The mushroom bodies (MBs) are known to be the most prominent olfactory memory center in Drosophila. We could show that, based on the anatomical connections between the MB Kenyon cells and dopaminergic pPAM neurons, there is a functional KC-to-pPAM feedback loop in fruit fly larvae. Optogenetic activation of KCs in the absence of physical sugar reward (substitution assay) leads to the formation of a robust appetitive memory in larvae. However, if the pPAM neurons are ablated, the larvae fail to form sugar memory. Thus, it is hypothesized that a comparable functional feedback loop from the KCs to negative DANs could lead to the formation of aversive memories. Interestingly, training the larvae in our substitution assay and testing them on a salt plate does not lead to a significant memory expression. However, in such a test situation, recall of appetitive memory should not be affected, as larvae are not exposed to a rewarding stimulus in the test. Nevertheless, considering that salt exposure results in the expression of an aversive memory, if such is formed during training, it can be suggested that both types of memory are expressed during test, and by that overall memory expression is nullified. Strikingly, larvae express robust appetitive memory on salt test plate after substitution learning when the pPAM neurons are ablated. It is not clear whether pPAM ablation impairs acquisition of aversive memory during training or its expression during test, however this phenotype is not due to inversion in the quality of salt in pPAM ablated animals. Given that rewarding and punishing DANs signal to different MB compartments (punishing DANs from the DL1 cluster -> peduncle and vertical lobe, rewarding DANs from the pPAM cluster -> medial lobe, the complex MB circuitry involving e.g. KC-to-KC connections and feed-across VL-to-ML loops, and the theory of KC-to-MBON synaptic depression, it is tempting to speculate that after ablation of pPAM neurons other so far unidentified functional signaling loops (feedback or feedforward) prevail in order to change the balance in depriving approach/avoidance behavior eliciting synaptic connections, and thus trigger searching behavior in presence of a punishing stimulus. However, appetitive memory expression during exposure to an aversive stimulus was so far not described, so dissection of the process underlying such behavior would be of great benefit for understanding the functional connectivity and neuronal mechanisms involved in appetitive vs. aversive learning. Interestingly, our latest results suggest that optogenetic activation of KCs might have an effect on naïve salt preference. So far, there are no studies on the role of the MBs in naïve gustatory behavior. Recently, MB signaling was shown to be involved in naïve olfactory behavior, so their effect also on high order taste perception is conceivable. Thus, during learning there might be an interference between the avoidance and approach mediating compartments within the MB circuitry on one hand. On the other hand, this could be coupled to a change in the naïve preference/perception for and the negative valence of salt, and/or motivational drive of the larvae.
