A new circuit underlying the renewal of appetitive Pavlovian responses: Commentary on Brown and Chaudhri (2022)

Imagine returning to a favourite bakery after foregoing sweet treats to start the new year. Would the sights and smells of freshly baked goods override your health-guided intentions? People often experience cravings when they encounter cues and contexts that remind them of their preferred indulgences. So-called cue reactivity is a core component of psychopathologies that contend with relapse as a barrier to recovery, such as binge eating and alcohol use disorder. Reactivity to cues and contexts associated with palatable foods or alcohol are enduring relapse triggers (Gearhardt & DiFeliceantonio, 2022; Valyear et al., 2022), even in people that have undergone surgical treatment interventions (Budak & Thomas, 2009; Ho et al., 2018). Uncovering the neural circuits that subserve the capacity for cues and contexts to encourage the pursuit of rewards, like palatable foods and alcohol, informs efforts to understand and prevent relapse. In the current issue of the European Journal of Neuroscience, Brown and Chaudhri (2022) describe a role for prefrontal inputs to the thalamus in the renewal of appetitive Pavlovian responding (Brown & Chaudhri, 2022). Renewal is a learning process often used to study relapse. In appetitive Pavlovian renewal, subjects are trained to respond to a conditioned stimulus (CS) that predicts a reinforcer (e.g. food, alcohol) in a distinct training context; responses are usually head entries into a receptacle wherein the reinforcer is delivered. CS responding is subsequently extinguished by presenting the CS without the reinforcer in a different extinction context. After responding decreases to low levels in extinction, subjects are returned to the original training context for the renewal test where conditioned responding to the CS is assessed in the absence of the reinforcer. Generally, responding to the CS renews to high levels during the renewal test. There are clear parallels between renewal and relapse episodes that occur when people encounter contexts associated with palatable foods or alcohol after achieving abstinence in a different context (e.g. treatment facility). Brown and Chaudhri (2022) demonstrate that optogenetic activation of the projection from the infralimbic (IL), but not prelimbic (PL), cortex to the paraventricular nucleus of the thalamus (PVT) reduces renewal. This effect is so pronounced that it reduced responding during the renewal test to extinction levels. In other words, subjects maintained extinction-level responding as though a context change never occurred. Other measures of conditioned responding, like the latency, duration and probability of CS responses, confirmed that renewal was robustly attenuated by stimulating the IL-to-PVT circuit. Reducing renewal is a meaningful result from a translational perspective. For people attempting to abstain from palatable foods or alcohol, environments associated with prior consumption are powerful relapse triggers. A behavioural intervention designed to reduce cue reactivity is cue exposure therapy, which involves repeatedly presenting cues in a controlled setting. Although cue exposure therapy can reduce cravings in people (Collins & Brandon, 2002; Toro et al., 2003), its effectiveness varies with changes in context. How might the IL-to-PVT circuit contribute to relapse? The IL has been implicated in addiction, mainly for its role in extinction retrieval (Peters et al., 2009). Despite a return to the training context, which evokes memories of rewards consumed (Figure 1, left), IL-to-PVT stimulation seems to promote the retrieval of extinction learning, thus discouraging the renewal of CS responding (Brown & Chaudhri, 2022) (Figure 1, right). The IL is also involved in processes separate from extinction. For example, optogenetic stimulation of the IL (Do-Monte et al., 2015) and its projection to the nucleus accumbens (NAc) shell (Villaruel et al., 2022) can suppress appetitive behaviour in the absence of extinction training. Further, Brown and Chaudhri (2022) showed that the IL-to-PVT circuit supports self-stimulation behaviour, indicating that activity in this circuit is reinforcing (Brown & Chaudhri, 2022). Thus, the IL-to-PVT circuit is implicated in numerous behavioural processes, including extinction retrieval, suppression of appetitive responding and reinforcement. The IL and PL both send glutamatergic projections to the PVT, but the psychological consequences of activating these inputs differs. Optogenetic stimulation of the IL-to-PVT circuit wholly attenuated renewal, whereas the same stimulation of the PL-to-PVT circuit left renewal intact (Brown & Chaudhri, 2022). Further, optogenetic stimulation of the IL-to-PVT, but not PL-to-PVT circuit, supported self-stimulation behaviour (Brown & Chaudhri, 2022). How might the IL and PL differentially modulate PVT activity to influence behaviour? The IL and PL similarly innervate the PVT, sending the densest projections to the posterior, rather than anterior, aspects of the PVT (Li & Kirouac, 2012), ruling out PVT subregion differences as an explanation for divergent behavioural consequences of stimulating these circuits (Gao et al., 2020). Disparate behavioural consequences of stimulating IL and PL inputs to the PVT may arise from recruitment of separate populations of PVT neurons. It was recently discovered that two subsets of cells in the PVT, which are intermixed throughout the anterior and posterior aspects of the PVT, differentially innervate the dorsomedial and ventromedial NAc (Dong et al., 2017; Li et al., 2021). Further, it was shown that PVT cells projecting to the dorsomedial NAc are highly collateralized, sending dense projections to diverse brain areas including the amygdala, hippocampus and hypothalamus (Li et al., 2021). Differently, PVT inputs to the ventromedial NAc shell are considerably less collateralized, projecting densely to only a few brain areas (Li et al., 2021). Brown and Chaudhri (2022) raise the possibility that the reinforcing properties of IL-to-PVT stimulation, not shared with PL-to-PVT stimulation, arise from a downstream projection from the PVT to the NAc shell, which coincides with the finding that mice self-stimulate PVT inputs to the dorsomedial NAc shell (Lafferty et al., 2020). It will be important for future experiments to identify whether IL and PL projections to the PVT synapse on cells that preferentially project to the dorsomedial and ventromedial NAc shell, respectively. Distinct contributions of IL-to-PVT projections in extinction retrieval, suppression of appetitive responding and reinforcement might also arise from diverse downstream collateral projections from the PVT to other areas. Neural circuits serving to promote the renewal of Pavlovian appetitive responding are likely engaged during relapse episodes that occur in contexts associated with palatable foods or alcohol. Brown and Chaudhri (2002) demonstrate that activity in the IL-to-PVT, but not PL-to-PVT, circuit suppresses renewal and supports self-stimulation. It is possible that IL and PL projections engage distinct cell populations in the PVT, consequently recruiting differential neural circuits and influencing separable psychological processes. Identifying a point of divergence in what are likely multi-region and heavily collateralized networks is critical. Probing circuits emanating from the PVT and impinged upon by the IL and PL will uncover the function of complex networks and their roles psychopathology. Hopefully, describing expansive multi-synaptic circuits will identify targets for therapeutic interventions that complement cue exposure therapy for the treatment of psychopathology. Milan D Valyear: Conceptualization; writing - original draft; writing - review and editing. Jonathan Philip Britt: Conceptualization; writing - review and editing. The authors have no conflicts of interest to declare. The peer review history for this article is available at https://publons.com/publon/10.1111/ejn.15924.

Récupéré en direct depuis OpenAlex et désinversé. Les résumés ne sont pas conservés dans cette base de données : les index inversés représentent 8,6 Go des 9,3 Go de texte de la base, et le serveur dispose de 13 Go libres.