Internal state dependent control of feeding behaviour via hippocampal ghrelin signalling

Abstract

Hunger is an internal state that not only invigorates behaviour towards feeding, but also acts as a contextual cue for the higher-order control of anticipatory feeding-related behaviour. The ventral hippocampus is a brain region important in encoding context, but how internal contexts such as hunger are represented in hippocampal circuits is not known. Pyramidal neurons in the ventral hippocampus, and in particular within the ventral CA1/subiculum border (vS) express the receptor for the peripheral hunger hormone ghrelin, and ghrelin is known to cross the blood brain barrier and directly influence hippocampal circuitry. However, what role vS neurons play during feeding related behaviour, and how ghrelin influences this role has not been directly investigated. In this study, we used a combination of whole-cell electrophysiology, optogenetics and molecular knockdown together with in vivo calcium imaging in mice to investigate the role of vS during feeding behaviour across different states of hunger. We found that activity of a unique subpopulation of vS neurons that project to the nucleus accumbens (vS-NAc) were active specifically when animals approached and investigated food, and that that this activity inhibited the transition to begin eating. Increases in peripheral ghrelin reduced vS-NAc activity during this anticipatory phase of feeding behaviour, by increasing the influence of synaptic inhibition. Furthermore, this effect required postsynaptic GHSR1a expression in vS-NAc neurons, suggesting a direct role of ghrelin signalling. Consistent with this role of hippocampal ghrelin signalling, removal of GHSR1a from vS-NAc neurons impaired ghrelin-induced changes in feeding-related behaviour. Together, these experiments define a ghrelin-sensitive hippocampal circuit that informs the decision to eat based on internal state.

A hippocampus-accumbens code guides goal-directed appetitive behavior

“…Activity of dHPC neurons, and especially those projecting to NAc, thus seems to mirror the activity within NAc: Inhibition of NAc D1R medium spiny neurons (MSNs) has been observed during lick behavior 74,92,93 and was shown to be both necessary 93,94 and sufficient 93,95 for consummatory licking. We found that the origin of these signals may be localized upstream, similar to recent findings of an inhibitory permissive drive for licking in ventral HPC (vHPC) projections to the NAc 74,75 and that inhibiting these projections facilitates licking 74 . This suggests a similar involvement of dHPC-and vHPC-accumbens projections for consummatory licking.…”

“…Contrasting this consumption-related inhibition, we found appetitive lick-related dHPC→NAc excitation and optogenetic induction of lick-related behaviors. This dichotomy likely reflects previously described differences in the NAc's role for consummatory and appetitive behaviors 61 : vHPC→NAc activity was shown to increase during active food investigation 75 and around the time of lever pressing to obtain a reward 96 , while optogenetic stimulation of this projection facilitated nose poke and lever pressing behaviors 31,34 . Given that optogenetic stimulation of D2R MSNs in the NAc shell was also found to produce repetitive jaw movements 97 , it is likely that D1R and D2R MSNs, both of which are targeted by HPC axons 30,98,99 , might play dichotomous roles in appetitive and consummatory behaviors 30,100 .…”

“…This view is also supported by a recent study on odor-outcome learning that found robust odor coding in dCA1, while vCA1 odor responses only emerged after acquiring positive valence 73 . In line with this, much of the past literature on spatial reward learning has focused on the prominent NAc afferents from vHPC 30,31,[74][75][76] . Only recently have the anatomically weaker dHPC afferents to the NAc started to receive attention.…”

Build instructions for a long term behavioural enclosure for measuring motivational switching in mice

“…The recorded behavioural sequences allowed analysing motivation transition likelihoods 10,11 (Figure 12H). As expected from the efficient organization of a small number of multi-trial blocks per hour, transition types within the foraging environment were most frequent (food-to-food, food-to-drink, drink-to-drink and drink-to-food).…”

“…On a slower timescale, a probabilistic behavioural satiety sequence of feeding-grooming-resting, occurring over the course of an hour has been described in rodents and crayfish 8,9 . Transitions between sequential motivated behaviours in the home cage environment are seldom measured at a high temporal resolution [10][11][12] . This may be because motivational switching can be highly disordered in the home cage where the goal objects are constantly accessible.…”