Priyattam J. Shiromani - US grants

Narcolepsy is a debilitating neurological sleep disorder in which the primary symptoms are excessive daytime sleepiness, sleep attacks, frequent and sudden attacks of cataplexy, and hypnagogic hallucinations. The disease appears to be a disorder of REM sleep because some narcoleptic symptoms are components of REM sleep which are somehow encroaching into the waking state. Currently, it is hypothesized that the inappropriate intrusions of REM components are due to hyperactive central cholinergic mechanisms influencing excitability of pontine neuronal mechanisms controlling the various components of REM sleep. We are proposing a combination of neuroanatomical, neurophysiological and receptor binding studies to examine more closely the mechanisms by which the brainstem cholinergic system influences REM sleep. In the first experiment we will use a combination of retrograde (WGA-HRP and Fluorogold, a fluorescent retrograde marker), anterograde (triated amino acid transport) and choline acetyltransferase immunolabelling techniques to determine the source of the cholinergic input to the medial pontine reticular formation, an area which has traditionally been implicated in REM sleep generation. In the second experiment we will monitor the sleep-related discharge of the brainstem cholinergic neurons located in the pedunculopontine and lateral dorsal tegmental nuclei. Immunohistochemical ChAT labelling and antidromic studies will serve to strengthen the possibility that cholinergic neurons were recorder. Thirdly, we will conduct a series of experiments to determine whether a genetic strain of rats with cholinergic hyperactivity demonstrate increased REM sleep. In these studies we will correlate REM sleep levels with muscarinic receptor binding in several brain regions including the brainstem. These studies will enable us to examine the receptor mechanisms involved in REM sleep generation.

Narcoleptic patients suffer from excessive daytime sleepiness, cataplexy, hypnagogic hallucinations and sleep attacks. Another identifying feature of narcolepsy is th entry into REM sleep directly from wakefulness. These symptoms are serious enough to disrupt the patients will being and may also be responsible for life threatening accidents. The disease appears to be a disorder of mechanisms controlling REM sleep because some narcoleptic symptoms, such as cataplexy and hypnagogic hallucinations are components of REM sleep, and in narcolepsy these components inexplicably encroach into the waking state. there is considerable evidence that acetylcholine plays an important role in triggering REM sleep, and the studies proposed in this grant application focus on delineating the cholinergic receptor subtype and neuronal elements related to this phase of sleep. The first experiment will serve to clarify the cholinergic receptor pharmacology involved in REM sleep generation. Pharmacological agents specific to the M1 and non-M1 receptors and nicotinic agents will be infused into the medial pontine reticular formation (mPRF), an area which has been shown to be important for REM sleep generation. The changes in REM sleep will be noted and comparisons between muscarinic and nicotinic mechanisms will be made. The second study will examine the glutaminergic pathway in the pons that along with the cholinergic pathway might represent the final common pathway involved in the atonia of REM sleep. Different retrograde tracers will be injected into the mPRF and the caudal medulla and the presence of the tracer will be examined in chemically identified glutaminergic somata in the dorsolateral pontine tegmentum. The third study will utilize fos-immunohistochemistry to reveal neuronal elements in the brainstem and forebrain related to REM sleep. The fos gene is triggered in response to neuronal stimulation and the presence of the protein allows for mapping of neuronal pathways. The protein might act as a third messenger serving to regulate the functional response of the cell. These studies will establish a groundwork for future work related to the transcriptional processes occuring during REM sleep.

DESCRIPTION: (Applicant's abstract) Narcoleptic patients suffer from excessive daytime sleepiness, cataplexy, hypnagogic hallucinations and sleep attacks. Other identifying feature of narcolepsy is the entry into REM sleep directly from wakefulness. The disease appears to be a disorder of mechanisms controlling REM sleep because some narcoleptic symptoms, such as cataplexy and hypnagogic hallucinations are components of REM sleep, and in narcolepsy these components inexplicably encroach into the waking state.There is considerable evidence that acetylcholine plays important role in triggering REM sleep, and we have contributed some of this data. Specifically, it is now clear that a group of cholinergic neurons in the lateral dorsal tegmental and pedunculopontine tegmental Pffl nuclei are important for REM sleep onset. We now have data that a subset of these cholinergic neurons show immediate- early gene expression, as assessed by the presence of the c-fos protein, Fos, in conjunction with cholinergically-induced REM sleep. The function of c-fos expression in these neurons is unclear at this time but, among various possibilities, such cascades, in pools of REM sleep related neurons, might reset the "need" for REM sleep. For instance, narcoleptics are urged to take periodic naps and this is extremely effective in alleviating their excessive sleepiness and cataplectic attacks. The first series of experiments will determine where immediate early gene expression occurs and under what circumstances. This is necessary in order to accomplish the long-term goal of determining what is being coded during sleep and REM sleep. Fos-like immunoreactivity (Fos-LI) will be examined in response to pharmacological manipulations of REM sleep. Microinjections of cholinergic drugs and the noradrenergic alpha- I antagonist, idazoxan, will be made into the medial pontine reticular formation (mPRF), an area which interacts with the LDT-PFT cholinergic neurons to produce REM sleep. Fos-L1 will be examined in neurotransmitter specific neurons. In another study, the role of NMDA and non-NMDA receptors in the LDTP~ will be assessed by examining changes in REM sleep and Fos-LI in neurotransmitter specific neurons. The second series of experiments will examine the age-related development of muscarinic receptors and choline acetyltransferase, the acetylcholine synthesizing enzyme, in a line of rats bred for increased cholinergic function. The changes in cholinergic markers will be correlated with development of REM sleep. Cur studies, in the adult rats, have shown that compared to age-matched controls, the cholinergic rats show 35% more REM sleep, 22% more LDT-P~ cholinergic soma, and II more pontine muscarinic receptors.Collectively, these studies are a logical next step in our efforts to understand how changes in the cholinergic system influence REM sleep.

DESCRIPTION (Adapted from applicant's abstract): The preoptic area has long been identified as a region that is important for induction of sleep. It contains populations of sleep-active neurons, and lesion in this area cause long-lasting insomnia. However, the circuitry that is critical for inducing sleep has never been identified. We recently identified as specific cell group, the VLPO, that contains neurons that show activation, as monitored by expression of the immediate early gene (IEG) product Fos, specifically during sleep. We hypothesize that these sleep-related neurons in the VLPO innervate the main components of the ascending arousal system in the brainstem, providing profound inhibition that allows the onset of sleep. We plan to test this hypothesis with four sets of experiments: (1) We will trace the efferent connections of the VLPO using a combination of anterograde tracers and double retrograde fluorescent tracing with immunocytochemistry, to determine whether sleep-active neurons in the VLPO innervate monoaminergic and cholinergic brainstem cel groups, whether these descending projections are collateralized or arise from separate neuronal populations, and whether the neurons participating in this projection use GABA and galanin as neurotransmitters, (2) We will trace the afferents to the VLPO and the neurotransmitters they employ, to determine the inputs that drive the sleep-active neurons, (3) We will examine the effects of unilateral lesions of the VLPO on ipsilateral EEG delta (0.1-4 Hz) power. Finally, (4) we will determine whether unilateral stimulation of the VLPO with endogenous sleep-inducing substances, including prostaglandin D2, adenosine, delta sleep-inducing peptide, or a sleep-inducing lipid, produces ipsilateral EEG slowing and VLPO Fos expression. These experiments will provide critical evidence to determine whether the VLPO has the appropriate connections and physiological responses to support a major role in inducing sleep.

DESCRIPTION: Sleep and wakefulness are regulated for the most part by circadian and homeostatic processes. The homeostatic process is more powerful and can override circadian influences. The nature of the homeostatic process is unknown, but might be molecular in nature. Recent studies with the immediate-early gene, c-fos, have shown that this gene is expressed differentially in discrete brain regions in response to sleep and wakefulness. These findings have led to the hypothesis that c-fos might be representative of a molecular cascade that transduces a signal involving sleep. If this cascade is compromised, as with the application of c-fos antisense (which blocks new c-Fos protein synthesis), or in animals lacking the c-fos gene (null c-fos), then there should be less sleep. Our recent findings support this hypothesis. The proposed studies will examine whether the diminished sleep in aging might, in part, result from a decline in c-fos to AP-1 binding. Four specific aims will directly test this hypothesis. Specific aim 1 will test the hypothesis that old rats (24 months) show less sleep and slow wave activity (0.3-4.0 Hz) for a given amount of prior wakefulness compared to young rats (2 months). Rats will be kept awake (by gentle handling) for various time periods (0, 6, 12 hr) and then allowed recovery sleep. The investigators predict that young rats will have more sleep, including increased slow wave activity (0.3-4 Hz), compared to old rats. Specific aim 2 will test the hypothesis that old (24 months) rats have reduced c-Fos and AP-1 binding in wake-active populations in the basal forebrain in response to wakefulness compared to young (2 months) rats. Rats will be kept awake as in Specific aim 1 and killed following 6, 12 hr of wakefulness. Immunohistochemistry and western blot assays will qualify c-fos levels. Gel-shift assays will examine the AP-1 complex in young vs. old rats. Specific aim 3 will test the hypothesis that old rats demonstrate a reduced response to adenosine in the basal forebrain compared to young rats. Adenosine will be dialyzed in the basal forebrain in young and old rats and the increase in sleep will be measured. The investigators predict that for the same dose of adenosine, older rats will show less sleep compared to young. Specific aim 4 will test the hypothesis that adenosine induces c-fos expression and resultant AP-1 binding in the basal forebrain. The investigators predict that the c-fos induction and AP-1 binding are much stronger in young versus old rats. Their preliminary results indicate that somnogens such as adenosine induce c-Fos (measured using immunohistochemistry and on tissue slices, and via western blot analysis of basal forebrain tissue homogenate) and AP-1.

DESCRIPTION (Adapted from applicant's abstract): This is an application for a K02 Scientist Research Award. The applicant's research on the neurobiology of sleep has progressed from single-cell recording of pontine neurons to in-vivo microinjection to anatomical networking related to REM sleep, and now intracellular molecular changes associated with sleep. Data from these studies have begun to shape a first-order map of sleep-wake control. To date the PI has concentrated on determining a circuit diagram of sleep-wake control because then investigators would know where to look for molecular changes. The PI proposes to continue his studies to define the neural circuitry, but also extend these studies to include the intracellular molecular events. Thus, it will be possible to understand sleep at both the network and single cell level. The proposed career plan consists of neuroanatomy studies under the guidance of Dr. Clifford B. Saper. The PI will utilize anterograde and retrograde tracers in conjunction with double and single labelling immunohistochemistry to identify the neural circuitry underlying sleep-wakefulness. The applicant has never used anterograde tracers (because of unfamiliarity with microiontophoretic injection and labelling procedures) in his previous studies, so the training is necessary. Newer retrograde tracers are now available which the PI will learn to use. Training will also involve cell counting and tracing procedures, morphometric analysis, and computer digitization and plotting techniques. The PI will also learn in-situ hybridization (ISH) and other molecular procedures (northern blot analysis, PCR, gel shift assays). These procedures will be used because in studying sleep, where non-REM and REM sleep last for only a few minutes, it is necessary to monitor markers that are rapidly induced. Protein markers (such as c-Fos) require the animal to remain in a state for at least 30-60 min so that sufficient levels of the protein are produced to be visualized using immunohistochemical procedures. The applicant has never used ISH or other molecular procedures. However, clearly the studies are going in that direction. The PI will be trained in the molecular studies by Dr. William Schwartz.

A pronounced decline in sleep occurs with aging. This deficit in sleep with aging presents a major problem for elderly individuals, causing daytime sleepiness and impairing alertness. However, the mechanism for this change in the sleep cycle with aging is not understood. In the previous cycle of this Program Project Grant, the hypothesis was explored that the deficit in sleep in aged individuals was due to a deterioration in the circadian pacemaker, i.e, suprachiasmatic nucleus (SCN). However, results from Czeisler's group indicate that older subjects show no change in period of the temperature rhythm. These findings raise the possibility that there is no change in the SCN's pacemaking ability, but that other coupling pathways might decline with age. One possibility is that homeostatic elements regulating sleep- wakefulness might be impaired with aging. Our studies will directly address the hypothesis that the defect in sleep caused by aging is a lack of responsiveness in the ventrolateral preoptic nucleus (VLPO) neurons to the accumulation of adenosine. Specific aim 1 will test the integrity of the VLPO neurons in young and old rats. We predict that in older animals for a given amount of prior wakefulness there is less c-fos expression in VLPO. This would indicate levels in VLPO neuron. We predict that for a given amount for prior wakefulness older rats will have less galanin mRNA levels in VLPO compared to young rats. Specific aim 2 will test the hypothesis that there may be reduced responsiveness of VLPO neurons to accumulation of adenosine because of a reduction in expression of adenosine receptors by VLPO neurons or their afferents during aging. Specific aim 3 will test the hypothesis that preoptic neurons themselves are hyporesponsive to adenosine in aged animals.

DESCRIPTION (provided by applicant): Narcolepsy has now been linked to a lesion of neurons containing the neuropeptide hypocretin (HCRT), also named orexin. However, the loss of these neurons can only be determined upon autopsy. Cerebrospinal fluid (CSF) measurements of HCRT represent a potential marker of an underlying pathology, i.e., lesion of HCRT neurons. However, for this marker to be clinically significant it is vital that we determine the relationship between number of HCRT neurons and CSF HCRT levels. Narcoleptic patients have reduced CSF levels of HCRT, a finding consistent with the loss of HCRT neurons. However, new data from Mignot's group are beginning to show that in about 10 percent of narcoleptics the CSF levels of HCRT are normal. This raises important questions about loss of HCRT neurons, and the utility of CSF levels of HCRT as an assay to diagnose narcolepsy. For instance, if a human narcoleptic has normal CSF HCRT levels does it mean that HCRT neurons are intact? Or does it suggest that the disease has not progressed to a point where a certain threshold number of HCRT neurons have not been destroyed? Because such questions cannot be easily addressed using current murine knockout, rat knockout or canine models of narcolepsy, we will utilize the hypocretin-saporin (HCRT2-SAP) neurotoxin method to lesion hypocretin and adjacent lateral hypothalamic neurons and monitor sleep and CSF HCRT-1 levels in rats. We have designed a set of aims to provide critical data to the clinician, but also provide a framework for integrating the hypocretin neurons within an overall model of sleep regulation. Specific aim 1 will test the hypothesis that a significant decrease in CSF HCRT levels is evident only after a threshold number of HCRT neurons are lost. Specific aim 2 will determine the association between CSF HCRT levels and sleep. Specifically, we will investigate whether narcoleptic-like behavior, characterized by sleep onset REM sleep periods (SOREMPs), increased REM sleep at night, and hypersomnia is evident ahead of a decline in CSF HCRT levels. Specific aim 3 will test the hypothesis that loss of a specific population of HCRT neurons is responsible for the symptoms of narcolepsy. Specific aim 4 will determine the afferents and efferents of this population of HCRT neurons.

How the brain regulates states of consciousness is not known, but first-order circuits detailing interactions between wake-active and sleep-active neurons have been proposed. Here we focus on a portion of the wake circuit that involves the basal forebrain (BF) and link it to the neuropeptide hypocretin (HCRT). The basal forebrain (BF)has been implicated in regulating wakefulness (Szymusiak, 2000), and contains neurons that are preferentially wake/REM-active or sleep-active (Szymusiak, 2000). The sleep-active neurons are GABAergic (Manns et al.,-2003) and could oppose the wake-active neurons (Jones, 2004). The wake-active neurons are thought to be driven by ascending influences from other arousal populations (Semba, 2000), and are hypothesized to shut-off by the localized build-up of adenosine (AD) (Porkka- Heiskanen et al., 1997). Although in the BF adenosine levels rise with wakefulness and then fall with sleep, would adenosine levels rise with prolonged waking if the cholinergic neurons were absent? This is an important question that surprisingly has never been investigated, but which can easily be answered by using 192-lgG-saporin to lesion the BF cholinergic neurons and measuring adenosine in the BF that is devoid of cholinergic neurons. Here we propose four aims utilizing overlapping methodologies and transgenic rats to test a specific circuit. Aim 1 will test the hypothesis that in the absence of the BF cholinergic neurons adenosine will not build in the BF in response to waking. Aim 2 will test the hypothesis that in the absence of the BF cholinergic neurons adenosine or the adenosine A1 receptor agonist CHA will not induce sleep. Aim 3 will link the HCRT system with the BF cholinergic system by demonstrating that ascending influences from this prominent arousal system drives the BF. We will directly test this possibility by demonstrating that in 192-lgG sap lesioned rats HCRT will be less effective in evoking wakefulness. In aim 4 we will utilize the ataxin-HCRT transgenic rats to test this circuit. We will measure AD levels in the BF of the transgenic rats and hypothesize that in response to prolonged waking they will be lower compared to wild-type control rats, since the HCRT neurons are lost and not driving the BF neurons. Then 192-lgG saporin will be used in the transgenic rats to lesion the BF cholinergic neurons and we hypothesize that with both the BF cholinergic and HCRT neurons lost, the rats should have more sleep compared to single lesions.

DESCRIPTION (provided by applicant): Narcolepsy is now considered a neurodegenerative disorder and as with other diseases where CNS neurons die it is necessary to explore new strategies to transfer genes to restore function. Here we propose developing a gene transfer approach that will serve as a neurobiological tool to understand the networking underlying narcolepsy and also to ultimately reverse symptoms. We have created a replication-defective HSV-1 amplicon vector to transfer the gene for mouse preprohypocretin together with reporter genes into hypocretin null mice. Our very strong preliminary data shows abundant and robust expression of hypocretin in the lateral hypothalamus of hypocretin null mice along with unambiguous decline of narcoleptic symptoms. We are proposing an integrated series of in vitro and in vivo aims that will serve as a foundation for a more comprehensive effort to utilize the gene transfer approach to reverse the symptoms of narcolepsy in hypocretin null mice. Appropriate experiments with controls are proposed to strengthen the conclusions. Here, we are focusing on transferring the gene for mouse preprohypocretin because this neuropeptide can be easily identified using simple immunohistochemical procedures. We will then migrate to transferring the gene for the receptor, a much more difficult task since there is no good antibody that will allow for verification of the gene transfer. Our overall strategic intent is to transfer the gene for the hypocretin 2 receptor in canine narcolepsy, thereby replacing a mutated receptor gene with a healthy one. PUBLIC HEALTH RELEVANCE Narcolepsy is now considered a neurodegenerative disorder and it is necessary to explore new strategies to treat the disease. The significance of this project is that it will develop a gene transfer approach that will serve as a neurobiological tool to understand the networking underlying narcolepsy and also to ultimately restore some function.

DESCRIPTION (provided by applicant): Gene transfer has proven to be an effective neurobiological tool in a number of neurodegenerative diseases and we have used it to correct a sleep disorder. We have focused on narcolepsy, a neurodegenerative sleep disorder linked to the loss of neurons containing the neuropeptide hypocretin, also known as orexin. To restore orexin levels we have inserted the gene for orexin into surrogate neurons and blocked narcoleptic behavior in two reliable and valid mice models of narcolepsy. The effects were site specific and depended on the connectivity of the surrogate neurons. We now propose to further narrow the site-specificity by confining expression of orexin only in MCH neurons and selectively activating them during waking. The MCH neurons are still viable in human narcolepsy and are connected to the same downstream targets as the orexin neurons. These neurons are normally silent during waking and we hypothesize that by selectively activating them during waking we will block cataplexy and lengthen waking bouts. In aim 1 we will insert the genes for channelrhodopsin-2 (ChR2), a light sensitive cation channel, and orexin only in MCH neurons (MCH promoter driven). Optogenetic stimulation will drive the MCH-ChR2-Orexin containing neurons and its effects on cataplexy and wake duration will be determined during both the day and night cycles. Aim 2 will utilize a new emerging methodology that relies on Designer Receptors Exclusively Activated by Designer Drugs to activate the MCH neurons. Experiments with appropriate controls, including orexin receptor antagonist are proposed to strengthen the conclusions. C-Fos will identify activation of the MCH neurons and an in vitro calcium imaging study will determine functionality of the genetically inserted ChR2 and hM3Dq receptors. These studies will for the first time identify neurons that can be selectively activated to block narcoleptic behavior. PUBLIC HEALTH RELEVANCE: Current pharmacological approaches for treating narcolepsy lack specificity since the drugs bathe the entire brain and body. The proposed studies will utilize optogenetics and DREADD to identify neurons that can be selectively activated to block narcoleptic behavior and lengthen wake bouts. We believe that the gene transfer approach can serve as a methodological tool to quickly and cost-effectively identify these neurons.

DESCRIPTION (provided by applicant): We have demonstrated the first successful use of neuronal-orexin gene transfer to ameliorate symptoms of narcolepsy in two established animal models of the disease. The effects were site specific and depended on the connectivity of the surrogate neurons. Since neuronal orexin can be secreted at multiple distal sites, some of which may regulate narcoleptic behavior whereas others do not, we propose a localized gene delivery method by expressing orexin in astroglia. Astroglial-orexin will re-establish the ligand-receptor link in a localized area that already contains the orexin receptors. Astroglial-orexin gene deliver will serve as a cost-effective neurobiological tool to understand the networking underlying sleep. In aim 1 the gene for orexin will be inserted into astroglia (GFAP promoter driven) in three specific brain regions, two of which are implicated in network sleep models (TMN, pons) and the third, striatum, will serve as control. Our preliminary results in the orexin-ataxin-3 mice model indicate that in the TMN glial-orexin (rAAV-GFAP-orexin) completely rescues waking but not cataplexy. In the pons it produces a 78% reduction in cataplexy but not waking. Thus, there is site- specificity of the glial-orexin. In aim 2 astroglial-orexin will be driven by optogenetic stimulation to determine whether such stimulation further stimulates waking or blocks cataplexy, especially in ineffective sites in aim 1. Experiments with appropriate controls, including orexin receptor antagonist are proposed to strengthen the conclusions. In-vitro calcium imaging study will determine functionality of the genetically inserted ChR2 receptors. To the best of our knowledge this is the first use of astroglial-orexin in sleep disorders. Gene transfer is extremely cost-efficient, it is used clinically and it will demonstrate that genetic pharmacology can be used to elucidate the network in sleep disorders.

DESCRIPTION (provided by applicant): There is considerable amount of data on arousal neurons whereas there is a paucity of knowledge regarding neurons that make us fall asleep. Indeed, current network models of sleep- wake regulation list many arousal neuronal populations compared to only one sleep group located in the preoptic area. There are neurons outside the preoptic area that are active during sleep, but they have never been selectively manipulated. Indeed, none of the sleep-active neurons have been selectively stimulated. To close this knowledge gap the proposed studies will use optogenetics to selectively manipulate neurons containing melanin concentrating hormone (MCH). The MCH neurons are located in the posterior hypothalamus intermingled with the orexin arousal neurons. Our very strong preliminary data indicate that optogenetic stimulation of MCH neurons excites sleep active neurons, decreases activity of wake active neurons and increases both non-REM sleep (NREM) and REM sleep (REMS) in wildtype mice (J Neuroscience, 2013), MCH-Cre mice and rats. MCH neuron stimulation increases sleep during the animal's normal active period, which is compelling evidence that stimulation of MCH neurons has a powerful effect in counteracting the strong arousal signal from all of the arousal neurons. Effects of MCH neuron stimulation versus inhibition will be tested in conditions that alter the animal's internal drive to stay awake (24h fasting) or sleep (6h sleep deprivation). In one aim, electrophysiology studies will monitor activity of sleep-active or wake-active neurons in the lateral hypothalamus, preoptic area and pons during normal sleep-wake cycles and during optogenetic stimulation, thereby identifying activity at the single neuron level during natural and optogenetically induced sleep. Neuroanatomy studies will show that MCH neurons project to multiple targets indicating powerful influence of these neurons in orchestrating shifts in vigilance states. The MCH neurons represent the only group of sleep-active neurons that when selectively stimulated induce sleep. From a translational perspective this is potentially useful in sleep disorders, such as insomnia, where sleep needs to be triggered against a strong arousal drive. These aims will provide a framework for integrating the MCH neurons within an overall model of sleep-wake regulation.

Current models of sleep-wake regulation are ?neuron-centric?. However, there are astrocytes in the brain, and they outnumber neurons. A single astrocyte contacts hundreds of dendrites, and tens of thousands of synapses (Bushong et al., 2002, Halassa et al., 2007). The concept of a tripartite synapse has emerged where astrocytes actively control neuronal activity and synaptic transmission. We provided the first evidence that selectively activating astrocytes in the posterior hypothalamus increases both NREM and REM sleep in mice. This is the first time that optogenetic or DREDD activation of cells other than neurons has been shown to increase sleep. Most importantly, sleep was increased at night, the normal wake period in nocturnal mice. This indicates that astrocytes can impart a load that can induce sleep. What this means is that astrocytes can impart a load on neurons throughout the brain, thus providing for the first time an explanation for the waxing and waning of sleep. This will support the hypothesis of ?local use dependent sleep? throughout the brain. We will determine the gliotransmitter that is released, and also monitor the activity of adjacent sleep-wake neurons in response to optogenetic stimulation. Aim 1 will use microdialysis to test the hypothesis that adenosine accumulates in response to optogenetic stimulation of ChR2-positive astroglia. We are focusing on adenosine based on the substantial evidence that it is released from astrocytes and its linkage with sleep. However, the other potential gliotransmitters will also be measured. This aim will also use the adenosine A1 receptor antagonist, CPT, to block the sleep induced by optogenetic stimulation of astrocytes. Aim 2 will test the hypothesis that in response to optogenetic stimulation of ChR2-positive astroglia the arousal neurons in the posterior hypothalamus are inhibited, and this is blocked by the adenosine A1 receptor antagonist. This aim will also determine the site-specificity of the effect in mice by activating the ChR2-containing astrocytes in areas implicated in sleep-wake regulation (VLPO, basal forebrain, and dorsolateral pons). The overall impact of this project is that it mechanistically connects astrocytes, adenosine, A1 receptor, and local neuronal activity with sleep. Inclusion of astrocytes in circuit models will lead to better explanation of sleep homeostasis, which is something that current ?neuron-centric? models have failed to do.