Modulation of early olfactory processing by an octopaminergic reinforcement pathway in the honeybee

Abstract

Processing of olfactory information in the antennal lobes of insects and olfactory bulbs of vertebrates is modulated by centrifugal inputs that represent reinforcing events. Octopamine release by one such pathway in the honeybee antennal lobe modulates olfactory processing in relation to nectar (sucrose) reinforcement. To test more specifically what role octopamine plays in the antennal lobe, we used two treatments to disrupt an octopamine receptor from Apis mellifera brain (AmOAR) function: (1) an OAR antagonist, mianserin, was used to block receptor function, and (2) AmOAR double-stranded RNA was used to silence receptor expression. Both treatments inhibited olfactory acquisition and recall, but they did not disrupt odor discrimination. These results suggest that octopamine mediates consolidation of a component of olfactory memory at this early processing stage in the antennal lobe. Furthermore, after consolidation, octopamine release becomes essential for recall, which suggests that the modulatory circuits become incorporated as essential components of neural representations that activate odor memory.

Figure 1.

Summary of MAS and dsRNA injection protocols. a, MAS dual-injection protocol. Four injection combinations were tested as indicated. One hour after surgery, saline or MAS was injected into the AL. Injection was followed 10 min later by six acquisition trials with the conditioned odor (C), which were separated by 30 sec intertrial intervals. Ninety minutes later, subjects were given a second application of saline or MAS. Ten minutes later, subjects were tested with C, a molecularly similar (S), and a molecularly dissimilar (D) odor in a randomized order. C, S, and Dtrials were 3 min apart. b, dsRNA injection protocol. b. 1, One hour after surgery, saline or dsRNA was injected into the AL of honeybee brain. After 24 hr, bees were conditioned with odor C. Ninety minutes later, they were tested with C, S, and D odors in randomized order. b. 2, One hour after surgery, bees were preconditioned with odor C. Two hours later, injection buffer or dsRNA was injected into the AL. Twenty-four hours later, bees were tested with C, S, and D odors. c, Schematic diagram of the honeybee head after a small window had been cut in the head capsule. CE, Compound eye; Oc, ocelli; An, antenna; Md, mandibles; OL, optic lobes; MB, mushroom body. d, Chemical structures of odorants used in the present study.

Figure 4.

Partial sequence of AmOAR and its alignment with other octopamine receptors. a, Partial nucleotide and amino acid sequence of AmOAR (GenBank accession number AYZ63366). Transmembrane regions VI and VII are underlined. Amino acid residues selected for peptide synthesis are indicated by the box. b, Amino acid alignment of AmOAR receptor with the D. melanogaster octopamine receptor (Dm-OAR; AF065443), B. mori octopamine receptor (Bm-OAR; X95607), L. stagnalis octopamine receptor (Lm-OAR; u62771), and A. kurodai octopamine receptor (Ak-OAR; AF117654). Residues conserved among all five octopamine receptors are shaded in gray. A dash is used to indicate a gap in the sequence. Amino acids conserved among Am-OAR and Dm-OAR are boxed. c, Amino acid alignment of AmOAR receptor with A. mellifera dopamine receptor D1 (AmDOP1; CAA73841), Drosophila dopamine D1 receptor (DAMB; AAB08000), Drosophila dopamine D1-like receptor (D1-like; AAA85716), and Drosophila dopamine D2 receptor (DopR2; Q24563). Residues conserved among all five proteins are shaded in gray. Residues conserved among all dopamine receptors are indicated by dashed lines. Residues conserved among AmOAR and AmDOP1 are boxed.

Figure 2.

The effect of MAS dose on acquisition and/or recall testing. a, Acquisition to the conditioned odor (C: 1-octanol). Four nanoliters of saline—MAS solution (20, 200, or 2000 μm) was injected into each AL 10 min before conditioning. Subjects were conditioned as in Figure 1a. b, Ninety minutes after conditioning, 4 nl of saline—MAS was injected again in each AL, and 10 min later, subjects were tested with odors 1-octanol (C), 1-hexanol (S), and geraniol (D) in a randomized manner. Sample size: saline (n = 42), MAS 20 μm (n = 51), MAS 200 μm (n = 55), MAS 2000 μm (n = 43). Asterisks indicate significant differences of respective points from control group (^*p < 0.05; ^**p < 0.005; ^***p < 0.0001). ns, Not significantly different from control group.

Figure 3.

The effect of injection combination on acquisition and/or recall testing. a, Acquisition to conditioned odor (1-octanol). Subjects received a single application of saline or saline—MAS. Ten minutes later, subjects were conditioned as in Figure 1a. b, Ninety minutes after conditioning, 4 nl of saline or saline-MAS was applied again in each AL, followed by the C, S, and D test procedure. Sample size: saline—saline (n = 42), MAS—saline (n = 55), saline—MAS (n = 48), MAS—MAS (n = 55). Asterisks indicate significant differences of respective points from control group (^*p < 0.05, ^***p < 0.0001). ns, Not significantly different from control group.

Figure 5.

AmOAR dsRNA impairs OAR protein translation in AL of honeybee. a, Western blotting was performed on protein fractions from control AL homogenates. I, Nitrocellulose membrane was probed with anti-AmOAR antiserum. A 78 kDa protein (arrow) was recognized by the anti-AmOAR antiserum. II, Nitrocellulose membrane was probed with anti-AmOAR antiserum preadsorbed with corresponding peptide resulted in significant decrease in intensity of 78 kDa band (p < 0.0001). The same blots were stripped and reprobed with anti-HRP antiserum (for reference). b, Western blotting was performed on protein fractions from control and 24 hr-dsRNA-injected AL homogenates. Nitrocellulose membrane was probed with anti-AmOAR antiserum. A significant reduction in 78 kDa band was observed in dsRNA injected AL homogenate (p < 0.0001). The blot was stripped and reprobed with anti-HRP antiserum (for reference) to confirm that equivalent amount of protein was loaded in each lane.

Figure 6.

The effect of AmOAR dsRNA on acquisition and/or recall testing. a, Acquisition to conditioned odor (C). In this experiment, 2 hr after surgery, subjects received 4 nl of injection buffer—AmOAR dsRNA (500 pg) in each AL. Twenty-four hours later, subjects were conditioned with odor C. b, Ninety minutes later, subjects were tested with C, S, and D in a randomized manner. Asterisk indicates significant difference of respective point from control group (^*p < 0.01; ^***p < 0.0001). Sample sizes: saline (n = 31), AmOAR dsRNA (n = 31).

Figure 7.

The effect of preconditioning on AmOAR dsRNA-mediated recall response. a, Conditioning of subjects in 2 control groups. Two hours after conditioning, in one group, injection buffer was injected into each AL. In the second group, 4 nl of AmOAR dsRNA (500 pg) was injected into each AL. b, Twenty-four hours later, subjects were tested with C, S, and D in a randomized manner. Asterisk indicates significant difference of respective point from control group (^*p < 0.05). Sample sizes: control (n = 36), AmOAR dsRNA (n = 35). ns, No significant difference.

Figure 8.

Other dsRNA types do not block acquisition response. On acquisition to conditioned odor (C): two hours after surgery, subjects received 4 nl of either injection buffer—Drosophila dsRNA (500 pg), buffer—yeast tRNA (500 pg), or buffer—E. coli tRNA (500 pg) in each AL. Twenty-four hours later, subjects were conditioned with C. Asterisk indicates significant difference of respective response from control group (^***p < 0.0001). Sample size:control (n = 23), Drosophila dsRNA (n = 37), yeast tRNA (n = 26), E. coli tRNA (n = 43). ns, No significant difference was detected among these four groups.

Figure 9.

MAS and AmOAR dsRNA do not block acquisition when applied to OL. a, Acquisition response to conditioned odor (C). Two hours after surgery, subjects received 4 nl of saline—MAS solution (200 μm) in each OL. Ten minutes later, subjects were conditioned with six acquisition trials of 30 sec intertrial interval. b, Ninety minutes after conditioning, 4 nl of saline or MAS was applied again in each AL. Ten minutes later, subjects were tested with odors C, S, and D in a randomized manner. c, Two hours after surgery, subjects received 4 nl of injection buffer—AmOAR dsRNA (500 pg) in each OL. Twenty-four hours later, subjects were conditioned with odor C. d, Ninety minutes later, subjects were tested with C, S, and D in a randomized manner. Sample size: saline—saline (n = 43), MAS—MAS (n = 43), buffer (n = 18), AmOAR dsRNA (n = 23). ns, No significant difference was detected among control and treated group.

Figure 10.

Effect of time intervals after injection of MAS or AmOAR dsRNA on the acquisition response. a, Two hours after surgery, subjects received 4 nl of saline or MAS solution (200 μm) in each AL. Twenty minutes later, subjects were conditioned with six acquisition trials of 30 sec intertrial interval. Twenty-four hours later, subjects were reconditioned with odor C. Sample size:saline (n = 42) 20 min after injection; MAS (n = 22) 20 min after injection; saline(n = 12) 24 hr after injection; MAS (n = 20) 24 hr after injection. Asterisk indicates significant difference between 20 min and 24 hr MAS after injection response (^*p < 0.02). b, Two hours after surgery, subjects received 4 nl of AmOAR dsRNA solution (500 pg) in each AL. Twenty-four hours later, subjects were conditioned with six acquisition trials of 30 sec intertrial interval. Forty-eight hours later, subjects were reconditioned with odor C. Sample size: injection buffer (n = 7) 24 hr after injection; AmOAR dsRNA (n = 14) 24 hr after injection; injection buffer (n = 10) 48 hr after injection; AmOAR dsRNA (n = 14) 48 hr after injection. ns, No significant difference was detected between 24 and 48 hr after injection.