New preprint on the function of dopamine reward neurons
We just posted a new preprint on bioRxiv on the function of dopamine reward neurons in flies!
Dopamine neurons (DANs) have a well-known role in helping us learn the value of cues that predict reward. This role is conserved in Drosophila, which has provided a powerful model to dissect the mechanisms underlying the role of DANs in reward learning. But with all the focus on how DANs mediate reward learning, less is known about how they regulate innate behaviors - behaviors that don’t require learning. In mammals, dopamine invigorates ongoing behaviors and has a role in promoting feeding behavior, so we set out to investigate whether reward DANs in Drosophila also regulate feeding and other innate behaviors.
Image of reward DANs in the fly brain stained green (created from data in the FlyLight database).
We began by using optogenetic manipulations to examine how reward-encoding DANs in Drosophila (called “PAM” DANs) influence sugar feeding. While we expected that PAM DAN activation would promote feeding, we were surprised to find the opposite: their activation strongly suppressed feeding in a variety of conditions. Through optogenetic and behavioral studies, we show that although PAM DAN activation robustly drives reward learning, this activation is innately aversive to flies and drives aversive locomotor and feeding responses. Activating smaller subsets of PAM DANs revealed that the effects on innate and learned behaviors are dissociable, with individual subsets being capable of driving learned attraction but not innate aversion.
These results raise some intriguing questions. First, what is the circuit mechanism by which DANs cause opposing effects on innate versus learned behaviors? Based on known circuitry and previous work, it’s likely that the innate aversion and learned attraction elicited by DANs arise from distinct effects – direct activation versus synaptic plasticity – on the same target neurons. We performed experiments to examine how these two mechanisms interact, revealing that the pathway for innate aversion overrides the mechanism for learned attraction.
Second, why would it be adaptive for reward DANs to cause opposing effects on innate and learned behaviors? One possibility is that this opposing relationship represents a mechanism to compare current reward (encoded by immediate DAN activity) with expected reward based on reward-associated cues that were previously learned (encoded by DAN-evoked synaptic plasticity). This model implies that a separate pathway mediates innate attraction to rewards and overrides DAN-mediated aversion.
Overall, we think these results uncover new and interesting aspects regarding the function of the most well-studied “reward neurons” in the fly brain, showing that they have distinct roles in guiding innate and learned behavior.
This study was led by Fiorella Lozada-Perdomo (former Research Specialist) with the help of Yuzhen Chen (Research Specialist), Ruby Jacobs (Research Specialist), Joyce Yeo (summer undergraduate from Colby), Maia Yang (former Emory undergraduate), and Janhavi Bhalerao (former PhD rotation student). Congratulations to all the authors!
