Effect of Xiaoyaosan on major depressive disorder

Regarding the efficacy of XYS in ameliorating MDD, the disease showed significant improvement according to the traditional Chinese medicine curative index [6]. XYS also significantly decreased the Hamilton depression scores in MDD patients [7]. Controlled, randomized, and double-blinded trials showed that the total effective rate of XYS in treating MDD was 91.38% in 58 patients, being significantly higher than the value of 32.69% for the placebo in 52 patients [8]. XYS treatment was associated with a significantly decreased depression inventory score [8] (Table 1).

The total effective rate of XYS–antidepressant combinations in 12 trials was not significantly improved compared with those of other antidepressants in seven trials. However, the cure rate was increased in one of the seven trials. The five remaining trials exhibited improved effective rates in the combination groups. Nine trials demonstrated that XYS reduced side effects, four trials revealed decreased relapse rates, and four trials showed delays in onset time (Table 2).

XYS exhibited fewer side effects than antidepressants, but similar effectiveness. The combinations of XYS and antidepressants advanced the onset time and reduced adverse side effects in most of the trials. The methodological quality of most included trials was generally “poor”. So, defects in sample sizes, blindness, and randomization were observed in most of the trials evaluated. Therefore, biases including performance and detection biases could not be ignored [21] (Table 3).

5-HT deficiency was the prevailing hypothesis for MDD [22, 23]. SSRIs were widely used, and accounted for about 60–80% of the total market share of antidepressants [24]. XYS upregulated the 5-HT contents in the cerebral cortex of a chronic restraint stress (CRS)-induced rat depression model [25], and increased the 5-HT contents in the hippocampus of rats with postpartum depression [26]. XYS could be a regulator of monoamine neurotransmitters [27].

The hypothalamic–pituitary–adrenal (HPA) axis is governed by secretion of corticotropin-releasing hormone (CRH) from the hypothalamus to activate secretion of adrenocorticotropic hormone (ACTH) from the pituitary gland. Corticoids (cortisol in humans and corticosterone in rodents) are stimulated from the adrenal cortex and interact with their receptors, such as glucocorticoid receptors, for negative feedback control [28]. HPA hyperactivity results from deficits in the negative feedback regulation of the axis based on the failure of glucocorticoid receptor activation to decrease plasma levels of cortisol [29]. XYS downregulated CRH-1 and upregulated CRH-2 expression in the hypothalamus of a CRS-induced depressive rat model [30]. XYS decreased the expression of CRH-1 mRNA in paraventricular nuclei and increased GR expression in the hippocampus of a chronic unpredictable mild stress-induced depressive rat model [31]. Therefore, homeostasis of CRH receptors might be involved in improvement of the disequilibrium in the HPA system.

HPA hyperactivity was observed in 30–50% of all acutely depressed patients [32]. Mitochondrial dysfunctions affected important functions in MDD pathogenesis [23]. Small deletions of mitochondrial DNA were observed in muscles from patients with MDD [33]. Alterations in nuclear DNA-encoded mitochondrial mRNA and proteins in the cerebellum of MDD patients were also reported [34]. MDD patients with serious somatic complaints exhibited low ATP production rates in biopsied muscles [35]. These studies provide concrete evidence for the clinical relevance of an association between low ATP supply arising through mitochondrial dysfunction and MDD. XYS was reported by our group to ameliorate depressive-like behaviors in rats by regulating mammalian target of rapamycin (mTOR), suggesting that XYS may exert its anti-depressive effects through regulation of energy metabolism [36].

Inflammatory pathways were suggested to be involved in the pathophysiology of MDD through increased blood and cerebrospinal fluid concentrations of pro-inflammatory cytokines as well as acute phase proteins and their receptors [37]. Cytokines interact with mitochondria to increase the production of reactive oxygen species (ROS). Increased expressions of pro-inflammatory mediators, neurotoxic factors, and ROS contributed to the development of MDD [23]. XYS has been widely used for treating inflammatory diseases and depression comorbidities in hepatitis [38]. Recently, we found that XYS significantly reduced the serum levels of tumor necrosis factor-α and interleukin-6 in rats with depressive-like behaviors induced by chronic unpredictable mild stress (unpublished data). MDD was associated with neuronal atrophy and neuronal cell loss, especially in the hippocampus and cerebral cortex [39]. Decreased brain-derived neurotrophic factor (BDNF) was strongly associated with an increased risk for MDD [40]. A clinical meta-analysis showed that BDNF levels were associated with changes in depression [41]. BDNF was downregulated in the hippocampus of a CRS-induced rat depression model [42]. Reports from our group and others indicated that XYS increased BDNF expression in the hippocampus [36, 42, 43]. These results suggest that XYS improves MDD by upregulating BDNF in specific encephalic regions.

Epigenetic alterations were found in the frontal cortex of suicide victims with depression [44]. Antidepressants exerted some of their effects by causing epigenetic alterations [45]. Observed dysfunctions of biological clocks were related to MDD [46]. Patients with depression often showed altered circadian rhythms, sleep disturbances, and variations in diurnal moods [47]. The degree of circadian misalignment was correlated with the severity of depressive symptoms [47]. The actions of XYS on epigenetic modifications and circadian rhythms are significant, because this medicinal formula is efficacious in the treatment of sleeping and mood disorders.

Hypotheses for MDD include 5-HT depletion, neurotrophin deficiency, neuroinflammation, mitochondrial dysfunction, HPA hyperactivity, epigenetic variation, and circadian dysrhythmia. However, the pathophysiology of MDD has rarely been studied and the published hypotheses are far from mutually exclusive.

The theory of inadequate monoamine neurotransmission, in which antidepressants increase monoamine availability and produce long-term adaptive changes in monoaminergic receptor sensitivity [48], is insufficient to explain MDD. Lowered plasma tryptophan reduced 5-HT synthesis and aggravated MDD symptoms [49]. N-acetylserotonin, an intermediate product of melatonin formation from 5-HT, is a specific agonist of BDNF receptors, and 5-HT is a substrate for melatonin biosynthesis. Melatonin deficiency contributed to primary and depression-associated insomnia as well as disturbances in circadian rhythms [50].

Neuroinflammation, which is characterized by increased production of interferon-γ, interleukin-6, and tumor necrosis factor-α, and induction of indoleamine 2,3-dioxygenase (IDO) in the blood and brain, plays a role in depression [37]. Activation of IDO reduces plasma tryptophan and brain 5-HT and increases the levels of tryptophan catabolites (TRYCATs), such as quinolinic and picolinic acids. Inflammation increases CRH and ACTH secretion. Cortisol levels are increased to activate liver tryptophan 2,3-dioxygenase, which further decreases plasma tryptophan and increases TRYCAT production. TRYCATs generate ROS, cause mitochondrial dysfunctions, and interfere with energy metabolism. They also potently activate NMDA receptors and induce pro-inflammatory responses and neuron apoptosis. These findings imply a shift from tryptophan and 5-HT depletion toward the detrimental effects of TRYCATs. IDO links the neuroinflammation and neurotoxicity of TRYCATs, which jointly promote the development of depressive symptoms [49].

Psychosocial stresses arising from life events can potentially induce continuous increases in stress hormones, which impair negative-feedback mechanisms and lead to continuous hyperactivity of the HPA axis. Pro-inflammatory cytokines are also potential activators of the HPA axis, thereby increasing the secretion of glucocorticoids, which are markers of glucocorticoid resistance. Glucocorticoids augment the alternative pathway for IDO-catalyzed tryptophan and decrease the amount of 5-HT available in synapses by increasing the expression of the serotonin transporter gene. Prolonged increases in glucocorticoids desensitize their receptors on immune cells, such as macrophages. Activation of macrophages in the periphery and brain occurred and pro-inflammatory cytokines were released in MDD patients [37]. Psychosocial stresses decrease the levels of BDNF and other neurotrophic/growth factors, while increasing the glucocorticoid concentration. Multiple interaction pathways exist between pro-inflammatory immune functions, brain and neuronal structures, brain serotonergic systems, and the HPA axis. The HPA axis is a key integrative component that links primary biological and psychosocial theories [36].

Dysfunction of the hippocampus, cerebellum, insula, frontal cortex, and temporal cortex could eventually contribute to the pathogenesis of MDD. The integrated model conjectures a general-purpose co-processor, whose effects depend on the specific brain centers to which individual modules are connected [51]. The disparate modules and different ideas on MDD emphasize the internal relationships among the different hypotheses (Figure 2).