Authors: Yu-Bing Wang, Ning-Ning Song, Yu-Qiang Ding, Lei Zhang
Categories: Articles from the Specail Issue Neural plasticity, Depression, Monoamine, Neural circuit, Neural plasticity, Neurotrophic factor, Synapse
Source: IBRO Neuroscience Reports
Authors: Yu-Bing Wang, Ning-Ning Song, Yu-Qiang Ding, Lei Zhang
Depression is one of the most common mental disorders, which can lead to a variety of emotional problems and even suicide at its worst. As this neuropsychiatric disorder causes the patients to suffer a lot and function poorly in everyday life, it is imposing a heavy burden on the affected families and the whole society. Several hypotheses have been proposed to elucidate the pathogenesis of depression, such as the genetic mutations, the monoamine hypothesis, the hypothalamic-pituitary-adrenal (HPA) axis hyperactivation, the inflammation and the neural plasticity changes. Among these models, neural plasticity can occur at multiple levels from brain regions, cells to synapses structurally and functionally during development and in adulthood. In this review, we summarize the recent progresses (especially in the last five years) on the neural plasticity changes in depression under different organizational levels and elaborate different treatments for depression by changing the neural plasticity. We hope that this review would shed light on the etiological studies for depression and on the development of novel treatments.
Neuropsychiatric diseases, such as schizophrenia, depression and autism, deteriorate the mental health of patients, impairing their cognition and communication with others and imposing a huge burden on the affected families and society. Among these diseases, depression is a global epidemic declared by the World Health Organization (WHO) with more than 300 million people suffering from it. According to the WHO, major depressive disorder (MDD), a leading reason of disability around the world, will rank as the first cause of burden of disease worldwide by 2030 (Malhi and Mann, 2018). The pathogenesis of depression includes complicated interactions between social, psychological and biological factors. Several hypotheses have been developed to explain the etiology of depressive disorder, such as the monoamine hypothesis, hypothalamic-pituitary-adrenal (HPA) axis changes, inflammation and so on. It is apparently impossible to cover every aspect of these hypotheses, we will in this review focus on the neural plasticity which has been shown to be central in depression.
Neural plasticity refers to the capacity of the nervous system to modify itself, functionally and structurally, in response to experience or injury (stimuli), which could be observed during development and in adults (De Berardis et al., 2017). Plasticity occurs at various organizational levels of the nervous system. Thus, neural plasticity will be categorized into nervous tissue plasticity (e.g., hippocampus, prefrontal cortex, amygdala and so forth) (Liu et al., 2017b), neuronal or glial plasticity, synaptic plasticity, etc.
An increasing number of studies have been devoted to unveiling the pathological mechanisms underlying depression. In this review, we summarized the recent progresses on the neuroplasticity changes in depression under different organizational levels and then elaborated different treatments of depression and the underlying mechanism of neuroplasticity. The latter sections involve the development, strengths and weaknesses of the existing therapies. We also discussed the future directions for depression research and treatment.
Structural magnetic resonance imaging (MRI) studies have shown that volumes of brain regions related to emotion, learning and memory, such as the hippocampus and amygdala, were reduced in MDD patients compared with healthy subjects. Moreover, increased regional grey matter (GM) was found in patients receiving treatment like electroconvulsive therapy (Enneking et al., 2020, Nuninga et al., 2020, Takamiya et al., 2019, Thomann et al., 2017). In the prefrontal cortex (PFC), marked decreases in cortical thickness were found in the orbital and medial areas of the PFC and anterior cingulate cortex (ACC) (Schmaal et al., 2017). The extent of volume reduction is associated with the severity of depression, duration of illness and time length of treatment (Duman et al., 2016). However, in a whole-brain experiment, greater GM volume in the dorsolateral PFC was observed in the adolescents with depression compared with healthy youth (Straub et al., 2019). This may be due to a compensatory and neuroplastic response that attenuated depressive symptom severity in adolescence (Bansal et al., 2018). The volume reduction of the hippocampus in depression may result from the disruption and atrophy of neurons and glia, and may be associated with the changes in synaptic plasticity (Duman and Aghajanian, 2012).
Question whether adult neurogenesis plays a causal role in depression still remains open, but there is evidence that antidepressants work through an active neurogenesis (Brooker et al., 2017, Micheli et al., 2018, Song et al., 2017). Adult neurogenesis takes place in two specific neurogenic niches, the dentate gyrus (DG) of the hippocampus and the subventricular zone adjacent to lateral ventricles (Anacker et al., 2018, Micheli et al., 2018). A morphological examination reported that adult human brain has undetectable levels of hippocampal neurogenesis (Sorrells et al., 2018), but it has also been mentioned that about 700 new neurons are born per day in the human DG (Tartt et al., 2022) and the presence of a substantial number of immature dentate granule cells in the adult human hippocampus via low-frequency de novo generation and protracted maturation (Zhou et al., 2022). Using proliferation and survival cell markers such as Ki67 and doublecortin (DCX) or applying bromodeoxyuridine (BrdU) pulse labeling, impaired neurogenesis was observed in rodent models with depression-like behaviors (Morais et al., 2017). Also, neurogenesis in the DG was inhibited by high levels of glucocorticoids in depression, while adrenalectomy could promote it (Liu et al., 2017a).
Chronic or excessive stress is highly associated with the onset and development of depressive disorders (Ding and Dai, 2019). Accumulating evidence indicates that stress-induced microglial dysregulation is also involved in depression. For instance, postmortem analysis on depressed suicides demonstrated elevated microglial activity in the cingulate cortex (Steiner et al., 2011, Torres-Platas et al., 2014). Aberrant microglial activity may impair brain network homeostasis and plasticity mainly via modulation of neuronal transmission by secreting cytokines and regulating synaptic and network remodeling (Li et al., 2022). In addition, the early-life inflammation causes adolescent depression-like symptoms by altering the long-term neuronal spine engulfment capacity of microglia via the Toll-like receptor 4/NF-κB signaling pathway (Cao et al., 2021). Priming of microglia with intracerebroventricular injection of IFN-γ in mice compromised adult hippocampal neurogenesis and resulted in depression-like behaviors (Zhang et al., 2020b). Conversely, administration of microglial activation inhibitor-minocycline could prevent IFN-α-induced neurogenesis deficiency and depression-like behaviors in mice (Zheng et al., 2015). Microglia can be activated by a pro-inflammatory environment, which is composed of a diversified group of molecules including pathogens-associated molecular patterns, damage-associated molecular patterns, cytokines, inflammasomes and so on (Fig. 1B).Fig. 1Neural plasticity changes in depression. A. In the adult healthy brain, NPCs give rise to new neurons, microglia stay silent, astrocytes extend branches and oligodendrocytes form myelin sheath wrapping axons, while under depressive situation induced by stress, atrophy (decrease of gray matter volume and increase of ventricle volume) of the brain with network dysfunction and lower level of neurotrophic factors (e.g., BDNF, NGF, etc.) are found, probably due to reduced neurogenesis. Besides, spine density of neurons is lower in depression. Moreover, hyperactivation of microglia, morphological changes of astrocytes and disruption of myelin sheath are observed in patients with depression and depressive rodent models. BDNF, brain-derived neurotrophic factor; NGF, nerve growth factor; NPCs, neural progenitor cells. B. The potential molecular mechanisms for microglia activation in depression. DAMPs, damage-associated molecular patterns; IFN, interferon; IL, interleukin; NLRP1/3, nucleotide-binding domain, leucine-rich repeat, pyrin domain containing protein 1/3; P2X7R, P2X7 receptor, two-transmembrane ATP-gated ionotropic purinoreceptor; PAMPs, pathogen-associated molecular patterns; TNF, tumor necrosis factor. C. The potential molecular mechanism for astrocyte atrophy in depression.Fig. 1
Astrocytes are potentially the key mediators in the synaptic dysfunction implicated in depression (Zhou et al., 2019). In the PFC of chronic restraint stress (CRS) induced depression-like mouse model, GFAP^+^ astrocytes showed significant atrophy of distal processes (Codeluppi et al., 2021). It has been reported that decreased glycogen content is associated with chronic stress-induced atrophy of hippocampal astrocyte and depression-like behaviors. However, chronic stress might promote both the breakdown and synthesis of hippocampal glycogen (Zhao et al., 2017). Thus, it can be inferred that the compensation of glycogen synthesis for breakdown is insufficient (Fig. 1C). Reductions of astrocyte number and astrocyte dysfunction in frontolimbic structures, which may lead to an aberrant synaptic connectivity and misbalance in neurotransmission, have been reported in MDD patients and suicide victims (Fullana et al., 2020, Peng et al., 2015).
In the central nervous system (CNS), oligodendrocytes form myelin sheaths that enwrap axon fibers, which constitute the white matter (WM). Complex behavioral dysfunction such as depression may occur when WM (e.g., the corpus callosum, uncinate fasciculus and cingulum) integrity is impaired (Bhatia et al., 2018, Williams et al., 2019). In a chronic unpredictable mild stress (CUMS) model of depression in mouse, significant reductions in myelin and oligodendrocyte-related proteins and WM deficits were observed, which were later ameliorated after treatment with the antidepressant desvenlafaxin (Wang et al., 2014). Also, in a rat depressive model, WM atrophy and myelinated fiber disruption were found (Gao et al., 2017). As such, the disturbance of oligodendrocyte lineage cell proliferation and differentiation in the WM may contribute to the depression phenotype (Zhou et al., 2021a). The myelination is affected by adolescent-onset depression and the change of myelin content of WM tracts is more sensitive in depressed female adolescents than in male (Ho et al., 2021).
Synaptic plasticity is one of the most fundamental and crucial functions in the brain (Liu et al., 2017b). The length of the basal dendrites, the dendritic spine density, postsynaptic density (PSD) thickness, the synaptic cleft width, synaptic interface curvature and other dendritic spine morphological characteristics may change during depression progression.
Generally, synaptogenesis and functional plasticity are tightly associated with the shape (i.e., thin, stubby or mushroom) and size of a dendritic spine. Usually, thin spines are highly dynamic and potential to change in response to activity, whereas stubby spines are transitional structures that will possibly transform into mature mushroom spines (de la Fuente Revenga et al., 2021). In the CRS mouse model, increased stubby spine density in pyramidal neurons of the basolateral amygdala (BLA) was observed, while the total spine density in most superficial pyramidal neurons of the medial PFC was reduced (Leem et al., 2020). In another study, CRS increased the diameter of dendritic spine heads and the number of mushroom spines in BLA-projection neurons that did not target dorsomedial PFC (Zhang et al., 2019c). Other studies showed that synaptic density and spine density in brain regions including the PFC are decreased in depression, and antidepressant treatments can reverse these deficits (Bollinger et al., 2020, Moda-Sava et al., 2019, Savalia et al., 2021, Yang et al., 2018a). Synaptic plasticity impairment shown by downregulated expression of synaptophysin, PSD95 and GAP-43 were found in the hippocampus of depressive rat model (Huang et al., 2021a). As mentioned above, the engulfment microglia are a key factor in depression-associated spine alteration (Cao et al., 2021).
Impaired neurotransmission of serotonin (5-HT), norepinephrine (NE), dopamine (DA) and endocannabinoids, enhancement of glutamate (Glu) release via N-methyl-D-aspartate receptor (NMDAR) and reduction of γ-aminobutyric acid (GABA) levels were described in chronic stress-induced depression models (D'Alessio et al., 2020, Hei et al., 2019).
Neurotrophic factors are a family of secreted proteins that promote different activities during development and in the adult nervous system, including gliogenesis, neurogenesis and synaptic plasticity (Zhou et al., 2019). Lower serum levels of neurotrophic factors are often observed in both animal models of depression and patients with depression. In human, a gender bias with less reduction of neurotrophic factors in male has also been recorded (Pallavi et al., 2013), which might be one of the reasons why depression is more common in women than in men. Increase of brain-derived neurotrophic factor (BDNF) has been reported as a response to antidepressant treatment (Abdullina et al., 2022, Kinoshita et al., 2018, Molendijk et al., 2014, Prowse and Hayley, 2021, Taniguti et al., 2019, Yang et al., 2020). Substantial evidence suggested that alterations in neurotrophic factors, majorly the BDNF and nerve growth factor (NGF), underlie impaired neuroplasticity (Levy et al., 2018).
In addition to the role in the growth, differentiation and survival of neurons, BDNF has also been shown to represent an essential factor in the regulation of synaptic plasticity and neurogenesis (Lu et al., 2014, von Bohlen Und Halbach and von Bohlen Und Halbach, 2018). In vitro, BDNF is sufficient to promote spine formation and dendritic outgrowth in B27-deprived primary hippocampal neurons (Park et al., 2016). In vivo, BDNF infusion in the rat hippocampus initiated synaptic strengthening and induced long-term potentiation (Bramham, 2007, Ying et al., 2002). Around 15% of mothers suffer from postpartum depression, which causes lower level of BDNF in the plasma (Lee et al., 2021).
As the first discovered neurotrophin, NGF has been implicated in neurogenesis in the striatum (Zhu et al., 2011) and in regulating hippocampal plasticity (Conner et al., 2009). Dysregulation of NGF is involved in the pathophysiology of depression. In rodent models, decreased NGF expression was observed (Mondal and Fatima, 2019). In the postmortem hippocampal tissue of individuals with depression who committed suicide, NGF and its binding receptor-TrkA were found to be reduced at both protein and mRNA levels (Erbay et al., 2021). On the contrary, some studies showed increased or unchanged NGF levels after stress exposures (Alleva et al., 1996, Lang et al., 2004). And these variable changes may be due to differences of type and intensity of stressors. Antidepressant (tianeptine) treatment restored NGF level in the hippocampus of a mouse model (Alfonso et al., 2006). It is noteworthy that NGF itself could act as an antidepressant, which improves depression-like behaviors in rats by modifying gene expression in amygdala and hippocampus (Overstreet et al., 2010).
Efficient communications between distinct brain regions guarantee normal brain function but the brain network is damaged in MDD patients with cognitive impairment (Yang et al., 2021). Under the guidance of the human connectome project (Glasser et al., 2016), substantial efforts have been made to understand the brain connectome related to depression in the past decade. Some large-scale brain networks, including default mode networks and salience networks (Evans et al., 2018, Sahib et al., 2020), are considered as potential neural substrates of depression (Kaiser et al., 2015). Depression has been associated with dysfunction of neural circuits, including decreased resting state functional connectivity within and between the PFC and limbic networks, decreased global intrinsic connectivity between the medial and lateral PFC and all other regions of the brain and decreased inter- and intra-hemispheric integration within and across dorsolateral PFC, medial PFC/ACC, hippocampus and parietal regions (Price and Duman, 2020).
Being part of the limbic system, the hippocampus, connected with emotion-related brain regions including PFC and amygdala, is the most commonly studied structure in depression field. Stress and other negative stimuli that induce depression can result in changes of hippocampal plasticity in different ways, such as volumetric changes, neurogenesis and apoptosis (Liu et al., 2017b). From MRI data, increased resting state functional connectivity between hippocampus and subgenual ACC was observed in depression patients, and this increase was blunted by ketamine administration (Morris et al., 2020). Intriguingly, negative stimuli-induced hippocampus hyperactivation could be normalized after antidepressant treatment (Delaveau et al., 2011).
As an important center of thinking and behavior regulation in the brain, the PFC is also related with depression (Treadway et al., 2015). Recent functional MRI studies showed that the global brain connectivity with global signal regression (GBCr) in the PFC was decreased in depression, and the dysconnectivity between the PFC and the rest of the brain was reversed after ketamine treatment (Abdallah et al., 2017a, Abdallah et al., 2017b, Kraus et al., 2020). According to distinct functional contributions and anatomical connectivity, the PFC is divided into two the dorsolateral part (dlPFC) and the ventromedial sector (vmPFC). Hyperactivity in the vmPFC and hypoactivity in the dlPFC were found during progression of depression, but after psychotherapy or medication, the situation in both subregions was reversed (Koenigs and Grafman, 2009, Mayberg et al., 2005). In addition, impaired neuroplasticity was also found in the dlPFC of patients with depression (Noda et al., 2018).
Located at the heart of the telencephalon, the amygdala is associated with affective modulation, memory encoding and social behavior (Baxter and Murray, 2002). Reduced functional connectivity between the basal amygdala and the medial orbitofrontal cortex, which is related to reward circuit, was found in depression patients. And the dorsolateral amygdala had reduced functional connectivity with the lateral orbitofrontal cortex, involving in non-reward, was also observed (Cheng et al., 2018). In MDD patients with comorbid anxiety, the amygdala functional connectivity is much lower (He et al., 2019). However, it has been reported that the MDD subjects have higher functional connectivity to the right PFC from the amygdala, which is decreased after eight weeks of fluoxetine antidepressant treatment (Zhang et al., 2020a). This discrepancy may be due to differences of severity of illness and age of the first onset or relapse.
The dysfunction of ventral striatum, a limbic structure participating in reward processing, is involved in anhedonia in depression (Goff and Tottenham, 2015). Lower ventral striatal activation was found in MDD patients, and this hyporesponsiveness was normalized after a treatment with the antidepressant S-citalopram (Stoy et al., 2012).
As the brain’s ‘antireward centre’, the LHb targets not only the noradrenergic and dopaminergic, but also the serotonergic systems. Dysfunction of the LHb may result in several psychiatric disorders, especially MDD (Hu et al., 2020). And the past decade has witnessed the spectacular progresses about the implication of LHb in MDD. Increased excitatory synaptic transmission in VTA-projecting LHb neurons is found to be the first cellular basis of MDD mediated by the LHb (Li et al., 2011). Through a quantitative proteomic screen, upregulation of CamKIIβ is identified as the first key molecular mechanism underlying the synaptic hyperactivity of the LHb and depression-like behaviors (Li et al., 2013). PP2A activity is shown to mediate stress-induced internalization of GABAB receptor and GIRK, which may increase excitability of LHb neurons and lead to depression-like behaviors (Lecca et al., 2016). Recently, astroglial Kir4.1 in the LHb is shown to drive neuronal bursts in depression, which could be blocked by antidepressant ketamine (Cui et al., 2018, Yang et al., 2018b).
The exact mechanisms by which antidepressants improve mood remain unclear, but most antidepressants were developed based on monoaminergic neurotransmission (Malhi and Mann, 2018). Chronic administration of conventional antidepressants, blocking the reuptake or breakdown of monoamines, can increase synaptic plasticity in different ways, including enhancement of neurogenesis in the adult hippocampus, increase of neurotrophic factor expression and promotion of synapse formation (Castrén and Hen, 2013, Duman and Aghajanian, 2012).
The mechanism of action of TCA premised that decreased monoamine functions in the brain lead to depression (Chen et al., 2021b). An increase of neural progenitor cells and dividing cells in the DG was observed in the postmortem brains of MDD patients treated with clomipramine and nortriptyline compared with untreated MDD patients (Boldrini et al., 2009). Chronic imipramine treatment promoted cell proliferation in the subgranular zone of the hippocampus (Schiavon et al., 2010). These first-generation antidepressants were effective by enhancing serotonergic or noradrenergic mechanisms or both. Nevertheless, the TCA also blocked histaminic and cholinergic receptor sites, which caused unwanted side effects like weight gain, constipation, drowsiness and so on (Feighner, 1999). As such, these earliest developed antidepressants are in minimal use today.
As one of the second-generation antidepressants, SSRIs specifically bind serotonin transporter and do not lead to anticholinergic or cardiovascular adverse effects associated with the TCA (Parker and Brown, 2000). The discovery of SSRI in the late 20th century was a fundamental step in advancing the treatment of depression and other mental disorders (Muscat et al., 2021). SSRIs, targeting the central 5-HT system, are the first-line antidepressants widely used in current general practice (Song et al., 2017). In mouse depression models, SSRIs show prominent effects on increasing BDNF in astrocytes (Kinoshita et al., 2018), synaptogenesis (Li et al., 2021b) and neurogenesis (Ribeiro et al., 2021). However, long-term treatment with SSRI can also produce side effects such as sexual dysfunction, nausea and headaches that patients do not tolerate (Moret et al., 2009). It is worth mentioning that some SSRIs, including fluoxetine, may aggravate neuronal plasticity, which is contradictory and seldom studied (Wu et al., 2020b).
Like SSRIs, SNRIs also regulate neurogenesis and plasticity in rodent model of depression (Mostany et al., 2008). Similarly, chronic venlafaxine (one type of SNRIs) treatment proved to be able to prevent the harmful effects of restraint stress on BDNF protein expression and hippocampal neurogenesis (Xu et al., 2006).
Typical antidepressants targeting monoamines take weeks to months for a positive response and are ineffective in about 30% of the patients (Cameron et al., 2018), suggesting that chronic administration is required to modulate neural plasticity. In the past 20 years, ketamine, a non-competitive N-methyl-D-aspartic acid receptor (NMDAR) antagonist, has shown to be promising (rapid and sustained antidepressant effects) as a novel treatment modality for treatment-resistant depression, which is appropriate for almost any age. The putative mechanism of ketamine’s antidepressant activity might be associated with the blockage of NMDARs (Wang et al., 2022b), which produces downstream effects including activation of postsynaptic α-amino-3-hydroxy-5-methyl-4-isoxazole-propionic acid receptors (AMPARs) and mTOR signaling, pyramidal cell disinhibition (Widman and McMahon, 2018), inactivation of glycogen synthase kinase 3 (GSK3) and eukaryotic elongation factor 2 (eEF2) and enhanced level of BDNF (Zanos and Gould, 2018). At cellular level, ketamine is able to prevent the reduction of spine density and impairment of synaptic functions in animal model of depression (Duman and Aghajanian, 2012). Nevertheless, ketamine might produce side effects like dissociations, increased blood pressure and addiction, and other glutamate receptor modulators would be the less problematic alternatives (Witkin et al., 2019).
The known toxicity of ketamine expedited the discovery of other NMDAR modulators with rapid-acting antidepressant activity. And rapastinel is one of such compounds, which produces rapid antidepressant effects without the psychotomimetic side effects of ketamine (Liu et al., 2017a). Rapastinel shows significant antidepressant-like properties in multiple animal models by its ability to enhance synaptic plasticity processes linked to learning and memory, which is further supported by the increase of mature dendritic spines in the DG and medial PFC in rat (Moskal et al., 2017). However, rapastinel failed to meet the primary and key secondary endpoints in Phase III MDD trials in 2019 (https://firstwordpharma.com/story/4752081).
Scopolamine, a non-selective muscarinic acetylcholine receptor antagonist, is another rapid antidepressant, producing antidepressant effects as fast as 3–5 days after intravenous administration at a very low dose (4 μg/kg) (Furey and Drevets, 2006). Moreover, intramuscular infusion of scopolamine works even faster within several hours (Gillin et al., 1991). Similar to the action of ketamine, scopolamine works by promoting synaptogenesis, increasing glutamate neurotransmission and activation of mTOR signaling in the PFC (Voleti et al., 2013). The durability of antidepressant effects of scopolamine is limited, but psychedelics drugs have been reported to produce effects for many months (Witkin et al., 2019). As scopolamine can cause memory impairments and bradycardia, it is suggested to use it in conjunction with another antidepressants, which will minimize the negative effect of scopolamine (Podkowa et al., 2018, Podkowa et al., 2016).
There are two main groups of phenethylamines, such as substituted amphetamine 2,5-dimethoxy-4-iodoamphetamine (DOI) and mescaline, and tryptamines such as N,N-dimethyltryptamine (DMT), lysergic acid diethylamide (LSD) and psilocybin (de la Fuente Revenga et al., 2021). Recent studies convincingly show that a single dose of psychedelics such as psilocybin promotes rapid structural plasticity in pyramidal neurons of the frontal cortex (Hesselgrave et al., 2021, Shao et al., 2021). Furthermore, in culturing neurons, DMT, DOI and LSD show remarkable effects on enhancing neuritogenesis, spinogenesis and synaptogenesis possibly by BDNF/TrkB/mTOR signaling (Ly et al., 2018).
Melatonin, an amphiphilic neurohormone primarily produced by the pineal gland (Boutin et al., 2020), was proven to play an important role in the regulation of circadian rhythms, sleep, immune function, neuroendocrine secretion and neuroprotection (Wang et al., 2022a). Besides its function as one of the main circadian output markers of the brain, melatonin could also enhance multiple levels of neuroplasticity in animal studies (Valdés-Tovar et al., 2018). Recently, melatonin has been demonstrated to increase neurogenesis in the DG (Vega-Rivera et al., 2020) and attenuate autophagy impairment through Foxo3a regulation (Ali et al., 2020). Elevation of BDNF and TrkB expression in the hippocampus of a sub-CRS mouse model was observed by treatment of fluoxetine combined with melatonin (Li et al., 2018), and melatonin treatment alone is sufficient to significantly increase the BDNF level in the hippocampus of an lipopolysaccharide (LPS)-induced depressive mouse model (Taniguti et al., 2018). However, the antidepressant effects of melatonin remain not fully explored in human studies (Won et al., 2021).
Agomelatine, both a melatonin receptor (i.e., MT1 and MT2) agonist and serotonin receptor (i.e., 5-HTR2B and 5-HTR2C) antagonist, is used to treat MDD (Norman and Olver, 2019) with relatively mild side effects. Except for effectively upregulating MAP2 and BDNF in the hippocampus and PSD95 in the amygdala, agomelatine can also downregulate the level of spinophilin in the hippocampus (Ladurelle et al., 2012, Lu et al., 2018).
Ramelteon, a synthetic analog of melatonin, has been reported to successfully ameliorate geriatric MDD without any side effects (Furuya et al., 2012). In addition, ramelteon could normalize the altered expression of the circadian genes and neurotrophic factors in insomnia patients with comorbid anxiety and depression (Satyanarayanan et al., 2020).
Neu-P11, also known as piromelatine, is a novel MRA that plays a GABA-enhancing role. It is worth noting that Neu-P11 works no matter the administration took place in the morning or in the evening, while melatonin was effective only when administered in the afternoon in models of depression in rats (Tian et al., 2010). In a chronic mild stress model, Neu-P11 showed antidepressant-like effects by increasing hippocampal BDNF, Creb and p-Creb expression (Fu et al., 2016).
Brain stimulation is a compelling tool that can be used to change neuroplasticity or reverse maladaptive plasticity, with robust clinical effects and a few side effects (George et al., 2022).
Since been used from 1938, ECT is a technique for treating several psychiatric disorders including medication-resistant depression (Trifu et al., 2021). Structural changes including increased white and grey matter volume that are particularly involved in the pathophysiology of depression such as in hippocampal subregions and amygdala were observed after ECT (Gryglewski et al., 2019). Neuroplastic changes have been detected as early as after a single ECT stimulus. ECT brings about synaptogenesis, neurogenesis, dendrogenesis, angiogenesis as well as gliogenesis (Dukart et al., 2014, Gbyl et al., 2021, Hellsten et al., 2005, Jansson et al., 2009, Nuninga et al., 2020, Singh and Kar, 2017). Although it can quickly reverse the symptoms of depression, ECT is associated with some rarely-reported medical and cognitive side effects (Dominiak et al., 2021).
In human, TMS exerts direct effects on both mood and sleep (Collins et al., 2021). Repetitive TMS (rTMS) on the dorsomedial PFC shows safety and profound effect in treatment-resistant depression (Bakker et al., 2015, Schulze et al., 2016). Low-frequency rTMS (1 Hz) increased dendritic length and branch number of mature granule cells and dendritic complexity of newly generated neurons in the DG of mice, which is associated with the antidepressant-like activity of rTMS (Cambiaghi et al., 2020). High-frequency rTMS (e.g., 10, 15 or 25 Hz) is widely used to treat depression, which could ameliorate depression by reducing excitability of cortical pyramidal cells, preventing neuronal loss, promoting neurogenesis and facilitating synaptic plasticity, partly through the p11/BDNF/Homer1a pathway (Zuo et al., 2020). Intermittent θ-burst stimulation, a novel rTMS technique, shows antidepressant effects when applied on the left dlPFC via facilitating normalization of neuronal circuit function by increasing neuroplasticity (Zavorotnyy et al., 2020).
As one of the main healing arts in oriental medicine, acupuncture has long been used in East Asian countries, especially China, and has been shown to have an effect on nervous system disorders (Shin et al., 2017). The procedure of acupuncture involves inserting stainless steel needles into the skin at acupoints to physically stimulate the body (Han et al., 2021). As an alternative treatment for MDD, acupuncture has effects through modulating the activity of PFC (Zhang et al., 2021a) or perhaps the corticostriatal motivation/reward circuitry (Wang et al., 2017). Jiang and co-workers verified that acupuncture could alleviate depression-like behaviors in mouse model by regulating PKA/Creb signaling, which is related to neurogenesis (Jiang et al., 2017).
Based on traditional Chinese medicine acupuncture, electroacupuncture sends electrical currents to stimulate the acupoint by using a needle connected to an electrical device. Having a faster onset of action than other treatments for depression, electroacupuncture can normalize aberrant amygdala networks in depressed patients (Duan et al., 2020). In depressive rat models, electroacupuncture stimulation at Baihui (GV20) and other acupoints could ameliorate depression-like behaviors by restoring hippocampal synaptic plasticity via modulation of 5-HT receptors (Chen et al., 2020, Han et al., 2018), by activating ERK signaling pathway (Li et al., 2017), or probably by anti-neuroinflammation (Yue et al., 2018).
Invasive VNS is a neural-modulation therapy but has surgery-related side effects which limit its application. Thus, transcutaneous auricular VNS (taVNS) was developed and is a relatively non-invasive treatment for MDD and treatment-resistant depression patients, which can regulate the functional activities of brain networks (Li et al., 2019). The regulation effects include increased amygdala-dlPFC connectivity (Liu et al., 2020), decreased connectivity between the medial hypothalamus and rostral ACC (Tu et al., 2018) and so on. In addition, VNS has been found to change the NGF and BDNF levels in the brain (Rosso et al., 2020).
DBS induces changes in gene expression and mRNA splicing which promote neurogenesis and plasticity (Pohodich et al., 2018). Transcranial direct current stimulation (tDCS), a safe and portable non-invasive neuromodulatory treatment, involves the application of a weak (0.5–2 mA) direct electric current to the scalp with generally mild adverse effects (Kenney-Jung et al., 2019). Human and animal experiments have suggested that tDCS might influence brain functions through promoting neural plasticity, which could serve as a first-line treatment for MDD (Woodham et al., 2021).
Anhedonia is one of the key symptoms of depression. Cognitive behavioral therapy has been demonstrated to be the first-line treatment for mild to moderate depression (MMD) (Butler et al., 2006), as it could reverse the reductions of the hippocampus and nucleus accumbens (NAc) volume and abnormalities of NAc-based functional connectivity (Meng et al., 2021). Positive emotional learning can induce rehabilitation in depressive rats by facilitating NMDAR-mediated synaptic plasticity (Burgdorf et al., 2017).
It is well known that exercise is beneficial for depressive symptoms and body functions (Xie et al., 2021). Exercise is an effective method in decreasing the severity of postpartum depression (Özkan et al., 2020). High-intensity aerobic exercise can acutely increase the serum BDNF in MMD individuals (Ross et al., 2019). Lee and Baek suggested that exercise may exert antidepressant effects by increasing neurogenesis through activating Wnt signaling pathway (Lee and Baek, 2017). Running may reverse the changes in synaptic plasticity and synaptic transmission in hippocampal CA1 pyramidal neurons induced by depression (Gómez-Galán et al., 2016). Tai chi improves the quality of life (Liu et al., 2019). Yoga decreases depressive symptoms possibly via increasing the activity of GABA system (Scott et al., 2019, Streeter et al., 2020), while swim reverses CUMS-induced depressive behaviors and neuroinflammation, and increases collapsing response mediator protein-2 (Crmp2) mediated neuroplasticity (Xie et al., 2022). Lactate produced during exercise also plays a role in anti-depression (Karnib et al., 2019). Lactate serves as a neuronal energy substrate and signaling molecule involved in memory consolidation and synaptic plasticity. Recent studies show that lactate produces antidepressant-like effects in animal models probably via enhancing adult hippocampal neurogenesis (Carrard et al., 2021, Carrard et al., 2018).
In CRS mice, overexpression of fat mass- and obesity-associated gene (Fto) in the hippocampus by stereotaxically injecting adeno-associated virus effectively reverses depression-like behaviors and enhances synaptic plasticity by modulating CamKII/Creb signaling pathway (Shen et al., 2021). Associated with WM integrity, ADAM metallopeptidase with thrombospondin type 1 motif 18 (ADAMTS18) plays an important role in the CNS and knockout of this gene exhibits higher dendritic branching complexity and spine density in DG granular cells and lower levels of depression-like behaviors compared to their wild-type littermates (Zhu et al., 2019). Normally expressed at low level in mature neurons, Twist1 is elevated in the medial PFC of chronic social defeat stress mice. And knockdown of Twist1 alleviates depression-like behaviors and prevents morphological defect of dendrites of layer II/III pyramidal neurons in the medial PFC, while overexpression of it increases the susceptibility to stress (He et al., 2021). Neuropeptide FF receptor 2 (NPFFR2) binds to the pain-modulating neuropeptide FF, which is distributed in the CNS of all mammals. NPFFR2 overexpression provokes depression-like behaviors in mice. As such, NPFFR2 knockout mice show good resilience to LPS-induced depressive symptoms mainly by downregulating 5-HTR1A in the ventral hippocampus (Yu et al., 2021). Mammalian STE20-like kinase 1 (Mst1) is a key factor associated with the pathophysiology of stress. Activating Mst1 causes synaptic plasticity impairment, neuronal dysfunction and neuroinflammation, while knockdown of Mst1 significantly improves hippocampal synaptic plasticity and suppresses microglial activation and the p38 pathway in CUMS mouse models (Chen et al., 2022, Yan et al., 2021). In CRS mice, Pten is upregulated in the serotonergic neurons (5-HTNs) in the raphe nucleus, while selectively deletion of Pten in the 5-HTNs increases dendritic branching and density of PSD95^+^ puncta (Chen et al., 2021b). Similarly, Pten is increased in the PFC of CRS mice with more depression-like behaviors, while inactivation of Pten is sufficient to prevent these behaviors (Wang et al., 2021b). Thus, Pten is suggested to be an intrinsic regulator of neuronal activity of 5-HTN and pyramidal neuron in the PFC, which makes it a potential and novel target in treating depression. In addition, Task3, Pdcd4, Trek1 and legumain may also be new targets for antidepressant development (Fullana et al., 2019, Li et al., 2021c, Wu et al., 2021, Zhang et al., 2021b).
In traditional Chinese medicine theory, “stagnation of liver qi” is regarded as the cause of depression. And the combination of Radix Bupleuri and Radix Paeoniae Alba has antidepressant effects by relieving “stagnation of liver qi”. From proteomic analysis on a CUMS rat model, the mechanism of this combined treatment might be linked to neuroplasticity and neuroprotection (Chen et al., 2021c). In rodent studies, Radix Polygalae, traditionally used for treating various psychiatric disorders in East Asia, improves depression-like symptoms probably by promoting autophagy and inhibiting neuroinflammation (Zhou et al., 2021b). Apart from antioxidant and immune modulative activities, Gastrodia elata Blume also has antidepressant, neuroprotective and neuroplasticity-promoting functions (Huang et al., 2021b). Oral treatment of Gastrodia elata water extract reverses depression-like behaviors in a mild social defeat stress-induced depression mouse model via regulating monoaminergic neurotransmission and gut microbiota. Quercitrin, a well-known natural flavone compound present in many herbs, rapidly alleviates depression-like behaviors in LPS-induced depressive mice by enhancing p-Creb/BDNF/PSD95/Synapsin1 signaling and inhibiting PI3K/Akt/NF-κB signaling (Sun et al., 2021). Neural Fuyuan Formula is another TCM that is effective in treating post-stroke depression in rat model, and it has recently been reported to work by promoting neural plasticity through BDNF/TrkB pathway (Cai et al., 2021). Besides, iridoids from Gardeniae fructus attenuates depression by enhancing synaptic plasticity via AMPAR-mTOR signaling (Xia et al., 2021b). Moreover, both the Danggui Buxue Decoction and Baihe Dihuang Decoction can ameliorate depression-like behaviors in rat depression models (Wang et al., 2021a, Zhao et al., 2021). Xia, Hirshler and their colleagues have summarized the neuroplasticity mechanisms of Chinese herbal medicines in antidepressant actions (Hirshler and Doron, 2017, Xia et al., 2021a). Once the safety, efficacy and tolerability of these traditional herbal medicines are confirmed, they can serve as natural alternatives to conventional antidepressants.
Taken together, depression caused by stress and negative stimuli largely impairs neural plasticity in many ways, including volumetric changes, neurogenesis, synaptogenesis, spinogenesis, glial plasticity, network connectivity and expression of genes associated with neuroplasticity. Based on the changes of neural plasticity in depression (Fig. 1A), different treatments were developed to relief depressive symptoms and the effects and related neuroplastic mechanisms of the treatments were listed in Table 1. Indeed, the etiology of depression is complex and could not be simplified by neural plasticity dysfunctions as the only targetable factor. As one of the most prevalent mental illnesses, the precise mechanism underlying depression still remains not yet clearly understood. Hence, further studies exploring the exact mechanisms about neuroplasticity changes in depression are needed to develop novel therapies for depression in the future.Table 1Treatment of depression and the underlying mechanism of neuroplasticity.Table 1Therapy/ TreatmentNeural plasticity changesDetection ApproachModelMechanism and influence of neuroplasticity****ReferencesBehavioralVolumetricConnectivityElectrophysiologicalCellularMolecularTCAAmitriptylineSucrose preference↑ Immobility time↓Relative p-NF-H labeling area in CA3↑Behavioral tests (SPT, TST) IHC (hippocampus) WBMouse-CUMSInducing cytoskeletal remodeling in the hippocampal CA3 region(Sanna et al., 2017)DesiparmineImmobility time↓DCX+ granule cells number↑ Dendritic extension↑ Terminal dendrites number↑ Spine density↑ V-shaped neurons in the DG↓Cdkn1c, p16↓ Klf15, Mbd1, TrkB, GAP-43, Disc1, Reelin↑Behavioral test (FST) IHC (hippocampus) Real-time PCRMouse-DEXAltering neurogenesis and neuronal morphology and restoring the expression of genes relevant to neuronal plasticity(Conti et al., 2017)ImipramineSocial avoidance↓ Immobility time↓Stubby spine density in pyramidal neurons of the basolateral amygdala↓ Total spine density in most superficial pyramidal neurons on layer II/III of the mPFC↑p-Creb, p-CamKII and PSD95 in BLA↓ but in mPFC↑Behavioral tests (SIT, TST) Golgi staining IHC (Amygdala and PFC)Mouse-CRSBlocking differential alteration of dendritic spine structure in amygdala and PFC(Leem et al., 2020)SSRIS-CitalopramLatency to feed↓ Immobility time↓DCX+ cell number (in SGZ and GCL)↑Behavioral tests (NSFT, TST) IHC (hippocampus) WB (hippocampus)Mouse-CUSIncreasing neurogenesis partly mediated by CB2 receptors(Ribeiro et al., 2021)FluoxetineImmobility time↓BDNF+ astrocyte number↑BDNF, p-Creb↑Behavioral test (TST) IHC (hippocampus) Luciferin-luciferase assay WB (cultured astrocytes)MouseIncreasing BDNF in astrocytes by increasing ATP gliotransmission(Kinoshita et al., 2018)FluoxetineImmobility time↓ Sucrose preference↑Iba1 intensity (DG, CA3, CTX)↓ GFAP intensity (DG, CA3)↓ Spine number↓IL-1β, IL-6, TNF-α, Nlrp3, Caspase1, p-mTOR, p-eEF2, 5-HT2A, HDAC1↓ IL-10, BDNF, SNAP25↑Behavioral tests (FST, SPT, TST) IHC (hippocampus) Golgi staining ELISA (Serum, hippocampus) WB (Hippocampus)Mouse- LPSIncreasing HDAC1-eEF2 activity and synaptogenesis and reducing inflammatory effects(Li et al., 2021a)ParoxetineImmobility time↓ Grooming time↑CORT, NK-κB, IL-1β, IL-18, IL-33↓Behavioral tests (FST, ST, TST) Fluorescence determination (plasma CORT) Real- time PCR (PFC)Mouse-ARSAttenuating HPA axis activation, modulating of the PI3K/Akt2 pathway and reduced neuroinflammation(de Oliveira et al., 2022)VortioxetinemRNA level of BDNF in the hippocampus↑ mRNA level of Arc in the PFC and c-Fos, Zif268 in the ventral hippocampus↓Real-time PCR WBRat-Acute swim stressIncreasing total BDNF expression in the hippocampus and modulating the expression of immediate early genes Arc and Zif268(Brivio et al., 2019)SNRILevomilnacipramcortical thickness of the right postcentral gyrus, supramarginal gyrus and lateral occipital cortex↑ Volume of left insula↓MRIHuman-LLDIncreasing cortical thickness of the right postcentral gyrus, supramarginal gyrus and lateral occipital cortex and reducing the left insula(Krause-Sorio et al., 2020)VenlafaxineOpen arm time↑ Immobility time↓Dendritic spine density↑ Dendritic branch length↑ Vesicle number↑SYN1, PSD95, SYP, SynGap↓Behavioral tests (EPM, FST) Electron microscopy Golgi staining IHC (hippocampus) WBRat-CRS, CORTRegulating the synaptic plasticity of hippocampal neurons(Zhao et al., 2021)KetamineKetamineBDNF↑IHC (hippocampus) Real-time PCR (hippocampus) WB (hippocampus)Rat-CUSEnhancing BDNF gene expression in the hippocampus via phosphorylation of HDAC5(Choi et al., 2017)KetamineMADRS score↓GBCr (lPFC, caudate and insular)↑ GBCr (cerebellum)↓MADRS rs-FC MRIHuman-TRD, MDDNormalizing the dysconnectivity between the PFC/subcortex and the rest of the brain(Abdallah et al., 2017a, Abdallah et al., 2017b)KetamineMADRS score↓Normalized connectivity between the insula and the DMNMADRS functional MRIHuman-MDDNormalizing the interaction between the DMN and SN(Evans et al., 2018)KetamineSucrose preference↑NMDAR fEPSP↓Ki67 + cell (in GCL and hilus)↑GluA4↓Behavioral tests (SPT) Electrophysiology IHC (DG) WB (hippocampus)Rat-DEXEnhancing DG proliferation(Michaëlsson et al., 2019)KetamineOpen arm time↑ Sucrose preference↑ Immobility time↓Rescue microcircuit dysfunctionSpine formation↑ Synaptogenesis↑ Spine density↑PSD95↑Behavioral tests (EPM, SPT, TST) IHC (mPFC) Two-photon imagingMouse-CORT, CRSRestoring lost spines and rescuing coordinated ensemble activity in PFC microcircuits(Moda-Sava et al., 2019)KetamineHAMD score↓FC between the cerebellum and SN↓ FC between cerebellum and the striatum↓ Normalized FC in the somatomotor network and DMNHAMD rs functional MRIHuman-MDDEnhancing neurofunctional plasticity between distinct neural networks(Sahib et al., 2020)KetamineImmobility time↓ Sucrose preference↑IL-1β, IL-6, TNF-α, NF-κB, MDA↓ SOD, GSH, BDNF↑Behavioral tests (FST, SPT) Chemical colorimetric analysis ELISA (hippocampus) WB (hippocampus)Rat-CPSPReducing proinflammatory cytokines, regulating oxidative stress and increasing BDNF in the hippocampus(Yang et al., 2020)KetamineCGI-S, HAMD, MADRS and SDQ score↓ PAS score↑fronto-striatal FC↑CGI-S, HAMD, MADRS, PAS, SDQ rs functional MRIHuman-TRDIncreasing fronto-striatal FC(Domany et al., 2019, Fava et al., 2020, Mkrtchian et al., 2021)KetamineMADRS score↓Subgenual ACC activation↓MADRS functional MRIHuman-TRDBlunting aberrant subgenual ACC hyper-reactivity to positive incentives(Morris et al., 2020)KetamineSucrose preference↑pSer880-GluA2 (synaptic membrane from hippocampus)↓ mGlu2↑Behavioral test (SPT) Real-time PCR (hippocampus) WB (hippocampus)Rat-CUMSIncreasing activity-dependent glutamatergic synaptic transmission(Elhussiny et al., 2021)R-KetamineSpine density (CA3, DG and prelimbic area of the mPFC)↑Golgi stainingMouse-CSDSAmeliorating the decrease spine density in the mPFC and hippocampus(Zhang et al., 2019a)S-KetamineImmobility time↓Volume of CA1. stratum radiatum area↑Total length of microvessel in CA1. stratum radiatum area and molecular layer of DG↑ Number of nonperforated and perforated synapses↑Behavioral test (FST) Electron microscopy (CA1 and DG)Rat-FSLCoregulating hippocampal vascularization and synaptogenesis(Ardalan et al., 2017)S-KetamineSpine density, mushroom spines and long-thin spines numbers (CA1 pyramidal neurons)↑GluN2A, GluN2B, p-cofilin↓ Homer3↑Rapid Golgi-Cox staining WB (hippocampus)RatIncreasing CA1 basal dendritic spine density via induction of long-thin spines and remodeling hippocampal structure in terms of rapid de novo spine growth (spinogenesis)(Treccani et al., 2019)(2 R, 6 R)-HNKImmobility time↓ Sniffing time↑ Latency to feed↓EPSCs frequency (layer V pyramidal neurons) ↑p-mTOR↑Behavioral tests (FST, FUST, NSFT) Electrophysiology WB (mPFC)MouseIncreasing synaptic function in the mPFC(Fukumoto et al., 2019)S-NorketamineImmobility time↓ Sucrose preference↑EPSC amplitude↓Spine density (pyramidal neurons in the mPFC, CA3 and DG)↑Behavioral tests (FST, SPT) Electrophysiology Golgi stainingMouse-CSDSIncreasing spine density in mPFC and hippocampus via a novel AMPAR activation-independent mechanism(Yang et al., 2018a)GLYX-13/ RapastinelImmobility time↓ Sniffing time↑ Latency to feed↓Spine density (apical dendritic tuft of layer V pyramidal neurons in the mPFC)↑p-mTOR, p-p70S6K, p-4EBP1, p-ERK, p-Akt, GluR1, SYN↑Behavioral tests (FST, FUST, NSFT) Electrophysiology Golgi staining (mPFC) WB (mPFC)RatEnhancing synaptic plasticity via mTOR pathway(Liu et al., 2017a)ScopolamineImmobility time↓ Sucrose preference↑AMPAR-fEPSP amplitude↑membrane GluA1 in CA1↑ p-mTOR, p-GluA1 in the hippocampus↑Behavioral tests (FST, SPT) Electrophysiology WB (hippocampus)Rat-LHInducing synaptic plasticity and mTOR pathway activation via a potential M2-PKA mechanism(Dong et al., 2018)Immobility time↓ Latency to feed↓ Sucrose preference↑Firing frequency (layer V pyramidal neurons in the mPFC)↑ SKC current amplitude↓Behavioral tests (FST, NSFT, SPT) Electrophysiology ELISA (mPFC and CPu)Rat-CUSEnhancing activity-dependent plasticity by regulating M1 receptor-SKC pathway(Bambico et al., 2020)Psychedelics/ 5-HT2AR agonistsPsilocybinImmobility time↓ Sucrose preference↑ Sniffing time↑fEPSP amplitude↑Behavioral tests (FST, SPT, FUST) Electrophysiology (hippocampus)Mouse-CMMSRestoring synaptic strength in cortico-mesolimbic reward circuits independent of 5-HTR2A(Hesselgrave et al., 2021)MelatoninMelatoninImmobility time↓TNF-α↓ BDNF↑Behavioral tests (FST, TST) ELISA (hippocampus)Mouse-LPSReducing TNF-α release and increasing BDNF expression(Taniguti et al., 2018)MelatoninImmobility time↓ Sucrose preference↑Iba1 + microglia number↑ Iba1 + microglia soma size↓ Ki67 + cell proliferation↑ BrdU+ newborn cells number↑Cx3cl1, Cx3cr1↑Behavioral tests (FST, SPT) IHC (DG) ELISA (Cx3cl1)Mouse-CUMSIncreasing neurogenesis and reversing the decreased immunoreactivity to Cx3cl1 and Cx3cr1 in the DG(Vega-Rivera et al., 2020)MelatoninImmobility time↓ Sucrose preference↑Beclin1, Lc3bII, Atg4/5↑ p-Foxo3a↓ Foxo3a↑Behavioral tests (FST, SPT, TST) IHC (hippocampus and CTX) WB (hippocampus and CTX)Mouse-LPSAttenuating autophagy impairment through Foxo3a regulation(Ali et al., 2020)MRAAgomelatineImmobility time↓ Grooming time↑ Sucrose preference↑ Latency to feed↓Iba1 + cell density↓ Neuron density↑ DCX+ cell density↑Restored circadian rhythm of melatoninBehavioral tests (FST, ST, SPT, NSFT) ELISA (plasma-melatonin) IHC (hippocampus)Rat-CCLReducing neuroinflammation and enhancing neurogenesis(Atanasova et al., 2021, Tchekalarova et al., 2018)AgomelatineImmobility time↓ Sucrose preference↑Iba1 + microglia number↑BDNF, TrkB, p-ERK1/2, Beclin, LC3II, Bcl2↑ Bax, NF-κB/p65, iNOS, nNOS↓Behavioral tests (FST, SPT) IHC (DG) WB (hippocampus)Mouse-CRSPromoting autophagy and decreasing apoptosis in the hippocampus(Chen et al., 2021a)BSTAcupuncture/ FluoxetineMADRS score↓rsFC between inferior ventral striatum and mPFC, ventral rostral putamen, amygdala and dorsal caudate and middle temporal gyrus↑ rsFC between right ventral rostral putamen and right dlPFC and right dorsal caudate and cerebellar tonsil↓MADRS functional MRIHuman-MDDModulating the corticostriatal motivation/reward circuitry(Wang et al., 2017)AcupunctureNormalized aberrant intrinsic FC within amygdala networksfunctional MRIHuman-MDDModulating abnormal amygdala networks(Duan et al., 2020)EAImmobility time↓ Sucrose preference↑fEPSP slope↑Sert↓Behavioral tests (FST, SPT) Electrophysiology (hippocampus) WB (hippocampus)Rat-WKYRestoring hippocampal synaptic plasticity via modulation of 5-HT receptors(Han et al., 2018)EASucrose preference↑Number and size of autolysosomes in CA1 neurons↓Lc3↓Behavioral test (SPT) IHC (left CA1) TEM (left CA1) WB (left CA1)Rat-CUMSInhibiting autophagy with a decreased number of autolysosomes(Zhang et al., 2020c)EAOpen arm time↑ Sucrose preference↑Median length of synapses↑ Median optical density↑GluR1, GluR2, Stargazin, SYP, PSD95, GAP-43↑ Pick1↓Behavioral tests (EPM, SPT) IHC (hippocampus) Nissl staining (hippocampus) Real-time PCR (hippocampus) WB (hippocampus)Rat-CUMSUpregulating expression of the AMPAR and protecting neural plasticity(Jiang et al., 2020)EA/ScopolamineSucrose preference↑ Latency to feed↓Spine density (layer V pyramidal neurons in PFC)↑BDNF, mTORC1, p-mTORC1, PSD95, SYN1, GluR1↑Behavioral tests (SPT, NSFT) Golgi staining WB (PFC)Rat-CUMSIncreasing synaptic plasticity in the PFC(Gao et al., 2022)EA/SSRISucrose preference↑Postsynaptic density↑ Reversed mitochondria lesionBehavioral test (SPT) Electron microscopy iTRAQRat-CUMSProtecting synaptic and mitochondrial functions in hippocampus(Zhang et al., 2019b)ECTBDI, HAMD, MMSE and VFT score↓ Improved moodAmygdala activity↓ FC strength of LAG↑ ALFF in dmPFC↑ dmPFC-PCC rsFC↑BDI, HAMD, MMSE, VFT rs functional MRIHuman-MDDIncreasing local activity and enhancing functional plasticity of the dmPFC and modulating FC of the LAG(Bai et al., 2019, Mo et al., 2020, Redlich et al., 2017, Wei et al., 2018)ECTHAMD score↓Cortical thickness in superior temporal gyrus and hippocampus and DG volume↑HAMD structural MRIHuman-TRDIncreasing regional GM volume and maybe increasing neurogenesis(Gbyl et al., 2021, Nuninga et al., 2020, Takamiya et al., 2019, Yrondi et al., 2019)ECTHAMD score↓Hippocampus, amygdala and putamen volume and temporal, parietal and insular cortices thickness (right hemisphere)↑ACC activity↑HAMD structural MRIHuman-MDD, TRDIncreasing hippocampus, amygdala and putamen volume, cortical thickness and ACC activity(Cao et al., 2018, Enneking et al., 2020, Gryglewski et al., 2019, Gyger et al., 2021)ECTHAMD, MADRS and QIDS-SR score↓Glu/Gln in the subgenual ACC↑ Creatine in the ACC↑ NAA in the right hippocampus↓ Glu/Gln in the left hippocampus↓HAMD, MADRS, QIDS-SR MRSHuman-MDDIncreasing restorative and neurotrophic processes(Njau et al., 2017)ECTHAMD score↓GM volume of left cerebellar area VIIa crus I↑ Medial temporal lobe and Perigenual ACC volume↑ Limbic GM↑rsFC (temporoparietal, prefrontal and cortical midline structures)↓ Hypothalamic rsFC↑NAA, Glu/Gln↑HAMD structural MRI (VBM, rsFC) MRS (hippocampus)Human-MDD, TRDReconciling the neuroplasticity and neuroinflammation, increasing regional GM and modulating rsFC(Cano et al., 2017, Depping et al., 2017, Thomann et al., 2017)MSTSSI score↓Cortical-evoked activity↑LTP↑SSI TMS-electroencephalographyHuman-TRDReducing suicidal ideation with concomitant decreases in cortical inhibition(Sun et al., 2018)taVNSHAMD and SDS score↓FC between the medial hypothalamus and rostral ACC↓ FC between the right rostral ACC and lingual gyrus↑ FC between the left rostral ACC and insula, dlPFC, middle cingulate cortex and precuneus↑GABA/Glu↓HAMD, SDS functional MRIHuman-MDD, TRDModulating the FC between the rostral ACC and medial hypothalamus, lingual gyrus, dlPFC, middle cingulate cortex and precuneus(Li et al., 2019, Tu et al., 2018)TMSImmobility time↑ Sucrose preference↑NeuN+ cell number in CA1 and DG↑ Ki67 + cell number in DG and SVZ ↑ DCX+ cell number in DG↑Bax↓ Bcl2, SYP, PSD95, NR2A/B, p11, BDNF, homer1a, p-TrkB↑Behavioral tests (FST, SPT, TST) IHC (hippocampus) WB (hippocampus)Mouse-CUMSPromoting neurogenesis and facilitating synaptic plasticity partly involved the p11/BDNF/Homer1a pathway(Zuo et al., 2020)CBTHAMD score↓GM volume of hippocampus and NAc↑rsFC between NAc and PCC↑HAMD MRIHuman-MMDModulating brain structure and FC plasticity within the NAc(Meng et al., 2021)DAR antagonistAmisulprideStriatal activation↑ Corticostriatal FC between the NAc and midcingulate cortex↑ (Reward learning↑)functional MRIHuman-MDDAcute enhancement of dopaminergic transmission potentiating reward-related striatal activation and corticostriatal FC(Admon et al., 2017)NMDAR antagonistIfenprodilImmobility time↓ Sucrose preference↑Rheb, p-mTOR, p-P70S6K, P706SK, BDNF, GluA1, IL-1β, IL-6, TNF-α↑ p-eEF2↓Behavioral tests (FST, SPT) ELISA (IL-1β, IL-6, TNF-α) WB (hippocampus and mPFC)Rat-CUMSModulating neuroplasticity by activating mTOR signaling and modulating proinflammatory cytokines(Yao et al., 2020)Lanicemine (AZD6765)Latency to feed↓ Immobility time↓SYN (in PFC)↑Behavioral tests (NSFT, TST) WB (PFC and hippocampus)Mouse-CUSEnhancing PI3K/Akt/mTOR/GSK3β signaling(Neis et al., 2020)Memantine/ SertralineImmobility time↓BDNF↑Behavioral test (FST) ELISA (BDNF)Rat-CUSIncreasing mRNA levels of BDNF and TrkB in PFC and hippocampus(Amidfar et al., 2017)NeuropeptideCPGImmobility time↓BDNF mRNA↑Behavioral test (TST) Real-time PCR (frontal CTX)Mouse-ASCEnhancing BDNF gene expression in the CTX(Abdullina et al., 2022)AgmatineImmobility time↓Ki67 + cell and number in the DG↑ PCNA+ cell number in the ventral DG↑ DCX+ cell number in the DG↑ Dendritic length (DCX+ immature neurons in the DG)↑Behavioral test (TST) IHC (hippocampus)Mouse-CORTIncreasing cell proliferation and dendritic arborization in the DG(Olescowicz et al., 2018)GSB-106Immobility time↓ Sucrose preference↑SYP, Creb, p-Creb↑Behavioral tests (FST, SPT) WB (hippocampus)Mouse-CSDSRestoring hippocampal neuroplasticity probably regulated by TrkB signaling(Gudasheva et al., 2021)ExerciseRunningImmobility time↓ Sucrose preference↑BOLD activation in PFC, corpus callosum and hippocampus↓ BOLD activation in amygdala, thalamus, midbrain and pontine↑Behavioral tests (FST, SPT, TST) functional MRIMouse-CUMSAmeliorating brain BOLD signals and adjusting the mood-regulating circuit(Huang et al., 2017)SwimmingImmobility time↓Sirt1, miR-124/134/138, Creb, BDNF, Akt, p-Akt, GSK3β, p-GSK3β, GAP-43, SYN↑Behavioral tests (FST, SPT, TST) Real-time PCR (hippocampus) WB (hippocampus)Mouse-CUMSIncreasing expression of hippocampal plasticity-related proteins(Liu et al., 2018b)SwimmingOpen arm time↑ Immobility time↓ Sucrose preference↑Crmp2, α-tubulin↑ p-Crmp2, Iba1, Caspase1, Cleaved caspase1↓Behavioral tests (EPM, SPT, TST) IHC (hippocampus) WB (hippocampus)Mouse-CUMSIncreasing the level of Crmp2 and α-tubulin in the hippocampus(Xie et al., 2022)YogaBDI score↓GABA↑BDI MRS (thalamus)Human-MDDIncreasing the activity of the GABA system(Streeter et al., 2020)LactateImmobility time↓ Sucrose preference↑BrdU+ and BrdU+NeuN+ cell number in DG↑Behavioral tests (FST, SPT, TST) IHC (hippocampus, culture cells)Mouse-CORTEnhancing adult hippocampus neurogenesis(Carrard et al., 2021)Gene modulationFto-overexpressionImmobility time↓ Sucrose preference↑Number of branches and dendritic spine density↑Fto, p-CamKII, p-Creb, PSD95, SYN↑Behavioral tests (FST, SPT) Electron microscopy Golgi staining Real-time PCR WB (hippocampus)Mouse-CRSEnhancing hippocampal synaptic plasticity by modulating CamKII/Creb signaling pathway(Shen et al., 2021)Mst1-knockdownRecognition index↑ Sucrose preference↑fEPSP amplitude↑Spine density (GNs in the DG)↑PSD95, SYP↑Behavioral tests (NORT, SPT) Electrophysiology Golgi staining WB (hippocampus)Mouse-CUMSIncreasing synaptic plasticity and efficacy(Chen et al., 2022)NPFFR2-knockoutImmobility time↓ Sucrose preference↑5-HT1AR↑ TLR4, TNF-α↓Behavioral tests (FST, SPT) Real-time PCR (hippocampus)Mouse-LPSDownregulating 5-HT1AR in the ventral hippocampus (serotonergic plasticity)(Yu et al., 2021)Pdcd4-knockoutOpen arm time↑ Immobility time↓ Sucrose preference↑Spine density (GNs in the DG)↑BDNF↑Behavioral tests (EPM, FST, SPT) ELISA (hippocampus-BDNF) Golgi staining Real-time PCR WB (hippocampus)Mouse-CRSBoosting BDNF expression(Li et al., 2021c)Pten-CKO/inhibitionOpen arm time↑ Immobility time↓ Latency to feed↓EPSCs amplitude↑ EPSCs frequency↑Diameter of 5-HTNs↑ Dendritic length of 5-HTNs↑ Puncta density↑ Ki67 + cell number in DG↑ p-Akt+Tph2 + /Tph2 + ↑p-Akt, Tph2, Sert, Map2, Map1a, Pfn2, Wasf1, BDNF, p-mTOR, PSD95, SYN1, Arc, p-4EBP1↑Behavioral tests (EPM, FST, NSFT, SPT, TST) Electrophysiology IHC (midbrain and hippocampus) WB (midbrain and hippocampus)Mouse-CSIncreasing dendritic branching, density of PSD95 + puncta in the dendrites and excitatory synaptic inputs of 5-HTNs(Chen et al., 2021b)Pten-knockdown/inhibitionImmobility time↓ Sucrose preference↑Dendrite atrophy↓ Soma size and total length of dendrites and dendritic branching of the PFC neurons↑Pten, p-Pten, p-ERK1/2, p-MEK1↓ p-Akt↑Behavioral tests (FST, SPT, TST) ELISA (CORT) IHC Real-time PCR WB (PFC)Mouse-CRS, DEXMediating neuron atrophy(Wang et al., 2021b)TCMα-CyperoneImmobility time↓ Sucrose preference↑Spine density↑ROS production, p-NF-κB, Nlrp3, Cleaved Caspase1, Cleaved IL-1β, Cleaved IL-18, GSDMD-N↓ Sirt3, PSD95, SYN1↑Behavioral tests (FST, SPT, TST) Golgi staining IHC (hippocampus) WB (hippocampus)Mouse-CUMSNeuroplasticity enhancement stimulated by Sirt3 via suppressing Nlrp3 inflammasome(Xia et al., 2020)Chaihu Shugan SanImmobility time↓ Sucrose preference↑Synaptic density (hippocampus)↑miR-124↓ Gria3, Mapk14↑Behavioral tests (FST, SPT) miRNA microarray Real-time PCR TEM WBRat-CUMSPromoting neuroplasticity in the hippocampus by downregulating miR-124 expression and by releasing the inhibition of the Mapk14 and Gria3 signaling pathways(Liu et al., 2018a)CrocinImmobility time↓ Sucrose preference↑GHSR, p-PI3K, p-Akt, p-mTOR, BDNF, GluR1, PSD95↑Behavioral test (FST, SPT, TST) WB (hippocampus)Mouse-PNSRestoring hippocampal synaptic plasticity-associated proteins through GHSR-PI3K signaling(Wu et al., 2020a)CurcuminImmobility time↓ Sucrose preference↑Iba1 + cell number↓ Iba1 + cell size↓ Dendritic spine and synapse densities within mPFC neurons↑IL-1β, IL-6, TNF-α, NF-κB↓Behavioral tests (FST, SPT) Electrophysiology ELISA (IL-1β, IL-6, TNF-α, NF-κB) Golgi staining IHC (mPFC) Real-time PCR WB (mPFC)Rat-CUMSIncreasing dendritic spine and synapse densities of mPFC neurons and repressing the inflammatory responses(Fan et al., 2018a)CurcuminImmobility time↓ Struggling time↑Aromatase in the limbic system↑ Serum oestradiol level↑ 5-HT, 5-HT1A/2A, Tph2, BDNF, ERK1/2↑ MaoA↓Behavioral test (FST) ELISA (oestradiol, ERK1/2) Real-time PCRRat-OVXModulating 5-HT and BDNF level(Abd-Rabo et al., 2019)CurcuminImmobility time↓ Latency to feed↓ Sucrose preference↑8-OHDG+ cell number↓ TUNEL+ cell number↓ Spine density and total dendritic length (GNs in the DG)↑CORT (serum), Nox2, 4-HNE↓ Nrf2, NQO-1, HO-1, p-Creb, BDNF, PSD95, SYP↑Behavioral tests (FST, NSFT, SPT) Golgi staining IHC Real-time PCR TUNEL staining WB (hippocampus)Rat-CUMSIncreasing spine density and total dendritic length of GNs in the DG by mitigation of oxidative stress and the activation of Nrf2 signaling pathway(Liao et al., 2020)Ginsenoside Rg1Immobility time↓ Sucrose preference↑Dendritic spine and synapse density (ventromedial PFC neurons)↑ Synapse density (BLA neurons)↑miR-134↓ Limk1, p-cofilin, Creb, p-Creb, BDNF↑Behavioral tests (FST, SPT) Electron microscopy Golgi staining Real-time PCR WB (ventromedial PFC and BLA)Rat-CUMSIncreasing spinogenesis and synaptogenesis within the ventromedial PFC and BLA by modulating the miR-134 signaling pathway and activating the Creb-BDNF system(Fan et al., 2018b, Yu et al., 2018)QuercitrinImmobility time↓ Sucrose preference↑IL-1β, IL-10, TNF-α, p-PI3K, p-Akt, p-NF-κB, p-ERK, p-p38, p-JNK↓ p-Creb, CBP, BDNF, PSD95, SYN1↑Behavioral tests (FST, SPT, TST) WB (hippocampus)Mouse-LPSInhibiting PI3K/Akt/NF-κB signaling and Enhancing p-Creb/BDNF/PSD95/SYN1 signaling in the hippocampus (improving neuroplasticity)(Sun et al., 2021)Radix PolygalaeImmobility time↓ Latency to feed↓Iba1 + cell number↓ Autophagosome number↑p62, p-Akt, p-mTOR, p-ULK1, NLRP3, Cleaved caspase1, IL-1β, IL-6, IL-18, TNF-α↓ Beclin1, LC3II, p-AMPK, GFAP↑Behavioral tests (FST, NSFT, SPT) IHC (PFC) Real-time PCR TEM (PFC) WB (PFC)Rat-CRSPromoting autophagy and inhibiting neuroinflammation(Zhou et al., 2021b)SiNiSanImmobility time↓ Sucrose preference↑Nissl+ cell number in the PFC and CA1↑ SYP+ area in the PFC and CA1↑PSD95, GAP-43, SYP, p-ERK1/2, PKC, CaSR↑Behavioral tests (FST, SPT) IHC Nissl staining WB (hippocampus and PFC)Rat-MS, CUMSImproving synaptic plasticity by activation of the CaSR-PKC-ERK signaling pathway(Shen et al., 2021)Acetyl-L-carnitineImmobility time↓ Social interaction ratio↑Dendritic length of stellate neurons in the medial amygdala↑Behavioral tests (FST, SIT) Golgi staining (amygdala)Mouse-CRSPromoting structural plasticity of the medial amygdala(Lau et al., 2017)AtorvastatinImmobility time↓BDNF↑ TNF-α↓Behavioral tests (FST, TST) ELISA (hippocampus and PFC)Mouse-LPSReducing TNF-α release and increasing BDNF expression in the hippocampus and PFC(Taniguti et al., 2019)ClozapineImmobility time↓ Sucrose preference↑BrdU+ , BrdU+PSA-NCMA+ and BrdU+NeuN+ cell number in DG↑ Total dendritic length of neurons in the DG and PFC↑BDNF (in the hippocampus and PFC), NCAM1, SYN1, Drd2 (in the PFC)↑Behavioral tests (FST, SPT) Golgi staining IHC (hippocampus) Real-time PCRRat-CUMSImproving adult neurogenesis, cell survival and neuronal reorganization(Morais et al., 2017)DiazepamImmobility time↓ Latency to feed↓Spine density (layer I apical dendrites in the PFC)↑ FosB+ cell number↓Csf1, C3↓Behavioral tests (FST, NSFT) Real-time PCR (PFC) IHC (mPFC)Mouse-CUSModulating neuronal activity and limiting microglia-mediated neuronal remodeling in the mPFC(Bollinger et al., 2020)EstradiolImmobility time↓ Latency to feed↓ Sucrose preference↑Dendrites density in the hippocampus↑ Dendrites density in the PFC↑p-cofilin1↑Behavioral tests (FST, NSFT, SPT, TST) IHC (hippocampus and PFC) Real-time PCR WB (hippocampus and PFC)Mouse-OVX, CRSPromoting the phosphorylation of cofilin1 and reducing the loss of dendrites and dendritic spines(Xu et al., 2021)IbrutinibImmobility time↓ Sucrose preference↑Spine number (apical dendrites from the hippocampus)↑PI3K, p-GSK3β, NF-κB, IL-1β, IL-6, TNF-α, Iba1, p-p38, Nlrp3, Caspase1, GFAP (in DG and CA3), p-eEF2↓ Snap25, PSD95, SYP, NR2A/B, BDNF↑Behavioral tests (FST, SPT) ELISA (hippocampus) Golgi staining IHC (hippocampus and CTX) WB (hippocampus)Mouse-LPSIncreasing BDNF level and spinogenesis and reducing neuroinflammation(Li et al., 2021b)LithiumImmobility time↓ Time in target quadrant↑ Preference index↑NeuN+ cell number↑ Reactive astrogliosis↓ Spine density, dendritic arborization, node and branch termination number (pyramidal neurons) ↑IL-2, IL-6, IL-1β, GFAP↓ Pparg, Ocln, Tjp1↑Behavioral tests (FST, MWM, NORT) IHC (hippocampus, CTX) Nissl and Golgi staining Real-time PCR WB (hippocampus, CTX)Rat-OVXSubduing neuroinflammation to maintain pyramidal cells arborization(Rana et al., 2022)NaHSImmobility time↓ Latency to feed↓PSD95, SYP1, Hkii, Pkm2, Ldha, Pdk1↑ PDH↓Behavioral tests (FST, NSFT, TST) WB (hippocampus)Rat- β2mIncreasing hippocampal synaptic plasticity as a result of enhancement of hippocampal Warburg effect(Yang et al., 2022)Phencynonate hydrochlorideSucrose preference↑ Immobility time↓Spine density (pyramidal neurons of the mPFC and CA3)↑NR1, NR2B↓ Kalirin7↑Behavioral tests (SPT, TST) Golgi staining WB (hippocampus)Rat-CUMSIncreasing dendritic spine density in the prelimbic and CA3 regions and regulating Kalirin7 and NMDAR level in the hippocampus(Zhu et al., 2021)↑, increase; ↓, decrease; 5-HTN, serotonergic neuron; ACC, anterior cingulate cortex; ALFF, amplitude of low-frequency fluctuations; AMPAR, α-amino-3-hydroxy-5-methyl-4-isoxazole-propionic acid receptor; ARS, acute restraint stress; ASC, antidepressants sensitive catalepsy; ATP, adenosine triphosphate; β2m, β2-microglobulin; BDI, Beck depression inventory; BDNF, brain-derived neurotrophic factor; BLA, basolateral amygdala; BOLD, blood oxygen level-dependent; BST, brain stimulation treatments; CBT, cognitive behavioral therapy; CCL, chronic constant light; CGI-S, clinical global impression-severity; CKO, conditional knockout; CMMS, chronic multimodal stress; CORT, corticosterone; CPG, cycloprolylglycine; CPSP, chronic postsurgical pain; CPu, caudate-putamen; CRS, chronic restraint stress; CS, chronic stress; CSDS, chronic social defeat stress; CTX, cortex; CUS, chronic unpredictable stress; CUMS, chronic unpredictable mild stress; DAR, dopamine receptors; DCX, doublecortin; DEX,dexamethasone; DG, dentate gyrus; dm, dorsomedial; DMN, default mode network; dl, dorsolateral; EA, electroacupuncture; ECT, electroconvulsive therapy; ELISA, enzyme linked immunosorbent assay; EPM, elevated plus maze; EPSCs, excitatory post-synaptic currents; FC, functional connectivity; fEPSP, field excitatory post-synaptic potential; FSL, Flinders Sensitive Line; FST, forced swim test; FUST, female urine sniffing test; GABA, γ-aminobutyric acid; GBCr, global brain connectivity with global signal regression; GCL, granule cell layer; Glu, glutamate; Gln, glutamine; GM, gray matter; GNs, granule neurons; HAMD, Hamilton depression rating scale; HNK, hydroxynorketamine; HPA, hypothalamic-pituitary-adrenal; IHC, immunohistochemistry; iTRAQ, isobaric tag for relative and absolute quantitation; l, lateral; LAG, left angular gyrus; LH, learned helplessness; LLD, late-life depression; LPS, lipopolysaccharide; LTP, long-term potentiation; m, medial; MADRS, Montgomery-Åsberg depression rating scale; MDD, major depressive disorder; MMD, mild to moderate depression; MMSE, mini mental state examination; MRA, melatonin receptor agonist; MRI, magnetic resonance imaging; MRS, magnetic resonance spectroscopy; MS, maternal separation; MST, magnetic seizure therapy; MWM, Morris water maze; NAA, N-acetylaspartate; NAc, nucleus accumbens; NMDAR, N-methyl-D-aspartic acid receptor; NORT, novel object recognition test; NSFT, novelty suppressed feeding test; OVX, ovariectomy; p-, phospho-; PAS, Positive Affect Scale; PCC, posterior cingulate cortex; PCR, polymerase chain reaction; PFC, prefrontal cortex; PNS, prenatal stress; PSD95, postsynaptic density protein-95; QIDS-SR, quick inventory of depressive symptomatology-self report; rs, resting state; SDQ, Symptoms of Depression Questionnaire; SDS, Sheehan disability scale; SGZ, subgranular zone; SIT, social interaction test; SN, salience network; SPT, sucrose preference test; SNRI, Serotonin and noradrenaline reuptake inhibitors; SSI, Scale for Suicidal Ideation; SSRI, selective serotonin reuptake inhibitors; ST, Splash test; SVZ, subventricular zone; SYN, synapsin; SYP, synaptophysin; taVNS, transcutaneous auricular VNS; TCA, tricyclic antidepressants; TCM, traditional Chinese medicine; TEM, transmission electron microscopy; TMS, transcranial magnetic stimulation; TRD, treatment-resistant depression; TST, tail suspension test; TUNEL, TdT mediated dUTP nick end labeling; VBM, voxel-based morphometry; VFT, verbal fluency test; VNS, vagus nerve stimulation; WB, western blot; WKY, Wistar Kyoto.
Therapies with high remission rate and low recurrence rate have to be developed urgently. As precision medicine become popular over recent years, it is encouraged to carry out personalized medicine in treating depression, whose heterogeneity is well-known. For instance, individualized functional MRI guidance may be of great advantage to increase the efficacy of rTMS in depression (Luber et al., 2017). Traditional antidepressants take weeks to months to demonstrate full effectiveness, but adverse effects occur more quickly, most of which subside within the first few days to weeks of therapy. Amidst these side effects, sexual dysfunction, nausea, weight changes and sleep disruption are the commonest and can be explained by the drug’s mechanism of action (Khawam et al., 2006). Ketamine shows rapid antidepressant effects, but also brings some psychotomimetic side effects and delirium (Mihaljevic et al., 2020). In general, animal and human studies demonstrated that short-term use of melatonin is safe with only mild adverse effects, while long-term safety of melatonin requires further investigation (Andersen et al., 2016). Brain stimulation treatments are relatively safer, but may still stand a chance of causing cognitive defects. Modification and improvements in treatment techniques could minimize such adverse effects (Andrade et al., 2016). Although exercise plays an exquisite role in mitigating depressive symptoms mainly by rebuilding brain structure, it remains to be improved that the existing exercise intervention programs are not fully applicable for depressive population (Zhao et al., 2020). TCM is a more holistic medicine, and may have its own unique features for emotional regulation. The effective ingredients of herbs worth further examinations.
Genetic factors play a key role in the neural plasticity of depression, and the relationship between genes and depression etiology/antidepressant efficacy is of special interest for scientists. Recent genome-wide association studies (GWAS) of depression have identified many susceptibility genes (Howard et al., 2018, Wray et al., 2018). Among them, neuronal growth regulator 1 (Negr1), member of the lgLON protein family, is implicated in axon extension and synaptic plasticity (Sanz et al., 2015). Besides, nuclear receptor subfamily 4 group A member 2 (Nr4a2, also Nurr1), a transcription factor, regulates genes of critical importance for dopaminergic neurotransmission (Sacchetti et al., 2001). Shisa family member 9 (Shisa9), enriched within the PSD, influences the electrophysiological properties of AMPARs and thus modulates short-term plasticity at specific excitatory synapses (von Engelhardt et al., 2010), and may be functionally involved in the synaptic anchoring of AMPARs through its interaction with PSD proteins (Haering et al., 2014). Seizure protein 6 homolog (Sez6), highly expressed in adult brain, plays roles in modulating neuronal morphology (Mitsui et al., 2013), establishing excitatory synaptic connectivity and controlling dendritic elongation and branching (Gunnersen et al., 2007). Furthermore, deleted in colorectal cancer (Dcc) has also been associated with MDD (Okbay et al., 2016), which controls dopamine connectivity in adolescence (Pokinko et al., 2017). Additionally, microRNAs and long non-coding RNAs are also involved in neural plasticity as well as episodes and management of depression, which has yet to be elucidated. As prevention is always better than cure, understanding the mechanisms underlying pathogenesis of depression will be beneficial to not only treat but also solve this illness.
This work was supported by the Shanghai Pujiang Program (20PJ1413300), the 10.13039/100007219Natural Science Foundation of Shanghai (22ZR1466300), the 10.13039/501100012226Fundamental Research Funds for the Central Universities, Shanghai Municipal Science and Technology Major Project (2018SHZDZX01), Zhangjiang (ZJ) Lab, and Shanghai Center for Brain Science and Brain-Inspired Technology.
Ethical approval was not required as this study did not involve any human or animal experiments.
The authors declare no conflict of interest.