Challenging current models of valence encoding in the mammalian brain

In an ever-changing environment, our brains evolved to assign valence to stimuli in order to survive. A positive (rewarding) valence stimulus elicits approach, whereas a negative (aversive) valence stimulus elicits avoidance. One of the key regions involved in reward and aversion is the nucleus accumbens (NAc), mostly composed of medium spiny neurons, divided into those expressing dopamine receptor D1 and dynorphin, and those expressing D2 and enkephalin. D1 and D2 neurons were assumed to encode opposing valence, but recent data challenged this model. In fact, though this region is canonically associated with reward and aversion, to date, it is still not known how valence is encoded in the NAc, nor the role of endogenous opioids (dynorphin and enkephalin) in this process.

Therefore, the main goal of this project is to determine how NAc neurons encode valence and what is the contribution of endogenous opioids for this mechanism.

To understand how valence is encoded in this region, we will record the neuronal activity of rodents performing behavioral tasks with opposing valences. Moreover, we will measure the release of endogenous opioids (dynorphin and enkephalin) during behavior using novel opioid fluorescent sensors. This approach will unravel how NAc neurons encodes valence, the role of endogenous opioids, and how these signals are decoded in the circuit to drive behavior with unprecedent temporal and spatial resolution.

Understanding how valence is encoded in this region enables a better grasp of the pathological mechanisms underlying disorders that present NAc dysfunction and deficits in reward and aversion, namely depression and addiction.

We have performed recordings of NAc neurons activity during exposure to positive and negative valence stimuli and during Pavlovian conditioning, i.e. during the process of learning to associate a neutral stimulus with a positive or negative outcome. Our results show that both D1- and D2-neurons are co-activated similarly in positive and negative valence conditions, contradicting the classical model of functional opposition. Moreover, valence is encoded at a population level, rather than at individual neuron level for both D1- and D2-neuronal populations.
We also found that both populations present a similar pattern of activity during Pavlovian learning, in agreement with a concurrent rather than opposing functioning of the two populations.
In parallel, using novel opioid fluorescent sensors (developed by Lin Tian and colleagues), we were able to detect dynorphin and enkephalin release in freely behaving animals. We observe that these opioids are released distinctly during different stages of Pavlovian learning, and are currently trying to understand why, and what is the impact of this opioidergic signaling in the learning process. This is a major breakthrough in the field, since opioids were technically very difficult to detect in vivo with spatial and temporal resolution.
Our team has also identified different neuronal pathways that are preferentially recruited during exposure to positive or negative valence stimuli. This can allow us to identify novel genetic markers to label specific valence ensembles. In addition, we are investigating some of these hits in more detail, by performing pharmacological and genetic manipulations in order to see if one can alter rewarding/aversive responses.
Moreover, we showed that optogenetic activation of D2-neurons during different stages of a positive valence task can have opposing roles in behavior, suggesting that the contribution of NAc neurons for rewarding/aversive behaviors is more complex than anticipated.

Our data shows that both D1- and D2-neuronal populations of the NAc are similarly co-activated during positive and negative valence stimuli and associated cues, contradicting the classical model of functional opposition of these two populations. Until the end of the project, we expect that we will be able to determine how NAc neurons represent positive and negative valence information. To complement this work, we are also investigating if one can distinguish valence neurons based on their molecular signature. If this is the case, one expects to identify novel markers that can be used to label and manipulate these neurons.
Moreover, in this project we validated novel sensors to detect endogenous opioids in vivo, which now opens new possibilities to better understand the role of opioidergic signaling in behavior.

Periodic Reporting for period 3 - VALENCE (Challenging current models of valence encoding in the mammalian brain)

Download Download the content of the page