Slow Blood Vessel Rhythms Controlled by Noradrenaline: A Hidden Fluid Pump Connecting Sleep, Brain Waste Removal, and Dementia Development

The vasomotor clearing system was developed based on a neuro-modulatory circuit that initially controlled vigilance and now has tight regulation of cerebral circulation. The LC is a small nucleus located in the pons that projects to many areas of the forebrain. It is active to varying degrees based on the state of arousal: when an organism is awake, LC neurons produce tonic activity at 1–5 Hz to maintain arousal; when the organism is asleep, particularly in deep non-rapid eye movement (NREM) sleep, LC neurons become inactive due to intrinsic membrane properties, GABAergic input from the preoptic region, glutamatergic input from the nucleus paragigantocellularis, and electrotonic coupling between LC neurons to produce infra-slow (0.01–0.1 Hz) oscillations that correspond to the slowest frequency bands of the brain during sleep (cerebral delta waves and astrocyte Ca²⁺ waves) that provide a temporal basis for state dependent transport of fluids [22,23].

LC activity influences cerebral blood flow and is primarily achieved through volume transmission. Neurotransmitters such as norepinephrine are released from varicosities of the LC and diffuse over tens of micrometers to influence the smooth muscle cells of blood vessels (VSMCs), astrocytes, pericytes and other neurons [24]. Volume transmission of neurotransmitter activity is a relatively slow process and is well suited to the mechanical characteristics of blood vessels, which tend to be most responsive to low frequency stimulation [25].

LC activity influences the activity of both alpha-1 (α₁)- and beta (β)-adrenergic receptors on VSMCs. Activation of α₁-receptors results in the activation of phospholipase C (PLC) and subsequent increase in intracellular calcium (Ca²⁺) resulting in contraction of VSMCs [26]. Low frequency (infralow) LC activity can result in slow vasomotion (oscillations of 2–5% of the diameter of the blood vessel at 0.01–0.1 Hz). Although the amplitude of the oscillations is relatively small, they have significant effects on the resistance of the blood vessel due to the relationship of resistance and radius of the blood vessel (resistance ∝ R⁴, where R is the radius of the blood vessel) [27]. The activity of β-receptors on VSMCs and astrocytes, however, will decrease the amplitude and possibly the phase of vasomotor cycles [28].

Astrocytes are responsible for translating neuro-modulatory signals into both structural and hydraulic changes. Astrocytes express both α₁- and β-receptors and respond to NE with transient increases in intracellular Ca²⁺ that will alter the shape of the astrocyte and end-foot, and the distribution of AQP4 channels [29]. These alterations in the geometry and water permeability of perivascular structures will directly affect hydraulic resistance of the perivascular space. Pericytes in the capillary bed, although slower to react than astrocytes, do contract in response to NE and distribute microvascular resistance, affecting flow to different regions of the brain [30]. Together, the LC, astrocytes, and pericytes form a distributed adrenergic network that controls the vasomotor activity of blood vessels throughout the brain. The effectiveness of the adrenergic network depends on the temporal relationships between LC oscillations, cortical–thalamic oscillations, and the mechanical responses of the blood vessels and the astrocytes. During N2-N3 sleep, LC infralow activity is temporally aligned with slow waves in the cortex, spindles of the thalamus-cortex, and ripples in the hippocampus, while the blood vessels undergo rhythmic dilation, the astrocytes produce waves of Ca²⁺, and the neurons stop firing [31,32]. This hierarchical temporal structure provides a framework for the brain to maximize clearance of waste under conditions of reduced energy use and increased need for clearance of waste [33]. This system is part of larger feedback systems. Impairment of LC function due to accumulated toxins and inflammatory mediators can lead to impaired function of the adrenergic system and reduced efficiency of clearance [34]. Reduced efficiency of clearance leads to reduced levels of LC-derived norepinephrine, which, in turn, reduces the ability of the brain to clear waste [33].

Slow traveling contraction waves are generated in the blood vessel wall by synchronized Ca²⁺ transients produced in VSMCs by periodic activation of α₁-adrenergic receptors [35]. At the same time, astrocytes modify the geometry of perivascular spaces and the distribution of AQP4 channels, thus regulating hydraulic conductivity. The combination of the oscillations of the blood vessel and glial elements generates coherent low-frequency pressure waves that propagate along the blood vessel and facilitate convective transport of fluid from superficial blood vessels to the central regions of the brain [36].

Thus, LC activity functions not only as a trigger for vasomotor waves but also as a regulator of the sensitivity of astrocytes to adrenergic stimuli and the polarity of AQP4 channels, thus continuously adjusting the efficiency of the clearance pathway [37,38].

Blood vessels transform neuro-modulatory signals into mechanical forces that drive perivascular flow. Pial and penetrating arterioles with rich walls of VSMCs are the major effectors. Slow oscillations of the diameter of blood vessels induced by low frequency NE production drive pressure waves that propagate along the blood vessel wall [39]. These waves cause displacement of CSF into the PVS and mixing of CSF with interstitial fluid (ISF); together they create convective currents of fluid through the brain [40]. Studies modeling the blood vessels suggest that optimal frequencies for producing net convective flow are in the range of 0.01–0.1 Hz [41]. PVS’s formed around penetrating arterioles small (10–30 µm), highly compliant tubes bounded by the vascular basement membrane and the end-feet of astrocytes are ideal for converting sustained pressure waves into coherent flow [41]. The interaction between pressure waves caused by vasomotion and local mechanical impedance (i.e., the product of vessel stiffness and the extracellular matrix) can enhance the transfer of fluid from the periarterial to the perivenous regions of the brain [42].

The geometry of PVSs (i.e., branching, tapering, and interconnecting) further modulates flow patterns. There are several oscillatory drivers of flow: cardiac pulses (5–10 Hz) predominate near the proximal ends of the arteries and quickly die away with distance; respiratory rhythms (0.2–0.3 Hz) modulate the pressure inside the skull and can synchronize slower waveforms; slow vasomotion (0.01–0.1 Hz) is a low frequency, high impact driver of flow that can extend flow over great distances [11,43,44]. Depending on phase, cardiac and respiratory components can either augment or dampen the gradients of flow caused by vasomotion. The interaction between cardiac/respiratory components and vasomotion can result in a multiscale, tunable clearance system that can adapt to changing physiological needs [45]. Astrocytes play a critical role in regulating the hydrodynamic environment of the blood vessels. Astrocytic end-feet that are rich in AQP4 channel proteins link the pressure in the PVSs to flow in the interstitium. Proper polarization of AQP4 channels to the end-feet of the astrocytes lowers resistance to flow and enhances convective transport of fluid from the blood vessels to the interstitium. However, depolarization of AQP4 channels in astrocytes in disease or aging can disrupt proper alignment of the channels and increase resistance to flow [46]. Astrocytes can dynamically adjust the morphology of their end-feet to modulate the geometry of the perivascular space and flow through the space, depending on the concentration of Ca²⁺ in the cell [47]. Astrocytes can also modulate the resistance of capillary beds to flow by contracting in response to adrenergic stimulation and redistributing flux to different parts of the brain. Venules, although poorly muscularized, can also participate in the generation of flow by passively oscillating in diameter at phases related to the waveforms generated by vasomotion, facilitating the exit of ISF from the brain. Optimal coordination of arteriovenous flow is necessary for effective clearance of waste; failure to achieve coordination can significantly reduce clearance [48].

Dysfunction of this axis can be a key underlying mechanism for various disease processes. Loss of LC neurons can lead to decreased oscillatory NE output, decreased vasomotor responsiveness to NE, increased stiffness of the blood vessel wall leading to decreased amplitude of vasomotion, depolarization of AQP4 channels leading to increased resistance to flow, and disrupted sleep–wake cycles disrupting the timing of vasomotion [49]. Together, these changes can lead to impaired perivascular flow, accumulation of toxic solutes, and enhanced neuroinflammation seen in aging, neurodegenerative disorders, and traumatic brain injury [50]. In order that the functional coherence of the system should be made clear, Table 1 is designed to indicate the major components of the noradrenergic vasomotion axis and their coordinated roles in effecting the cerebrovascular and perivascular fluid dynamics. It is intended to synthesize the neural, vascular and glial layers in a coherent mechanistic explanation, showing how the rhythms of norepinephrine derived from the LC are manifested in terms of oscillatory vasomotion, convective flow and adaptive clearing of solutes.

Direct perturbation of LC activity is direct proof that noradrenergic vasomotion causes perivascular transport. The optogenetic stimulation of LC neurons at sleep-like frequencies (~0.05 Hz), which are low enough to be termed “infralow”, induces simultaneous oscillations in arteriole diameter and significantly increases tracer uptake into the tissue; but tonic or high-frequency stimulation does not. This clearly demonstrates that it is the temporal pattern of neuronal activity, and not the frequency of the neuronal firing, that is causal [58,59].

Similarly, chemo-genetic activation also increases vasomotion and transport, and chemo-genetic inhibition suppresses both by >50%. Both of these effects are maintained even after removal of the sympathetic innervation of the vessels, and without an increase in systemic pressure, demonstrating that the central nervous system controls vascular tone via noradrenaline [60,61,62]. Pharmacological studies have shown that alpha-1 adrenergic receptor signaling is required for vasomotion and perivascular transport. Alpha-1 blockade abolishes vasomotion and perivascular flow even though the LC is still firing; beta-blocking has little effect [63,64].

Activation of the LC in mice with reporter genes for amyloid-beta and tau accelerate their clearance; inhibition of the LC delays clearance, with clearance being proportional to the amplitude of the vasomotion [65].

Clearance peaks during NREM slow waves, showing that the physiological state affects the coherence of the pump [66,67].

Measurements of perivascular pressure demonstrate that LC activity generates oscillatory pressure gradients of several mmHg that are phase-locked to vasomotion and will produce convective flow. Models of the biophysics of perivascular flow, that include the amplitude–frequency parameters measured in vivo, accurately model in vivo tracer kinetics. Models that assume only diffusion cannot accurately model the in vivo tracer kinetics, establishing that the LC-driven vasomotion is sufficient to explain brain-wide clearance [68].

Optical imaging with two-photon microscopy shows that low-frequency (0.01–0.1 Hz) oscillations in arteriole diameter consistently precede the movement of tracer molecules along the periarteriolar space by several seconds, demonstrating the temporal precedence of vasomotion relative to the movement of tracer molecules [69]. Similar dynamic contrast MRI studies show that whole-brain CSF influx oscillate at the same periodicities and peak during NREM slow-wave activity. Blocking the alpha-1 adrenergic receptors abolishes vasomotion and reduces CSF influx; stimulating the LC or administering alpha-1 agonists restores both [70,71,72]. Sedatives–hypnotics (zolpidem and propofol) decrease LC activity and abolish infralow vasomotion, although cortical slow-waves remain intact. As a result of the abolished vasomotion, the decreased LC activity eliminates the influx of tracer molecules into the perivascular space by ~60%, and eliminates the clearance of metabolites from the interstitial space, indicating that sleep EEG is not the driving force behind the clearance of solutes from the brain, but rather the adrenergic rhythm is [73]. Re-establishing alpha-1 tone restores vasomotion and clearance of solutes from the interstitial space, without affecting sleep stages, thereby establishing that the adrenergic tone is the primary modifiable variable controlling clearance [74].

Biophysical studies of structural perturbations have demonstrated that the efficiency of the pump is dependent on the integrity of the substrate. Decreasing vascular compliance by arterial stiffening results in decreased amplitude of vasomotion and decreased influx of tracers into the perivascular space, even with preserved LC activity. Increasing hydraulic resistance by reducing AQP4 polarity or function in astrocytes reduces perivascular flow; reversing AQP4 polarity or function restores perivascular flow [75,76]. Therefore, vasomotion requires coherent coordination between LC activity, vascular compliance, and astrocyte polarity. Models of the biophysics of perivascular flow that incorporate vessel elasticity, PVS geometry, and fluid viscosity accurately model the convective velocities measured in vivo when vasomotion is modeled. Impairment of any one parameter predicts deficits in the age-related and disease-related clearance [77]. Stimulating sensory input at low frequency can entrain cortical slow-wave activity, enhance vasomotion, and increase influx of tracer molecules into the perivascular space, demonstrating that the pump can be externally entrained [78,79]. Studies using two-photon microscopy and MRI resolved different scales of resolution: microscopy revealed localized micro-meter scale oscillations in arteriole diameter; MRI resolved whole-brain convective exchange. Collectively, they showed that glymphatic transport arises from a frequency-layered system in which cardiac pulsation generates localized mixing within the micro-vessels, respiratory-rhythms modulate intracranial pressure, and low-frequency vasomotion provides the persistent convective-bias [80,81,82,83].

Studies of disease models indicate that disruption of vasomotion precedes pathological development. In models of Alzheimer’s disease, degeneration of the LC occurs early, accompanied by decreased vasomotor amplitude and decreased influx of tracer molecules into the perivascular space; restoring either LC activity or alpha-1 tone restores clearance of solutes and delays pathological development [84,85]. Traumatic brain injury disrupts rhythmic activity of the LC and glymphatic flow; restoration of rhythmic activity of the LC restores glymphatic flow and promotes recovery [71].

Vascular stiffening, chronic sleep loss, inflammatory responses, and metabolic stress all converge on the same phenotype: decreased amplitude of vasomotion → impaired clearance of solutes [86,87,88]. Restoration of any one of these components (LC output, vascular compliance, AQP4 polarity) restores clearance [89].

Although human studies are indirect, they provide strong evidence for conservation of this system. Blood-volume and flow were imaged using functional MRI to show that infra-slow rhythms in blood volume and flow occur with sleep stage; phase-contrast MRI was used to show that CSF oscillations at 0.01–0.1 Hz occur with sleep stage and peak during NREM; neuromelanin-sensitive MRI was used to show that degeneration of the LC correlates with cognitive decline and impaired dynamics of perivascular space [90,91].

This system is very plastic. Enhancement of LC activity, enhancement of alpha-1 signaling, restoration of AQP4 polarity, or improvement of vascular compliance all enhance clearance, whereas suppression of LC activity, blockade of adrenergic receptors, or stiffening of arteries impede clearance [92]. Interventions of sensory, behavioral, or electrophysiological nature that enhance slow-wave or infralow rhythmic activity can modulate vasomotion non-invasively [93].

In addition to influencing clearance of solutes, vasomotion also influences other functions, such as perivascular immune trafficking, BBB permeability, and regulation of extracellular vesicles, placing vasomotion as a key regulator of neurovascular-immune signaling [69,94,95].

A hierarchical order of evidence now exists:

Collectively, these studies establish that brain clearance is an actively regulated, state-dependent neurovascular process, not a passive byproduct of circulation or sleep. Failure of this system promotes neurodegeneration, and modifiability of this system presents a potentially powerful therapeutic strategy [43]. An outline of this process is suggested in Figure 1.

While preclinical data have provided compelling evidence for a role of the LC in driving vasomotor mechanisms, human validation is currently limited. The majority of evidence has been derived from optogenetic, chemo-genetic, and tracer-based studies with high levels of causal resolution, but relatively low levels of translational relevance [99,100]. Therefore, near-term progress toward the development of a predictive clinical physiology model of perivascular transport will depend upon the establishment of quantitative correlations between noradrenergic rhythmicity and measurable hemodynamic or perivascular parameters in humans. Potential methods to achieve this goal include using neuromelanin-sensitive magnetic resonance imaging (MRI) to evaluate the structural integrity of the LC, dynamic contrast-enhanced MRI to measure glymphatic flux, and diffusion-based measures of perivascular anisotropy. While indirect, these measures may provide estimates of threshold values—vascular elasticity, astrocytic permeability, oscillation coherence—whose deterioration can indicate transition from reversible to irreversible dysfunction. In order to transform the model from a conceptual framework into a predictive clinical physiology model, it will be necessary to integrate multiple variables. A significant number of translational gaps exist between rodent models of LC function and human physiology despite methodological convergence across optogenetic and chemo-genetic systems [101,102]. These differences arise due to differences in the anatomy of the LC between rodents (≈150,000 neurons) and primates (≈400,000 neurons), the length of vascular pathways, and the frequency of intrinsic oscillations (≈0.01 Hz in primates vs. 0.05–0.20 Hz in rodents), which affect the relative contributions of neural drive, vascular compliance, and hydrodynamic forces [103,104]. Additionally, many studies using optical and chemo-genetic methods use anesthesia (isoflurane, urethane) during experimentation, which depresses noradrenergic tone, increases BBB permeability, and affects vascular elasticity—factors that are known to alter vasomotor amplitude and coherence [105].

As such, human evidence remains primarily correlative and is generated by slow-fMRI, phase-contrast MRI, and neuromelanin-sensitive imaging studies demonstrating spatiotemporal vasomotor coherence, but lacking the ability to isolate the contribution of noradrenergic mechanisms [106]. Non-invasive neuromodulation methods, including focused ultrasound and transcutaneous vagal stimulation, may offer the potential to modulate LC tone indirectly, but the physiological specificity and long-term safety of these methods remain unclear [107]. As such, translational progress will be dependent upon the development of cross-species computational models that incorporate multimodal imaging (PET, fMRI, elastography) and longitudinal biomarkers (plasma NE metabolites, soluble AQP4 fragments) to determine if the quantitative relationships demonstrated in animal models are preserved in humans [108,109]. Table 2 illustrates the major experimental methodologies used to establish noradrenergic vasomotion as a primary driver of perivascular transport.

LC is the first and most susceptible pillar of the vasomotor pump to degrade, often decades prior to clinical symptom development. Because LC has fewer than 50,000 neurons and a high metabolic demand due to its extensive arborization, it is particularly sensitive to mitochondrial dysfunction and impaired mitophagy as well as oxidative stress [96]. Systemic inflammatory mediators such as IL-6 and TNF-α and breakdown of the blood–brain barrier stimulate the activation of pontine microglia and exacerbate LC neurodegeneration. Selective early loss of LC neurons occurs in those neurons that project to forebrain vasculature and astro-glial territories and thus decreases the noradrenergic drive responsible for coordinating clearance [110,111].

With the decrease in the number of LC neurons, there is a reduction in the precision of infralow oscillations. The remaining LC neurons begin to fire abnormally due to the disruption of their hyperpolarization-activated cyclic nucleotide-gated channels, reduced T-type calcium channel currents and altered GABAergic modulation, which disrupts the hierarchical relationship between LC output, cortical slow wave and astrocytic calcium rhythm and destroys the temporal framework necessary for clearance to occur efficiently. Lower levels of alpha1-adrenergic receptor stimulation lead to decreased vasomotor gain, whereas lower beta-adrenergic receptor stimulation leads to the loss of stabilizing feedback mechanisms for oscillatory amplitude and phase [112,113,114].

Declining levels of noradrenergic stimulation propagate dysfunction throughout the glial–vascular unit. Astrocytes lose the adrenergic-evoked cytoskeletal dynamics and AQP4 clustering; pericytes demonstrate blunted contractile responses leading to reduced capillary level resistance tuning; and microglia enter a chronic state of activation leading to the release of cytokines that are toxic to LC neurons, creating a self-amplifying loop of degeneration [115]. Sleep becomes increasingly fragmented; slow-wave activity decreases; and the coherence of vasomotion during NREM sleep deteriorates. Factors in the environment impact upon this vulnerable node; chronic sleep loss impacts the inhibitory tone; hypertension and metabolic syndrome negatively impact pontine perfusion; and mild traumatic brain injury (mTBI) results in persistent microglial activation. Regardless of the condition causing it, the consistent finding is that reduced LC inflow will predict the failure of perivascular transport and neurodegenerative progression [116,117,118].

Regardless of the integrity of neuro-modulatory inputs, failure of vasomotion occurs if the vascular network cannot translate rhythmic input into mechanical movement. Chronic aging, hypertension and metabolic disorder result in the stiffening of the cerebrovascular network through fragmentation of the internal elastic lamina, loss of elastin and collagen cross-linking [119]. VSMCs transition from a contractile to a synthetic phenotype characterized by a reduction in the expression of alpha-actin and diminished Ca2+ dynamics, resulting in a decreased magnitude and fidelity of diameter oscillations. These structural changes result in reduced sensitivity to adrenergic transduction, resulting in turn in a narrow, unresponsive dynamic range. Oxidative stress, advanced glycosylation end products, and increased oxygen tension contribute to the remodeling of the extracellular matrix and calcification of the vascular wall, further decreasing vascular compliance [120]. Imaging using two-photon microscopy demonstrated that the vasomotor amplitude of penetrating arterioles in aged mice was less than 30% of the vasomotor amplitude seen in young adult mice, which correlated with the decreased perivascular flow and tracer penetration [69]. Decreased compliance results in a disruption of the frequency-layered relationship between vasomotion and cardiac pulsatility and respiratory oscillations. As vessels become stiffened, the low-frequency components of vasomotion are lost; the high-frequency components of cardiac waves become mechanically decoupled; and the multiscale pressure gradients required for convective transport are collapsed [121]. Arterio-venous phase coherence is lost, and the pressure differentials across the parenchyma flatten, leading to the necessity for the system to rely on diffusion for transport, which is inadequate for clearing amyloid-β, tau, α-synuclein, or inflammatory mediators. Remodeling of the microvasculature adds additional impedance to transport. Chronic hypertension causes arteriolar rarefaction, and increases the tortuosity, formation of micro-aneurysms and basement membrane thickening—all of which reduce the volume of the perivascular space, increase resistance and alter the distribution of microvascular load. Dropout of pericytes also disrupts the stability of capillary tone and the hydraulic architecture downstream [122,123,124].

Regardless of whether the LC output is maintained, the mechanical degradation significantly impairs the transmission of vasomotor waves [125].

Vascular stiffening is a feed-forward mechanism that promotes additional damage to the neuro-modulatory system. Stiffer vessels transmit greater pressure shocks to perivascular cells and neurons, contributing to the breakdown of the blood–brain barrier, inflammation of the brain and progressive damage to the LC [126,127].

Therefore, there is a self-reinforcing loop where the decrease in LC tone impairs mechanical transduction, and the stiffening of the vasculature continues to erode neuromodulation, ultimately leading to the degradation of the clearance pump [128,129].

Finally, if the astro-glial interface is damaged regardless of the preservation of neuro-modulatory input and mechanical transduction, then the exchange of CSF–ISF will be impaired. With age, and with the occurrence of disease, the morphological and functional polarity, morphology and signaling of astrocytes is destroyed, converting the perivascular interstitial gateway into a high resistance bottle-neck [130].

The primary defect lies in the depolarization of AQP4. Astrocytic reactivity, cytokines produced by microglia and destruction of the dystrophin-dystroglycan complex results in the displacement of AQP4 from the astrocytic end-feet into non-perivascular membranes, greatly increasing hydraulic resistance [131,132]. The degree of depolarization is sufficient to produce a decrease in the rate of tracer penetration of >50%, regardless of the presence of intact vasomotion [133].

Astrocytic Ca²⁺ signaling is also compromised. There is a reduction in the expression of IP₃ receptors and diminished store-operated Ca²⁺ entry, both of which limit the extent of end-foot remodeling and the response of astrocytes to rhythmic neuro-modulatory inputs, thereby limiting the ability to dynamically adjust perivascular resistance during oscillatory flow [134,135].

Activation of microglia and the infiltration of other immune cells contribute to the worsening of the impairment. Activated microglia and infiltrating immune cells release metalloproteases and reactive oxygen species that alter the stiffness and topography of the basement membrane, redirecting flow pathways, creating zones of stasis and enhancing solute retention [136,137]. Activated perivascular macrophages enhance the inflammatory loops and distort the localization of AQP4 and glial function [138].

Impairment of glial function ripples throughout neural homeostasis. Impaired glymphatic flow reduces the delivery of ions, neuromodulators and extracellular vesicles; synaptic homeostasis is impaired; antigen clearance is impaired; and neuroinflammation becomes self-sustaining [11,139].

In virtually all neurodegenerative diseases, a similar triad is present: decreased LC drive, vascular stiffening and astrocytic depolarization shifts the system from an active convective pump to a diffusion limited architecture incapable of clearing amyloid-β, tau, α-synuclein, or inflammatory mediators [140,141,142]. Traumatic brain injury follows the same path, initially involving acute AQP4 depolarization and basement membrane remodeling followed by chronic LC dysfunction and vascular stiffening [143]. However, significant aspects of the pump can be reversed. Restoration of LC tone (chemo-genetic or pharmacologic) can restore vasomotion even in aged mice. Pharmacological or genetic intervention to target vascular stiffness (e.g., NO donors, crosslink breakers or TGF-β pathway modulators) can improve vascular compliance and mechanical transduction. Repair of dystrophin-complex defects or reprogramming of astrocytes to restore AQP4 polarity can restore perivascular water flux. Early multimodal interventions (combining neuro-modulatory, vascular and astro-glial repair) have shown synergistic effects and suggest that restoring coherence within the system may prevent or partially reverse disease progression [144,145]. Table 3 intends to outline the step-wise collapse and emerging reversibility of the pump at the neuron, vessel and glia interfaces.

As described above, vasomotor collapse is an end point for a variety of diseases; however, the initial pathophysiology differs among them. For example, degeneration of the LC occurs prior to significant amyloid deposition within the cortex in patients with Alzheimer’s disease, causing early disruption of adrenergic rhythmicity and resulting in both acute vasomotor dysfunction and disruption of aquaporin-4 function [70]. Similarly, α-synuclein pathology within the locus coeruleus and pons disrupts noradrenergic signaling, although later than that observed in Alzheimer’s disease, in patients with Parkinson’s disease [153]. In contrast, in cerebral small vessel disease, vascular injury (such as arteriolosclerosis, endothelial inflammation, or decreased vascular compliance) progressively disrupts the neural modulation of vascular response by disconnecting neural input from the mechanical transduction of the vascular response. Across all these conditions, chronic sleep fragmentation, hypertension, systemic inflammation, and metabolic syndrome are common factors that promote the progressive loss of the same axis of function: noradrenergic function, vascular stiffness, and astro-glial polarization, although each condition has its own unique temporal and spatial entry points [154].

Similarly, the loss of slow vasomotor control also contributes to a broad spectrum of neurovascular disorders through its role in disrupting cerebrovascular autoregulation [155]. Disrupted cerebrovascular autoregulation results in compromised blood flow regulation and leads to increased risk of stroke, cerebral vasculopathy, post ischemic edema and breakdown of the BBB, which result from impaired neurovascular coupling, microvascular hypoxia, and perivascular congestion without structural occlusion [156,157,158]. Therefore, vascular stiffness, endothelial dysfunction, and impaired waste clearance are united by their disruption of a single pathway of vasomotor–adrenergic function, thus establishing a direct link between neurovascular disease and neurodegenerative disease mechanisms [159].

The central driver is the LC. The most upstream and often the most transformative way to reactivate the pump is to restore LC rhythmicity. There are two primary adrenergic strategies for enhancing vascular responsiveness to LC output: (1) alpha-1 agonists increase vasomotor amplitude, and (2) beta-modulators increase oscillatory coherence between vascular responses to LC output. Delivery systems have been developed to deliver therapeutic agents directly to the spinal cord via the use of nanotechnology, thereby limiting side effects from systemic administration [165,166,167]. Chemo-genetic stimulation has demonstrated the ability to reinstate infra-slow oscillations and to improve clearance in both aged and diseased model systems [59]. Optogenetic entrainment of LC activity has been shown to enhance phase coherence and tracer transport when stimulated in synchrony with slow cortical waves. Although non-invasive stimulation techniques have yet to demonstrate similar efficacy to chemo-genetic and optogenetic stimulation, they represent a potential avenue for stimulation of LC activity in a clinically relevant manner [168].

Preservation and/or replacement of LC neurons represents another approach to maintain stable output. Antioxidants, mitophagy enhancers, and anti-inflammatory agents have been used to prolong the life span of LC neurons, whereas stem-cell derived noradrenergic progenitor cells have been shown to partially restore norepinephrine output in preclinical models [169,170]. Modulation of brain state has also been shown to support the maintenance of LC rhythmicity. For example, closed-loop sensory stimulation, synchronized to cortical slow waves, has been shown to strengthen endogenous oscillations, and pharmacologically inducing a deeper level of NREM sleep (e.g., GABA_B agonist, orexin antagonist) has been shown to enhance clearance conditions [171].

Gene-based therapies, including CRISPR modification of mitochondrial regulators and pacemaker channels (i.e., HCN, T-type Ca²⁺), represent an evolving new generation of therapies aimed at maintaining LC rhythmicity [172].

Since vessels convert LC rhythmicity into mechanical force, restoring vascular compliance is critical to the functioning of the pump. Most therapeutic strategies seek to reverse ECM-induced stiffening of vessels using lysyl oxidase inhibitors, matrix metalloproteinase (MMP) modulators, and advanced glycation end-product breakers in metabolic diseases [173,174]. Increasing the bioavailability of nitric oxide (NO) via NO donors or phosphodiesterase inhibitors increases vessel dilation and long-term elasticity and expands the dynamic range of vasomotor responses and wave propagation [175].

Furthermore, reintroducing a contractile VSMC phenotype (via modulation of myocardin, serum response factor, Notch, or PI3K-Akt pathways) has been shown to restore Ca²⁺ sensitivity and oscillatory fidelity to vessels. In addition, epigenetic interventions that reverse senescent chromatin states have been shown to further enhance contractile behavior [176]. Mechanical conditioning (e.g., sustained blood-pressure reduction or pulsed-flow conditioning) of vessels synchronizes vascular mechanics with cardiac and respiratory rhythms and restores vasomotor propagation in experimental models [177].

Advanced nanoparticle delivery systems enable selective remodeling of cerebral vessels. Elastin-stabilizing peptide or ECM-modifying enzyme nanoparticles delivered to cerebral vessels enable the localized restoration of compliance [178]. Increased elasticity of cerebral vessels reduces pathological pulse transmission, protects micro-vessels and perivascular glia, and restores the multiscale pressure gradients necessary for convective transport [179]. Clinical trials examining AGE-breakers, endothelial gene therapy, and NO-based interventions are underway and provide evidence for the translational feasibility of vascular rejuvenation. Emerging non-invasive conditioning paradigms (intermittent hypoxia, visceral pressure stimulation) are showing promise in enhancing glymphatic function [180].

The astro-glial end-foot is the last barrier to be crossed for CSF–ISF exchange; failure of this interface can prevent clearance, regardless of whether LC rhythmicity and vascular mechanics are otherwise normal. Restoration of this interface begins with repolarization of AQP4. This can be achieved by pharmacologically enhancing the expression of dystrophin and syntrophin, and by remodeling ECM anchors such as laminin and agrin [181]. Other anti-inflammatory strategies include suppressing IL-1β and TNF-α signaling, reducing astrogliosis, and stabilizing end-foot architecture. Additional therapeutic avenues exist through genetic and cell-based methods. Viral vectors that enhance AQP4 trafficking, CRISPR tools that modify AQP4 localization, and astrocyte reprogramming via STAT3, NF-κB, or Sox9 pathways facilitate restoration of perivascular polarity in models of glymphatic dysfunction [182,183].

Additionally, transplantation of glial progenitors capable of expressing polarized AQP4 is being investigated as a means to repair perivascular organization. Restoration of dynamic astrocytic Ca²⁺ signaling (via modulation of IP₃ receptors or store-operated Ca²⁺ entry) enables real-time adjustments to perivascular resistance in response to neuro-modulatory rhythmicities [184]. Remodeling of the perivascular ECM (using enzymatic degradation of pathological collagen or hyaluronan, or through the use of biomimetic scaffolds) decreases hydraulic resistance and reopens narrow spaces. Protection of the basement membrane against damage caused by microglial NADPH oxidase or MMPs is also critical to maintaining the structural integrity of the astro-glial interface over time [185].

In addition to supporting solute clearance, repair of the astro-glial interface supports synaptic homeostasis, neuromodulator distribution, extracellular vesicle transport, and immune surveillance. Anti-inflammatory biologics that target IL-1β and TNF-α have been shown to normalize AQP4 polarity and glymphatic flow in aged models and combination astro-glial–vascular–neuro-modulatory therapies have been shown to exhibit additive improvements in clearance [186,187]. Collectively, these results suggest a unifying therapeutic approach, in which restoration of LC rhythmicity, vascular mechanics, and astro-glial polarity converge to restore the brain’s intrinsic vasomotor flow. Figure 2 illustrates how each of these interventions contribute to an integrated recovery process.

Even though the therapeutic methods presented earlier provide a logical structure for recovering vasomotor regulation, most of these treatments are still at an early stage of translational development. Those that have the greatest potential for rapid development, based upon the existing use of clinical agents such as adrenergic receptor modulators, nitric oxide-based vasodilators, phosphodiesterase inhibitors and anti-inflammatory biologics, are those whose safety and systemic effects have been previously documented [188]. On the other hand, more complex approaches, for example, gene editing methods, nanoparticle-based targeting systems and cell replacement for the LC or astro-glial cells, need to undergo a significant amount of validation in terms of long-term safety and specificity, as well as immunological tolerance [189].

There is a unique risk for each therapeutic strategy: stimulating LC activity can negatively impact cardiovascular stability or sleep architecture; augmenting vascular compliance can disrupt autoregulation in sensitive blood vessels; and therapies targeted towards the astrocytes carry a risk of non-specific genomic or immune related side effects [175]. While non-invasive neuro-modulatory methods such as focused ultrasound and vagal nerve stimulation are encouraging, they also require further definition of safe dose levels and longer term safety information [190].

The reasonable pathway for translating these findings into humans will occur in stages. First, standardization of imaging and fluid biomarker assessments of LC health, vascular compliance and astro-glial polarity will need to be established to serve as surrogates for assessing the functionality of the pump [191]. Next, early phase studies utilizing already approved agents will be required to determine if small changes in neuro-modulatory, vascular and glial tone result in detectable increases in perivascular flow or sleep-dependent clearance. Only after safety and biological effectiveness are demonstrated should more invasive approaches such as gene editing, closed-loop neurostimulation, or cell-based repair be advanced to human trials. Ultimately, identifying which component of the clearance network is failing (i.e., neuro-modulatory, vascular, glial, or a combination) in individual patients will be critical to developing lower risk therapeutic approaches that target only the patient’s predominant failure mechanism [161].

Clearance is a regulated, state dependent process mediated by LC controlled vasomotion; it is not simply a passive hydraulic process. Noradrenergic rhythm, vascular biomechanical, astrocyte polarity, and perivascular geometry are part of a single tunable clearance system [194,195].

LC infra-slow oscillations (0.01–0.1 Hz,) which coordinate with cortical slow waves and astrocyte calcium oscillations, modulate vasomotion and AQP4 clustering via α1- and β-adrenergic pathways [196,197]. Studies have shown that longitudinal changes in LC degeneration, vascular stiffening and AQP4 depolarization occur prior to accumulation of amyloid, tau, and α-synuclein and therefore indicate that clearance failure is an upstream cause of neurodegenerative pathology [198,199]. These three factors define Alzheimer’s, Parkinson’s, small vessel disease, TBI, and post-infection syndromes [200].

Therefore, in order to intervene early in patients, future research needs to develop methods to quantify phase dynamics, nonlinear feedbacks and tipping points between neuro-modulatory, vascular and glial subsystems using multiscale modeling [201,202,203].

Translation of the NVP requires development of tools to assess and manipulate pump function in humans [70,204].

Ultra-slow fMRI can detect sub-0.1 Hz vasomotor oscillations; 4D phase contrast MRI can resolve perivascular flow; and elastography can determine arterial compliance and therefore permit multi parametric pump mapping when used in conjunction with ASL, VASO, and neuromelanin sensitive MRI [205,206,207]. PET ligands for adrenergic receptors and astro-glial markers as well as biochemical indicators (NE metabolites and soluble AQP4) provide dynamic biomarkers [101,208].

Focused ultrasound to the LC, vagus nerve stimulation and closed loop phase locked stimulation synchronized to slow waves represent examples of neuro-modulatory interventions to activate the pump. Therapies targeted at the vascular compartment (AGE breakers, lysyl oxidase inhibitors, PDE inhibitors, and eNOS activators) aim to restore compliance and wave propagation. These are complemented by low frequency oscillatory conditioning and nanoparticle mediated ECM remodeling [209,210,211,212].

Interventions directed at the astro-glial compartment (antisense dystrophin correction, CRISPR based AQP4 relocalization, IL-1β/TNF-α anti-biologics and STAT3/Sox9 reprogramming) restore perivascular polarity and water flux [213,214]. The combined use of biomarkers will allow for early identification, stratification, and real-time monitoring of the effectiveness of therapy [215,216].

In addition, machine learning models that incorporate cognitive markers, sex-specific predictors, amnesia profiles, and LC-personality interactions will further refine individualized risk and resilience assessments [217,218,219,220].

The NVP provides a new perspective on neurodegenerative diseases as reversible systems-level failures. Combination therapies that target the LC rhythmicity (α1/β2 agonists and NE reuptake modulators), vascular mechanics (AGE breakers, nitroxyl donors), and astro-glial polarity (AQP4 traffickers, anti-inflammatory agents) can potentially restore pump function synergistically [169,221,222].

Gene editing strategies that increase the resilience of the LC or reduce VSMC senescence can also be incorporated into neuro-modulatory devices. Closed loop systems that can sense LC activity or vasomotor oscillations and deliver adaptive stimulation represent a potential next step in maintaining clearance continuously [223]. Identifying early signs of LC, vascular or astro-glial decline allows for the implementation of prevention strategies (sleep optimization, vascular maintenance, intermittent neuro-modulatory stimulation) decades before the first cognitive symptoms occur [224,225].

Classifying neurological diseases according to failure of clearance networks (neuro-modulatory, vascular, glial or mixed) may provide a common classification for all neurodegenerative, autoimmune, post-infectious and cognitive disorders [226,227].

Restoration of perivascular flow may extend the cognitive lifespan of individuals by reducing metabolic waste, inflammation and network fragility [47].

The various mechanisms provided by clearance to couple brain function to the removal of metabolic waste products provide several ways in which the brain’s metabolic waste removal affects its function. First, clearance helps to maintain an appropriate concentration of ions (K⁺ and Ca²⁺) in the extracellular environment to keep neurons excited and promote coherent oscillations in the network necessary for memory formation [55]. Second, clearance clears away the metabolic byproducts produced by astrocytes, including lactate and reactive oxygen species, providing mitochondria with the resources they need to sustain the high energy levels required for long term potentiation [230]. Third, clearance limits the amount of excess neurotransmitters and inflammatory mediators that accumulate in the synaptic and peri-synaptic space, preventing the synapse from becoming saturated and reducing the noise-to-signal ratio necessary to transmit information accurately. Therefore, clearance becomes essential for maintaining network stability and is important for the fidelity of computations performed by neuronal circuits [231].

Clearance of synaptic by-products from the brain produced by waking behavior is one of the clearest examples of how clearance relates to cognition. Ongoing neuronal activity will continually produce extracellular metabolites, reactive oxygen species, and misfolded proteins. These by-products, if cleared inefficiently, will negatively impact the ability of receptors to traffic properly, alter ion homeostasis, and affect the physical and chemical composition of the extracellular matrix [232,233]. Thus, the slow perivascular convective flow caused by vasomotion promoted by neuromodulator-dependent changes in blood vessel diameter will remove these by-products and maintain the local ionic environment required for proper synaptic transmission. However, failure to clear these by-products efficiently will lead to problems in network organization, leading to problems in synchronization, excitability, and signal-to-noise ratios that do not require significant structural damage and manifest clinically as initial difficulties in attention, working memory, and cognitive flexibility [234].

Clearance processes also interact with synaptic scaling and plasticity. The spatio-temporal distribution and rate of turnover of matrix metalloproteinases, neurotrophic factors, and extracellular vesicles carrying microRNAs and synaptic regulatory proteins are regulated by perivascular flow [235]. These regulators control the reformation of dendritic spines, the strength of long-term potentiation, and synaptic pruning. Therefore, the vasomotor clearance of waste products exerts effects on synaptic reorganization that are similar to the effects of synaptic activity itself. Therefore, alterations in perivascular flow can alter the threshold of plasticity and reduce learning efficiency, establishing a link between clearance kinetics and the time courses of learning and memory acquisition throughout the life span [236].

AQP4 dependent water transport and astrocyte derived calcium signals provide additional connections between clearance processes and neural activity. AQP4 polarization establishes a gradient in perivascular pressure that controls the fraction of the volume of the extracellular space available for the diffusion of neurotransmitters, the extent of ephaptic coupling, and the degree of cross talk between synapses [54]. Additionally, noradrenergic input to astrocytes regulates the geometry of the perivascular space, thereby regulating the hydraulic resistance and creating optimal conditions for synaptic signaling. Together, these results show that glia–perivascular interaction is an active player in the processing of cognitive information, rather than a mere passive supporter [130].

These results have recently been expanded to include oscillatory network dynamics, particularly those that relate to attention and memory. Oscillations in vasomotor pressure in the infra-slow range (0.01–0.1 Hz) are synchronized with infra-slow fluctuations in cortical excitability and with theta–gamma coupling in the hippocampus [237]. This indicates that the pressure waves caused by clearance may play a role in coordinating large scale populations of neurons through either the regulation of the ionic composition of the extracellular space or through the regulation of network gain mediated by astrocytes. Thus, disruptions in the synchronization of these processes could disrupt the coordination of different regions of the brain and the temporal relationships between the activity of neurons in different regions, which are necessary for working memory, cognitive control, and flexible behavior [238].

Furthermore, the relationship between clearance and the systems level stability of recurrent networks indicates that perturbations of clearance can add thermal “noise” to network computations, disrupting attractor basins, inducing imperfect transitions between cognitive states and altering the repertoire of possible network configurations [239]. The mechanisms described above may also be responsible for the early and subtle cognitive dysfunction seen in individuals who are pre-clinically neuro-degenerative, where mild impairments in high frequency clearance processes are sufficient to cause detectable cognitive dysfunction without significant structural change. Thus, the vasomotor pump represents a latent factor in cognitive plasticity that may influence learning, attention, and memory even in the absence of overt pathology [199].

Given the relationship between clearance and sleep, the vasomotor pump is not only activated during sleep but is also integrated into the structural and temporal organization of sleep itself. The implications of this relationship are broad and, in addition to memory consolidation, clearance dynamics also influence synaptic reorganization and metabolic re-equilibration, all of which are fundamental to the maintenance of cognitive health and brain longevity [240].

During NREM sleep, the LC undergoes a transition from tonic firing to infra-slow oscillatory firing caused by intrinsic membrane currents and reciprocal inhibition between inhibitory cells. The LC oscillations are timed to the cortical down-states, thalamocortical spindles, and hippocampal sharp wave ripples, a process which forms a hierarchical temporal sequence that is suitable for memory replay and synaptic renormalization [241]. The vasomotor pump is part of this hierarchy. The adrenergic oscillations caused by the LC create slow vasomotion and the pressure gradients created by the vasomotion are at their greatest during cortical down-states when the population activity of neurons is at its lowest and when the energy requirements of the organism are at their lowest. Therefore, perivascular flow is enhanced at times when clearance is energetically advantageous and least disruptive to the ongoing synaptic processing [242].

This temporal relationship extends beyond the removal of metabolites. The perivascular flow influences the distribution of neuromodulators and extracellular vesicles, and therefore influences the timing and location of the signaling cascades involved in synaptic plasticity and memory consolidation [243]. In addition, clearance dynamically regulates the concentration of K⁺ and H⁺ in the extracellular space, both of which are strong regulators of the excitability of neurons and the oscillations of networks. Together, these processes indicate that clearance mechanisms are not only permissive for sleep-dependent plasticity, but are also active components in the cellular and network events that facilitate memory consolidation [244].

Optogenetic and pharmacological manipulations of LC firing during sleep have provided evidence for this perspective. Artificial restoration of the infra-slow noradrenergic oscillations causes an enhancement of perivascular flow and improved memory consolidation of spatial memories, whereas disruption of the LC oscillations impairs clearance and memory retention independent of the total duration of sleep [71]. These results establish a mechanistic link between clearance and the mnemonic functions of sleep and suggest that therapeutic manipulation of the vasomotor pump may enhance memory consolidation in the aged or diseased.

In addition to NREM sleep, clearance dynamics may also be involved in the synaptic remodeling that occurs during REM sleep. Although LC activity is low during REM sleep, the persistence of pressure oscillations caused by residual vascular compliance and respiratory coupling suggests that there is still some clearance occurring in this state [245]. Although the residual clearance may be insufficient to support widespread signaling and remodeling processes that occur during NREM sleep, it may be sufficient to support localized signaling and remodeling processes that differ in character from those occurring during NREM sleep. Therefore, clearance may be involved in all stages of the sleep cycle [6].

The noradrenergic vasomotor pump is not only involved in removing metabolic waste products. It is also deeply involved in neuroimmune communication, extracellular signaling, and systemic physiology. Pathophysiological disturbances in this system extend beyond traditional neurodegenerative disease to include a variety of psychiatric disorders, infections, and systemic inflammatory conditions [246].

Perivascular flow serves as a primary route for antigen drainage and immune surveillance. The flow transports solutes and immune mediators from the brain parenchyma to the meningeal and deep cervical lymphatics, thereby determining the degree of antigen presentation, adaptive immune priming, and communication between the central and peripheral immune compartments [247]. Slowed clearance of antigens will slow antigen trafficking, distort cytokine gradients, and alter the recruitment of immune cells, and may contribute to chronic neuroinflammation, impaired clearance of pathogens, and aberrant immune tolerance. These mechanisms help to explain the dysregulation of neuroimmune responses in a number of conditions, including multiple sclerosis, viral encephalitis, and post-infectious cognitive syndromes. Clearance also influences systemic physiology through the molecular cargo transported through clearance. Extracellular vesicles containing microRNAs, growth factors, and peptide signaling molecules move along perivascular flow and enter the peripheral circulation via the meningeal lymphatics, where they can modulate peripheral immune states, metabolic pathways, and possibly vascular tone. These findings indicate that the brain may influence the state of remote organs through clearance-dependent signaling [227,248]. In the CNS, clearance determines the spatial and temporal distributions of extracellular vesicles and signaling mediators involved in synaptic plasticity, neurogenesis, and glio-transmission. Therefore, altered clearance dynamics can affect the operation of particular signaling pathways and influence adult neurogenesis in the subventricular zone, microglial synaptic pruning, and long-range astrocyte–neuron communication [249]. These phenomena illustrate how clearance not only contributes to the maintenance of homeostatic constancy, but also to the adaptation and plasticity of behavior. There is now a growing literature which suggests that clearance dynamics may be involved in psychiatric pathology. There is evidence that disturbances in noradrenergic tone, decreased vascular compliance, and hyperpolarized AQP4 expression are present in major depression, schizophrenia, and bipolar disorder [250]. These disturbances in clearance would be expected to compromise trans-cellular and extra-cellular ionic homeostasis, disrupt synaptic signaling, and disrupt immune signaling cascades. These disruptions may make the networks more unstable and prone to abnormal patterns of activity. Therefore, modulating clearance dynamics may represent a new area of therapeutic intervention in psychiatric disorders, many of which are presently defined primarily in terms of neurotransmitter and receptor abnormalities [251].

The systemic implications of clearance clearly go beyond the CNS, and age-related declines in perivascular flow not only directly impair cerebral homeostasis, but also impair the trafficking of extracellular vesicles and cytokines involved in systemic inflammatory states (“inflammaging”), vascular stiffness, and immuno-senescence. Enhancing clearance may therefore have advantages that extend beyond the CNS, and potentially improve systemic metabolic health, immune competency, and vascular resilience [252].

Collectively, these findings indicate that a new paradigm may exist. The vasomotor pump can be viewed as a bidirectional signaling nexus between the brain and the body, and serves as a means of clearing solutes as well as a conduit for inter-organ communication. This connectivity may allow for coordinated regulation of neural, immune, and systemic physiological processes, and presents a fundamentally integrating model for understanding a wide range of diseases and conditions that are presently considered to be distinct neurological, psychiatric, and systemic diseases [253].

No new data were created or analyzed in this study. Data sharing is not applicable to this article.