|Home | About | Journals | Submit | Contact Us | Français|
Adenosine is a modulator of brain function uniquely positioned to integrate excitatory and inhibitory neurotransmission. The past few years brought a wealth of new data fostering our understanding how the adenosine system is involved in the pathogenesis of neurological diseases. Thus, dysregulation of the adenosine system is implicated in epileptogenesis and cell therapies have been developed to locally augment adenosine in an approach to prevent seizures. While activation of inhibitory adenosine A1 receptors is beneficial in epilepsy, chronic pain and cerebral ischemia, inhibition of facilitatory A2A receptors has profound neuroprotective effects, which are currently exploited in clinical trials in Parkinson’s disease. A new era of adenosine-based therapies has begun, with the prospect to cover a wide range of neurological diseases.
The purine ribonucleoside adenosine controls many brain functions in physiological and pathophysiological conditions via activation of high-affinity A1 or A2A, low-affinity A2B, or low-abundance A3 adenosine receptors (ARs) [1,2]. Coupling of the receptors to either inhibitory (A1, A3) or stimulatory (A2A, A2B) G-proteins and differential spatial distribution within the brain  allow a high degree of complexity in the effects of adenosine and permit the modulation of other neurotransmitter or modulator systems . Due to these multifaceted properties of adenosine, an adenosine-based pharmacopoeia has been established for a variety of conditions . This review covers literature from the past two years, focusing on selected newer trends on the role of adenosine in neurological disease and translation of recent research findings into adenosine-based therapies.
Inhibitory neuromodulation by adenosine is largely mediated by activation of A1Rs that are coupled to inhibitory Gi or Go containing G-proteins [1,2]. The consequences of A1R activation are stimulation of adenylyl cyclase, activation of inwardly rectifying K+ channels, inhibition of Ca2+ channels and activation of phospholipase C. As a net result, the release of various neurotransmitters, in particular glutamate, dopamine, serotonin and acetylcholine, is inhibited. Accordingly, preclinical studies have demonstrated that activation of A1Rs has profound antiepileptic  and neuroprotective  functions.
The highest expression of A2ARs is found in striatal neurons, which are involved in the control of motor function and habit formation . Inhibitors of the A2AR have profound neuroprotective functions  and prevent apoptosis . Recent findings indicate that synaptic activation of A2ARs can subsequently downregulate A1Rs or its responses . An additional layer of complexity in adenosine receptor function is created by heteromerization with each other (e.g. A1/A2A heteromers) or with other neurotransmitter receptors (e.g. A2A/D2 heteromers) . Thus, adenosine, by its capability to activate receptors with opposing functions (A1 versus A2), is uniquely positioned as an upstream regulator for the fine-tuning and integration of excitatory and inhibitory functions within the CNS.
Recent findings indicate that synaptic levels of adenosine are largely regulated by astrocytes [10–14]. The elegant experiments from Phil Haydon’s group have documented that – under physiological conditions – the major source of synaptic adenosine is derived from astrocytic vesicular release of ATP, followed by extracellular degradation to adenosine via a cascade of ectonucleotidases [15••]. Using transgenic mice with a defective astrocytic vesicular release system for ATP, the authors demonstrated a reduction of the tonic suppression of neuronal networks [15••]. Intracellularly, adenosine is rapidly metabolized by phosphorylation to AMP via adenosine kinase (ADK), the key enzyme of adenosine metabolism , which, in adult brain, is predominantly expressed in astrocytes . Thus, ADK is ideally located to control the astrocyte-based adenosine cycle (Fig. 1) by driving the influx of adenosine into the cell via bi-directional nucleoside transporters .
Given the complex nature and ubiquitous distribution of the adenosine system, any imbalance is expected to lead to neurological disease. The following sections describe recent findings on the role of the adenosine system in neurological diseases (Table 1).
Adenosine is an endogenous anticonvulsant . The past few years have brought new insights in our understanding that dysfunction of astrocyte-mediated gliotransmission , and in particular dysfunction of the adenosine system [16,18] can cause seizures , or augment the spread of kainic acid- or traumatic brain injury- induced seizures and associated neuronal cell loss [18,19]. The adenosine system is critically involved in regulating proliferation and hypertrophy of astrocytes leading to astrogliosis , a hallmark of epilepsy. Consequently, changes in the homeostasis of the astrocyte-based adenosine cycle are to be expected in epileptic brain. Thus, upregulation of the astrocyte-based enzyme ADK has been identified as a consequence of astrogliosis, leading to a reduced tone of the endogenous anticonvulsant adenosine . During epileptogenesis, upregulation of ADK coincides with the emergence of spontaneous chronic seizures . Likewise, high neuronal expression of ADK in the immature mouse brain  may explain increased susceptibility to hypoglycemia-induced seizures in the immature brain . The hypothesis that upregulated ADK might be a culprit of epileptogenesis is further supported by transgenic mice overexpressing ADK in brain; these animals display reduced seizure thresholds and spontaneous seizures [22•]. These findings suggest, that local restoration of the inhibitory adenosine tone should prevent seizures. Indeed, intraventricular implants of cells engineered to release adenosine, prevent seizures in a rat model of epilepsy . More recently, it was demonstrated that intrahippocampal implants of ADK deficient stem cell derived neural precursors suppress kindling epileptogenesis [23•]. These findings suggest, that focal augmentation of the adenosine system has the potential not only to suppress seizures, but also to prevent epileptogenesis. Thus, adenosine-augmenting cell and gene therapies might lead to improved treatment options for patients suffering from intractable epilepsy .
A rapid increase in adenosine as an acute response to cerebral ischemia [25,26••] likely represents an endogenous neuroprotective mechanism, which might be involved in the initial phase of ischemic preconditioning leading to a cerebroprotective state known as ischaemic tolerance . This is a complex phenomenon, in which reprogramming of the genome  and astrocytes appear to play key roles . Recently, the astrocytic enzyme ADK was shown to be downregulated within a time frame of 3 hours following middle cerebral artery occlusion (MCAO) in the mouse . As expected, downregulation of ADK led to a marked increase in ambient adenosine, which might be a mechanistic cue for the acute endogenous protective response of the brain to injury . Indeed, ADK expression levels appear to play a key role in regulating neuronal vulnerability. Transgenic mice overexpressing ADK are characterized by increased susceptibility to MCAO-induced cell death . Conversely, transplantation of ADK-deficient neural or glial progenitor cells into the striatum of mice one week prior to MCAO, led to a striking reduction of the injured brain volume . These findings indicate that susceptibility to ischemia-induced cell death is tightly balanced by the tone of ambient adenosine.
Adenosine has complex effects on pain and can either be pro- or antinociceptive depending on the site of application or the receptor(s) involved . Most attention has been devoted to pain pathways involving the A1R, which dominates in mediating the antinociceptive effects of adenosine, as has been documented in animal models of neuropathic and inflammatory pain . Apart from direct administration of adenosine or A1 agonists, ADK inhibitors have been considered for the treatment of chronic pain . More recently, antagonism of A2ARs has been suggested as a novel strategy for the management of chronic pain . Thus, the A2A selective antagonist SCH 58261 produced antinociception in two acute thermal pain tests . Following intraplanar formalin injection, A2AR knockout mice displayed reduced biting and flinching responses . These findings suggest that the A2AR has significant potential as a novel target for pain control.
Due to its high abundance in striatopallidal neurons, the adenosine A2AR has emerged as the prime non-dopaminergic target in the search for improved therapeutics for Parkinson’s disease (PD) [38•]. Most importantly, A2ARs can form functional heteromeric complexes with dopamine D2Rs and A1Rs, and thus have the capability to integrate and fine-tune neurotransmission in the basal ganglia [9,39,40]. Due to reciprocal antagonistic interactions between A2A and D2Rs, the blockade of the adenosine A2AR in striatopallidal neurons leads to a reduction of the postsynaptic effects of dopamine deficiency and thus to an amelioration of the motor deficits in PD . In addition, antagonism of the A2AR has direct neuroprotective effects, with the potential to slow down disease progression [38•]. Consequently, A2A antagonists have been developed for clinical trials in PD. The A2A antagonist KW-6002 has shown potential in phase IIb and III clinical trials in patients in an advanced stage of PD . The initial clinical results are promising and illustrate a translational drug discovery approach, which is based on the discovery of A2A/D2 interactions.
Epidemiological studies indicate that caffeine consumption is inversely correlated with the incidence of Alzheimer’s disease (AD) . In line with these observations, long-term caffeine treatment in AD transgenic mice resulted in improved reference and working memory, and in reduced brain β-amyloid production . Likewise, caffeine and selective A2AR antagonists prevented β-amyloid induced cognitive deficits in mice . The potential of adenosine-based therapies for AD is just emerging and – in contrast to PD – has not yet been translated into clinical trials.
Similar to its role and therapeutic potential in PD and AD, the A2AR has gained recent interest in the pathophysiology of Huntington’s disease (HD) . Since A2ARs stimulate glutamate outflow and inflammatory gliosis, it is of interest to note that A2ARs are mainly localized on those neurons, which degenerate early in HD; consequently, changes in A2AR expression and signalling have been reported in models of HD, leading to the concept that the aberrant A2AR phenotype may represent a novel biomarker for HD . This conclusion is supported by findings in a mouse model of HD, demonstrating a transient increase in A2AR density and A2AR-dependent cAMP production at early presymptomatic ages, i.e. well before the onset of symptoms . These findings suggest that modulation of A2ARs might be an interesting therapeutic strategy for HD.
Currently, schizophrenia is best explained by the dopamine and glutamate hypotheses of schizophrenia. However, each hypothesis alone can only incompletely explain the complex pathophysiology of schizophrenia. The newly proposed adenosine hypothesis of schizophrenia [46•] might constitute a critical link between the two hypotheses.
The dopamine hypothesis of schizophrenia is largely based on effects of dopamine receptor antagonists and agonists to suppress and to promote psychotic symptoms, respectively. Thus, the efficacy of many antipsychotics correlates with their ability to block dopamine D2Rs [47,48]. In addition, increased dopamine turnover, and enhanced amphetamine-induced dopamine-release have been documented in schizophrenia patients [46•]. These findings are compatible with a deficiency in adenosine, which can enhance dopamine release due to the inhibitory effect of A1Rs on dopamine release , and potentiate amphetamine-induced locomotion and dopamine release as suggested by effects of A1R antagonists . Interactions between A2ARs and D2Rs allow further opportunity for mutual modulation between the adenosine and dopamine systems. These mechanisms could provide a rationale for an antipsychotic-like profile of adenosine receptor agonists.
According to the glutamate hypothesis, reduced N-methyl-D-aspartate receptor (NMDAR) function may contribute to the cognitive and negative symptoms of schizophrenia . NMDAR blockade can give rise to behavioral dysfunction and psychotic-like behavior, while co-agonists of the NMDAR, such as D-serine and glycine, improve cognition and negative symptoms in schizophrenia . Similarly, disruption of the glycine transporter 1 has promnesic and antipsychotic potential . Adenosine inhibits glutamate release and the post-synaptic action of excitatory neurotransmitters upon neuronal hyperpolarization via A1Rs . Thus, in animal models of schizophrenia, AR agonists were effective against behavioral as well as neurophysiological effects induced by NMDAR antagonists, thus indicating their potential antipsychotic efficacy in humans [46•].
The examples outlined above in a variety of neurological disorders indicate that adenosine concentrations need to be under tight control. Any increases or decreases in ambient adenosine above or below certain thresholds is expected to lead to characteristic pathologies via an imbalance of adenosine receptor mediated secondary effects. Due to interactions with glutamatergic and dopaminergic neurotransmission, adenosine can be regarded as a “master regulator” to integrate and fine-tune various neurotransmitter systems in the brain (Fig. 2). ADK and astrocytes as the key regulators of synaptic adenosine have therefore important upstream functions. Thus, therapeutic modulation of ADK has great promise for future treatment strategies for the wide range of pathologies discussed here (Table 1). To further promote ADK-based therapies, a detailed understanding how the enzyme is regulated in vivo is urgently needed.
The work of the author was supported by grant R01 NS047622-01 from the NIH, by the Good Samaritan Hospital Foundation and by the Epilepsy Research Foundation through the generous support of Arlene & Arnold Goldstein Family Foundation.
Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
Papers of particular interest, published within the period of review, have been highlighted
• as of special interest
• • of outstanding interest