be associated with depression in two distinct aspects. Firstly, OXTR DNAm is altered following exposure to stress. Specifically, maternal adverse psychosocial experiences during pregnancy have been found to downregulate umbilical cord blood DNAm of OXTR, including 19 CpG sites. Furthermore, a higher incidence of maternal depression symptoms was found to be associated with lower OXTR DNA methylation in cord blood [86]. In addition, females in the high early life adversity (ELA) group were found to have higher OXTR DNAm of intron 1 CpG 5 than the low ELA group, although this difference was not detected in males [87]. Secondly, OXTR DNAm has been found to increase the risk of depression. Lower OXTR DNAm of the CpG sites (cg08535600) within exon 1 or higher DNAm of the CpG sites (cg11589699) within intron 3, was associated with a higher risk for depression following abuse [88]. In addition, lower OXTR DNAm at CpGs chr3:8810078 and 8810069 was significantly associated with a higher postpartum depression score [89]. These findings suggest that OXTR DNAm alterations may be a critical factor in the development of MDD and may be a useful biomarker for identifying individuals at risk for depression.

OXTR SNP genotypes have been reported to be associated with depression in humans in three ways. Firstly, OXTR SNPs may increase the risk of depression. For instance, rs53576 is reported to increase with the risk of comorbid depression and disruptive behavior disorders in youth [12], as well as susceptibility to depression within the context of interpersonal risk factors [90]. Secondly, in certain conditions, OXTR SNPs may cause more severe depressive symptoms. Significant association between rs237885 and a higher prenatal depression score [91]. rs53576 is associated with a higher level of youth depressive symptoms for individuals with a history of maternal depression [92]. Compared to the AA/AG carriers, the GG homozygotes of rs2254298 induce more possibility of aggravated depressive symptoms due to childhood adversity [93]. Thirdly, OXTR SNPs also act as a media of intergenerational transmission of psychopathology. In a 3-year longitudinal study, it was observed that parental OXT levels remained relatively stable over time and were associated with a lower risk of allelic variations on the OXTR (rs2254298, rs1042778). These allelic variations were found to be predictive of OXT levels in offspring [94].

In conclusion, OXT, OXTR, and the epigenetics of OXTR demonstrate significant alterations in depression. Future use of new detection methods to clarify OXT changes throughout the MDD could assist us in resolving these issues.

4. Immune dysfunction in depressive disorder

The immune system is a complex network, composed of multiple immune cells (e.g., neutrophils, macrophages, T cells, microglia), organs (e.g., lymph glands, thymus), and other substances (e.g., inflammatory cytokines, chemokines). This intricate network plays a crucial role in sensing internal and environmental pathogens, and it participates in a wide variety of physiological processes to defend against pathogens and maintain body homeostasis. Traditionally, the brain has been considered an immune privileged organ when compared to the peripheral immune system due to the lower expression of major histocompatibility complex (MHC) class II molecules in the brain. Additionally, the blood-brain barrier (BBB) creates a physical barrier that prevents peripheral pathogens or immune cells from entering the brain [95].

MDD is a common psychiatric disorder that is typically triggered by various stressors [1]. Although the exact cause of MDD remains uncertain, researchers have established that abnormalities in both the peripheral and central immune systems are common among patients with MDD. Kronfol and colleagues were among the first to report lower lymphocyte mitogenic activity in depression [96], and subsequent studies have confirmed this observation. Based on these findings, it is hypothesized that stress can lead to immune dysfunction, which in turn contributes to the development and manifestation of MDD.

4.1. Stress induces immune dysfunction in MDD

Stress, a non-specific response to any demand for change, is a major contributor to depression [1], [97]. Mild stress triggers the release of corticotropin-releasing hormone (CRH) and arginine vasopressin (AVP) from hypothalamic neurons, which stimulate adrenocorticotropic hormone (ACTH) release from the anterior pituitary gland [98], [99]. ACTH then acts on the adrenal gland to stimulate glucocorticoid release. AVP is also produced locally in the adrenal medulla and can stimulate release of glucocorticoid, underscoring its role as a positive HPA regulator. Glucocorticoids have immunosuppressive effects that inhibit leukocyte trafficking and activation, as well as the production of inflammatory cytokines. Glucocorticoids can cross the BBB and activate pituitary glucocorticoid receptors (GR) in the hypothalamus to inhibit CRH and ACTH signals ( Fig. 1). This negative feedback pathway maintains the homeostasis of the HPA axis. The HPA axis homeostasis rapidly modulates the adaptive response to a stressor, and efficiently terminates and resets for the next stress [100].

Fig. 1.

Stress-HPA axis-immune crosstalk in depression. Hypothalamus neurons will be activated by stressful stimuli to release CRH and vasopressin. These hormones jointly promote the ACTH release from the anterior pituitary, and ACTH further acts on the adrenal gland to promote cortisol release. Excessive cortisol could disturb the homeostasis of the HPA axis, resulting in the immune dysfunction, including elevations of inflammation cytokines and BBB impairment. Then, these inflammation cytokines could cross through the damaged BBB into CNS. Inflammation cytokines and immune signal would activate central microglia and astrocyte to release inflammation cytokine, leading to a vicious circle. All of these damage factors will lead to the neuro dysfunction, result in depression. Solid red arrows represent activation, while dotted red arrows indicate passing through BBB.

However, excessive stress can activate immune cells in both the peripheral and central immune systems, such as macrophages and microglia, resulting in the release of pro-inflammatory cytokines, including IL-6, IL-1beta, and TNF-alpha [101]. This inflammatory response can be transmitted to the brain through the parasympathetic nervous system and damaged blood-brain barrier, creating a vicious cycle [102], [103], [104]. Cortisol and pro-inflammatory cytokines can also activate the kynurenine pathway [105]. This leads to the production of 3-hydroxy-kynurenine (3-OH-KYN) and quinolinic acid (QUIN) [106], these metabolites can be taken up by glial cells in the brain and cause neurodegeneration [106]. QUIN has been found to increase the stimulation of hippocampal NMDA receptors, leading to apoptosis and hippocampal atrophy [106],which is associated with depression-like behavior in rodents [107], [108] and elevated levels have been reported in the cingulate cortex of suicidal patients with acute idiopathic depression [109]. Moreover, IDO activity can competitively inhibit the metabolism of tryptophan to serotonin [105], which plays a crucial role in depression and is targeted by SSRIs to increase its activity [110]. Although the relationship between serotonin and depression remains complex and unclear, the broad anti-inflammatory effects of SSRIs may explain their efficacy as antidepressants [111].

The interplay between the stress response, hypothalamic-pituitary-adrenal (HPA) axis, and immune system provides insights into the pathophysiological mechanisms of depression. Chronic stress, corticosteroid supplementation, and repetitive systemic inflammatory stimulation are among the various stimuli that can induce depressive symptoms through the stress-HPA axis-immune crosstalk (Fig. 1) [7]. Depressive symptoms are frequently observed in immune disorders, such as sepsis, inflammatory bowel disease, rheumatoid arthritis, and systemic lupus erythematosus, due to immune dysfunctions that impact the central nervous system and neuronal function.

The dysregulation of the HPA axis homeostasis is a common feature of depression, including abnormalities in cortisol elevation, circadian rhythm, hyperactivity, and glucocorticoid function modulators, as well as the downregulation of corticotrophin-releasing hormone (CRH) signal-related genes and single nucleotide polymorphisms (SNPs) [112]. These abnormalities result in hypercortisolemia and glucocorticoid resistance, which contribute to the development of depression. Although strategies to treat major depressive disorder (MDD) by modulating the HPA axis and extrahypothalamic targets have been explored, including cortisol synthesis inhibitors, glucocorticoids, and vasopressin receptor antagonists [112], their efficacy remains limited and further research is warranted.

4.2. Peripheral immune dysfunction in MDD

Depression is associated with two aspects of peripheral immune dysfunction. Firstly, Depression induces an immunodeficient state, with reduced T cell lymphoproliferative response, Th cell number, and natural killer cell activity [113], and increased neutrophils, CD4 + T cells, and monocytes in peripheral blood [114]. These findings are similar to those observed in sepsis survivors, with increased pathogen-specific memory CD8 + T cells and immunoparalysis [115]. These results suggest that Immunodeficiency is an outcome of depression, increasing vulnerability to environmental stress due to compromised immunity and broken HPA axis [116]. Secondly, pro-inflammatory cytokines are abnormally expressed in depression, including C-reactive protein (CRP), interleukin (IL)−6, IL-1beta, IL-18, IL-2, tumor necrosis factor alpha (TNF-alpha), and chemokine ligand 7 (CXCL7), all of which were found to have abnormally high levels in depression patients’ peripheral blood [117], [118], [119]. Furthermore, Th17 and IL-17A were reported to be increased in the plasma of patients with MDD [120], [121]. On the basis of these findings, researchers have examined the antidepressant role of anti-inflammatory agents, such as adalimumab, etanercept, infliximab, and tocilizumab, which have received the most attention and have demonstrated antidepressant effects [122].

4.3. Central immune dysfunction in MDD

4.3.1. BBB

The blood-brain barrier (BBB) serves as the primary defense mechanism of the CNS against the infiltration of circulating toxins, pathogens, and immune cells. It is composed of tightly sealed endothelial cells, pericytes, and astrocytes, which regulate the exchange of essential nutrients, gases, and signaling molecules [123]. Prior studies have shown that individuals with MDD exhibit abnormal concentrations of urate and albumin in both cerebrospinal fluid and blood, indicating BBB impairment [124]. The degree of BBB destruction has also been found to positively correlate with the severity of depressive symptoms [124]. Furthermore, BBB impairment can lead to increased permeability and infiltration of peripheral immune signals into the CNS, potentially contributing to neuroinflammation and the development of depression [123].

The destruction of the BBB in individuals with depression is influenced by two primary factors. Firstly, depression is associated with increased levels of pro-inflammatory cytokines in the blood [117], [118], [119] that increase permeability by down-regulating endogenous cell adhesion molecules [123]. This leads to the recruitment of more inflammatory cytokines and aggravation of BBB destruction. Secondly, the protein functionality of the BBB itself has been impaired. Individuals with MDD exhibit a significant reduction in the coverage of blood vessels by endfeet of AQP4-immunoreactive (IR) astrocytes, indicating alterations in AQP4 functions in the BBB [125]. Moreover, chronic social stress can lead to the loss of tight junction protein claudin-5 (CLDN5), which destroys the integrity of the BBB and allows peripheral inflammatory cytokines to enter the CNS [126].

The impairment of the BBB in depression may be critical to the development and progression of the disorder by allowing for the infiltration of peripheral immune signals into the CNS. Further research is essential for a comprehensive understanding of the mechanisms underlying BBB impairment in depression and its potential as a therapeutic target.

4.3.2. Microglia

Microglia, the critical resident immune cells in the CNS, monitor the physiological environment through motile protrusions. Microglia play a crucial role in neuronal events such as phagocytizing dead cells and debris, remodeling synapses, and regulating neurotransmission. Emerging evidence suggests that microglia are also involved in depression-related immune dysfunction [104], [127].

Inflammatory cytokines and immune signals that infiltrate the CNS under stressful conditions can trigger the activation of microglia. Postmortem studies have revealed that MDD patients have higher levels of microglial marker expression in specific brain regions, such as the medial frontal gyrus, superior temporal gyrus, thalamus, and subventricular zone [128]. Moreover, MDD suicidal patients have more activated microglia in the dorsal anterior cingulate cortex [129]. Activated microglia are known to release inflammatory cytokines, such as IL-6, IL-1β, and TNF-α, which can cause neuronal dysfunction and contribute to depressive symptoms [130]. Abnormal immune activation of microglia can also result in increased synaptic elimination and dendritic spine engulfment, leading to cognitive memory impairment and a worsening of depressive symptoms [131], [132]. Activating microglia in the dorsal striatum has been reported to result in depression through a prostaglandin-dependent decrease in neuronal excitability[133]. Postmortem studies have also revealed that the density of quinolinic acid-positive microglia in the subgenual anterior cingulate cortex is increased [134]. The quinolinic acid，which can increase reactive oxygen species (ROS) and nitrogen radicals, leading to oxidative stress [130]. This, in turn, can increase the immune reaction in the central nervous system (CNS) and further exacerbate the activation of microglia, leading to a vicious cycle.

Under physiological conditions, microglia play a crucial role in the process of synaptic pruning, which maintains the integrity of neural circuits. However, under pathological conditions, activated microglia can lead to excessive synaptic phagocytosis, resulting in the loss of synapses and consequent cognitive impairment [135], [136], [137]. In mice with chronic social defeat stress-induced depression, there is evidence of excessive synaptic pruning, which is attributed to microglia-dependent synaptic phagocytosis [138]. Moreover, in chronic stress-induced depression animal models, there is an upregulation of Neural Colony Stimulating Factor 1 (CSF1) expression on microglia in the medial prefrontal cortex (mPFC), which contributes to synaptic deficits and depression development [139]. It has also been reported that patients with depression exhibit widespread cognitive memory impairments [140]. The excessive phagocytosis of synapses by activated microglia may be a crucial factor that leads to these cognitive memory deficits.

In summary, microglia play a critical role in detecting depression-related stressors and eliciting neuroinflammation, excitotoxicity, oxidative stress and synaptic impairment, all of which contribute to depression development. However, determining whether microglia are the primary or secondary effectors following stress-induced immune system changes in a specific region or circuit is critical. Additionally, a better understanding of how stress-induced changes in microglial cellular pathways lead to depression is required.

4.3.3. Astrocyte

Astrocytes are a highly versatile type of glial cells found in the brain. These cells are involved in a range of functions, including the formation of the BBB and immune responses within the brain. Specifically, astrocytes are able to release cytokines and other signaling molecules that can activate immune cells, thereby contributing to the immune response [141], [142]. Furthermore, astrocytes closely interact with neurons and microglia. They can form tripartite synapses with neurons, which play an important role in maintaining stable neural activity [143], [144]. Abnormalities in astrocyte function have been implicated in depression, as observed in both animal models and patients.

Firstly, the density of astrocytes is abnormal in depressive disorders. In MDD patients who did not receive medical treatment, the density of hippocampal astrocytes was reduced [145], and astrocytes in the mediodorsal thalamus and caudate nucleus were reduced in depressive suicide patients [146]. The mean areal fraction and packing density of astrocytes varied with the age of depression patients, decreased in young patients but increased in elderly patients [147].

Secondly, in depressive disorders, the morphological properties of astrocytes, such as a larger cell body, longer and more ramified processes, and a reduction in gap junctions between astrocytes and oligodendrocytes, are altered in the thalamus, caudate, and anterior cingulate [146], [148], [149].

Thirdly, the function of astrocytes is perturbed in the context of depression. Patients with depression exhibit abnormal glutamate signaling and gene expressions related to glutamatergic transmission [150], [151], and elevated levels of glutamate in both the plasma and brain have been observed [152], [153]. In normal conditions, astrocytes convert glutamate to glutamine via glutamine synthetase, with some of the glutamine being degraded and the rest taken up by neurons and converted back to glutamate [154]. However, impairment of neurons or astrocytes can lead to the accumulation of excessive glutamate in the synaptic cleft, resulting in the over-activation of glutamate receptors in neurons and neuro-excitotoxicity [152], [155], [156]. Nonetheless, whether the damage to astrocyte function is a cause or a consequence of depression remains unclear based on current evidence, and this warrants further exploration in future studies.

Fourthly, astrocyte-dependent mechanisms regulate depressive-like behaviors, suggesting that astrocyte abnormalities may contribute to the onset of depression. In mice, reduced expression of multiple endocrine neoplasia type 1 (MEN1) in astrocytes enhances the activation of NF-B and production of IL-1, resulting in depressive-like behaviors [157]. Additionally, the absence of the glucocorticoid receptor (GR) in astrocytes reduces Ca2 + activity in response to stress, decreases ATP release from astrocytes through lysosome exocytosis via the PI3K-Akt signaling pathway, and leads to depressive-like behaviors [158]. In a rat model of depression, upregulation of the potassium channel (Kir4.1) in astrocytes of the lateral habenula (LHb) is associated with bidirectional regulation of neuronal activity and depressive-like behaviors [159].

In summary, excessive stress disrupts the homeostasis of the HPA axis and enhances the activity of the extrahepatic enzyme 2,3-indolimine dioxygenase (IDO), leading to a decrease in serotonin levels. Additionally, stress-induced immune dysfunction signals are transmitted to the CNS, leading to the activation of microglia and astrocytes and resulting in neuronal damage. Collectively, these events create a cascade of neurobiological changes caused by an overactive immune system, ultimately culminating in the development of depression (Fig. 1).

5. OXT and immune system in MDD

As previously mentioned, Major Depressive Disorder (MDD) has been associated with immunocompromise [113] and elevated levels of pro-inflammatory cytokines [117], [118], [119]. Evidence suggests that certain anti-inflammatory drugs have anti-depressive effects [122]. Multiple studies have demonstrated significant anti-inflammatory effects of OXT in various diseases and systems, including the central nervous system, cardiovascular, lung, intestine, and urinary [160], [161]. Oxytocin receptor (OXTR) is widely expressed in the hypothalamus, pituitary gland, adrenal gland [40], as well as in a variety of immune elements such as microglia, astrocytes, and thymus[9], all of which are affected in MDD. OXT can activate OXTR expressed on the HPA axis to suppress the abnormal immune response associated with MDD [162], [163], [164]. Additionally, OXT can regulate T cell function [165], which has the potential to treat immunocompromise in MDD ( Fig. 2).

Fig. 2.

Interaction between OXT signal and immune system in anti-depression. Stress will activate OXT neurons in PVN and SON to produce endogenous OXT. Some OXT can be released via volume transmission or axonal projection to the PFC, VTA, HIP, BLA. Some OXT can be released into periphery circulation via posterior pituitary. Endogenous or exogenous OXT targets OXTR expressed in HPA axis and immune cells to suppress abnormal immune response, resulting in improvement of depressive symptoms. Purple arrow represents OXT alleviating depression via immune system. Black blunt arrow represents the inhibition effect.

5.1. Effect of OXT in regulating stress and HPA axis

5.1.1. Animal-based studies

Acute stress triggers a cascade of physiological responses that serve as a self-protective mechanism against potential threats, with OXT release being a critical component of this response [166]. Under normal conditions, OXT, similar to AVP and CRH, stimulates the pituitary gland to secrete ACTH, with the pituitary gland being sensitive to OXT. Even a low concentration of OXT, 1 nM, suffices to induce the release of ACTH in rat anterior pituitary cells [164].

However, the effects of OXT under normal and stress-stimulated conditions appear to be contradictory. The stressor-elicited increase in the expression of the immediate early gene c-fos mRNA in brain regions involved in the regulation of the HPA axis, such as the PVN, ventrolateral septum, and dorsal hippocampus, was attenuated by concomitant administration of OXT [167]. In stress-stimulated animals, exogenous OXT administration inhibits the activity of the HPA axis, inhibits the secretion of ACTH and corticosterone levels [167], [168], [169]. OXT also enhances the buffering effect of social support against stress responsiveness and promotes positive social interaction, making it a crucial component in the treatment of depression [170]. Moreover, OXT modulate the HPA axis over a prolonged period of time. Rats that received subcutaneous OXT injections over five days displayed decreased corticosterone concentrations that persisted for up to 10 days [171].

There are several plausible explanations for these results. Although the pituitary gland is sensitive to OXT, its efficacy in promoting pro-ACTH secretion is low. OXT induces a modest amount of cortisol release (approximately 30% of that induced by AVP) in rat pituitary adenoma cells [172], [173]. In stressed animals, elevated catecholamine secretion due to sympathetic excitation efficiently modulates corticotropin-releasing hormone (CRH) secretion via hypothalamic β-adrenergic receptors. As a result, OXT application can impede catecholamine secretion, leading to a reduction in ACTH and cortisol secretion [168].

5.1.2. Human-based studies

The consistent anti-cortisol effects of OXT in humans’ contrast with the mixed results observed in animal studies. In both normal and stressed subjects, OXT reduces cortisol secretion in a dose-dependent manner, accompanied by a decrease in ACTH levels [14], [174]. The consistency of these results in humans may be attributed to several reasons: 1) OXT's poor efficacy in promoting ACTH secretion, similar to animals [173]. Unlike animals, prior to OXT injection or blood sample collection, human subjects are informed of the experiment, which can induce a stressful event and increase catecholamine secretion. This psychological factor, induced by the "inform," may provide a plausible explanation for theconsistent anti-cortisol effects of OXT in humans. Supporting this notion, several studies have demonstrated that OXT reduces stress response and cortisol levels in individuals who are stressed and have poor emotion regulation, but not in healthy individuals [175], [176]; 2) OXT can act directly on the adrenal gland and inhibit the cortisol release [14]; 3) Since the OXT level in human portal blood is nearly 300 times higher than the peripheral circulation, a high level of OXT will suppress the pro-ACTH secretory effect of AVP by competitively binding to vasopressin receptors [173].

Taken together, these findings suggest that OXT has the potential to modulate the HPA axis and may be a promising target for the treatment of stress-related disorders. However, further research is needed to elucidate the precise mechanisms underlying these effects and to determine the optimal therapeutic interventions.

5.2. OXT and microglia

Microglia are a type of resident macrophage in the brain that have been shown to be activated in depression [129], [130]. Prior research has indicated that microglia express the OXTR [177], [178], which is upregulated upon lipopolysaccharide-induced microglial activation in a time-dependent manner [177]. Additionally, OXT has been demonstrated to dose-dependently attenuate LPS-stimulated expression of major histocompatibility complex class II, a marker of microglial activation, in cultured microglia [178]. OXT has also been found to reduce the activation state of microglia in primary microglia, BV2 microglia cell lines [177], and MG6 microglia cell lines [179], as well as decrease the expression of pro-inflammatory cytokines such as TNFα and IL-1β in activated microglia [177], [179]. These effects may be attributed to the inhibition of phosphorylation of ERK/p38 [177] and the endoplasmic reticulum (ER) stress-related eIF-2-ATF4 pathway [179]. In animal studies, it has been observed that adult mice lacking OXTR exhibit significant activation of microglia in the amygdala and lateral septal nuclei[180], two brain regions often implicated in depression [181], [182]. Conversely, exogenous administration of OXT can inhibit LPS-induced activation of microglia in the prefrontal cortex of adult mice and reduce the production of pro-inflammatory cytokines [177]. Furthermore, OXT has been found to reduce perinatal brain injury in mice and zebrafish by targeting microglial activation [183]. Specifically, Intranasal administration of OXT has also been shown to inhibit microglial activation and inflammatory cytokine release in the hippocampus and improve depressive behavior in a mouse model of autism [184]. Collectively, these cross-species in vivo and in vitro findings suggest that OXT exerts a negative regulatory effect on microglial activation, which may represent a potential mechanism underlying its antidepressant effects, given the important role of microglia in depression.

5.3. OXT and astrocytes

Astrocytes are also an essential part of immune system in the brain. In the LPS-induced depression model, astrocytes are extensively activated, accounting for inflammatory response and depressive symptoms [185]. OXTR is detected in astrocytes, and exogenous OXT application will inhibit the activation of Nucleotide-binding oligomerization domain-like receptor protein 3 (NLRP3) in astrocytes and reduce the inflammatory response [80], [186]. In addition, stress increased the H2S production by astrocytes and the subsequent retraction of astrocytic processes around OXT neurons [187]. Some studies showed that OXT could target astrocytes to change astrocytic GFAP plasticity via modulating the p-ERK/PKA pathway [188] and regulating GABA and glutamate release [189], [190]. OXT has a similar mechanism as the rapid antidepressant ketamine, which regulates astrocyte morphological atrophy [191] and strengthens synaptic connections by increasing glutamate-mediated neurotransmission [192]. Based on the success of ketamine in anti-depression therapy [193], it is worth to further exploring the role of OXT in this field.

5.4. OXT and T cell in brain

T cells, a type of lymphocyte immune cell, is originate from bone marrow progenitors, migrate to the thymus for maturation, and are then released into the peripheral immune organs. Under physiological conditions, T cells are present only in the meninges and choroid plexus of the brain. However, CD4 + or CD8 + T cells can be found in the CNS parenchyma after exposure to stress [194]. T cells recognize multiple antigens from pathogens, and differentiate into effector cells to clear these pathogens. Some effector cells have an immunological memory and help the immune system respond quickly to similar pathogens [195]. In immune disorders, T cell differentiation and immunological memory are disturbed. For instance, sepsis could cause a lymphopenia. The survivor of sepsis showed increased pathogen-specific memory CD8 + T cells. These pathogen-specific memory CD8 + T cells could inhibit the function of normal CD8 + T cells, resulting in immunoparalysis [115]. This phenomenon is similar to that of depression, which shows an immunocompromised state [113] and an increase in T cells [114].

Interestingly, evidence showed that OXTR is expressed in all T cell subsets in the thymus [196], and the thymus can also produce OXT [197]. OXT secreted by the thymus causes the clonal deletion of self-reactive T cells, which is beneficial for the immature T cells to not be disturbed by neuropeptides from other sources, which may promote the differentiation of T cells [197]. In addition, carbetocin, an OXTR agonist, attenuates the T cell inhibition and enhances T cell activation. Carbetocin is even more effective than hydroxychloroquine or lopinavir in enhancing T cell activation [198]. Thus, OXT may have the potential effect of improving immune function in depression via T cells.

6. Conclusion and future directions

In this review, we summarized the relation between OXT, the immune system, and depressive disorders, and emphasized the neuro-immune crosstalk in MDD. Immune dysfunction is one of the causes of MDD, so restoring the homeostasis of the immune system is supposed to be pivotal in improving and treating depressive symptoms. As an important endogenous peptide, OXT is involved in multiple modulation functions in the CNS, including the anti-inflammatory effect and neurotransmission regulation. The regulatory role of OXT signals in central immune system has been intensively investigated [9]. In addition, OXT and OXTR play a crucial role in emotion regulation, and disruption of OXT or OXTR increases the likelihood of developing depressive disorders [10], [11], [12], [13]. Exogenous OXT has antidepressant effects through complex neural mechanisms, including regulation of the "stress-HPA axis-immune crosstalk" (Fig. 2) [14], [177], [186]. Future research evaluating the interactions between OXT and OXTR with the immune system in depressive disorders will therefore be beneficial for the diagnosis, prevention, and treatment of MDD.

OXT is a relatively large hydrophilic molecule with poor penetration across the BBB [199], though some studies showed the possibility of OXT crossing the BBB [200]. Meanwhile, the OXT half-life in the human peripheral circulation is ∼3 min [201]. Increasing the penetrability and stability of OXT will enhance its clinical use for diseases of the central nervous system, such as MDD. Intranasal spray is widely used to promote neuropeptides transporting into the brain [202], and exogenous OXT was detected in cerebrospinal fluid and many brain regions, including hippocampus and amygdala with the intranasal application [203], [204]. Some studies are also aiming to develop the structure of OXT to improve its stability and efficacy [205] and develop non-peptide agonists of OXTR [18], which may facilitate the translational research on OXT in the treatment of MDD.

CRediT authorship contribution statement

Junliang jiang: Conceptualization, Writing – original draft, Date curation, Resources, Software; Miaoxian Yang, Mi Tian: Resources, Software; Chen Zhong, Lei Xiao and Ye Gong: Supervision, Writing – review & editing.

Declaration of Competing Interest

We declare that we have no financial and personal relationships with other people organizations that can inappropriately influence our work, there is no professional or other personal interest of any nature or kind in any product, service or company that could be construed as influencing the position presented in, or the review of, the manuscript entitled.

Data availability

Data will be made available on request.