Neural basis of hunger-driven behaviour in Drosophila

Abstract

Hunger is a motivational state that drives eating and food-seeking behaviour. In a psychological sense, hunger sets the goal that guides an animal in the pursuit of food. The biological basis underlying this purposive, goal-directed nature of hunger has been under intense investigation. With its rich behaviour-al repertoire and genetically tractable nervous system, the fruit fly Drosophila melanogaster has emerged as an excellent model system for studying the neural basis of hunger and hunger-driven behaviour. Here, we review our current understanding of how hunger is sensed, encoded and translated into foraging and feeding behaviours in the fruit fly.

Full text

royalsocietypublishing.org/journal/rsob
Review
Cite this article: Lin S, Senapati B, Tsao C-H.
2019 Neural basis of hunger-driven behaviour
in Drosophila.Open Biol. 9: 180259.
http://dx.doi.org/10.1098/rsob.180259
Received: 12 December 2018
Accepted: 4 March 2019
Subject Area:
neuroscience
Keywords:
Drosophila, neural circuits, hunger,
feeding behaviour, food-seeking behaviour
Author for correspondence:
Suewei Lin
e-mail: sueweilin@gate.sinica.edu.tw
A contribution to the special collection
commemorating the 90th anniversary of
Academia Sinica.
Neural basis of hunger-driven behaviour
in Drosophila
Suewei Lin1,2, Bhagyashree Senapati1,2 and Chang-Hui Tsao1
1
Institute of Molecular Biology, Academia Sinica, Taipei, Taiwan, Republic of China
2
Molecular and Cell Biology, Taiwan International Graduate Program, Academia Sinica and Graduate Institute of
Life Sciences, National Defense Medical Center, Taipei, Taiwan, Republic of China
SL, 0000-0001-7079-7818
Hunger is a motivational state that drives eating and food-seeking behaviour.
In a psychological sense, hunger sets the goal that guides an animal in the
pursuit of food. The biological basis underlying this purposive, goal-directed
nature of hunger has been under intense investigation. With its rich behaviour-
al repertoire and genetically tractable nervous system, the fruit fly Drosophila
melanogaster has emerged as an excellent model system for studying the
neural basis of hunger and hunger-driven behaviour. Here, we review our
current understanding of how hunger is sensed, encoded and translated into
foraging and feeding behaviours in the fruit fly.
1. Introduction
Hunger is an internal state elicited by lack of nutrients and energy in the body.
It is difficult to establish if the subjective feeling of hunger is unique to human,
but the foraging and consumptive behaviours evoked by hunger are nearly uni-
versal among mobile animals. Since foraging for food is costly with respect to
energy and physical risk, hunger can be considered a guidance signal ensuring
that animals only seek food when there is a need [1]. Defects in the hunger
system can lead to malnutrition, obesity, eating disorders and even death.
Hunger has very broad effects on an animal’s behaviour. It can increase an ani-
mal’s risk tolerance [2], elevate its locomotion [3–5], change its sensitivity to
external stimuli [6,7] and affect its decision-making [8,9]. Effectively, hunger
sets a goal—acquiring food—for an animal, and the animal can use whatever
means are available to it to achieve that goal. Given these broad effects of
hunger, unravelling its underlying neural mechanisms can be challenging.
Compared to the extreme complexity of the mammalian brain, the nervous
system of the fruit fly Drosophila melanogaster is simpler and genetic tools for
labelling defined neuronal populations in the fly are more developed. These
features have made the fly an excellent model for establishing fundamental
neural principles of hunger. In this article, we review what has been learned
from both fly larvae and adult flies in terms of how nutrient needs are
sensed, how these needs are encoded as diverse hunger and satiety signals,
and how these signals are translated by both peripheral and central neural cir-
cuits to elicit foraging and feeding behaviour. Since neural mechanisms in
larvae and adult flies may not always be the same, we always clarify if a con-
clusion is based on larval or adult studies.
2. Sensing nutrient needs
Flies change their food preferences in response to lack of calories, amino acids
or salts. This nutrient-specific hunger-driven behaviour implies the existence of
internal sensors for specific food components, and several such molecular sen-
sors have been identified in the fly. These sensors monitor nutrient levels and
&2019 The Authors. Published by the Royal Society under the terms of the Creative Commons Attribution
License http://creativecommons.org/licenses/by/4.0/, which permits unrestricted use, provided the original
author and source are credited.

regulate fly feeding behaviour accordingly when a nutrient
concentration falls below a normal level.
Sugar is a major nutrient in food, and glucose is a primary
source of energy in the body. Circulating sugar level is a good
estimate for body energy state. Therefore, it is not surprising
that the fly nervous system is equipped with the ability
to directly sense glucose levels. Multiple lines of evidence
show that flies can detect the nutrient value of sugar indepen-
dently of its taste. Adult flies lacking both Gr5a and Gr64a, the
taste receptors for sugar, still exhibit a preference for sucrose
over plain agarin atwo-choice assay after prolonged starvation
[10]. This sugar preference appears to be correlated with
haemolymph sugar content, and it can be enhanced by a
drug that reduces circulating glucose and trehalose in the
haemolymph. Flies are also capable of learning the nutritional
value of sugar. Pairing sugar with odours can condition adult
flies to form an appetitive olfactory memory that lasts for days.
This memory becomes less robust if sweet but non-nutritious
sugar is used. However, the robustness of the memory can be
restored if the non-nutritious sugar is supplemented with a
tasteless but nutritious substance [11,12].
Several tissues or cell types in the fly are considered to be
glucose-sensitive. Insulin-producing cells (IPCs) in the brain
and the corpora cardiaca that releases adipokinetic hormone
(AKH) control feeding and foraging behaviours in both
larvae and adult flies [4,13– 16], and glucose contrastingly
regulates their activities. Although direct evidence is still lack-
ing, it is proposed that the glucose sensitivity of these cells is
mediated by a mechanism similar to that of mammalian
pancreatic a- and b-cells [17,18]. Fat body in the fly is a
nutrient-sensing tissue equivalent to mammalian liver and adi-
pose. Larval fat body has also been shown to sense glucose via
a G protein-coupled receptor, Bride of sevenless (BOSS) [19].
BOSS protein contains a fragment homologous to a trehalose
taste receptor, but it responds specifically to glucose, i.e. not
to trehalose or sucrose. Furthermore, boss mutant adult flies
and flies in which boss has been specifically knocked down in
fat body exhibit increased food intake [20]. In addition to
sensing glucose, gustatory receptor 43a (Gr43a) functions as
a fructose receptor in the adult fly brain [21]. Gr43a is expressed
in two to four neurons per hemisphere of the central brain of
adult flies. These neurons respond to fructose, but not other
tested sugars (including sucrose, glucose and trehalose). This
fructose response is Gr43a-dependent and is completely
abolished in gr43a mutant flies. Although glucose and treha-
lose are the main haemolymph sugars in insects, large
increases in fructose concentration in the haemolymph have
been observed after flies have eaten a sugar meal, even if the
sugar consumed is glucose or sorbitol. Thus, haemolymph
fructose can serve as an estimate for carbohydrate consump-
tion. Consistent with their role as internal nutrient sensors,
brain Gr43a neurons promote sugar intake in hungry flies
and suppress sugar feeding in satiated flies [21].
Amino acids are another primary food nutrient. When they
encounter food lacking essential amino acids, Drosophila larvae
initially consume it as normal food but, after approximately 1 h,
food intake is reduced by 20–25% compared with larvae that
eat normal food. This delayed reduction in food intake is
mediated by a small group of amino acid-sensitive dopamin-
ergic neurons in the brain [22]. These neurons are activated
when larval brain explant is perfused with an imbalanced
amino acid mix, and this response requires the amino acid trans-
porter slimfast and the intracellular amino acid-sensor GCN2. In
addition to neurons, larval fat body has also been demonstrated
to use slimfast and TOR/S6 K to sense amino acids and secrete
endocrine signals to regulate functions of the nervous system
[23,24]. Adult flies also exhibit nutrient-specific hunger, exhibit-
ing a preference for protein-rich food when they are deprived of
amino acids [25,26]. Adult fat body and some neurons in the
adult brain, including a group of dopaminergic neurons, have
been shown to respond to protein starvation, but their sensory
mechanisms remain unclear [13,27,28]. Moreover, the amino
acid-sensitive TOR/S6 K pathway also appears to regulate
amino acid state-dependent food preferences in adult flies,
but the neurons on which TOR/S6 K signalling acts have not
yet been identified [25,26].
Food components are complex. Our understanding of how
the nervous system detects the need for specific nutrients
remains profoundly insufficient. Whether there are additional
internal sensors in the fly for other nutrients such as salt, lipids
or minerals, as well as the molecular nature of these sensors, are
exciting questions for future research.
3. Neural coding of hunger and satiety
states
A diverse array of neuronal signals induced by hunger and
satiety states has been identified in the fly (figure 1). Combi-
nations of these signals can be considered representations of
hunger states. Most of these signals are neuropeptides, which
are modulatory and can potentially work long-range to con-
trol multiple neural circuits in the nervous system.
Insulin-like peptides (DILPs) and Unpaired 2 (Upd2) are
the fly equivalents of mammalian insulin and leptin. The fly
has eight different DILPs. Expression of the eight DILPs is
tissue-specific and dynamic during development [29– 31].
There is only one known insulin receptor in the fly, but insulin
DILP
AKH
Upd2
PTP
CCHa2
FIT
Upd1
NPF
sNPF
AstA
Hugin
LK
CRZ
DSK
5HT
SIFamide
IPCs
fat body
MIP
54
13
87
Taotie neurons
EB R4 neurons
94
40
72
86
86
Gr43a neurons
71
55
55
90
hunger signals satiety signals
activation
inhibition
activatin or inhibition
79
91
38
42
72
42
37
81
65
Figure 1. Hunger and satiety signals and their interactions. The numbers on
the lines indicate the references in which evidence for the indicated inter-
action is presented. The shapes that outline the numbers denote whether
the evidence supporting the indicated interaction are from larval studies
(circle), adult studies (square) or both (diamond). Gr43a neurons, Taotie
neurons and EB R4 neurons were identified in adult flies.
royalsocietypublishing.org/journal/rsob Open Biol. 9: 180259
2

signalling has been shown to regulate diverse biological pro-
cesses, including cell growth, longevity, feeding and food
foraging behaviour [6,15,29,32–34]. DILP2, 3 and 5 are co-
expressed in adult and larval IPCs [35,36]. Secretion of DILPs
from the IPCs is hunger-dependent [24,37]. When larvae are
nutrient-deprived, DILPs accumulate in the IPCs, and their
concentrations decline in the haemolymph. DILPs function as
satiety signals that attenuate feeding motivation. Overexpres-
sion of DILPs in larval IPCs reduces feeding activity in
starved larvae. By contrast, downregulation of DILP release
promotes feeding in larvae and increases their acceptance of
low-quality foods, suggesting elevated feeding motivation
[38]. In adult flies, DILPs have also been shown to act as satiety
signals modulating neural circuits that control feeding and
food-seeking behaviour [6,16,32,39]. The nutrient state-depen-
dent control of DILP release in larvae requires Upd2, which is
secreted from fat body [37]. Under starvation conditions, upd2
transcript levels are greatly reduced. In the fed state, Upd2 acts
on its receptor Domeless to activate the JAK/STAT signalling
pathway in a population of GABAergic neurons that project
onto the larval IPCs. Active JAK/STAT signal lowers the neur-
onal activity of these GABAergic neurons and releases their
inhibitory effect on the IPCs, leading to DILP release. Human
leptin fully rescues upd2 mutant fly phenotypes, suggesting
that they are functional homologues. Intriguingly, despite the
direct stimulatory effect of Upd2 on DILP release, knockdown
of upd2 in fat body mainly impairs body growth and has a
minimal effect on feeding and food-seeking behaviour in
both larvae and adult flies [37,40]. Recently, Unpaired 1
(Upd1), another fly leptin-like peptide, was revealed to fulfil
these roles of Upd2 upon upd2 knockdown in the adult fly
[40]. Unlike Upd2, which is secreted from fat body, Upd1 is
derived from a small cluster of neurons in the brain. Fed flies
have higher levels of upd1 mRNA and less detectable proteins
in Upd1-expressing neurons compared to hungry flies,
suggesting that the nutrient-rich state may increase both tran-
scription and release of Upd1. Importantly, pan-neuronal
knockdown of upd1 in fed flies increases food intake and
heightens their responses to food odour. Thus, Upd1 and
Upd2 might work in concert to signal satiety states in the fly.
AKH is the fly analogue of mammalian glucagon. Akh is
expressed in the corpora cardiaca from late embryo to adult
stages [4]. Starvation induces AKH release into haemolymph
to signal hunger [18]. AKH induces utilization of stored
energy by stimulating lipolysis, glycogenolysis and trehalose
release in larval and adult fat bodies [18,41]. AKH and DILP
signals are mutually inhibitory. Ablation of IPCs increases
AKH expression, whereas ablation of the corpora cardiaca
enhances DILP3 expression in both larvae and adult flies
[42]. Adult flies lacking AKH are more resistant to starvation
and do not exhibit starvation-induced hyperactivity [4], pheno-
types regulated by AKH via a small group of octopaminergic
neurons in the brain [33]. AKH also regulates adult feeding be-
haviour by directly activating four interoceptive SEZ neurons
(ISNs) in the suboesophegeal zone (SEZ), as well as by redu-
cing bitter sensitivity in adult flies [16,43].
Another hunger signal in the fly, Neuropeptide F (NPF), is
a homologue of mammalian Neuropeptide Y (NPY) [44]. NPF
is expressed in the brain and endocrine cells of the midgut in
both larvae and adult flies. The role of midgut NPF is less
clear, but brain-derived NPF appears to facilitate feeding and
foraging behaviour in starved flies. In larvae, NPF is expressed
in four neurons in the brain. Ablation of these neurons results
in reduced feeding behaviour, whereas broad overexpression
of NPF in the nervous system prolongs the feeding phase of
third instar larvae [45]. Furthermore, food-deprived larvae
exhibit a higher tolerance for feeding on low-quality or noxious
foods. Blocking neurotransmission of NPF-expressing neurons
impairs this starvation-induced tolerance [15,38]. The adult
brain contains approximately 30 NPF-positive neurons [46].
Activation of these NPFneurons increases gustatorysensitivity
of adult flies to sugar and promotes food intake [43,47].
The adult NPF neurons express Domeless and are inhibited
by the satiety signal Upd1 [40]. They also respond to olfactory
inputs, and these responses are positively correlated with the
attractiveness of the presented odour [48]. Therefore, NPF
neurons may integrate sensory and internal state information
to regulate foraging behaviour. NPF also mediates hunger-
dependent expression of food memory. Starved adult flies
can be taught to associate sugar with odours that have no
innate appetitive value, and the resulting olfactory memory
is only expressed when flies are hungry [49]. Stimulation of
NPF neurons is sufficient to mimic the hunger state and pro-
mote sugar-odour memory expression in food-satiated adult
flies. Consistently, npfr mutant adult flies fail to express this
memory when they are food-deprived, behaving like non-
hungry flies [50].
Flies have another NPY-like peptide, short neuropeptide F
(sNPF). NPF and sNPF are not evolutionarily closely related
[51]. Nevertheless, sNPF also regulates multiple aspects of
hunger-driven behaviour. Overexpression or RNAi knock-
down of snpf pan-neuronally promotes or suppresses feeding
in adult flies, respectively [52]. sNPF regulates adult feeding
behaviour partly by suppressing the synaptic release of
bitter-sensing gustatory receptor neurons [43]. Furthermore,
starvation increases sNPF signalling in odorant receptor neur-
ons (ORNs) to elevate their sensitivity to food odours, thereby
promoting foraging behaviour in adult flies [6]. sNPF has also
been demonstrated to directly stimulate the expression of
DILPs in larval and adult IPCs [53,54]. This scenario seems to
contradict the proposed role of sNPF as a hunger signal. How-
ever, upregulation of DILPs by sNPF might relate more
to regulating metabolism and growth, rather than feeding
motivation. Given that sNPF is broadly expressed in the fly
nervous system, it is also possible that sNPF functions as a
co-transmitter in most neural circuits, including those involved
in hunger-driven behaviours [51].
Allatostatin A (AstA) is another satiety signal that inhibits
adult feeding behaviour when flies are fed [47,55,56]. Its recep-
tors DAR-1 and DAR-2 are homologues of the mammalian
galanin receptors [57,58]. Activation of AstA-expressing neur-
ons reduces the proboscis extension reflex (PER) of hungry flies
to sucrose as well as their food intake [47]. The same manipu-
lation has no effect on PER in fed flies, suggesting that the AstA
signal might already be maximized when flies are well fed.
Consistent with the notion that AstA signal is high under fed
conditions, flies on a low-nutrient diet (1% sucrose) exhibit
downregulated expression of AstA and its receptor dar-2 in
IPCs and AKH-producing cells in the corpora cardiaca [55].
Giving flies a high sugar diet (cornmeal) after nutrient restric-
tion strongly increases the expression of AstA and dar-2.
However, when a high-protein diet (yeast) is given, AstA
expression is only modestly increased and dar-2 expression is
unchanged. Interestingly, flies fed ad libitum normally prefer
sucrose over yeast food, but when AstA neuronal activity is
genetically increased in these flies, they shift their preference
royalsocietypublishing.org/journal/rsob Open Biol. 9: 180259
3

to yeast food,suggesting that AstA might be a nutrient-specific
satiety signal for sugar [55].
Myoinhibitory peptides (MIPs) have also been identified as
a potent satiety signal in the adult fly brain [59]. The mip gene
encodes five MIPs (MIP1-5) that are closely related [60]. Some
MIPs have N-terminal sequences similar to mammalian Gala-
nin [61]. Silencing MIP-expressing neurons makes food-
satiated flies behave like hungry flies, showing elevated taste
sensitivity to sugar, heightened olfactory sensitivity to food
odours and increased food intake [59]. Similar phenotypes
are also observed in mip mutant flies. By contrast, activation
of MIP neurons in starved flies reduces their food intake,
body weight and sensory sensitivity toward food. MIPs are
expressed in about 70 neurons in the central nervous system,
but it is not clear which neurons mediate the satiety effect or
how MIP neurons sense the satiety state.
Hugin is a fly homologue of mammalian neuropeptide U
[62]. It is expressed in 20 neurons in the larval SEZ with
axonal projections to the ventral nerve cord, the pharynx, the
protocerebrum, as well as the ring gland (the master larval
neuroendocrine organ), where AKH-producing cells reside
[27]. The connectivity patterns of hugin neurons in the adult
brain are less well characterized, but are suggested to be similar
to those of the larval brain [27]. Transcription of the hugin gene
is affected by diet. Levels of hugin mRNA are decreased in both
food-deprived larvae and larvae grown on a sugar-rich but
amino acid-deficient diet, indicating that hugin may function
as an amino acid-specific satiety signal. High hugin mRNA
levels in larvae are correlated with low food intake and reduced
food-seeking behaviour [27]. Furthermore, activation of hugin
neurons reduces yeast food intake, whereas blocking them pro-
motes feeding in both larvae and adult flies [27,63]. A subset of
hugin neurons (hugin PC neurons) in the larval brain that pro-
ject to the protocerebrum have also been shown to be part of
the bitter gustatory pathway, linking bitter sensation to
median neurosecretory cells, including IPCs [64,65]. Therefore,
hugin neurons might represent a hub in the fly brain for inte-
gration of taste and internal nutrient level signals, facilitating
the regulation of feeding motivation.
Leucokinin (LK) is an insect neuropeptide that was initially
identified as regulating body water homeostasis [66]. It has
since also been shown to control feeding behaviour and metab-
olism [67]. There is no known mammalian counterpart for LK,
but its receptor (LKR) is homologous to the vertebrate tachyki-
nin receptor [68]. LK is expressed in three sets of neurons in the
adult fly nervous system: a pair of LHLK neurons in the lateral
horn (LH), a pair of SELK neurons in the SEZ, and 11 –12 pairs
of ABLK neurons in the abdominal ganglia (AB) [69,70].
During feeding, lk and lkr mutant adult flies consume larger
meals at a lower frequency than control flies [67]. These
mutant flies also exhibit reduced food intake in long-term
but not short-term feeding assays [71]. However, conditional
activation or silencing of LK neurons in adult flies both result
in decreased foodintake [72], so the neural mechanisms under-
lying the contribution of LK to feeding behaviour are not
fully understood. Apart from its impact on feeding behaviour,
the LK pathway has been shown to mediate postprandial sleep
in adult flies [73]. Flies sleep more after eating, and this post-
prandial sleep is mainly induced by protein and salt but
not sucrose consumption. Activation of LK neurons reduces
postprandial sleep, whereas lk knockdown has the opposite
effect. These results suggest that LK may function as a nutri-
ent-specific hunger signal. Furthermore, adult ABLK neurons
express insulin receptor (InR) and the 5HT1B receptor for the
hunger signal serotonin (see below) [72,74], and adult IPCs
express LKR [71]. Knockdown of InR in adult ABLK neurons
elevates LK levels in the cell bodies, but whether this is due
to decreased LK release or increased LK production has not
been determined. By contrast, knockdown of 5HT1B reduces
LK levels, but again it is not clear whether the reduced LK
level is due to increased LK release or reduced LK production.
Although high-concentration serotonin treatment (10 mM) has
been shown to silence ABLK neurons, this outcome could be an
indirect consequence of strong neuronal bursting [72]. Notably,
the 5HT1B receptor is not expressed in adult LHLK or SELK
neurons, suggesting that the LK neurons in the brain and ven-
tral ganglia are differentially regulated and may have different
functions [71,72,75]. Complex results have also been observed
when LKR is knocked down in adult IPCs, with LK signal
seeming to negatively regulate DILP2 expression, but posi-
tively regulate DILP5 expression and DILP3 release [71].
Taken together, LK is likely to be part of the hunger regulatory
network, but its detailed role remains to be resolved.
Corazonin (CRZ) is evolutionarily related to AKH [76]. In
both larval and adult brains, CRZ is predominantly expressed
in bilateral clusters of dorsolateral neurons (DLPs), which also
express sNPF [77,78]. Chronic activation of CRZ neurons leads
to increased food intake in adult flies [47], whereas knockdown
of crz in these neurons has the opposite effect [79]. The CRZ
receptor (CrzR) is expressed in adult salivary glands and fat
body, and knockdown of CrzR in these peripheral tissues
results in decreased food intake. However, knockdown of
CrzR also has complex effects on the expression of multiple
genes involved in regulating metabolism and stress [79]. Con-
sequently, further studies are required to reveal the precise role
of CRZ on hunger-driven behaviour.
Drosulfakinin (DSK) is the fly homologue of mammalian
cholecystokinin. In addition to being expressed in several
neuron groups in the fly brain, DSK is co-expressed with
DILPs in a subset of larval and adult IPCs [80]. DSK appears
to function as a satiety signal together with DILPs. RNAi
knockdown of dsk specifically in these IPCs increases food
intake in both larvae and adult flies, even when the proffered
food is bitter or otherwise less palatable. A similar phenotype
has been observed when these IPCs or all DSK-expressing
cells are silenced. Furthermore, downregulation of dsk
expression in adult IPCs leads to increased levels of dilp2,3
and 5transcript, suggesting compensatory regulation among
these satiety-mediating neuropeptides [80].
A peptide hormone named female-specific independent of
transformer (FIT) was recently discovered to be a protein-
specific satietysignal in adult flies [13]. Protein but not sucrose
or lipid consumption increases the expression of FIT. The
protein-induced satiety effect on feeding is impaired in flies
lacking FIT, whereas FIT overexpression reduces the preference
of flies for protein food as well as their overall food intake after
starvation. FIT is expressed in fat cells within the head of adult
flies but not in the brain. However, it has been demonstrated
that FIT secreted from fat cells executes its satiety effect by pro-
moting DILP release from IPCs. FIT has not been detected in
larvae, suggesting that it is an adult-specific satiety signal.
Another neuropeptide proposed to mediate hunger and
satiety control is CCHamide-2 (CCHa2), an insect neuropep-
tide without a known counterpart in mammals. However,
conflicting results have been obtained regarding the role of
CCHa2 in feeding control. In one study, CCHa2 was found
royalsocietypublishing.org/journal/rsob Open Biol. 9: 180259
4

to be expressed in larval fat body and in the midgut [81]. In that
study, CCHa2 expression levels decreased in starved larvae
and levels could be recovered by feeding yeast food or glucose
to the larvae. Moreover, perfusion of CCHa2 onto larval brain
explants can directly activate IPCs, in which the CCHa2 recep-
tor (CCHa2-R) is enriched. Also, ccha2-r mutant larvae exhibit
normal feeding activity, but their feeding phase during the
third instar stage is prolonged. These results suggest that
CCHa2 is a nutrient-stimulated satiety signal similar to Upd2
[37]. However, in other studies, CCHa2 has only been detected
in the midgut but not fat body in both larvae and adult flies
[82,83]. Furthermore, both larvae and adult flies lacking
CCHa2 show reduced feeding activity, suggesting that
CCHa2 is a hormone that stimulates feeding [83]. These con-
flicting results may partly be due to the pleiotropic functions
of CCHa2 pathways, so more research is required to further
elucidate the roles of CCHa2 in feeding regulation.
SIFamide (SIFa) has been shown as a hunger signal in adult
flies. The receptor of SIFamide (SIFaR) is homologous to ver-
tebrate gonadotropin inhibitory hormone receptor (GnIHR),
but the sequences of SIFa and GnIH are not closely related
[84,85]. SIFa is expressed in four neurons in the fly brain and
their neurites innervate broadly in the brain, particularly den-
sely in the antennal lobe, central complex and SEZ [86]. SIFa
neuronal activity is elevated in starved flies. Acute activation
of SIFa neurons alone is sufficient to increase the response of
fed flies to food odour and to enhance their food intake,
making them behave like hungry flies. However, although
knockdown of SIFa completely abolishes starvation-induced
sensitization of olfactory projection neurons to food odour,
silencing SIFa neurons does not affect food intake in starved
flies, suggesting that hunger may elicit recruitment of compen-
satory pathways to promote feeding.SIFa neurons are inhibited
by satiety-encoding MIP neurons. Intriguingly, they are also
positively regulated by another satiety signal, hugin [86]. The
biological significance of hugin regulation in this scenario is
not clear, but it highlights the complexity of hunger regulation.
Apart from neuropeptides, 6-pyruvoyltetrahydropterin
(PTP) and serotonin are known to be satiety and hunger
signals, respectively, in adult flies. PTP represents a curious
case illustrating how peripheral nutrient-sensing tissue can
communicate with central neural circuits that control feeding
behaviour [87]. PTP is anintermediate in the biosynthesis path-
way that produces the enzymatic cofactor tetrahydrobiopterin
(BH4). There are three enzymes in this pathway—Purple (Pr),
Punch (Pu) and Sepiapterin reductase (Sptr). Knockdown of
pr or pu in fat body increases food intake, which can be rescued
by feeding flies BH4. Intriguingly, knockdown of sptr in fat
body has no effect on feeding behaviour. Instead, increased
food intake has been observed when sptr is knocked down in
NPF neurons. BH4 appears to inhibit NPF release through an
unknown mechanism. Furthermore, expression of pr and pu
in fat body is diet-dependent. Flies on a low-nutrient diet exhi-
bit reduced pr and pu expression. Therefore, when flies are well
fed, increased levels of Pr and Pu have been proposed to elevate
synthesis of PTP, which is released from fat body, circulated to
the brain and taken up by NPF neurons. In the NPF neurons,
PTP is converted by Sptr to BH4, which inhibits the release of
NPF that promotes feeding behaviour.
In contrast to the satiety function of BH4, activation of a
small subset of serotoninergic neurons mimics starvation and
induces a potent hunger response in food-satiated adult flies
[88]. There are 25 of these hunger-inducing serotoninergic
neurons per adult fly brain hemisphere and they project their
neurites into broad areas. Artificial activation of these serotoni-
nergic neurons in fed flies causes the flies to ingest the same
amount of food as flies starved for 24 h. They also exhibit the
same intensity of PER in response to sucrose solution and the
same level of preference for nutritious sugar over sweet-only
sugar as starved flies. However, serotoninergic neuronal acti-
vation does not mimic starvation-induced hyperactivity,
which is mediated byoctopamine [89], highlighting the modu-
lar nature of the hunger control mechanism. Serotonin may
promote hunger-driven behaviours partly by inhibiting DILP
expression in IPCs. IPCs express the serotonin receptor
5HT1A, and knockdown of 5HT1A in IPCs is reported to
increase the expression of DILP2 and DILP5 [90,91].
Taotie neurons and a group of ellipsoid body (EB) R4 neur-
ons in the adult fly brain have also been suggested to encode
hunger [92–94]. The neurotransmitters/modulators released
by these neurons are currently unknown, but Taotie neurons
are likely be peptidergic [94]. Taotie neurons form a smallcluster
in the pars intercerebralis (PI) that also harbours neurosecretory
cells including IPCs. Taotie neurons do not overlap with these
IPCs, and their activation induces persistence of hunger signal.
Fed flies with activated Taotie neurons consume the same
amount of food as flies starved for at least 12 h. These flies also
prefer nutritious sugar and yeast (i.e. just like the starved
flies). Activation of Taotie neurons also reduces the release of
DILPs from IPCs, suggesting that Taotie neurons promote
hunger-driven behaviours partly through inhibiting the satiety
signal of DILPs [94]. EB R4 neurons express sodium/solute co-
transporter-like 5A11 (SLC5A11) [93]. When flies are starved,
transcription of SLC5A11 increases in the R4 neurons, leading
to enhanced neuronal excitability by inhibiting the Drosophila
KCNQ potassium channel [92]. Overexpression of SLC5A11 or
activation of R4 neurons is sufficient to drive feeding behaviour
in food-satiated flies, whereas silencing these neurons has the
opposite effect. However, how starvation regulates SLC5A11
expression in R4 neurons remains to be established.
The large number of neuromodulators involved in hunger-
driven behaviour highlights the complexity of this motivational
system, even in small insects. These neuromodulators do not
work alone. Instead, they interact with each other to generate
coordinated outputs (figure 1). Many of these modulators
converge on IPCs, regulating their release of DILPs and pre-
sumably also DSK. Remote control of IPCs by fat body, as
well as mutual inhibition between hunger- and satiety-mediat-
ing neuromodulators, are also prominent features of the control
of this motivational state. Systems approaches that allow visu-
alization and manipulation of multiple neuromodulators
simultaneously will be needed to understand the dynamics of
this intricate regulatory network. Furthermore, developing an
understanding of the links between nutrient sensors and this
regulatory network remains an important avenue for future
research. In the following sections, we discuss how these
hunger and satiety neuromodulators act on both central and
peripheral neural circuits to control feeding and foraging
behaviours.
4. Hunger-based modulation of feeding
circuits
Consumption of appropriate nutrients to fulfil body require-
ments is the ultimate goal of hunger motivation. Starvation
royalsocietypublishing.org/journal/rsob Open Biol. 9: 180259
5

sharpens the sensitivity of flies to food components, increases
their preference for nutritious food and dampens theirresponse
to bitter compounds. Recent studies have begun to unravel the
neural mechanisms underpinning these hunger modulations.
4.1. Hunger sensitizes sugar taste
Flies detect food components via a repertoire of taste-specific
gustatory receptor neurons (GRNs). Sugar-sensing GRNs
expressing gustatory receptor 5a (Gr5a) in adult flies show
enhanced responses to sucrose upon starvation (figure 2a).
The enhancement requires presence of the dopamine receptor
DopEcR in the Gr5a neurons, and a single dopaminergic
neuron (TH-VUM) is likely to be the source of dopamine in
this case [43,95,96]. TH-VUM neurites broadly innervate the
SEZ, also where Gr5a neurons innervate. The neuronal activity
of TH-VUM is positively correlated with the starvation dur-
ation. Silencing TH-VUM decreases PER of starved flies to
sucrose,whereas increasing the excitabilityof TH-VUM elevates
PER to sugar in both fed and starved flies. This modulation of
sugar sensitivity is likely initiated by the hunger signal NPF
[43]. Activation of NPF neurons also increases the sugar
response of Gr5a neurons. However, this enhancedeffect disap-
pears in flies with hypomorphic mutation of DopEcR. The link
betweenNPF neurons and TH-VUM or Gr5a neurons is unclear.
Knockdown of npfr in the TH-VUM neuron does not affect
sugar sensitivity, suggesting that the effect of NPF on the
TH-VUM-to-Gr5a GRN pathway is indirect.
4.2. Hunger desensitizes bitter taste
A parallel neural pathway regulates starvation-induced dam-
pening of the response to bitter foods (figure 2a). Hungry
larvae and adult flies show a higher tolerance to food contain-
ing bitter substances [15,43], and in adult flies this increase in
bitter tolerance correlates with the decreased bitter response
of Gr66a-positive bitter-sensing GRNs [43]. This hunger-
dependent modulation of Gr66a neurons requires sNPF, but
not NPF. Silencing a subset of sNPF neurons called lateral
neurosecretory cells (LNCs) increases bitter sensitivity in
hungry flies. LNCs project their axons to the SEZ, which is
also innervated by Gr66a neurons. However, the effect of
sNPF is likely to be indirect, acting through as yet unidentified
GABAergic neurons. Furthermore, activation of hunger-med-
iating AKH neurons decreases bitter sensitivity in an sNPF-
dependent manner, suggesting that AKH neurons may func-
tion upstream of or parallel to the sNPF pathway [43]. Two
pairs of SEZ-innervating OA-VL neurons have also been
shown to regulate starvation-induced bitter insensitivity in
adult flies [97]. These OA-VL neurons release both octopamine
and tyramine to directly potentiate Gr66a neurons in fed flies.
The neural pathways upstream of the OA-VL neurons remain
to be determined, but given that starvation strongly decreases
their neuronal activity, sNPF-regulated GABAergic neurons
are likely candidates.
4.3. Hunger modulates other neurons involved in sugar
feeding
In addition to the neurons in the taste circuits, several other
neurons have been identified as regulating sugar feeding.
A pair of Fdg neurons in adult flies appear to act as command
neurons, activation of which is sufficient to generate the full
spectrum of the feeding motor programme, from proboscis
extension to pharyngeal pumping and proboscis retraction
(figure 2a) [98]. Fdg neurons are activated by sugar taste, but
only when a fly is starved. Probably downstream of the Fdg
neurons are several motor neurons in the SEZ that innervate
muscles involved in feeding behaviour. Activating these neur-
ons individually triggers a subset of the feeding motor
TH-
VUM
DA
NPF
hunger
GABA
N.
LNCs
AKH
sNPF
GABA OA-VL
OA
TA
IN1
ACh
ISNs
Ir76b
a
.
a
.
s
p
e
c
i
f
i
c
DA-
WED PLP
FB-
LAL
protein intake sugar intake
protein hunger
(lack of Gln)
yeast taste
food intake
DILPs
DH44
R1
DH44
nutritious sugar
a. a.
hunger
DILPs
NPFR
NPF
feeding
rate
feeding
tolerance
OA-
VUM1
OA-
VUM2
Oamb
Oct
b
3
R
OA OA
Fdg
DopEcR
Gr5a
sweet taste
Gr66a
bitter taste
(b)(c)
(a)
Figure 2. Hunger-based control of feeding circuits. (a) In adult flies, hunger
modulates GRNs (orange, green and red) and SEZ neurons (light blue) to pro-
mote food intake. Starvation increases the release of NPF, which indirectly
activates the dopaminergic TH-VUM neurons that in turn potentiate sweet
taste-responsive Gr5a neurons via the dopamine receptor DopEcR. Starvation
also increases the release of AKH, which indirectly activates sNPF-releasing
LNCs. sNPF then activates as yet unknown GABAergic neurons that inhibit
the bitter taste-responsive Gr66a neurons. The same GABAergic neurons
may also inhibit OA-VL neurons that can potentiate Gr66a neurons by releas-
ing tyramine (TA) and octopamine (OA). In addition, starvation potentiates
yeast taste-responsive Ir76 neurons, Fdg command neurons, cholinergic IN1
neurons, and AKHR-expressing ISNs to promote food intake. Starvation also
positively regulates DH44 neurons that promote feeding (via DH44R1-expres-
sing neurons) in response to post-ingestive amino acid and nutritious sugar
signals. Dashed lines indicate the regulation is indirect or its underlying
mechanism is not fully understood. a.a., amino acids. (b) In the adult central
brain, protein starvation (and particularly lack of glutamine) activates dopa-
minergic DA-WED neurons, which activate FB-LAL neurons to promote protein
intake while inhibiting PLP neurons that promote sugar consumption. (c)In
fly larvae, hunger increases feeding tolerance through an NPF pathway, while
also enhancing feeding rate via OA neurons. Two populations of OA neurons,
OA-VUM1 and OA-VUM2, act, respectively, through OAMB- and Octb3R-
expressing neurons to contrastingly regulate feeding rate. Additionally,
DILPs (release of which is inhibited by hunger) also negatively regulate feed-
ing tolerance and feeding rate. DILPs regulate feeding tolerance via NPFR
neurons, but whether they regulate feeding rate through the OA neurons
remains to be determined.
royalsocietypublishing.org/journal/rsob Open Biol. 9: 180259
6

programme [99–101]. The upstream mechanisms relaying
taste and internal state information to Fdg neurons are not
clear. Twelve cholinergic interneurons (IN1) in the SEZ form
synapses with sugar-sensing neurons (figure 2a) [102], but
their relationship with Fdg neurons has not been established.
IN1 activity is also modulated by hunger state. However,
unlike Fdg neurons that respond to sugar taste, IN1 is activated
by sucrose ingestion. Also, activation of IN1 does not directly
trigger feeding behaviour, but it does increase the probability
of sugar ingestion when a drop of sucrose solution is presented
in close proximity to a fly. Four interoceptive SEZ neurons
(ISNs) have also been identified as controlling sugar feeding
in adult flies (figure 2a) [16], but whether ISNs connect to
Fdg neurons remains undetermined. Activation of ISNs
induces sugar consumption even in well-fed flies. However,
interestingly, ISNs do not respond to sugar taste but encode
information about hunger state. The activity of ISNs increases
significantly in starved flies, and this modulation requires
AKHR expressed in ISNs. Furthermore, application of
hunger-promoting AKH to explant brains activates ISNs.
Apart from AKH-mediated regulation, ISNs are indirectly
inhibited by satiety-inducing DILPs, demonstrating a push
and pull mechanism for satiety-state control. Intriguingly,
ISNs are repressed by high haemolymph osmolarity caused
by desiccation and they negatively regulate water consumption
[16]. Thus, ISNs represent a signal convergence node in the
fly brain where hunger and thirst motivations compete for
behavioural expression.
4.4. Protein hunger modulates fly responses to amino
acids
Like the sugar-sensing GRNs, gustatory neurons for yeast taste
are regulated by internal amino acid levels (figure 2a) [103].
Ir76b-expressing GRNs in the proboscis of adult flies respond
to yeast taste and are required for yeast intake. Amino acid-
deprivation significantly increases the response of these
GRNs to yeast. Amino acid-deprivation has no effect on the
response of Gr5a neurons to sucrose, highlighting the nutrient
specificity of hunger control. How amino acid hunger regulates
these yeast-responsive GRNs remains to be identified. It would
be interesting to investigate if protein-specific satiety signals,
such as FIT and hugin, are involved in this regulation.
Amino acid starvation also promotes yeast feeding by
regulating central brain circuits. Two dopaminergic neurons
(DA-WED) in each hemisphere of the adult brain and that
innervate the vedge neuropil are proposed to encode protein
hunger (figure 2b) [28]. Silencing these DA-WED neurons
decreases yeast intake but increases sucrose consumption,
whereas activating these neurons enhances yeast intake but
reduces sucrose consumption. Therefore, like overall hunger
and thirst, nutrient-specific hunger motivations may also
compete for behavioural expression. Amino acid starvation,
especially glutamine starvation, increases the activity of
DA-WED neurons and, remarkably, also drastically lengthens
their medial branches in a form of structural plasticity. The
medial branches appear to contact the FB-LAL neurons,
whose activity drives persistent protein intake. By contrast,
the lateral branches of DA-WED neurons form synapses with
PLP neurons that drive sugar intake. The dopamine released
from DA-WED neurons activates FB-LAL neurons via the
DopR2 receptor, but it inhibits PLP neurons via the DopR1
receptor. As yet, how DA-WED neurons sense the need for
amino acids has not been elucidated.
Some neurons responsible for feeding behaviour regulate
both sugar and amino acid consumption. Six neurons expres-
sing diuretic hormone 44 (DH44) in the adult fly brain
regulate consumption of nutritious sugar and essential amino
acids (figure 2a) [104,105]. DH44 is a homologue of mamma-
lian corticotrophin-releasing hormone (CRH). These DH44
neurons are located in the PI region. The DH44 neurons are
directly activated by nutritious sugars and three specific
amino acids: L-glutamate, L-alanine and L-aspartate. Stimu-
lation of downstream neurons expressing DH44 receptor 1
(DH44 R1) leads to rapid proboscis extension, even in the
absence of food. Therefore, DH44 neurons have been proposed
to function as a post-ingestive nutrient sensor that facilitatesthe
consumption of nutritious sugar and amino acids [104,105].
Neither activation nor silencing of DH44 neurons has an
effect on the amount of food ingested, suggesting that they
do not mediate general signals of hunger or satiety. Interest-
ingly, although promotion of feeding behaviour by DH44 is
independent of internal nutritional status, flies only prefer
nutritious sugars over sweet-only sugars when they are
starved. Therefore, parts of the DH44 pathway only seem to
work in the hungry state. How the DH44 pathway is regulated
by starvation is not currently known. However, DH44 neurons
express hugin receptor, so they may be directly modulated by
the satiety signal hugin [65].
4.5. Hunger modulates larval feeding circuits
Hunger also changes the feeding behaviour of fly larvae.
Starved larvae have higher feeding rates and increased toler-
ance for low-quality and noxious foods [15,38,106]. These
hunger-driven increases in feeding rate and feeding tolerance
are regulated by different neural circuits (figure 2c) [106].
Blocking NPF neurons in starved larvae specifically abolishes
increased feeding tolerance without affecting the feeding rate,
and the opposite scenario is observed when octopaminergic
(OA) neurons are blocked. Blocking the neurotransmission of
NPF receptor (NPFR)-expressing neurons also impairs the
increased feeding tolerance of starved larvae, whereas overex-
pression of npfr in NPFR neurons makes larvae feed on low-
quality or noxious food even when they are not food-deprived
[15,38]. Down- or upregulation of insulin signalling in NPFR
neurons also leads to increased or decreased acceptance,
respectively, of low-quality or noxious food. Therefore, NPF
and DILPs work together on NPFR neurons to regulate feeding
tolerance [15,38]. The OA neurons regulate feeding rate
through two populations of downstream neurons that express
distinct OA receptors, Octb3R and OAMB [106,107]. Starved
larvae with reduced Octb3R activity cannot increase their feed-
ing rate [106]. By contrast, knockdown of oamb increases
feeding rate in larvae fed ad libitum [107]. Therefore, the
downstream neurons expressing these two OA receptors con-
trastingly modulate larval feeding rate. Interestingly, two
distinct populations of OA neurons in the larval SEZ that
receive gustatory inputs have also been found to contrastingly
regulate feeding rate [106]. Laser ablation of five OA-VUM1
neurons caused increased feeding rates in fed larvae, whereas
laser ablation of six OA-VUM2 neurons attenuated the elevated
feeding rate of starved larvae. Accordingly, the OA-VUM1
and OA-VUM2 neurons are thought to operate upstream of
the OAMB- and Octb3R-expressing neurons, respectively.
royalsocietypublishing.org/journal/rsob Open Biol. 9: 180259
7

Furthermore, apart from regulating feeding tolerance through
NPFR neurons, IPCs and DILPs negatively regulate feeding
rate in both fed and starved larvae [38]. However, whether
they act through the OA neurons remains to be determined.
5. Hunger-based modulation of olfactory
circuits
Flies mainly rely on olfaction when foraging for food. Hungry
flies are more sensitive to and more attracted by food odours
[6,32,50]. As for their control of feeding behaviour, hunger
and satiety signals modulate both peripheral and central
neural circuits to regulate the responses of flies to food odour.
5.1. Hunger sensitizes food odour responses
Starvation directly enhances the sensitivity of odorant receptor
neurons (ORNs) that detect food odour (figure 3a) [6]. Starved
but not fed adult flies exhibit strong seeking behaviour when
they smell the food odour emanating from 1% apple cider vine-
gar (ACV).ACV-responsive ORNs that trigger odour attraction
show an enhanced response to ACV upon starvation. Knock-
down of sNPF or its receptor (sNPFR) in these ORNs
eliminates this enhanced response in hungry flies and the
flies exhibit reduced ACV-seeking behaviour. By contrast,
overexpression of sNPFR in ACV-responsive ORNs is suffi-
cient to increase their ACV response and evoke ACV-seeking
behaviour in fed flies. The sNPF signalling in these ORNs
appears to be regulated by satiety-inducing DILPs. ACV-
responsive ORNs express InR, activation of which inhibits
sNPFR expression. When flies are starved, low levels of
DILPs results in high sNPFRexpression in the ACV-responsive
ORNs, with the increased sNPF signalling in turn enhancing
the sensitivity of these ORNs. Therefore, global insulin signal-
ling and the local sNPF pathway work cooperatively in
peripheral sensory neurons to tune the sensitivity of flies to
food odour.
5.2. Hunger desensitizes responses to ‘bad’ smells
High concentrations of ACV also activate ORNs expressing
Or85a that signals negative valence to render them less attrac-
tive to adult flies. Just as starvation dampens bitter GRNs in
feeding circuits [43], food deprivation also reduces the ACV
response in Or85a neurons (figure 3a) [39]. RNA profiling
experiments have revealed that transcripts of Drosophila tachy-
kinin receptor (DTKR) are increased in the antennae of
starved flies. Knockdown of DTKR in the Or85a neurons of
starved flies increases the response of those neuronsto high con-
centrations of ACV. The same manipulation also reduces
the odour-seeking response of hungry flies. The source of the
DTKR ligand tachykinin (TK) is local interneurons (LNs) in
the antennal lobe, where the axons of Or85a neurons also
project to. TK knockdown in the LNs results in the same pheno-
type as DTKR knockdown in Or85a neurons. Furthermore,
enhancing insulin signalling in Or85a neurons increases their
activity and decreases the attraction of hungry flies to a high-
concentration ACV. Although the direct link between insulin
signalling and the expression of DTKR has not been established,
it may be that reduced insulin signalling in starved flies leads to
increased expression of DTKR, which suppresses the response
of Or85a neurons to high concentrations of ACV upon receiving
TK released from the LNs. Starvation therefore fine-tunes
the sensitivity of flies to food odour through bi-directional
regulation of ORNs with opposite valences.
5.3. Hunger modulates responses to learned
food-associated olfactory cues
In addition to modulating the innate responses of flies to
food odour, satiety state also gates their expression of sugar-
rewarded olfactory memory [50,108]. As we have already
mentioned, starved adult flies can be trained to form an appe-
titive olfactory memory by pairing odours with sugar reward.
After training, flies only approach the sugar-predictive odours
when they are hungry. Starvation promotes this learned
approach behaviour via the hunger signal NPF. NPF executes
its function by inhibiting two pairs of dopaminergic neurons
(MP1) innervating the mushroom body (MB), an olfactory
memory centre in the fly brain [50]. The adult MB has five
MBON-
a
3
MVP2
MBON-
a
1
M4/6 MBONs
food seeking
LNs
hunger
attraction
ORNs
Kenyon cells
food seeking
food
odour
hunger signals
satiety signals
PPL1-
a
3
PAM-
b
2
b¢
2a
PAM-
b¢
2a
PPL1-
a¢2a2
PPL1-
g
2
a¢1
MP1
NPF
AstA
NPF
insulin
sNPF
insulin
AstA 5HT
sNPF
insulin
AstA
NPF
sNPF
5HT
sNPF
sNPF
NPF
avoidance
ORNs
InR
sNPFR
sNPF InR
DILPs
DTKR TK
GABA
DAN MBON
MBON-
b
2
b¢2a
MBON-
a¢2
MBON-
g
2
a¢1
(b)
(a)
Figure 3. Hunger-based control of olfactory circuits. (a) Starvation inhibits the
release of DILPs, reducing insulin signalling in both attraction and avoidance
ORNs. The reduced insulin signalling leads to increased expression of sNPFR
and DTKR in the attraction and avoidance ORNs, respectively. In the attraction
ORNs, sNPFR receives sNPF secreted from the same ORNs, which in turn
enhances their attraction to food odours. By contrast, in avoidance ORNs,
DTKR receives TK from LNs in the antennal lobe and consequently repress the
synaptic output of avoidance ORNs. (b) Five MBON pathways (blue) that innerv-
ate different zones of the KC axons promote odour-driven food-seeking
behaviour. These MBON pathways are regulated by their corresponding dopa-
minergic neurons (DANs; green), and these DANs receive different
combinations of hunger and satiety signals. The MP1-MVP2-M4/6 MBON path-
way also mediates hunger control of learned food-seeking behaviour.
royalsocietypublishing.org/journal/rsob Open Biol. 9: 180259
8

lobes, which are axonal bundles of approximately 2500 intrin-
sic neurons called Kenyon cells (KCs). The MB lobes can be
subdivided into 15 zones innervated by 20 types of dopamin-
ergic neurons and 21 types of output neurons (MBONs). The
axons of different dopaminergic neuronal types and the den-
drites of different MBONs innervate distinct zones of the MB
lobes [109]. The KCs receive direct input from olfactory projec-
tion neurons, representing the third-order neurons in the
olfactory pathway. Different MBONs receive odour infor-
mation from different zones along the MB lobes, and the
dopaminergic neurons are thought to modify the KC-MBON
synapses in zone-specific ways [109,110]. Some MBONs
encode positive or negative valences, and olfactory learning
is thought to tip over the balanced collective outputs of these
positives and negatives, resulting in odour-driven approach
or avoidance behaviour [110,111]. Knockdown of NPF in
MP1 neurons or artificially activating MP1 neurons impairs
the learned approach to the sugar-predictive odours in
hungry flies, whereas blocking MP1 neuronal output promotes
the learned approach in fed flies. Therefore, MP1 neuronal
activity blocks learned food-seeking behaviour, and NPF can
remove this blockage by downregulating MP1 neuronal
activity. The neurons downstream of MP1 are GABAergic
MBONs called MVP2 neurons. Dopamine released from MP1
neurons depresses the odour-evoked responses of MVP2 neur-
ons. The axons of MVP2 are projected to the tips of the MB
horizontal lobes where they release GABA to inhibit another
group of negative-valence MBONs called the M4/6 neurons
[112]. Thus, MP1-MVP2-M4/6 neurons represent a multi-
inhibitory neural circuit that represses odour-driven learned
food-seeking behaviour in fed flies. When MP1 neuronal
activity is downregulated by NPF upon starvation, high
MVP2 activity inhibits M4/6 neurons to permit learned food-
seeking behaviour (figure 3b).
5.4. Hunger modulates MB circuits to tune innate
responses to food odour
The MB not only controls the responses of adult fliesto learned
odours, it also regulates their innate food-seeking behaviour
evoked by food odour (figure 3b). A recent study shows that
5 of the 21 types of MBONs are required for hungry flies to
seek food odours [32]. The MP1-MVP2-M4/6 pathway has
also been identified as regulating innate food-seeking behav-
iour. Moreover, MP1 neurons are not only regulated by NPF
but also by two other hunger signals: sNPF and serotonin.
Knockdown of NPFR, sNPFR or the serotonin receptor 5H2A
in MP1 neurons impairs innate food-seeking behaviour. Four
other MBONs—MBON-a3, MBON-b2b02a, MBON-a02 and
MBON-g2
a
01—and their corresponding dopaminergic neur-
ons also regulate innate food-seeking behaviour. Blocking
these MBONs and dopaminergic neurons diminishes innate
food-seeking behaviour in hungry flies, and activation of the
dopaminergic neurons is sufficient to evoke food-seeking
behaviour in fed flies. Importantly, results from RNAi knock-
down of various receptors for different hunger and satiety
signals support the idea that the MB-innervating dopaminergic
neurons are directly regulated by many of these signals,
making the MB an integrative centre for hunger and satiety sig-
nals in the fly brain. This idea is further supported by a recent
study showing that the MBcircuit also regulates fat storage and
food intake [113].
6. Hunger-based modulation of locomotion
circuits
Starvation increases locomotion, presumably to facilitate
food-seeking behaviour [4,89]. This starvation-induced hyper-
activity in adult flies requires the AKH hunger signal [4,33].
Flies lacking AKH or its receptor AKHR do not exhibit star-
vation-induced hyperactivity. AKH stimulates locomotion by
acting on a small number (two to four neurons per hemisphere)
of AKHR-expressing neurons in the SEZ [33]. Silencing
these AKHR neurons abolishes the hyperactivity induced by
starvation, whereas activating them accelerates the onset of
hyperactivity upon starvation. These AKHR neurons are
specific for controlling locomotion since manipulating them
has no effect on feeding behaviour. These AKHR neurons
also express InR and their activity is suppressed by DILPs.
Thus, like ISNs that regulate feeding behaviour, these AKHR
neurons are directly influenced by both hunger and satiety sig-
nals [16,33]. These AKHR neurons are octopaminergic, but
their downstream neurons await identification.
7. Intricacy of foraging and feeding
behaviour: a future challenge
Recently, detailed quantitative methods measuring proboscis
extensions to food (sips) and consumed food volume in close
to real time in adult flies have revealed the intricate microstruc-
ture of feeding modulated by starvation [8,102,114]. Adult fly
feeding is highly rhythmic, with most sips lasting for 0.16 s,
with an inter-sip interval (ISI) of 0.08 s when consuming 10%
sucrose [114]. The sips can be further organized into bursts,
each defined as three or more consecutive sips separated by
inter-burst intervals (IBI) shorter than double the median ISI.
Although flies increase food consumption when starved, the
duration of sips and the ISI remain nearly constant. However,
starved flies exhibit higher numbers of sips per burst and a
shorter IBI. Interestingly, the dynamics of sips per burst and
IBI are differentially modulated by hunger. Four hours of star-
vation significantly shortens IBI without affecting the number
of sips. An additional four hours of starvation (8 h in total)
does not further curtail IBI, but it significantly increases the
number of sips per burst. Furthermore, 4 h starvation is suffi-
cient to lengthen the time flies spend on a food patch during
each visit (activity bout duration), but 8 h starvation is
needed to increase the length of the feed bursts. When 8 h-
starved flies eat to satiation, increased IBI and decreased
activity bout duration occur within 10 min after the meal
starts, whereas a decrease in number of sips per burst can
only be detected near the end of the meal. Similar regulation
of the dynamics of feed microstructure has also been observed
in protein-starved flies eating yeast [8]. Furthermore, detailed
analysis of adult female flies foraging on distributed yeast
patches uncovered modulation of foraging decisions by
internal amino acid-deprivation state [8]. When amino acid
demand is low (e.g. in virgin females), amino acid-deprived
flies spend more time on a yeast patch during each yeast
visit, but their rate of yeast encounters and probability of stop-
ping at a yeast patch are not different from amino acid-satiated
flies. All three of these parameters are increased when mated
female flies (experiencing high amino acid demand) are
deprived of amino acids. Amino acid deprivation also
royalsocietypublishing.org/journal/rsob Open Biol. 9: 180259
9

modulates the exploratory and exploitatory behaviour of flies.
Mated female flies pretreated with amino acid-rich food typi-
cally explore large area of yeast patches and almost never
return to yeast patches they have just visited. By contrast,
amino acid-deprived mated female flies show reduced global
exploration and have a much higher likelihood of revisiting
the same yeast patch [8]. This local food searching behaviour
appears to rely on idothetic cues independent of vision and
olfaction [115]. These detailed analyses of feeding and foraging
behaviour have revealed the sophisticated nature of hunger in
tuning rich repertoires of behaviour modules. An important
future challenge is to understand mechanistically how this
level of control is achieved.
8. Concluding remarks
Studying hunger and the behaviours it elicits will not only
lead to a better understanding of this highly conserved pri-
mary motivation, but will also provide insights into many
general and fundamental questions in neuroscience, such as
how the nervous system senses and encodes bodily require-
ments, how the brain integrates external stimuli and
internal states, how a neural circuit gates information flow,
and how goal-directed behaviour emerges from the brain.
Studies of hunger-driven behaviour in the fruit fly have
helped to reveal the biological nature of hunger at the mol-
ecular and neural circuit levels. Several important principles
can be deduced from these studies. First, the brain is
equipped with sensors that detect specific nutrients, and
these centralized sensors and those in peripheral tissues
work in concert to assess bodily requirements. Second,
hunger and satiety states are encoded by a large number of
neuromodulators, many of which are highly conserved
across animal species. These neuromodulators interact to
form an intricate regulatory network that presumably gives
rise to coherent feeding and foraging behaviour. Third,
these hunger and satiety neuromodulators can temporarily
and reversibly reconfigure both central and peripheral
neural circuits so that the same inputs can lead to different
outputs in a state-dependent manner. However, our quest
for a complete understanding of hunger-driven behaviour
has only just begun. Sophisticated behavioural tracking/
analysis techniques have revealed intricate behavioural
microstructures modulated by hunger. What are the neural-
circuit underpinnings of these behavioural microstructures?
How are they influenced by hunger and satiety neuromodu-
lators? And, most importantly, how are these microstructures
and their control circuits integrated to give rise to coordinated
feeding and foraging behaviour? These are important and
very challenging questions. Nevertheless, recent technological
breakthroughs—including whole-brain functional imaging of
live flies, advanced behavioural tracking, improved geneti-
cally encoded sensors of neurotransmitter release and
neuromodulator activity, and real-time optogenetic control
of defined neuronal types—have brought the answers to
these questions closer to our grasp.
Data accessibility. This article has no additional data.
Competing interests. We declare we have no competing interests.
Funding. This study was supported by a grant from Taiwan Ministry of
Science and Technology (105-2628-B-001-005-MY3) to S.L..
References
1. Toates F. 1986 Motivational systems. Cambridge, UK:
Cambridge University Press.
2. Sih A. 1980 Optimal behavior: can foragers balance
two conflicting demands? Science 210, 1041–1043.
(doi:10.1126/science.210.4473.1041)
3. Dietrich MO, Zimmer MR, Bober J, Horvath TL. 2015
Hypothalamic Agrp neurons drive stereotypic
behaviors beyond feeding. Cell 160, 1222– 1232.
(doi:10.1016/j.cell.2015.02.024)
4. Lee G, Park JH. 2004 Hemolymph sugar
homeostasis and starvation-induced hyperactivity
affected by genetic manipulations of the
adipokinetic hormone-encoding gene in Drosophila
melanogaster.Genetics 167, 311– 323. (doi:10.
1534/genetics.167.1.311)
5. Browne LB, Evans DR. 1960 Locomotor activity of
the blowfly as a function of feeding and starvation.
J. Insect Physiol. 4, 27–37. (doi:10.1016/0022-
1910(60)90065-2)
6. Root CM, Ko KI, Jafari A, Wang JW. 2011 Presynaptic
facilitation by neuropeptide signaling mediates
odor-driven food search. Cell 145, 133– 144.
(doi:10.1016/j.cell.2011.02.008)
7. Dethier VG, Chadwick LE. 1948 Chemoreception in
insects. Physiol. Rev. 28, 220– 254. (doi:10.1152/
physrev.1948.28.2.220)
8. Corrales-Carvajal VM, Faisal AA, Ribeiro C. 2016
Internal states drive nutrient homeostasis by
modulating exploration– exploitation trade-off. Elife
5, 1–29. (doi:10.7554/eLife.19920)
9. Sternson SM. 2013 Hypothalamic survival circuits:
blueprints for purposive behaviors. Neuron 77,
810–824. (doi:10.1016/j.neuron.2013.02.018)
10. Dus M, Min S, Keene AC, Lee GY, Suh GSB. 2011
Taste-independent detection of the caloric content
of sugar in Drosophila.Proc. Natl Acad. Sci. USA
108, 11 644 –11 649. (doi:10.1073/pnas.
1017096108)
11. Burke CJ, Waddell S. 2011 Remembering nutrient
quality of sugar in Drosophila.Curr. Biol. 21,
746–750. (doi:10.1016/j.cub.2011.03.032)
12. Fujita M, Tanimura T. 2011 Drosophila evaluates and
learns the nutritional value of sugars. Curr. Biol. 21,
751–755. (doi:10.1016/j.cub.2011.03.058)
13. Sun J, Liu C, Bai X, Li X, Li J, Zhang Z, Zhang Y, Guo
J, Li Y. 2017 Drosophila FIT is a protein-specific
satiety hormone essential for feeding control. Nat.
Commun. 8, 14161. (doi:10.1038/ncomms14161)
14. Hong S-H et al. 2012 Minibrain/Dyrk1a regulates
food intake through the Sir2-FOXO-sNPF/NPY
pathway in Drosophila and mammals. PLoS Genet.
8, e1002857. (doi:10.1371/journal.pgen.1002857)
15. Wu Q, Zhao Z, Shen P. 2005 Regulation of aversion
to noxious food by Drosophila neuropeptide Y- and
insulin-like systems. Nat. Neurosci. 8, 1350– 1355.
(doi:10.1038/nn1540)
16. Jourjine N, Mullaney BC, Mann K, Scott K. 2016
Coupled sensing of hunger and thirst signals
balances sugar and water consumption. Cell 166,
855–866. (doi:10.1016/j.cell.2016.06.046)
17. Kre
´neisz O, Chen X, Fridell Y-WC, Mulkey DK. 2010
Glucose increases activity and Ca
2þ
in insulin-
producing cells of adult Drosophila.Neuroreport 21,
1116–1120. (doi:10.1097/WNR.
0b013e3283409200)
18. Kim SK, Rulifson EJ. 2004 Conserved mechanisms of
glucose sensing and regulation by Drosophila
corpora cardiaca cells. Nature 431, 316– 320.
(doi:10.1038/nature02897)
19. Kohyama-Koganeya A, Kim Y-J, Miura M,
Hirabayashi Y. 2008 A Drosophila orphan G protein-
coupled receptor BOSS functions as a glucose-
responding receptor: loss of boss causes abnormal
energy metabolism. Proc. Natl Acad. Sci. USA 105,
15 328–15 333. (doi:10.1073/pnas.0807833105)
20. Kohyama-Koganeya A, Kurosawa M, Hirabayashi Y.
2015 Differential effects of tissue-specific deletion of
BOSS on feeding behaviors and energy metabolism.
PLoS ONE 10, e0133083. (doi:10.1371/journal.pone.
0133083)
21. Miyamoto T, Slone J, Song X, Amrein H. 2012 A
fructose receptor functions as a nutrient sensor in
the Drosophila brain. Cell 151, 1113– 1125. (doi:10.
1016/j.cell.2012.10.024)
royalsocietypublishing.org/journal/rsob Open Biol. 9: 180259
10

22. Bjordal M, Arquier N, Kniazeff J, Pin JP, Le
´opold P.
2014 Sensing of amino acids in a dopaminergic
circuitry promotes rejection of an incomplete diet in
Drosophila.Cell 156, 510– 521. (doi:10.1016/j.cell.
2013.12.024)
23. Colombani J, Raisin S, Pantalacci S, Radimerski T,
Montagne J, Le
´opold P. 2003 A nutrient sensor
mechanism controls Drosophila growth. Cell 114,
739–749. (doi:10.1016/S0092-8674(03)00713-X)
24. Ge
´minard C, Rulifson EJ, Le
´opold P. 2009 Remote
control of insulin secretion by fat cells in Drosophila.
Cell Metab. 10, 199–207. (doi:10.1016/j.cmet.2009.
08.002)
25. Vargas MA, Luo N, Yamaguchi A, Kapahi P. 2010 A
role for S6 kinase and serotonin in postmating
dietary switch and balance of nutrients in
D. melanogaster.Curr. Biol. 20, 1006– 1011.
(doi:10.1016/j.cub.2010.04.009)
26. Ribeiro C, Dickson BJ. 2010 Sex peptide receptor and
neuronal TOR/S6 K signaling modulate nutrient
balancing in Drosophila.Curr. Biol. 20, 1000–1005.
(doi:10.1016/j.cub.2010.03.061)
27. Melcher C, Pankratz MJ. 2005 Candidate gustatory
interneurons modulating feeding behavior in the
Drosophila brain. PLoS Biol. 3, e305. (doi:10.1371/
journal.pbio.0030305)
28. Liu Q, Tabuchi M, Liu S, Kodama L, Horiuchi W,
Daniels J, Chiu L, Baldoni D, Wu MN. 2017 Branch-
specific plasticity of a bifunctional dopamine circuit
encodes protein hunger. Science 356, 534–539.
(doi:10.1126/science.aal3245)
29. Brogiolo W, Stocker H, Ikeya T, Rintelen F, Fernandez
R, Hafen E. 2001 An evolutionarily conserved
function of the Drosophila insulin receptor and
insulin-like peptides in growth control. Curr.
Biol. 11, 213–221. (doi:10.1016/S0960-9822(01)
00068-9)
30. Garelli A, Gontijo AM, Miguela V, Caparros E,
Dominguez M. 2012 Imaginal discs secrete insulin-
like peptide 8 to mediate plasticity of growth and
maturation. Science 336, 579– 582. (doi:10.1126/
science.1216735)
31. Colombani J, Andersen DS, Le
´opold P. 2012
Secreted peptide Dilp8 coordinates Drosophila tissue
growth with developmental timing. Science 336,
582–585. (doi:10.1126/science.1216689)
32. Tsao C-H, Chen C-C, Lin C-H, Yang H-Y, Lin S. 2018
Drosophila mushroom bodies integrate hunger and
satiety signals to control innate food-seeking
behavior. Elife 7, e35264. (doi:10.7554/eLife.35264)
33. Yu Y, Huang R, Ye J, Zhang V, Wu C, Cheng G, Jia J,
Wang L. 2016 Regulation of starvation-induced
hyperactivity by insulin and glucagon signaling
in adult Drosophila.Elife 5, 1– 19. (doi:10.7554/
eLife.15693)
34. Loh K et al. 2017 Insulin controls food intake and
energy balance via NPY neurons. Mol. Metab. 6,
574–584. (doi:10.1016/j.molmet.2017.03.013)
35. Birse RT, So
¨derberg JAE, Luo J, Winther AME, Na
¨ssel
DR. 2011 Regulation of insulin-producing cells in
the adult Drosophila brain via the tachykinin
peptide receptor DTKR. J. Exp. Biol. 214,
4201–4208. (doi:10.1242/jeb.062091)
36. Ikeya T, Galic M, Belawat P, Nairz K, Hafen E. 2002
Nutrient-dependent expression of insulin-like
peptides from neuroendocrine cells in the CNS
contributes to growth regulation in Drosophila.Curr.
Biol. 12, 1293–1300. (doi:10.1016/S0960-
9822(02)01043-6)
37. Rajan A, Perrimon N. 2012 Drosophila cytokine
unpaired 2 regulates physiological homeostasis by
remotely controlling insulin secretion. Cell 151,
123–137. (doi:10.1016/j.cell.2012.08.019)
38. Wu Q, Zhang Y, Xu J, Shen P. 2005 Regulation of
hunger-driven behaviors by neural ribosomal S6
kinase in Drosophila.Proc. Natl Acad. Sci. USA 102,
13 289–13 294. (doi:10.1073/pnas.0501914102)
39. Ko KI, Root CM, Lindsay SA, Zaninovich OA,
Shepherd AK, Wasserman SA, Kim SM, Wang JW.
2015 Starvation promotes concerted modulation of
appetitive olfactory behavior via parallel
neuromodulatory circuits. Elife 4, 1– 17. (doi:10.
7554/eLife.08298)
40. Beshel J, Dubnau J, Zhong Y. 2017 A leptin analog
locally produced in the brain acts via a conserved
neural circuit to modulate obesity-linked behaviors
in Drosophila.Cell Metab. 25, 208– 217. (doi:10.
1016/j.cmet.2016.12.013)
41. Bharucha KN, Tarr P, Zipursky SL. 2008 A glucagon-
like endocrine pathway in Drosophila modulates
both lipid and carbohydrate homeostasis. J. Exp.
Biol. 211, 3103–3110. (doi:10.1242/jeb.016451)
42. Buch S, Melcher C, Bauer M, Katzenberger J,
Pankratz MJ. 2008 Opposing effects of dietary
protein and sugar regulate a transcriptional target
of Drosophila insulin-like peptide signaling.
Cell Metab. 7, 321– 332. (doi:10.1016/j.cmet.2008.
02.012)
43. Inagaki HK, Panse KM, Anderson DJ. 2014
Independent, reciprocal neuromodulatory control of
sweet and bitter taste sensitivity during starvation
in Drosophila.Neuron 84, 806– 820. (doi:10.1016/j.
neuron.2014.09.032)
44. Brown MR, Crim JW, Arata RC, Cai HN, Chun C, Shen
P. 1999 Identification of a Drosophila brain-gut
peptide related to the neuropeptide Y family.
Peptides 20, 1035–1042. (doi:10.1016/S0196-
9781(99)00097-2)
45. Wu Q, Wen T, Lee G, Park JH, Cai HN, Shen P. 2003
Developmental control of foraging and social
behavior by the Drosophila neuropeptide Y-like
system. Neuron 39, 147– 161. (doi:10.1016/S0896-
6273(03)00396-9)
46. Shao L, Saver M, Chung P, Ren Q, Lee T, Kent CF,
Heberlein U. 2017 Dissection of the Drosophila
neuropeptide F circuit using a high-throughput
two-choice assay. Proc. Natl Acad. Sci. USA 114,
E8091–E8099. (doi:10.1073/pnas.1710552114)
47. Hergarden AC, Tayler TD, Anderson DJ. 2012
Allatostatin-A neurons inhibit feeding behavior in
adult Drosophila.Proc. Natl Acad. Sci. USA 109,
3967–3972. (doi:10.1073/pnas.1200778109)
48. Beshel J, Zhong Y. 2013 Graded encoding of
food odor value in the Drosophila brain. J. Neurosci.
33, 15 693 –15 704. (doi:10.1523/JNEUROSCI.2605-
13.2013)
49. Krashes MJ, Keene AC, Leung B, Armstrong JD,
Waddell S. 2007 Sequential use of mushroom body
neuron subsets during Drosophila odor memory
processing. Neuron 53, 103– 115. (doi:10.1016/j.
neuron.2006.11.021)
50. Krashes MJ, DasGupta S, Vreede A, White B,
Armstrong JD, Waddell S. 2009 A neural circuit
mechanism integrating motivational state with
memory expression in Drosophila.Cell 139,
416–427. (doi:10.1016/j.cell.2009.08.035)
51. Na
¨ssel DR, Wegener C. 2011 A comparative review
of short and long neuropeptide F signaling in
invertebrates: any similarities to vertebrate
neuropeptide Y signaling? Peptides 32, 1335– 1355.
(doi:10.1016/j.peptides.2011.03.013)
52. Lee K-S, You K-H, Choo J-K, Han Y-M, Yu K. 2004
Drosophila short neuropeptide F regulates food
intake and body size. J. Biol. Chem. 279,
50 781–50 789. (doi:10.1074/jbc.M407842200)
53. Lee K-S et al. 2008 Drosophila short neuropeptide F
signalling regulates growth by ERK-mediated insulin
signalling. Nat. Cell Biol. 10, 468– 475. (doi:10.
1038/ncb1710)
54. Kapan N, Lushchak OV, Luo J, Na
¨ssel DR. 2012
Identified peptidergic neurons in the Drosophila
brain regulate insulin-producing cells, stress
responses and metabolism by coexpressed
short neuropeptide F and corazonin. Cell. Mol. Life
Sci. 69, 4051–4066. (doi:10.1007/s00018-012-
1097-z)
55. Hentze JL, Carlsson MA, Kondo S, Na
¨ssel DR, Rewitz
KF. 2015 The neuropeptide allatostatin A regulates
metabolism and feeding decisions in Drosophila.Sci.
Rep. 5, 11680. (doi:10.1038/srep11680)
56. Chen J, Reiher W, Hermann-Luibl C, Sellami A,
Cognigni P, Kondo S, Helfrich-Fo
¨rster C, Veenstra JA,
Wegener C. 2016 Allatostatin A signalling in
Drosophila regulates feeding and sleep and is
modulated by PDF. PLoS Genet. 12, e1006346.
(doi:10.1371/journal.pgen.1006346)
57. Birgu¨l N, Weise C, Kreienkamp HJ, Richter D. 1999
Reverse physiology in Drosophila: identification of a
novel allatostatin-like neuropeptide and its cognate
receptor structurally related to the mammalian
somatostatin/galanin/opioid receptor family. EMBO
J. 18, 5892–5900. (doi:10.1093/emboj/18.21.5892)
58. Lenz C, Søndergaard L, Grimmelikhuijzen CJ. 2000
Molecular cloning and genomic organization of a
novel receptor from Drosophila melanogaster
structurally related to mammalian galanin
receptors. Biochem. Biophys. Res. Commun. 269,
91–96. (doi:10.1006/bbrc.2000.2251)
59. Min S et al. 2016 Identification of a peptidergic
pathway critical to satiety responses in Drosophila.
Curr. Biol. 26, 814–820. (doi:10.1016/j.cub.2016.
01.029)
60. Williamson M, Lenz C, Winther AME, Na
¨ssel DR,
Grimmelikhuijzen CJP, Winther ME. 2001 Molecular
cloning, genomic organization, and expression of a
B-type (cricket-type) allatostatin preprohormone
from Drosophila melanogaster.Biochem. Biophys.
Res. Commun. 281, 544– 550. (doi:10.1006/bbrc.
2001.4402)
royalsocietypublishing.org/journal/rsob Open Biol. 9: 180259
11

61. Blackburn MB, Wagner RM, Kochansky JP, Harrison
DJ, Thomas-Laemont P, Raina AK. 1995 The
identification of two myoinhibitory peptides, with
sequence similarities to the galanins, isolated from
the ventral nerve cord of Manduca sexta.Regul.
Pept. 57, 213– 219. (doi:10.1016/0167-
0115(95)00034-9)
62. Melcher C, Bader R, Walther S, Simakov O, Pankratz
MJ. 2006 Neuromedin U and its putative Drosophila
homolog hugin. PLoS Biol. 4, e68. (doi:10.1371/
journal.pbio.0040068)
63. Schoofs A, Hu¨ckesfeld S, Schlegel P, Miroschnikow
A, Peters M, Zeymer M, Spieß R, Chiang A-S,
Pankratz MJ. 2014 Selection of motor programs for
suppressing food intake and inducing locomotion in
the Drosophila brain. PLoS Biol. 12, e1001893.
(doi:10.1371/journal.pbio.1001893)
64. Hu¨ckesfeld S, Peters M, Pankratz MJ. 2016 Central
relay of bitter taste to the protocerebrum by
peptidergic interneurons in the Drosophila brain.
Nat. Commun. 7, 12796. (doi:10.1038/
ncomms12796)
65. Schlegel P et al. 2016 Synaptic transmission
parallels neuromodulation in a central food-intake
circuit. Elife 5, e16799. (doi:10.7554/eLife.16799)
66. O’Donnell MJ, Rheault MR, Davies SA, Rosay P,
Harvey BJ, Maddrell SH, Kaiser K, Dow JA. 1998
Hormonally controlled chloride movement across
Drosophila tubules is via ion channels in stellate
cells. Am. J. Physiol. 274, R1039– R1049.
67. Al-Anzi B, Armand E, Nagamei P, Olszewski M,
Sapin V, Waters C, Zinn K, Wyman RJ, Benzer S.
2010 The leucokinin pathway and its neurons
regulate meal size in Drosophila.Curr. Biol. 20,
969–978. (doi:10.1016/j.cub.2010.04.039)
68. Radford JC, Davies SA, Dow JAT. 2002 Systematic
G-protein-coupled receptor analysis in Drosophila
melanogaster identifies a leucokinin receptor with
novel roles. J. Biol. Chem. 277, 38 810–38 817.
(doi:10.1074/jbc.M203694200)
69. de Haro M, Al-Ramahi I, Benito-Sipos J, Lo
´pez-Arias
B, Dorado B, Veenstra JA, Herrero P. 2010 Detailed
analysis of leucokinin-expressing neurons and their
candidate functions in the Drosophila nervous
system. Cell Tissue Res. 339, 321–336. (doi:10.
1007/s00441-009-0890-y)
70. Alvarez-Rivero J, Moris-Sanz M, Estacio-Go
´mez A,
Montoliu-Nerin M, Dı
´az-Benjumea FJ, Herrero P.
2017 Variability in the number of abdominal
leucokinergic neurons in adult Drosophila
melanogaster.J. Comp. Neurol. 525, 639– 660.
(doi:10.1002/cne.24093)
71. Zandawala M, Yurgel ME, Texada MJ, Liao S, Rewitz
KF, Keene AC, Na
¨ssel DR. 2018 Modulation of
Drosophila post-feeding physiology and behavior by
the neuropeptide leucokinin. PLoS Genet. 14,
e1007767. (doi:10.1371/journal.pgen.1007767)
72. Liu Y, Luo J, Carlsson MA, Na
¨ssel DR. 2015
Serotonin and insulin-like peptides modulate
leucokinin-producing neurons that affect
feeding and water homeostasis in Drosophila.
J. Comp. Neurol. 523, 1840– 1863. (doi:10.1002/
cne.23768)
73. Murphy KR et al. 2016 Postprandial sleep mechanics
in Drosophila.Elife 5, 1–19. (doi:10.7554/eLife.
19334)
74. Okusawa S, Kohsaka H, Nose A. 2014 Serotonin
and downstream leucokinin neurons modulate
larval turning behavior in Drosophila.J. Neurosci.
34, 2544–2558. (doi:10.1523/JNEUROSCI.3500-
13.2014)
75. Zandawala M, Marley R, Davies SA, Na
¨ssel DR. 2018
Characterization of a set of abdominal
neuroendocrine cells that regulate stress physiology
using colocalized diuretic peptides in Drosophila.
Cell. Mol. Life Sci. 75, 1099–1115. (doi:10.1007/
s00018-017-2682-y)
76. Veenstra JA. 1994 Isolation and structure of
the Drosophila corazonin gene. Biochem. Biophys.
Res. Commun. 204, 292– 296. (doi:10.1006/bbrc.
1994.2458)
77. Choi YJ, Lee G, Hall JC, Park JH. 2005 Comparative
analysis of Corazonin-encoding genes (Crz’s) in
Drosophila species and functional insights into
Crz-expressing neurons. J. Comp. Neurol. 482,
372–385. (doi:10.1002/cne.20419)
78. Lee G, Kim K-M, Kikuno K, Wang Z, Choi Y-J, Park
JH. 2008 Developmental regulation and functions of
the expression of the neuropeptide corazonin in
Drosophila melanogaster.Cell Tissue Res. 331,
659–673. (doi:10.1007/s00441-007-0549-5)
79. Kubrak OI, Lushchak OV, Zandawala M, Na
¨ssel DR.
2016 Systemic corazonin signalling modulates stress
responses and metabolism in Drosophila.Open Biol.
6, 160152. (doi:10.1098/rsob.160152)
80. So
¨derberg JAE, Carlsson MA, Na
¨ssel DR. 2012
Insulin-producing cells in the Drosophila brain also
express satiety-inducing cholecystokinin-like
peptide, drosulfakinin. Front. Endocrinol. (Lausanne).
3, 109. (doi:10.3389/fendo.2012.00109)
81. Sano H et al. 2015 The nutrient-responsive hormone
CCHamide-2 controls growth by regulating insulin-
like peptides in the brain of Drosophila
melanogaster.PLoS Genet. 11, e1005209. (doi:10.
1371/journal.pgen.1005209)
82. Li S, Torre-Muruzabal T, Søgaard KC, Ren GR, Hauser
F, Engelsen SM, Pødenphanth MD, Desjardins A,
Grimmelikhuijzen CJP. 2013 Expression patterns of
the Drosophila neuropeptide CCHamide-2 and its
receptor may suggest hormonal signaling from the
gut to the brain. PLoS ONE 8, e76131. (doi:10.1371/
journal.pone.0076131)
83. Ren GR et al. 2015 CCHamide-2 is an orexigenic
brain-gut peptide in Drosophila.PLoS One 10,
e0133017. (doi:10.1371/journal.pone.0133017)
84. Jørgensen LM, Hauser F, Cazzamali G, Williamson
M, Grimmelikhuijzen CJP. 2006 Molecular
identification of the first SIFamide receptor.
Biochem. Biophys. Res. Commun. 340, 696– 701.
(doi:10.1016/j.bbrc.2005.12.062)
85. Ubuka T, Tsutsui K. 2014 Evolution of gonadotropin-
inhibitory hormone receptor and its ligand. Gen.
Comp. Endocrinol. 209, 148– 161. (doi:10.1016/j.
ygcen.2014.09.002)
86. Martelli C et al. 2017 SIFamide translates hunger
signals into appetitive and feeding behavior in
Drosophila.Cell Rep. 20, 464– 478. (doi:10.1016/j.
celrep.2017.06.043)
87. Kim D-H, Shin M, Jung S-H, Kim Y-J, Jones WD.
2017 A fat-derived metabolite regulates a
peptidergic feeding circuit in Drosophila.PLoS Biol.
15, e2000532. (doi:10.1371/journal.pbio.2000532)
88. Albin SD, Kaun KR, Knapp J-M, Chung P, Heberlein
U, Simpson JH. 2015 A subset of serotonergic
neurons evokes hunger in adult Drosophila.
Curr. Biol. 25, 2435–2440. (doi:10.1016/j.cub.2015.
08.005)
89. Yang Z, Yu Y, Zhang V, Tian Y, Qi W, Wang L. 2015
Octopamine mediates starvation-induced
hyperactivity in adult Drosophila.Proc. Natl Acad.
Sci. USA 112, 5219–5224. (doi:10.1073/pnas.
1417838112)
90. Luo J, Becnel J, Nichols CD, Na
¨ssel DR. 2012 Insulin-
producing cells in the brain of adult Drosophila are
regulated by the serotonin 5-HT1A receptor. Cell.
Mol. Life Sci. 69, 471–484. (doi:10.1007/s00018-
011-0789-0)
91. Luo J, Lushchak OV, Goergen P, Williams MJ, Na
¨ssel
DR. 2014 Drosophila insulin-producing cells are
differentially modulated by serotonin and
octopamine receptors and affect social behavior.
PLoS ONE 9, e99732. (doi:10.1371/journal.pone.
0099732)
92. Park J-Y, Dus M, Kim S, Abu F, Kanai MI, Rudy B,
Suh GSB. 2016 Drosophila SLC5A11 mediates
hunger by regulating K
þ
channel activity. Curr. Biol.
26, 1965–1974. (doi:10.1016/j.cub.2016.05.076)
93. Dus M, Ai M, Suh GSB. 2013 Taste-independent
nutrient selection is mediated by a brain-specific
Na
þ
/solute co-transporter in Drosophila.Nat.
Neurosci. 16, 526– 528. (doi:10.1038/nn.3372)
94. Zhan YP, Liu L, Zhu Y. 2016 Taotie neurons regulate
appetite in Drosophila.Nat. Commun. 7, 13633.
(doi:10.1038/ncomms13633)
95. Marella S, Mann K, Scott K. 2012 Dopaminergic
modulation of sucrose acceptance behavior in
Drosophila.Neuron 73, 941– 950. (doi:10.1016/j.
neuron.2011.12.032)
96. Inagaki HK, Ben-Tabou de-Leon S, Wong AM,
Jagadish S, Ishimoto H, Barnea G, Kitamoto T, Axel
R, Anderson DJ. 2012 Visualizing neuromodulation
in vivo: TANGO-mapping of dopamine signaling
reveals appetite control of sugar sensing. Cell 148,
583–595. (doi:10.1016/j.cell.2011.12.022)
97. LeDue EE, Mann K, Koch E, Chu B, Dakin R, Gordon
MD. 2016 Starvation-induced depotentiation of
bitter taste in Drosophila.Curr. Biol. 26,
2854–2861. (doi:10.1016/j.cub.2016.08.028)
98. Flood TF, Iguchi S, Gorczyca M, White B, Ito K,
Yoshihara M. 2013 A single pair of interneurons
commands the Drosophila feeding motor program.
Nature 499, 83– 87. (doi:10.1038/nature12208)
99. Manzo A, Silies M, Gohl DM, Scott K. 2012 Motor
neurons controlling fluid ingestion in Drosophila.
Proc. Natl Acad. Sci. USA 109, 6307– 6312. (doi:10.
1073/pnas.1120305109)
100. Tissot M, Gendre N, Stocker RF. 1998 Drosophila
P[Gal4] lines reveal that motor neurons involved in
feeding persist through metamorphosis.
royalsocietypublishing.org/journal/rsob Open Biol. 9: 180259
12

J. Neurobiol. 37, 237–250. (doi:10.1002/(SICI)1097-
4695(19981105)37:2,237::AID-NEU4.3.0.CO;2-F)
101. Gordon MD, Scott K. 2009 Motor control in a
Drosophila taste circuit. Neuron 61, 373–384.
(doi:10.1016/j.neuron.2008.12.033)
102. Yapici N, Cohn R, Schusterreiter C, Ruta V, Vosshall
LB. 2016 A taste circuit that regulates ingestion by
integrating food and hunger signals. Cell 165,
715–729. (doi:10.1016/j.cell.2016.02.061)
103. Steck K, Walker SJ, Itskov PM, Baltazar C, Moreira
J-M, Ribeiro C. 2018 Internal amino acid state
modulates yeast taste neurons to support protein
homeostasis in Drosophila.Elife 7, 1– 29. (doi:10.
7554/eLife.31625)
104. Dus M, Lai JS-Y, Gunapala KM, Min S, Tayler TD,
Hergarden AC, Geraud E, Joseph CM, Suh GSB. 2015
Nutrient sensor in the brain directs the action of the
brain-gut axis in Drosophila.Neuron 87, 139– 151.
(doi:10.1016/j.neuron.2015.05.032)
105. Yang Z, Huang R, Fu X, Wang G, Qi W, Mao D, Shi
Z, Shen WL, Wang L. 2018 A post-ingestive amino
acid sensor promotes food consumption in
Drosophila.Cell Res. 28, 1013– 1025. (doi:10.1038/
s41422-018-0084-9)
106. Zhang T, Branch A, Shen P. 2013 Octopamine-
mediated circuit mechanism underlying controlled
appetite for palatable food in Drosophila.Proc. Natl
Acad. Sci. USA 110, 15 431 –15 436. (doi:10.1073/
pnas.1308816110)
107. Branch A, Zhang Y, Shen P. 2017 Genetic and
neurobiological analyses of the noradrenergic-like
system in vulnerability to sugar overconsumption
using a Drosophila model. Sci. Rep. 7, 1– 7. (doi:10.
1038/s41598-017-17760-w)
108. Krashes MJ, Waddell S. 2008 Rapid consolidation to
a radish and protein synthesis-dependent long-term
memory after single-session appetitive olfactory
conditioning in Drosophila.J. Neurosci.
28, 3103–3113. (doi:10.1523/JNEUROSCI.5333-
07.2008)
109. Aso Y et al. 2014 The neuronal architecture
of the mushroom body provides a logic for
associative learning. Elife 3, e04577. (doi:10.7554/
eLife.04577)
110. Aso Y et al. 2014 Mushroom body output neurons
encode valence and guide memory-based action
selection in Drosophila.Elife 3, e04580. (doi:10.
7554/eLife.04580)
111. Owald D, Felsenberg J, Talbot CB, Das G, Perisse E,
Huetteroth W, Waddell S. 2015 Activity of defined
mushroom body output neurons underlies learned
olfactory behavior in Drosophila.Neuron 86,
417–427. (doi:10.1016/j.neuron.2015.03.025)
112. Perisse E, Owald D, Barnstedt O, Talbot CB,
Huetteroth W, Waddell S. 2016 Aversive learning
and appetitive motivation toggle feed-forward
inhibition in the Drosophila mushroom body.
Neuron 90, 1086–1099. (doi:10.1016/j.neuron.
2016.04.034)
113. Al-Anzi B, Zinn K. 2018 Identification and
characterization of mushroom body neurons that
regulate fat storage in Drosophila.Neural Dev. 13,
18. (doi:10.1186/s13064-018-0116-7)
114. Itskov PM, Moreira J-M, Vinnik E, Lopes G, Safarik S,
Dickinson MH, Ribeiro C. 2014 Automated
monitoring and quantitative analysis of feeding
behaviour in Drosophila.Nat. Commun. 5, 4560.
(doi:10.1038/ncomms5560)
115. Kim IS, Dickinson MH. 2017 Idiothetic path
integration in the fruit fly Drosophila melanogaster.
Curr. Biol. 27, 2227–2238. (doi:10.1016/j.cub.2017.
06.026)
royalsocietypublishing.org/journal/rsob Open Biol. 9: 180259
13