The monoamine deficiency theory has been the primal causative model of major depression and has accompanied the rise of antidepressants. The current pharmacologic arsenal is based on this theory as the main antidepressants, such as selective serotonin reuptake inhibitors, serotonin–noradrenaline reuptake inhibitors, and monoamine oxidase (MAO) inhibitors, are designed to increase the availability of monoamines (serotonin, noradrenaline, dopamine). According to this theory, it is thought that the depletion of these neurotransmitters in the central nervous system (CNS) constitutes the core of depression pathophysiology (14).

Evidence of curcumin being able to influence levels of monoamines in the central nervous system has emerged from animal and in vitro studies conducted over the past two decades. In a study, Bhutani et al. (15) showed that curcumin reversed the depressive-like behavior induced by chronic stress on mice and enhanced the serotoninergic and dopaminergic transmission alongside an inhibition of the MAO-A. Kulkarni et al. (16) had similar results in a study on rats in which curcumin dose-dependently increased serotonin and dopamine as well as inhibited the monoamine oxidase enzymes. More recent studies also showed that curcumin could elevate norepinephrine, serotonin, and dopamine in the frontal cortex, hippocampus, and striatum in rats (17–22).

Wang et al. (23), in an animal study on the effects of curcumin on serotonin (5-HT) receptors, stated that the antidepressant-like effect that they could notice was related to the serotoninergic system, possibly due to an interaction with 5-HT1A/1B and 5-HT2C receptors. Similarly, Xu et al. (24) conducted a study in which the antidepressant effect of curcumin seemed related to the expression of 5-HT1A receptors, as 5-HT1A receptor mRNA levels across all hippocampal subfields were increased after curcumin administration. These results have been replicated by more recent studies like by Li et al. (25) or Lian et al. (26) in which they noticed an upregulation of 5-HT1A receptor expression after curcumin administration in chronically stressed mice and a prevention of its antidepressant-like effect with the administration of a 5-HT1B receptor antagonist.

Taken together, these animal studies strongly support the hypothesis that curcumin can modulate monoaminergic systems in pre-clinical rodent models.

In 1959, Crane (27) made an observation that the anti-tuberculosis agent d-cycloserine, which acts as a glutamatergic modulator, could possess an antidepressant effect, yet this observation gained little attention during the following decades until 2000, when Berman et al. (28) reported that ketamine, an antiglutamatergic anesthetic agent, could induce a rapid antidepressant effect in severely depressed patients. This triggered vigorous research to understand how modulating glutamate signaling could be beneficial in depression and to develop new antidepressants targeting glutamate neurotransmission. It is worth noting that ketamine has now proven its efficacy in treating depression (29).

Glutamate is the major excitatory neurotransmitter in the central nervous system and has a vital role in the regulation of synaptic plasticity. By binding to its receptors, notably NMDA receptors (NMDAR), glutamate will modulate post-synaptic plasticity as well as exert longer-term changes in synaptic strength and neuroplasticity. However, the abnormal elevation of NMDAR signaling leads to deleterious effects on neurons (30). As such, excitotoxicity is the phenomenon associated with excessive glutamate release and subsequent overactivation of NMDA receptors and has long been described (31).

A number of reports suggest that the glutamate system and excitotoxicity are involved in MDD pathophysiology. For example, glutamate levels have been shown to be elevated in the plasma, cerebrospinal fluid, and brains of patients with depression (32).

Chronic stress is believed to lead to detrimental changes within glutamate synapses, including reduced extracellular glutamate clearance by glia, especially in the pre-frontal cortex, leading to increased extrasynaptic glutamate levels and excitotoxicity, potentially contributing to synaptic loss (33). Hence, in a study conducted by Lin et al. (34), curcumin was shown to inhibit the liberation of glutamate in the rat pre-frontal cortex, counteracting this phenomenon in a similar way (yet greater) to that of fluoxetine (suggesting that “classic” antidepressants like fluoxetine also act on the glutamate pathway).

A study by Zhang et al. (35) suggested that the antiglutamatergic action of curcumin could be mediated by the GluN2B subunit of NMDA receptors. Also, they noted that the administration of a sub-effective (that did not produce antidepressant effect when administered alone) dose of curcumin produced an antidepressant-like effect when paired with a sub-effective dose of fluoxetine, leading to the hypothesis of a synergistic interaction between NMDA and 5-HT receptors. These results have been clarified by further in vitro studies showing that curcumin could reverse glutamate-induced neurotoxicity on hippocampal cells and downregulate the expression of the GluN2B subunit of NMDA receptors (36, 37). The antagonization of the GluN2B subunit seems to have a crucial role in the effect of anti-NMDAR drugs such as ketamine, as the activation of this subunit inhibits the synthesis of certain synaptic proteins, such as brain-derived neurotrophic factor (BDNF), thus altering synaptic function (38). Hence, the antagonization of GluN2B subunit permits an enhancement of synaptic function. For instance, memantine, which is also an antagonist of NMDA receptors, after acute administration, had no effect on hippocampal BDNF and failed to show immediate antidepressant behavioral effects in an animal depression model (39). However, the prolonged administration of memantine was associated with an increase in BDNF levels and hippocampal cell proliferation while displaying antidepressant-like effects in other animal studies (40, 41). In studies displaying the action of curcumin on the glutamatergic system, these effects were associated with increased levels of BDNF (23, 42) that might account for its antidepressant efficacy.

Collectively, those results suggest that an inhibitory effect of curcumin on excitotoxicity may be one of the mechanisms underlying its antidepressant effects.

As stated in the introduction, most of the first studies trying to unravel the beneficial effects of curcumin came from research on cancer and inflammatory diseases, and many anti-inflammatory effects of curcumin are now being acknowledged (55). A number of studies have indeed shown that curcumin could inhibit TNF production by macrophages and downregulate its expression by modulation of its transcription factors. Notably, it has been shown that curcumin was able to down-modulate the activation of nuclear factor kappa beta (NF-κβ) (56).

A study conducted by Wang et al. (57) in 2014 has reported the effects of curcumin on inflammation and depressive symptomatology in mice treated with lipopolysaccharide (LPS). This study showed that administrating curcumin reverted the depressive-like behavior and attenuated LPS-induced microglial activation and overproduction of pro-inflammatory cytokine (interleukin-1β and tumor necrosis factor-α). It could also inhibit LPS-induced NF-κβ activation in the hippocampus and pre-frontal cortex (PFC). In addition, curcumin could counteract the increase in the levels of nitric oxide synthase and cyclooxygenase-2 mRNA in the hippocampus and pre-frontal cortex. Jangra et al. (58) reported similar results, as the administration of curcumin in LPS-treated mice improved the depressive-like symptoms, reversed the depletion of glutathione level in the hippocampus, and decreased the level of pro-inflammatory cytokines (IL-1β and TNF-α) in the hippocampus.

Fan et al. (59) showed that chronic unpredictable mild stress exposure in mice induced microglia activation and overexpression of the cytokines IL-1β, IL-6, and TNF-α within the medial pre-frontal cortex, effects which were paralleled with neuronal structural changes. They showed that curcumin administration produced antidepressant-like actions and reversed the inflammatory responses and neuronal structural abnormalities, thus concluding that inhibiting the IL-1β pathway could account for curcumin efficacy.

With regards to studies of human patients, a clinical trial conducted by Yu et al. (60) showed that curcumin decreased the inflammatory cytokines IL-1β and TNF-α in depressed patients in comparison with the placebo group. Moreover, it has been shown in other clinical trials that curcumin can lower TNF-α, IL-6, and CRP levels in patients' plasma (56, 61, 62).

All in all, what we can draw from these studies is that curcumin permits a decrease in inflammation, acting on pro-inflammatory cytokine pathways, notably TNF-α, IL-6, and IL-1β in the hippocampus and pre-frontal cortex.

Clinical depression is characterized by inflammatory cascades as indicated by elevated concentrations of pro-inflammatory cytokines in the serum and in the CNS tissue. Given the known action of curcumin on some immunoinflammatory pathways, the pivotal role of IL-1β and microglia activation in the pathophysiology of depression, and the relevance of these processes to the therapeutic properties of curcumin, we can highlight a probable role for the inflammasomes.

Inflammasomes are crucial components of the innate immune response that initiate immunological reactions against microbial infections, tissue injury, and other aggressions. The key inflammasome in depression is the NLRP3 inflammasome complex, which, upon activation, consists of a NOD-like receptor protein containing pyrin domain (NLRP3) associated with an adaptor protein ASC and an effector caspase-1 (63).

NLRP3 inflammasome activation is related to microglial activation. Upon initially recognizing an environmental stressor, the microglia can indeed enter in an active state and are thereafter more prone to induce a prolonged inflammatory response following the upcoming exposure to those stressors. NLRP3 inflammasome has drawn interest as an explanatory element as to why primed microglia cells display sensitivity to environmental stressors. In fact, NLRP3 inflammasome acts as a transducer of neuroinflammatory responses. A lowered threshold for NLRP3 activation induces an increased production of inflammatory cytokines, such as IL-1β and IL-18, and then causes a persistent neuroinflammation (64). This chronic inflammation results from chronic exposure to noxious threats that negatively affect the homeostatic feedback mechanisms. In sum, NLRP3 inflammasome is a molecular mechanism that translates psychological stressful stimuli into inflammatory responses.

The activation of the NLRP3 inflammasome relies on a two-step mechanism. It requires a priming signal to initiate the transcription of the NLRP3 inflammasome to above a certain threshold, while a second signal will then promote the formation of the NLRP3 inflammasome by recruiting ASC and pro-caspase-1 proteins.

The priming of NLRP3 transcription is mediated by NF-κβ activation, which can result from the stimulation of membrane-bound receptors by TNF or pathogen-associated molecular patterns like LPS (63) or damage-associated molecular patterns whose presence indicates cellular or metabolic stress. It is worth noting that there were some reports indicating that corticosterone may also stimulate the transcription of NLRP3 via glucocorticoid receptor activation, involving the NF-κβ pathway but also independently of NF-κβ activation (65, 66).

Another way of activating NLRP3 inflammasome is the reduction of intracellular potassium (K+) levels via the activation of the P2X7 purinergic receptors. Extracellular ATP is released by neurons and astrocytes following excitotoxicity. Activation of these receptors causes potassium efflux in microglia, which is crucial for the activation of the NLRP3 inflammasome (67). Besides that, P2X7 receptor is being investigated as a potential drug target in mood disorders (68). Reactive oxygen species (ROS) can also promote NLRP3 activation (69). An activated NLRP3 inflammasome will then cleave pro-IL-1β and pro-IL-18 into active IL-1β and IL-18.

There have been a few studies highlighting the role of NLRP3 inflammasome in depression. Animal studies investigating this theory showed that chronic stress could stimulate NLRP3 activation in rodent brains and that, accordingly, in the absence of the NLRP3 inflammasome activation, stress did not produce a depressive behavior, anhedonia, or social impairment in those mice (70–73). It has also been shown that NLRP3 is activated in mononuclear cells in patients with depressive disorder and that some antidepressants compounds seem to show a decrease in NLRP3 activation (74).

Concerning curcumin, in an in vitro study conducted by Li et al. (75) on rat hippocampal cells, excessive glutamate release induced IL-1β secretion and generation of ROS. Curcumin was able to inhibit the generation of ROS and NLRP3 inflammasome activation assessed by NLRP3 and caspase-1 expression, which resulted in a reduction of IL-1β secretion. In addition, curcumin could prevent glutamate-induced cell apoptosis. In the same vein, Fan et al. (76) showed that curcumin administration could prevent IL-1β-induced apoptosis in stressed rats in the ventromedial pre-frontal cortex and alleviate depressive behavior.

In a study conducted by Zhang et al. (77), rats were exposed to chronic unpredictable stress, and this exposure led to depressive-like behaviors. Curcumin successfully corrected the depressive-like behaviors in those stressed rats. Additionally, curcumin could effectively decrease the mRNA expression of proinflammatory cytokines (IL-1β, IL-6, and TNF-α) and suppress NF-κβ activation as well as inhibit the expression of P2X7 receptors, thus preventing NLRP3 activation. These results were in accord with a study conducted by Wang et al. (78), in which curcumin could inhibit the activation of the P2X7 receptor and thus inhibit the inflammatory response and the microglial activation, thus preventing NLRP3 activation.

It is also worth quoting a recent study conducted by Ozkartal et al. (73) that highlighted the role of nitric oxide synthase (NOS) in the activation of NLRP3 inflammasome; in this study, the administration of NOS inhibitors in chronically stressed mice prevented the activation of NLRP3 inflammasome and the production of IL-1β.

In some reports, depression has been associated with an excessive activity of nitric oxide synthase (79). Also, some antidepressant medications in current use, such as paroxetine, inhibit NOS activity (80). In some other studies, suicidal and depressed patients had elevated levels of plasma NO and its metabolites (81, 82). The overactivation of NOS pathway and subsequently increased nitric oxide has been described as an important pathophysiological factor in neuroinflammation and neurotoxicity processes involved in stress and depression (79, 83) as well as an important modulating factor in the production of neurotransmitters such as noradrenaline, serotonin, and dopamine (79), thus explaining why NOS inhibitors can exert antidepressant and anxiolytic-like effects (73, 84).

In human cultured neurons exposed to quinolinic acid, quinolinic acid increased NOS activity and consequently increased the nitrite levels. In those neurons, curcumin could counteract this increase of NOS activity (85). Studies investigating the influence of curcumin on NOS pathway in animal models have shown that, in rat (86) or pig (87) stress models, curcumin could inhibit NOS hyperactivation and the subsequent increase in hippocampal NO. As previously cited in the study conducted by Wang et al. (57) on mice with LPS-induced depressive symptomatology, curcumin could attenuate the LPS-induced microglial activation and NF-κβ activation in the hippocampus and PFC as well as decreased the levels of NOS in the hippocampus and PFC.

These data suggest that NLRP3 activation can play a major role in inflammatory-related depressive symptoms and that its inhibition by curcumin might be a key feature of its effectiveness.

Indoleamine 2,3-dioxygenase (IDO) is an enzyme that is strongly activated by proinflammatory cytokines (such as TNF-α, IL-1, and IL-6) (88). This enzyme has an important role in depression by catabolizing tryptophan (the primary precursor of serotonin) into kynurenine pathway metabolites: IDO leads to tryptophan depletion and thus inhibits serotonin synthesis. As stated before, the role of serotonin in depression is already admitted. However, research shows that kynurenine pathway metabolites can impact several mechanisms associated with depression, such as pro-apoptotic changes in the CNS or generation of free radicals (89). Preclinical studies have shown that IDO activation results in depressive-like symptoms and that IDO inhibitors could relieve depression-like behaviors in mice (90).

There are two main kynurenine pathway metabolites studied in depression: one of them is quinolinic acid, an NMDA agonist, that is thought to be excitotoxic, thus leading to synaptic loss. It can also disrupt the oxidant/antioxidant balance by increasing the generation of free radicals and induce lipid peroxidation (88). Excess quinolinic acid levels were found in depression (91), however inconsistently (92). The levels of quinolinic acid were still shown to predict the response to antidepressant medications such as ketamine (93). The second main kynurenine pathway metabolite is kynurenic acid, considered to be neuroprotective. Its balance with quinolinic acid is considered to be protective against the excitotoxicity of the latter (94). Moreover, studies reported lower levels of kynurenic acid in depression (92).

In vitro studies demonstrated that curcumin could counteract the inflammation-induced overexpression of IDO (95), and more recently, a study conducted by Zhang et al. (77) showed that curcumin could inhibit the stress-induced overexpression of IDO in rats and normalize the quinolinic acid/tryptophane ratio.

Another important co-factor of the increase of IDO is cyclo-oxygenase 2 (COX-2) (96). COX-2 is an enzyme responsible for the production of prostaglandin E2 (PGE2), which is a pro-inflammatory chemical messenger. Studies have shown an increase in PGE2 production and COX-2 expression in depressed patients (97, 98). The peripheral blood cells of patients with recurrent depression also exhibited an increased expression of the genes encoding for COX-2 (99). A number of clinical trials have yielded promising results for the use of COX-2 inhibitors as an augmentation of antidepressant treatments (100). Their use could also help decrease inflammatory cytokine levels (101).

Regarding curcumin, studies have shown that it can downregulate COX-2 expression and PGE2 synthesis in vitro (102) and in animal models of depression (103), highlighting its potential role as an alternative natural COX-2 inhibitor option. It is relevant to mention that COX-2 can induce NOS activity and vice versa (104). The 2014 Wang et al. (57) study mentioned before (see the previous section) illustrates the interplay between NOS, IDO, and inflammatory cytokines.

Considering the possible role of COX-2 in depression and its role in IDO synthesis and the role of IDO in depression that we illustrated, these data suggest interesting mechanisms of action for the use of curcumin as an augmentation therapy.

As the largest mucosal surface in the human body, the intestinal epithelium acts as an interface with the external environment (105). In some situations, the permeability of the intestinal epithelium can be altered in a state also called “leaky gut.”

Lipopolysaccharides (LPS) are large molecules found in the outer membrane of gram-negative bacteria. In a state of “leaky gut,” there will be an increased translocation of those gram-negative bacteria from the intestine into the systemic circulation and, with them, the LPS (106, 107). LPS can then stimulate toll-like receptors, generating an inflammatory process. This will result in pro-inflammatory cytokine secretion and neuroinflammation (108) and also activation of IDO with the previously described consequences (109–111).

Some studies have reported increased serum IgM and IgA levels directed against the LPS of gram-negative enterobacteria in depressed patients, reflecting an increased translocation of LPS from those gram-negative enterobacteria, with the authors of such studies concluding that depressive disorder was accompanied with intestinal hyperpermeability (107, 112).

In animals, LPS induces depressive behaviors (110). Thus, a number of studies investigating the anti-inflammatory effects of curcumin were carried out using LPS models. In such models, curcumin reversed LPS-induced behavioral and molecular changes. Moreover, a handful of in vitro studies have shown that curcumin could increase the expression of tight junction proteins, thus preventing the disruption of tight junction organization and decreasing LPS increase due to paracellular permeability (113).

These data suggest that one antidepressant mechanism of action of curcumin could be its ability to counteract this state of gut permeability and subsequent LPS-induced inflammation.

The HPA axis is a central system in the body's stress response, and abnormalities in its activity have long been noted in MDD (114). Studies have shown that depression was associated with impairment in the responsiveness to glucocorticoids and a subsequent hypersecretion of CRH. This phenomenon is known as glucocorticoid resistance and can, in turn, prime the inflammatory response (114). We have indeed highlighted the role of corticosterone in the activation of NLRP3 inflammasome (65, 66). Depression is also associated with an increased size and activity of the pituitary and the adrenal glands (115).

Based on these assumptions, animal models of depression related to stress and elevated cortisol levels have been developed. In such models, curcumin has been shown to alleviate the depressive symptoms and other physiological alterations induced by cortisol.

Li et al. (25) showed that, in stressed rats, curcumin could restore the normal level of corticosterone. Rinwa and Kumar later found similar results (116). Researchers showed that curcumin could protect against corticosterone-induced neurotoxicity and downregulation of mRNA levels of serotoninergic receptors in a rodent model (5-HT1A, 5-HT2A, and 5-HT4) (117). Curcumin could also normalize the adrenal gland size in stressed rats (118). Furthermore, curcumin decreased the salivary cortisol concentrations in depressed patients compared with the placebo group in the clinical trial of Yu et al. (60).

Insulin receptors are present throughout the brain, and they play a central role in regulating its use of glucose for energy (119). Moreover, some of insulin's actions are specific for the CNS, like promoting neuronal survival or synaptic plasticity and regulation of brain functions including memory, cognition, learning as well as attention (120, 121). Insulin also inhibits norepinephrine and serotonin reuptake and downregulates alpha-2 receptor expression in hypothalamic neurons (122).

Consequently, there has been a hypothesis suggesting that insulin resistance played a role in depressive disorder pathophysiology, which was reinforced by epidemiological studies. For example, adults with insulin resistance are more at risk for the development of future depression (123), and a meta-analysis has shown an association between insulin resistance and depression incidence (124). Significantly, insulin resistance interacts with other depressogenic processes. Given the insulin functions in the brain, it has also been shown that reduced brain sensitivity to insulin can manifest as impaired neuroplasticity and disturbances in neurotransmitter's release and uptake (125). Insulin resistance has also been shown to be associated with altered dopamine signaling in rodents (126).

Inflammation and oxidative stress are both involved in the pathogenesis of insulin resistance (127, 128). In turn, insulin has antioxidant properties which become disrupted following insulin resistance (129). As described earlier, oxidative stress resulting from the disruption of insulin functions can, in turn, trigger the production of pro-inflammatory cytokines. The desensitization of glucocorticoid receptors that can be associated with insulin resistance (130) may potentiate this depressogenic cascade.

There is an emerging research field trying to reposition some antidiabetic medications as new means in depression management like GLP1 functional agonists, pioglitazone, or even metformin (131).

Concerning curcumin, Shen et al. (132) have conducted a study on mice with unpredictable chronic mild stress in which they showed that curcumin could attenuate insulin resistance alongside a decrease in depressive-like symptoms. There were also other studies showing the positive effects of curcumin on insulin resistance in in vitro, animal, or even human studies (133).