Homeostasis is a biological principle for regulation of essential physiological parameters within a set range. Behavioural responses due to deviation from homeostasis are critical for survival, but motivational processes engaged by physiological need states are incompletely understood. We examined motivational characteristics and dynamics of two separate neuron populations that regulate energy and fluid homeostasis by using cell type-specific activity manipulations in mice. We found that starvation-sensitive AGRP neurons exhibit properties consistent with a negative-valence teaching signal. Mice avoided activation of AGRP neurons, indicating that AGRP neuron activity has negative valence. AGRP neuron inhibition conditioned preference for flavours and places. Correspondingly, deep-brain calcium imaging revealed that AGRP neuron activity rapidly reduced in response to food-related cues. Complementary experiments activating thirst-promoting neurons also conditioned avoidance. Therefore, these need-sensing neurons condition preference for environmental cues associated with nutrient or water ingestion, which is learned through reduction of negative-valence signals during restoration of homeostasis.

Neural circuits for essential natural behaviors are shaped by selective pressure to coordinate reliable execution of flexible goal-directed actions. However, the structural and functional organization of survival-oriented circuits is poorly understood due to exceptionally complex neuroanatomy. This is exemplified by AGRP neurons, which are a molecularly defined population that is sufficient to rapidly coordinate voracious food seeking and consumption behaviors. Here, we use cell-type-specific techniques for neural circuit manipulation and projection-specific anatomical analysis to examine the organization of this critical homeostatic circuit that regulates feeding. We show that AGRP neuronal circuits use a segregated, parallel, and redundant output configuration. AGRP neuron axon projections that target different brain regions originate from distinct subpopulations, several of which are sufficient to independently evoke feeding. The concerted anatomical and functional analysis of AGRP neuron projection populations reveals a constellation of core forebrain nodes, which are part of an extended circuit that mediates feeding behavior.

Hypothalamic hypocretin/orexin (hcrt/orx) neurons coordinate sleep-wake cycles, reward seeking, and body energy balance. Neurochemical data suggest that hcrt/orx cells contain several transmitters, but what hcrt/orx cells release onto their projection targets is unknown. A major pathway by which hcrt/orx neurons are thought to promote arousal is through projections to tuberomammillary histamine (HA) neurons. To study the impact of the electrical activity in hcrt/orx cells on HA neurons, we genetically targeted the light-activated excitatory ion channel channelrhodopsin-2 (ChR2) to the plasma membrane of hcrt/orx cells, and performed patch-clamp recordings from HA cells in acute mouse brain slices. Stimulation of ChR2-containing fibers with millisecond flashes of blue light produced fast postsynaptic currents in HA neurons, with a high connection probability (Ϸ60% of HA cells were connected to Ϸ40% of hcrt/orx cells expressing ChR2). These inputs depended on tetrodotoxin-sensitive action potentials, had kinetics typical of glutamatergic responses mediated by AMPA receptors, were blocked by the AMPA receptor blocker CNQX, and displayed multiple forms of short-term plasticity (depression in Ϸ70% trials, facilitation in Ϸ30% trials, both often in the same cell). Furthermore, stimulation of hcrt/orx axons at physiological frequencies rapidly and reversibly increased action potential firing in HA cells, an effect that was abolished by blockade of AMPA receptors. These results provide the first functional evidence that hcrt/orx neurons are capable of fast glutamatergic control of their projection targets, and suggest that variations in electrical activity of hcrt/orx axons can induce rapid changes in long-range signals generated by HA neurons.