Cerebrovascular-mediated dynamic alterations in neurovascular coupling: a key pathological mechanism of depression

Abstract

Neurovascular coupling (NVC) is a vital regulatory mechanism that synchronizes neural activity with vascular responses to support brain function. Although the precise mechanisms of NVC remain incompletely elucidated, its dysfunction is increasingly implicated in the pathogenesis of various neurological disorders. This review synthesizes recent advancements in understanding the vascular cascade, emphasizing key dynamic regulators of NVC, including mechanical forces and diffusible signals mediated by blood flow. We explore the intricate bidirectional interactions between the vasculature and neurons, highlighting their interdependent roles in neurovascular regulation. Using major depressive disorder (MDD) as a case study, we further discuss emerging evidence linking vascular dysfunction and impaired NVC to MDD pathophysiology. These insights position NVC as a promising therapeutic target for emotional disorders, underscoring the pivotal roles of hemodynamic signaling in neurovascular regulation.

Introduction

The brain maintains one of the body’s most sophisticated vascular networks, with its dense microvasculature playing an indispensable role in supporting neuronal function and metabolic homeostasis [1]. NVC represents a critical physiological process that coordinates bidirectional communication between neural activity and vascular responses, thereby maintaining precise metabolic-vascular homeostasis [2, 3]. During sensory processing or cognitive tasks, localized neuronal activation triggers rapid vasodilation, resulting in increased cerebral blood flow (CBF), blood volume (CBV), and tissue oxygenation [2, 4]. This neurovascular synchrony not only facilitates efficient delivery of oxygen and nutrients but also actively modulates neuronal excitability to maintain cerebral homeostasis. Beyond perfusion, NVC contributes to neuronal signaling [5, 6], stabilizes and optimizes cerebrovascular architecture [7], regulates brain temperature [8, 9], and drives cerebrospinal fluid (CSF) production and circulation [10–12], ultimately preserving brain homeostasis.

In recent years, the vascular system’s involvement in psychiatric disorders has gained increasing attention, offering a unifying framework to explain various pathophysiological abnormalities [13–16]. Perturbations in NVC have been consistently documented across multiple neuropsychiatric conditions, including MDD [17], autism spectrum disorder [18], and related mental illnesses [19, 20]. As the world’s most prevalent psychiatric condition [21, 22], MDD demonstrates particularly strong associations with both cardiovascular/metabolic comorbidities [23] and stress-induced microvascular dysfunction [17]. Notably, preclinical models reveal that cerebrovascular abnormalities often precede neuronal dysfunction in depression pathogenesis, highlighting the need to better understand the mechanistic relationships between vascular regulation, NVC integrity, and mood disorder development. This review explores the mechanisms underlying neurovascular communication and evaluates the role of NVC dysfunction in MDD, emphasizing its potential as a therapeutic target.

Cascade reaction of blood vessels in NVC

The structural diversity of cerebral vasculature implies distinct functional roles for different segments in regulating CBF. Arterioles and capillaries demonstrate the most pronounced relative CBV changes [1], highlighting their central importance in CBF modulation. Neuronal activity triggers vessel dilation via ion channels [24] and receptors [25], maintaining stable cerebral perfusion [26] (Table 1). The spatial pattern of vasodilation—ranging from focal microvascular expansion to widespread blood flow redistribution—emerges from discrete signaling pathways that propagate regulatory signals in a cascading manner.

Table 1.

Functions of different subtypes of channels and receptors in NVC

Channels	Functional relevance		Cell types	References
Kir2.1	Sense neuronal activity and propagate hyperpolarization in adjacent ECs		ECs	[147]
NMDAR	Detect synaptic activity and regulate functional hyperemia		ECs	[27, 147]
TRPA1	Instigate a propagating retrograde signal		ECs	[29, 134]
TRPV4	Induce vasodilation in upstream arteries		ECs	[148, 149]
IP3R	Mediating Ca2+ release		ECs	[33]
SKCa	Conduce the outward current and modulate CBF		ECs	[150]
IKCa	Conduce the outward current and modulate CBF		ECs	[150]
KATP	Sense neural activity and transform electrical signals		PCs	[32]
Kir	Sensitive to voltage and induce hyperpolarization		PCs	[151]
BKCa	Sensitive to voltage and induce hyperpolarization		PCs, SMCs	[28, 152–154]
SKCa	Sensitive to voltage and induce hyperpolarization		PCs, SMCs	[152, 155, 156]
sGC	Sense NO and open BKCa channel		PCs, SMCs	[152, 157]
P2X7	Sense ATP and induce depolatization		PCs	[158]
P2X4	Sense ATP and increase the intracellular Ca2+		PCs	[159]
Kir2	Sense neuronal activity and induce hyperpolarization		SMCs	[160]
NMDAR1	Mediating synaptic-like transmission with axons		SMCs	[25]
Na+, K+-ATPase	Sense the external K+ and induce hyperpolarization		SMCs	[161, 162]
KATP		Modulating intracellular Ca2+ oscillation	SMCs	[163]

Beyond mural cell-mediated vasodilation, endothelial cells (ECs) contribute significantly to NVC by integrating neurotransmitter and electrical signals [27–29]. Recent findings have identified an electrochemical (E-Ca) coupling mechanism in brain capillary ECs that facilitates the retrograde transmission of endothelial hyperpolarization signals from capillaries to upstream arterioles, adjusting blood supply to meet downstream demand [30]. Specifically, as sensors of neural activity and generators of hyperpolarized signals, pericytes (PCs) form a distributed electrical communication system that transmits signals upstream to ECs [31, 32], inducing IP3R-mediated Ca²⁺ release and propagating hyperpolarization via Kir 2 activation in adjacent ECs [33]. This mechanism establishes a spatiotemporal framework for cerebral blood flow regulation, integrating diverse signaling pathways across vascular compartments. Notably, perisynaptic capillary-to-arteriole signaling exhibits variability in onset time, directionality, and vessel diameter changes [34], underscoring the complexity of NVC dynamics. The interplay between capillary- and arteriole-level signaling cascades orchestrates cerebral perfusion with high spatiotemporal precision, ensuring optimal neuronal function.

Dynamic regulatory signals from vasculature to neurons

Mechanical forces

The myogenic response and cardiac cycle contribute to the autoregulation of CBF by modulating shear stress, maintaining cerebral perfusion. Arteriole diameter oscillations at ~ 0.1 Hz generate local physiological pressures [35] that elicit transient neuronal responses through mechanosensitive ion channels [36]. Neuronal Piezo-1 receptors respond to vascular pressure fluctuations, interacting bidirectionally with transient receptor potential (TRP) ion channels to modulate neurotransmitter release [37]. Piezo-2 detects rhythmic vascular pulsations, inducing spontaneous spike activity in mitral cells of the olfactory bulb, with similar pulsation-driven neuronal activation observed in the hippocampus and prefrontal cortex [38]. Furthermore, neurons distinguish varying mechanical stimuli from blood flow via voltage-gated channels (e.g., TRPV1, TRPV4), exhibiting transient or sustained responses. Notably, axons demonstrate a higher responsiveness to mechanical stimuli (73%) compared to somata and dendrites [36]. Interestingly, elevated cerebral blood flow/pressure in the rat cortex enhances the firing of somatostatin interneurons while suppressing pyramidal neuron activity [39]. This neuron-specific negative feedback regulates vascular responses to balance energy supply with demand, preventing both hyperperfusion and hypoperfusion damage in brain tissue.

Likewise, astrocytes exhibit the ability to sense mechanical stimuli derived from vascular dynamics. Emerging evidence reveals a sequential vessel-to-astrocyte-to-neuron signaling pathway, where astrocytes serve as critical intermediaries. For example, astrocytic TRPV4 channels respond to PA tone, initiating vascular responses while simultaneously modulating neuronal activity via adenosine release, as shown in vitro [39, 40]. Under physiological conditions, where cerebral autoregulation mechanisms operate efficiently, graded adjustments in neuronal activity—triggered by adenosine release—may encode varying degrees of vascular tone. This could provide a mechanism for the cortex to monitor changes in brain perfusion pressure. This interaction suggests that NVC ensures metabolic homeostasis, allowing neurons to adapt their activity in response to cerebral perfusion. This dynamic equilibrium maintains a balance between energy supply and demand.

Diffusible factors

Diffusible factors play a critical role in vascular-neuronal communication by freely crossing the blood–brain barrier and directly influencing neural activity.

Nitric oxide

NO is a gas neurotransmitter that well-characterized and essential for NVC in humans [41]. Neurophysiologists often focus on neuronal nitric oxide synthase (nNOS)-derived NO [42]. While, endothelial- and blood-derived NO can diffuse up to 100 μm from blood vessels [43, 44] and modulate neuronal function by binding to soluble guanylate cyclase (sGC) and calcium-dependent potassium channels [45]. Endothelial NO has been implicated in synaptic plasticity within the cortex and striatum [46, 47], generated tonic NO signals [48] which are crucial for hippocampal LTP [49]. Experiments with eNOS-/- mice demonstrate endothelial NO’s role in regulating neurogenesis and angiogenesis [50] (Table 2). Insufficient endothelial NO secretion triggers compensatory neuronal mechanisms [51], while nNOS upregulation may offset eNOS deficiency to preserve cognition, emotion, and motor control [52]. Collectively, these findings highlight NO as a key integrator of vascular and neuronal signaling, ensuring the bidirectional communication necessary for maintaining NVC.

Table 2.

Recent advances in eNOS deficiency and phenotype research

Knockout Strategies	Neuronal phenotype	Behavioural phenotype	References
eNOS±	–	Depressive-like behavior	[17]
eNOS−	Impairment of neocortical long-term potentiation	–	[46]
eNOS−/−	Reduce the occurrence of LTP in striatal	–	[47]
eNOS−	Generate tonic NO signals to induce hippocampal LTP	–	[49]
eNOS−	Decrease the CBF responses in the cortex evoked by whisker stimulation and by administration of ATP	–	[164]
eNOS−	Involved in an NMDAR-independent form of LTP in the cortex	–	[46]
eNOS−/−	Neuroinflammation	Spatial memory impaired	[165]
eNOS−/−	Demyelinate of cortical, corpus callosum and hippocampus in middle-ages mice	Gait behavior defect and association recognition memory disorder	[166]
eNOS−/−	Influence progenitor cell proliferation, neuronal migration and neurite outgrowth after stroke	Functional recovery disorder after stroke	[50, 167]
eNOS−/−	Decrease in retinal neovascularization and the expression of VEGF	–	[168, 169]

Hormones

Hormones, potent bioactive molecules secreted by endocrine glands, regulate physiological functions across the body and brain [53, 54]. Recent research underscores their role in synaptogenesis, neurogenesis, and NVC [55–57]. Several hormones and their receptors are directly involved in NVC. Angiotensin II (Ang II), a major component of the renin–angiotensin–aldosterone system (RAAS), secreted by renal juxtaglomerular cells, binds to receptors widely distributed in the brain. It regulates vasoconstriction, aldosterone secretion, and sympathetic activity, playing a critical role in fluid balance. Dysregulation of Ang II causes endothelial dysfunction [58, 59] through AT1R- and NADPH oxidase-mediated oxidative stress [60], leading to nNOS expression increased and subsequent NVC impairment [61].

Another key hormone, Insulin-like growth factor-1 (IGF-1), is produced primarily by the liver and exerts protective vascular effects. Age-related IGF-1 decline impairs CBF and NVC responses [62] by disrupting endothelial IGF1R signaling and NO-dependent vasodilation [63, 64]. Additionally, endothelial IGF-1 can also protective neuron [65] and prevent the disruption of BBB [66], maintaining normal NVC responses. Crucially, the disruption of hormone homeostasis triggers neuroprotective adaptations that include the Nrf2/ ARE antioxidant pathway and Trx/ Prx redox buffering system, which collectively counteract oxidative stress [67, 68]. These adaptive responses also regulate endothelial-mediated mitochondrial biogenesis, enhance SOD2 activity, and confer resistance to oxidative damage [69, 70].

Beyond these, signaling molecules such as miRNAs, lncRNAs, and exosomes circulate in body fluids [71] and mediate communication between neurons, ECs, and PCs [72]. These molecules may play a yet-undetermined role in NVC, warranting further investigation.

Caveolae

Caveolae, cholesterol- and lipid-rich membrane invaginations, serve as regulatory hubs for ion channels (e.g., Cav1.2, Nav1.5, NCX1) [73, 74] and mechanotransduction pathways [75]. These structures respond to mechanical stimuli, including membrane stretch, by flattening to provide mechanical protection [76] and modulating signal transduction through protein interactions [77]. While most brain endothelial cells suppress caveolae to control cellular transport and ensure the integrity of the blood–brain barrier [78–80], their role in NVC has been largely unexplored. Recent findings indicate that inhibiting endothelial caveolae in brain arterioles reduces the NVC response by approximately 50%, independent of the eNOS-NO pathway. Simultaneous inhibition of both caveolae and the NO pathway abolishes the NVC response entirely [81]. This suggests that caveolae are essential platforms for clustering ion channels and receptors involved in NVC. Identifying the specific ion channels localized within endothelial caveolae will be critical for elucidating their contribution to cerebrovascular regulation.

Brain temperature

Brain temperature is regulated by a dynamic interplay between metabolic activity, CBF, and systemic body temperature. Even minor fluctuations in brain temperature can disrupt vascular and neuronal homeostasis. Substantial evidence indicates that such changes elicit corresponding vascular responses. For example, as brain temperature decreases, hemoglobin affinity for oxygen increases and reduces cerebral oxygen saturation [82, 83], inducing mild blood–brain barrier leakage in the hypothalamus and piriform cortex, and disrupting cerebral water homeostasis [84]. Conversely, an increase in cortical temperature amplifies vascular oscillations, leading to enhanced power within the 0.05–0.25 Hz frequency range [85]. Positron emission tomography (PET) studies further demonstrate that localized increases in CBF correlate with regional temperature shifts and alterations in oxygen metabolism [86].

Beyond effects on vascular, temperature fluctuations influence synaptic function. Elevated temperature potentiates synaptic transmission via TRPV4-mediated mechanisms [87, 88], which increase presynaptic vesicle release probability and synaptic inhibition, ultimately suppressing spiking activity in thalamocortical neurons [89]. Most importantly, neuronal activity exerts feedback effects on vascular function. For instance, activation of TRPC4 channel in Preoptic region stimulates the hypothalamic warm-sensitive neurons, thereby increasing CBF to facilitate heat dissipation [90]. Recent MRI findings suggest an inverted U-shaped relationship between cortical temperature and evoked neuronal and hemodynamic responses [85], demonstrating that cooling delays neurovascular coupling (NVC). These observations underscore the critical role of brain temperature in modulating neurovascular interactions. Collectively, the evidence underscores the importance of blood flow in neural information processing and its impact on NVC dynamics (Fig. 1).

Fig. 1.

Mutual communication in neurovascular coupling. Cascade reactions in NVC. Synaptic activity triggers the release of extracellular K⁺, NO, and glutamate, which act on ion channels and receptors in PCs, SMCs, and endothelial cells, leading to NO release and hyperpolarization. Additionally, peripheral hormones can interact with endothelial cells via blood circulation or diffuse to neurons, binding to their receptors. Vascular oscillation-induced mechanical forces and blood flow-regulated brain temperature also modulate neuronal activity

Clinical evidence of NVC dysfunction in MDD patients

Imaging study

Mounting evidence links NVC impairment to MDD, with neuroimaging studies consistently reporting altered CBF, reduced arteriolar dilation, and delayed hemodynamic responses [3]. In MDD patients, abnormalities in the fractional amplitude of low-frequency fluctuations (fALFF), disrupted activity in the default mode network (DMN), and shortened time-to-peak (TTP) in hemodynamic response function (HRF) parameters have been documented [91]. MRI studies reveal significant reductions in CBF in key regions implicated in mood regulation, including the prefrontal cortex [92, 93], anterior cingulate cortex [94], thalamus, and left superior temporal gyrus [95]. Notably, alterations in prefrontal-limbic functional connectivity (FC) in severe depression have been linked to suicidal ideation [96]. Additionally, abnormal directional connectivity in the cortical-subcortical-cerebellar network may lead to unbalanced integrating the emotional-related information for MDD, and further exacerbating depressive symptoms [97].

Antidepressant interventions further support the vascular basis of MDD pathophysiology; escitalopram has been shown to enhance CBF in the left temporal lobe, frontal lobe, and anterior cingulate cortex [98], while fluoxetine improves fMRI-BOLD responses in MDD patients [99, 100]. Most neuroimaging studies operate on the fundamental assumption that reduced CBF in MDD reflects neuronal hypometabolism; however, a more direct interplay between cerebrovascular circulation and neuronal activity may underlie these findings. The precise mechanisms underlying NVC impairment in MDD remain elusive, prompting further exploration of potential vascular changes at both structural and functional levels.

Structural changes in the brain vascular network

The cerebral vasculature may function as an interoceptive system, integrating homeostatic reflexes and allostatic responses—including motivational behaviors and emotional states—to maintain physiological equilibrium. Chronic stress induces profound morphological changes in the brain's vascular network, contributing to the pathophysiology of neuropsychiatric disorders. Recent advancements in optical coherence tomography angiography (OCTA) have facilitated clinical assessments of vascular changes in MDD. For instance, MDD patients exhibit abnormal retinal vascular fractal dimensions [101], and symptom severity correlates with retinal vascular diameter [102]. Swept-source optical coherence tomography angiography (SS-OCTA) studies further reveal reduced choroidal vascular density and decreased macular vascular density in the superficial retinal capillary plexus of MDD patients [103].

In elderly individuals with MDD, cortical gray and white matter exhibit increased vascular segment density and perivascular space expansion [104]. Moreover, depression severity correlates with elevated serum levels of angiogenesis inhibitors, such as endostatin [105], suggesting that pathological vascular remodeling plays a central role in mood disorders. Postmortem stereological studies support these findings, reporting increased neurovascular cells and potential vascular alterations in the basolateral amygdala of MDD patients [106]. These findings collectively indicate that MDD is associated with structural cerebrovascular abnormalities, reinforcing that vascular integrity is intricately linked to emotional and cognitive function.

Blood–brain barrier damage

The blood–brain barrier (BBB), a crucial neurovascular interface, maintains central nervous system homeostasis and ensures normal NVC function. Emerging evidence from both MDD patients and animal models indicates BBB disruption [13, 84, 107]. Dynamic contrast-enhanced (DCE) MRI and structural imaging studies have revealed significantly elevated mean volume transfer constants (K trans) in the olfactory region, caudate nucleus, and thalamus of MDD patients compared to healthy controls, suggesting a direct association between BBB leakage and depressive symptom severity [108]. Notably, BBB damage appears more pronounced in female MDD patients, with a ~ 50% reduction in Claudin5 mRNA levels in the prefrontal cortex (PFC) and marked vascular morphological alterations [107]. These findings suggest that chronic stress induces region-specific BBB breakdown, contributing to MDD pathology and potentially explaining gender differences in disease prevalence and severity.

Endothelial cell dysfunction

Endothelial dysfunction, characterized by abnormal endothelial activity, is a key contributor to vascular pathophysiology and has been detected in depression [109]. MRI studies have demonstrated increased relative transit time heterogeneity (RTH) in the capillaries of the bilateral prefrontal cortex, ventral anterior cingulate cortex, and left insular cortex in MDD patients, along with a decreased cerebral metabolic rate of oxygen (nCMRO₂), indicating capillary dysfunction and impaired NVC within the ventral circuit [110]. Additionally, flow-mediated dilation (FMD) assessments in middle-aged and elderly MDD patients have revealed a correlation between microvascular dysfunction and neurocognitive decline, reinforcing the link between endothelial dysfunction and cognitive impairment in MDD [111]. Evidence suggests that early-life stress impairs visceral microcirculation and endothelial function [112], increasing susceptibility to cardiovascular and cerebrovascular diseases later in life [113–115]. These findings underscore the importance of endothelial cell protection and repair as a potential therapeutic strategy for restoring NVC integrity in MDD.

Vascular inflammatory response

Autoimmune diseases exhibit high comorbidity with MDD, suggesting that chronic stress may provoke immune responses targeting the brain [116]. The shared psycho-immune-neuroendocrine (PINE) network model provides a mechanistic framework linking vascular inflammation and depression [117]. MDD patients display elevated plasma levels of pro-inflammatory cytokines, including interleukin (IL)-6, tumor necrosis factor (TNF)-α, platelet-derived growth factor (PDGF), granulocyte–macrophage colony-stimulating factor (GM-CSF), and IL-8, particularly in response to psychosocial stressors [118]. These inflammatory responses are primarily mediated by serum- and glucocorticoid-regulated kinase 1 (SGK1) [119], autonomic imbalances leading to excessive catecholamine release [120], and disruptions in serotonergic and glutamatergic neurotransmission [91], further exacerbating neurovascular dysfunction. These findings highlight the critical role of chronic stress-induced vascular inflammation in MDD, implicating oxidative stress, neuroinflammation, metabolic dysregulation, and neurotransmitter imbalances in the disorder's pathophysiology.

NVC impairment in models of depression: insights from preclinical studies

The above-mentioned research proves that chronic stress-induced vascular dysfunction disrupts NVC, leading to metabolic deficits and energy shortages in MDD patients. Although the precise mechanisms remain incompletely understood, recent advances in animal models offer promising insights into the intersection of neurovascular and metabolic dysregulation in MDD (Fig. 2).

Fig. 2.

Neurovascular coupling dysfunction in MDD patients. Chronic stress disrupts endothelial ion channels and functions, reduce the area of astrocyte endfoot and vessel, leading to PCs loss and BBB breakdown. The compromised BBB allows inflammatory factors and catecholamines from hyperactive HPA axis signaling to infiltrate the brain parenchyma, inducing ROS accumulation and synaptic loss. Reduced cerebrovascular density and CBF lead to ATP deficiency, impairing action potential firing and circuit dysfunction

Cerebrovascular damage and angiogenesis

A fully developed and functionally intact cerebral vascular network is essential for normal NVC function [121]. However, preclinical studies have revealed region-specific differences in NVC, though the underlying mechanisms remain unclear. Wu et al. mapped the distribution and spatial relationships of cerebral blood vessels, neurons, and PCs in mice, providing insights into these mechanisms [122]. Key brain regions involved in information processing, such as the somatosensory and primary auditory cortices, exhibit higher vascular density and more responsive NVC compared to other regions. In contrast, with its sparse capillary network, the hippocampus demonstrates weaker NVC responses and is more vulnerable to hypoxia [123]. These findings highlight the susceptibility of specific brain regions to stress and suggest potential mechanisms linking cerebrovascular network damage and NVC impairment in MDD.

Single-cell RNA sequencing of brain endothelial cells in mice subjected to chronic social stress revealed upregulation of angiogenesis-related genes and VEGF signaling in capillaries, indicating their potential role in vascular repair following stress exposure [124]. Additionally, human and animal studies have demonstrated that antidepressant treatment promotes endothelial cell proliferation, particularly in the hippocampus and hypothalamus [125–127]. Electroconvulsive therapy [128] and antidepressants promote VEGF- and BDNF-mediated angiogenesis via the SIRT1/FOXO1 pathway, thereby supporting neurogenesis and exerting antidepressant effects [129]. Collectively, these findings underscore the critical role of vascular integrity in maintaining NVC and suggest that targeting angiogenesis may offer a promising therapeutic approach.

BBB damage

Chronic social stress and subchronic variable stress have been shown to alter BBB integrity in emotion-related brain regions in mice [107], suggesting a key pathological link between stress exposure and psychiatric disorders. Chronic stress increases VEGF levels, and VEGF/VEGFR2 signaling compromises the paracellular barrier. Pharmacological inhibition of VEGFR2 mitigates chronic stress-induced BBB damage and anhedonia, further supporting its role in MDD pathophysiology [130]. In addition, aberrant Claudin5 expression and its promoter histone modifications contribute to BBB leakage in stress-exposed mice [131]. Besides, the decrease of the coverage area of the astrocyte endfoot and vessels is involved in the leakage of BBB. Such as the abnormal expression of CB1 receptor in astrocyte endfoot is linked to perivascular endogenous cannabinoid signal, which obstruct the expression of the vascular-related genes and compromised BBB stability. This mechanism has been associated with stress susceptibility and emotional disorders [132].

BBB breakdown permits peripheral inflammatory mediators such as IL-6 and TNF-α to infiltrate the nucleus accumbens (NAc) and hippocampus, exacerbating neuroinflammation. Interestingly, overexpression of Claudin5 or inhibition of histone deacetylase 1 (HDAC1) reverses BBB dysfunction and neuroinflammation, improving depressive-like behaviors in mice [13]. However, these interventions do not ameliorate stress-induced adrenal hypertrophy or weight loss, indicating that while BBB integrity protects against inflammatory neurotoxicity, it remains a downstream response to autonomic nervous system and HPA axis hyperactivation. These findings suggest that targeting BBB stabilization may alleviate MDD symptoms but should be complemented by broader therapeutic strategies addressing systemic stress responses.

Endothelial dysfunction

Cerebrovascular endothelial cells play a central role in NVC by expressing ion channels and receptors that translate neuronal activity into vascular responses (Table 1). Subchronic stress (7 days) has been shown to downregulate Kir 2.1 mRNA in parenchymal arterioles and reduce Kir current density in smooth muscle, impairing Kir channel function and NVC [133]. Endothelial TRPA1, a key sensor of inflammation and oxidative stress, has been implicated in MDD pathophysiology. Pharmacological TRPA1 antagonists reverse despair-like behaviors in mice, suggesting its potential as a therapeutic target [134]. Additionally, endothelial GPR4 (H⁺ receptor) and Gαq/11 proteins mediate cerebrovascular CO₂ sensitivity, and their dysfunction contributes to anxiety-like behaviors and respiratory disturbances in stress-exposed mice [20]. These findings indicate that stress-induced inflammation disrupts endothelial ion channel function, representing a key mechanism underlying vascular and NVC dysfunction in MDD. However, research on endothelial ion channels and receptors under chronic stress remains limited, hindering a full understanding of their role in NVC dysfunction.

Reduced CBF and energy homeostasis imbalance

Chronic stress-induced circulatory dysfunction and energy deficits are well-documented in preclinical models of depression [17]. When vascular dysfunction leads to insufficient energy supply, AMPK acts as a cellular energy sensor, reducing neuronal excitability to adjust local energy availability [135, 136]. However, chronic stress has been shown to reduce AMPK levels and disrupt synaptic integrity in the cortex of depression model mice [137]. While exogenous ATP supplementation has emerged as a potential therapeutic approach for MDD [138, 139], excessive extracellular ATP can cause neuronal hyperactivation and exacerbate metabolic stress [140]. Thus, targeting vascular homeostasis to balance energy supply and demand may offer a more effective treatment strategy.

Neuronal subtypes have distinct energy demands [141], such as glutamatergic neurons (vGlut1⁺) account for ~ 80% of gray matter energy consumption due to their role in neurotransmission [142]. When mitochondrial respiratory chain inhibitors are applied, hippocampal parvalbumin (PV) neurons—but not pyramidal neurons—exhibit firing deficits [143], suggesting that PV neurons are particularly susceptible to energy depletion. In energy-deficient conditions, which neuronal populations are most vulnerable? Moreover, metabolic pathways differ between neuronal somata (which rely primarily on glycolysis) and synaptic terminals (which depend on oxidative phosphorylation) [144]. In response to stress or energy shortages, do neurons undergo metabolic reprogramming to maintain function? Investigating such adaptive mechanisms may provide critical insights into energy homeostasis and novel treatment strategies for MDD.

Conclusion

NVC embodies the integrated function of neurons and blood vessels in information processing, extending beyond metabolic supply to include dynamic signaling that influences neural activity. Despite its significance, our understanding of NVC remains limited, with several fundamental questions unanswered. Notably, NVC exhibits remarkable regional heterogeneity—manifesting as nonspecific, absent, or even inverted responses in certain brain areas [145, 146]. The underlying causes of this spatial variability warrant systematic investigation. Are these differences solely attributable to vascular architecture, or do they reflect selective vulnerabilities of specific neurovascular unit components or cerebrovascular segments? Resolving these questions will not only advance our understanding of NVC’s physiological role in brain function but also elucidate its pathological alterations in psychiatric disorders, ultimately clarifying the basis of regional and cellular variations in neurovascular interactions.

This review has systematically examined cerebrovascular impairments observed in both MDD patients and animal models. Key findings include structural alterations in the vascular network, BBB disruption, endothelial dysfunction, and neuroinflammatory processes- all of which contribute to NVC deficits. It is well known that depression represents a complex, chronic systemic disorder involving multiple organ systems and interconnected pathological pathways. A comprehensive understanding of its pathophysiology requires interdisciplinary collaboration between researchers and clinicians to elucidate the disorder’s multifactorial risk profile. Through a vascular-focused perspective, we have investigated how cerebrovascular dysfunction serves as a critical mediator of NVC impairment and contributes to MDD pathogenesis. These insights underscore the promising therapeutic potential of vascular-targeted interventions for psychiatric disorders, opening new avenues for treatment development.

Abbreviations

AMPK

AMP-activated protein kinase

BBB

Blood–brain barrier

BDNF

Brain-derived neurotrophic factor

BK_Ca

Large conductance, calcium-activated potassium channel

[Ca²⁺]i

Intracellular Ca²⁺ concentration

CBF

Cerebral blood flow

CBV

Cerebral blood volume

ECs

Endothelial cells

K_ATP

ATP-sensitive potassium channel

Kir

Inward rectifier potassium channels

MDD

Major depressive disorder

NO

Nitric oxide

NVC

Neurovascular coupling

PA

Parenchymal arteriole

PCs

Pericytes

P2X

P2X purinoceptor

sEPSCs

Spontaneous excitatory postsynaptic currents

sGC

Soluble guanylyl cyclase

SK_Ca

Small conductance, calcium-activated potassium channel

SMCs

Smooth muscle cell

IGF-1

Insulin-like growth factor-1

IK_Ca

Intermediate conductance, calcium-activated potassium channel

IP₃R

Inositol triphosphate 3 receptor

IP-TNTs

Interpericyte tunneling nanotubes

TRPA1

Transient receptor potential A1

TRPV4

Transient receptor potential vanilloid 4

VEGF

Vascular endothelial growth factor

Author contributions

LY, YC conceived and designed project. XY, MZ, prepared the figures. AH, XH, FG, NW prepared the reference. XY, YZ, JZ, JW wrote the manuscript. XY, LY, YC, JY, YX, TW, MZ helped revise the manuscript. All authors read and approved the final manuscript.

Funding

This work was supported, in part, The National Science Foundation of China (8227142405), The National Science Foundation of China (82374586), the National Comprehensive Traditional Chinese Medicine Reform Demonstration Zone Science and Technology Collaborative Development Project (GZY-KJS-SD-2024-046), The Shanghai institute of Traditional Chinese Medicine for mental health (SZB2024101), the Shandong Traditional Chinese Medicine Technology Development Project (NO. M-2022198), Shandong University of Traditional Chinese Medicine Students’ Innovation and Entrepreneurship Training Program Project (2024059).

Data availability

Not applicable.

Declarations

Ethics approval and consent to participate

Not applicable.

Consent for publication

With the submission of this manuscript, we would like to undertake that all authors of this paper have read and approved the final version submitted.

Competing interests

The authors declare no competing interests.