Major depressive disorder (MDD) is responsive to serotonin transporter reuptake inhibitors (SSRIs), and a functional polymorphism in the serotonin transporter (SERT) putative promoter region impacts vulnerability to develop MDD. Paradoxically, mice lacking SERT (knock-out; SERT-KO) display an increased anxiety-like phenotype, which has been demonstrated to have developmental origins; however, the extent to which systems adaptations that have occurred in SERT-KO mice recapitulates a broader anxious depressive-like phenotype at the behavioral and molecular levels is not known. We investigated SERT-KO in a panel of behavioral tests, and analyzed WT and KO gene expression in amygdala (AMY) and cingulate cortex (CC) by microarray. We then compared the SERT-KO gene expression results to human MDD and Control microarray in AMY and anterior cingulate cortex (ACC), areas shown to be functionally, structurally and molecularly affected in MDD. Gene expression changes were confirmed by real-time quantitative polymerase chain reaction (qPCR). RESULTS: SERT-KO behaviors across several tests denote a robust anxious depressive-like syndrome, which is reminiscent of the human MDD syndrome. Some gene expression changes were conserved between mouse and human in AMY and ACC/CC as measured by microarray (29 genes; 19 genes), and a subset were selected for qPCR validation. Differential gene expression confirmed by qPCR in both mouse and human (2 genes in AMY) included an upregulation in AMY of adenylate cyclase VII (ADCY7), a gene previously implicated in MDD. Increased expression of two genes that display significant coregulation in mouse and human suggests the recruitment of a conserved functional unit related to cyclic adenosine monophosphate (cAMP) signaling, a signal transduction pathway implicated in MDD. CONCLUSIONS: The SERT-KO mouse recapitulates behavioral and selected molecular features of a rodent syndrome homologous to human MDD. Therefore, it provides a useful model for investigating molecular mechanisms in AMY which are relevant to the pathology of MDD. These results support altered cAMP signaling pathway as a cross-species conserved feature of the pathophysiology of MDD.

Work is presented aimed at understanding the function of the basal ganglia in reward-related learning. Behavioral, fMRI, and computational techniques are used to examine basal ganglia activation during the reinforcement of physical (i.e., motoric) and cognitive (i.e., non-motoric) actions. In a single experiment design, participants received positive and negative reinforcement for performing actions in one of four possible directions depending on a color cue stimulus. During different phases of the experiment, participants performed either hand movements, eye movements, or covert attention shifts. Behavioral and fMRI data collected during the task were used to test predictions from simulated Reinforcement Learning (RL) agents trained on the same sequences of stimulus, action, and outcome experienced by the human participants. Behavioral data showed that participants were able to learn the three types of action equally well and at similar rates, providing behavioral evidence that a common algorithm might be involved. Further, RL simulations fit the learning of the three types of action equally well, suggesting that RL might be that common algorithm. A deconvolution analysis of striatal fMRI BOLD data suggested that: (1) the striatum computed reward prediction errors for both physical and cognitive actions, and (2) this computation was localized to different regions of the striatum depending on the type of action that was being rewarded. The localization of these computations replicates prior findings implicating those regions in action-specific voluntary control, but extends them to include the fact that these regions compute action-specific reward-prediction errors. Together, these data suggest that distinct circuits linking neocortex with the basal ganglia are involved in RL-related computations for the actions controlled by those circuits. The results point to a theoretical framework in which the basal ganglia mediate the reinforcement of actions whose control is delegated to local neocortical regions. Implications for theories of learning, theories of cognitive control, and mapping RL to the basal ganglia are discussed. Finally, the possibility that phasic dopamine might mediate the prediction error signals is considered in view of some theoretical challenges such as its non-specificity, its asymmetric representation of negative reinforcement, and its presence during non-rewarding events.

Early life experience has a powerful influence on later stress reactivity, which is demonstrated by the animal model, repeated brief postnatal maternal separation. In this classic paradigm, rat pups undergo a 15-minute daily separation (MS15) from their dam for approximately one to two postnatal weeks. A substantial literature has demonstrated that adult rats with a developmental history of MS15 are significantly less stress reactive compared to controls, as evidenced by decreased stress-induced hormone release. Conversely, the effects of early life experience on brain circuits that control stress responses are virtually unknown. Descending preautonomic circuits govern the output of the autonomic nervous system, which mediates physiological responses to stress (e.g., increased heart rate and decreased digestion). These circuits begin in the paraventricular nucleus of the hypothalamus (PVN) and limbic forebrain and synaptically innervate preganglionic neurons in the brainstem dorsal vagal complex (DVC) and spinal cord that ultimately innervate body organs. A previous study from our laboratory has demonstrated that MS15 alters the developmental assembly of gastric preautonomic circuits (Card et al., 2005). These findings led us to hypothesize that MS15 rats would display altered circuit strength of gastric preautonomic circuits later in development, as assessed in juvenile rats. Indeed, the study described in Chapter 2 demonstrated that MS15 enhances the circuit strength of gastric preautonomic circuits originating within the PVN in juvenile rats. This enhanced circuit strength suggests that the function of preautonomic PVN pathways might also be altered by MS15. Thus, we hypothesized that MS15 rats would display altered stress-induced activation of the PVN to DVC pathway. The study described in Chapter 3 revealed that MS15 rats display decreased stress-induced Fos activation within the PVN and within a specific population of DVC neurons. Therefore, studies within this dissertation revealed that M15 alters the circuit strength of PVN preautonomic pathways and alters stress-induced activation of brainstem preautonomic pathways. These findings suggest that MS15 rats would display attenuated autonomic responses to stress and may provide insights into how early life experience shapes later stress reactivity.

As a foveate animal, the primate must redirect its gaze with saccadic eye movements to subject different objects to high resolution analysis. Though beneficial in extending the range of visual analysis, the saccade-and-fixate oculomotor strategy poses a problem to the visual system as it performs its analyses. Each saccade results in a whole-field displacement of the visual image across the retina. Nevertheless, we experience a stable visual percept, implying a brain mechanism for visuo-spatial correction. The experiments reported here examine the neural mechanisms underwriting this correction. In the first study, we sought to understand how the frontal eye field (FEF) gains access to information about ipsilateral space. Information about all of space, not just the contralateral hemifield, is a prerequisite for omnidirectional processes such as spatial remapping, a putative mechanism of visual stability. We found that one source of ipsilateral information is the superior colliculus (SC) on the opposite side of the brain. In the second study, we set out to test a major prediction of one theory of visual stability. This theory invokes the function of neurons with shifting receptive fields (RFs) as a mechanism for achieving transaccadic visual stability. Shifting RFs effectively sample the same region of space twice, presaccadically and postsaccadically, and a percept of stability may rely on how well the samples match. This theory has the salient prediction that neurons in areas where shifting RFs are found should be sensitive to changes that occur to stimuli during saccades. We tested this prediction by recording from FEF neurons while monkeys performed a task during which a probe changed along a particular dimension during a saccade. We found that FEF neurons are indeed sensitive to intrasaccadic alterations of visual stimuli. In a third and final study, we sought to bridge the neuron-behavior gap by recording from FEF neurons while monkeys performed a visual stability judgment task that probed their capacity to detect changes occurring during saccades. We found that monkeys are clearly able to discern whether a stimulus is stable or unstable during a saccade and moreover that FEF neural activity is predictive of monkey psychophysical performance.

The high-affinity choline transporter (CHT) supplies the substrate, choline, for the synthesis of acetylcholine (ACh) within cholinergic neurons. Choline uptake mediated by this protein has been studied for over 30 years and many of the regulatory mechanisms governing its function are well characterized. Early studies, as well as more recent investigations, focused on specific populations of cholinergic axons in the brain, namely the cholinergic innervations of cortical, striatal, and hippocampal regions. Details of the expression and subcellular localization of the high-affinity choline transporter within the projections of the pedunculopontine (PPT) and laterodorsal (LDT) tegmental cholinergic neurons have not been examined. The studies described herein compare the cholinergic axons within two limbic regions that are innervated by the ascending projections from these brainstem nuclei. These experiments were designed to characterize: 1) the relative amount and pattern of subcellular localization of the high-affinity choline transporter protein in the axon varicosities of this projection system; 2) the co-expression of the high-affinity choline transporter and the vesicular acetylcholine transporter in these two populations of axon varicosities; and 3) the organization and possible collateralized projections of the cholinergic neurons that provide cholinergic innervation to these regions. The results of these studies indicate that the expression and localization of the high-affinity choline transporter differs only subtly across brain regions innervated by the brainstem tegmental cholinergic neurons, and suggest that these differences may be accounted for by a pattern of specific innervation arising from distinct subsets of pedunculopontine and laterodorsal tegmental cholinergic neurons.

Glutamate mediates the majority of fast excitatory neurotransmission in the vertebrate brain. Glutamate receptors (GluRs) transduce signals in two ways: metabotropic GluRs signal via intracellular G proteins, whereas ionotropic GluRs (iGluRs) open intrinsic ion channels in response to agonist binding. NMDA receptors (NMDARs) are glutamate- and glycine-gated iGluRs that play critical roles in spatial learning, contextual fear memory acquisition, synapse elimination, and chronic pain. Their particularly high calcium (Ca2+) permeability and strongly voltage-dependent channel block by external magnesium (Mg2+) distinguish NMDARs from other iGluRs. Mg2+ channel block of NMDARs inhibits current influx through the majority of agonist-bound, open NMDARs at resting membrane potentials (Vms), but this block is relieved by depolarization. Thus, significant current flow through NMDARs requires presynaptic activity (glutamate release) and postsynaptic activity (depolarization to relieve Mg2+ channel block), conferring on NMDARs a coincidence-detection capability that is central to their physiological importance. To mediate this and other important functions, NMDARs require tight regulation of the voltage-dependent Mg2+ block that provides crucial control of NMDAR-mediated current flow and Ca2+ influx. NMDARs are typically composed of NR1 and NR2 subunits. The four NR2 subunits (NR2A-D) contribute to four diheteromeric NMDAR subtypes (NR1/2A-NR1/2D), which differ in many respects, including the magnitudes of channel block by Mg2+, Ca2+ permeability, and single-channel conductance. Previously-gathered data from our lab demonstrates that the subtype specificity of Mg2+ block is principally conferred by a single amino acid site in the third transmembrane region (M3) of NR2 subunits. This NR2 S/L site contains a serine in NR2A and NR2B subunits and a leucine in NR2C and NR2D subunits. Surprisingly, the NR2 S/L site does not line the pore. I created several structural homology models of NMDARs to generate hypotheses regarding how the NR2 S/L site conveys its effects to the pore. I tested these hypotheses experimentally and found that the NR2 S/L site interacts with an NR1 subunit tryptophan in the pore-loop to regulate Mg2+ block properties. I further determined that the NR2 S/L site greatly contributes to the subtype variation in single-channel conductance, and likely plays a role in the subtype variation in Ca2+ permeability.

We see the world as it unfolds in both space and time. Neuroscience research so far, however, has largely focused on the spatial aspects of vision, including orientation and size. No less important to a comprehensive understanding of brain function is an understanding of how visual input is transformed into knowledge about the timing of events in the world. To begin to address this issue, we recorded the activity of single neurons in the frontal eye field (FEF), an area of prefrontal cortex thought to help mediate conscious visual perception. For comparison, we recorded from two portions of the superior colliculus (SC) in the midbrain. The superficial SC receives inputs from the retina and early visual areas, and intermediate SC is associated with early visual processing and the control of eye movements. In two experiments, we measured visual responses to time-varying stimuli and tested whether the magnitudes or latencies of the responses might be used by the brain as a source of timing information. First, we measured visual responses in individual neurons while two consecutive flashes of light were presented during passive fixation. We found that when stimulus intervals were brief (~200 milliseconds), neurons responded robustly to the first flash but not the second one ("neuronal adaptation"). As intervals lengthened, neurons fired more robustly for the second flash. Thus, information about time was implicit in the size of successive visual responses. We then asked if this timing information is exploited by the brain. We recorded activity in FEF and SC while monkeys performed a time interval discrimination task. We evaluated the "magnitude hypothesis", stemming from our adaptation findings, and the "latency hypothesis", which predicts that time intervals are encoded by the relative latencies of visual responses. We found that performance in the task was best described by the magnitude hypothesis; larger visual responses were associated with longer passages of time. We conclude that neuronal adaptation may play a functional role in time perception. Thus, the timing of visual events in the world, at short naturalistic timescales, is partly encoded by the magnitude--not just the latency--of neuronal activity.

Cognitive impairments are a core feature of schizophrenia and the best predictor of functional outcome, though current pharmacotherapies offer only limited cognitive improvement. Cognitive deficits span multiple domains and thus may reflect an overarching alteration in cognitive control, the ability to adjust thoughts or behaviors to achieve goals. Cognitive control depends on the dorsolateral prefrontal cortex (DLPFC), which exhibits altered activity in schizophrenia. DLPFC dysfunction is thought to be due, at least partially, to alterations in interneurons, which are regulated by levels of the GABA synthesizing enzyme glutamic acid decarboxylase (GAD). A deficit in the 67 kDa isoform (GAD67), responsible for the majority of cortical GABA synthesis, has been widely replicated in the DLPFC of subjects with schizophrenia and is particularly prominent in the parvalbumin (PV)-containing subclass of interneurons. However, little is known about the relationship of DLPFC GAD67 mRNA levels and medication use, substance abuse, and illness severity and chronicity; translation of the transcript into protein; or protein levels in axon terminals, a key site of GABA production and function. Additionally, alterations in other GABA neurotransmission markers, including lower PV and GABA membrane transporter 1 (GAT1), are also present, and thought to result from lower GAD67 in PV neurons, though this hypothesis has not been directly tested. Accordingly, here we measured GAD67 mRNA, tissue-level protein, and axon terminal protein in PV cells, and examined whether lower PV and GAT1 mRNA are consequences of lower GAD67 protein in PV neurons. GAD67 mRNA levels were significantly 15% lower in schizophrenia subjects, but transcript levels were not associated with medication use, substance abuse, predictors or measures of disease severity, or illness chronicity. GAD67 protein levels were significantly 10% lower in total gray matter and 49% lower in PV axon terminals. These data provide an extensive characterization of the GAD67 deficit in schizophrenia, and provide novel evidence of a functional impairment in PV neurons that may underlie cognitive deficits. Additionally, PV and GAT1 mRNAs were not altered in two mouse models with lower GAD67 expression, suggesting that lower GAD67 is unlikely to be the cause of reduced PV and GAT1 mRNA in schizophrenia.

We scan the visual world by making rapid eye movements (saccades) and serially focusing on objects of interest. Despite abrupt retinal image shifts, we see the world as stable. Remapping contributes to visual stability by updating the internal image with every saccade. Neurons in macaque lateral intraparietal cortex (LIP) and other brain areas update information about salient objects around the time of a saccade. Information about salient objects is transferred from neurons that currently encode their screen locations to other neurons that will encode their locations after the saccade. The depth of information transfer remains to be thoroughly investigated. Area LIP, as part of the dorsal visual stream is regarded as a spatially selective area. Yet there has been increasing evidence that LIP neurons also encode object features. We sought to determine whether LIP remaps shape information. Such insight is required for understanding what information is retained from each glance and how the visual percept is built (transsaccadic perception). First, we presented shapes in the future location of the receptive field around the time of the saccade and tested for shape selectivity during remapping. Second, we presented the same shapes within the receptive field and tested for shape selectivity in the fixation task. Finally, we compared selectivity in the two tasks. We found that LIP neurons automatically encode and remap shape information. Selectivity in the two tasks was comparable. Our results provide critical evidence for the idea that remapping may be a mechanism for transsaccadic perception of features.

We spend a large portion of life as the object of our own thoughts. Daily we reflect on all sorts of recent and not so recent decisions, and the products of those reflective thoughts serve to guide future goals, actions, and thoughts. The process of thinking about thinking, or metacognition, has garnered scrutiny in psychology studies for decades and recently in some imaging and neurological studies, but its neuronal basis remains unknown. Moreover, metacognition is largely thought a uniquely human ability, and only very recently has some evidence suggested other species may harbor metacognitive skills. To begin investigating neuronal mechanisms underlying metacognition, we performed two experiments. First, we tested whether rhesus macaques exhibited evidence for metacognition. We trained monkeys to perform a visual oculomotor metacognition task. In each trial, monkeys made a decision then made a bet. To earn maximum reward, monkeys had to monitor their decision and then make a bet to indicate whether the decision was correct or incorrect. We found the monkeys behavior was best explained by a metacognitive strategy, and we ruled out possible alternative strategies to perform the task such as reliance on visual stimuli or saccadic reaction times. Second, we tested whether neurons exhibited activity correlated with metacognition. While monkeys performed the task we recorded from single neurons in three frontal cortical areas known to play roles in higher cognitive functions: the frontal eye field, lateral prefrontal cortex, and the supplementary eye field. Our predictions were that frontal eye field neuronal activity would correlate with making the decisions but not the bets, and that lateral prefrontal cortex and supplementary eye field neuronal activity would correlate with linking the decisions to the bets the putative metacognitive signals. We found signals in all three brain areas correlated with making decisions and correlated with making bets. The supplementary eye field was the only area of the three that exhibited strong signals correlated with metacognitive monitoring, and these signals appeared early and were sustained throughout the task. Our results identify the supplementary eye field as a likely contributor to metacognitive monitoring.