The Cerebral Lymphatic System: Function, Controversies, and Therapeutic Approaches for Central Nervous System Diseases

Abstract

Abstract

The meninges serve as critical barriers that maintain immune homeostasis in the central nervous system (CNS) and play vital roles in immune surveillance and defense. Traditionally, the brain has been regarded as an “immune-privileged” organ owing to the absence of conventional lymphatic vessels. However, the rediscovery of meningeal lymphatic vessels (MLVs) has revealed a mechanism for the directional transport of cerebrospinal fluid (CSF) to the deep cervical lymph nodes (dCLNs), demonstrating that the brain possesses a distinct fluid communication pathway with the peripheral system that is independent of blood circulation. Additionally, the identification of the glymphatic system has revealed a perivascular mechanism for solute exchange between the CSF and brain parenchyma, primarily mediated by the astrocytic water channel protein aquaporin-4 (AQP4). These discoveries have significantly expanded our understanding of brain fluid dynamics and CNS homeostasis. This review provides a comprehensive overview of the structure, regulation, and function of MLVs and the glymphatic system, which together constitute lymphatic system of the brain. We also discuss recent evidence, particularly conflicting perspectives, on the role of meningeal immunity in various central nervous system (CNS) disorders, such as multiple sclerosis, Parkinson’s disease, and epilepsy. Furthermore, we explore the therapeutic potential of targeting the brain lymphatic system to treat these conditions. Given their critical roles in CNS homeostasis, MLVs and the glymphatic system have emerged as promising therapeutic targets, potentially offering novel treatment strategies for currently incurable neurological diseases.

Graphical Abstract

The figure illustrates the mechanisms and modulators of the cerebral lymphatic system, including recent insights and influencing factors (left); its involvement in six neurological disorders (middle); and associated therapeutic strategies (right)

Introduction

The brain, the most complex and functionally advanced component of the nervous system, regulates vital human activities. Historically, the brain has been regarded as an “immune-privileged” organ owing to the absence of conventional lymphatic vessels composed of lymphatic endothelial cells. However, Wilhelm His and Gustav Schwalbe first identified a connection between the CSF and the lymphatic system when Berlin Blue was injected into the central nervous system (CNS) of dogs and observed that a network of dural lymphatics facilitated the drainage of interstitial fluid (ISF) from the CNS into peripheral lymph nodes. In the twenty first century, the presence of dural lymphatics has been confirmed in both rodents and humans (Ahn et al. 2019; Louveau et al. 2015), establishing that the brain microenvironment communicates with the peripheral lymphatic system via CSF drainage into the dCLNs. Additionally, recent studies have identified a glymphatic system within the brain responsible for nutrient transport and metabolic waste clearance through a paravascular pathway primarily regulated by the astrocytic water channel protein AQP4 (Iliff et al. 2012). Collectively, these systems contribute to the maintenance of intracerebral homeostasis (Louveau 2018).

The onset and progression of CNS disorders are intricately linked to the maintenance of intracerebral homeostasis. The glymphatic system and meningeal lymphatics play crucial roles in various neurological diseases, including multiple sclerosis (MS; Brosnan & Raine 2013; Maggi et al. 2018), Parkinson’s disease (PD; Ding et al. 2021), cerebral small vessel disease (CSVD; Esposito et al. 2019), epilepsy (Feldman et al. 2018), major depressive disorder (MDD; Roomruangwong et al. 2017), and traumatic brain injury (TBI; Bolte et al. 2020; Piantino et al. 2022). Dysfunction of the glymphatic system and meningeal lymphatics contributes to disease pathogenesis by facilitating the accumulation of neurotoxic molecules, disrupting brain tissue integrity, and modulating immune and inflammatory responses. However, the impact of these systems is not necessarily unidirectional. For instance, in MS, meningeal lymphatics exacerbate neuroinflammation (Hsu et al. 2019; Yang et al. 2024), while on the other hand, they exert protective effects by influencing oligodendrocytes and astrocytes, thereby improving the disease prognosis (Louveau et al. 2018). Moreover, the glymphatic system and meningeal lymphatics are implicated in secondary brain injury in various CNS disorders, particularly in pathological processes such as tau protein aggregation and cerebral edema (Jessen et al. 2015; Mestre et al. 2020). However, conflicting findings regarding the role of these systems in cerebral edema require further investigation. Thus, targeting the cerebral lymphatic system may be a promising therapeutic strategy for treating CNS disorders.

This review aims to explore the physiological functions of meningeal lymphatics in intracerebral homeostasis, the role of the lymphatic system in ISF–CSF exchange, and the implications of these systems in the pathophysiology and therapeutic approaches for various CNS disorders.

Structure and Function of Meningeal Lymphatic Vessels

Meningeal lymphatic vessels express all classical lymphatic endothelial cell markers, including vascular endothelial growth factor receptor 3 (VEGFR3), prospero homeobox protein 1 (Prox1), podoplanin (gp38), lymphatic vessel endothelial hyaluronan receptor 1 (LYVE-1), CD31, and C-C motif chemokine ligand 21 (CCL21; Louveau et al. 2015). Based on their anatomical location, MLVs are classified into the dorsal intracranial surface and ventral base (Hu et al. 2020), each exhibiting distinct structural characteristics and drainage functions.

Ventral basal MLVs are extensively branched, exhibit distinct lymphatic endothelial features such as valves, and demonstrate a clear structural distinction between capillaries and collecting lymphatics, making them particularly suitable for CSF absorption (Hu et al. 2020). Furthermore, their proximity to the subarachnoid space, particularly in areas lacking arachnoid separation at the cribriform plate, facilitates direct CSF drainage. With ongoing research, the pathways involved in MLV-mediated fluid drainage have been elucidated. CSF is primarily produced by the choroid plexus epithelium and flows through the ventricles and subarachnoid space (Spector et al. 2015). The CSF then exchanges with ISF through the PVS of the penetrating arteries and is ultimately drained by arachnoid granulations and MLVs. Macromolecules in the subarachnoid space are transported to dCLNs and superficial cervical lymph nodes via MLVs (Hu et al. 2020).

Emerging evidence suggests that different meningeal layers exhibit varying degrees of involvement in autoimmune processes in the CNS (Merlini et al. 2022). Specifically, effector T cell infiltration and activation predominantly occur in the leptomeninges, whereas the dura mater appears to be relatively excluded from direct CNS autoimmune involvement (Merlini et al. 2022). This is demonstrated in experimental autoimmune encephalomyelitis models, in which myelin basic protein (MBP) injection in mice leads to intense inflammation in the leptomeninges, characterized by substantial infiltration of effector T cells and myeloid cells (Merlini et al. 2022). In contrast, the dura mater exhibits a minimal inflammatory response, likely due to a lack of intrinsic antigen-presenting capacity (Merlini et al. 2022). The dura mater plays a crucial role in meningeal immunity by facilitating CSF drainage and metabolite clearance and modulating central immune cell dynamics.

MLVs, characterized by relatively loose endothelial junctions, function in concert with CNS immune cells to establish a tightly regulated immuno-lymphatic interface (Hu et al. 2020; Mogensen et al. 2021). Vascular endothelial growth factor C (VEGFC) enhances CSF drainage into dCLNs and induces MLVs remodeling by promoting lymphangiogenesis and upregulating neuroprotective signaling pathways, as revealed by single-nucleus RNA sequencing of brain cells (Boisserand et al. 2024). VEGFC administration increases MLVs diameter and density, leading to improved CSF lymphatic drainage, suppression of pro-inflammatory microglial responses, and enhancement of neurotrophic signaling (Boisserand et al. 2024). Notably, VEGFC preconditioning elevates the levels of brain-derived neurotrophic factor, a key regulator of neuronal differentiation, maturation, and synaptic plasticity, which are critical for CNS development and neuroprotection following injury or disease (Boisserand et al. 2024). Conversely, genetic deletion of VEGFC or VEGFR3, pharmacological inhibition of VEGFR signaling via tyrosine kinase inhibitors (e.g., sunitinib), or the use of VEGFC/D traps results in impaired meningeal lymphatic development and reduced CSF drainage into dCLNs (Antila et al. 2017). These findings suggest that VEGFC-mediated modulation of MLVs function has therapeutic potential for neurological disorders, including enhanced clearance of pathological proteins, such as glutamate, amyloid-beta (Aβ), and tau proteins, immune cells, and inflammatory mediators, thereby preventing their pathological accumulation in the CNS.

Glymphatic System: CSF–ISF Communication

The traditional theory of CSF homeostasis presents several challenges. For instance, studies have demonstrated that labeled substances with different molecular weights, such as albumin (69 kDa) and polyethylene glycol (900 kDa), are cleared from the brain within nearly identical timeframes, despite a fivefold difference in their diffusion coefficients (Cserr et al. 1981). The rapid exchange of ISF and CSF within the brain cannot be adequately explained by the conventional diffusion theory, suggesting the presence of a distinct metabolite clearance mechanism unique to the brain that differs from that in other tissues and organs. Based on this hypothesis, Iliff et al. first proposed the existence of a CNS lymphatic-like system in a murine model. Their findings revealed a fluid transport pathway in the CNS that parallels the peripheral lymphatic system (Iliff et al. 2012).

This system facilitates efficient CSF–ISF exchange through the paravascular pathway. The CSF enters this system via the PVS surrounding the choroidal arteries (Iliff et al. 2012). As the CSF follows the branching of blood vessels, it crosses the astrocytic end-feet via AQP4 channels and enters the PVS, where it contributes to nutrient delivery and metabolic waste clearance (Hannocks et al. 2018). After mixing with ISF and undergoing molecular exchange within the brain parenchyma, CSF exits the brain through multiple drainage routes, including MLVs (Kress et al. 2014) (Fig. 1).

Fig. 1.

Mechanisms of CSF–ISF communication and the role of AQP4 involved. Bottom, schematic diagram of a coronal brain section showing the lymphatic pathway along the cerebral vessels; Top, schematic showing the case of CSF–ISF communication in the case of AQP4 polarization and depolarization, respectively. Right, the CSF communicates with the ISF via a paravascular pathway that allows cerebrospinal fluid from the subarachnoid space to enter the lymphatic system via the PVS. Waste products are drained from the ISF within the parenchyma into the CSF, further cleared by the drainage of functional MLVs. The blue dashed box shows the structure of AQP4, a water channel protein composed of six helical transmembrane domains and two highly conserved regions (Asn-Pro-Ala) that form narrow hemipores. Left, mislocalization or depolarization of AQP4 fails to support CSF–ISF exchange, contributing to glymphatic system dysfunction

AQP4 is a water channel protein isoform exclusively expressed in astrocytes and ependymal cells of the ventricular lining. It shows a polarized distribution on over 60% of the astrocytic end-feet surrounding the blood vessels (Nielsen et al. 1993). Under physiological conditions, AQP4 exhibits polarized localization, which is crucial for maintaining the efficiency of glymphatic clearance (Simon et al. 2022). It is predominantly positioned at the plasma membrane of perivascular astrocytic end-feet, facing the PVS, and anchored by the dystrophin-associated protein complex (DAPC). This configuration enables selective water transport and reduces the hydrodynamic resistance of the plasma membrane, facilitating fluid movement from the perivascular space into the ISF and supporting the transport of small molecules and ions (Jessen et al. 2015). Additionally, advanced imaging techniques, such as large-volume three-dimensional scanning electron microscopy and optoelectronic microscopy, reveal gaps in astrocytic coverage of penetrating cortical arterioles in mice. These structural discontinuities permit direct communication between the vasculature and the subcellular structures of adjacent cells, allowing the passage of macromolecular compounds (Zhang et al. 2024a, b) (Fig. 1). In contrast, under pathological conditions, such as neuroinflammation, trauma, and aging, AQP4 depolarization is associated with glymphatic dysfunction.

A growing body of research indicates that AQP4 not only provides structural support to astrocytic and ependymal cell membranes but also serves as a crucial regulator of fluid dynamics within the brain (Ho & Yang 2024). It facilitates fluid exchange between various compartments, including the vasculature, interstitial space, intracellular space, and CSF (Ho & Yang 2024). Moreover, AQP4 plays a significant role in CSF production and circulation. Disruptions in AQP4 localization or polarization at astrocytic end-feet can impair glymphatic drainage, potentially contributing to hydrocephalus and neurodegenerative conditions such as Alzheimer’s disease and idiopathic normal pressure hydrocephalus (Reeves et al. 2020; Simon et al. 2022). Overall, glymphatic system functionality is highly dependent on AQP4 expression and its precise localization within astrocytic end-feet (Palazzo et al. 2019).

In summary, the glymphatic system mediates ISF–CSF exchange, whereas MLVs facilitate CSF drainage to the dCLNs, thereby maintaining intracranial homeostasis. Together, these components form the cerebral lymphatic system, which serves as a crucial interface between the CNS and the peripheral immune system.

Glymphatic Circulation: Influencing Factors

Glymphatic function is critically dependent on CSF circulation and the bulk flow of ISF within the brain, both of which are intrinsic components of the glymphatic system (Chong et al. 2022; Jiang-Xie et al. 2024). Arterial pulsation serves as the primary driving force of CSF circulation, and the rhythmic expansion and contraction of arteries during the cardiac cycle (systole and diastole) dictate the overall direction of CSF flow (Enzmann and Pelc 1991). Notably, the peak velocity of CSF movement coincides with arterial pulsation, providing microscopic evidence that arterial pulsation is a key regulator of CSF inflow (Mestre et al. 2018).

Studies utilizing in vivo two-photon microscopy have demonstrated that unilateral internal carotid artery ligation reduces arterial pulsation and subsequently slows the rate of perivascular CSF–ISF exchange (Mestre et al. 2018). Conversely, systemic administration of the adrenergic agonist dobutamine significantly enhances penetrating artery pulsation, leading to a concomitant increase in the rate of CSF–ISF exchange (Mestre et al. 2018). In addition to arterial pulsation, changes in arterial diameter and fluctuations in venous blood volume resulting from the systolic-diastolic modulation of vascular smooth muscle cells can markedly influence intracerebroventricular CSF flow and metabolite clearance (Mateo et al. 2017). Respiratory movements also contribute to CSF dynamics by modulating intracranial vascular function. The absence of valves in the intracranial venous system renders it highly sensitive to central venous pressure, which fluctuates in response to changes in intrathoracic pressure during the respiratory cycle (Wilson 2016). Additionally, respiratory-induced alterations in the arterial wall are shown to affect CSF circulation (Mestre et al. 2018).

Using two-photon microimaging, Xie et al. observed that the volume fraction of the interstitial space was significantly higher during natural sleep and anesthesia than during wakefulness, suggesting that the glymphatic function is markedly enhanced during sleep (Xie et al. 2013). In awake mice, intracerebroventricular injection of an adrenergic receptor antagonist inhibited central norepinephrine signaling, leading to an increase in ISF volume and enhanced glymphatic activity (Xie et al. 2013). This finding highlights that norepinephrine is a critical neuromodulator of the glymphatic function, linking arousal states to lymphatic circulation. However, recent research measuring the clearance and movement of fluorescent molecules in the brains of male mice has revealed that brain clearance is significantly reduced during sleep and anesthesia, contradicting previous beliefs (Miao et al. 2024). During non-rapid eye movement (NREM) sleep, a coordinated interplay between electrophysiological oscillations, hemodynamic changes, and CSF flow dynamics has been observed, suggesting that neuronal activity, cerebral blood flow, and CSF circulation are functionally interdependent during sleep (Fultz et al. 2019) (Fig. 2). Studies using transcranial optogenetic stimulation to generate synthetic neural network oscillations demonstrated a significant enhancement of CSF perfusion into the ISF, further supporting the hypothesis that neural activity regulates ISF bulk flow (Jiang-Xie et al. 2024).

Fig. 2.

The illustration highlights cellular and material changes within the brain parenchyma and MLVs in both healthy (left side) and diseased or aged states (right side). Pathological changes in the brain parenchyma include AQP4 depolarization, impaired demyelination, Aβ and tau protein deposition, oxidative stress, vascular leakage, loss of dopaminergic neurons, and release of toxic molecules

Aging is strongly correlated with the accumulation of metabolic waste products in the brain, which may be attributable to age-related impairments in the glymphatic function. Ma et al. labeled CSF in the lateral ventricles with an infrared tracer and measured its accumulation in the cervical lymph nodes over a fixed period. They found that 18-month-old mice exhibited significantly slower CSF outflow kinetics compared to 2-month-old mice (Ma et al. 2017). Furthermore, aged mice display markedly reduced clearance of Aβ injected into the brain parenchyma, confirming age-associated glymphatic dysfunction (Kress et al. 2014). Aging is also accompanied by structural and functional changes in cerebral arterioles, including decreased pulsatility of small intracortical arterioles (Kress et al. 2014) and increased vessel tortuosity, both of which increase the resistance to CSF flow within the PVS (Bennett et al. 2024). Additionally, aging is associated with diminished perivascular AQP4 expression and widespread loss of AQP4 polarization, which further impairs glymphatic clearance (Iliff et al. 2012). In aged mice, ligation of the dCLNs exacerbates severe glymphatic dysfunction, accompanied by increased depolarization of astrocyte-activated AQP4, neuronal damage, induced neuroinflammation, behavioral deficits, and memory impairment (Zhu et al. 2024a, b) (Fig. 2).

Given the central role of glymphatic circulation in waste clearance and brain homeostasis, the modulation of its function has emerged as a promising therapeutic strategy for neurodegenerative diseases and other CNS disorders. In the following sections, we explore current research efforts aimed at targeting the glymphatic system for disease prevention and treatment.

The Role of the Cerebral Lymphatic System in CNS Diseases

Parkinson’s Disease

Parkinson’s disease (PD), the second most prevalent neurodegenerative disorder after Alzheimer’s disease, is pathologically characterized by the early degeneration of dopaminergic neurons in the substantia nigra pars compacta, aberrant aggregation and propagation of α-synuclein (αSyn) and tau proteins, and formation of Lewy bodies (Kalia & Lang 2015). Emerging evidence suggests that the cerebral lymphatic system influences PD progression by modulating the accumulation of pathological proteins and neuroinflammatory responses in the brain.

The glymphatic system facilitates metabolic waste clearance, including misfolded αSyn, via CSF–ISF exchange (Iliff et al. 2012). AQP4, a critical water channel in the glymphatic system, plays a pivotal role in PD pathogenesis (Ding et al. 2021). Reduced expression or depolarization of AQP4 has been observed in both patients with PD and A53T-αSyn-overexpressing mice, impairing glymphatic function and diminishing αSyn clearance (Lapshina & Ekimova 2024). This dysfunction is correlated with αSyn accumulation in the brain parenchyma, which is a hallmark of progressive PD. Notably, AQP4-knockout mice exhibit αSyn aggregation in the dorsal striatum, cerebral cortex, and substantia nigra pars compacta, accompanied by dopaminergic neuronal loss, likely because of impaired macromolecular clearance by the glymphatic system (Lapshina & Ekimova 2024).

The glymphatic system also modulates PD progression through inflammatory pathways. In the 1-methyl-4-phenyl-1, 2, 3, 6-tetrahydropyridine (MPTP)-induced PD model, AQP4-deficient mice exhibited exacerbate neuroinflammation and dopaminergic neuron (DN) loss compared with wild-type mice (Chi et al. 2011) (Fig. 2). Mechanistically, interferon-gamma (IFNγ) interacts with leucine-rich repeat kinase 2 (LRRK2) to induce neuroinflammation, whereas LRRK2 R1441G-mediated AQP4 phosphorylation leads to depolarization, disruption of IFNγ clearance, and exacerbation of neurodegeneration (Huang et al. 2024). CD4⁺ T cells upregulate major histocompatibility complex class II (MHC-II) expression in CNS myeloid cells during αSyn pathology, and genetic ablation of CD4⁺ T cells ameliorates αSyn-induced neurodegeneration (Fig. 3). Notably, AQP4 dysfunction in PD models reduces CD4⁺CD25⁺ regulatory T cells, thereby amplifying microglial inflammation (Jiang et al. 2024). Furthermore, AQP4 single nucleotide polymorphisms (SNPs) rs162009 and rs2075575 are linked to susceptibility to PD and may serve as genetic markers for glymphatic impairment and cognitive decline in patients with PD (Sun et al. 2023; Jiang et al. 2023).

Fig. 3.

The scheme depicts alterations in the MLVs and cervical lymph nodes in healthy states (left side of the illustration) and central nervous system disorders (right side of the illustration). Pathological changes in MLVs are characterized by augmented MHC-II expression, immune cell proliferation, reduced tau and Aβ clearance, and abnormal secretion of immune molecules. In cervical lymph nodes, these changes include lymph node atrophy, lymphocyte depletion, and abnormal antibody secretion

Meningeal lymphatics have a dual effect on PD progression. Impaired meningeal lymphatic drainage exacerbates pathological protein accumulation (Da Mesquita et al. 2018). Fibrillar αSyn activates pro-inflammatory nuclear factor kappa B (NF-κB) signaling via toll-like receptor (TLR)-dependent and -independent mechanisms or binds to receptors for advanced glycation end products (RAGE) on microglia, driving cytokine production (Bearoff et al. 2023). Meningeal lymphatics mitigate the αSyn burden by draining it to the dCLNs, whereas cervical lymphatic ligation promotes αSyn accumulation and DN loss. Additionally, pathological αSyn deposition disrupts meningeal lymphatic function by inducing meningeal inflammation or compromising endothelial tight junctions (Ding et al. 2021; Zou et al. 2019) (Fig. 3). Paradoxically, some clinical studies reported unchanged αSyn levels in the brains of patients with PD despite meningeal lymphatic dysfunction, potentially due to compensatory mechanisms. First, AQP4-dependent perivascular glymphatic pathways may sustain αSyn clearance via CSF–ISF exchange, particularly when AQP4 polarization remains partially intact (Lapshina & Ekimova 2024). Second, meningeal lymphatic dysfunction may enhance microglial phagocytosis or upregulate molecular chaperones (e.g., heat shock proteins) to counteract αSyn aggregation (Da Mesquita et al. 2021). Finally, Genetic heterogeneity, including AQP4 SNPs (rs162009 and rs2075575), may further obscure αSyn pathology due to inter-individual variability in glymphatic efficiency (Jiang et al. 2023; Sun et al. 2023). These observations underscore the need to evaluate meningeal lymphatic contribution to αSyn metabolism within the broader neuroimmune microenvironment (Fig. 3).

Conversely, meningeal lymphatics exacerbate neuroinflammation by facilitating the infiltration of peripheral immune cells (Williams et al. 2021). CD4⁺ T cells migrate along the meningeal lymphatics via the CCR7-CCL21 signaling axis to the dCLNs (Leser et al. 2025), where they acquire CNS-specific phenotypes. Upon reentering the CNS through the blood–brain barrier (BBB) or choroid plexus–CSF pathways (Cervellati et al. 2020; Salvador et al. 2024), these cells secrete pro-inflammatory cytokines (e.g., TNF-α and IL-17), activate microglia, and perpetuate a neurotoxic feedback loop (Fig. 2). This cascade amplifies neuroinflammation, edema, apoptosis, and neuronal death, highlighting the complex interplay between meningeal lymphatic function and immune dysregulation in PD (Da Mesquita et al. 2018).

Multiple Sclerosis

Multiple sclerosis (MS) is a chronic autoimmune disorder characterized by progressive neurological damage. Its pathogenesis involves motor and cognitive impairment resulting from demyelination, neuronal injury, and astrocyte activation within the CNS (Filippi et al. 2018) (Fig. 2). These pathological changes are believed to be driven by a complex interplay between BBB disruption, microglial activation, localized and diffuse inflammation, and the release of pro-inflammatory cytokines (Magliozzi et al. 2006). Emerging evidence suggests that the cerebral lymphatic system influences MS progression by modulating immune cell trafficking, priming inflammatory responses, and altering the function of oligodendrocytes.

Studies have demonstrated glymphatic system dysfunction in patients with MS. In acute MS lesions, retraction of glial cells delineating the PVS occurs, accompanied by astrocyte damage (Brosnan & Raine 2013). Patients with MS exhibited significantly reduced diffusivity along the PVS, as measured by diffusion tensor imaging analysis along the perivascular space (DTI-ALPS), and dilation of the perispinal veins, both of which are indicative of glymphatic system impairment. In addition, a decreased DTI-ALPS score is associated with greater clinical disability and longer disease duration. These findings suggest that glymphatic system dysfunction may contribute to MS progression (Brosnan & Raine 2013; Carotenuto et al. 2022; Bateman & Bateman 2024).

Inflammatory mediators, such as reactive oxygen species (ROS), produced in the meninges of patients with MS, enhance microglial proliferation and activation, leading to the acquisition of an infiltrative phenotype and subsequent CNS damage (Yang et al. 2024; Hsu et al. 2019). Several studies indicated that the glymphatic system plays a crucial role in microglial clearance, implicating it in MS-related neuroinflammation (Yang et al. 2024; Hsu et al. 2019). In secondary progressive MS, elevated levels of immune cells and inflammatory molecules, including tumor necrosis factor (TNF) and interferon (IFN), within the meninges and CSF contribute to microglial activation and demyelination (Gardner et al. 2013). Peripheral immune-mediated destruction of CNS structures is a key pathogenic mechanism underlying autoimmune diseases (Carotenuto et al. 2022). Impaired lymphatic fluid flow and reduced drainage at the regional level result in the accumulation of inflammatory cells and neurotoxic elements, exacerbating cortical demyelination (Carotenuto et al. 2022). Toxic molecules or metabolic byproducts also play crucial roles in MS pathogenesis. Microbial dysbiosis facilitates the infiltration of bacterial toxins into the brain parenchyma. Among these, ε-toxin (ETX), produced by Clostridium perfringens, has been found to be elevated in the brain tissue of patients with MS. ETX binds to the cerebral vasculature and induces oligodendrocyte death (Cekanaviciute et al. 2018; Berer et al. 2017). Lymphatic system dysfunction impairs the clearance of toxic molecules and metabolites from the ventricles and deep gray matter, contributing to gray matter pathology (De Meo et al. 2021) (Fig. 2).

MLVs have a dual effect on MS progression. MLVs exacerbate MS by promoting inflammatory responses. MLVs facilitate the migration of meningeal T cells to dCLNs via the CCR7-CCL21 signaling pathway (Louveau et al. 2018). In an experimental autoimmune encephalomyelitis mouse model, cribriform plate lymphatic vessels developed, whereas classical inflammation-induced lymphangiogenesis was absent in meningeal lymphatics (Hsu et al. 2019; Louveau et al. 2018). Increased T-cell density has been observed within and around the meningeal lymphatic system, accompanied by enhanced drainage of meningeal immune cells to downstream lymph nodes (Hsu et al. 2019; Louveau et al. 2018). MLVs ablation achieved through nasal lymphatic disruption using Visudyne, lymphatic vessel ligation at deep cervical lymph node input, or inhibition of VEGFR3 with MAZ51 attenuates CD4⁺ T cell infiltration, reduces spinal cord demyelination, and subsequently improves experimental autoimmune encephalomyelitis prognosis (Louveau et al. 2018; Hsu et al. 2019). This protective effect may stem from MAZ51-mediated inhibition of lymphangiogenesis in the cribriform plate, thereby limiting the recruitment of pathogenic CNS antigen-specific T cells (Hsu et al. 2019). Furthermore, impaired lymphatic drainage diminishes T cell-antigen-presenting cell (APC) interactions and disrupts T cell phenotype differentiation, leading to altered cytokine and chemokine signaling (Louveau et al. 2018). Although MLVs play a fundamental role in immune cell drainage, their influence on T cell differentiation and immune activation appears to be more pronounced (Fig. 3).

Conversely, MLVs may exert neuroprotective effects in MS by modulating the function of oligodendrocytes and astrocytes. Oligodendrocytes, which are responsible for myelin formation in the CNS, are depleted in MS lesions (Brosnan & Raine 2013). In a well-established cuprizone-induced demyelination model, meningeal lymphatic ablation led to altered glial gene expression, a reduction in mature oligodendrocyte populations, impaired myelin regeneration, extensive axonal loss, and increased apoptosis, all of which had significant implications for demyelinating disorders such as MS (Das Neves et al. 2024). Additionally, MLVs dysfunction has been associated with astrocytic overexpression of Cst3, suggesting increased oxidative phosphorylation stress prior to demyelination, which may contribute to subsequent demyelination (Das Neves et al. 2024) (Fig. 2).

Epilepsy

Epilepsy is one of the most prevalent neurological disorders, exerting a multitude of effects on the cerebral nerves, cognition, and psychology (Fisher et al. 2014). It is characterized by its spontaneity and enduring predisposition to generate epileptic seizures. However, 75% of patients have responded inadequately to treatment (Fisher et al. 2014).

Most patients with epilepsy have significant GS impairment. In children, a significant increase in PVS counts is observed in various forms of epilepsy, such as idiopathic generalized epilepsy, suggesting that impaired GS function leads to slowed clearance of interstitial fluid and fluid retention in the PVS (Liu et al. 2020a, b; Salimeen et al. 2021). Notably, contradictory experimental results regarding PVS changes, ranging from no significant alterations to increased counts, have been reported in patients with febrile seizures and epilepsy (Salimeen et al. 2021; Spalice et al. 2020). Additionally, some studies have revealed that PVS morphology and size in patients with typical Rolandic epilepsy are comparable to those observed in migraine patients, indicating a likely congenital origin rather than epileptogenic changes (Boxerman et al. 2007). In adults, glymphatic malformations and changes in perivascular lymphatic dysfunction appear to be more prominent, with many epilepsy patients exhibiting marked asymmetrical PVS distribution, particularly ipsilateral to the seizure onset zone. This has been observed in focal epilepsy (Feldman et al. 2018), post-TBI epilepsy (Hlauschek et al. 2024), temporal lobe epilepsy (Lee et al. 2022), and frontal lobe epilepsy (Zhou et al. 2024). The DTI-ALPS index was significantly reduced in temporal lobe epilepsy patients with hippocampal sclerosis compared to healthy controls. However, the index showed no significant association with clinical parameters, such as age at epilepsy onset, disease duration, or anti-seizure medication dosage (Lee et al. 2022). Moreover, in mouse models, status epilepticus (SE) was associated with substantial glymphatic impairment, which positively correlated with the degree of post-SE cerebral edema (Liu et al. 2021a, b). Alleviation of cerebral edema was shown to facilitate earlier restoration of glymphatic function following SE (Liu et al. 2021a, b).

The mechanisms by which glymphatic system dysfunction affects epilepsy include impaired solute clearance, which leads to ion homeostasis disruption, accumulation of toxic substances, and cerebral edema (Fig. 2). During epileptic seizures, the reduced excretion of K⁺ and Ca⁺ ions in the ISF affects the electrophysiological activity of the neurons. AQP4 interacts with molecules such as glutamate transporter 1 to modulate the CSF–ISF osmotic pressure and ion exchange. Dysregulation and mislocalization of AQP4 in patients with epilepsy affect ion exchange, thereby exacerbating epileptic symptoms (Rabinovitch et al. 2019). The deposition of phosphorylated tau and serum proteins leads to axonal damage and neuronal hyperexcitability (Noé & Marchi 2019). Glutamate serves as a precursor for the reduction of inhibitory neurotransmitter synthesis and is thus closely linked to the onset of epilepsy (Hlauschek et al. 2024). Glymphatic system-induced phosphorylation impairment of tau and glutamate deposition also increases neuronal excitability (Hlauschek et al. 2024). Post-epileptic cerebral edema, accompanied by glymphatic system dysfunction and enhanced AQP4 polarization, can be alleviated by glibenclamide treatment or Trpm4 knockout. This intervention significantly facilitates early recovery of impaired glymphatic function and ISF drainage following SE, as evidenced by reduced PT deposition, decreased neuronal degeneration, and improved cognitive function (Liu et al. 2021a, b). Notably, among patients with focal epilepsy, those with poor responses to anti-seizure medications (ASMs) exhibited more severe glymphatic system dysfunction than good responders, suggesting that the glymphatic system may enhance ASM responsiveness and contribute to treatment outcomes in patients with focal epilepsy (Kim et al. 2024).

On the other hand, MLVs play a significant role in regulating immune cells. Leukocyte infiltration in the PVS induces inflammation in the central nervous system after seizures, leading to neuronal damage. Inhibition of the inflammatory receptor CCR2 alleviates the deleterious effects of status epilepticus (Alemán-Ruiz et al. 2023). Impaired MLVs drainage allows brain-derived antigens to reach the periphery, activating cytotoxic CD8⁺ T cells to mediate neuroinflammation, thereby causing neuronal damage (Noé & Marchi 2019). CD8⁺ T-cell-driven limbic encephalitis leads to temporal lobe epilepsy, suggesting that MLVs may play a crucial role in epilepsy development via CD8⁺ T cell regulation (Pitsch et al. 2021).

Major Depressive Disorder

Major depressive disorder (MDD) is a globally prevalent mood disorder characterized by cognitive and behavioral impairments, anhedonia, and suicidal ideation (Monroe & Harkness 2022). It manifests as a wide range of functional disturbances, including dysregulation of sleep, cognition, appetite, and mood, making it one of the most pressing mental health challenges (Monroe & Harkness 2022).

Sleep disturbances are a core symptom of depression (Stickley et al. 2019), and their alleviation is closely associated with the remission of depressive symptoms (Fang et al. 2019). The glymphatic system, which is primarily active during sleep (Xie et al. 2013), has been implicated in the causal relationship with MDD pathophysiology (Nedergaard & Goldman 2020; Zhang et al. 2024a, b). Global blood oxygen level-dependent (gBOLD) signals derived from resting-state functional magnetic resonance imaging (rs-fMRI) provided a noninvasive method for quantifying glymphatic dynamics (Liu et al. 2018). Compared to healthy controls, patients with MDD exhibited weaker gBOLD–CSF coupling, which may reflect glymphatic dysfunction (Zhang et al. 2024a, b).

Astrocytes, a key component of the glymphatic system, are reduced in both number and activity in MDD (Wallensten et al. 2021). The glymphatic system influences the pathogenesis of MDD through mechanisms involving oxidative stress, monoaminergic dysregulation, and neuroinflammation (Fig. 2). The glymphatic system facilitates the clearance of reactive oxygen species (ROS) into the peripheral lymphatic system, which serves as the primary pathway for ROS elimination (Harrison et al. 2020). Given that the brain accounts for 20% of total oxygen consumption, it is particularly vulnerable to ROS-induced damage (Bhatt et al. 2020). In MDD, increased BBB permeability exacerbates neuroinflammation by enhancing the leakage of astrocyte-derived extracellular vesicles (Wallensten et al. 2021). Impaired glymphatic clearance of ROS can activate microglia, leading to the production of pro-inflammatory cytokines that induce apoptosis, thereby contributing to the pathophysiology of MDD (Roomruangwong et al. 2017). While low ROS levels promote neuronal growth, excessive oxidative stress triggers neuroinflammatory damage, apoptosis, and ultimately, MDD progression (Bhatt et al. 2020). Chronic unpredictable mild stress (CUMS)-induced glymphatic dysfunction may act as a bridge between monoamine dysregulation and neuroinflammation, thereby playing a central role in the pathogenesis of depression (Roomruangwong et al. 2018). The glymphatic function is closely linked to monoamine neurotransmitter levels (Liu et al. 2020a, b). During sleep, reductions in monoamine concentrations enhance glymphatic clearance, whereas during wakefulness, increased monoamine release, particularly norepinephrine (NE), inhibits glymphatic function (Liu et al. 2020a, b). Recent studies have shown that CUMS induces excessive NE release, which subsequently disrupts AQP4 polarization in astrocytes, thereby impairing glymphatic function (Yao et al. 2023b, a). This impairment contributes to oxidative stress and neuroinflammation, ultimately leading to the development of depression-like symptoms in affected individuals.

Sex-dependent differences in stress sensitivity are associated with meningeal lymphatic dysfunction in mice. Studies have demonstrated that subchronic variable stress selectively impairs meningeal lymphatic function in female mice, whereas enhancement of meningeal lymphatic function reduces stress susceptibility and mitigates stress-induced alterations in the medial prefrontal cortex (mPFC) and ventral tegmental area (VTA; Dai et al. 2022). Notably, both the mPFC and VTA are critical brain regions involved in emotion regulation (Tye et al. 2013). These findings suggest that meningeal lymphatics play a pivotal role in the sex-dependent modulation of stress susceptibility and resilience.

Cerebral Small Vessel Disease

Diseases such as aging, hypertension, and diabetes influence the pathogenesis of cerebral small vessel disease (CSVD) via the cerebral lymphatic system. Aging is considered a key factor in the regulation of cerebral lymphatic function (Tian et al. 2022). Age-related dysfunction in the glymphatic system and MLVs exacerbates cognitive decline and promotes neurodegenerative pathologies, further contributing to CSVD development (Tian et al. 2022). Type 2 diabetes mellitus, a prevalent metabolic disorder in middle-aged and elderly populations, is a significant risk factor for CSVD (Staals et al. 2014). Compared to non-diabetic rats, rats with type 2 diabetes mellitus exhibit a threefold slower clearance of CSF tracers, as demonstrated by Gd-DTPA contrast-enhanced imaging and fluorescent tracer studies. Impairment of the glymphatic system-induced inhibition of hippocampal and hypothalamic ISF clearance is directly associated with cognitive deficits (Jiang et al. 2017). Clustering analyses further revealed that type 2 diabetes mellitus significantly increased the perivascular space, likely due to diabetes-induced microvascular and macrovascular damage, which in turn heightened CSVD susceptibility (Jiang et al. 2017; Doubal et al. 2010) (Fig. 2).

The formation of cerebral edema in CSVD is influenced by the function of the glymphatic system. Stroke is the leading cause of long-term disability and the second leading cause of mortality worldwide (Steinmetz et al. 2024). Ischemic stroke (IS) accounts for 63% of all stroke subtypes globally and is the primary contributor to non-fatal and fatal burdens in adults (Steinmetz et al. 2024). IS is frequently accompanied by cerebral edema, which exerts mechanical stress on brain tissues and capillaries, exacerbating damage (Simard et al. 2009). Recent studies have shown that the glymphatic system provides a pathway for lymphatic exchange between the CSF and mesenchymal cells, facilitating the rapid influx of CSF into brain tissue and serving as the earliest source of Na⁺ and fluid in cerebral edema (Mestre et al. 2020). Moreover, increased PVS flow is observed during stroke, potentially due to the release of K⁺ and other vasoactive molecules, along with endothelial nitric oxide depletion, leading to significant contraction of parenchymal and pial arterioles and the generation of a pressure gradient. This mechanism may play a crucial role in the development of cerebral edema (Mestre et al. 2020). AQP4 exacerbates post-ischemic cerebral edema by promoting CSF transport to the brain parenchyma. AQP4-knockout mice (Mestre et al. 2020) or treatment with AQP4 inhibitors, such as TGN-020 (Hirt et al. 2017; Sun et al. 2022), significantly reduced CSF tracer influx into the PVS and ISF within 15 min of ischemic onset, thereby mitigating cerebral edema, reducing mortality, and restoring motor function (Mestre et al. 2020; Hirt et al. 2017; Sun et al. 2022). However, some studies found that IS also induced glymphatic system dysfunction. Three hours post-IS onset, reduced CSF inflow and impaired clearance of low-molecular-weight compounds were linked to decreased arterial pulsatility and PVS compression, resulting in glymphatic system impairment (Gaberel et al. 2014). Notably, glymphatic system function was restored 24 h post-IS onset, coinciding with the reperfusion of the middle cerebral artery (Gaberel et al. 2014). Additionally, AQP4 reactivity and depolarization increased at 3 and 14 days after diffuse microinfarcts, disrupting cerebral water homeostasis and impairing edema clearance (Wang et al. 2012). Thus, although glymphatic system dysfunction may limit the extent of AQP4-mediated cerebral edema, its role in stroke-related pathophysiology remains complex.

Glymphatic system impairment also occurs in the acute phase following subarachnoid hemorrhage (SAH), leading to the accumulation of metabolic waste and neurotoxic factors (De Rooij et al. 2013). Fibrinolytic plasminogen activators enhance lymphatic function by dissolving blood clots in the PVS, but their effect on prognosis remains unclear (Gaberel et al. 2014; Goulay et al. 2017). Interestingly, AQP4-knockout models demonstrated increased cerebral edema and neuroinflammation (Liu et al. 2021a, b). However, conflicting findings have been reported regarding the role of AQP4 in SAH-induced cerebral edema. AQP4 expression was significantly elevated in the posterior cerebral arteries following SAH, whereas venous AQP4 levels remain unchanged, potentially due to astrocyte activation (Liu et al. 2021a, b; Chen et al. 2016). Higher AQP4 expression levels in patients with SAH were correlated with more severe cerebral edema (Chen et al. 2016). Notably, atorvastatin-mediated AQP4 inhibition reduced the severity of cerebral edema, Caspase-3 expression, and mortality in SAH models, suggesting potential therapeutic benefits in early brain injury following SAH (Chen et al. 2016). These findings indicate that AQP4’s involvement in SAH-associated cerebral edema is multifaceted and necessitates further investigation.

MLVs play a pivotal role in various CSVDs through immune regulation and drainage. In IS, MLVs mediate immune responses in the nervous system, promoting neuroinflammation by activating macrophages via the VEGFC/VEGFR3 signaling pathway, ultimately contributing to cerebral infarction (Esposito et al. 2019). Additionally, thrombospondin-1 (THBS1) is significantly elevated in the CSF of patients, and THBS1-CD47 ligand–receptor interactions were shown to induce apoptosis in meningeal lymphatic endothelial cells, leading to MLVs dysfunction and poor prognosis in SAH (Wang et al. 2023). In intracerebral hemorrhage (ICH), increased meningeal lymphangiogenesis is observed in the later stages, persisting for up to 60 days. Impaired MLVs function hampers hematoma clearance from the brain parenchyma, whereas MLVs enhancement improves neurological outcomes and accelerates hematoma resolution (Tsai et al. 2022). Furthermore, cilostazol-induced enhancement of meningeal lymphatic drainage following ICH promoted erythrocyte clearance and facilitated hematoma regression (Kimura et al. 2014).

Overall, the interplay between glymphatic system dysfunction and CSVD pathogenesis remains incompletely understood, with variability in experimental timelines, therapeutic interventions, and individual AQP4 responses potentially contributing to the conflicting findings (Hirt et al. 2017; Wang et al. 2012; Gaberel et al. 2014; De Rooij et al. 2013; Chen et al. 2016). Conversely, MLVs consistently exhibit varying degrees of dysfunction, while their enhancement is generally associated with improved CSVD outcomes (Esposito et al. 2019; Wang et al. 2023).

Traumatic Brain Injury

Traumatic brain injury (TBI) is a form of organic brain tissue damage caused by external collisions, blunt trauma, or penetrating injuries (Bolte et al. 2020). Enhancing basic and clinical research on TBI and improving therapeutic interventions are crucial for reducing the high mortality and disability rates associated with severe TBI (Bolte et al. 2020).

Meningeal lymphatic dysfunction is closely related to the development of TBI, and the two influence each other. Intracranial pressure rises significantly within 2 h of TBI onset, leading to dysfunction in key CSF absorption sites within the subarachnoid space, suggesting meningeal lymphatic impairment (Bolte et al. 2020). This dysfunction persists for at least one month after injury, exacerbating neuroinflammation and cognitive deficits (Willis et al. 2020). TBI induces both morphological and functional changes in MLVs, including increased expression of lymphatic vessel endothelial hyaluronan receptor 1 (Lyve-1), a significant increase in the number of endothelial loops and buds within the lymphatic system, and an overall trend toward increased lymphatic vessel diameter (Bolte et al. 2020). Impaired lymphatic drainage following TBI results in the accumulation of Aβ, cellular debris, and other neurotoxic substances, perpetuating immune activation in post-injury brain tissue (Wu et al. 2024). This sustained neuroinflammatory response contributes to the prolonged pathology observed after TBI. Furthermore, pre-existing meningeal lymphatic dysfunction exacerbates neuroinflammation and cognitive impairment following trauma (Bolte et al. 2020). For instance, compared with mice with TBI alone, those with pre-existing meningeal lymphatic injury exhibited significant upregulation of complement-related gene expression, increased glial fibrillary acidic protein immunoreactivity, and elevated ionized calcium-binding adapter molecule 1 immunoreactivity within 24 h after TBI, indicating exacerbation of glial cell damage (Bolte et al. 2020). These findings suggest that promoting meningeal lymphatic recovery after TBI may be an effective strategy for mitigating neuroinflammation and improving neurological outcomes in TBI patients.

The glymphatic system is closely associated with both the initial and secondary pathophysiological processes of TBI. TBI disrupts the function of key regulatory structures, such as the hypothalamus, pineal gland, and brainstem, which impairs glymphatic function by affecting sleep regulation (Piantino et al. 2022). Additionally, post-TBI elevation of blood fibrinogen levels allows fibrinogen to cross the BBB, where it can be deposited and converted into fibrin at the vascular–astrocyte end-foot interface. This process disrupts the paravascular pathways of the glymphatic system, leading to the continuous accumulation of metabolic waste products, ultimately contributing to neurodegeneration (Sulimai & Lominadze 2020). Neurodegenerative features commonly observed in patients with TBI include neurofibrillary tangles composed of tau protein aggregates and Aβ amyloid plaques (Iwata et al. 2002; Zanier et al. 2018) (Fig. 2). Glymphatic circulation was significantly impaired following TBI, with reductions detectable as early as one day post-injury and persisting for at least 28 days post-injury. In severe cases, CSF clearance can decrease by up to 60% (Jessen et al. 2015). Notably, in a moderate TBI animal model, tau protein aggregates were found within the perivenous interstitial space of large veins, aligned with the established paravascular pathways of the glymphatic system. These findings suggest that tau accumulation is closely linked to glymphatic clearance deficits (Jessen et al. 2015). Similarly, studies in veterans with blast-related mild TBI (mTBI) have identified an increased burden of MRI-visible perivascular spaces in the frontal cortex, which is identified as a potential neuroimaging marker of perilymphatic dysfunction. A greater PVS volume was correlated with poorer cognitive performance (Braun et al. 2024; Jung et al. 2024). These findings further support the notion that glymphatic dysfunction, which leads to impaired tau clearance, plays a crucial role in secondary neuronal injury and neurofibrillary tangle formation following TBI.

Cerebral edema is a critical sequela of TBI, and glymphatic dysfunction is implicated as a key contributor to early stage cytotoxic edema (Eide & Ringstad 2024). TBI-induced sympathetic activation and increased noradrenergic tone are identified as major factors disrupting glymphatic function (Hussain et al. 2023). Driven by CSF influx and osmotic pressure gradients, outflow pathways from the venous interstitial space are inhibited by osmotic forces that lead to ISF retention and the development of brain edema (Wu et al. 2021).

Modulating the Cerebral Lymphatic System for Neurological Disorders

Direct Targeting of AQP4

Specifically, targeting AQP4 offers a direct approach to modulating the glymphatic system. In AQP4 antibody-positive patients, immunosuppression has reduced the risk of relapse (Paul et al. 2023). The PPA regimen (prazosin, propranolol, and atipamezole) may alleviate post-traumatic brain edema by blocking adrenergic receptors and enhancing lymphatic drainage (Hussain et al. 2023). Furthermore, AQP4 neutralizing antibodies have been shown to significantly reduce cerebral edema within 24 h and mitigate hippocampal neuronal loss and memory deficits by day 21 (Li et al. 2023). Moreover, AQP4 localization at the blood–spinal cord barrier has been shown to be regulated by calmodulin, which binds to its C-terminal domain during hypoxia-induced swelling, exacerbating CNS edema (Kitchen. et al. 2020; Salman et al. 2017). Inhibiting calmodulin or using AQP4 inhibitors (e.g., TGN-020) has been demonstrated to effectively reduce edema (Vandebroek and Yasui 2020; Sun et al. 2022). Trifluoperazine, a calmodulin inhibitor, has been shown to prevent AQP4 localization to the barrier, reduce edema, and promote functional recovery (Xing et al. 2023).

Indirect Modulation of Glymphatic System

Astrocytic modulation influences the glymphatic system. Knockdown of the transient receptor potential melastatin 4 (TRPM4) has been shown to reduce astrocytosis and microglial proliferation post-status epilepticus, alleviating cerebral edema and neuronal damage (Rolin et al. 2024; Wang et al. 2025a, b). As a result, targeted inhibition of astrocyte proliferation has been shown to increase AQP4 expression and reduce inflammation (Nakano-Kobayashi et al. 2023). Antisense oligonucleotides (ASOs) are promising for the treatment for Huntington’s disease (HD), where HTT gene dysfunction disrupts astrocyte and glymphatic system functions via NF-κB signaling (Dhuri et al. 2020; Träger et al. 2014). Intravenous mesenchymal stem cell (MSC) therapy has been shown to restore glymphatic system function by enhancing ASO distribution in BACHD mice while suppressing mutant HTT (mHTT; Wang et al. 2021; Yen et al. 2020). Consequently, omega-3 polyunsaturated fatty acids (n-3 PUFAs) have been shown to inhibit astrocyte activation, while preserving AQP4 polarization, improving glymphatic system function, and alleviating depressive and cognitive symptoms (Ren et al. 2017). Interstitial space expansion during slow-wave sleep has been demonstrated to correlate with δ-wave activity (Xie et al. 2013). Similarly, transcranial direct current stimulation (tDCS) has been shown to enhance δ-wave activity through the IP3/Ca²⁺ pathway, increasing the CSF–ISF exchange (Smith et al. 2020; Wang et al. 2025a, b).

Regulation via Proteins and Cytokines

Specifically, Per2 expression is inversely correlated with AQP4 polarization. Melatonin maintains AQP4 polarity via circadian regulation (Yao et al. 2023; Cruz-Sanabria et al. 2023). AQP4 polarity also depends on the dystrophin complex (Si et al. 2024), which is degraded by matrix metalloproteinase-9 (MMP-9; Huang 2018). MMP-9 inhibition (GM6001) has been demonstrated to restore perivascular AQP4 and β-dystroglycan levels (Yuan et al. 2024; Zhu et al. 2024a, b). Notably, ultrasound therapy, including continuous theta burst stimulation (cTBS), has been shown to enhance the glymphatic system by upregulating VEGFC, which promotes MLVs dilation and drainage (Yulug et al. 2017; Liu et al. 2017; Song et al. 2020). Similarly, 40 Hz light flicker, acting via the adenosine-A2AR pathway, has been demonstrated to improve lymphatic inflow and AQP4 distribution (Sun et al. 2024; Li et al. 2020; Poderoso et al. 2019).

Modulating MLVs in Neurological Disorders

VEGFC is critical to MLVs development and has therapeutic potential for neurological disorders (Boisserand et al. 2024; Xu et al. 2023). VEGFC-VEGFR3 signaling has been demonstrated to enhance MLVs function, alleviating cognitive impairment, neuroinflammation, and neuronal damage in sepsis-associated encephalopathy (Deng et al. 2015; Leppänen et al. 2013). Impaired MLVs drainage worsens microglial activation in Alzheimer’s disease, thereby reducing the efficacy of immunotherapy. However, VEGFC delivery has been shown to improve Aβ clearance, suggesting that MLV-targeted therapy can enhance AD immunotherapy (Da Mesquita et al. 2021). More importantly, BO nanomicelles (BO-MSs) have been demonstrated to improve MLV permeability and enhance Aβ₄₂ oligomer clearance within minutes (Wu et al. 2023; Ye et al. 2024). Similarly, near-infrared light has been demonstrated to enhance the metabolic homeostasis of mitochondria in meningeal lymphatic endothelial cells, promote cell adhesion and growth, improve the connectivity of meningeal lymphatic endothelial cells, and favor lymphatic transport (Foo et al. 2020; Wang et al. 2024). In a mouse model of chronically elevated intracranial pressure, elevated intracranial pressure has been demonstrated to limit CSF acquisition by MLVs and affect MLV development, maintenance, and functional drainage, whereas stimulation of Piezo1 ion channels in mice enhances MLV coverage, drainage, and cerebral CSF perfusion (Coste et al. 2010; Li et al. 2022; Warren et al. 2024). Some drugs ultimately act on both the glymphatic system and MLVs. In 5XFAD mice (a model of early dementia), focused ultrasound combined with microbubbles (FUS-MB) inhibited microglia and astrocyte proliferation or activation and increased lymphoid clearance of Aβ through meningeal arterioles and veins to the CSF space, and into the dCLN via meningeal lymphatic vessels. This treatment significantly improved working memory, showing potential for preventing cognitive impairment (Daini et al. 2021; Lee et al. 2020; Yang et al. 2023).

Controversies and Discussion

While GS primarily suppresses neuroinflammation in PD (Chi et al. 2011; Huang et al. 2024; Jiang et al. 2024; Sun et al. 2023), MLVs exacerbate neuroinflammation in PD and MS (Leser et al. 2025; Cervellati et al. 2020; Salvador et al. 2024; Da Mesquita et al. 2018; Louveau et al. 2018; Hsu et al. 2019). On the other hand, some studies have found that damage to MLVs in PD patients exacerbates pathological protein deposition (Da Mesquita et al. 2018; Bearoff et al. 2023; Ding et al. 2021; Zou et al. 2019). However, a subset of these studies reported that αSyn levels remained unchanged (Lapshina & Ekimova 2024; Jiang et al. 2023; Sun et al. 2023).

Conflicting findings regarding epilepsy are primarily observed in pediatric patients. Studies have indicated that the number of PVS associated with different types of epilepsy may increase, decrease, or remain relatively unchanged (Liu et al. 2020a, b; Salimeen et al. 2021; Spalice et al. 2020; Boxerman et al. 2007). Many of these studies relied solely on MRI scans to demonstrate alterations in PVS (Liu et al. 2020a, b, 2021a; Spalice et al. 2020; Boxerman et al. 2007; Feldman et al. 2018; Hlauschek et al. 2024). Disruption of GS structures may lead to compensatory restoration of lymphatic drainage through mechanisms such as MLVs, which could explain the divergence from mainstream findings in studies on childhood epilepsy (Spalice et al. 2020). Furthermore, most of these studies had sample sizes of approximately 60 participants (Liu et al. 2020a, b, 2021a; Salimeen et al. 2021; Feldman et al. 2018; Hlauschek et al. 2024). In contrast, in the study in which contradictory results emerged, the sample size was only 30 cases (Spalice et al. 2020). Therefore, a more comprehensive retrospective study is needed to validate these findings.

In CSVD, GS may exacerbate brain edema after IS. However, other studies suggest that severe GS dysfunction associated with IS may instead impair water clearance. This apparent contradiction may be attributable to temporal variations in cerebral lymphatic system function. Early activity of the cerebral lymphatic system primarily facilitates the development of brain edema (Mestre et al. 2020; Hirt et al. 2017; Sun et al. 2022), whereas subsequent dysfunction later in life exacerbates impaired water clearance and further contributes to neurological impairment (Doubal et al. 2010; Wang et al. 2012; Gaberel et al. 2014). Further experimental studies are required to validate this hypothesis. Additionally, various experimental models were employed in these studies, including the Middle Cerebral Artery Occlusion Model (MCAO; Mestre et al. 2020; Hirt et al. 2017; Sun et al. 2022), minimally invasive injection of gadolinium chelate into the cisterna magna (Gaberel et al. 2014), acute stroke patients (Doubal et al. 2010), a mouse model of multiple diffuse microinfarcts induced by unilateral internal carotid artery injection of cholesterol crystals (Wang et al. 2012), among others.

Research on AQP4 in SAH has yielded mixed results. Inhibition of AQP4 with atorvastatin in a rabbit model ameliorated early brain damage after SAH (Chen et al. 2016). AQP4 knockdown in a rat model exacerbated brain edema and neurological deficits (Liu et al. 2021a, b). In contrast, measurement of lymphatic system perfusion using MRI after minimally invasive injection of gadolinium chelates into the large pool using a mouse model indicated that AQP4 may not be directly associated with brain edema or increased intracranial pressure (Gaberel et al. 2014). We hypothesize that this may be related to animal model reactivity, experimental methods for AQP4 inhibition, and methods to detect AQP4.

Based on the above studies, we propose the following hypotheses. First, in most diseases, MLVs and GS function interdependently, i.e., they jointly mediate anti-inflammatory and drainage functions. However, in certain diseases, the two systems may exhibit opposing roles (e.g., PD, MS). We hypothesize that the MLVs and GS may have compensatory roles for each other and cannot be studied in complete isolation. For instance, findings show that impairment of AQP4 in the aging brain is accompanied by increased AQP4 expression, which has been viewed as a physiological compensation, and it is possible that a similar compensatory mechanism may apply to MLVs damage (Lapshina & Ekimova 2024; Jiang et al. 2023; Sun et al. 2023). Other factors such as genetic heterogeneity, including AQP4 SNPs (rs162009 and rs2075575) (Jiang et al. 2023; Sun et al. 2023), enhanced phagocytosis by microglia, or upregulation of molecular chaperones (e.g., heat shock proteins) (Da Mesquita et al. 2021) may compensate for impaired MLV-mediated clearance via GS mechanisms. Most of the above studies target GS or MLVs individually. However, given their possible interdependence, the respective roles of GS and MLVs in drainage and neuroinflammation should not be evaluated in isolation. Therefore, researchers should simultaneously investigate the inflammatory and clearance-related roles of both GS and MLVs in the same model to better elucidate their complex mechanisms of action within the CNS. Second, differences in experimental methods may cause errors in the results. In the above studies, the main methods for observing the lymphatic system were MRI and DTI-ALPS. The exclusive use of MRI can reflect structural changes and cannot directly assess GS dysfunction (Liu et al. 2020a, b, 2021a; Boxerman et al. 2007; Spalice et al. 2020; Feldman et al. 2018; Hlauschek et al. 2024), which in turn contributes to errors. Furthermore, potential compensatory interactions between MLVs and GS may further confound the findings. Therefore, we advocate the combined use of MRI and DTI-ALPS to demonstrate the structural and functional aspects of GS changes and their impact on disease. Third, modeling methods and animal models are different. These studies applied different modeling methods, animal models, or patient samples, and, as a result, their findings were inevitably inconsistent, considering that different species may exhibit varied physiological responses (Mestre et al. 2020; Hirt et al. 2017; Sun et al. 2022; Doubal et al. 2010; Wang et al. 2012; Chen et al. 2016; Liu et al. 2021a, b; Gaberel et al. 2014). Among them, in the experiments conducted using drugs, unknown drug effects may have confounded experimental outcomes (Huang et al. 2024; Liu et al. 2021a, b; Kim et al. 2024; Hirt et al. 2017; Sun et al. 2022; Gaberel et al. 2014; Goulay et al. 2017; Chen et al. 2016). Additionally, other sources of error include small sample sizes, lack of patient follow-up, and variability in observation time points, all of which contribute to inconsistencies.

Conclusion and Outlook

The cerebral lymphatic system, comprising the MLVs and the glymphatic system, plays a crucial role in metabolic waste clearance and the regulation of neuroinflammation. It has become a major focus in neurological disease research. In this review, we summarize recent physiological and functional findings regarding the cerebral lymphatic system, including factors influencing the glymphatic system, pathological alterations, and the pathogenic mechanisms of various neurological disorders. We also discuss the therapeutic targets and approaches aimed at modulating this system, along with conflicting findings across studies.

Many central nervous system (CNS) disorders are closely associated with cerebral lymphatic dysfunction. The cerebral lymphatic system consistently supports recovery in MDD and TBI, with the glymphatic system aiding in the clearance of reactive oxygen species (ROS) in MDD and tau protein clearance and edema resolution in TBI. No significant contradictions were observed between these conditions. However, research in PD, MS, CSVD, and epilepsy has yielded contradictory findings and unexplained discrepancies under different conditions. We analyze contradictory studies on these diseases and find that the reasons for these discrepancies include studying MLVs and GS in isolation while ignoring the compensatory interactions between them; using different research models and observational methods within the same disease; small sample sizes; and lack of patient follow-up. Addressing these uncertainties is critical for optimizing CSF–ISF exchange and developing precise clinical guidelines for therapeutic interventions. These findings also underscore the complex and multifactorial nature of CNS pathogenesis, emphasizing the need for evidence-based research to elucidate disease mechanisms. Despite these challenges, there is no doubt that meningeal lymphatic vessels and glymphatic system represent promising therapeutic targets, offering potential breakthroughs in treating currently incurable neurodegenerative diseases.

Several cerebral lymphatic system-targeted therapies have been developed in recent years, with promising results. However, disease-specific therapeutic strategies, particularly for MS and TBI, remain limited. Notably, we found that the cerebral lymphatic system shares common pathogenic mechanisms with different neurological diseases, suggesting that similar experimental models and therapeutic targets may be applicable to multiple conditions. In addition, research into the glymphatic system and MLVs has progressed unevenly across different disorders. For example, studies on MLVs involvement in epilepsy are scarce compared to those on the glymphatic system. Future investigations should explore the involvement of MLVs in epilepsy by leveraging insights from lymphatic research on other diseases. This review not only synthesizes recent advances in the structure, function, and regulatory factors of the cerebral lymphatic system but also highlights its involvement in major CNS disorders. These findings provide key insights for the development of more effective therapeutic strategies, ultimately paving the way for improved clinical outcomes in neurological diseases.

Abbreviations

CNS

Central nervous system

AQP4

Aquaporin-4

CSF

Cerebrospinal fluid

ISF

Interstitial fluid

PVS

Perivascular space

BBB

Blood–brain barrier

MLVs

Meningeal lymphatic vessels

VEGFR3

Vascular endothelial growth factor receptor 3

dCLNs

The deep cervical lymph nodes

Aβ

Amyloid-beta

IFN

Interferon

TNF

Tumor necrosis factor

ROS

Reactive oxygen species

DTI-ALPS

Diffusion tensor imaging analysis along the perivascular space

αSyn

α-Synuclein

THBS1

Increased levels of thrombospondin 1

VEGFC

Vascular endothelial growth factor C

MS

Multiple sclerosis

PD

Parkinson's disease

CSVD

Cerebral small vessel disease

MRI

Magnetic resonance imaging

IS

Ischemic stroke

SAH

Subarachnoid haemorrhage

MDD

Major Depressive Disorder

TBI

Traumatic brain injury

Author Contributions

JXP and YQF: conceived, drafted, edited, rigorously reviewed the manuscript and drafted illustrations; PY: critically revised and integrated the final manuscript; WFL: reviewed and edited the manuscript; ZFL: contributed significantly to the revision; AS: Supervised and gave valuable input to the manuscript; JSD: reviewed and revised the manuscript; FM: gave valuable input to the manuscript; FL: responsible for revision of illustrations; STQ: Supervised and gave valuable input to the manuscript; YB: Supervised and gave valuable input to the manuscript.

Funding

This work was supported by the Clinical Research Program of Nanfang Hospital, Southern Medical University (Grant Numbers 2023CR023, 2018CR029, 2019CR020); and the Guangdong Basic and Applied Basic Research Foundation (Grant Number 2024A1515030069).

Data Availability

No datasets were generated or analysed during the current study.

Declarations

Conflict of interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Ethical Approval

Not applicable.

Illustration Attribution

All illustrations presented in this manuscript were originally created by the authors using Adobe Illustrator 2015 (version 19.0.0). All graphical elements (including schematics, cellular diagrams) were developed de novo based on theoretical frameworks in this study.