Paul A. Rosenberg - US grants

Adenosine is an inhibitory neuromodulator that has been postulated to be involved in behavioral state control in several regions of the brain, including the cortex, thalamus, basal forebrain and lateral dorsal tegmental nucleus/pedunuculopontine tegmental nucleus (LDT/PPT). Adenosine inhibits firing of cholinergic cells of the adenosine levels in the extracellular space in any brain region. Recently it has been found that monoamines, peptides and metabotropic glutamate agonists regulate adenosine levels by a mechanism dependent upon cycle AMP metabolism. In addition, in cortical cultures, and in the hippocampal and LDT/PPT slice preparations, all derived from the rat, the cyclic nucleotide phosphodiesterase inhibitor zaprinast evokes significant extracellular adenosine accumulation. Since the LDT/PPT receives glutamatergic, monoaminergic, and peptidergic inputs, our specific hypothesis is that mechanisms linking adenosine accumulation with neurotransmitter control of cyclic nucleotide metabolism may be important in regulating extracellular adenosine in the LDT/PPT. The specific aims of this project are to: 1) determine the origin of zaprinast evoked extracellular adenosine accumulation; 2) identify phosphodiesterases present in the LDT/PPT; 3) characterize the effect of phosphodiesterase inhibitors on cyclic nucleotide metabolism and adenosine accumulation in the LDT/PPT; 4) characterize the effects of endogenous neuromodulators of cyclic nucleotide metabolism and adenosine accumulation in the LDT/PPT. This work should improve our understanding of how adenosine levels are regulated at a molecular level, how elevations in adenosine levels might be produced in brain regions involved in state control and whether such changes play a causal role in the determination of behavioral state.

Although much has been learned about the transmitter systems regulating behavioral state, we know little about the actual molecular mechanisms causing the brain to sleep and to wake up. This project is directed at defining these mechanisms. Cholinergic neurons of the laterodorsal tegrnental/pedunculopontine tegmental nucleus, which project to the basal forebrain (BF), as well as intrinsic cholinergic neurons of the BF, both contain nitric oxide synthase (NOS), and are active during waking. We found that: 1) NO evokes the release of adenosine from forebrain neurons in culture by a cGMP-independent mechanism; 2) A NOS inhibitor dialyzed into the BF suppressed NREM sleep following sleep deprivation in rats, and a nitric oxide donor dialyzed into the BF increased NREM sleep. These observations are the basis for one of two major hypothesis motivating this project, which is that NO, released during wakefulness in the BF by cholinergic neurons, is a major stimulus to the release of adenosine during waking, and thus is a key factor regulating the extracellular level of this important somnogen. In addition, we found that: 1) The high affinity cGMP degrading enzyme cyclic nucleotide phosphodiesterase (PDE) 9A is expressed in large neurons of the BF; 2) An inhibitor of cGMP degrading PDEs dialyzed into the BF increased NREM sleep. On the basis of these observations, we hypothesize that NO also has important effects relevant to the regulation of behavioral state that are mediated by the NO/cGMP signaling pathway, and are independent of the adenosine-releasing effects of NO. The specific aims of this project are to: 1) Use mierodialysis with behavioral and electroeneephalographie monitoring to test for the role of NO in homeostatic sleep regulation in the rat using NOS inhibitors and NO donors. 2) Determine the anatomic and cellular localization of the components of the NO/cGMP signal transduction system (cGMP hydrolyzing cyclic nucleotide phosphodiesterases, soluble guanylyl cyclase, NO synthase, protein kinase G) in regions of the brain relevant for sleep/wake regulation. 3) Characterize the electrophysiological effects of NO on neurons of the BF and ventrolateral preoptic nucleus (VLPO) using a basal forebrain/preoptic area BF/POA) slice preparation.

DESCRIPTION (provided by applicant): The precise regulation by glutamate transporters of glutamate concentrations in and around excitatory synapses is critical for the normal function of excitatory synapses. During cerebral ischemic injury, the third leading cause of death of adults in the United States, extracellular glutamate concentration rises, leading to excitotoxicity caused by excess activation of glutamate receptors. The long-term goal of this research is to understand how glutamate transporters regulate synaptic and perisynaptic glutamate concentrations normally, and how ischemia disrupts glutamate homeostasis to produce excitotoxicity. Although glutamate transporters are well known to be expressed in astrocytes, the identity of the glutamate transporter expressed in the presynaptic terminal of excitatory synapses was unknown, representing a major gap in our knowledge and understanding of excitatory synapses. We discovered that GLT1, previously thought to be exclusively expressed in astrocytes in the mature brain, is expressed in axon terminals in the hippocampus. A major hypothesis of this proposal is that GLT1 is the major glutamate transporter expressed in excitatory terminals throughout the forebrain. The function of GLT1 expressed in axon terminals as opposed to GLT1 expressed in astrocytes is unclear. We hypothesize that expression of GLT1 in neurons is important for the normal function of excitatory synapses, by preventing spillover onto perisynaptic glutamate receptors as well as cross-talk between excitatory synapses. We further hypothesize that the release of glutamate from excitatory terminals is important in the pathogenesis of excitotoxic injury in ischemia. We have generated a mutant mouse in which a critical exon in the GLT1 gene has been flanked with loxP sites, allowing the use of Cre-loxP recombination technology to produce cell-type specific knockouts. In this project we will compare the effects of deletion of GLT1 in astrocytes or in neurons, on glutamate homeostasis and synaptic function under normal conditions and in models of excitotoxic injury. The specific aims are to: 1) Perform phenotypical, morphological, and physiological analysis of mouse lines in which GLT1 is deleted in astrocytes or neurons. 2) Characterize the role of GLT1 expressed in different cell types in the pathogenesis of ischemic injury. 3) Determine whether GLT1 is expressed in excitatory terminals in regions outside the hippocampus. The expression of GLT1 in excitatory terminals has important implications for our understanding of the physiology of excitatory synaptic transmission, synaptic plasticity, and ischemic injury. Using mouse lines that will be produced for this project, we hope to gain important insights into the role of neuronal expression of GLT1 into the normal and abnormal regulation of glutamate at the excitatory synapse. PUBLIC HEALTH RELEVANCE: The long-term goal of this research is to understand how glutamate transporters normally regulate glutamate concentrations at and around the excitatory synapse, and how ischemia disrupts this regulation to produce excitotoxicity-death of neurons caused by excess release of glutamate and activation of glutamate receptors. We discovered that the glutamate transporter GLT1 is expressed in excitatory axon terminals, but the function of glutamate transporters in excitatory terminals, as opposed to astrocytes, is unclear. In this project we make mouse lines in which GLT1 expression is deleted in neurons or in astrocytes to compare the effects of expression of GLT1 in these different locations on glutamate homeostasis and synaptic function under normal conditions and in models of excitotoxic injury.

? DESCRIPTION (provided by applicant): Schizophrenia is a chronic, devastating, psychiatric disorder characterized attributed to abnormalities in dopamine and glutamate signaling. Excitatory circuits control the activity of dopamine neurons, and it is thought that abnormalities in these circuits produce the positive, negative and cognitive features of schizophrenia. Glutamate homeostasis refers to the control of glutamate levels in and around excitatory synapses, and we now have evidence that glutamate homeostasis might be important in controlling the circuits involved in schizophrenia. Glutamate transporters control brain glutamate homeostasis, and the major glutamate transporter in the brain is GLT-1, primarily expressed in astrocytes. Others and we have found that GLT-1 is also expressed in excitatory presynaptic terminals. To understand the function of GLT-1 expressed in neurons, we generated a conditional GLT-1 knockout mouse in which we have used synapsin-cre to accomplish the selective inactivation of GLT-1 in neurons. We have performed extensive behavioral phenotyping of this mouse, including testing responses to amphetamine, which are highly modulated by excitatory signaling and therefore likely, we thought, to be affected by glutamate dyshomeostasis. Previous work by many groups has demonstrated the phenomenon of sensitization to amphetamine, in which behavioral or neural (i.e. dopamine release) effects increase with repeated administration. Amphetamine sensitization is thought to model the cellular processes that underlie the positive symptoms of schizophrenia. Remarkably, we found that inactivation of GLT-1 in neurons produced significant decrease in the acute and sensitized locomotor responses to amphetamine. In addition, we have found improved performance of the nGLT-1 KO in novel object recognition and light-dark emergence. Defects on NOR and LDE may reflect impaired working memory and increased anxiety, components of the cognitive and negative domain of symptoms of schizophrenia. These observations have led us to hypothesize that the nGLT-1 KO may demonstrate resilience to the biochemical and circuit disturbances associated with schizophrenia. We hypothesize further that the phenotype that we observe in the nGLT-1 KO partially stems from changes in ambient glutamate within regions of the brain dependent upon neuronal GLT-1 for glutamate homeostasis. We propose to characterize these phenotypes further to establish whether GLT-1 may be valid target for therapeutic drug discovery. Toward that end, we will: 1) characterize the behavioral phenotype of the nGLT-1 KO mouse subjected to subchronic PCP administration to model symptom domains observed in schizophrenia; 2) characterize the biochemical phenotype of nGLT-1 KO mice; 3) determine the effect of neuronal knockout of GLT-1 on glutamate homeostasis in the nucleus reticularis slice preparation.