needs also activate a dopamine wanting sys-
tem. To this end, we studied whether starva-
tion of individual foragers enhanced dopamine
brain levels. Marked foraging bees of con-
trolled age were captured at an artificial feeder
and harnessed individually [see supplemen-
tary materials (SM) for details]. Bees were
housedinanincubatorfor1or2hoursand
then frozen in liquid nitrogen for quantifying-
dopamine and serotonin brain levels. The inter-
action between the factors“biogenic amine”and
“starvation time”was significant (Fig. 3A;F1,159=
7.25,P<0.01),asonlydopamine(Tukeytests,P<
0.05), not serotonin, levels increased with starva-
tion time. This increase is consistent with the
activation of an individual wanting system.
We reasoned that manipulating dopamine
levels would change the hedonic value of sucrose
solution and thus individual appetitive respon-
siveness. To test this hypothesis, foragers cap-
tured upon landing at a feeder were individually
enclosed within syringes, which allowed them
to be fed with 30ml of 1.0 M sucrose solution
via a tip inserted in the syringe hub ( 21 ). In
this way, we reduced individual appetitive
responsiveness to be able to observe behavioral
changes due to artificial dopamine increase.
After feeding, bees were topically exposed to
dopamine, 6,7-ADTN, or the solvent DMF. A
sham group was also included. Bees were then
individually harnessed and left at rest (see SM
for details). They were then stimulated with six
increasing concentrations of sucrose solution
delivered to the antennae (0.1, 0.3, 1, 3, 10, and
30%, w/w) ( 22 , 23 ). The appetitive proboscis
extension response to these stimuli was quan-
tified, and an individual sucrose responsive-
ness score was calculated on the basis of the
number of sucrose concentrations that elicited
proboscis extension. Appetitive responsiveness
varied between treatments (Fig. 3B; Kruskal-
Wallis test;H5,225= 57.55,P< 0.0001). Whereas
the two control groups (sham and DMF) showed
similarly low responsiveness scores (mean rank
comparison, not significant), the highest dose of
both dopamine and 6,7-ADTN significantly in-
creased sucrose scores (P< 0.0001). We con-
firmed this finding in bees availing themselves of
an even higher energy supplement as they were
individually fed with a mixture of honey, pollen,
and sucrose solution ( 24 – 26 ) before the sucrose-
responsiveness assay. Enhancing dopaminelevels in these bees rescued their appetitive
responsiveness (fig. S2;H3,411= 130.08,P<
0.0001), as their sucrose scores were sig-
nificantly higher than those of the fed controls
(P< 0.05).
We then tested whether manipulating dopa-
mine would affect learning and retrieval of
odorants predicting food. We took advantage
of the olfactory conditioning of the proboscis
extension response in which harnessed bees
learn to associate a neutral odorant with a
reward of sucrose solution ( 27 , 28 ). Foragers
captured upon landing at a feeder were handled
as in the previous experiments and topically
exposed to DMF, the higher dose of dopamine,
or the dopamine receptor antagonist flupen-
tixol ( 13 , 29 ). An additional control was left
untreated (sham). Each group was trained to
discriminate 1-nonanol from 1-hexanol ( 30 )
(see SM for details).
All four groups learned the olfactory discrim-
ination (fig. S3). To compare group performance,
we calculated for each bee an acquisition score
over the four last trials, which could vary be-
tween 2 and−2, with a higher positive score
indicating successful learning. Scores varied510 29 APRIL 2022•VOL 376 ISSUE 6592 science.orgSCIENCE
Fig. 2. Pharmacological blockade of dopaminergic signaling in honey bee
foragers decreases the frequency of foraging and increases the time
between successive visits to the feeder.Honey bees received a topical
application on the thorax of either fluphenazine, 6,7-ADTN, or DMF or remained
untreated (sham). The frequency of foraging bouts (A) and the time spent away
from the feeder between two consecutive visits (return time) (B) were quantified
during two 30-min periods, one before and one 20 min after topical exposure
(6,7-ADTN:n= 22; fluphenazine:n= 24; DMF:n= 18; sham:n= 22). Treatment
with fluphenazine significantly decreased the frequency of foraging bouts and
increased the return time, thus revealing a negative impact on foraging
motivation. (CtoE) Pharmacological blockade of dopaminergic signaling in
honey bee foragers increases the time spent in the hive at the expense of
foraging. Neither the time spent flying from the feeder to the hive (C) nor from
the hive to the feeder (D) was affected by the pharmacological treatments. On
the contrary, blockade of DA signaling increased the time spent in the hive at
the expense of the foraging time (E) (6,7-ADTN:n= 22; fluphenazine:n= 20; DMF:
n= 22; sham:n= 13). In all panels, the values shown correspond to the mean ± SEM
**P< 0.001; ns, not significant; Tukey tests.RESEARCH | REPORTS