Establishing precise neuronal connections is crucial for normal brain function. In many parts of the brain, this is accomplished by refining initially diffuse neuronal connections during development. In contrast to our understanding of the mechanisms by which excitatory connections are refined, the refinement of inhibitory circuits is poorly understood. In this thesis, I investigated the refinement of inhibitory connections in the lateral superior olive (LSO), a mammalian auditory brainstem nucleus involved in sound localization. The glycinergic, and during development also GABAergic, projection from the medial nucleus of the trapezoid body (MNTB) to the LSO is tonotopically organized with a single-cell-level precision in adults. I asked whether and by what mechanisms the precise organization emerges during development. I found that the refinement of this inhibitory pathway is achieved by a large degree of functional elimination of exuberant inputs and strengthening of maintained inputs. Elimination of inputs occurred in a frequency-specific manner, resulting in a sharper inhibitory topography. The topographic refinement occurred while the MNTB inputs act excitatory, raising the possibility that depolarizing action of GABA and glycine is a mechanism underlying the rearrangement of inhibitory synaptic connections. In parallel with the elimination, synaptic responses elicited by the maintained MNTB inputs increased over 10-fold. This was due to a moderate increase in quantal size and a large increase in the number of release sites formed by each MNTB fiber. I further investigated whether olivocochlear efferent neurons that project from the brainstem back to the cochlea play a role in the topographic sharpening. I found that the refinement of the MNTB-LSO pathway is impaired in animals with compromised efferent projections. In both efferent-lesioned rats and mice lacking the ?9 nicotinic acetylcholine receptor subunit, LSO neurons received a greater number of weak MNTB inputs than in control animals. These results indicate that normal olivocochlear efferent projections are necessary for the refinement of the MNTB-LSO pathway. I discuss the possibility that the effect of efferent manipulation could be due to altered levels or temporal patterns of spontaneous activity before hearing onset.

Hox genes encode anterior-posterior identity during central nervous system development. Few studies have examined Hox gene function at lumbosacral (LS) levels of the spinal cord, where there is extensive information on normal development. Hoxd10 is expressed at high levels in the embryonic LS spinal cord, but not the thoracic (T) spinal cord. To test the hypothesis that restricted expression of Hoxd10 contributes to the attainment of an LS identity, and specifically an LS motoneuron identity, Hoxd10 was ectopically expressed in T segments in chick embryos via in ovo electroporation. Electroporations were carried out at early neural tube stages (stages 13-15) and at the onset of motoneuron differentiation (stages 17-18). Regional motoneuron identity was assessed after the normal period of motor column formation (stages 28-29). Subsets of motoneurons in transfected T segments developed a molecular profile normally shown by anterior LS LMCl motoneurons, including Lim 1 and RALDH2 expression. In addition, motoneurons in posterior T segments showed novel axon projections to two muscles in the anterodorsal limb, the sartorius and anterior iliotibialis muscles. These changes are accompanied by a significant reduction in the number of T motoneurons at stage 29. Analyses of Hoxd10 electroporated embryos at the onset of motor column formation (stage 18) suggest that early and high levels of Hoxd10 expression led to the death of some early differentiating motoneuron. Despite these adverse effects, our data indicate that Hoxd10 expression is sufficient to induce LS motoneuron identity and axon trajectories characteristic of motoneurons in the LS anterior spinal cord. Equivalent changes in motoneuron identity were not found with the ectopic expression of Hoxd9, a gene normally expressed in T as well as LS segments. In an additional series of experiments, Hoxd10 was overexpressed in LS segments via in ovo electroporation at early neural tube stages. Analyses at stage 29 indicated proportionate increases in LMCl (Lim 1+, RALDH2+) motoneurons, and proportionate decreases in LMCm and MMC motoneurons (Isl 1+) motoneurons. These findings suggest that Hoxd10 specifically promotes the development and/or survival of LMCl motoneurons.

In the mammalian brain, thalamocortical circuits perform the initial stage of processing before information is sent to higher levels of the cerebral cortex. Substantial changes in receptive field properties are produced in the thalamocortical response transformation. In the whisker-to-barrel thalamocortical pathway, the response magnitude of barrel excitatory cells is sensitive to the velocity of whisker deflections, whereas in the thalamus, velocity is only encoded by firing synchrony. The behavior of this circuit can be captured in a model which contains a window of opportunity for thalamic firing synchrony to engage intra-barrel recurrent excitation before being 'damped' by slightly delayed, but strong, local feedforward inhibition. Some remaining aspects of the model that require investigation are: (1) how does adaptation with ongoing and repetitive sensory stimulation affect processing in this circuit and (2) what are the rules governing intra-barrel interactions. By examining sensory processing in thalamic barreloids and cortical barrels, before and after adaptation with repetitive high-frequency whisker stimulation, I have determined that adaptation modifies the operations of the thalamocortical circuit without fundamentally changing it. In the non-adapted state, higher velocities produce larger responses in barrel cells than lower velocities. Similarly, in the adapted barrel, putative excitatory and inhibitory neurons can respond with temporal fidelity to high-frequency whisker deflections if they are of sufficient velocity. Additionally, before and after adaptation, relative to putative excitatory cells, inhibitory cells produce larger responses and are more broadly-tuned for stimulus parameters (e.g., the angle of whisker deflection). In barrel excitatory cells, adaptation is angularly-nonspecific; that is, response suppression is not specific to the angle of the adapting stimulus. The angular tuning of barrel excitatory cells is sharpened and the original angular preference is maintained. This is consistent with intra-barrel interactions being angularly-nonspecific. The maintenance of the original angular preference also suggests that the same thalamocortical inputs determine angular tuning before and after adaptation. In summary, the present findings suggest that adaptation narrows the window of opportunity for synchronous thalamic inputs to engage recurrent excitation so that it can withstand strong, local inhibition. These results from the whisker-to-barrel thalamocortical response transformation are likely to have parallels in other systems.

The monosynaptic stretch reflex circuit consists of two neural cell types, sensory neurons and alpha-motoneurons. Ia afferents form synaptic connections with motoneurons projecting to the same or synergistic muscles, but not with motoneurons projecting to unrelated muscles. These synaptic connections form appropriately from the outset suggesting that they may be controlled by expression of specific adhesion molecules in matching sensory and motor neurons. Recently, two ETS-family transcription factors (Er81 and PEA3) were shown to be expressed in subsets of motoneurons and muscle sensory neurons. The expression patterns of these factors suggested that ETS genes might regulate the formation of synaptic connections between Ia afferents and motoneurons. This thesis explores the roles of Er81 and PEA3 in the formation of the stretch reflex circuit inferred from a study of Er81 and PEA3 null-mutant mice. Analysis of Er81 null-mutant mice revealed that Er81 controls a late step in Ia afferent axon guidance. Ia afferents induce the development of muscle spindles in the periphery and project axons into the spinal cord, but fail to grow axon collaterals into the ventral spinal cord where normally strong monosynaptic connections are formed with motoneurons. Consequently, monosynaptic Ia afferent inputs to motoneurons are greatly reduced in these mice. This severe phenotype precluded determination of whether or not the pattern of remaining Ia afferent inputs was normal. Intracellular recordings from quadriceps and obturator motoneurons in PEA3 null-mutants, however, revealed that functionally appropriate patterns of Ia afferent input to motoneurons develop normally in the absence of PEA3. PEA3 mutant mice demonstrated a role for PEA3 in the formation of a specific motor pool. Cutaneous maximus muscle motoneurons normally express PEA3. In PEA3 mutants, the majority of these motoneurons fail to migrate and coalesce appropriately into a discrete motor pool. These motoneurons also fail to project axons into the c. maximus muscle. Consequently, the muscle is atrophic. Thus, Er81 and PEA3 contribute to key developmental stages in the formation of the stretch reflex circuit: the growth of Ia afferents axons toward motoneurons and the formation of appropriate motor pool targets.

The NMDA receptor is the subject of intense study due to its critical role in many neuronal processes and neuropathologies. This receptor is modulated by a wide variety of endogenous and exogenous agents, including reducing and oxidizing (redox) agents. Despite a wealth of physiological information, details of the structural basis of modulation are only beginning to emerge. It has been proposed that the amino terminal domain (ATD) of NMDA receptor subunits may serve as a modulatory domain, as several agents appear to have sites of action in this region of the receptor. NR1/NR2A receptors contain cysteines in the ATD of both NR1 and NR2 that confer unique redox sensitivity to these receptors; however, the ATD redox sensitivity of NR1/NR2A receptors remains largely unexplored. The goal of this dissertation was to explore the impact of reducing and oxidizing agents on NMDA receptor function, focusing on the amino terminal domain redox sites. Here we demonstrate that a clinically efficacious neuroprotective agent, ebselen, is active as an oxidizing agent of the NMDA receptor. Additionally, these studies demonstrate a novel modulation of NR1/NR2A redox mutants by the polyamine spermine and explore a relationship between redox and spermine modulation of NR1/NR2A mutant receptors.

Spontaneous and sound-driven activity, glutamatergic synaptic transmission and Ca2+ signaling are critical for formation, maturation, refinement and survival of neuronal circuits including the auditory system. The present study investigated the mechanisms by which glutamatergic inputs from the cochlear nucleus regulate intracellular calcium concentration ([Ca2+]i) in developing lateral superior olive (LSO) neurons, using Ca2+ imaging in fura-2AM labeled brainstem slices. AMPA/kainate receptors primarily mediated Ca2+ responses elicited by single stimuli and contributed to Ca2+ responses elicited by low and high frequency bursts by approximately 75% and 50% respectively. Both AMPAR and kainate receptors were Ca2+ impermeable and increased [Ca2+]i via membrane depolarization and activation of voltage gated calcium channels (VGCCs). NMDARs contributed approximately 50% to Ca2+ responses independent of the stimulus frequency. Their high contribution to Ca2+ responses was consistent with their contribution (30-60%) to EPSPs triggered by stimulation of AVCN-LSO synapses. mGluRs contributed to Ca2+ responses only under high frequency stimulation (>20Hz). Group I mGluR-mediated Ca2+ responses had two components: release from internal stores and influx from the extracellular milieu. The influx was mediated by a channel sensitive to Ni2+, La3+ and 2-APB, consistent with it being a member of the TRP family. During development, the contribution of this channel decreased and it was lost after hearing onset, suggesting that it might be downregulated by auditory experience. In summary, distinct temporal patterns of synaptic activity in the LSO activate distinct GluR types and each receptor type employs a distinct Ca2+ entry pathway. This could possibly lead to activation of distinct intracellular cascades and distinct gene expression programs (West et al., 2001) that may be involved in distinct developmental aspects.

Parkinsons disease (PD) is a progressive neurodegenerative disorder resulting in part from loss of nigrostriatal dopamine (DA) neurons. Treatments act only to relieve symptoms. It is therefore essential to develop treatments that slow or reverse the neurodegenerative process. Here, I explored exercise as a potential treatment against a 6-hydroxydopamine (6-OHDA) rat model. 6-OHDA causes selective loss of DA neurons, a PD model. Forced limb use after 6-OHDA ameliorates behavioral and striatal DA effects. Further, exercise increases trophic factors, such as GDNF, that have neuroprotective qualities in this model. I explored the effects of forced limb use prior to 6-OHDA on the effects of the toxin and GDNF levels in the striatum. I demonstrated that prior forced limb use attenuated the behavioral deficits and loss of DA typical of 6-OHDA and increased GDNF in the striatum of animals exposed to forced use. The protective effect of exercise could reflect a decrease in the vulnerability of DA neurons, a regeneration of axons, or sprouting of axon terminals from undamaged neurons. Thus, I investigated the hypothesis that casting induced neuroprotection was due to the preservation of DA cells and terminals. Here, I demonstrated that forced limb use protected from 6-OHDA induced loss of DA neurons and terminals. These findings suggest that exercise exerts its effects by decreasing the vulnerability of DA neurons and terminals to 6-OHDA. Because casting increased GDNF, I next examined the effects of GDNF on 6-OHDA neurotoxicity during the 8 wk period after 6-OHDA. Using phenotypic markers of the nigrostriatal system, a non-DA cellular marker, and striatal DA content, I demonstrated these markers in the striatum and SN were not protected at 2 wks after 6-OHDA but recovered by 8 wks. No loss of DA cells in the SN or DA content in the striatum was observed in animals pretreated with GDNF. These data suggest that GDNF prevents 6-OHDA-induced DA cell death, but that weeks are required before these cells begin to normally express phenotypic markers. In conclusion, exercise may function to enhance the brains ability to produce trophic factors, which may slow or halt the degenerative process in PD.

Neuroinflammation is a major component of several neurodegenerative disorders such as Alzheimers disease, HIV-associated dementia, Multiple sclerosis and Parkinsons disease. Although the primary pathology underlying each of these diseases is vastly different, neuroinflammation, consisting of chronic activation of brain macrophages, is a significant factor that contributes to neuronal damage in all these conditions. Were it possible to image chronic activation of brain macrophages, it would be possible to monitor developing neuroinflammation and assess the efficacy of therapies that are targeted at modulating CNS inflammation. The goal of this thesis is to determine if activated brain macrophages can be imaged in vivo using positron emission tomography. We propose to take advantage of increased expression of the peripheral benzodiazepine receptor in activated brain macrophages and hypothesize that ligands that bind specifically to this receptor will label activated brain macrophages in vivo using positron emission tomography. The peripheral benzodiazepine receptor is normally expressed at low levels in the central nervous system in astrocytes and brain macrophages and is hypothesized to increase specifically on brain macrophages in neuroinflammation. We show that PK11195, a specific ligand to the peripheral benzodiazepine receptor, shows increased binding to brain macrophages in HIV encephalitis and that PK11195 can be used to image brain macrophages in vivo using positron emission tomography in a macaque model of HIV encephalitis. PK11195 binding is also increased in activated brain macrophages in Alzheimers disease and shows age dependent increases in transgenic mice models of Alzheimers disease. Finally we compare binding characteristics of DAA1106, a novel peripheral benzodiazepine receptor ligand, with PK11195 to show that DAA1106 binds with greater affinity in rat models of neuroinflammation both in brain tissues as well as in vivo. These data suggest that ligands of the peripheral benzodiazepine receptor specifically label activated brain macrophages and may be used to image neuroinflammation in vivo using positron emission tomography.

The effects of early life stressors on the developing infant can be severe. These effects include, but are not limited to, disordered attachment behavior in childhood, behavioral problems throughout childhood and adolescence, and increased risk for developing psychiatric illnesses later in life. Recent work in monkey models of early life stress suggests that the timing of the stressor is of paramount importance in determining the exact nature of the long-term behavioral disturbances. In monkeys, when stress is experienced at one-week of age, a pattern of decreased social affiliation and increased anxious behaviors develops that is not unlike children with the inhibited form of reactive attachment disorder. Conversely, if the stress is experienced at one-month of age, a pattern of increased social affiliation and increased anxious behaviors develops that resembles children with the disinhibited form of reactive attachment disorder. The work presented in this thesis describes two more features in this monkey model that depend on the timing of the early life stress. First, analysis of the acute behavioral response to the early life stress indicates that infants experiencing this stress at one-week of age immediately respond by increasing their level of nonsocial contact comforting behaviors, while infants experiencing the stress at one-month of age immediately respond by increasing their level of social comforting behaviors. It will be shown that the levels of each of these behaviors can predict aspects of the longer term behavior when measured between two and three months of age in their social rearing group environments. Next, it will be shown that gene expression in the amygdala, a brain region that controls socioemotional behaviors, also critically depends on the age at which early life stress is experienced. Furthermore, the expression of Guanylate Cyclase 1 á 3 in this brain region will be shown to correlate with certain differing aspects of behavior again measured between two and three months of age. Collectively, these studies may offer clues in determining very early which children are at risk for psychopathology following early life stress, and represent early work in determining a potential therapeutic target for the risks that ensue.

Preclinical and clinical evidence indicates that the serotonergic system regulates processing in the vestibular nuclei and in pathways linking balance function with emotional responses and affect. Previous studies conducted in this laboratory demonstrated that the serotonergic innervation of the vestibular nuclei is derived largely from the dorsal raphe nucleus (DRN), and revealed that the DRN also sends a nonserotonergic projection to the vestibular nuclei. The purpose of these experiments was to characterize the organization of the serotonergic and nonserotonergic components of the DRN projection to the vestibular nuclei. In Chapter 3, we describe retrograde tracing experiments that examined whether DRN cells send collateralized projections to the vestibular nuclei and central amygdaloid nucleus (CeA), regions involved in the clinical linkage between disorders of balance control and anxiety, and concluded that a subset of the serotonergic and nonserotonergic projections to the vestibular nuclei also project to CeA. Chapter 4 describes experiments with the anterograde tracer biotinylated dextran amine (BDA) that identified the terminal distribution of DRN projections within the vestibular nuclei. This study revealed that DRN projections descending in the ventricular plexus and the medial longitudinal fasciculus terminate within distinct vestibular terminal fields. In Chapter 5, BDA was used in combination with the serotonergic neurotoxin 5,7-dihydroxytryptamine (5,7-DHT) to selectively anterogradely trace nonserotonergic DRN projections to the vestibular nuclei. These experiments demonstrated that nonserotonergic DRN projections descend exclusively within the ventricular plexus and terminate primarily within the periventricular aspect of the vestibular nuclei. The purpose of the experiments in Chapter 6 was to map the distribution of serotonergic DRN terminals within the vestibular nuclei; 5,7-DHT was injected directly into DRN and silver staining was used to visualize the resulting pattern of terminal degeneration. It appears that projections from serotonergic DRN neurons terminate within both medial and lateral regions of the vestibular nuclei. Based on these findings, we conclude that major differences exist in the course of descent and termination patterns of serotonergic and nonserotonergic DRN projections to the vestibular nuclei, indicating that serotonergic and nonserotonergic cells give rise to distinct DRN projection systems that may selectively modulate processing within specific functional domains of the vestibular nuclei.