My lab studies the neuronal circuitry of the prefrontal cortex (PFC) that underlies higher-order cognitive functions. We are particularly interested to see how alterations in synaptic transmission might be related to schizophrenia and drug addiction.
An important cognitive feature of higher organisms is their ability to temporarily structure their behavior and to actively hold in mind information relevant for goal-attainment. This so called “working memory≵ is closely related to the functions of the PFC and its innervation by dopaminergic fibers. Pathological dysfunctions that disrupt the intrinsic circuitry of the PFC and impair working memory have been implicated in numerous mental illnesses, most notably schizophrenia. In addition, the importance of the PFC for executive functions and behavioral flexibility have also made it a focus for studies into the mechanisms that underlie drug addiction.
In order to understand the network properties that underlie the persistent neuronal activity required for working memory, projects in my lab utilize a combination many different techniques that include behavioral tests of cognitive flexibility, electrophysiological recordings in-vivo and in vitro, high-resolution calcium-imaging, pharmacological manipulations, and quantitative immunohistochemistry.
Repeated exposure to a sub-anesthetic dose of ketamine (30 mg/kg) reduces parvalbumin immunofluorescence in the adult medial PFC, mimicking the loss of GABAergic inhibitory interneurons seen in schizophrenia.
Network activity in the form of an Up-state and synaptic stimulation (top insert) in a fast-spiking interneuron. Recordings were made in co-cultures from transgenic GAD67- GFP+ animals. Expression of the green fluorescent protein (GFP) occurs selective in fast-spiking, parvalbumin-positive neurons.
Schizophrenia is a severe neurodevelopmental brain disorder that usually produces a lifetime of disability and emotional distress for affected individuals, and it places a considerable burden on the US economy. Among the clinical manifestations of schizophrenia, the deficits in cognitive abilities are a major determinant of functional outcome in schizophrenic patients. Understanding the cellular basis of these symptoms will aid the development of new treatment strategies. We are particularly interested in the functional role of inhibition in the PFC circuit and how the activity of GABAergic interneurons (which have been identified as a main locus of change in schizophrenia) is modulated by dopamine (Kroener et al., 2007). Several animal models of cognitive dysfunction replicate the functional loss of GABAergic interneurons that is seen in the PFC of schizophrenic patients; however, the physiological mechanisms which lead to this and the implications for network function are yet not well understood.
The so-called NMDA receptor hypofunction theory of schizophrenia provides a framework for the complex interactions of the dopaminergic, glutamatergic, and GABAergic systems in the pathophysiology of the disease. The theory suggests that schizophrenia is associated with a functional loss of NMDA receptors, specifically on GABAergic interneurons, which leads to a loss of inhibition and a secondary overstimulation in the glutamatergic and monoaminergic neurotransmitter systems. Administration of noncompetitive NMDA receptor antagonists such as ketamine or phencyclidine (PCP) in healthy subjects induces a schizophrenia-like syndrome including positive and negative symptoms, as well as cognitive dysfunction. Moreover, these compounds have been shown to exacerbate symptoms in schizophrenic patients. In animal models, acute or subchronic application of NMDA antagonists produces hyperlocomotion and stereotypies as well as deficits in social interactions, which are thought to correspond to aspects of the positive and negative symptoms of schizophrenia, respectively. These animals also show cognitive deficits such as impairments in memory and learning. Hypofunction of NMDA receptors on GABA neurons, especially those that contain the calcium-binding protein parvalbumin, disrupts the synchronization of neural circuits in the gamma-frequency range by altering inhibitory control of pyramidal cell networks. This synchronous activity is correlated with working memory and attentional processes, and thus its disruption in schizophrenia is likely related to the cognitive deficits observed in the disease.
One current line of research in our lab tries to elucidate the changes in excitatory (glutamatergic) synaptic transmission onto interneurons in a developmental NMDA receptor-hypofunction model. Using organotypic co-cultures we also investigate how these alterations affect recurrent activity in small networks of the PFC and related structures such as the ventral tegmental area (the source of the mesocortical dopamine innervation) or the hippocampus. Finally, we examine how NMDA receptor hypofunction affects the ability of DA to modulate the signal-to-noise ratio in these cortical networks (Kroener et al., 2009).
Synaptic barrages during persistent activity. The movie shows calcium transients that result from synaptic activity in a distal dendrite of a pyramidal cell in the PFC. The cell was filled with the calcium indicator Oregon Green 488 BAPTA 1 and up-states were evoked by stimulation of the ventral tegmental area.
Projects in my laboratory also examine how drugs of abuse (specifically cocaine and alcohol) can alter PFC function. A characteristic feature of addiction to alcohol is inappropriate decision making and loss of control over drinking in spite of negative outcomes. This is consistent with accumulating evidence that addiction involves alterations in the ability of the PFC to exert supervisory control over impulsive behaviors related to drug-seeking and relapse to drug taking. Current studies in my lab use an animal model of alcohol addiction to study changes in glutamatergic synaptic transmission and NMDA receptor function in the PFC. Chronic ethanol exposure induces homeostatic increases in NMDA receptors, which may affect the interplay between backpropagating action potentials and localized calcium-spikes required for spike timing-dependent plasticity, a physiologically relevant model of synaptic plasticity. Thus changes at the NMDA receptor could alter integrative properties and synaptic plasticity in PFC pyramidal neurons. These pathological neuroadaptations may contribute to a loss of response inhibition in the PFC during the development and maintenance of alcohol addiction.
Finally, we try to understand how stimulus associations may be “relearned” and new behaviors may be formed to aid the recovery from drug addiction. To this end we study drug extinction in rats that have learned to self-administer cocaine. These animals have learned a certain stimulus-reward association (i.e. a light or tone signals availability of cocaine) and they perform a simple task (a lever press) to receive the reward. At some point the stimulus (i.e. the tone or light) does not result in drug delivery anymore, and during the process of extinction the animal gradually (re)learns to suppress its behavior; that is to stop pressing the lever. We are trying to induce cortical plasticity during extinction in the hope to be able to speed up the process of relearning. Interventions that facilitate the extinction of drug-seeking behavior may prove useful in the treatment of drug addicts and may help them to transition into prolonged abstinence.
Calcium imaging of pairing action potential (AP) bursts with EPSPs in a layer 5 pyramidal neuron. A) Dendritic Ca2+ transients after extracellular stimulation of an EPSP alone (green), three APs alone (blue; 200 Hz), or during pairing of EPSPs and AP bursts (black), imaged in a basal dendrite 125 nm from the soma. The red trace shows the calculated linear sum of the EPSP plus AP response. Supralinear Ca2+ influx that results from pairing backpropagating APs with EPSPs as demonstrated here might result in spike-timing dependent plasticity and specifically LTP. B) Somatic voltage during EPSP or AP bursts, respectively. C) Confocal image of the cell recorded in A and B. White boxes indicate dendritic locations in the basal dendrites where Ca2+ spike could be recorded.