Octopamine Shifts the Behavioral Response From Indecision to Approach or Aversion in Drosophila melanogaster

Abstract

Animals must make constant decisions whether to respond to external sensory stimuli or not to respond. The activation of positive and/or negative reinforcers might bias the behavioral response towards approach or aversion. To analyze whether the activation of the octopaminergic neurotransmitter system can shift the decision between two identical odor sources, we active in Drosophila melanogaster different sets of octopaminergic neurons using optogenetics and analyze the choice of the flies using a binary odor trap assay. We show that the release of octopamine from a set of neurons and not acetylcholine acts as positive reinforcer for one food odor source resulting in attraction. The activation of a subset of these neurons causes the opposite behavior and results in aversion. This aversion is due to octopamine release and not tyramine, since in Tyramine-β-hydroxylase mutants (Tβh) lacking octopamine, the aversion is suppressed. We show that when given the choice between two different attractive food odor sources the activation of the octopaminergic neurotransmitter system switches the attraction for ethanol-containing food odor to a less attractive food odor. Consistent with the requirement for octopamine in biasing the behavioral outcome, Tβh mutants fail to switch their attraction. The execution of attraction does not require octopamine but rather initiation of the behavior or a switch of the behavioral response. The attraction to ethanol also depends on octopamine. Pharmacological increases in octopamine signaling in Tβh mutants increase ethanol attraction and blocking octopamine receptor function reduces ethanol attraction. Taken together, octopamine in the central brain orchestrates behavioral outcomes by biasing the decision of the animal towards food odors. This finding might uncover a basic principle of how octopamine gates behavioral outcomes in the brain.

Keywords: Tβh; attraction; aversion; decision making; ethanol attraction; food odor; octopamine.

Figure 1

Activation of tyraminergic/octopaminergic/cholinergic neurons results in attraction. (A) Schematic of the site attraction assay (modified after Schneider et al., 2012). Two LEDs with the same light intensity but different wavelengths illuminate two similar odor traps with the same flicker frequency of 2 s at 40 Hz, 16 s at 8 Hz and 2 s at 0 Hz. If not otherwise indicated, this pattern was used for activation. The attraction index (AI) describes the fly number in the blue-illuminated odor trap minus the fly number in the yellow-illuminated odor trap in comparison to the total number of flies. (B) Neuronal activation using the UAS-ChR2 under the control of the decarboxylase 2 (dTdc2)-Gal4 driver elicits site attraction for the blue illuminated odor trap (AIs of the control and experimental group: 0.21 ± 0.14 and 0.6 ± 0.12, respectively; n = 34, 31). Removing the activation of cholinergic neurons by combining the dTdc2-Gal4 driver with the ChAT-Gal80 driver eliminates site attraction (AIs for the control and experimental group: 0.08 ± 0.16 and −0.05 ± 0.17, respectively; n = 25, 23). (C) Activation of the UAS-ChR2 transgene in a dTdc2-Gal4-dependent manner in Tβh^nM18 mutants elicits site aversion (AIs of the control and experimental group: −0.05 ± 0.14 and −0.48 ± 0.15, respectively; n = 15, 12, and AIs at 8 Hz in the control and experimental group: −0.09 ± 0.11 and −0.55 ± 0.08, respectively; n = 20, 19). (D) Removal of UAS-ChR2 transgene activation by expression of ChAT-Gal80 eliminates aversion (AIs of the control and experimental group: −0.07 ± 0.14 and −0.03 ± 0.13, respectively; n = 16, 16). Errors are SEM. The letter “a” indicates differences from random choice as determined by the one-sample sign test. Student’s T-test was used to determine differences between two groups. *P < 0.05 and **P < 0.05. For data, see Supplementary Table S1.

Figure 2

Octopamine release causes aversion. (A) Neuronal activation under the control of the 6.2-Tβh-Gal4 driver elicits site aversion (AIs of the control and experimental group: 0.02 ± 0.14 and −0.41 ± 0.13, respectively; n = 22, 19). Removal of the activation of dopaminergic neurons by combining the 6.2-Tβh-Gal4 driver with the TH-Gal80 driver does not influence aversion (AIs for the control and experimental group: 0.1 ± 0.12 and −0.49 ± 0.14, respectively; n = 14, 14). Removing ChAT positive neurons by using the ChAT-Gal80 in combination with the 6.2-Tβh-Gal4 driver resulted in loss of aversion (AIs of control and experimental group: −0.14 ± 0.17 and −0.08 ± 0.21 respectively; n = 13, 11). (B) Activation of the UAS-ChR2 transgene under the control of 6.2.-Tβh-Gal4 in Tβh^nM18 mutants eliminates aversion (AIs of the control and experimental group: −0.17 ± 0.1 and −0.19 ± 0.08, respectively; n = 20, 19). For activation, the flicker frequency of 2 s at 40 Hz, 16 s at 8 Hz and 2 s at 0 Hz was used. Errors are SEM. Differences from random choice were determined using the one-sample sign test and are indicated by the letter “a”. Student’s T-test was used to determine differences between two groups. *P < 0.05. For data, see Supplementary Table S2.

Figure 3

Frequency-dependent activation results in attraction. Different flicker patterns indicated above the panel were used to activate neurons in a dTdc2-Gal4-dependent manner (A,C) and 6.2-Tβh-Gal4-dependent manner (B,D). Only activation with a flicker of 2 s at 20 Hz, 16 s at 8 Hz followed by 2 s of silence continues to elicit site attraction (A, AIs of the control and experimental group: −0.08 ± 0.14 and 0.55 ± 0.13, respectively; n = 27, 23). (B) The 8 Hz pattern is sufficient to elicit site aversion when used to activate neurons (AIs of the control and experimental group: −0.11 ± 0.12 and −0.53 ± 0.11, respectively; n = 20, 14). (C) Activation with UAS-ReaChR in a dTdc2-Gal4 dependent manner leads to site attraction (2 s at 40 Hz, 16 s at 8 Hz activation pattern: AIs of the control and experimental group: −0.06 ± 0.15 and 0.6 ± 0.08, respectively; n = 18, 18; 2 s at 20 Hz, 16 s at 8 Hz activation pattern: AIs of the control and experimental group: −0.17 ± 0.1 and 0.6 ± 0.08, respectively; n = 13, 7; 16 s at 8 Hz activation pattern: AIs of the control and experimental group: 0.15 ± 0.1 and 0.47 ± 0.11, respectively; n = 12, 8). (D) Activation with UAS-ReaChR in a 6.2-Tβh-Gal4-dependent manner results in site aversion (2 s at 40 Hz, 16 s at 8 Hz activation pattern: AIs of the control and experimental group: −0.12 ± 0.1 and −0.42 ± 0.08, respectively; n = 23, 20; 2 s at 20 Hz, 16 s at 8 Hz activation pattern: AIs of the control and experimental group: −0.17 ± 0.2 and −0.61 ± 0.1, respectively; n = 23, 16; 16 s at 8 Hz activation pattern: AIs of the control and experimental group: −0.05 ± 0.17 and −0.77 ± 0.05, respectively; n = 18, 17). Errors are SEM. Differences from random choice as determined by the one-sample sign test are labeled with the letter “a”. Student’s T-test was used to determine differences between two groups with significance levels as follows: *P < 0.05, ***P < 0.001. For data, see Supplementary Table S3.

Figure 4

Octopaminergic neurotransmitter shifts behavioral outcomes. (A) Neuronal activation using the blue light-sensitive UAS-ChR2 transgene in a dTdc2-Gal4-dependent manner results in attraction to the blue light-illuminated odor trap. Addition of 5% ethanol to the blue light-illuminated trap did not change this attraction significantly (AIs of the control and experimental group: 0.66 ± 0.07 and 0.55 ± 0.1, respectively, n = 25, 17). (B) Addition of 5% ethanol to the yellow light-illuminated trap results in a significant reduction of the attraction to the blue light-illuminated food odor trap (AIs of the control and experimental group: 0.68 ± 0.09 and 0.28 ± 0.15, respectively, n = 20, 18). (C) Flies normally prefer the 5% ethanol-enriched food odor trap over the plain food odor trap, and Tβh^nM18 mutants differ significantly from the control in their behavior and do not show attraction. Pre-feeding control flies with 10% ethanol-enriched sucrose solution results in a similar degree of attraction to the 5% ethanol-containing food odor trap. Ethanol pre-feeding in Tβh^nM18 mutants results in a significant attraction to the 5% ethanol-containing food odor trap (AIs for w¹¹¹⁸: 0.44 ± 0.06 and with EtOH: 0.22 ± 0.08; for w¹¹¹⁸, Tβh^nM18: −0.07 ± 0.07 and with EtOH: 0.38 ± 0.06; n = 33, 31, 17, 20). The errors are SEM. The differences from random choice were determined using the one sample- sign test, and significance is indicated by the letter “a”. Student’s T-test was applied to determine differences between two groups, and ANOVA followed by Tukey post hoc analyses were used for comparisons among more than two groups. Significant differences are indicated as follows: *P < 0.05, **P < 0.01 and ***P < 0.001. For data, see Supplementary Table S4.

Figure 5

Octopamine is required for ethanol attraction. (A) Feeding with 53 mM octopamine restores the loss of attraction to ethanol-containing food odors in Tβh^nM18 mutants (AIs for w¹¹¹⁸: 0.4 ± 0.05 and with octopamine: 0.45 ± 0.11; for w¹¹¹⁸, Tβh^nM18: 0.06 ± 0.1 and with octopamine: 0.39 ± 0.07; n = 25, 21, 19, 22). (B) Blocking OA receptors by feeding 3 mM epinastine eliminates the attraction of control flies similarly to the observed loss of attraction in Tβh^nM18 mutants (AIs for w¹¹¹⁸: 0.45 ± 0.08 and with epinastine: 0.11 ± 0.11; for w¹¹¹⁸, Tβh^nM18: 0.02 ± 0.15; n = 29, 28, 19). (C) Tβh^nM18 mutants fed with 50 mM clonidine show a significant attraction to ethanol-containing food odors (AIs for w¹¹¹⁸: 0.33 ± 0.04 and with clonidine: 0.27 ± 0.03; for w¹¹¹⁸, Tβh^nM18: 0.1 ± 0.08 and with clonidine: 0.47 ± 0.05; n = 30, 28, 29, 31). (D) Feeding 200 nM naphazoline to Tβh^nM18 mutants restores the loss of attraction to ethanol-containing food odors and significantly reduces the attraction of the control group (AIs for w¹¹¹⁸: 0.59 ± 0.04 and with naphazoline: 0.31 ± 0.08; for w¹¹¹⁸, Tβh^nM18: 0.11 ± 0.09 and with naphazoline: 0.45 ± 0.07; n = 32, 33, 29, 33). (E) Blocking TA receptors by feeding with 25 mM yohimbine does not alter the attraction to ethanol-containing food odors (AIs for w¹¹¹⁸: 0.38 ± 0.08 and with yohimbine: 0.41 ± 0.07; for w¹¹¹⁸, Tbh^nM18: 0.01 ± 0.11 and with yohimbine: 0.09 ± 0.1; n = 40, 40, 21, 24). (F) Increasing tyramine levels in control flies do not significantly alter attraction, but they significantly induce attraction in Tβh^nM18 mutants (AIs for w¹¹¹⁸: 0.42 ± 0.05 and with tyramine: 0.32 ± 0.09; for Tβh^nM18: −0.01 ± 0.09 and with tyramine: 0.38 ± 0.08). Errors are SEM, and the letter “a” indicates differences from random choice as determined by the one-sample sign test. Student’s T-test was used to determine differences between two groups and ANOVA followed by the Tukey post hoc test for more than two groups. *P < 0.05, **P < 0.01, ***P < 0.001. For data, see Supplementary Table S5.

Figure 6

Model for octopamine induced attraction and aversion. The circles highlight the Tbh positive neurons targeted by the dTdc2-Gal4 driver in the adult brain (Schneider et al., 2012). The Tbh is the rate limiting enzyme for octopamine synthesis (Monastirioti et al., 1996). Activation of these neurons using UAS-ChR2 elicits site attraction (Figure 1). When the Gal4 expression of the driver is restricted by co-expression of ChAT-GAL80 driver the flies choose both odor traps equally (Figure 1). The neurons that are not affected by the ChAT-Gal80 repressor are indicated in open circles. They only express Tbh and activation of these neurons results in indifferences (Figure 1). The ChAT and Tbh expressing neurons are indicated by gray circles and they are responsible—when activated—to mediate attraction. Within these set of positive neurons, the activation of the VUMa4 neuron (indicated by a circle that is half black and half gray) results in aversion (Figure 2). The VUMa4 neurons are Tbh positive (Schneider et al., 2012). Since the ChAT-Gal80 repressor eliminates the aversion caused by activation of the 6.2 Tbh-Gal4 driver, the VUMa4 neuron is also ChAT positive (Figure 2). The activation of the VUMa4 neuron results in aversion, but when the VUMa4 neuron is activated together with other octopaminergic neurons, attraction occurs. This result is consistent with the idea that the activation of a second set of octopaminergic neurons can overrule the octopamine-induced aversion by the VUMa4 neuron. For data, see Supplementary Table S6.