Adenosine A₁ and A_2A Receptors in Sleep Disorders: Mechanisms and Therapeutic Implications

Abstract

Sleep–wake regulation is controlled by circadian and homeostatic processes, with adenosine acting as a key molecular mediator of homeostatic sleep pressure. Extracellular adenosine accumulates during wakefulness as a result of neuronal energy metabolism, particularly in the basal forebrain, and declines during recovery sleep, thereby reflecting the physiological need for sleep. The sleep-promoting effects of adenosine are mediated primarily by two G protein–coupled receptor subtypes, the adenosine A₁ receptor and the adenosine A_2A receptor. The adenosine A₁ receptor, coupled to inhibitory G_i/o proteins and widely expressed in the cortex, hippocampus, thalamus, and basal forebrain, suppresses wake-promoting neuronal activity and facilitates slow-wave activity during non-rapid eye movement sleep. In contrast, the adenosine A_2A receptor, coupled to stimulatory Golf proteins and enriched in the striatum and nucleus accumbens, promotes sleep by activating neurons in the preoptic hypothalamus and engaging the indirect basal ganglia pathway. Despite these well-established roles, the contributions of dysregulation of the adenosine A₁ receptor and the adenosine A_2A receptor to specific sleep disorders remain incompletely understood. This review examines how signaling of the adenosine A₁ receptor and the adenosine A_2A receptor is altered in insomnia, obstructive sleep apnea, narcolepsy, and restless legs syndrome, and evaluates the therapeutic potential of receptor-selective strategies for adenosine receptor–targeted treatment.

INTRODUCTION

Sleep is a fundamental physiological process that is essential for metabolic homeostasis, neural network stability, and brain function (Huang et al., 2024; Porkka-Heiskanen et al., 2002). Adequate sleep supports cognitive performance, synaptic plasticity, and neuronal recovery, whereas chronic sleep disruption is associated with neurological, psychiatric, metabolic, and cardiovascular disorders (Irwin, 2019; Lee et al., 2023). Sleep disorders are therefore a major public health concern. Insomnia affects approximately 10-12% of adults (Morin and Jarrin, 2022), obstructive sleep apnea (OSA) affects 9-38% of the population (Benjafield et al., 2019; Senaratna et al., 2017), and restless legs syndrome (RLS) occurs in about 3% of adults (Broström et al., 2023). Despite their high prevalence, current pharmacological treatments have limited long-term efficacy and safety.

Sleep–wake regulation is controlled by the interaction between circadian and homeostatic processes, which determine the timing and intensity of sleep (Borbély, 1982). Multiple neural circuits and neurotransmitter systems, including GABAergic, monoaminergic, cholinergic, and orexinergic pathways, regulate sleep–wake states (Saper and Fuller, 2017). Although these systems have been extensively studied, the molecular basis of the homeostatic drive for sleep remains unclear (Borbély et al., 2016). Among the molecular factors involved in sleep regulation, adenosine is a key neuromodulator that links cellular energy metabolism to sleep–wake control. Extracellular adenosine levels increase during wakefulness and decrease during sleep, reflecting homeostatic sleep pressure (Bjorness and Greene, 2009; Gao et al., 2024; Porkka-Heiskanen et al., 1997).

Adenosine exerts its physiological effects through four P1 G protein–coupled receptor subtypes: A₁, A_1A, A_1B, and A₃. The existence of distinct adenosine receptor subtypes was first suggested by the demonstration that adenosine stimulates cAMP accumulation in brain tissue (Sattin and Rall, 1970). Subsequent pharmacological studies further refined two major subtypes based on their opposing effects on adenylyl cyclase activity: A₁ receptors (A₁Rs), which inhibit cAMP production, and A₂ receptors (A₂Rs), which stimulate it (van Calker et al., 1979). These receptor subtypes differ in their tissue distribution, G protein coupling, and downstream signaling pathways (Martinez-Gallego and Rodriguez-Moreno, 2025; Trincavelli et al., 2010). Among them, A_2B receptors (A_2BRs) and A₃ receptors (A₃Rs) appear to have limited or no established roles in sleep–wake regulation under physiological conditions, whereas A₁R and A_1AR are highly expressed in the central nervous system (CNS) and are key regulators of sleep homeostasis, as demonstrated by pharmacological and genetic studies (Lazarus et al., 2019a; Trincavelli et al., 2010). The identification of caffeine as a non-selective antagonist of A₁R and A_1AR provided early functional evidence linking these receptors to sleep–wake regulation (Fredholm et al., 1999). More recently, the use of A₁R and A_1AR knockout models has enabled the delineation of their distinct contributions to sleep homeostasis (Huang et al., 2024; Johansson et al., 2001).

Nevertheless, how dysregulation of A₁R and A_1AR signaling contributes to specific sleep disorders remains unclear. This review therefore examines A₁R and A_1AR signaling in four major sleep disorders, insomnia, OSA, narcolepsy, and RLS evaluates the therapeutic potential of receptor-selective pharmacological strategies.

ADENOSINE IN SLEEP-WAKE HOMEOSTASIS

Sleep–wake regulation is governed by the interaction between a circadian process (Process C), driven by the suprachiasmatic nucleus (SCN), and a homeostatic process (Process S), which reflects the accumulation of sleep pressure during wakefulness and its dissipation during sleep (Borbély, 1982; Borbély et al., 2016). Among the proposed mediators of Process S, adenosine is the most strongly supported candidate (Bjorness and Greene, 2009; Gao et al., 2024). As a metabolic product of adenosine triphosphate (ATP) hydrolysis, extracellular adenosine increases during wakefulness and decreases during sleep, reflecting neuronal energy use (Borbély et al., 2016; Porkka-Heiskanen et al., 1997). This increase is region-specific, as sleep deprivation selectively elevates adenosine levels in the basal forebrain (BF), while other brain regions show minimal change (Porkka-Heiskanen et al., 2000). These findings provide direct in vivo evidence that adenosine encodes sleep pressure and links neuronal metabolism to sleep homeostasis.

Extracellular adenosine levels are controlled by the balance between production and clearance. Adenosine is generated intracellularly from AMP by cytosolic 5′-nucleotidases and extracellularly through ATP breakdown by CD39 and CD73 (Gao et al., 2024; Lazarus et al., 2019a), while clearance is mainly mediated by adenosine kinase (ADK) in astrocytes and adenosine deaminase (ADA), which converts adenosine to inosine (Bjorness and Greene, 2009; Ren et al., 2021). During prolonged wakefulness, inducible nitric oxide synthase (iNOS) is activated in wake-active BF neurons, and nitric oxide inhibits ADK, leading to further accumulation of extracellular adenosine (Kalinchuk et al., 2010; Reichert et al., 2022). This increase is translated into sleep-promoting signals through specific neural circuits. In the BF, glutamatergic neurons are the primary source of activity-dependent adenosine release, and their ablation impairs sleep homeostasis (Peng et al., 2020). Beyond the BF, adenosine activates neurons in the ventrolateral preoptic area (VLPO) and median preoptic nucleus (MnPN), and modulates thalamocortical activity associated with non-rapid eye movement (NREM) sleep (Huang et al., 2024). In addition, prostaglandin D₂ promotes sleep by increasing adenosine release and activating A_2AR signaling in the BF, highlighting its role as a common mediator of sleep-promoting pathways (Porkka-Heiskanen et al., 2002). Together, these mechanisms establish adenosine as a central mediator of homeostatic sleep pressure, acting mainly through A₁R and A_2AR signaling, as illustrated in Fig. 1.

Fig. 1.

Adenosine A₁ and A_1A Receptor Signaling Pathways in Sleep Regulation. Extracellular adenosine (orange circles) accumulates during wakefulness and acts on two G protein-coupled receptor subtypes expressed on neuronal membranes. (Left) The adenosine A₁ receptor (A₁R) signaling suppresses arousal through distinct presynaptic and postsynaptic mechanisms. Presynaptically, A₁R couples to inhibitory Gi proteins, and Gβγ-mediated inhibition of voltage-gated calcium channels (VGCC) reduces Ca²⁺ influx and suppresses neurotransmitter release from nerve terminals. Postsynaptically, A₁R activation engages Gi-dependent inhibitory signaling through two complementary pathways: Gβγ-mediated activation of G protein-coupled inwardly rectifying potassium channels (GIRK), which increases K⁺ efflux and causes membrane hyperpolarization with reduced neuronal firing, and Gαi-mediated inhibition of adenylyl cyclase, which lowers cyclic AMP (cAMP) and protein kinase A (PKA) activity, thereby decreasing neuronal excitability. Together, these presynaptic and postsynaptic effects suppress wake-promoting neurons in the lateral hypothalamus (LH; orexin neurons), basal forebrain (BF; cholinergic neurons), and tuberomammillary nucleus (TMN; histaminergic neurons), contributing to arousal suppression and sleep pressure accumulation. (Right) The adenosine A_1A receptor (A_1AR) couples to stimulatory Gs proteins, whose Gαs subunit activates adenylyl cyclase to increase cAMP and PKA activity. This excitatory signal is transduced through three key sleep-promoting circuits: in the ventrolateral preoptic area and median preoptic nucleus (VLPO/MnPN), A_1AR activation excites sleep-active GABAergic and galaninergic neurons that suppress TMN histaminergic neurons; in the nucleus accumbens (NAc), A_1AR activation of D₂R/A_1AR-expressing indirect-pathway neurons promotes slow-wave sleep; and in the striatum, activation of A_1AR-expressing neurons drives inhibitory projections to external globus pallidus parvalbumin neurons (GPe PV neurons), collectively contributing to sleep promotion. Together, A₁R-mediated suppression of arousal systems and A_1AR-mediated activation of sleep-promoting circuits coordinately regulate sleep–wake homeostasis. Black arrows indicate signal activation; blunt-ended lines indicate inhibition; red arrows indicate excitatory circuit projections; blue lines indicate inhibitory circuit projections. A₁R, adenosine A₁ receptor; A_1AR, adenosine A_2A receptor; BF, basal forebrain; cAMP, cyclic adenosine monophosphate; D₂R, dopamine D₂ receptor; GABA, gamma-aminobutyric acid; GIRK, G protein-coupled inwardly rectifying potassium channel; GPe PV, external globus pallidus parvalbumin neurons; Gs, stimulatory G protein; Gi, inhibitory G protein; Gα, G protein alpha subunit; Gβγ, G protein beta-gamma subunit; LH, lateral hypothalamus; MnPN, median preoptic nucleus; NAc, nucleus accumbens; PKA, protein kinase A; TMN, tuberomammillary nucleus; VGCC, voltage-gated calcium channel; VLPO, ventrolateral preoptic area.

ADENOSINE RECEPTORS IN SLEEP HOMEOSTASIS

Four adenosine receptor subtypes, namely A₁R, A_1AR, A_1BR, and A₃R, differ in their distribution and signaling properties. Among these, A₁R and A_1AR are the primary receptors implicated in sleep–wake regulation, whereas A_1BR and A₃R have limited or no established roles under physiological conditions, as summarized in Table 1.

Table 1.

Adenosine receptor subtypes and their roles in sleep–wake regulation

Receptor	Distribution	G-protein Coupling	Role in Sleep Regulation	Representative Ligands	References
A1R	• CNS: cortex, hippocampus, thalamus, cerebellum, BF, basal ganglia, striatum, astrocytes • Peripheral: heart, kidney, liver, adipose tissue, vascular smooth muscle, spinal cord	Gi/o	• Enhances NREM SWA (0.5-4.5 Hz) • Inhibits BF, orexin, and histaminergic arousal neurons • Gi/o coupling reduces cAMP and activates GIRK channels, causing neuronal hyperpolarization • A1R KO results in a loss of SWA rebound following sleep deprivation	• Agonists: CPA, CCPA, GR79236, AR-001 • Antagonists: DPCPX, DPX, Caffeine (non-selective)	Bjorness and Greene, 2009; Kim et al., 2024; Lazarus et al., 2019b; Reichert et al., 2022; Thakkar et al., 2003
A2AR	• CNS: Striatum, nucleus accumbens, olfactory bulb, olfactory tubercle, astrocytes • Peripheral: platelets, immune cells, vasculature, lung, heart	Gs/Golfα	• Activates NAc D2/A2AR MSNs to drive GABAergic inhibition of VP, inducing SWS • Excites VLPO/MnPN GABAergic neurons to suppress TMN histaminergic arousal output • Golfα–cAMP–PKA signaling enhances excitability of sleep-promoting circuits • A2AR KO abolishes the ethanol induced increase in NREM sleep	• Agonists: CGS21680 • PAMs: A2AR positive allosteric modulators • Antagonists: ZM-241385, SCH58261, SCH-442416, Caffeine (non-selective)	Huang et al., 2024; Kumar et al., 2013; Lazarus et al., 2019b; Lin et al., 2023
A2BR	• CNS: Low CNS expression; hippocampus, striatum, thalamus • Peripheral: vasculature, mast cells, lung, intestine	Gs/Gq	No established physiological role in sleep regulation	Not well characterized for sleep	Lazarus et al., 2019a; Trincavelli et al., 2010
A3R	• CNS: Low CNS expression; hippocampus, cerebellum • Peripheral: lung, liver, testis, mast cells, spleen, kidney	Gi/o	Insufficient evidence for a direct role in sleep–wake regulation	Not well characterized for sleep	Lazarus et al., 2019a; Trincavelli et al., 2010

A₁R, adenosine A₁ receptor; A_2AR, adenosine A_2A receptor; A_2BR, adenosine A_2B receptor; A₃R, adenosine A₃ receptor; BF, basal forebrain; cAMP, cyclic adenosine monophosphate; CNS, central nervous system; GIRK, G-protein–gated inwardly rectifying potassium channel; KO, knockout; LH, lateral hypothalamus; MnPN, median preoptic nucleus; MSNs, medium spiny neurons; NAc, nucleus accumbens; NREM, non-rapid eye movement sleep; PKA, protein kinase A; PAM, positive allosteric modulator; SWA, slow-wave activity; SWS, slow-wave sleep; TMN, tuberomammillary nucleus; VLPO, ventrolateral preoptic nucleus; VP, ventral pallidum.

A₁Rs are widely expressed in the CNS, particularly in the cortex, hippocampus, thalamus, and BF, and are coupled to inhibitory G_i/o proteins that reduce neuronal excitability (Trincavelli et al., 2010). A₁R exerts differential effects depending on its synaptic localization. Presynaptically, A₁R activation inhibits voltage-gated Ca²⁺ channels (VGCCs), thereby reducing Ca²⁺ influx and suppressing neurotransmitter release from presynaptic terminals, ultimately attenuating excitatory and wake-promoting synaptic transmission (Dunwiddie and Masino, 2001; Reichert et al., 2022). In contrast, postsynaptic A₁R activation, primarily mediated via Gβγ subunits, activates G protein–coupled inwardly rectifying potassium (GIRK) channels, leading to increased K⁺ efflux and membrane hyperpolarization, which reduces neuronal excitability and firing (Gulia, 2012; Reichert et al., 2022). In addition, postsynaptic A₁R signaling via Gαi inhibits adenylyl cyclase, resulting in decreased cAMP levels and PKA activity, thereby further suppressing neuronal excitability. Collectively, these presynaptic (reduced neurotransmitter release) and postsynaptic (membrane hyperpolarization and intracellular signaling suppression) mechanisms act in concert to promote slow-wave activity (SWA) during NREM sleep (Reichert et al., 2022). At the circuit level, A₁R activation suppresses cholinergic neurons in the BF, orexin neurons in the lateral hypothalamus (LH), and histaminergic neurons in the tuberomammillary nucleus (TMN), thereby reducing arousal (Chikahisa and Sei, 2011; Huang et al., 2024). Functionally, A₁R signaling is critical for sleep homeostasis, as A₁R knockout mice show reduced SWA rebound after sleep deprivation without changes in total sleep duration (Lazarus et al., 2019b; Reichert et al., 2022). Pharmacological blockade of A₁R similarly reduces SWA and attenuates the homeostatic response to sleep loss (Urry and Landolt, 2015).

In contrast, A_2ARs are highly expressed in the striatum, nucleus accumbens (NAc), and olfactory bulb, and are coupled to stimulatory G proteins that enhance cAMP–PKA signaling and neuronal excitability (Liu and Gao, 2007; Reichert et al., 2022; Sims et al., 2013). A_2AR activation promotes both NREM and rapid eye movement (REM) sleep, partly through enhanced GABAergic inhibition of TMN histaminergic neurons (Bjorness et al., 2009; Chikahisa and Sei, 2011). Pharmacological and genetic studies support this role, including increased sleep following A_2AR activation and loss of ethanol-induced sleep responses in A_2AR knockout mice (Fang et al., 2017; Hong et al., 2005; Satoh et al., 1999; Scammell et al., 2001).

A_2AR-mediated sleep promotion involves multiple circuits. In the striatum, activation of A_2AR-expressing neurons increases NREM sleep via projections to globus pallidus neurons (Huang et al., 2024). In the NAc, A_2AR-expressing indirect pathway neurons regulate the transition from wakefulness to sleep, as demonstrated by the loss of caffeine-induced wakefulness after A_2AR deletion (Thakkar et al., 2008). In the preoptic hypothalamus, A_2AR activation enhances inhibitory output from VLPO and MnPN neurons, facilitating sleep onset (Bjorness et al., 2009; Kumar et al., 2013).

Taken together, A₁R and A_2AR regulate sleep through complementary mechanisms: A₁R controls homeostatic sleep pressure by suppressing arousal, whereas A_2AR promotes sleep by activating sleep-promoting circuits and facilitating the transition from wakefulness to sleep.

A₁R AND A_1AR IN SLEEP DISORDERS

Sleep disorders arise from diverse pathophysiological mechanisms that disrupt sleep–wake regulation. Growing evidence indicates that dysregulation of A₁R and A_1AR signaling contributes to the development and maintenance of major sleep disorders. This section addresses the mechanisms and therapeutic implications of A₁R and A_1AR in four major sleep disorders - insomnia, OSA, narcolepsy, and RLS - as summarized in Table 2.

Table 2.

A₁R and A_2AR in sleep disorders: mechanisms and therapeutic strategies

Sleep Disorder	Pathophysiology	Mechanism	Therapeutic Strategy	References
Insomnia	ADA overactivity depletes adenosine, causing hypofunction of both A₁R- and A2AR-mediated sleep-promoting pathways	• Reduced A₁R inhibition disinhibits BF, LH, and TMN arousal neurons • Insufficient A2AR activation impairs preoptic sleep circuit recruitment	• AR-001 (A₁R agonist): shortens sleep onset latency and prolongs total sleep time • A2AR PAMs: enhance endogenous adenosine signaling with lower cardiovascular risk than direct agonists	Kim et al., 2024; Kumar et al., 2013; Lin et al., 2023; Oishi et al., 2008; Ren et al., 2021; Thakkar et al., 2003
Obstructive Sleep Apnea (OSA)	CIH alters adenosine signaling: A₁R downregulation/desensitization and A2AR hyperactivation	• A₁R downregulation destabilizes brainstem respiratory control • A2AR hyperactivation promotes hippocampal neuroinflammation and cognitive impairment	• GR79236, CCPA (A₁R agonists): reduce apnea frequency and protect hippocampal neurons • SCH58261 (A2AR antagonists): reduce neuroinflammation and cognitive impairment	Carley et al., 1997; Coelho et al., 2006; Johansson et al., 2001; Li et al., 2022; Ma et al., 2025
Narcolepsy (NT1)	Orexin neuron loss disrupts adenosinergic arousal regulation; BF adenosine fails to accumulate normally during sleep deprivation in orexin-deficient models	• Residual A₁R inhibition further weakens surviving orexin neurons • A2AR activation of the indirect striatopallidal pathway additionally suppresses wakefulness	• A2AR antagonism reinforces wake-promoting circuits • Caffeine (A₁R/A2AR antagonist): provides indirect clinical support • Modafinil (DAT inhibitor): provides wake promotion via dopaminergic system	Ferre et al., 2023; Mochizuki et al., 2004; Murillo-Rodriguez et al., 2008; Scammell, 2015
Restless Legs Syndrome (RLS)	BID selectively reduces the A₁R/A2AR ratio at corticostriatal terminals, inducing a hypoadenosinergic state	• A₁R downregulation drives corticostriatal glutamate hyperactivity and striatal hyperdopaminergia • A2AR upregulation impairs D₂R signaling, producing sensorimotor dysregulation and PLMS	• Dipyridamole (ENT inhibitor): raises extracellular adenosine and reduces IRLS scores and PLMI in clinical trials • SCH-442416 (A2AR antagonist): selectively targets presynaptic A2ARs while sparing postsynaptic D₂LR signaling	Ferre et al., 2017, 2023; Garcia-Borreguero et al., 2021; Quiroz et al., 2016

ADA, adenosine deaminase; A₁R, adenosine A₁ receptor; A_2AR, adenosine A_2A receptor; BF, basal forebrain; BID, brain iron deficiency; CIH, chronic intermittent hypoxia; D2LR, dopamine D2 long-form receptor; D2R, dopamine D2 receptor; DAT, dopamine transport; ENT, equilibrative nucleoside transporter; IRLS, International Restless Legs Syndrome Study Group severity scale; LH, lateral hypothalamus; LTP, long-term potentiation; NT1, narcolepsy type 1; PAM, positive allosteric modulator; PLMI, periodic limb movement index; PLMS, periodic limb movements during sleep; TMN, tuberomammillary nucleus; VLPO, ventrolateral preoptic area.

Insomnia

Insomnia is characterized by persistent dissatisfaction with sleep quality or quantity and commonly presents as difficulty initiating sleep, difficulty maintaining sleep, or early-morning awakening (Patel et al., 2018). It is associated with a state of hyperarousal in which arousal-promoting circuits remain pathologically active, leading to impaired sleep initiation and fragmented sleep architecture (Ren et al., 2021). Clinical evidence supports a role for adenosinergic dysfunction, as insomnia patients show reduced morning plasma adenosine levels and increased ADA activity; because ADA irreversibly converts adenosine to inosine, its upregulation reduces adenosine availability and weakens inhibitory control over wake-promoting neuronal networks (Ren et al., 2021). In contrast to the physiological accumulation of adenosine in the BF observed during sleep deprivation in healthy individuals, insomnia is characterized by a paradoxical reduction in adenosine bioavailability attributable to elevated ADA activity, representing a pathological disruption of homeostatic adenosinergic signaling rather than its physiological expression (Ren et al., 2021). Reduced adenosinergic tone impairs both A₁R- and A_2AR-mediated regulation of sleep circuits. A₁R normally suppresses arousal by inhibiting BF cholinergic neurons, LH orexin neurons, and TMN histaminergic neurons (Liu and Gao, 2007; Oishi et al., 2008; Thakkar et al., 2003). In insomnia, increased ADA activity reduces extracellular adenosine bioavailability, thereby diminishing A₁R-mediated inhibitory drive on these wake-promoting circuits and contributing to sustained hyperarousal (Ren et al., 2021). Notably, A₁R-mediated sleep induction is not limited to cholinergic BF neurons, as A₁R agonist infusion induces sleep even after near-complete lesion of these neurons, indicating a contribution of noncholinergic BF neurons (Martinez-Gallego and Rodriguez-Moreno, 2025; Thakkar et al., 2003). Pharmacological evidence strongly supports this mechanism: the selective A₁R agonist AR-001 produces pronounced hypnotic effects, significantly shortening sleep onset latency and increasing total sleep time, while suppressing neuronal activation in TMN and LH and enhancing activity in the VLPO (Kim et al., 2024). Similarly, β-lapachone suppresses neuronal activity in wake-promoting regions and enhances VLPO activation through A₁R signaling, further supporting the therapeutic potential of A₁R-targeted approaches (Lee et al., 2024).

A_2AR signaling promotes sleep by activating sleep-promoting GABAergic neurons in the VLPO and MnPN, suppressing TMN histaminergic output, and regulating the transition from wakefulness to sleep through striatal circuits (Ferre et al., 2007; Huang et al., 2024). In insomnia, reduced adenosine levels may limit A_2AR-mediated recruitment of these circuits, impairing suppression of arousal systems. Pharmacological studies further support this mechanism. Positive allosteric modulators (PAMs) of A_2AR enhance endogenous adenosine signaling and induce slow-wave sleep (SWS) in both wild-type mice and psychiatric disease models, including mania-like and schizophrenia-like models (Lin et al., 2023). Unlike direct A_2AR agonists, PAMs require endogenous adenosine, providing a physiologically gated mechanism that may reduce cardiovascular and off-target effects (Lin et al., 2023).

Taken together, insomnia is associated with reduced adenosinergic inhibitory drive involving both A₁R and A_2AR pathways. A₁R hypofunction weakens suppression of wake-promoting circuits, whereas reduced A_2AR signaling limits activation of sleep-promoting networks. Pharmacological strategies targeting these mechanisms, including A₁R agonists and A_2AR PAMs, represent promising alternatives to current insomnia therapies. In addition, the flavonoid luteolin exerts hypnotic effects through direct interaction with both A₁R and A_2AR, increasing NREM sleep and reducing sleep latency, suggesting that dual-target approaches may provide additional therapeutic benefit (Kim et al., 2019). However, clinical translation remains challenging due to receptor subtype selectivity, peripheral cardiovascular effects, and the complex, region-dependent actions of adenosine signaling.

Obstructive Sleep Apnea (OSA)

OSA is the most prevalent form of sleep-disordered breathing, characterized by recurrent episodes of upper airway obstruction during sleep, resulting in intermittent hypoxia, hypercapnia, and sleep fragmentation (Ma et al., 2025). Under hypoxic conditions, extracellular adenosine levels increase markedly as a protective metabolic response to cellular energy failure, driven by accelerated ATP hydrolysis and CD73-mediated AMP dephosphorylation (Ma et al., 2025). In OSA, recurrent cycles of chronic intermittent hypoxia (CIH) maintain elevated extracellular adenosine levels, thereby profoundly altering both A₁R and A_1AR signaling. Sustained adenosine exposure under CIH triggers distinct, receptor-specific adaptive changes. Prolonged A₁R activation leads to homologous desensitization and downregulation through classical GPCR regulatory mechanisms: agonist-occupied A₁Rs are phosphorylated by G protein–coupled receptor kinases (GRKs), followed by β-arrestin recruitment, clathrin-mediated internalization, and subsequent lysosomal degradation. These processes reduce surface receptor density and attenuate Gi-mediated inhibitory signaling, as demonstrated in the rat hippocampus under acute and sustained hypoxia (Coelho et al., 2006; Zhang et al., 2020). In contrast, A_1AR expression and signaling are persistently enhanced under CIH through transcriptional upregulation driven by hypoxia-inducible factor-1α (HIF-1α), which is stabilized under low-oxygen conditions and promotes increased A_1AR-driven cAMP–PKA activity in the limbic system and hippocampus (Ma et al., 2025; Zhang et al., 2020). This imbalance between A₁R and A_1AR signaling is likely to contribute to the neuroinflammatory and pro-apoptotic processes associated with OSA-related hippocampal injury. A₁R contributes to OSA pathophysiology through both peripheral and central mechanisms. At the carotid body, A₁R activation inhibits L-type Ca²⁺ channels and reduces hypoxia-induced catecholamine release, providing negative feedback that modulates chemoreflex sensitivity (Ma et al., 2025). The balance between A₁R-mediated inhibition and A_2AR-mediated excitation determines overall chemosensory output, and caffeine-induced blockade of adenosine receptors reduces hypersensitized responses under chronic hypoxia (Li et al., 2022). In the central nervous system, A₁R knockout mice lack hypoxia-induced respiratory depression, indicating that A₁R-mediated inhibition of brainstem respiratory circuits is essential for hypoxic adaptation (Heitzmann et al., 2016; Johansson et al., 2001). In addition, sleep fragmentation attenuates hypercapnic ventilatory responses through an A₁R-dependent mechanism, further impairing respiratory control (Liu et al., 2011). Despite A₁R downregulation under CIH, pharmacological activation of A₁R with GR79236 reduces spontaneous apneas in animal models, indicating retained therapeutic potential (Carley et al., 1997). A₁R signaling also exerts neuroprotective effects: selective activation with CCPA attenuates hippocampal neuronal apoptosis, preserves synaptic plasticity, and improves spatial memory under CIH conditions (Li et al., 2022).

In contrast, A_2AR signaling is persistently enhanced under CIH, particularly in limbic and hippocampal regions. A_2AR activation engages the PI3K/Akt/HIF-1α pathway, which initially promotes metabolic adaptation and suppresses inflammation during acute hypoxia (Ma et al., 2025). However, under sustained CIH, prolonged A_2AR activation becomes detrimental, promoting hippocampal neuronal apoptosis, impairing long-term potentiation (LTP), and increasing neuroinflammation and oxidative stress (Lin et al., 2023). Pharmacological evidence supports this transition: blockade of A_2AR with SCH58261, or genetic deletion of A_2AR, attenuates CIH-induced hippocampal injury and preserves synaptic integrity, whereas activation with CGS21680 exacerbates neuronal damage (Li et al., 2022).

Narcolepsy

Narcolepsy is a chronic neurological disorder characterized by excessive daytime sleepiness (EDS) and unstable transitions between wakefulness and REM sleep (Rogers and Dreher, 2002). It is classified into narcolepsy type 1 (NT1), caused by extensive loss of hypothalamic orexin (hypocretin) neurons affecting up to 90-95% of the orexin-producing population, and narcolepsy type 2 (NT2), which presents with similar symptoms but lacks a clearly defined etiology (Scammell, 2015). Loss of orexin neurons in NT1 disrupts stabilization of sleep–wake states, resulting in inappropriate transitions between wakefulness, NREM sleep, and REM sleep (Scammell, 2015). Unlike insomnia, OSA, and RLS, the role of adenosine receptor signaling in narcolepsy remains largely inferential, as most mechanistic insights are derived from non-narcoleptic or orexin-deficient models.

Adenosine signaling interacts closely with the orexin system in regulating arousal. A₁R activation strongly inhibits orexin neurons in the LH by reducing excitatory synaptic input and suppressing neuronal firing through G_i protein–dependent mechanisms (Liu and Gao, 2007). A₁R activation also inhibits voltage-dependent Ca²⁺ currents, further attenuating orexin-mediated wake-promoting output, and these effects are not reproduced by A_2A or A_2B receptor blockade, confirming A₁R as the primary mediator of adenosinergic inhibition in orexin neurons (Liu and Gao, 2007). In narcolepsy, where orexin neurons are severely depleted, residual A₁R-mediated inhibition may further weaken the already compromised orexinergic arousal drive. However, in vivo evidence indicates a dissociation between adenosine signaling and narcoleptic sleep instability. Microdialysis studies show that basal forebrain adenosine does not accumulate normally during sleep deprivation in orexin-lesioned rats (Murillo-Rodriguez et al., 2008). Consistent with this, orexin knockout mice exhibit normal homeostatic sleep rebound and EEG delta power responses after sleep deprivation (Mochizuki et al., 2004). These findings suggest that A₁R-mediated homeostatic sleep regulation remains largely intact and functionally dissociated from the instability of sleep–wake transitions in narcolepsy.

In contrast, A_2AR signaling plays a key role in regulating arousal through striatal circuits. A_2ARs are highly expressed on striato-pallidal medium spiny neurons in the NAc and broader striatum, where tonic activation suppresses wakefulness by engaging inhibitory GABAergic projections within the indirect pathway (Huang et al., 2024; Thakkar et al., 2008). In narcolepsy, where orexin-dependent stabilization of wakefulness is impaired, this A_2AR-mediated suppression may further destabilize wake states and promote inappropriate sleep transitions. In addition, A_2AR signaling modulates dopaminergic neurotransmission through antagonistic interactions with D₂R, reducing D₂R signaling efficacy and potentially impairing dopamine-dependent maintenance of wakefulness (Ferre et al., 2023). Pharmacological evidence indirectly supports this mechanism: caffeine, a non-selective A₁R and A_2AR antagonist, has long been used as a wake-promoting agent in narcolepsy, and its alerting effects are mediated primarily through A_2AR antagonism (Reichert et al., 2022).

Modafinil and its R-enantiomer, armodafinil, are currently the most widely prescribed wake-promoting agents for excessive daytime sleepiness (EDS) in both NT1 and NT2 (Mann et al., 2026; Scammell, 2015). The primary mechanism of modafinil involves inhibition of the dopamine transporter (DAT), resulting in increased extracellular dopamine levels and subsequent activation of downstream arousal systems, including noradrenergic and histaminergic pathways (Hersey and Tanda, 2024; Minzenberg and Carter, 2008; Szabadi, 2025). Unlike amphetamine-class stimulants, modafinil does not induce substantial monoamine release and is associated with a lower abuse potential. Although adenosinergic mechanisms have been implicated in sleep–wake regulation, it remains unclear whether selective modulation of A₁R or A_1AR can recapitulate or enhance the wake-promoting effects of modafinil in narcolepsy, highlighting the need for further investigation in orexin-deficient models.

Taken together, the role of adenosine receptor signaling in narcolepsy remains largely indirect. A₁R-mediated inhibition of orexin neurons and A_2AR-mediated suppression of striatal arousal circuits may each contribute to arousal dysregulation, but direct experimental evidence is limited. The lack of normal adenosine accumulation during sleep deprivation and the preservation of homeostatic sleep responses in orexin-deficient models indicate that adenosinergic homeostatic mechanisms are functionally dissociated from narcoleptic sleep instability (Mochizuki et al., 2004; Murillo-Rodriguez et al., 2008). Although A_2AR antagonism has been proposed as a potential therapeutic strategy to reinforce wakefulness, no receptor-selective adenosinergic agents have been directly evaluated in established narcolepsy models. Future studies using receptor-selective pharmacological or genetic approaches in orexin-deficient models are required to determine the therapeutic potential of A₁R- and A_2AR-targeted interventions.

Restless Legs Syndrome

RLS is a common sensorimotor disorder characterized by an urge to move the legs with discomfort, worsening during rest and at night (Ferre et al., 2019). Symptoms often include periodic limb movements during sleep (PLMS) and nocturnal hyperarousal, reflecting related mechanisms: impaired sensorimotor integration and disrupted sleep–wake homeostasis (Ferre et al., 2019). Current evidence suggests that brain iron deficiency (BID) and altered adenosine signaling form a linked pathological pathway, with BID inducing a hypoadenosinergic state (Ferre et al., 2019).

BID reduces the A₁R/A_1AR ratio in VGLUT1-positive corticostriatal terminals, while VGLUT2 thalamostriatal terminals remain unaffected, indicating pathway-specific changes (Ferre et al., 2017, 2023). A₁Rs are downregulated in cortex and striatum alongside D₂R downregulation, whereas A_1AR upregulation occurs in severe BID (Earley et al., 2017; Quiroz et al., 2016). This imbalance shifts glutamatergic signaling toward facilitation, impairing sensorimotor filtering. Reduced A₁R signaling also enhances A_2AR activity, further increasing glutamate release (Ferre et al., 2017, 2023). In vivo, BID rats show abnormal glutamate release after low-frequency stimulation, which is reproduced by A₁R antagonism and blocked by dipyridamole (Ferre et al., 2017). This glutamatergic activity activates cholinergic interneurons and drives dopamine release, producing a presynaptic hyperdopaminergic state despite D₂R downregulation (Ferre et al., 2017, 2019).

A₁R downregulation in cortex and arousal systems may also explain hyperarousal and impaired sleep onset (Ferre et al., 2019), as reduced A₁R signaling weakens inhibition of wake-promoting neurons and increases nocturnal wakefulness. Postsynaptic A_2ARs are upregulated in striatopallidal neurons and form heteromers with dopamine D₂ long-form receptors (D₂LR), where A_2AR activation reduces D₂LR signaling (Ferre et al., 2023). However, postsynaptic A_2AR blockade may worsen symptoms by enhancing D₂LR activity. SCH-442416 has been identified as a selective antagonist with lower affinity for postsynaptic A_1AR–D₂LR complexes (Ferre et al., 2023).

Clinically, dipyridamole, an ENT1/ENT2 inhibitor, improves RLS symptoms by increasing extracellular adenosine. It reduces International Restless Legs Syndrome Study Group rating scale (IRLS) scores and the periodic limb movement index (PLMI) and increases total sleep time and SWS in both open-label and randomized studies (Garcia-Borreguero et al., 2018, 2021). However, small sample sizes limit interpretation. Overall, BID-induced hypoadenosinergic signaling—characterized by A₁R downregulation and A_2AR upregulation—links iron deficiency to RLS symptoms and supports adenosine-based therapeutic strategies.

CONCLUSIONS

This review demonstrates that A₁R and A_2AR serve functionally distinct yet complementary roles in the regulation of sleep homeostasis. A₁R primarily modulates homeostatic sleep pressure through regulation of SWA and suppression of wake-promoting circuits, whereas A_2AR facilitates the transition from wakefulness to sleep by activating sleep-promoting networks. This functional division provides a mechanistic framework for understanding how receptor-specific dysfunction contributes to distinct sleep disorder phenotypes.

Across the four disorders examined, a common pattern emerges in which the direction of adenosinergic dysregulation determines pathological outcomes. Hypoadenosinergic states, such as increased ADA activity in insomnia or receptor imbalance in RLS, impair sleep-promoting drive. In contrast, hyperadenosinergic states, as observed in OSA, induce A₁R downregulation while chronically overactivating A_2AR, shifting receptor signaling from protective to maladaptive. In narcolepsy, the adenosinergic system remains largely intact, but its interaction with the depleted orexin system renders the contribution of each receptor subtype to sleep instability largely inferential.

This disorder-specific understanding of receptor dysregulation directly informs therapeutic strategies. Rather than broadly targeting adenosine signaling, effective interventions require receptor-selective and directionally appropriate approaches tailored to each condition. However, several challenges remain, including receptor subtype selectivity, peripheral cardiovascular effects, and the region-specific complexity of adenosinergic signaling across sleep–wake circuits. Future studies focusing on A₁R- and A_2AR-mediated mechanisms, particularly using cell-type-specific approaches, disease-relevant animal models, and translational biomarkers, will be essential to refine mechanistic insight and support the rational development of targeted therapies for sleep disorders with unmet clinical needs.