Neuroglia’s signaling: Orchestrating neuronal gamma synchrony?

Introduction

Emerging evidence suggests that sensory-induced brainwave entrainment within the gamma frequency range (30–80 Hz, typically at 40 Hz) can be used as a therapeutic tool to slow progression of Alzheimer’s disease highlighting that gamma oscillations play a role in healthy brain function [1,2]. Gamma oscillations occur naturally in the brain and play a central role in cognitive functions. Their disruption has been linked to various neurodevelopmental and neurodegenerative pathological conditions including schizophrenia, autism spectrum disorder, attention deficit/hyperactivity disorder, Alzheimer’s disease, and Parkinson’s disease (for review see ref: [3]).

Although gamma oscillations originate from neuronal activity, several lines of work suggest that astrocytes also contribute to their regulation. In this review, we examine recent advances on astrocytic modulation of the extracellular environment and propose mechanisms by which astrocytes support gamma oscillations in the healthy brain. Furthermore, we explore how astrocytic dysfunction may contribute to neurological disorders characterized by abnormalities in gamma oscillations.

Gamma oscillations, PV interneurons and cognitive function

Gamma oscillations reflect synchronized neuronal activity and are thought to play a pivotal role in sensory and cognitive processing. Neuronal synchrony was initially associated with large-amplitude slow brain rhythms that dominate during sleep. These slow oscillations diminish during arousal, giving rise to the term “desynchronization” [4]. However, fast, lower-amplitude gamma oscillations were later identified during wakefulness, particularly in relation to sensory processing and cognition.

Gamma oscillations are believed to emerge from the interplay between excitatory-inhibitory feedback loops (Figure 1a) [5]. Parvalbumin (PV) interneurons are strongly implicated in generating these oscillations through perisomatic inhibition of excitatory neurons via GABA_A receptors [5–8]. When PV interneurons fire in synchrony, they produce rhythmic inhibitory postsynaptic potentials (IPSPs) in pyramidal neurons, resulting in the characteristic rhythmic discharge pattern of gamma oscillations. At the local circuit level, gamma oscillations depend on a fine-tuned balance between excitatory and inhibitory inputs [9]. This balance relies on an appropriate inhibitory tone and ion homeostasis, which are hallmarks of healthy neuronal networks. Disruptions, such as hyperexcitability, may therefore impede gamma oscillations (Figure 1b).

Originally, gamma oscillations were described in the context of sensory processing, where sensory evoked gamma oscillations encode stimulus features, such as visual contrast, and integrate these features through synchrony among neurons encoding different aspects of the same stimulus [10]. Beyond sensory processing, gamma oscillations are now implicated in higher cognitive functions, including selective attention and memory formation [11–15]. A key feature of gamma oscillations is their role in precise timing of spiking, which may support spike-timing dependent plasticity [16].

The phase-locking and coherence of gamma oscillations across cortical and subcortical regions – measures of synchrony within neuronal assemblies - modulate cognitive processes like selective attention [11,12]. For instance, gamma-coupling with theta oscillations in olfactory bulb is heightened during attention switching depending on odors [13]. Such cross-frequency coupling, particular between gamma to slower oscillations, may facilitate coordinated communication between neuronal assemblies, defined as groups of neurons linked by spatio-temporal activity patterns [5].

Gamma oscillations have been widely established as a neural correlate of higher cognitive function. Their significance is thought to lie in their ability to coordinate and integrate neuronal assemblies across both proximal and distant brain regions [17]. For example, distinct gamma frequency bands have been shown to mediate communication between entorhinal cortex and hippocampus, which is critical for spatial and object learning [15]. Both excitatory and inhibitory long-range neurons have been implicated in mediating long range gamma synchronization [5], and a recent study highlighted the role of long-range inhibitory neurons in driving interhemispheric gamma synchrony [18]. While some have questioned whether gamma coherence is simply a correlation of inter-regional communication [19], recent in vivo studies have provided compelling evidence for a causal role of gamma oscillations in facilitating inter-hemispheric and inter-regional communication [14,15].

In summary, gamma oscillations play an active role in sensory processing and inter-regional synchronization, both of which are essential for higher cognitive function. Given that the generation of gamma oscillations depends on feedback inhibition by PV interneurons, factors influencing the inhibitory tone are crucial for the regulation and proper functioning of gamma oscillations.

Astrocytes affect inhibitory tone by regulating the extracellular milieu

Astrocytes are glial cells that perform numerous critical functions in the central nervous system. They maintain the blood-brain barrier and regulate the extracellular environment within the brain parenchyma by keeping a tight control of extracellular ion concentrations and taking up neurotransmitters. For instance, astrocytes clear glutamate and K⁺ ions during synaptic transmission and control extracellular K⁺ in a brain-state dependent manner [20,21]. Specifically, K⁺ uptake by astrocytes steepens the somatic gradient of K⁺ across the neuronal cell membrane leading to more efficient hyperpolarization following excitation without affecting inhibitory potentials [20]. The overall effect is a reduction in postsynaptic excitation. Additionally, astrocytes can act as a reservoir of Cl⁻, thereby directly modulating inhibitory neurotransmission [22,23]. Through these mechanisms, astrocytes influence neuronal excitability by shaping the ionic composition of the extracellular fluid.

Astrocytic K⁺ uptake is facilitated by intracellular Ca²⁺ elevations, which in astrocytes are triggered by G_q coupled receptor activation and subsequent IP₃-mediated release of Ca²⁺ from the endoplasmic reticulum [20]. These astrocytic Ca²⁺ events are elicited by neuromodulators such as norepinephrine, acetylcholine, and dopamine, which govern brain states [24,25], but are in particular linked to adrenergic α₁ receptor activation [26,27]. Since NE levels correlate with arousal and state transitions [28], astrocytic Ca²⁺ rises are consistently observed during heightened arousal, such as startle responses or sleep-wake transitions [29–32]. Thus, following arousal, astrocytes might terminate excessive excitation by enhancing K⁺ clearance in a NE- and Ca²⁺-dependent manner.

Another downstream effect of NE signaling through astrocytic Ca²⁺ is ATP release [33,34]. This has directed attention to the role of astrocytes in adenosine-mediated feedback inhibition [35–37]. In hippocampal slices from adult mice, NE reduces synaptic strength and presynaptic release probability via adrenergic α_1A signaling in astrocytes but not neurons [37]. This pathway was dependent on adenosine receptor A1. A similar pathway in zebrafish [36] revealed a mechanism, where NE induces a biphasic response – an initial excitatory phase mediated by β-adrenergic receptors followed by an α₁-mediated inhibitory phase resulting in passivity [36]. This delayed inhibitory phase aligns with slower astrocytic responses compared to neuronal signaling. While adenosine receptor A1 is implicated in this astrocyte-mediated inhibitory feedback [25,35–37], conflicting evidence challenges the notion that ATP released from astrocytes is converted to adenosine, extracellularly. For instance, one study demonstrated that IP₃ receptor 2 (IP₃R2) activation in astrocytes contributes to extracellular ATP but not to adenosine levels [33]. Moreover, ATP from astrocytes has been suggested to enhance excitability [34,38], contrasting with adenosine’s inhibitory effects. Thus, the exact mechanism remains elusive and warrants further investigation.

While much of the literature focuses on astrocytic IP₃R2-mediated effect of NE, recent studies highlighted alternative pathways. For instance, adrenergic signaling in astrocytes has been shown to modulate excitation by releasing free fatty acids that influence K⁺ uptake in both neurons and astrocytes, causing a swelling of the latter [39]. Additionally, β-adrenergic signaling reduces glutamate spillover onto extrasynaptic NMDA receptors [40]; likely a result of NE-mediated astrocytic swelling [39]. These findings illustrate that NE signaling exerts profound effects on astrocytic intracellular signaling, ion buffering, and volume regulation.

Do astrocytes facilitate gamma oscillations?

As described, astrocytes respond to neuromodulators and modulate the extracellular ionic environment, which could modulate inhibitory tone critical for PV interneuron-generation of gamma oscillations. Interestingly, it was also demonstrated that vesicular release of the astrocytic S100B Ca²⁺-binding protein enhances hippocampal gamma oscillations [41], suggesting that gliotransmission modulate gamma oscillations. Blocking vesicular release in astrocytes using tetanus toxin reduced duration of gamma oscillations in hippocampal slices without directly affecting synaptic transmission [42]. This reduction in gamma power was observed in vivo during wakefulness only and was associated with impaired object recognition, while short-term memory and fear memory remained intact [42]. These findings align with the role of phase-amplitude coupling between gamma and theta oscillations in hippocampal-dependent learning [43,44]. In addition to paracrine signaling such as gliotransmission, vesicular trafficking also affects channels and transporters in the cell membrane and thus ion homeostasis. Thus, interference with astrocytic regulation of the extracellular milieu likely explains the observed disruptions.

Astrocytes influence neuronal excitability through various mechanism, including limiting glutamate spillover [40], modulating inhibitory synapses [22], IP₃R2-mediated K⁺ removal [20], and possible by adenosine-mediated inhibition [35–37]. As a single astrocyte contacts numerous neurons [45], they could adjust inhibitory tone at the local circuit or network level to support PV-interneuron mediated gamma oscillations. Astrocytes also regulate axonal conduction speed by modulating myelination [34,46], which influences gamma oscillation frequency by affecting synaptic delays and phase shift between excitatory and inhibitory neurons [5]. Long-range interneurons mediating cross-regional gamma synchrony may also be influenced by astrocytic regulation of conduction speed [5].

PV interneurons are essential for gamma oscillations, and astrocytes modulate inhibitory synapses through Cl⁻ regulation [20,22]. Computational models suggest that astrocytes enhance gamma synchrony by facilitating inhibitory synapses [47]. Experimental evidence supports this, as GABA_B receptor signaling in astrocytes contributes to gamma oscillations; conditional knockout of this receptor reduces gamma-theta coupling [48]. Conversely, stimulating astrocytes via channelrhodopsin (ChR2) disrupts gamma power, likely due to abnormal K⁺ efflux and increased neuronal excitability [49,50]. Together, these findings highlight astrocytes’ role in maintaining the extracellular environment necessary for gamma oscillations.

Astrocytic Ca²⁺ elevations in the somatosensory cortex accompany sensory-induced gamma oscillations, as demonstrated during electrical paw stimulation [51]. This response, likely mediated by NE, is stimulus-dependent, with gamma oscillations preceding the astrocytic activity. Mice lacking IP₃R2 exhibit exaggerated gamma power upon sensory stimulation, suggesting astrocytes normally dampen such gamma responses [51]. G_q-DREADD-mediated Ca²⁺ elevation in astrocytes also impairs sensory-induced gamma power, though this interpretation may be complicated by G_q-DREADDs inhibitory effects on astrocytic Ca²⁺ [31]. Additionally, astrocytes counteract the desynchronizing effects of norepinephrine [29]; conditional knockout of adrenergic α₁ receptor in astrocytes increased gamma response during voluntary running [29]. This inhibitory effect may reflect astrocyte-mediated feedback inhibition [35–37] or regulation of Cl⁻ and K⁺ concentrations [20,22], preventing hyperexcitability.

While astrocytes support gamma oscillation and phase-coupling, they also reduce gamma power following arousal. It is important to distinguish true gamma oscillations from desynchronized non-oscillatory spiking activity, which can inflate gamma power measurements [5,52]. High frequency broadbands (>50 Hz) often reflect spike contamination, whereas narrow-bands (~30–40 Hz) more accurately represent synchronized oscillatory activity (Figure 1b) [52]. This distinction may resolve discrepancies, as reductions in induced gamma power [29,51] might reflect astrocytes counteracting spiking rather than disrupting true gamma oscillations. This may also explain findings from sleep studies showing that arousal transitions are accompanied by elevations in astrocytic Ca²⁺ and gamma power, respectively [30,53]; gamma power may merely be a sign of desynchronized non-oscillatory activity. Future research on astrocytic involvement in gamma oscillations should make this distinction between a general increase in excitation and true gamma synchronization.

In summary, astrocytic response to NE may terminate excitation following arousal, returning the system to a resting state conducive to PV interneuron-mediated gamma oscillations. Astrocytes contribute to inhibitory tone through diverse mechanisms, preventing hyperexcitability and maintaining network health, thus facilitating gamma oscillations and ensuring proper neuronal communication (Figure 2a).

Are astrocytes involved in neurological disorders with gamma deficits?

If astrocytes mediate neuromodulatory effects on gamma synchrony, disruption in astrocytic function could contribute to diseases characterized by gamma oscillation deficits and broader neural dysfunction. While gamma deficits are implicated in many disorders, we will focus on schizophrenia and Alzheimer’s disease in the following section. Alterations in gamma oscillations are well-documented in schizophrenia (SCZ), a condition often attributed to disinhibition and hyperexcitability linked to PV interneuron deficits [54,55]. Cognitive impairments, such as working memory deficits, are core symptoms of schizophrenia and have been associated with reduced task-evoked gamma oscillations [56]. Patients also exhibit reduced gamma coherence [57,58] and reduced responsiveness to auditory gamma entrainment [59], yet paradoxically show increased resting-state gamma power alongside elevated alpha and beta power [60,61]. This broadband power increase likely reflects irregular excitability rather than true oscillatory activity as mentioned in the previous section [52], consistent with earlier observations that task-related (true) gamma oscillations are reduced. The link between hyperexcitability and gamma disruption in schizophrenia is further supported by the higher incidence of epilepsy in schizophrenia patients [62].

Astrocytes may directly contribute to the pathophysiology of schizophrenia [63]. Chimeric mice with glial cells derived from SCZ human subjects develop abnormal astrocyte morphology and some schizophrenia-like behavior [63]. Single-nucleus RNA sequencing of cortical tissue from schizophrenia patients revealed reduced astrocytic expression of genes involved in synapse adhesion and neurotransmitter uptake [64], potentially impairing astrocytic regulation of neuronal activity. Studies of cultured astrocytes derived from schizophrenia patients also implicate astrocytes in the disease pathology [65–67]. SCZ astrocytes show reduced K⁺ uptake [65] and neurons co-cultured with SCZ astrocytes exhibit increased glutamate sensitivity [66], suggesting that dysfunctional astrocytes contribute to increased excitability in schizophrenia.

Gamma oscillation deficits are also observed in Alzheimer’s disease (AD) where reduced theta-gamma coupling correlates with working memory deficits [68,69]. Notably, Alzheimer’s disease is also associated to increased seizure susceptibility [70] and widespread hyperexcitability [69,71,72], potentially stemming from impaired astrocytic K⁺ regulation [72]. Remarkably, studies show that gamma entrainment can alleviate cognitive symptoms and reduce amyloid pathology [1,2], possible through enhancement in glymphatic clearance involving astrocytic endfeet modulation [2]. Astrocytes are increasingly recognized as key players in Alzheimer’s disease pathology. Several Alzheimer’s risk genes are linked to astrocyte morphology, and their knockdown induces cognitive deficits [73]. Mouse models of Alzheimer’s disease show disrupted astrocytic Ca²⁺ responses [74,75], which have been linked with network hyperactivity [76]. Astrocytic Ca²⁺ signaling appears protective against Alzheimer’s disease progression by mitigating neuronal hyperexcitability [76]. This protection may involve Ca²⁺ dependent K⁺ buffering [20] or the downstream inhibitory effects of purinergic signaling [35–37]. Interventions targeting astrocytic dysfunction in AD models, including manipulation of astrocytic Ca²⁺ signaling, have successfully rescued disease phenotypes [77–79].

In summary, astrocytic dysfunction and dysregulated Ca²⁺ signaling are implicated in neurological disorders with gamma oscillation deficits, such as schizophrenia and Alzheimer’s disease. These findings highlight the importance of astrocytes in maintaining excitatory-inhibitory balance and gamma coherence (Figure 2b). Future studies exploring the impact of astrocytic Ca²⁺ disruption or restoration on gamma oscillations and disease phenotypes in vivo could provide critical insights into their mechanistic role in these conditions.

Therapeutic use of gamma stimulation

Recent studies have highlighted the potential of noninvasive magnetic resonance imaging (MRI) to elucidate the dynamics of glymphatic function, offering a valuable bridge between cellular mechanisms, such as astrocytic signaling and gamma oscillations, and macroscopic brain clearance. Gamma entrainment, achieved through sensory stimulation at ~40 Hz (e.g., auditory and visual flicker), has been shown to reduce amyloid-beta accumulation and mitigate neuroinflammatory responses in Alzheimer’s disease models [80,81]. Emerging evidence suggests that gamma stimulation enhances perivascular CSF influx possibly by synchronizing neuronal and vascular activity [82], thereby enhancing glymphatic transport to remove inflammatory substances. Improving glymphatic flow by gamma stimulation may not only increase amyloid-beta waste clearance but also reduce neuroinflammation [83]. Mapping these effects using contrast-enhanced MRI in awake animals could offer unprecedented insight into how oscillatory activity impacts CSF influx and waste clearance in vivo. Incorporating MRI-based techniques such as dynamic contrast-enhanced imaging or diffusion tensor imaging will strengthen the translational relevance of these findings, especially for neurodegenerative and inflammatory diseases where glymphatic dysfunction likely plays a central role [84].

Conclusion

Astrocytes are uniquely positioned to modulate neuronal network synchrony by responding to neuromodulators and regulating the extracellular environment, thereby influencing neuronal excitability and the inhibitory tone essential gamma oscillation regulation. Astrocytes likely create optimal conditions for gamma oscillations by reducing neuronal excitability under baseline conditions and following brain state transitions involving heightened arousal. Their extensive interactions with neurons allow astrocytes to simultaneously influence multiple neurons, facilitating the recruitment of assemblies and amplifying their role in shaping network dynamics. Through these mechanisms, astrocytes may support gamma oscillations critical for higher cognitive functions. Conversely, astrocytic dysfunction could contribute to hyperexcitability and impaired gamma oscillations, as observed in conditions such as schizophrenia and Alzheimer’s disease.