A limited number of endogenous NSCs are enriched in the DG and SVZ. These NSCs possess the ability to self-renew and differentiate into neurons and various other cell types. The proliferation, differentiation, and survival of these NSCs are influenced by multiple factors within the DG microenvironment (15), including neurotransmitters, growth factors, cytokines, and hormones secreted by other cell types (16, 17). These processes are believed to play a crucial role in maintaining the structural and functional integrity of the hippocampus.

Ischemic injury enhances the proliferation of endogenous NSCs during the acute phase post-stroke (18, 19). However, this increase in NSCs does not appear to significantly impact the hippocampal formation (20–22). Several potential explanations may account for this phenomenon. Firstly, NSC proliferation is typically restricted to the acute and/or subacute phases of stroke. Kathner-Schaffert et al. demonstrated a significant decline in endogenous NSCs proliferation six months following a stroke (23). Generally, this proliferation begins approximately one week post-stroke, peaks around days 10 to 14, and subsequently returns to baseline levels within four to five weeks after onset (19, 24). Secondly, only a small fraction of endogenous NSCs are capable of surviving due to the inhospitable microenvironment in the post-stroke brain (19). Thirdly, the stroke event can lead to a shift in neurogenesis toward astrogenesis (25), which, in theory, reduces the pool of newly generated neurons. Furthermore, an increased number of morphologically altered aberrant neurons, characterized by aberrant basal dendritic trees or abnormal locations within the hilus, can be observed in the hippocampus following a stroke (19, 26). Consequently, while hippocampal neurogenesis is activated post-stroke, it may represent an abnormal process both functionally and structurally (27).

Depression is a common mental health disorder characterized by complex and diverse pathological mechanisms. Numerous studies have demonstrated a significant reduction in neurogenesis within the hippocampus of patients with depression, which is closely associated with symptoms such as memory impairment and difficulties with concentration. Moreover, mouse models of major depressive disorder (MDD) show decreased proliferation and differentiation of neurons in the hippocampus (28–31).

The onset of depression is often influenced by a range of physiological and environmental factors, including chronic stress, inflammatory responses, and hormonal imbalances (32). Research has indicated that prolonged stress can lead to a decrease in neurotrophic factors in the hippocampus, such as brain-derived neurotrophic factor (BDNF). This reduction directly inhibits neurogenesis in the hippocampus, potentially exacerbating depressive symptoms (33). Furthermore, inflammatory responses are also believed to play a critical role in the pathophysiology of depression, with inflammatory markers identified as significant in the brains of individuals with depression, adversely affecting hippocampal structure and function, and further diminishing neurogenesis (34–36).

Interestingly, existing treatments for depression, such as antidepressant medications and psychotherapy, have been shown to promote neurogenesis in the hippocampus. For instance, selective serotonin reuptake inhibitors (SSRIs) enhance neurotransmitter transmission and facilitate neuronal growth by increasing the expression of BDNF. Consequently, these therapeutic approaches not only alleviate depressive symptoms but may also improve cognitive function in patients by enhancing neurogenesis within the hippocampus (8). These findings collectively suggest that decreased hippocampal neurogenesis may contribute to depressive symptoms.

Hippocampal neurogenesis plays a critical role in both the development of ischemic brain injury and depression. Indeed, accumulating literature has revealed the correlation between hippocampal neurogenesis and PSD (37, 38). Mariel Pietri et al. built a middle cerebral artery occlusion (MCAO) model of mice, which reliably replicates the core clinical features of stroke and associated PSD, including motor and cognitive deficits, neuronal loss, and impaired neurogenesis, alongside the observation of depressive-like behaviors ten weeks post-stroke (24).

It has been posited that PSD is a consequence of decreased hippocampal neurogenesis in stroke patients. However, due to the challenges in obtaining human hippocampal tissue samples, a reliable animal model is essential for further study. Moriyama et al. developed a PSD model by subjecting MCAO-treated mice to chronic mild stress (CMS), which resulted in reduced sucrose consumption (a marker of anhedonia) (39). Their findings indicate that CMS treatment impairs spontaneous neurogenesis in the DG approximately one month post-stroke, aggravates apoptotic injury, and disrupts circuit integration in mature neonatal cells, ultimately leading to the occurrence of PSD (40, 41). Additionally, Kim et al. found that MCAO mice subjected to CMS exhibited anxiety-like behavior, anhedonia, and despair-like symptoms (28). Zhang et al. established a PSD model using MCAO in conjunction with post-stroke isolated housing conditions, observing impairments in hippocampal neurogenesis at 3 days and memory function at 14 days post-stroke (42).

The hypothalamic-pituitary-adrenal (HPA) axis is a major neuroendocrine system responsible for the body’s response to stress. Hyperactivity of HPA axis is one of the most prominent biological findings in depression (43). Following a stroke, the hypothalamus detects ischemic injury signals and relays this information to the paraventricular nucleus, which subsequently produces corticotropin-releasing hormone (CRH). CRH stimulates the pituitary gland to release adrenocorticotropic hormone (ACTH), which promotes the synthesis and release of corticosteroids (44) (Figure 1). Elevated corticosteroid levels in the central nervous system (CNS) are significantly associated with PSD (45). Research has demonstrated that corticosteroids reduce neurogenesis and neuronal survival in the hippocampus (46).

The actions of corticosteroids are mediated by members of the nuclear receptor superfamily, specifically the glucocorticoid receptor (GR) and mineralocorticoid receptor (MR). These receptors bind to their ligands, functioning as transcription factors to directly regulate gene expression. They also produce rapid non-genomic effects on neuronal excitability and activation, thereby influencing cognitive processes (47, 48). In a study by Kim et al., a stress response was induced in mice using corticosterone, which resulted in reduced proliferation of hippocampal NSCs and NPCs, along with decreased expression of GR (49). Further confirmation of this association was supported through experiments involving GR siRNA knockdown and overexpression. Additionally, corticosteroids may act synergistically with neurotransmitters, neurotrophic factors, and other stress mediators, shaping both current and future responses to ischemic insults. Given its high density of GR and MR, the hippocampus is particularly sensitive to corticosteroid effects. Corticosteroids have been shown to decrease hippocampal neurogenesis and impair synaptic plasticity (46). Administration of the GR antagonist mifepristone can partially reverse PSD symptoms (50), further underscoring the importance of this pathway.

Neuroinflammation occurs as a response to ischemic injury and involves a complex process characterized by the activation of inflammatory cells and the release of cytokines. Inflammatory mediators, such as tumor necrosis factor-alpha (TNF-α), interleukins (IL-1α, IL-1β, IL-6), and interferon (51), along with inflammatory cells like microglia and astrocytes, contribute to protecting the brain from further damage and promoting healing. However, excessive inflammation may lead to additional brain injury (52). The link between neuroinflammation and the onset of PSD has been well documented (53, 54).

After a stroke, microglia and astrocytes quickly recognize ischemic stimuli and neuronal damage, triggering an inflammatory response. Activated microglia and astrocytes release pro-inflammatory or anti-inflammatory factors, thus modulating neuroinflammation (52). Pro-inflammatory (M1) microglia can exacerbating neuroinflammation by recruiting peripheral monocytes and activating astrocytes (55). Inhibiting neuroinflammation has been shown to improve neurogenesis in animal models (56, 57). Additionally, CMS can promote pro-inflammatory differentiation of microglia and increase astrocyte activation in the DG while simultaneously inhibiting neurogenesis (58). Reactive astrocytes not only participate in neuroinflammation but also play a significant role in regulating neurotrophic factors expression and neuronal transmission (52, 59). Interestingly, oligodendrocytes may also be involved in neuroinflammation through communication with reactive astrocytes (60). Furthermore, neuroinflammation induced by abnormal peripheral immune cell activation can impair neurogenesis (35) (Figure 2).

Dysbiosis of intestinal microbiota has emerged as a recent area of interest concerning the mechanisms underlying PSD. Some researchers propose that stroke alters gut microbiota (GM), leading to increased gut permeability and the subsequent release of systemic inflammatory factors. This process may exacerbate the severity of stroke-related damage and raise the likelihood of post-stroke complications by disrupting T cell homeostasis, inducing pro-inflammatory responses, and promoting oxidative stress (61–63) (Figure 2). However, whether GM dysbiosis is a direct causal driver of impaired hippocampal neurogenesis and depressive symptoms—or rather a secondary consequence of stroke-related physiological stress—has yet to be definitively established. Although preclinical interventions such as probiotic supplementation and fecal microbiota transplantation have shown promising antidepressant effects, their clinical applicability, mechanistic specificity, and long-term safety remain unresolved.

Pro-inflammatory cytokines play a crucial role in the regulation of cell apoptosis and necrosis, particularly in the vulnerable hippocampus (40, 64, 65). Accumulating evidence suggests that inflammation promotes the development of PSD (52, 66–68). Various cytokines have been associated with reduced hippocampal neurogenesis (69, 70). Ischemic insults increase TNF-α levels, which adversely affect hippocampal neurogenesis, while exogenous TNF-α administration exacerbates cell proliferation and survival within the hippocampus (71). Moreover, IL-1β can have both positive and negative effects on hippocampal neurogenesis, depending on concentration and context. Low to moderate levels of IL-1β enhance NSC proliferation and survival, whereas excessive IL-1β exerts detrimental effects. Clinical research indicates that PSD patients present with elevated serum IL-1β levels compared to non-PSD patients (72). Furthermore, increased serum IL-6 levels in the acute phase of ischemic stroke have been independently associated with the onset of depression two weeks and one year post-stroke (66). Currently, it is known that IL-6 and IFN-γ have regulatory effects on the tryptophan-serotonin pathway (73).

Neurotrophic factors play an indispensable role in the proliferation, survival, development, function, and plasticity of neurons. Key neurotrophic factors include nerve growth factor (NGF), BDNF, neurotrophin-3/4, neurotrophin-5, and neurotrophin-6 (74). Additionally, several cytokines, such as leptin, insulin-like growth factor-1 (IGF-1), transforming growth factor, fibroblast growth factor, and platelet-derived growth factor (75), are present in the central nervous system and can modulate neurogenesis. Among these, BDNF is considered to have the most significant and widespread neurotrophic effect, particularly in promoting and sustaining neurogenesis in the hippocampus. BDNF not only supports the survival and differentiation of newly generated neurons but also facilitates their integration into existing neural circuits. Ischemic injury typically increases BDNF expression (76, 77). However, studies have reported that plasma levels of BDNF in patients with acute stroke are 2.5 times lower than those in healthy controls. Furthermore, the level of BDNF decreased in the chronic stage (78). Research by Mayuri N. Tuwar et al. indicated that patients with acute ischemic stroke had low BDNF levels upon admission (average 2.94 ng/mL) within the first 24 hours. These levels significantly increased to an average of 6.79 ng/mL between 25 and 48 hours but subsequently declined to 2.77 ng/mL after 96 hours (79).

DNA methylation, a process known to regulate gene expression at the transcriptional level, may contribute to the reduction of BDNF following a stroke (80, 81) (Figure 3A). Hypoxia-induced hypermethylation of the BDNF gene can lead to the epigenetic silencing of CpG islands, which may correlate with the prevalence and persistence of PSD (82). The application of 5-Aza-2′-deoxycytidine (a DNA methyltransferase inhibitor) after stroke has shown to increase BDNF levels by modulating DNA methylation (83). Additionally, single nucleotide polymorphisms (SNPs), such as the rs6265 variant, alter the BDNF gene from a G allele to an A allele at position 196, resulting in the amino acid substitution of valine at codon 66 with methionine (84, 85). This SNP is associated with a reduction in BDNF secretion without affecting overall gene expression (86) (Figure 3B). MiRNA-210 has been identified in cellular experiments as a regulator of BDNF transcription and is observed at elevated levels in stroke patients, negatively correlating with BDNF concentrations (Figure 3C).

Yang et al. demonstrated that serum BDNF concentrations below 5.86 ng/mL on day 1 post-stroke were independently correlated with PSD, with an 11.5-fold increase in PSD risk at three months when BDNF levels fell below 10.2 ng/mL (87). A multicenter prospective study conducted in 2024 revealed that elevated serum BDNF levels in ischemic stroke patients were independently associated with a reduced risk of developing PSD at three months (88). Chen et al. confirmed that low BDNF concentrations could lead to diminished neurogenesis rates in the hippocampus (89). Results from the study by Sheng Huang et al. indicate that the intravenous injection of hippocampal exosomes derived from stroke mice exacerbates depression-like behaviors in recipient mice. This effect is potentially mediated through the modulation of neurogenesis by depression-related proteins, specifically proBDNF and its receptor p75NTR (90). Maintaining low BDNF levels post-stroke may compromise the survival, differentiation, and integration of new neurons derived from NSCs in the hippocampus.

Neurotransmitters, chemical substances released from presynaptic neurons, bind to receptors on postsynaptic neurons to modulate various neural functions. Key neurotransmitters involved in the development of PSD include glutamate, dopamine, 5-HT, and norepinephrine. Among these, the monoamines-dopamine, 5-HT, and norepinephrine are found to be reduced in patients experiencing PSD (91). The amine hypothesis, which supplements the lesion location hypothesis, suggests that cerebral lesions disrupt projections ascending from the midbrain and brainstem through the thalamus and basal ganglia, ultimately reaching the frontal cortex, thereby reducing monoamine synthesis (Figure 4). Additionally, the disorder of 5-HT may be linked to DNA methylation of the serotonin transporter gene SLC6A4, which plays a critical role in 5-HT recycling (92). Highly methylated SLC6A4 promoters have been associated with depression following a stroke, from two weeks to more than one year (92).

5-HT functions as a growth-promoting factor in the central nervous system during development and is essential for neuronal and synaptic plasticity in the adult brain (93). Furthermore, proliferation of NPCs and survival of adult neurons are positively correlated with 5-HT levels (11). As emotional behaviors are regulated by several neurotransmitters, particularly monoamines, the symptoms of depression may result from impaired monoamine function (94). The concentration of monoamines is significantly lower in patients with depression (95), a similar trend observed in those with PSD (91). Of note, Wang et al. found a significant decrease in 5-HT1A receptors and mRNA levels in the hippocampus of animals exhibiting PSD (96). Hence, we speculate that the reductions in monoamines and/or their receptors in the hippocampus may lead to a reduction in neurogenesis, which and then triggers the PSD. Research by Villa et al. indicates a significant reduction in dopamine levels in rats exhibiting depression-like states, with this change closely associated with decreased functionality of D2 receptors (97). Dopamine functions as a neurotransmitter in the hippocampus, paralleling the role of the 5-HT system in regulating hippocampal neurogenesis, which contributes to depressive-like behaviors (12, 98). The noradrenergic system is believed to have a direct impact on hippocampal neurogenesis. Notably, the multi-step process of hippocampal neurogenesis, which includes proliferation, maturation, and survival, is positively regulated by the serotonergic system, while only the early stages of adult neurogenesis are influenced by the noradrenergic system (11).

Glutamate-mediated excitotoxicity may present another factor contributing to reduced hippocampal neurogenesis. Following an acute stroke, glutamate levels in both the brain and plasma increase (99). In patients with depression, cerebrospinal fluid concentrations of glutamate are significantly higher compared to healthy controls (100). A magnetic resonance spectroscopy (MRS) study (101) has yielded similar findings in patients with PSD. As a primary neurotransmitter in the brain, glutamate plays a crucial role in various brain functions, including neurogenesis and mood regulation. Evidence suggests that glutamate and the activation of its receptors significantly contribute to neurogenesis (102). Conversely, abnormally high levels of glutamate, resulting from increased release and/or decreased clearance, can cause neuronal atrophy, inhibit neurogenesis, and lead to depression (103). These opposing effects on neurogenesis and neuronal survival depend on glutamate concentration. An abnormal increase in glutamate following a stroke may inhibit neurogenesis in the hippocampus and remodel dendrites and cytoarchitecture (104, 105), consequently contributing to the development of PSD.

The post-stroke neurobiological alterations do not occur in isolation. Complex “crosstalk” exists among the HPA axis, neuroinflammation, neurotransmitter systems and neuronutrients, collectively contributing to hippocampal neurogenesis impairment.

Aside from neurobiological factors, the social and psychological responses to post-stroke functional impairments are considered major contributors to PSD. Clinical researches have shown a higher risk of PSD with post-stroke functional impairments and lack of social support, both perceived and objective (130, 131) (Figure 5A). Wang et al. investigated predictors of PSD, revealing that neurological function had the most significant overall effect on PSD, with social support acting as the most substantial direct factor, while family function had the largest indirect effect (132). A cross-sectional study indicated a negative correlation between self-efficacy in participation and depression in stroke survivors (133). Increasing social support interventions would prevent or alleviate depressive symptoms (134).

Animal models of PSD are generally constructed by stroke model combined with CMS (28). The design of this combination includes mild or low-grade stressors, such as overnight illumination, food/water deprivation, social isolation and soiled cage (Figure 5B). These stressors aim to simulate the daily problems and difficulties faced by stroke patients. The modified CMS program has been widely recognized as an effective method to induce depression. The evaluation of depression and despair behaviors in these animals is conducted through sucrose consumption tests, forced swimming tests, and field trials (40, 94). CMS can exacerbate tissue damage following a stroke, leading to severe clinical symptoms of depression through mechanisms like neuronal loss, brain degeneration, and impaired neurogenesis (135, 136). In preclinical experiments, Zhang et al. found that chronic stress results in the depletion of the aNSC pool and impairs the development of new neurons in the DG (30). Sylvia F Fawzi et al. observed that norepinephrine and 5-HT levels were reduced in CMS rats, accompanied by a significant increase in corticosterone, leading to increased production of pro-inflammatory cytokines (IL-6 and TNF-α) (137). Additionally, CMS can also alter ischemia-induced neurogenic fate by promoting the differentiation of NPCs into glial lineage cells (41) with the Notch signaling pathway, which is especially expressed in sites of neurogenesis, including SGZ (138). This pathway affects the fate of NSCs, promoting differentiation into astrocytes while inhibiting differentiation into neurons and oligodendrocytes (139). Moreover, it affects the growth and extension of axons (140) and disrupts synaptic formation between mature and newborn neurons (141–143). Zhang et al. found that post-stroke social isolation-mediated PSD may reduce hippocampal neurogenesis and impair memory function, involving regulatory expression of transforming growth factor-β (42). Intranasal delivery of hyperforin alleviates the PSD by increasing hippocampal neurogenesis (42).

Studies have provided support for the close relationship between hippocampal neurogenesis and depression in adult mammals (32, 33). A decrease in proliferating and differentiated neurons has been observed in the hippocampus of depressed mice, a finding that is also noted in PSD models (28, 41). However, the specific mechanisms governing this process are not yet fully understood. It is currently widely believed that this relationship involves an interplay of neurobiological, behavioral, and social factors, including interactions among glucocorticoids, cytokines, neuronutrients, and neurotransmitters, alongside CMS-induced stress, leading to reduced hippocampal neurogenesis (144) (Figures 6A, B). These findings highlight the causal relationship between ischemia-induced neurogenesis and depression-like behavior, shedding light on the etiology of PSD and offering potential targets for treatment.

Pharmacotherapy remains the cornerstone of PSD management. Several drug classes have demonstrated efficacy, with their mechanisms increasingly linked to promoting neurogenesis.

This category encompasses promising but less-established interventions where clinical evidence is nascent and significant translational hurdles remain.