When we are short of energy, circulating neurohormones and modulators prime our body and brain to mobilise our remaining metabolic resources and to promote food-seeking and consummatory behaviours. Although the drive to relieve our hunger can be all encompassing, our brains retain the capacity to do other things if required.
Prior work from the Waddell group established that both hunger and thirst motivated resource seeking behaviours are under the control of the fly’s dopaminergic system (Krashes et al., 2009; Senapati et al., 2019), which is composed of about 200 neurons representing rewards and 10-fold fewer representing aversive things (Li et al., 2020). Their studies have shown that a key element of the deprivation-state specific control of the expression of food- and water-seeking memories is provided by neuropeptidergic inhibition of two particular types of aversive dopaminergic neurons (Krashes et al., 2009; Senapati et al., 2019). Using the dopaminergic system this way for motivational purpose raises the question of how the system retains the ability to signal aversion, when these neurons are inhibited to promote resource seeking.
In the new work led by postdoctoral fellow Eleonora Meschi, the group provide an answer to this question. Dr. Meschi discovered that endocrine release of the fly’s functional equivalent of glucagon, called adipokinetic hormone (AKH), compensates for the hunger-dependent suppression of aversive dopaminergic neuron activity, by facilitating the strength of input pathways that convey punitive information to these neurons.
Eleonora says, 'The project developed from my original and completely unexpected finding that hungry flies lacking glucagon/AKH had normal food-reinforced learning but poor shock-reinforced learning. By tracking AKH signalling from the bloodstream to the brain we discovered why this was the case. AKH regulates the activity of neural pathways that relay unpleasant signals, like painful shock and bitter taste, to dopaminergic neurons that write aversive memories.'
When hungry, flies release more AKH into their circulating haemolymph. Meschi demonstrated that AKH influences aversive learning behaviour by activating four large neurons in the base of the brain that express the AKH receptor. Interestingly, the cell bodies of these neurons reside outside of the glial covering of the brain, giving them direct access to the circulatory environment, and nutrient status, of the fly’s body. Using connectomics, research technician Georgia Dempsey and postdoctoral fellow Nils Otto, mapped the neurons downstream of the AKH response neurons, which revealed 135 neurons that could be clustered into 38 distinct neuronal types. Critically, these neurons project axons that ascend from the base of the brain to the top, where many of them were found to selectively synapse onto aversively reinforcing dopaminergic neurons.
With DPhil student Lucille Duquenoy, Dr. Meschi used live-imaging of a transgenic fluorescent dopamine sensor to show that hunger increased the shock-evoked release of dopamine from the aversive dopaminergic neurons and that this facilitation required AKH signalling. Further genetic experiments established that AKH signalling directs modulation of the ascending neurons so that they more efficiently relay punitive information, such as electric shock and bitter taste, from the periphery to the aversive dopamine neurons that are required for aversive learning.
Lucille comments, 'I have been using the new genetically-encoded GRAB dopamine sensors from Yulong Li's group to measure stimulus-evoked dopamine release from specific neurons in the fly brain. With Eleonora, I found that electric shocks produce more dopamine release from aversive neurons if the flies are chronically hungry and that shock-triggered dopamine release was greatly reduced in flies with no AKH signalling.'
In the published paper the Waddell group note that similar mechanisms may exist in the mammalian brain, between nutrient-responsive neurons in the hypothalamus, and neurons that project from there to dopaminergic neurons in the ventral tegmental area.
Read the paper here