Evidence for a vocal signature in the rat and its reinforcing effects: a key role for the subthalamic nucleus

Although rodents have a well-structured vocal form of communication, like humans and non-human primates, there is, to date, no evidence for a vocal signature in the well-known 50- and 22-kHz ultrasonic vocalizations (USVs) emitted by rats. Here, we show that rats can recognize the identity of the USV emitter since they choose to preferentially self-administer playback of 50-kHz USVs emitted by a stranger rat over those of their cagemate. In a second experiment, we show that only stranger, but not familiar, 50-kHz USVs reduce cocaine self-administration. Finally, to study the neurobiological substrate of these processes, we have shown that subthalamic nucleus (STN)-lesioned rats did not lever press much for any USV playback, whatever their emotional valence, nor did they seem able to differentiate familiar from stranger peer. Advocating for the existence of a vocal signature in rats, these results highlight the importance of ultrasonic communication in the socio-affective influence of behaviour, such as the influence of proximal social factors on drug consumption and confirm the role of the STN on this influence.

Instrumental aversion coding in the basolateral amygdala and its reversion by a benzodiazepine

Punishment involves learning the relationship between actions and their adverse consequences. Both the acquisition and expression of punishment learning depend on the basolateral amygdala (BLA), but how BLA supports punishment remains poorly understood. To address this, we measured calcium (Ca2+) transients in BLA principal neurons during punishment. Male rats were trained to press two individually presented levers for food; when one of these levers also yielded aversive footshock, responding on this punished lever decreased relative to the other, unpunished lever. In rats with the Ca2+ indicator GCaMP6f targeted to BLA principal neurons, we observed excitatory activity transients to the footshock punisher and inhibitory transients to lever-presses earning a reward. Critically, as rats learned punishment, activity around the punished response transformed from inhibitory to excitatory and similarity analyses showed that these punished lever-press transients resembled BLA transients to the punisher itself. Systemically administered benzodiazepine (midazolam) selectively alleviated punishment. Moreover, the degree to which midazolam alleviated punishment was associated with how much punished response-related BLA transients reverted to their pre-punishment state. Together, these findings show that punishment learning is supported by aversion-coding of instrumental responses in the BLA and that the anti-punishment effects of benzodiazepines are associated with a reversion of this aversion coding.

Opposing roles for striatonigral and striatopallidal neurons in dorsolateral striatum in consolidating new instrumental actions

Comparatively little is known about how new instrumental actions are encoded in the brain. Using whole-brain c-Fos mapping, we show that neural activity is increased in the anterior dorsolateral striatum (aDLS) of mice that successfully learn a new lever-press response to earn food rewards. Post-learning chemogenetic inhibition of aDLS disrupts consolidation of the new instrumental response. Similarly, post-learning infusion of the protein synthesis inhibitor anisomycin into the aDLS disrupts consolidation of the new response. Activity of D1 receptor-expressing medium spiny neurons (D1-MSNs) increases and D2-MSNs activity decreases in the aDLS during consolidation. Chemogenetic inhibition of D1-MSNs in aDLS disrupts the consolidation process whereas D2-MSN inhibition strengthens consolidation but blocks the expression of previously learned habit-like responses. These findings suggest that D1-MSNs in the aDLS encode new instrumental actions whereas D2-MSNs oppose this new learning and instead promote expression of habitual actions.

The dopamine dip amplitude is a quantitative measure of disappointment

Great expectations can lead to greater disappointment when the desired results are not achieved. Although it is well-documented that a better-than-expected reward is encoded quantitatively via midbrain dopaminergic (DA) activity, it has been less addressed experimentally whether disappointment, a worse-than-expected outcome is also expressed quantitatively by the DA signal. We show that the degree of disappointment is quantified by the magnitude of a transient decrease in the extracellular DA concentration (DA dip) in the ventral striatum of mice. We set up a lever press task on a fixed ratio (FR) schedule requiring five lever presses as an effort for a food reward (FR5). Mice occasionally checked the food magazine without a reward before completing the task. The percentage of this premature magazine entry (PME) increased as the number of lever presses approached five. This behavioral readout shows rising expectations with increasing proximity to task completion, leading to greater disappointment in the mice. Fiber photometry of the extracellular DA dynamics in the ventral striatum using a fluorescent protein (genetically encoded GPCR-activation-based-DA sensor: GRAB-DA2m) revealed that the amplitude of the DA dip following a PME was correlated with the percentage of the PME, demonstrating a monotonic relationship between the DA dip amplitude and the degree of disappointment. Computational modeling of the lever press task implementing temporal difference errors and state transitions replicated the positive correlation between the PME frequency and DA dip amplitude during improvement of the FR5 task. We propose that the degree of disappointment during making an effort for a goal is represented monotonically by the magnitude of the DA dip in the ventral striatum, which may guide behavioral adjustment.

Instrumental aversion coding in the basolateral amygdala and its reversion by a benzodiazepine

Punishment involves learning the relationship between actions and their adverse consequences. Both the acquisition and expression of punishment learning depend on the basolateral amygdala (BLA), but how BLA supports punishment remains poorly understood. To address this, we measured calcium (Ca2+) transients in BLA principal neurons during punishment. Male rats were trained to press two individually presented levers for food; when one of these levers also yielded aversive footshock, responding on this punished lever decreased relative to the other, unpunished lever. In rats with the Ca2+indicator GCaMP6f targeted to BLA principal neurons, we observed excitatory activity transients to the footshock punisher and inhibitory transients to lever-presses. Critically, as rats learned punishment, activity around the punished response transformed from inhibitory to excitatory and similarity analyses showed that these punished lever-press transients resembled BLA transients to the punisher itself. Systemically administered benzodiazepine (midazolam) selectively alleviated punishment. Moreover, the degree to which midazolam alleviated punishment was associated with how much punished response-related BLA transients reverted to their pre-punishment state. Together, these findings show that punishment learning is supported by aversion-coding of instrumental responses in the BLA and that the anti-punishment effects of benzodiazepines are associated with a reversion of this aversion coding.

Stress Promoted Nicotine Intake in a Rat Model of Tobacco Smoking

Epidemiological documents show an association of tobacco smoking rates and perceived stress levels. This study, using an animal model of nicotine self-administration, investigated effects of stress on nicotine-taking behavior. Sprague-Dawley rats were trained to intravenously self-administer nicotine. Thirty minutes before test sessions, animals were challenged with an intraperitoneal administration of a pharmacological stressor yohimbine. In the low nicotine-taking rats, yohimbine challenge enhanced lever-press responses and thereby nicotine intake. In contrast, no such effect was observed in the high nicotine-taking rats. After yohimbine challenge, nicotine intake in those originally low nicotine-taking rats remained at the heightened level. These findings demonstrate that exposure to stress facilitates nicotine self-administration in the rats previously consuming less nicotine and makes them to become high nicotine-taking subjects. The results of this study suggest that stressful life events may be effective in increasing tobacco smoking in light to moderate smokers and therefore increase the prevalence of nicotine dependence. As such, reducing stress levels in daily life may prove to be an effective approach to the prevention of nicotine addiction.

Evidence for a vocal signature in the rat and its reinforcing effects

While the term 'language' is used for human and non-human primates, 'vocal communication' is rather used for rodents or other species. The main difference is that there is, to date, no evidence for a vocal signature in the well-known 50- and 22-kHz ultrasonic vocalizations (USV) emitted by rats. Here, we show that rats can recognize the identity of the USV emitter since they self-administer preferentially playback of 50-kHz USV emitted by a stranger rat over those emitted by their cage-mate. In a second experiment, we show that the familiarity with the USV emitter also modulate the effect of USV playback during cocaine self-administration, since only stranger, but not familiar, 50-kHz USV decrease drug intake. Finally, to study the neurobiological substrate of those processes, we have tested the effects of the subthalamic nucleus (STN) lesion on these various conditions. STN-lesioned rats did not lever press much for any USV playback, whatever their emotional valence, nor did they seem able to differentiate familiar from stranger peer. Advocating for the existence of a vocal signature in rats, these results highlight the importance of ultrasonic communication in socio-affective influence of behavior, such as the influence of proximal social factors on drug consumption and confirm the role of the subthalamic nucleus on this influence.

Unveiling the neural correlates of habit in the dorsal striatum

We have recently reported sustained inhibition in the dorsomedial striatum (DMS) and sustained excitation in the dorsolateral striatum (DLS) during execution of a lever press sequence in a discrete-trials task promoting habit. This sustained dorsostriatal activity was present early on, and did not clearly change in step with improved performance over ten training sessions. Early onset of sequence-related neural activity could have resulted from rapid habitual learning promoted by presentation of lever cues, predicting reward availability and delivery. To test this hypothesis, we compared DLS and DMS spiking activity in the discrete trials habit-promoting task with two task variants that promote goal-directed behavior. Comparison of the three tasks revealed that mean neuronal spiking activity was generally sustained across the lever press sequence in the task promoting habit and characterized by overall excitation in DLS and inhibition in DMS relative to baseline. In contrast, mean activity differences in DLS and DMS were much less prominent, and most changes occurred transiently around individual lever presses, in the tasks promoting goal-directed behavior. These results indicate that sequence delineation cues, such as the lever cues in these studies, promote habitual behavior and that this habitual behavior is encoded in the striatum by cue-triggered sustained DLS excitation and DMS inhibition that likely reflects cue-elicited behavioral chunking.

Effect of Voluntary Adolescent Alcohol Consumption on Encoding of Decision-Related Variables in Prefrontal Cortex

ABSTRACTThe medial and orbitofrontal regions of prefrontal cortex (PFC) have been implicated in guiding optimal behavior and updating the economic value of rewards that result from choice behaviors. Both regions mature through adolescence into early adulthood and are thus vulnerable to exposure to neurotoxins, such as alcohol, during this critical developmental window. We sought to examine how voluntary alcohol consumption during adolescence would alter long-term PFC function and subsequent decision-making behavior in adulthood. Male and female rats were given adolescent intermittent ethanol (AIE) exposure to provide voluntary access to alcohol during the period of PFC maturation. In adulthood, we assessed the long-term effects on decision-making behavior using a risk task in adulthood, while concurrently recording neural activity in orbitofrontal cortex (OFC) and medial prefrontal cortex (mPFC). While control animals’ preferences for risky rewards increased with the likelihood of their delivery, AIE animals showed an overall reduction in their preferences for the risky option with higher levels of alcohol consumption, suggesting reduced discriminability of uncertain rewards and a shift away from the potential for reward omission. During task performance, neurons in mPFC and OFC responded to events (lever press, reward delivery). In mPFC, neurons with phasic increases at the time of lever press showed a reversal from larger elevations for risky presses in control animals to larger elevations for certain presses as prior alcohol consumption increased. Neurons in mPFC generally showed less discrimination of reward outcome with increased alcohol consumption as well. In OFC, responses to lever press were largely unaffected by AIE exposure. However, encoding of reward size in OFC showed differential effects in males and females. With higher alcohol consumption in males, OFC neurons showed largest excursions from baseline activity in response to largest reward, and smallest excursions for reward omission. This discrimination was reduced as prior alcohol consumption increased. In females, neurons with increased reward-activity, showed an overall higher level of activity due to stronger responses to certain rewards that were selected more frequently. Collectively the results show diminished capacity of PFC to encode decision-related elements to guide adaptive behavior and further clarify the lasting impact of adolescent alcohol use on neural function and behavior, even in the absence of continued use.