Article

Endocrine signals fine-tune daily activity patterns in

Drosophila
Highlights Authors
d AKH induces activity under non-starving conditions via the Dennis Pauls, Mareike Selcho,
octopaminergic system Johanna Ra €derscheidt, ...,
Ronald P. Kühnlein, Martin J. Mueller,
d AKH prevents increased activity during the night by signaling Christian Wegener
to the fat body

d AKH promotes increased activity during the day by signaling

Correspondence
to octopaminergic neurons dennis.pauls@uni-leipzig.de (D.P.),
christian.wegener@biozentrum.
d The regulation of activity involves octopamine feedback to uni-wuerzburg.de (C.W.)
AKH-producing cells
In brief
Pauls et al. show that adipokinetic
hormone fine-tunes daily activity patterns
via two signaling pathways that are
counter-acting each other with regard to
locomotor activity and rest. These
branches comprise octopaminergic
neurons in the CNS and the energy-
storing fat body.

Pauls et al., 2021, Current Biology 31, 4076–4087

September 27, 2021 ª 2021 Elsevier Inc.
https://doi.org/10.1016/j.cub.2021.07.002 ll
 ll

Article
Endocrine signals fine-tune
daily activity patterns in Drosophila
Dennis Pauls,1,2,9,* Mareike Selcho,1,2 Johanna Ra €derscheidt,1 Kelechi M. Amatobi,3 Agnes Fekete,3 Markus Krischke,3
Christiane Hermann-Luibl,1 Ayten Gizem Ozbek-Unal,1 Nadine Ehmann,2 Pavel M. Itskov,4,5 Robert J. Kittel,2
Charlotte Helfrich-Förster,1 Ronald P. Kühnlein,6,7 Martin J. Mueller,3 and Christian Wegener1,8,10,*
1Neurobiology and Genetics, Theodor-Boveri-Institute, Biocenter, University of Würzburg, Am Hubland, 97074 Würzburg, Germany
2Department of Animal Physiology, Institute of Biology, Leipzig University, Talstraße 33, 04103 Leipzig, Germany
3Pharmaceutical Biology, Julius-von-Sachs-Institute, Biocenter, University of Würzburg, Julius-von-Sachs-Platz 2, 97082 Würzburg,

Germany
4Easy Behavior, rua Sao Tome e Principe 23, 2765-282 Estoril, Portugal
5Department of Neuroscience, University of Uppsala, Husargatan 3, 75237 Uppsala, Sweden
6Institute of Molecular Biosciences, University of Graz, Humboldtstraße 50, 8010 Graz, Austria
7BioTechMed-Graz, A-8010 Graz, Austria, Field of Excellence BioHealth-University of Graz, Humboldtstraße 50, 8010 Graz, Austria
8Twitter: @FlyNeuro
9Twitter: @DerPaule2009
10Lead contact

*Correspondence: dennis.pauls@uni-leipzig.de (D.P.), christian.wegener@biozentrum.uni-wuerzburg.de (C.W.)

https://doi.org/10.1016/j.cub.2021.07.002

SUMMARY

Animals need to balance competitive behaviors to maintain internal homeostasis. The underlying mecha-
nisms are complex but typically involve neuroendocrine signaling. Using Drosophila, we systematically
manipulated signaling between energy-mobilizing endocrine cells producing adipokinetic hormone (AKH),
octopaminergic neurons, and the energy-storing fat body to assess whether this neuroendocrine axis
involved in starvation-induced hyperactivity also balances activity levels under ad libitum access to food.
Our results suggest that AKH signals via two divergent pathways that are mutually competitive in terms of
activity and rest. AKH increases activity via the octopaminergic system during the day, while it prevents
high activity levels during the night by signaling to the fat body. This regulation involves feedback signaling
from octopaminergic neurons to AKH-producing cells (APCs). APCs are known to integrate a multitude of
metabolic and endocrine signals. Our results add a new facet to the versatile regulatory functions of APCs
by showing that their output contributes to shape the daily activity pattern under ad libitum access to food.

INTRODUCTION the fat body by mobilizing carbohydrates in form of the hemo-

Foraging behavior is fundamental to fitness and survival of ani-
mals. When food is scarce, starving animals explore their envi-
ronment to locate food sources. Depending on the amount of
available energy and time of day, foraging can suppress other
essential behaviors like sleep. Upon food restriction, rodents
and other vertebrates increase their activity that can be inter-
preted as foraging.1,2 In the fruit fly, Drosophila melanogaster,
food deprivation induces locomotor hyperactivity (food
search),3,4 which is triggered by a signaling pathway comprising
the peptide hormone adipokinetic hormone (AKH)3,5–7 and the
biogenic amine octopamine (OA).4,8 AKH is produced and
released by endocrine cells in the corpora cardiaca (CCA; a ho-
molog of the vertebrate anterior pituitary gland and an analog of
mammalian pancreatic alpha cells)9–11 and is required to main-
tain hemolymph sugar and lipid levels.3,6,7,10 Upon starvation,
carbohydrate levels in the hemolymph decrease, which is
indirectly sensed by the AKH-producing cells (APCs) through
ATP-sensitive potassium channels and AMP-activated protein
kinase.5,12 Similar to glucagon, AKH acts on energy stores in
lymph sugar trehalose (hypertrehalosemic effect), and by mobi-
lizing lipid depots (adipokinetic effect).7,13–15
OA, an analog to vertebrate noradrenaline, acts as an insect
stress or arousal hormone.16 OA is produced by octopaminergic
neurons (OANs) with somata located mostly along the midline in
the brain and ventral nerve cord.17 Among other functions, OA
exerts myotropic effects on skeletal muscles18 and mediates
the adaptation of muscle metabolism to the physiological de-
mands of locomotion.19–22
AKH and OA are crucial for starvation-induced hyperactiv-
ity, because this behavior is lost in flies with impaired AKH
or OA signaling.3,4,6–8 AKH activates a subset of OANs
required for starvation-induced hyperactivity.4,8 These neu-
rons express receptors for AKH and Drosophila insulin-like
peptides (DILPs). DILPs act antagonistically to AKH and sup-
press starvation-induced hyperactivity via the OANs.8 The
AKH-dependent increase in locomotor activity upon starva-
tion represents a direct neuromodulatory and behavioral ef-
fect of AKH and is not an indirect consequence of altered
energy metabolism.8

4076 Current Biology 31, 4076–4087, September 27, 2021 ª 2021 Elsevier Inc.

Article
Studies in various insect species suggest that AKH and OA
play a role in the regulation of locomotor activity when insects
have free access to food.23–30 In Drosophila, genetic overexpres-
sion of AKH enhances locomotor activity when food is freely
available.31 Unlike for starvation-induced hyperactivity, howev-
er, it is unclear whether and how AKH:OAN signaling contributes
to shape daily activity under ad libitum feeding conditions, or
whether increased AKH signaling is a consequence of a locomo-
tor-induced increase in energy demands.
Here, we used the genetically tractable fruit fly to address the
role of AKH:OAN signaling for locomotor activity under non-
starved conditions. We find that under these conditions AKH
signaling to OANs participates in shaping diel activity levels.
Surprisingly, AKH-induced energy mobilization appears to
antagonize the AKH:OAN-dependent promotion of activity by
preventing abnormally high nocturnal activity levels during an
initial phase of starvation. In turn, OA modulates the activity of
APCs via feedback signaling comprising OAMBII receptors.
Taken together, our results indicate that AKH:OAN and AKH:fat
body signaling in Drosophila helps to balance and adjust loco-
motor activity in the conflict of foraging and rest.
RESULTS
To validate our experimental conditions, we first confirmed that
AKH is required for starvation-induced hyperactivity and affects
starvation resistance (Figure S1). As previously shown,3,6,7
starved control flies showed significant starvation-induced hy-
peractivity compared to fed controls (Figures S1A–S1D). This
starvation-induced hyperactivity was impaired in flies lacking
APCs. Increased lipid stores in fat body cells contribute to the
higher starvation resistance observed in flies lacking AKH
signaling.7,14,15 In line with this, Akh>hid,rpr32,33 flies showed
higher starvation resistance during activity recording, because
on day 3, more than 80% of flies were still alive, compared to
less than 15% of starved controls (Figure S1E).
Octopaminergic neurons modulate diurnal activity
levels in fed flies
In addition to AKH, OA signaling is essential for starvation-
induced hyperactivity.4 We first tested whether the OA neuronal
system is also involved in modulating locomotor activity in flies
that have permanent access to food. We ablated Tdc2-Gal4-
positive neurons (Tdc2Ns; Tdc2-Gal4 specifically marks OANs
and tyraminergic neurons34–38) by the expression of UAS-hid,rpr
to test whether Tdc2Ns are involved in the modulation of loco-
motor activity in fed flies.39 As expected, Tdc2>hid,rpr flies
showed significantly reduced locomotor activity during the
day, whereas activity levels during the night were unaffected
(Figures 1A–1C). We next blocked neuronal transmission of
Tdc2Ns using UAS-DOrk40 and UAS-shibirets41 to exclude any
side effects on activity due to cell ablation. In line with the previ-
ous results, both electrical and chemical silencing of synaptic
transmission reduced locomotor activity levels specifically dur-
ing the day (Figures 1D–1I). Next, we constitutively increased
excitability of Tdc2Ns by expressing the bacterial sodium chan-
nel NaChBac.42 Tdc2>NaChBac flies showed constant
arrhythmic activity throughout the 24-h day. Activity levels
were higher during the usual midday siesta and during night,
ll
but lower during the morning. This resulted in normal cumulative
activity levels during the day but increased activity levels during
rest phases (Figures 1J–1L). We next activated Tdc2Ns via chan-
nelrhodopsin (chopXXL),43 limited to the photophase to not inter-
fere with the circadian clock. Accordingly, Tdc2>chopXXL flies
showed reduced activity during the morning activity bout, and
by trend increased activity during siesta which is consistent
with the data from the NaChBac experiment, albeit in an attenu-
ated form (Figure S2).
Our results are consistent with an earlier study showing that
Tdc2R054 and TbHnm18 mutants increase sleep mainly during
the day44,45 but partially differ from earlier reports showing that
Kir2.1-mediated electrical silencing of Tdc2Ns consistently
decreased locomotor activity and increased night sleep.44 None-
theless, the results show that OA/TA signaling promotes loco-
motor activity and hence wakefulness during the night44,46 or
under starvation4,8 and, in addition, shapes the general locomo-
tor activity level during the day when flies have permanent ac-
cess to food.
AKH:OAN signaling is important for the modulation of
daytime activity
Our results and data from Tdc2 mutants44 showed that impaired
OA signaling reduced locomotor activity specifically during the
day. We thus hypothesized that OANs signal primarily during
the day but not during the night to modulate locomotor activity
under ad libitum feeding conditions. Indeed, similar to Tdc2Ns
ablation (Figures 1A–1C), Akh>hid,rpr and Tdc2>AkhR-RNAi
flies showed significant reduced locomotor activity during the
day but not during the night (Figures 2A–2F). To demonstrate
that the observed effect of AKHR knockdown in Tdc2Ns is spe-
cific and not due to a general impairment of motor capability, we
performed a startle-induced negative geotaxis climbing assay.47
Tdc2>AkhR-RNAi flies performed normally in the climbing assay,
suggesting that motor capability was not generally affected by
our genetic manipulation (Figure S3). Interestingly, although the
ablation of APCs led to a reduction in activity throughout the
day (Figures 2A–2C), the average activity profile in
Tdc2>AkhR-RNAi flies suggests that the reduction in activity is
mainly based on reduced activity immediately after the morning
activity bout, whereas locomotor activity around the evening
peak was normal (Figure 2F). This daytime-specific reduction
of locomotor activity was partially rescued when the expression
of AkhR-RNAi was restricted to Tdc2Ns in the brain, using
tshGal80 (Figures S4A and S4B).48,49 Expression of GFP in
Tdc2,tshGal80>10xmyr::GFP flies revealed a strong reduction
of labeled Tdc2Ns in the ventral nerve cord (VNC). From 39
Tdc2-Gal4-positive VNC neurons50 7 cells (7.4 ± 2.07, n = 5)
escaped tshGal80-dependent suppression (Figure S4C),
whereas the number of Tdc2Ns in the brain was not affected.
These results suggest that Tdc2Ns in the VNC are involved in
the AKH:OAN signaling. To confirm the expression of AKHR in
cells in the VNC, we performed real-time PCR. We found
AKHR to be expressed in isolated adult VNCs, as well as in adult
brains and the larval fat body (Figure S4D).
The results so far are consistent with data from the cockroach
Periplaneta americana, where AKHR is expressed in thoracic
and abdominal octopaminergic dorsal unpaired median neurons
and signals both via Gq and Gs to activate cyclic AMP (cAMP)
Current Biology 31, 4076–4087, September 27, 2021 4077
 ll
and Ca2+-dependent second messenger pathways.29,30 To
confirm AKH signaling to OANs located in the VNC, we bath-
applied AKH to isolated VNCs (detached from the brain) and re-
corded changes in intracellular cAMP or free Ca2+ levels in the
mesothoracic Tdc2Ns expressing the genetically encoded
Epac1-camps50A51 or GCaMP6m52 sensors.
All mesothoracic Tdc2Ns responded to the adenylate cyclase
agonist NKH477 or the nicotinic receptor agonist carbamylcho-
line with robust increase in cAMP/free Ca2+ levels showing that
neurons were functional in our preparations. Bath-application
of 10 5 M AKH induced a significant cAMP increase, whereas
application of hemolymph-like HL3.1 saline solution as vehicle
control had no effect (Figures S5A and S5B).53 In contrast, free
4078 Current Biology 31, 4076–4087, September 27, 2021
Article
Figure 1. Tdc2-Gal4-positive neurons
(Tdc2Ns) modulate activity during the day
under non-starvation conditions
(A–C) Activity levels are reduced during the day in
Tdc2>hid,rpr flies compared to genetic controls
(p < 0.05 to Gal4/+ and p < 0.001 to UAS/+).
(D–F) Neuronal silencing of Tdc2Ns revealed a
similar effect with reduced activity of experimental
flies during the day.
(G–I) Conditional thermogenetic silencing of
Tdc2Ns using UAS-shits1 at restrictive temperature
also reduced activity levels during the day (p <
0.001 to Gal4/+ and p < 0.01 to UAS/+).
(J–L) Constitutive activation of Tdc2Ns via UAS-
NaChBac elicited significantly increased levels of
activity of experimental flies during the night (all p <
0.001). Conditional activation of Tdc2Ns via
chopXXL is shown in Figure S2. n = 26–32, signifi-
cance levels: ns, not significant; p > 0.05; *p < 0.05;
**p < 0.01; ***p < 0.001. Error bars indicate SEM.
intracellular Ca2+ levels did not signifi-
cantly change upon AKH bath application
compared to HL3.1 vehicle control (Fig-
ure S5C). These results suggest that acti-
vated AKHR modulate OAN activity via
cAMP-dependent pathways. Of note,
the observed AKH-induced increase in
cAMP levels occurred with a time lag of
at least 15 min after bath application of
the peptide (Figure S5B). We next overex-
pressed AKHR in Tdc2Ns and repeated
cAMP imaging in the mesothoracic
cluster. Overexpression of AKHR did
not significantly increase cAMP levels
beyond the level observed in controls.
Although the response to exogenous
AKH occurred significantly faster, re-
sponses were still delayed with a mean
lag time of around 10 min (Figure S5B).
This suggests that AKH induces slow
cAMP elevations in OANs even if AKHR
is overexpressed in these neurons.
To provide further genetic evidence for
the modulatory role of AKHR in OANs, we
rescued AKHR expression specifically in
Tdc2Ns in an AKHR1 mutant background and monitored loco-
motor activity (Figures 3A–3C).14 Rescued flies showed signifi-
cantly increased locomotor activity during the usual activity
bouts in the morning and evening and during the early and late
night, compared to AKHR-deficient controls (Figures 3A–3C).
No difference in locomotor activity was apparent during midday
siesta.
Taken together, the results suggest that AKH modulates OANs
in the VNC (and potentially in the brain)4,8 to increase locomotor
activity during the day when flies have permanent access to
food. Yet, AKHR is most strongly expressed by the fat
body14,23 where it mediates the hypertrehalosemic effect of
AKH (Figure 7). To test whether AKH:fat body signaling affects

Article
locomotor activity under ad libitum feeding conditions, we first
knocked down AKHR expression specifically in the fat body.
NP5267>AkhR-RNAi flies showed significantly increased loco-
motor activity during the night (Figures 3D–3F). Further, specific
rescue of AKHR in the fat body in an AkhR1 mutant background
significantly reduced locomotor activity during morning and eve-
ning activity bouts as well as during midday siesta (phenocopy-
ing the impairment of AKH:OAN signaling; see Figures 1 and 2),
whereas locomotor activity appeared to be normal during the
night (Figures 3G–3I).
OAN signaling modulates AKH cells in the corpora
cardiaca
These results so far support a role for AKH:OAN signaling in loco-
motor activity under ad libitum feeding conditions. We next
wondered whether OANs provide feedback signals to the APCs
in the corpora cardiaca (CCA), and traced the projections of
Tdc2-positive cells in whole body sections of flies by Tdc2 immu-
nostaining (Figures 4A and 4B). Besides arborizations in the brain
and VNC, we found Tdc2-positive fibers projecting to the CCA-
hypocerebral ganglion complex (CCA-HG complex, also known
as stomodeal ganglion) (Figure 4B, inset) presumably via the sto-
modeal nerve (Figure 4B, asterisk). Thus, OA may be released in
vicinity of the AKH cells. Among the different OA receptors, OAM-
BII has been responsible for most OA actions,16 and increases
intracellular Ca2+ levels in heterologous expression systems as
well as in Drosophila neurons.54–56 Using primers targeting
different exons of the OAMBII gene,57 we confirmed the expres-
sion of the OAMB-K3 isoform (Figure 4D, OAMB-K3 contains
exon 7 but not 8) in the CCA-HG complex, whereas signals for
the OAMB-AS isoform (OAMB-AS contains exon 8 instead of 7)
were not detected (Figure 4C). To test whether APCs express
functional OA receptors, we expressed the genetically encoded
Ca2+ sensor GCaMP6m in APCs and imaged the effect of bath-
applied OA. For these experiments, we used the whole larval
ll
Figure 2. AKH:OAN signaling is important
for the modulation of daytime activity
(A–C) Akh>hid,rpr flies showed significantly
reduced activity (all p < 0.001) specifically during
the day. Starvation responses of Akh>hid,rpr flies
are shown in Figure S1.
(D–F) Knockdown of AKHR in Tdc2Ns (Tdc2-Gal4-
positive neurons) specifically affected locomotor
activity during the day (p < 0.01 to Gal4/+ and p <
0.001 to UAS/+). Data for general mobility are
shown in Figure S3. The identification of Tdc2Ns of
the ventral nerve cord as main target cells is shown
in Figures S4 and S5. n = 26–32, significance
levels: ns, p > 0.05; *p < 0.05; **p < 0.01; ***p <
0.001. Error bars indicate SEM.
CNS with attached ring glands because it
was not feasible to dissect intact adult
CCA-HG-brain preparations. Because at
low trehalose levels, the Ca2+ concentra-
tion is generally high in APCs, we adopted
the procedure of Kim and Rulifson5 to im-
age Ca2+ changes in these cells: prepara-
tions were first pre-incubated in 80 mM
trehalose solved in HL3.1 that decreased free intracellular Ca2+
concentration to basal level (Figure 4E: arrow at 100 s). Subse-
quent application of 1 mM OA + 80 mM trehalose (Figure 4E: ar-
row at 300 s) significantly increased intracellular Ca2+ levels indi-
cated by an increase of fluorescence intensity (Figure 4E, blue
line). In contrast, no changes in Ca2+ levels were visible after
bath application of the control solution (Figure 4E, gray line), sug-
gesting that APCs respond with increased intracellular Ca2+ levels
to OA (Figure 4F). A knockdown of OAMBII in these cells
(Akh>OAMBII-RNAi) resulted in significantly increased nocturnal
activity (Figures 4G–4I), suggesting that OANs modulate AKH
release as reported for the locust.58,59
AKH:fat body signaling prevents an increase of activity
during the night
Based on these results and previous studies that suggested he-
molymph trehalose levels to be at a trough at the end of night and
immediately after lights on,60 we hypothesized that AKH modu-
lates locomotor activity in a balanced state-dependent manner
through AKH:OAN and AKH:fat body signaling. Although
AKH:OAN signaling may promote activity during the day,
AKH:fat body signaling may stabilize rest during the night.
To test this hypothesis, we monitored locomotor activity of
NP5267>AkhR-RNAi for 3 days under starving conditions. We
assumed that flies lacking AKH:fat body signaling cannot
compensate for the lack of food with energy mobilization from
the fat body and hence will show an earlier onset of nocturnal
starvation-induced hyperactivity than control flies. All flies
showed starvation-induced hyperactivity during the experiment
(Figures 5A–5C). For NP5267>AkhR-RNAi flies, however, this
hyperactivity started earlier and became obvious during the
first night around 12 h after starvation onset (Figure 5A, red
arrow). In contrast, control flies did not became hyperactive until
the second day (Figures 5B and 5C, red arrows), with starvation-
induced hyperactivity fully developed first during the second
Current Biology 31, 4076–4087, September 27, 2021 4079
 ll
night in both genetic controls (Figures 5D–5F). Obviously,
endogenous AKH:fat body signaling prevented undue high
nocturnal activity levels in consequence of food search in control
flies during the first night, presumably by mobilizing energy from
the fat body into the hemolymph (Figures 5D–5F).
Impaired AKH-dependent carbohydrate mobilization is
compensated by dietary uptake
So far, our results suggest that AKH signaling modulates loco-
motor activity not only under starved conditions but also when
animals have permanent access to food. The effect was most
obvious during the night when feeding-related behaviors are
normally absent or strongly reduced.60 We assumed that during
these nocturnal non-feeding phases, flies usually compensate
the reduced dietary carbohydrate intake via the hypertrehalose-
mic effect of AKH. However, flies deficient of AKH:fat body
signaling are unable to maintain high trehalose levels.3,6,10
This may lead to increased AKH release, which in turn will pro-
mote foraging activity via the integrated increase of AKH:OAN-
signaling (Figure 5A) especially toward the end of the night,
analogous to, but less pronounced as, starvation-induced hy-
peractivity. To find out whether APCs of flies with impaired
4080 Current Biology 31, 4076–4087, September 27, 2021
Article
Figure 3. AKH balances day and night loco-
motor activity via different signaling routes
(A–C) Genetic rescue of AKHR in Tdc2Ns (Tdc2-
Gal4-positive neurons) rescued locomotor activity
during the day (all p > 0.05); however,
AkhR1,Tdc2 > AkhR,AkhR1 flies showed signifi-
cantly higher activity during morning and evening
activity bouts (all p < 0.001 at ZT0 and ZT12). In
addition, experimental flies showed significantly
increased activity levels during the night (all p <
0.001).
(D–F) Flies with fat body-specific knockdown of
AKHR showed increased levels of activity during
the night (p < 0.01 to Gal4/+ and p < 0.001 to
UAS/+).
(G–I) Specific rescue of AKHR in the fat body eli-
cited the opposite effect seen to the rescue in
OANs (see A–C). Expression of AKHR specifically
in the fat body rescued activity levels during the
night; in contrast, locomotor activity was signifi-
cantly reduced during the day (all p < 0.01). Note
that the experiments in (A)–(C) and (G)–(I) were
conducted at the same time. The controls shown
are identical in the two experiments. n = 13–32 for
behavioral data, significance levels: ns, p > 0.05;
*p < 0.05; **p < 0.01; ***p < 0.001. Error bars indi-
cate SEM.
AKH signaling are more active during
the night, we first used Ca2+-dependent
nuclear import of LexA (CaLexA),61 an
integrator of long-term Ca2+ activity. As
expected, CaLexA-driven GFP expres-
sion increased significantly upon 24 h
of starvation in AKH>CaLexA flies (Fig-
ures S6B, S6C, S6F, and S6G), showing
that the method is apt to monitor APC
activity.
On food, we found a weak and statistically insignificant in-
crease in CaLexA-dependent GFP expression in Akh>CaLexA
controls between end of the day (ZT10) and middle (ZT18) or
end (ZT22) of the night (Figures S6E–S6G). In Akh;amon-RNAi>-
CaLexA flies with reduced AKH signaling,10 we found a slight but
insignificant increase in CaLexA-driven GFP expression
compared to controls (Figures S6D, S6F, and S6G). These re-
sults suggest that APCs generally have a low activity on food
and speak against a significantly increased need for energy
mobilization in flies with impaired AKH signaling under our exper-
imental conditions. We therefore tested whether the loss of AKH-
dependent carbohydrate mobilization is compensated by dietary
carbohydrate uptake. First, we quantified hemolymph sugar
levels during the light phase under starved and ad libitum feeding
conditions using ultra performance liquid chromatography-tan-
dem mass spectrometry. Indeed, APC-ablated flies under fed
conditions showed similar hemolymph levels of sucrose, fruc-
tose, trehalose, and glucose, although trehalose and fructose
levels were somewhat reduced (Figures 6A–6D). Under starved
conditions, however, trehalose levels became depleted in
APC-ablated flies (Figure 6C), as previously shown.6 Glucose
levels were strongly, although not significantly, reduced

Article
(Figure 6D). In contrast, fructose and sucrose levels did not
significantly differ compared to controls and were generally
very low (Figures 6A and 6B), likely because the hemolymph ti-
ters of these sugars originate mostly from food uptake. These re-
sults support our assumption that flies deficient in AKH signaling
compensate impaired carbohydrate mobilization by food uptake
and explain the similar activity levels of APCs between controls
and AKH downregulated flies as determined with CaLexA.
Next, we quantified trehalose, glucose, and diacylglycerol
levels every 4 h during the day under ad libitum feeding condi-
tions. In line with the results in Figure 6C, trehalose levels of
Akh>hid,rpr flies were lower but not significantly reduced
compared to both controls (Figure 6E). As expected from previ-
ous measurements (Figure 6D), the glucose concentrations were
similar in both Akh>hid,rpr and control flies (Figure 6F). Yet, the
temporal pattern differed. Akh>hid,rpr flies showed constant
low levels of trehalose during the night, whereas control flies dis-
played increased levels of trehalose during rest phases (the
siesta and night) suggesting trehalose mobilization from the fat
body (Figure 6E). The glucose level stayed constant within a
certain range in controls, with a drop at the end of the night in
Akh>w control flies (Figure 6F). In Akh>hid,rpr flies, however,
the glucose titer was low at ZT2 and steadily increased until
ZT14 to then drop at ZT18 and increase again to a high level at
ZT22. This suggests feeding activity in Akh>hid,rpr flies espe-
cially around ZT22, given that Akh>hid,rpr flies are impaired in
mobilizing carbohydrates. Interestingly, no obvious differences
in hemolymph diacylglycerol levels were detectable between
the genotypes (Figure 6G). Strong increases of diacylglycerol
occurred at times when the glucose titer dropped even in
ll
Figure 4. AKH release is modulated by OAN
feedback signaling
(A and B) Whole body sections displayed arbori-
zations of Tdc2-positive neurons (white) in the
CCA-HG complex (inset in B). (B) A magnification
of the red squared region in (A). The CCA-HG
complex is highlighted in red in the inset. White,
anti-Tdc2; orange, anti-synapsin.
(C and D) Real-time PCR analysis confirmed
expression of the OA receptor isoform OAMBII-K3
but not OAMBII-AS (red #) in the CCA-HG complex
(blue #).
(E and F) Average changes in Ca2+ levels of larval
AKH cells to bath applied OA and HL3.1 (Akh>G-
CaMP6m). AKH cells were pre-incubated with
80 mM trehalose to reduce intracellular Ca2+ levels
to baseline (arrow at 100 s). AKH cells responded
to bath applied OA (arrow at 300 s) with a strong
increase in intracellular Ca2+ levels (blue line; F); in
contrast, significant responses were not obvious
in response to the vehicle control (gray line).
(G–I) Akh>OAMBII-RNAi flies showed significantly
increased activity during the night (all p < 0.001).
n = 29–32 for behavioral data, significance levels:
ns, p > 0.05; *p < 0.05; **p < 0.01; ***p < 0.001 ***.
CCA, corpora cardiaca; Ctrl, control; BR, brain; E,
exon; HG, hypocerebral ganglion; OA, octop-
amine; PV, proventriculus; ROI, region of interest;
VNC, ventral nerve cord. Error bars indicate SEM.
Akh>hid,rpr flies (Figure 6F), providing further evidence for an
AKH-independent lipid mobilization pathway.14
We next monitored feeding in individual flies on 5 mM sucrose
versus 10% yeast in a two-choice flyPAD assay62 (Figures 6H–6J
and S7). Generally, flies consumed significantly more sucrose and
yeast during the day than during the night (Figure S7). Sucrose
consumption was non-significantly elevated in Akh>hid,rpr flies
both during day and night. Analysis at a higher temporal resolution
revealed significantly increased sucrose feeding in Akh>hid,rpr
flies at the end of the night and after lights-on during morning ac-
tivity compared to controls (Figure 6H). A less pronounced in-
crease in sucrose feeding was further visible around the end of
the day (coinciding with the start of evening activity), as well as
during the night, although levels did not reach significance except
for ZT 10–11 (Figure 6H). Thus, genetic ablation of APCs led to an
earlier onset of sucrose feeding activity with a slight general in-
crease throughout the night. In contrast, total yeast consumption
was not significantly affected in Akh>hid,rpr flies during the day
and night (Figure S7). A significantly elevated level of yeast feeding
was also not visible at higher temporal resolution across the day
(Figure 6I). Likewise, total feeding was not affected by APC abla-
tion (Figure 6J). These results show that ablation of APCs exerts a
specific effect on sucrose but not yeast feeding. The temporal
course of sucrose feeding in Akh>hid,rpr flies (Figure 6H) is
compatible with their glucose hemolymph profile (Figure 6F) that
shows increasing concentrations during the day and a drop in
the middle of the night when feeding activity was basically absent
in Akh>hid,rpr flies (Figure 6H). Downregulation of OAMBII recep-
tor in APCs did not affect yeast feeding or total food consumption.
Similarly, sucrose feeding was largely unaltered compared to
Current Biology 31, 4076–4087, September 27, 2021 4081
 ll
Figure 5. Onset of starvation-induced hyperactivity
(A–F) NP5267>AkhR-RNAi flies showed an early onset of starvation-induced hyperactivity during the first night (after ~12 h of starvation; red arrow). In contrast,
AKH:fat body signaling in control flies allowed a short-term adaptation under starvation and prevented abnormally high night activity. Red arrows indicate first
obvious appearance of starvation-induced hyperactivity. n = 13–32; significance levels: ns, p > 0.05; *p < 0.05; **p < 0.01; ***p < 0.001. Error bars indicate SEM.
controls, although during the first hours after lights-on, a small in-
crease is visible in Akh>OAMBII-RNAi flies compared to controls
(Figure S7).
Taken together, our results propose that Akh>hid,rpr flies are
not able to mobilize trehalose from the fat body and rescue he-
molymph carbohydrate titers by increased food uptake at certain
times of the day.
DISCUSSION
The circadian activity and rest pattern of animals is determined
by a multitude of intertwined systems and mechanisms on
different levels. The general organization is largely set out by
the circadian clock and the sleep homeostat.63 Yet, internal
and external signals can dramatically alter activity/rest levels
and induce locomotor activity during normal resting phases to
meet physiological demands or to escape from harmful situa-
tions. A particularly well studied example is the starvation-
induced locomotor activity in Drosophila, which is triggered by
the action of the peptide hormone AKH on neurons that release
the arousal signal OA.3,4,8 On food, AKH signaling was thought to
not affect general activity levels, because AKH or AKHR mutants
show similar activity levels as controls7,64 and display normal
circadian activity patterns.3,65 Our study now suggests that the
unaltered activity levels and activity/rest patterns in AKH and
AKHR mutants are an outcome of the combined loss of both
branches of AKH signaling to OANs (regulating locomotor
4082 Current Biology 31, 4076–4087, September 27, 2021
Article
activity via feedforward and feedback signaling) and to the fat
body (regulating energy mobilization).
Based on our results and available literature, we propose the
following model for the modulatory action of AKH on balancing
activity under non-starving conditions (Figure 7). When flies
need to mobilize energy (e.g., due to increased physical activity),
AKH is released into the circulation to mobilize carbohydrate and
lipid fat body stores and simultaneously activates OANs in the
CNS.8,66 The resulting increase in hemolymph trehalose and
glucose negatively feeds back to APCs, reducing their secretory
activity via opening of ATP-sensitive potassium channels.5,67
This reduces AKH:fat body signaling, and prevents a strong ac-
tivity-promoting effect of time-delayed AKH:OAN signaling. Un-
der starved conditions, however, energy stores become
depleted, which leads to a strong and long-lasting increase in
the hemolymph AKH titer once carbohydrate levels fall, which
in turn stimulates hyperactivity via the activation of OANs.3,6–8
When AKH:fat body signaling is impaired, flies can no longer
efficiently mobilize fat body energy stores upon depletion of he-
molymph carbohydrate levels. During the day, hemolymph sugar
levels have then to be restored by feeding, as suggested by the
observed increase in trehalose and glucose titers during photo-
phase and the increased sucrose feeding during morning activity,
the main time of food intake.60 In other words, impairing AKH:fat
body signaling in flies with free access to food does not induce hy-
peractivity because flies can perform compensatory feeding.
However, flies are non-continuous feeders that feed little during
 ll
Article

Figure 6. Carbohydrate hemolymph titers and feeding activity

(A–D) Hemolymph sugar levels in Akh>hid,rpr flies and controls: while sucrose and fructose levels did not differ between genotypes and conditions (starved
versus fed), trehalose levels were depleted in experimental flies under starvation conditions (p = 0.078 to Gal4/+ and p = 0.063 to UAS/+). In addition, glucose
levels were reduced in Akh>hid,rpr flies between fed and starved conditions (p < 0.01), without any effect in control flies (p > 0.05; all N = 3–4).
(E) In control flies, hemolymph trehalose levels rise during the night. In contrast, trehalose levels stay low during the night in NP5267>AkhR-RNAi flies.
(F) Hemolymph glucose levels increase throughout 24 h in experimental flies indicating higher food intake compared to controls.
(G) No change was detectable on hemolymph lipid levels. APC cell activity analysis is shown in Figure S6.
(H–J) Time course of feeding for sucrose (H), yeast extract (I), or sucrose and yeast combined (J) in Akh>hid,rpr flies and controls. Significant differences between
APC-ablated flies and controls are visible for sucrose feeding mostly at the end of the night and the first hours after lights on, whereas feeding on yeast or total
food consumption (provided in Figure S7) was not significantly altered. Time course of feeding of Akh>OAMB-RNAi flies is shown in Figure S7. n = 35–36 for
feeding data, significance levels: ns, p > 0.05; *p < 0.05; **p < 0.01; ***p < 0.001 (p < 0.06 [*]). APC, AKH-producing cell. Error bars indicate SEM.

the night.60,68,69 During prolonged nocturnal rest phases, a loss of
AKH-dependent carbohydrate mobilization thus leads to
increased AKH release, which positively modulates OANs and
the observed increase in late nocturnal activity, and earlier and
more pronounced onset of morning feeding activity on food.
This scenario is in line with the upregulation of Akh expression
in AKHR mutants.8 It also fits with our observation of a drop in
glucose levels in the middle of the night, followed by an increase
at late night coinciding with the earlier onset of feeding activity in
flies with ablated APCs. These flies obviously compensate the
loss of carbohydrate mobilization by feeding during the nocturnal
phase, but without triggering increased nocturnal locomotor
activity because AKH:OAN signaling is also impaired. A time-
restricted increased feeding in AKH-deficient flies is also sug-
gested by the transcriptional upregulation of the orexigenic pep-
tides NPF and CCHa2 in AKH mutant flies, which was, however,
not coupled to increased total feeding.70 Wild-type flies increase
Akh mRNA levels 3 h after the dark-light or light-dark transition,
the time points at which starved flies showed pronounced starva-
tion-induced hyperactivity.71 Interestingly, times of increased

Current Biology 31, 4076–4087, September 27, 2021 4083

 ll
sucrose feeding in flies with ablated APCs occurred around ZT3
(significant increase, coupled with low glucose titers) and ZT15
(small and insignificant increase, coupled with high glucose titers)
in our study.
When AKH:OAN signaling is specifically impaired on food, flies
are still sufficiently able to mobilize their energy stores (via AKH:
fat body signaling) to meet the metabolic requirements during
rest at night, as evidenced by the observed normal nocturnal ac-
tivity patterns. However, when AKH:OAN signaling is impaired
during the light phase, the increased energy demand during
diurnal physical activity triggers AKH release that is then no
longer able to stimulate activity-promoting OA signaling, explain-
ing the observed reduced locomotor activity in Tdc2>AkhR-
RNAi flies.
The encountered effects of manipulating the two AKH
signaling branches are not very strong, and our study does
not suggest that AKH is a major factor in shaping circadian
locomotor activity patterns. However, AKH appears to be
one of presumably several factors that fine-tune locomotor
activity patterns to meet physiological requirements under di-
etary stress-free conditions. OANs and APCs receive a multi-
tude of converging and interconnected inputs, especially from
insulin-producing cells (IPCs), with extensive feedbacks (see
Koyama et al.,72 Ahmad et al.,73 and Braco et al.74). For
example, OANs in the SEZ express receptors for AKH and
DILPs and integrate the nutritional status by responding to
AKH and DILPs in an antagonistic fashion.8 In turn, APCs ex-
press a range of OA receptors74 and are negatively modu-
lated by IPCs. This opens the possibility for indirect signaling
from IPCs to OANs via AKH signaling in parallel to direct
IPC:OAN signaling.73,74 AKH signaling is also required for
the carbohydrate-dependent release of DILP375 and re-
presses expression of DILP2,3 and DILP5.72 Moreover,
feeding and diet, as well as gustatory perception, also influ-
ence activity and sleep,76–78 and AKHR is expressed by gus-
tatory neurons including those that express the sugar-sensing
receptor Gr5a.15 AKH production or release is also modulated
by diurnal myokine secretion from muscles.79 Thus, any
manipulation of AKH or OA signaling likely results in complex
neuroendocrine changes. This complexity may also underlie
differences in the effect and effect size observed between
different studies
4084 Current Biology 31, 4076–4087, September 27, 2021
Article
Figure 7. AKH signaling to OANs and the fat
body balances competing rest and feeding
drives
In response to a negative balance in hemolymph
carbohydrate homeostasis, AKH is released from
neuroendocrine cells in the corpora cardiaca and
functions via two signaling pathways: (a) AKH
signaling to OANs induces food-search behavior
(during the day); and (b) at the same time AKH
mobilizes energy from the fat body to restore he-
molymph sugar levels. (c) In addition, APC activity
is negatively regulated by OA-feedback signaling,
which inhibits the activation of locomotor activity
by OANs and thereby prevents abnormally high
activity levels during the night. AKH, adipokinetic
hormone; AKHR, AKH receptor; ATP, adenosine
triphosphate; OA, octopamine; OAN, octopami-
nergic neuron.
OA and AKH signaling are intertwined
OANs in the SEZ are important to integrate AKH and DILP signals
to modulate activity levels in response to the nutritional situa-
tion.8 Further wake-promoting OANs are found in the ASM clus-
ter in the supraesophageal zone.80 Consistent with findings from
the cockroach,29,30 we now provide evidence for a role of OANs
of the VNC in the AKH signaling pathway that modulates activity
levels under ad libitum feeding conditions. Tdc2>AkhR-RNAi
flies showed reduced activity levels during daytime, whereas
this phenotype was reduced in flies expressing tshGal80, which
blocks expression in OANs located in the VNC. Further, bath-
applied AKH induced a slow but significant increase in intracel-
lular cAMP in mesothoracic OANs. The slow increase might
suggest indirect action of AKH, yet transgenic (over)expression
of AKHR in the OANs only shortened the lag time by one-third.
Injection of AKH into the mesothoracic neuropil of the moth
Manduca sexta leads to increased activity of motoneurons,81
supporting an activity-promoting action of AKH in the VNC. Inter-
estingly, the AKH-induced increase in intracellular Ca2+ activity
of larval prothoracic gland cells also requires 15 min of peptide
bath application to set in.82 Clearly, further studies will be
required to clarify whether the observed slow cAMP increase
in OANs is due to direct or indirect action of AKH.
Our results provide several lines of evidence for a modulation
of APC activity by OA and the OAMBII-K3 receptor and may sug-
gest an involvement of OAMBII receptors in modulating locomo-
tor activity but not feeding. Yet, the modulation of APC activity
and the role of OA herein is very complex and includes various
peptides and biogenic amines.74,83 A recent cell-specific tran-
scriptomic study supports the expression of OAMBII in APCs
but also found expression of most other OA receptors.74 At the
moment, we can only speculate about the functional significance
of OA modulation of APCs and the role of OAMBII therein.
STAR+METHODS
Detailed methods are provided in the online version of this paper
and include the following:
d KEY RESOURCES TABLE
d RESOURCE AVAILABILITY
B Lead contact

Article
B Materials availability
B Data and code availability
d EXPERIMENTAL MODEL AND SUBJECT DETAILS
B Fly strains
d METHOD DETAILS
B Behavioral assays
B Live cell imaging
B Immunostainings
B CaLexA labeling
B PCR
B Ultra performance liquid chromatography-tandem
mass spectrometry (UPLC-MS/MS) measurements
d QUANTIFICATION AND STATISTICAL ANALYSIS
B Statistical Analysis
SUPPLEMENTAL INFORMATION
Supplemental information can be found online at https://doi.org/10.1016/j.
cub.2021.07.002.
ACKNOWLEDGMENTS
The authors thank Bryon Hughson, Wolf Huetteroth, and Pamela Menegazzi
for fruitful discussions and comments on the manuscript, Gertrud Gramlich
and Susanne Klühspies for excellent technical assistance, and Jay Hirsh and
the Bloomington Stock centre for providing flies. This work was supported
by intramural support of the University of Würzburg (to C.W.), the University
of Graz (to R.P.K.), and the Deutsche Forschungsgemeinschaft (SFB1047 ‘‘In-
sect timing’’, project B2 to C.W., DP1979/2-1 and DP1979/3-1 to D.P.,
PA3241/2-1 to M.S., FOR 2149/P03 to R.J.K., and FO207/14-1 to C.H.-F.).
AUTHOR CONTRIBUTIONS
D.P., M.S., J.R., R.J.K., R.P.K., and C.W. designed the experiments. D.P.,
M.S., J.R., K.M.A., M.K., C.H.-L., A.G.O.-U., N.E., P.M.I., and C.W. performed
the experiments. D.P., M.S., J.R., A.F., M.K., C.H.-L., P.M.I., C.H.-F., M.J.M.,
and C.W. analyzed the results. D.P. and C.W. wrote the article. All authors pro-
vided comments and approved the manuscript.
DECLARATION OF INTERESTS
P.M.I. is selling and provides service to flyPAD devices through https://www.
flypad.rocks. The other authors declare no competing interests.
INCLUSION AND DIVERSITY
We worked to ensure sex balance in the selection of non-human subjects.
While citing references scientifically relevant for this work, we also actively
worked to promote gender balance in our reference list. The author list of
this paper includes contributors from the location where the research was con-
ducted who participated in the data collection, design, analysis, and/or inter-
pretation of the work.
Received: February 16, 2020
Revised: February 24, 2021
Accepted: July 2, 2021
Published: July 29, 2021
REFERENCES
1. Hebebrand, J., Exner, C., Hebebrand, K., Holtkamp, C., Casper, R.C.,
Remschmidt, H., Herpertz-Dahlmann, B., and Klingenspor, M. (2003).
Hyperactivity in patients with anorexia nervosa and in semistarved rats:
evidence for a pivotal role of hypoleptinemia. Physiol. Behav. 79, 25–37.
ll
2. Hall, J.F., and Hanford, P.V. (1954). Activity as a function of a restricted
feeding schedule. J. Comp. Physiol. Psychol. 47, 362–363.
3. Lee, G., and Park, J.H. (2004). Hemolymph sugar homeostasis and starva-
tion-induced hyperactivity affected by genetic manipulations of the adipo-
kinetic hormone-encoding gene in Drosophila melanogaster. Genetics
167, 311–323.
4. Yang, Z., Yu, Y., Zhang, V., Tian, Y., Qi, W., and Wang, L. (2015).
Octopamine mediates starvation-induced hyperactivity in adult
Drosophila. Proc. Natl. Acad. Sci. USA 112, 5219–5224.
5. Kim, S.K., and Rulifson, E.J. (2004). Conserved mechanisms of glucose
sensing and regulation by Drosophila corpora cardiaca cells. Nature
431, 316–320.
6. Isabel, G., Martin, J.-R., Chidami, S., Veenstra, J.A., and Rosay, P. (2005).
AKH-producing neuroendocrine cell ablation decreases trehalose and in-
duces behavioral changes in Drosophila. Am. J. Physiol. Regul. Integr.
Comp. Physiol. 288, R531–R538.
7. Gáliková, M., Diesner, M., Klepsatel, P., Hehlert, P., Xu, Y., Bickmeyer, I.,
Predel, R., and Kühnlein, R.P. (2015). Energy Homeostasis Control in
Drosophila Adipokinetic Hormone Mutants. Genetics 201, 665–683.
8. Yu, Y., Huang, R., Ye, J., Zhang, V., Wu, C., Cheng, G., Jia, J., and Wang,
L. (2016). Regulation of starvation-induced hyperactivity by insulin and
glucagon signaling in adult Drosophila. eLife 5, e15693.
9. Wang, S., Tulina, N., Carlin, D.L., and Rulifson, E.J. (2007). The origin of
islet-like cells in Drosophila identifies parallels to the vertebrate endocrine
axis. Proc. Natl. Acad. Sci. USA 104, 19873–19878.
10. Rhea, J.M., Wegener, C., and Bender, M. (2010). The proprotein conver-
tase encoded by amontillado (amon) is required in Drosophila corpora car-
diaca endocrine cells producing the glucose regulatory hormone AKH.
PLoS Genet. 6, e1000967.
11. Park, S., Bustamante, E.L., Antonova, J., McLean, G.W., and Kim, S.K.
(2011). Specification of Drosophila corpora cardiaca neuroendocrine cells
from mesoderm is regulated by Notch signaling. PLoS Genet. 7,
e1002241.
12. Braco, J.T., Gillespie, E.L., Alberto, G.E., Brenman, J.E., and Johnson,
E.C. (2012). Energy-dependent modulation of glucagon-like signaling in
Drosophila via the AMP-activated protein kinase. Genetics 192, 457–466.
13. Ga€de, G., and Auerswald, L. (2003). Mode of action of neuropeptides from
the adipokinetic hormone family. Gen. Comp. Endocrinol. 132, 10–20.
14. Grönke, S., Müller, G., Hirsch, J., Fellert, S., Andreou, A., Haase, T.,
€ckle, H., and Kühnlein, R.P. (2007). Dual lipolytic control of body fat stor-
Ja
age and mobilization in Drosophila. PLoS Biol. 5, e137.
15. Bharucha, K.N., Tarr, P., and Zipursky, S.L. (2008). A glucagon-like endo-
crine pathway in Drosophila modulates both lipid and carbohydrate ho-
meostasis. J. Exp. Biol. 211, 3103–3110.
16. Roeder, T. (2005). Tyramine and octopamine: ruling behavior and meta-
bolism. Annu. Rev. Entomol. 50, 447–477.
17. Monastirioti, M. (1999). Biogenic amine systems in the fruit fly Drosophila
melanogaster. Microsc. Res. Tech. 45, 106–121.
18. Socha, R., Kodrı́k, D., and Zemek, R. (2008). Stimulatory effects of bio-
amines norepinephrine and dopamine on locomotion of Pyrrhocoris apte-
rus (L.): is the adipokinetic hormone involved? Comp. Biochem. Physiol. B
Biochem. Mol. Biol. 151, 305–310.
19. Mentel, T., Duch, C., Stypa, H., Wegener, G., Müller, U., and Pflüger, H.-J.
(2003). Central modulatory neurons control fuel selection in flight muscle of
migratory locust. J. Neurosci. 23, 1109–1113.
20. Pflüger, H.-J., and Duch, C. (2011). Dynamic neural control of insect mus-
cle metabolism related to motor behavior. Physiology (Bethesda) 26,
293–303.
21. Pauls, D., Blechschmidt, C., Frantzmann, F., El Jundi, B., and Selcho, M.
(2018). A comprehensive anatomical map of the peripheral octopaminer-
gic/tyraminergic system of Drosophila melanogaster. Sci. Rep. 8, 15314.
22. Selcho, M., and Pauls, D. (2019). Linking physiological processes and
feeding behaviors by octopamine. Curr. Opin. Insect Sci. 36, 125–130.
Current Biology 31, 4076–4087, September 27, 2021 4085
 ll
23. Lorenz, M.W., and Ga €de, G. (2009). Hormonal regulation of energy meta-
bolism in insects as a driving force for performance. Integr. Comp. Biol. 49,
380–392.
24. Kodrı́k, D., Socha, R., and Syrová, Z. (2003). Developmental and diel
changes of adipokinetic hormone in CNS and hemolymph of the flightless
wing-polymorphic bug, Pyrrhocoris apterus (L.). J. Insect Physiol. 49,
53–61.
25. Maxová, A., Kodrı́k, D., Zemek, R., and Socha, R. (2001). Diel changes in
adipokinetic response and walking activity of Pyrrhocoris apterus
(Heteroptera) in relation to physiological status and wing dimorphism.
Eur. J. Entomol. 98, 433–438.
26. Kodrı́k, D., Socha, R., Simek, P., Zemek, R., and Goldsworthy, G.J. (2000).
A new member of the AKH/RPCH family that stimulates locomotory activ-
ity in the firebug, Pyrrhocoris apterus (Heteroptera). Insect Biochem. Mol.
Biol. 30, 489–498.
27. Kodrı́k, D., Simek, P., Lepsa, L., and Socha, R. (2002). Identification of the
cockroach neuropeptide Pea-CAH-II as a second adipokinetic hormone in
the firebug Pyrrhocoris apterus. Peptides 23, 585–587.
28. Fassold, K., El-Damanhouri, H.I.H., and Lorenz, M.W. (2010). Age-depen-
dent cyclic locomotor activity in the cricket, Gryllus bimaculatus, and the
effect of adipokinetic hormone on locomotion and excitability. J. Comp.
Physiol. A Neuroethol. Sens. Neural Behav. Physiol. 196, 271–283.
29. Wicher, D., Agricola, H.-J., Söhler, S., Gundel, M., Heinemann, S.H.,
Wollweber, L., Stengl, M., and Derst, C. (2006b). Differential receptor acti-
vation by cockroach adipokinetic hormones produces differential effects
on ion currents, neuronal activity, and locomotion. J. Neurophysiol. 95,
2314–2325.
30. Wicher, D., Berlau, J., Walther, C., and Borst, A. (2006a). Peptidergic
counter-regulation of Ca(2+)- and Na(+)-dependent K(+) currents modu-
lates the shape of action potentials in neurosecretory insect neurons.
J. Neurophysiol. 95, 311–322.
31. Katewa, S.D., Demontis, F., Kolipinski, M., Hubbard, A., Gill, M.S.,
Perrimon, N., Melov, S., and Kapahi, P. (2012). Intramyocellular fatty-
acid metabolism plays a critical role in mediating responses to dietary re-
striction in Drosophila melanogaster. Cell Metab. 16, 97–103.
32. Kornbluth, S., and White, K. (2005). Apoptosis in Drosophila: neither fish
nor fowl (nor man, nor worm). J. Cell Sci. 118, 1779–1787.
33. Wing, J.P., Karres, J.S., Ogdahl, J.L., Zhou, L., Schwartz, L.M., and
Nambu, J.R. (2002). Drosophila sickle is a novel grim-reaper cell death
activator. Curr. Biol. 12, 131–135.
34. Cole, S.H., Carney, G.E., McClung, C.A., Willard, S.S., Taylor, B.J., and
Hirsh, J. (2005). Two functional but noncomplementing Drosophila tyro-
sine decarboxylase genes: distinct roles for neural tyramine and octop-
amine in female fertility. J. Biol. Chem. 280, 14948–14955.
35. Pech, U., Pooryasin, A., Birman, S., and Fiala, A. (2013). Localization of the
contacts between Kenyon cells and aminergic neurons in the Drosophila
melanogaster brain using SplitGFP reconstitution. J. Comp. Neurol. 521,
3992–4026.
36. Busch, S., Selcho, M., Ito, K., and Tanimoto, H. (2009). A map of octopa-
minergic neurons in the Drosophila brain. J. Comp. Neurol. 513, 643–667.
37. Sinakevitch, I., and Strausfeld, N.J. (2006). Comparison of octopamine-
like immunoreactivity in the brains of the fruit fly and blow fly. J. Comp.
Neurol. 494, 460–475.
38. Monastirioti, M., Linn, C.E., Jr., and White, K. (1996). Characterization of
Drosophila tyramine b-hydroxylase gene and isolation of mutant flies lack-
ing octopamine. J. Neurosci. 16, 3900–3911.
39. Pauls, D., von Essen, A., Lyutova, R., van Giesen, L., Rosner, R., Wegener,
C., and Sprecher, S.G. (2015). Potency of transgenic effectors for neuro-
genetic manipulation in Drosophila larvae. Genetics 199, 25–37.
40. Nitabach, M.N., Blau, J., and Holmes, T.C. (2002). Electrical silencing of
Drosophila pacemaker neurons stops the free-running circadian clock.
Cell 109, 485–495.
4086 Current Biology 31, 4076–4087, September 27, 2021
Article
41. Kitamoto, T. (2001). Conditional modification of behavior in Drosophila by
targeted expression of a temperature-sensitive shibire allele in defined
neurons. J. Neurobiol. 47, 81–92.
42. Nitabach, M.N., Wu, Y., Sheeba, V., Lemon, W.C., Strumbos, J., Zelensky,
P.K., White, B.H., and Holmes, T.C. (2006). Electrical hyperexcitation of
lateral ventral pacemaker neurons desynchronizes downstream circadian
oscillators in the fly circadian circuit and induces multiple behavioral pe-
riods. J. Neurosci. 26, 479–489.
43. Dawydow, A., Gueta, R., Ljaschenko, D., Ullrich, S., Hermann, M.,
Ehmann, N., Gao, S., Fiala, A., Langenhan, T., Nagel, G., and Kittel, R.J.
(2014). Channelrhodopsin-2-XXL, a powerful optogenetic tool for low-light
applications. Proc. Natl. Acad. Sci. USA 111, 13972–13977.
44. Crocker, A., and Sehgal, A. (2008). Octopamine regulates sleep in
Drosophila through protein kinase A-dependent mechanisms.
J. Neurosci. 28, 9377–9385.
45. Li, Y., Hoffmann, J., Li, Y., Stephano, F., Bruchhaus, I., Fink, C., and
Roeder, T. (2016). Octopamine controls starvation resistance, life span
and metabolic traits in Drosophila. Sci. Rep. 6, 35359.
46. Erion, R., DiAngelo, J.R., Crocker, A., and Sehgal, A. (2012). Interaction
between sleep and metabolism in Drosophila with altered octopamine
signaling. J. Biol. Chem. 287, 32406–32414.
47. Ali, Y.O., Escala, W., Ruan, K., and Zhai, R.G. (2011). Assaying locomotor,
learning, and memory deficits in Drosophila models of neurodegeneration.
J. Vis. Exp. (49), 2504.
48. Clyne, J.D., and Miesenböck, G. (2008). Sex-specific control and tuning of
the pattern generator for courtship song in Drosophila. Cell 133, 354–363.
49. Selcho, M., Pauls, D., El Jundi, B., Stocker, R.F., and Thum, A.S. (2012).
The role of octopamine and tyramine in Drosophila larval locomotion.
J. Comp. Neurol. 520, 3764–3785.
50. Schneider, A., Ruppert, M., Hendrich, O., Giang, T., Ogueta, M., Hampel,
S., Vollbach, M., Büschges, A., and Scholz, H. (2012). Neuronal basis of
innate olfactory attraction to ethanol in Drosophila. PLoS ONE 7, e52007.
51. Nikolaev, V.O., Bünemann, M., Hein, L., Hannawacker, A., and Lohse, M.J.
(2004). Novel single chain cAMP sensors for receptor-induced signal
propagation. J. Biol. Chem. 279, 37215–37218.
52. Chen, T.-W., Wardill, T.J., Sun, Y., Pulver, S.R., Renninger, S.L., Baohan,
A., Schreiter, E.R., Kerr, R.A., Orger, M.B., Jayaraman, V., et al. (2013).
Ultrasensitive fluorescent proteins for imaging neuronal activity. Nature
499, 295–300.
53. Feng, Y., Ueda, A., and Wu, C.-F. (2004). A modified minimal hemolymph-
like solution, HL3.1, for physiological recordings at the neuromuscular
junctions of normal and mutant Drosophila larvae. J. Neurogenet. 18,
377–402.
54. Han, K.A., Millar, N.S., and Davis, R.L. (1998). A novel octopamine recep-
tor with preferential expression in Drosophila mushroom bodies.
J. Neurosci. 18, 3650–3658.
55. Balfanz, S., Strünker, T., Frings, S., and Baumann, A. (2005). A family of oc-
topamine [corrected] receptors that specifically induce cyclic AMP pro-
duction or Ca2+ release in Drosophila melanogaster. J. Neurochem. 93,
440–451.
56. Burke, C.J., Huetteroth, W., Owald, D., Perisse, E., Krashes, M.J., Das, G.,
Gohl, D., Silies, M., Certel, S., and Waddell, S. (2012). Layered reward sig-
nalling through octopamine and dopamine in Drosophila. Nature 492,
433–437.
57. Lee, H.-G., Seong, C.-S., Kim, Y.-C., Davis, R.L., and Han, K.-A. (2003).
Octopamine receptor OAMB is required for ovulation in Drosophila mela-
nogaster. Dev. Biol. 264, 179–190.
58. Pannabecker, T., and Orchard, I. (1986). Octopamine and cyclic AMP
mediate release of adipokinetic hormone I and II from isolated locust
neuroendocrine tissue. Mol. Cell. Endocrinol. 48, 153–159.
59. Passier, P.C., Vullings, H.G., Diederen, J.H., and Van der Horst, D.J.
(1995). Modulatory effects of biogenic amines on adipokinetic hormone
secretion from locust corpora cardiaca in vitro. Gen. Comp. Endocrinol.
97, 231–238.

Article
60. Seay, D.J., and Thummel, C.S. (2011). The circadian clock, light, and cryp-
tochrome regulate feeding and metabolism in Drosophila. J. Biol. Rhythms
26, 497–506.
61. Masuyama, K., Zhang, Y., Rao, Y., and Wang, J.W. (2012). Mapping neural
circuits with activity-dependent nuclear import of a transcription factor.
J. Neurogenet. 26, 89–102.
62. Itskov, P.M., Moreira, J.-M., Vinnik, E., Lopes, G., Safarik, S., Dickinson,
M.H., and Ribeiro, C. (2014). Automated monitoring and quantitative anal-
ysis of feeding behaviour in Drosophila. Nat. Commun. 5, 4560.
ly, A.A., Daan, S., Wirz-Justice, A., and Deboer, T. (2016). The two-
63. Borbe
process model of sleep regulation: a reappraisal. J. Sleep Res. 25,
131–143.

64. Sajwan, S., Sidorov, R., Stasková, T., Zaloudı́ková, A., Takasu, Y., Kodrı́k,
D., and Zurovec, M. (2015). Targeted mutagenesis and functional analysis
of adipokinetic hormone-encoding gene in Drosophila. Insect Biochem.
Mol. Biol. 61, 79–86.
65. Bednárová, A., Tomc
ala, A., Mochanová, M., Kodrı́k, D., and Krishnan, N.
(2018). Disruption of Adipokinetic Hormone Mediated Energy
Homeostasis Has Subtle Effects on Physiology, Behavior and Lipid
Status During Aging in Drosophila. Front. Physiol. 9, 949.
66. Van der Horst, D.J. (2003). Insect adipokinetic hormones: release and inte-
gration of flight energy metabolism. Comp. Biochem. Physiol. B Biochem.
Mol. Biol. 136, 217–226.
67. Perry, R.J., Saunders, C.J., Nelson, J.M., Rizzo, M.J., Braco, J.T., and
Johnson, E.C. (2020). Regulation of Metabolism by an Ensemble of
Different Ion Channel Types: Excitation-Secretion Coupling Mechanisms
of Adipokinetic Hormone Producing Cells in Drosophila. Front. Physiol.
11, 580618.
68. Xu, K., Zheng, X., and Sehgal, A. (2008). Regulation of feeding and meta-
bolism by neuronal and peripheral clocks in Drosophila. Cell Metab. 8,
289–300.
69. Ro, J., Harvanek, Z.M., and Pletcher, S.D. (2014). FLIC: high-throughput,
continuous analysis of feeding behaviors in Drosophila. PLoS ONE 9,
e101107.
70. Gáliková, M., Klepsatel, P., Xu, Y., and Kühnlein, P. (2017). The obesity-
related Adipokinetic hormone controls feeding and expression of neuro-
peptide regulators of Drosophila metabolism. Eur. J. Lipid Sci. Technol.
119, 1600138.
71. He, Q., Du, J., Wei, L., and Zhao, Z. (2020). AKH-FOXO pathway regulates
starvation-induced sleep loss through remodeling of the small ventral
lateral neuron dorsal projections. PLoS Genet. 16, e1009181.
72. Koyama, T., Texada, M.J., Halberg, K.A., and Rewitz, K. (2020).
Metabolism and growth adaptation to environmental conditions in
Drosophila. Cell. Mol. Life Sci. 77, 4523–4551.
73. Ahmad, M., He, L., and Perrimon, N. (2020). Regulation of insulin and adi-
pokinetic hormone/glucagon production in flies. Wiley Interdiscip. Rev.
Dev. Biol. 9, e360.
74. Braco, J.T., Saunders, C.J., Nelson, J.M., and Johnson, E.C. (2020).
Modulation of metabolic hormone signaling via a circadian hormone and
ll
a biogenic amine in Drosophila melanogaster. bioRxiv. https://doi.org/
10.1101/2020.09.25.312967.
75. Kim, J., and Neufeld, T.P. (2015). Dietary sugar promotes systemic TOR
activation in Drosophila through AKH-dependent selective secretion of
Dilp3. Nat. Commun. 6, 6846.
76. Regalado, J.M., Cortez, M.B., Grubbs, J., Link, J.A., van der Linden, A.,
and Zhang, Y. (2017). Increased food intake after starvation enhances
sleep in Drosophila melanogaster. J. Genet. Genomics 44, 319–326.
77. Linford, N.J., Chan, T.P., and Pletcher, S.D. (2012). Re-patterning sleep ar-
chitecture in Drosophila through gustatory perception and nutritional qual-
ity. PLoS Genet. 8, e1002668.
78. Catterson, J.H., Knowles-Barley, S., James, K., Heck, M.M.S., Harmar,
A.J., and Hartley, P.S. (2010). Dietary modulation of Drosophila sleep-
wake behaviour. PLoS ONE 5, e12062.
79. Zhao, X., and Karpac, J. (2017). Muscle Directs Diurnal Energy
Homeostasis through a Myokine-Dependent Hormone Module in
Drosophila. Curr. Biol. 27, 1941–1955.e6.
80. Crocker, A., Shahidullah, M., Levitan, I.B., and Sehgal, A. (2010).
Identification of a neural circuit that underlies the effects of octopamine
on sleep:wake behavior. Neuron 65, 670–681.
81. Milde, J.J., Ziegler, R., and Wallstein, M. (1995). Adipokinetic hormone
stimulates neurones in the insect central nervous system. J. Exp. Biol.
198, 1307–1311.
82. Hughson, B.N., Shimell, M., and O’Connor, M.B. (2021). AKH signaling in
D. melanogaster alters larval development in a nutrient-dependent manner
that influences adult metabolism. Front. Physiol. 12, 619219.
83. Koyama, T., Terhzaz, S., Naseem, M.T., Nagy, S., Rewitz, K., Dow, J.A.T.,
Davies, S.A., and Halberg, K.A. (2020). A nutrient-responsive hormonal cir-
cuit controls energy and water homeostasis in Drosophila. bioRxiv. https://
doi.org/10.1101/2020.07.24.219592.
84. Hofbauer, A., Ebel, T., Waltenspiel, B., Oswald, P., Chen, Y.C., Halder, P.,
Biskup, S., Lewandrowski, U., Winkler, C., Sickmann, A., et al. (2009). The
Wuerzburg hybridoma library against Drosophila brain. J. Neurogenet. 23,
78–91.
85. Schindelin, J., Arganda-Carreras, I., Frise, E., Kaynig, V., Longair, M.,
Pietzsch, T., Preibisch, S., Rueden, C., Saalfeld, S., Schmid, B., et al.
(2012). Fiji: an open-source platform for biological-image analysis. Nat.
Methods 9, 676–682.
86. Selcho, M., and Wegener, C. (2015). Immunofluorescence and Genetic
Fluorescent Labeling Techniques in the Drosophila Nervous System.
Neuromethods 101, 39–62.
87. Wegener, C., Neupert, S., and Predel, R. (2010). Direct MALDI-TOF mass
spectrometric peptide profiling of neuroendocrine tissue of Drosophila.
Methods Mol. Biol. 615, 117–127.
88. Scha€bler, S., Amatobi, K.M., Horn, M., Rieger, D., Helfrich-Förster, C.,
Mueller, M.J., Wegener, C., and Fekete, A. (2020). Loss of function in
the Drosophila clock gene period results in altered intermediary lipid meta-
bolism and increased susceptibility to starvation. Cell. Mol. Life Sci. 77,
4939–4956.
Current Biology 31, 4076–4087, September 27, 2021 4087
 ll
Article

STAR+METHODS

KEY RESOURCES TABLE

REAGENT or RESOURCE SOURCE IDENTIFIER

Antibodies
Anti-GFP Invitrogen RRID: AB_221570
84
Anti-Synapsin RRID: AB_2313617
Goat anti-rabbit Alexa 488 Invitrogen RRID: AB_2536097
Goat anti-mouse DyLight 649 Jackson ImmunoResearch RRID: AB_2339069
Anti-Tdc2 Covalab, pab0822-p N/A
Chemicals, peptides, and recombinant proteins
Paraformaldehyde Merck Cat#104005
PBS Sigma-Aldrich Cat#P4417
Triton Sigma-Aldrich Cat#T8787
Gylcerol Sigma-Aldrich Cat#G9012
Agarose Carl Roth Cat#3810.2
Agarose low EEO AppliChem Cat#A2114
Sucrose Sigma-Aldrich Cat#S9378
Isopropanol Sigma-Aldrich Cat#563935
1,2-didecanoyl-glycerol Avanti Polar Lipids Cat#800810
Glucose-6,6-d2 Sigma-Aldrich Cat#282650
Trehalose-1,1-d2 Sigma-Aldrich Cat#700452
JumpStart REDTaq ReadyMix Reaction Mix Sigma-Aldrich Cat#P0982
DMSO Sigma-Aldrich Cat#276855
Octopamine Sigma-Aldrich Cat#O0250
AKH NovoPro Bioscience N/A
Carbamoylcholine chloride Sigma-Aldrich Cat#C4382
Normal goat serum Sigma-Aldrich Cat#G9023
Yeast extract AppliChem, Darmstadt, Germany Cat#A1552
Critical commercial assays
Drosophila Activity Monitors (DAM) Trikinetics, Waltham, USA https://trikinetics.com
flyPAD EasyBehavior, Estoril, Portugal https://www.flypad.rocks
Experimental models: Organisms/strains
33,34,39
UAS-hid,rpr N/A
ts ts 41
UAS-shibire (shi ) N/A
UAS-DORK Bloomington Drosophila Stock Center RRID: BDSC_6586
UAS-OambII-RNAi Bloomington Drosophila Stock Center RRID: BDSC_31171
UAS-NaChBac Bloomington Drosophila Stock Center RRID: BDSC_9467
13
UAS-AkhR-RNAi N/A
UAS-Epac1-camps50A Bloomington Drosophila Stock Center RRID: BDSC_25407
UAS-GCaMP6m Bloomington Drosophila Stock Center RRID: BDSC_42748
UAS-10xmyr::GFP Bloomington Drosophila Stock Center RRID: BDSC_32197
UAS-amonRNAi78b Bloomington Drosophila Stock Center RRID: BDSC_29009
UAS-CaLexA Bloomington Drosophila Stock Center RRID:BDSC_66545
AkhR1 13
N/A
AkhRrevA 13
N/A
13
UAS-AkhR N/A
AkhR1; UAS-AkhR 13
N/A
(Continued on next page)

e1 Current Biology 31, 4076–4087.e1–e5, September 27, 2021

 ll
Article

Continued
REAGENT or RESOURCE SOURCE IDENTIFIER
Akh-Gal4 Bloomington Drosophila Stock Center RRID: BDSC_25683
Tdc2-Gal4 Bloomington Drosophila Stock Center RRID: BDSC_9313
Tdc2-Gal4 Hirsh lab N/A
49
Tdc2-Gal4; tshGal80 N/A
1
AkhR ; Tdc2-Gal4 This paper N/A
NP5267-Gal4 Kyoto Stock Center RRID: DGRC_104929
Oligonucleotides
Primer: OAMB E4E5 fw CGTTCTG This paper N/A
GTCATCGGTGG
Primer: OAMB E4E5 rv CGTGGACATTATGCTTGG This paper N/A
Primer: OAMB E6E7 / E6E8 fw CACCG This paper N/A
CCGCAGAACAGAC
Primer: OAMB E6E7 rv CGCTACTGATTACCGCAT This paper N/A
Primer: OAMB E6E8 rv CGCTCGT This paper N/A
GACCCACATCC
Primer: AKHR fw GGACTCTACAACATTCGC This paper N/A
Primer: AKHR rv CCTCCTCCATTCAGCAGC This paper N/A
Primer: a-tubulin fw TCTGCGATT This paper N/A
CGATGGTGCCCTTAAC
Primer: a-tubulin rv GGATCGCACT This paper N/A
TGACCATCTGGTTGGC
Software and algorithms
R: A language and environment for statistical R Foundation for Statistical Computing, https://www.r-project.org
computing Vienna, Austria
GraphPad Prism 8 GraphPad Software, La Jolia, CA, USA https://www.graphpad.com
85
Fiji https://fiji.sc
OriginPro 2016G OriginLab Corporation, Northampton, MA, USA https://www.originlab.com/origin
Adobe Photoshop CS6 Adobe Systems, San Jose, CA, USA https://www.adobe.com/de/products/
photoshop.html
62
MATLAB ; The Mathworks, Natick, MA, USA https://www.mathworks.com/products/
matlab.html

RESOURCE AVAILABILITY

Lead contact
Further information and requests for resources and reagents should be directed to and will be fulfilled by the lead contact, Christian
Wegener (christian.wegener@biozentrum.uni-wuerzburg.de)

Materials availability
All Drosophila lines used in this study are available on request from the lead Contact, or from either Bloomington Stock Centre or the
Vienna Drosophila Resource Centre.

Data and code availability

All data reported in the present study have not been deposited in a public repository but are available from the lead contact on request
without restriction.
This paper does not report original code.
Any additional information required to reanalyze the data reported in this paper is available from the lead contact upon request.

EXPERIMENTAL MODEL AND SUBJECT DETAILS

Fly strains
Flies were cultured according to standard methods. Vials were kept under constant conditions at 25 C and 60% humidity in a 12:12
light:dark cycle. Genetic controls were obtained by crossing Gal4-driver and UAS-effector lines to w1118. Transgenic driver lines used

Current Biology 31, 4076–4087.e1–e5, September 27, 2021 e2

 ll
Article

in this study were Akh-Gal4, Tdc2-Gal4, Tdc2-Gal4;tshGAL80, and NP5267-Gal4. Transgenic UAS-effector lines in this study were
UAS-hid,rpr, UAS-shibirets (UAS-shits), UAS-DORK, UAS-NaChBac, UAS-OambII-RNAi, UAS-AkhR-RNAi, UAS-Epac1-camps50A,
UAS-GCaMP6m, 10xUAS-myr::GFP, UAS-amonRNAi78b, UAS-CaLexA. AkhR1 and AkhRrevA mutant lines were used.

METHOD DETAILS

Behavioral assays
Locomotor activity recording
3-7d old flies were recorded individually in Drosophila activity monitors (DAM, Trikinetics, Waltham, USA) under standard 12:12 light:
dark conditions with light intensities around 60-100lux at 20 C (unless otherwise specified). Flies were kept in small glass tubes with
2% agarose and 4% sugar on one side. Under starved conditions, flies were kept on 2% agarose to avoid dehydration. 32 flies of a
genotype (half male and female) and appropriate controls were measured simultaneously. Fly activity was defined by the number of
infrared light beam crosses within 12 hours. Activity patterns were separated into day (ZT0-ZT12; ZT0 = lights-on, ZT12 = lights-off)
and night (ZT12-ZT24), and data was pooled from six days and nights, respectively. For average activity profiles, fly activity was
defined as beam crosses per hour. Within these average activity profiles, asterisks label those time points at which experimental flies
differ significantly from both genetic controls. For thermogenetic inhibition or activation via UAS-shibirets experiments were per-
formed at 31 C under the same light regime.
Survival
Survival analysis was performed with the same DAM system raw data as above. The number of flies alive at the end of day was
counted based on present or absent beam crossings.
Startle-induced negative geotaxis (SING)
Negative geotaxis was performed to screen for general motor impairments. The experiments were performed as described in Ali
et al., 2011, with minor modifications: Groups of 10 male and female flies, respectively, were separated into Falcon tubes 5 minutes
before the experiment. Flies were tapped to the conic bottom of the Falcon tube and then recorded with a DMK22BUC09 video cam-
era with a Pentax C2514-M objective (The Imaging Source Europe GmbH, Germany). Data was averaged over 5 trials per group (with
an interspace period of 1 minute) and represents the percentage of flies reaching a label 8.5 cm above the bottom within 5 and 10 s.
flyPAD feeding assay
Feeding was automatically monitored using a 2-channel flyPAD system (https://www.flypad.rocks) adapted for long-term measure-
ment of food intake. The flyPAD relies on capacitative measurements.62 For each experimental run, four males and four females of
each genotype were in parallel anesthetised on ice, and then transferred to individual arenas in the late afternoon prior to the evening
activity phase. In the arenas, flies had the choice between 2% agarose (Carl Roth, Karlsruhe, Germany) containing 5 mM sucrose
(Sigma, ‘‘sucrose’’) or 10% yeast extract (AppliChem, Darmstadt, Germany, ‘‘yeast’’). Food was provided in glass tubes with an
enlarged diameter of 7mm to allow for long-term monitoring. The flyPADs were turned upside-down and placed on evenly-illuminated
white LED plates under control of a timer to generate a light regime of LD12:12 in a dark box at 21 C. Feeding (number of sips) was
analyzed in 30 min bins under exclusion of flies that did not feed during the entire experiment. The data was analyzed using custom
made MATLAB scripts.62

Live cell imaging

Imaging setup
For Ca2+ imaging, specimens were imaged with an AXIO Examiner D1 upright microscope (Carl Zeiss, Göttingen, Germany) with a
Zeiss W Plan-Apochromat 40x 1.0 or 20x 0.8 water immersion objective and a SPECTRA 4 hybrid solid state LED source (Lumencor,
Beaverton, USA). Images were captured with a PCO.edge 4.2 m sCMOS camera (PCO AG, Kelheim, Germany). For cAMP imaging,
we used an Axioskop 2FS plus microscope (Carl Zeiss, Göttingen, Germany) with a Visichrom high-speed polychromator system and
a Xenon arc lamp (75W, Visitron Systems, Puchheim, Germany). Images were recorded with a CCD camera (CoolSnap HQ, Roper
Scientific, Tuscon, USA) and a Photometrics DualView2 beam splitter allowing to separately record CFP and YFP emissions. Images
were analyzed using VisiView3.0 software (Visitron Systems, Puchheim, Germany).
Bath application experiments
Intracellular Ca2+ and cAMP levels were monitored by UAS-Epac1-camps50A51 and UAS-GCaMP6m.52 1-2 brains were dissected,
mounted in 405ml hemolymph-like HL3.1 saline solution53 in the imaging chamber and allowed to settle for 30-45 minutes. For Ca2+
and cAMP imaging in OANs, AKH (+ 0.1% DMSO) or HL3.1 (+0.1% DMSO; as vehicle control), were bath-applied after 100 s. Appli-
cation of carbamylcholine (carbachol; for Ca2+ imaging; Figure S6) and the water-soluble forskolin derivate NKH477 served as pos-
itive control (data not shown).
Prior to Ca2+ imaging in AKH cells, Ca2+ levels were reduced to background levels by applying trehalose (80mM) after 100 s.5
80mM trehalose was added to all subsequent solutions applied during the experiments. At 300 s, either HL3.1 (+0.1% DMSO) as
vehicle control or 1mM OA dissolved in HL3.1 (+0.1% DMSO) was bath-applied. Application of potassium chloride (end concentra-
tion: 80 mM) after 2980 s served as viability control.
Changes in the GCaMP fluorescence intensity within the region of interest were calculated as: DF/F0 = (Fn-F0)/F0, where Fn is the
fluorescence at time point n and F0 the baseline fluorescence intensity calculated from the first 30 s. For Epac1 imaging, the CFP/YFP
ratio was calculated for each time point: D(CFP/YFP) with CFP = CFPn/CFP0 and YFP = (YFPn-(CFPn*spillover into YFP)/YFP0, where

e3 Current Biology 31, 4076–4087.e1–e5, September 27, 2021

 ll
Article

CFPn or YFPn is the fluorescence intensity at a given time point, and CFP0 or YFP0 is the baseline fluorescence intensity calculated
from the first 30 s. CFP spillover was determined as 35.7% of the YFP signal.
Region of interests (ROI) were drawn over the cell bodies and the background was subtracted for every ROI. All imaging experi-
ments were performed during the light phase.

Immunostainings
Whole mounts
Immunofluorescence stainings of Tdc2 > 10xmyr::GFP and Tdc2,tshGal80 > 10xmyr::GFP preparations were performed as
described.86 In short, ventral nerve cords of 3-7 days old flies were dissected in HL3.1 (pH 7.2-7.4). Afterward, specimens were
fixated in 4% paraformaldehyde in PBS (0.1M; phosphate buffer saline) for two hours, washed four times in PBS with 0.3% Triton
X-100 (PBT), and blocked with 5% normal goat serum in PBT. Then, specimens were incubated with rabbit anti-GFP serum
(A6455, Invitrogen, USA; 1:1000) and mouse monoclonal antibodies against Synapsin (3C11;84; 1:50) in blocking solution for one
night at 4 C. On day two, preparations were washed six times with PBT and incubated for one night at 4 C with secondary antibodies
goat anti-rabbit Alexa 488 (Invitrogen, USA; 1:250) and goat anti-mouse DyLight 649 (Jackson ImmunoResearch, USA; 1:250).
Finally, specimens were rinsed six times in PBT and mounted in 80% glycerol in PBS. Until scanning with a Leica TCS SPE or
SP8 confocal microscope (Leica microsystems, Wetzlar, Germany), specimens were stored in darkness at 4 C.
Tissue sections
To visualize the arborisations of OANs on the CCA-HG complex, whole body sections were performed (as described in Pauls et al.21).
In short, whole flies with opened cuticle were fixated with 4% paraformaldehyde in PBS for two hours and afterward washed three
times with PBS. Subsequently flies were embedded in hot 7% low melting agarose. After hardening, specimens were cut with a vi-
bratome into 90-95 mm sections. Staining of the sections was continued as described in 6.4.1 in a 24 well plate. Rabbit anti-Tdc2
(pab0822-p, Covalab, France; 1:200) and mouse anti-Synapsin (3C11; 1:50) were used as primary antibodies visualized by goat
anti-rabbit Alexa488 (1:200) and goat anti-mouse DyLight649 (1:200) sera.

CaLexA labeling
For CaLexA labeling, equal numbers of 4-11 day old males and female flies were used. The proventriculus plus attached retrocerebral
complex (corpora allata and fused corpora cardiaca-hypocerebral ganglion CCA-HG complex) were dissected in HL3.1 saline as
previously described87 and immediately imaged with an AXIO Examiner D1 upright microscope (Carl Zeiss, Göttingen, Germany)
with a Zeiss W Plan-Apochromat 40x 1.0 water immersion objective, and a SPECTRA 4 hybrid solid state LED source with
475 nm excitation maximum (Lumencor, Beaverton, USA). Images were captured with a PCO.edge 4.2 m sCMOS camera (PCO
AG, Kelheim, Germany) at 16bit intensity resolution. All settings were kept constant throughout the experiments. Images were
analyzed using Fiji.85 A region of interest was drawn around the CCA-HG and in the direct vicinity (background). The mean intensity
of the CCA-HG was then calculated and the mean background was subtracted. In addition, the number of labeled cells was counted.

PCR
Total RNA was extracted from adult Canton-S flies using the Quick-RNA MicroPrep Kit (Zymo Research, Irvine, USA) according to
manufacturer’s instructions. Central brains, VNCs, and HG-CCA (for OAMB 15 flies per group; for AKHR five flies per group) were
dissected in HL3.1 medium, and transferred to 300 ml RNA lysis buffer on ice. Tissues were homogenized with a plastic pestle. Total
RNA was eluted in 8 ml RNase-free water. For cDNA synthesis, the QuantiTect Reverse Transcription Kit from QIAGEN (Hilden, Ger-
many) was used. All steps were performed following the manufacturer’s protocol. Genomic DNA was removed by adding 1 ml of
gDNA wipe-out buffer to 6 ml of the eluted RNA. Following incubation at 42 C for 2min, the samples were placed for 2min at 4 C
and 3 ml of a mastermix composed of 2 ml RT buffer, 0.5 ml RT primer Mix and 0.5 mL reverse transcriptase was added. Reverse tran-
scription was performed for 30 min at 42 C, followed by 3 min at 95 C and 2min at 4 C. Finally, 20 ml water were added; cDNA sam-
ples were stored at 20 C. To analyze OAMB and AKHR expression in the isolated tissues, cDNAs were PCR-amplified using spe-
cific primers and a JumpStart REDTaq ReadyMix Reaction Mix (Sigma-Aldrich, MO, USA). a-tubulin and water (‘‘no DNA’’) were used
as internal control. The PCR program consisted of 5min at 95 C, followed by 35 cycles of 30 s at 95 C, 30 s at 61 C (57,5 C for AKHR),
60 s at 72 C and finally 5 min at 72 C.

Ultra performance liquid chromatography-tandem mass spectrometry (UPLC-MS/MS) measurements

Haemolymph sampling
For each sample, 20 flies were perforated by a thin needle at the thorax and put in a perforated 0.5ml Eppendorf cap, which was fitted
into a 1.5ml Eppendorf cap. Haemolymph was collected within the 1.5ml Eppendorf cap after 1min of centrifugation in a benchtop
centrifuge at 5000 rpm, and immediately deep-frozen until analysis.
UPLC-MS/MS analysis of sugars and diacylglycerols
Sugars were analyzed using a Waters Acquity ultra-high-performance liquid chromatograph coupled to a Waters Micromass Quattro
Premier triple quadrupole mass spectrometer (Milford, MA, USA) with an electrospray interface. Chromatographic separation was
performed on a Waters Acquity BEH amide column (1.7 mm, 2.1 3 100mm; Waters, Milford, MA, USA) according to Waters applica-
tion note WA60126 with a modified flow rate of 0.2 ml/min. The analytes were eluted in a linear solvent-strength gradient (75% to 45%
acetonitrile containing 0.1% ammonium hydroxide in 10min) at a column temperature of 35 C. Sugars were detected in the negative

Current Biology 31, 4076–4087.e1–e5, September 27, 2021 e4

 ll
Article

electrospray mode at a source temperature of 120 C and a capillary voltage of 3.25 kV. Nitrogen was used as desolvation and cone
gas with flow rates of 800 l h–1 at 350 C and 25 l h–1. The mass spectrometer was operated in multiple reaction monitoring (MRM)
mode using Argon as collision gas at a pressure of approximately 3 3 10 3 bar. Cone voltage (CV) and collision energy (CE) were
optimized for maximum signal intensity of each individual compound during ionization and collision induced dissociation (CID) for
a dwell time of 50ms per transition. Glucose-6,6-d2 and Trehalose-1,1¢-d2 were used as internal standards for quantification of
mono- and disaccharides.
The following cone voltages and collision energies were used for each transition:

Glucose/Fructose: parent 179 m/z, daughter 89 m/z, CV 18V, CE 10eV,

Glucose-6,6-d2: parent 179 m/z, daughter 89 m/z, CV 18V, CE 10eV,
Sucrose/Trehalose: parent 341 m/z, daughter 179 m/z, CV 32V, CE 15eV,
Trehalose-1,1’-d2: parent 343 m/z, daughter 180 m/z, CV 32V, CE 15eV,

€bler et al.88 Prior to analysis, hemolymph samples were dried

Determination of diacylglycerols were performed according to Scha
and recovered in 75 ml isopropanol containing 1 mg/ml 1,2-didecanoyl-glycerol used as an internal standard.

QUANTIFICATION AND STATISTICAL ANALYSIS

Statistical Analysis
Data was analyzed for normal distribution using the Shapiro-Wilk test. For statistical comparison between genotypes, a pairwise t test
was used for normally distributed data, a pairwise Wilcoxon rank sum test was used for not normally distributed data (both including a
Bonferroni-Holm correction for pairwise comparisons). All statistical analyses were done with R Studio version 1.0.136. (https://www.
r-project.org). For the statistical comparison of sugar levels quantified by LC-MS, we used a simple t test based on the relatively small
sample size (N = 3-5, with n = 20 per N). Data were plotted using OriginPro 2016G, GraphPad Prism 8, R Studio and Photoshop CS6.
Activity levels are presented as boxplots, with 50% of values for a given genotype being located within the box. Box and whiskers
represent the entire set of data. Outliers are indicated as open circles. The median performance index is indicated as a thick line within
the boxplot. Data presented as barplots show the mean values and standard error of the mean (SEM). Significance levels between
genotypes shown in the figures refer to the raw p values obtained in the statistical tests.

e5 Current Biology 31, 4076–4087.e1–e5, September 27, 2021