Understanding interactions between brain reward and sleep systems in driving maladaptive behaviors

National Institute On Alcohol Abuse And Alcoholism

Investigators

Linked publications, trials & patents

Abstract

FY2023 was the second year of this project. We have continued following up on findings from the first publication related to this research program (See: Kesner et. al., 2022. PMID: 35478010). That publication describes our work developing pre-clinical sleep and behavioral models of cannabis withdrawal symptoms. This research suggests that brain regions that express cellular machinery related to eCB activity and are associated with both sleep and motivated behaviors may be important loci to assess the neural mechanisms governing withdrawal phenomena associated with cessation of cannabis, and potentially other misused substances. One such brain area that we feel is understudied is the medial septum (MS). The MS expresses cannabinoid receptor type 1 (CB1), which predominantly occupies presynaptic neuron terminals, and is the main target for the psychoactive properties and misuse liability of THC. Release of endocannabinoids (eCBs) from neurons in the MS would act in a retrograde manner to reduce the release of neurotransmitters from presynaptic terminals in the MS. There are several brain regions that have neurons that make CB1 and project to the MS. One region, the supramammillary nucleus (SuM), is known to be involved in sleep-wake processes, heavily expresses mRNA for CB1, and densely projects to the septal complex. We used an intersectional viral strategy to express Cre-recombinase in SuM neurons projecting to the MS in mice that have the gene encoding CB1 flanked by loxp sites. This results in a Cre-recombination-mediated deletion of CB1 on SuM neurons that project to MS. We compared sleep in these mice to wildtype littermates that received the same intracranial viral injections and found that SuM to MS CB1 knockout increases NREM sleep during the active phase of the light cycle, while decreasing REM sleep in the inactive phase. These data confirm that both eCB release and its action on CB1 receptors in the MS is important for sleep-wake processes. During the past year we have also performed similar studies targeting another brain region involved in sleep vigilance-state architecture, the ventral medial thalamus (VM). When removing CB1 from VM circuitry, we found mice were spending less time in NREM sleep during the dark phase. These results show differential contributions of CB1 to sleep depending on which sleep pathways it is acting on. Beyond its role in sleep, the MS has also been implicated in reward seeking behaviors, making it an interesting region to study in relation to synergy of sleep and reward related brain circuitry. The MS is comprised of neurons that primarily make and release the inhibitory neurotransmitter gamma-Aminobutyric acid, the excitatory neurotransmitter glutamate (GLU), or the modulatory neurotransmitter acetylcholine. Of these distinct subpopulations of neurons within the MS, the GLU neurons (MS-GLU) are particularly understudied and have recently been implicated in reward processes. We know that selective stimulation of these neurons using optogentics is reinforcing in mice, who will actively press a lever to earn stimulation specifically of MS-GLU neurons. These neurons in turn project to the ventral tegmental area (VTA), a brain region highly implicated in motivated behavior. This pathway will thus influence VTA dopaminergic neuron activity and dopamine release in the ventral striatum. Beyond these recent findings, little is known about how MS-GLU neurons respond during natural reward seeking behaviors or how modulation of their activity can influence these behaviors. Our group has performed a set of studies to elucidate the role of these neurons in these processes. We expressed either the excitatory designer receptor exclusively activated by designer drug (DREADD) hM3D-Gq, the inhibitory DREADD hM4D-Gi, or control mCherry protein exclusively on MS-GLU neurons. We trained mice to perform a basic reward seeking procedure where one lever press on an active lever (of two levers) resulted in delivery of a sucrose reward. We found that administration of the DREADD ligand, clozapine-n-oxide (CNO) resulted in reduced lever pressing in hM4D-Gi-expressing mice, but not in hM3D-Gq or control mice. Additionally, consummatory behaviors were increased in the hM4D-Gi group but not in other groups. These data suggest that taking MS-GLU neurons offline alters the appetitive and consummatory processes associated with goal directed behaviors in an opposite fashion. When we reversed the lever assignments, where the previously active lever did nothing, but pressing the other lever delivered reward, the hM3D-Gq group adapted their behavior to this new contingency much faster than the control mice, and the hM4D-Gi mice never appropriately adapted. These findings suggest augmenting MS-GLU activity plays a role in enhancing cognitive flexibility. we then used fiber photometry techniques to measure calcium activity in MS-GLU neurons. Calcium activity is a correlate of neuronal activity, so increased calcium levels indicate action potentials and general neuronal activity. In mice performing the same reward-seeking behaviors we found profoundly reduced activity in MS-GLU neurons during reward consumption. MS-GLU neuron activity also appears to encode some valence of different components of the appetitive behaviors; as activity dynamics are different depending on which lever the mouse presses and whether reward was present during consummatory behaviors. In addition, we have begun studying the interactions between MS-GLU neurons and canonical reward circuitry, specifically dopamine (DA) activity in the nucleus accumbens (NAc). We examined NAc-DA responses during 3 consecutive days while the animals performed a Pavlovian strategy switch paradigm, all while MS-GLU neurons were modulated via DREADDs. Analysis of these experiments is ongoing, but preliminary results indicate NAc-DA changes in response to the two cues in this paradigm differ depending on MS-GLU modulation via DREADDs. Of particular note, we see a far smaller NAc-DA response upon a stimulus that used to predict a reward (but now does not) in Gq mice compared to Gi or mCherry groups on Day 1 of the strategy switch paradigm. We also observe potentially larger increase in NAc-DA upon reward retrieval after a new stimulus that predicts a reward in the Gq group again on Day 1. These Day 1 observations are intriguing as they suggest increasing MS-GLU activity during the first day of a strategy switch paradigm helps the animal better disregard the old reward predictive cue, and better associate the new predictive cue with reward. In essence, increasing MS-GLU activity via Gq DREADDs seems to enhance the timeline in which reward prediction error may factor into the animal learning the new cue-reward pairing, and extinguish the old cue-reward pairing. These data are being prepared for article submission. Overall, we have begun homing in on brain circuitry that is involved in both motivation and arousal processes to assess the interaction between these processes in sleep and goal directed behavior.