Project 1: Dopamine regulation of lateral inhibition between striatal neurons gates the stimulant actions of cocaine This project uncovers a novel synaptic mechanism by which cocaine exerts its stimulant effect on locomotion. It showed that cocaine suppresses the lateral inhibition between neurons in the nucleus accumbens to disinhibit striatal neurons and promote locomotion. Striatal medium spiny neurons (MSNs) form inhibitory synapses on neighboring striatal neurons through axon collaterals. The functional relevance of this lateral inhibition and its regulation by dopamine remains elusive. We show that synchronized stimulation of collateral transmission from multiple indirect-pathway MSNs (iMSNs) potently inhibits action potentials in direct-pathway MSNs (dMSNs) in the nucleus accumbens. Dopamine D2 receptors (D2Rs) suppress lateral inhibition from iMSNs to disinhibit dMSNs, which are known to facilitate locomotion. Surprisingly, D2R inhibition of synaptic transmission was larger at axon collaterals from iMSNs than their projections to the ventral pallidum. Targeted deletion of D2Rs from iMSNs impaired cocaines ability to suppress lateral inhibition and increase locomotion. These impairments were rescued by chemogenetic activation of Gi-signaling in iMSNs. These findings shed light on the functional significance of lateral inhibition between MSNs and offer a novel synaptic mechanism by which dopamine gates locomotion and cocaine exerts its canonical stimulant response. Project 2: Enhanced GABA transmission drives bradykinesia following loss of dopamine D2 receptor signaling Dopamine (DA) depletion, caused by pharmacological agents or genetic models of DA neuron degeneration, produces a pronounced motor deficit that closely resembles the motor symptoms characteristic of Parkinsons disease (Bergman and Deuschl, 2002; Deumens et al., 2002; Lovinger, 2010; Palmiter, 2008; Schultz, 1982; Surmeier et al., 2014). However, this basic research has struggled to accurately and consistently relate the motor deficits produced by dopamine depletion to specific changes in striatal neuronal activity and circuit function. This is owing, in large part, to technical limitations such as the experimental variability inherent to 6-OHDA lesions (Deumens et al., 2002; Vandeputte et al., 2010) in addition to the biological complexity such as the heterogeneity of DA receptor expression within the basal ganglia (Surmeier et al., 2011). DA D2 receptors (D2Rs) are critical to motor output because global deletion of D2Rs leads to similar motor deficits as those seen after dopamine depletion (Baik et al., 1995; Kelly et al., 1998; Rubinstein et al., 1990). However, in the striatum alone, D2Rs are expressed on at least five different types of neurons: indirect-pathway MSNs (iMSNs), cholinergic interneurons, a subset of GABA interneurons, and on afferents to the striatum from dopamine and prefrontal cortex neurons (Bamford et al., 2004; Bello et al., 2011; Centonze et al., 2003; Maurice et al., 2004; Surmeier et al., 2011). It is unclear whether ubiquitous activation of all these striatal D2Rs or selective activation of a subset of D2Rs is responsible for generating the motor deficits of DA depletion models. Furthermore, it is unknown whether D2Rs expressed in the cortex and other non-striatal regions that receive projections from midbrain DA neurons also contribute to this motor phenotype. In this study, we generated a mouse line with a precise genetic manipulation that allows for reliable titration of the levels of D2Rs on iMSNs by crossing conditional Drd2loxP/loxP mice with mice expressing Cre recombinase selectivity in striatal iMSNs (Adora2A-Cre+/-). This mouse line, referred to as iMSN-Drd2KO, exhibits bradykinesia while showing normal DA release properties, indicating that downregulation of D2R function in iMSNs is sufficient to recapitulate the motor impairments of DA depletion models. This project showed that selective deletion of DA D2 receptors (D2Rs) from indirect-pathway medium spiny neurons (iMSNs) is sufficient to impair locomotor activity, phenocopying DA depletion models of Parkinsons disease, despite this mouse model having intact DA transmission. There was a robust enhancement of GABAergic transmission and a reduction of in vivo firing in striatal and pallidal neurons. Mimicking D2R signaling in iMSNs with Gi-DREADDs restored the level of tonic GABAergic transmission and rescued the motor deficit. These findings indicate that DA, through D2R activation in iMSNs, regulates motor output by constraining the strength of GABAergic transmission. Project 3: The predictability of seeking behaviors and alcohol drinking in two operant models of alcohol self-administration Alcohol use disorders (AUDs) cause significant health and societal problems, and nearly 17 million adults have an AUD (SAMHSA, 2013). Therefore, enhancing our understanding of the neural circuitry that drives and regulates drink alcohol is a critically needed advancement. Animal models that test an animals motivation for the reinforcing properties of alcohol are invaluable tools for capturing the endophenotypes associated with AUD and identifying the circuits responsible. Operant self-administration of ethanol is well suited for this purpose, in part because of the flexibility in the reinforcement schedule, but also because the animals seeking and drinking behaviors can be precisely measured across a session. Numerous procedures of operant ethanol drinking have been developed. However, several limitations are apparent. First, few studies used mice, limiting the potential to test transgenic mice or manipulate neural activity using chemogenetic or optogenetic approaches. Second, most studies rely on operant responding as the main measure of alcohol consumption and use previously published blood ethanol concentration (BEC) levels to estimate intoxication, rather than validating consumption by post-session sampling of BEC. As a consequence, it is very difficult to tell if the intoxication of a particular animal can be accurately predicted by its operant behaviors (i.e. lever pressing, etc.). Third, some procedures add sucrose to the alcohol solution, especially during initial training phase with sucrose fading as the operant behavior is acquired, which confounds the reinforcing effects of ethanol that are being studied. Finally, the approach of separating the seeking and consummatory phases of drinking by limiting alcohol access while the mouse works towards the set ratio prevents testing the alteration of alcohol dose (concentration) or taste within one session. For this project we compare two models of operant alcohol self-administration that use a CUP or a SIPPER to deliver the alcohol solution. We found that the model using the SIPPER provide more reliable operant behavior that can be better correlated with alcohol consumption and the levels of achieved BEC. We also described how this model can be used to test the sensitivity to quinine adulteration of the alcohol solution as a measure of aversive insensitive alcohol drinking. The manuscript is in the final stages of preparation for submission.
Showing the most recent 10 out of 29 publications
