nAChR-induced octopamine release mediates the effect of nicotine on a startle response in Drosophila melanogaster


  • Nicolás Fuenzalida-Uribe,

    1. Departamento de Biología Celular y Molecular, Pontificia Universidad Católica de Chile, Millennium Nucleus Stress and Addiction (NEDA), Santiago, CHILE
    Search for more papers by this author
  • Rodrigo C. Meza,

    1. Departamento de Fisiología, Facultad de Ciencias Biológicas, Pontificia Universidad Católica de Chile, Millennium Nucleus Stress and Addiction (NEDA), Santiago, CHILE
    Search for more papers by this author
  • Hernán A. Hoffmann,

    1. Facultad de Medicina, Pontificia Universidad Católica de Chile, Millennium Nucleus Stress and Addiction (NEDA), Santiago, CHILE
    Search for more papers by this author
  • Rodrigo Varas,

    1. Departamento de Fisiología, Facultad de Ciencias Biológicas, Pontificia Universidad Católica de Chile, Millennium Nucleus Stress and Addiction (NEDA), Santiago, CHILE
    Search for more papers by this author
  • Jorge M. Campusano

    Corresponding author
    • Departamento de Biología Celular y Molecular, Pontificia Universidad Católica de Chile, Millennium Nucleus Stress and Addiction (NEDA), Santiago, CHILE
    Search for more papers by this author

Address correspondence and reprint requests to Jorge M. Campusano, Departamento de Biología Molecular y Celular, Facultad de CienciasBiológicas, Pontificia Universidad Católica de Chile, PO Box 114-D, Santiago, Chile. E-mail:


Biogenic amines (BAs) play a central role in the generation of complex behaviors in vertebrates and invertebrates, including the fly Drosophila melanogaster. The comparative advantages of Drosophila as a genetic model to study the contribution of BAs to behaviors stumble upon the difficulty to access the fly brain to ask relevant physiological questions. For instance, it is not known whether the activation of nicotinic acetylcholine receptors (nAChRs) induces the release of BAs in fly brain, a phenomenon associated to several behaviors in vertebrates. Here, we describe a new preparation to study the efflux of BAs in the adult fly brain by in vitro chronoamperometry. Using this preparation we show that nAChR agonists including nicotine induce a fast, transient, dose-dependent efflux of endogenous BAs, an effect mediated by α-bungarotoxin-sensitive nAChRs. By using different genetic tools we demonstrate that the BA whose efflux is induced by nAChR activation is octopamine (Oct). Furthermore, we show that the impairment of a mechanically induced startle response after nicotine exposure is not observed in flies deficient in Oct transmission. Thus, our data show that the efflux of BAs in Drosophila brain is increased by nAChR activation as in vertebrates, and that then AChR-induced Oct release could have implications in a nicotine-induced behavioral response.

Abbreviations used

biogenic amines




DA reuptake transporter


DA decarboxylase


tyrosine decarboxylase 2




tyramine beta hydroxylase


monoamine vesicular transporter

Biogenic amines (BAs) play a central role in a wide range of behaviors such as the regulation of arousal state, associative learning, sexual behavior, reward processing and the control of motor function, both in vertebrates and invertebrates (Bainton et al. 2000; Kume et al. 2005; Draper et al. 2007; Schultz 2007; Sara 2009). As the aminergic systems are phylogenetically conserved, it is possible to study their contribution to the generation or modulation of many of these complex behaviors in simpler animal models that show fewer limitations from an ethical and/or technical point of view, including the fly Drosophila melanogaster. This invertebrate animal model has proven useful to study the molecular mechanisms underlying behaviors, in part because of its powerful genetic tools (Greenspan and Ferveur 2000; Sokolowski 2001; Helfrich-Förster 2005) and the fact that key elements of aminergic neurotransmission are evolutionary conserved from arthropods to humans. For instance, in the case of dopamine (DA), arguably the best studied of all BA species, several dopaminergic-related proteins have been described in Drosophila, including the biosynthetic enzymes tyrosine-hydroxylase (TH; Neckameyer and Quinn 1989) and l-DOPA decarboxylase (Tempel et al. 1984), the monoamine vesicular transporter (vMAT; Greer et al. 2005), the DA reuptake transporter (DAT; Pörzgen et al. 2001) and the D1-like and D2-like dopaminergic receptor subtypes (Blenau and Baumann 2001).

Evidence suggests that nAChRs are one of the key regulators of BA signaling in vertebrates. These are ionotropic receptors located in nerve terminals as well as in the cell bodies of aminergic neurons (Champtiaux et al. 2003; Albuquerque et al. 2009; Livingstone and Wonnacott 2009). Interestingly, nAChRs mediate fast synaptic neurotransmission at most central synapses in the Drosophila brain (Gundelfinger and Hess 1992) but their role as regulators of BA release has not been elucidated in this model. This is an important issue as it has been suggested that BA release and signaling play a key role in the generation or modulation of several behaviors in Drosophila. For instance, behavioral responses induced in flies after exposure to drugs of abuse, depend on DA and/or octopamine (Oct) signaling (Bainton et al. 2000; Hardie et al. 2007).

Oct is a BA highly represented in invertebrates and is considered the functional homolog of the vertebrate noradrenaline/adrenaline (Roeder 2005; Verlinden et al. 2010). At the molecular level, the Oct receptors share several functional and sequence homology to adrenergic/noradrenergic vertebrate receptors (Blenau and Baumann 2001; Evans and Maqueira 2005). Moreover, similarities –and also differences– have been described between the tyramine beta hydroxylase (TβH) and DA beta hydroxylase, responsible for Oct and noradrenaline biosynthesis, respectively (Monastirioti et al. 1996; Hess et al. 2008).

The possibility that nAChR activation regulates BA release has not been tackled in the past because of the technical difficulties associated to getting access to the small fly brain. However, in the last few years, new preparations have been developed to study the cellular and molecular determinants of neuronal communication. For example, by using fast-scan cyclic voltammetry and an optogenetic approach, the group of Jill Venton has been able to measure the release of DA (Vickrey et al. 2009) and serotonin (Borue et al. 2009), in a larval ventral nerve cord preparation.

Here, we describe a new preparation that allows us to evaluate for the first time the efflux of endogenous BAs from adult Drosophila brain. By using this new preparation we examined the hypothesis that activation of nAChRs induces the release of endogenous BAs in fly brain. Moreover, we show data suggesting that Oct release induced by nAChR activation is involved in the modulation of a startle response by nicotine in the fly.

Materials and methods

Drosophila stocks

Flies were maintained on a standard agar medium (yeast, sugar, agar, flour and propionic acid and nipagin), at 18°C under a 12/12 h light/dark cycle. The day before the experimental day, animals were brought to room temperature (24–25°C). The following transgenic animals were obtained from the Bloomington stock center (Indiana University, Bloomington, IN, USA): Tdc2-Gal4 (line #9313; Cole et al. 2005), DDC-Gal4 (line #7009; Li et al. 2000; Sitaraman et al. 2008), UAS-RNAiTβH(line #27667; JF02746 Transgenic RNAi Project, Harvard Medical School, Cambridge, MA, USA), UAS-RNAiDDC(line #27030; JF02356 Transgenic RNAi Project, Harvard Medical School) and UAS-TeTx.LC (line # 28997; Sweeney et al. 1995). TβH mutant flies were a generous gift from Dr. Vivian Budnik (University of Massachusetts Medical School, Worcester, MA, USA; Monastirioti et al. 1996). Wild-type Canton-S, UAS-shits1and DDC mutant flies were part of the Campusano lab stock (originally part of Dr. Diane O'Dowd lab stock at University of California Irvine, CA, USA).

Whenever needed, brains were obtained from Tdc2-Gal4/+;UAS-shits1/+ or Ddc-Gal4/+;UAS-shits1/+ flies and brought to 24°C or 32°C. These flies express a temperature-sensitive mutation in the dynamin gene in tyrosine decarboxylase 2 (Tdc2) or DA decarboxylase (DDC) positive neurons, respectively. Tdc2-Gal4/+;UAS-TeTx.LC/+ and DDC-Gal4/+;UAS-TeTx.LC/+ flies express a copy of the tetanus toxin gene in tyraminergic/octopaminergic and dopaminergic neuronal populations, respectively. The expression of this toxin has been shown to block synaptic vesicle recycling and ultimately chemical neurotransmission (Sweeney et al. 1995). Brains obtained from Tdc2-Gal4/+;UAS-RNAiTβH/+ animals, were used to reduce the Oct synthesis in Tdc2 positive neurons, as an alternative to the TβH mutant. When using mutants we carried out experiments in cantonized animals to avoid possible effects of genetic background on our results.

Fly brain dissection and preparation

One adult fly male (3–6 days old) was anesthetized under CO2 and its brain was removed. The tissue was pinned down to the bottom of a recording chamber (total volume 500 μL) and superfused (3 mL/min) with a standard HEPES-buffered Tyrode solution (in mM: NaCl 140, KCl 4.5, HEPES 10, MgCl2 1.0, CaCl2 2.5, glucose 11; pH 7.2) under continuous supply of air, at 24°C.

High-speed chronoamperometric recordings

Electrochemical measurement of BAs was carried out with a microcomputer-controlled high-speed chronoamperometric system (IVEC-10, Medical System Corp., Greenvale, NY, USA). Carbon fiber electrodes (tip diameter, 30 μm, Rocky Mountain Centre for Sensor Technology, Denver, CO, USA), covered with Nafion (Aldrich Chemical Co., Milwaukee, WI, USA) to increase the sensitivity for amines (Gerhardt et al. 1984), were positioned onto the mid-ventral side of the fly brain (dorsal to the antennal lobes, on top of the zone where the ellipsoid body is located) opposite to the inflow of saline solution. A 30-min stabilization period was defined before starting any manipulation.

A 0.7 V respect to a reference Ag-AgCl electrode was applied for 100 ms at 5 Hz. The resulting oxidation current was digitally integrated during the last 80 ms of each pulse, averaged for five cycles, displayed at 1 Hz and stored in the computer. The reduction current generated when the potential returned to 0 V was processed in the same manner. The IVEC-10 system expresses the integrated currents in arbitrary units, which are proportional to BA concentration. The sensitivity and linearity of the Nafion-coated electrodes were determined in vitro by calibrating their response to solutions containing DA, serotonin, tyramine (Sigma-Aldrich, Saint Louis, MO, USA) or Oct (Tocris Bioscience, Bristol, UK) (Fig. 1a). Final concentration for each BA was 2–12 μM in HEPES buffer at pH 7.2, with 0.1 M HCLO4 to prevent oxidation. These calibrations were performed every experimental day at 24°C and at 32°C whenever needed. Only electrodes showing linear responses (r > 0.99) to increasing concentrations of BAs were used. The threshold for BA determinations was typically 10–20 nM. Evoked BA efflux over baseline (ΔBAefflux) was computed from changes in the oxidation currents recorded by the carbon electrodes.

Figure 1.

(a) Reduction to oxidation currents ratio (red/ox ratio) for dopamine (DA, n = 7), octopamine (OCT, n = 7), tyramine (TYR, n = 7) and serotonin (5-HT, n = 7). Carbon fiber electrodes were exposed to 6 μM of each amine in an HEPES buffer solution. Bars represent the mean + SEM of the red/ox ratios. No differences were observed between the electrode responses to DA and 5-HT or TYR and OCT. However, differences were observed between the first and second group of amines (***p < 0.001). (b) Representative trace demonstrating how control experiments (no fly brain in the chamber) are carried out: after a 30 min of basal recording stabilization (‘Signal stabilization’), a carbon fiber electrode is exposed to the drugs of interest. No responses are detected after a buffer change, or after exposure to nicotine (Nic, 5 mM) or PNU-282987 (PNU, 5 mM). (c) Raw trace of a typical experiment in which after a period of stabilization (not shown), a fly brain is exposed to a buffer change and latter to nicotine (Nic, 5 mM). A response is only detected after nicotine stimulus. (d) Representative trace showing the response of a single brain to six nicotine (3 mM) stimuli applied consecutively every 3 min. S1-S6 indicate the first to the sixth consecutive stimulus. No exhaustion is evident in the responses.

All drugs were freshly prepared in the Hepes-Tyrode solution described above and diluted to the desired working concentration. None of the drugs used in this study, produced signals that could be mistakenly attributed to the oxidation or reduction of the amines of interest, at the concentrations reported (Fig. 1b). The nAChR agonists were applied by a fast bolus injection (200 μL) into the recording chamber from a pipette located at about 1 mm from the brain surface. Control experiments were performed to discard that the rapid exchange of solutions could affect the recordings (Fig. 1c). nAChR blockers were applied continuously through the superfusion system for at least 5 min before tested. A single brain was usually subjected to several manipulations, as we do not see exhaustion of the preparation after several stimuli (Fig. 1d).

Nicotine volatilization and the startle response to a mechanical stimulus

Experiments were carried out as in Bainton et al. (2000). Briefly, 10–12 male flies were placed in a test tube divided into three proportional regions (top, middle and bottom), where they were exposed to volatilized Nicotine (6 μg) or saline for 1 min. Afterwards, the tube was gently tapped down three times. The number of flies in each tube region was recorded 5 s after this mechanical stimulus. Then, the geotaxis index was calculated for each group of control or genetically deficient flies by subtracting the number of flies in the bottom region of the tube from the number of flies in the top region, divided by the total number of flies. Data are presented considering as 100% the startle response recorded in the animals of interest when exposed to volatilized saline. For these experiments we received the kind help of Dr. Ulrike Heberlein (HHMI/Janelia Farm, VA, USA).


Nicotine (tartrate salt) was purchased from Sigma-Aldrich. PNU-282987 (PNU) and α-bungarotoxin were purchased from Tocris Bioscience (Bristol, UK). Erysodine was a gift of Dr P. Iturriaga-Vásquez (Faculty of Sciences, Universidad de Chile, Santiago, Chile; Iturriaga-Vásquez et al. 2010).

Data analysis

Values given are mean ± SEM. Statistical comparisons were done using either Student's t-test or Dunn's Multiple Comparison post hoc test, after a Kruskal–Wallis test. The significance level was set at p < 0.05. All curve fitting and statistical calculations were performed with GraphPad Prism 5.0 (GraphPad Software, La Jolla, CA, USA). The dose-dependent activation of the currents by nAChR agonists was fitted to the variable slope model equation: Y = ΔBAmin + (ΔBAmax − ΔBAmin)/1 + 10 ((log EC 50-X) × Hill slope), where X is the log of agonist concentration (in mM), ΔBAmin and ΔBAmax are the minimum and maximal response detected, respectively, and EC50 is the agonist concentration that evoked the half-maximal response.

Biosafety issues

The experimental procedures were approved by the Bioethical and Biosafety Committee of the Facultad de Ciencias Biológicas, Pontificia Universidad Católica de Chile and were conducted in accordance with the guidelines of the National Fund for Scientific and Technological Research (FONDECYT) and the Servicio Agricola y Ganadero de Chile (SAG).


First, we evaluated the electrode ability to detect DA, Oct, serotonin and tyramine, at different concentrations. As shown in Fig. 1a, we were able to measure the four BAs assayed. Some differences were evident in the red/ox ratio recorded for the BAs at the range of concentrations that we later describe as relevant in the brain preparation (see below): the redox ratio for DA and serotonin is significantly smaller that the ratio recorded for Oct or Tyr (Fig. 1a).

We then positioned the electrode upon an isolated fly brain in the recording chamber and applied a fast bolus injection of KCl (300 mM, 200 μL). Previous calibrations show that the concentration of any drug effectively reaching the tissue is about 1 : 30 of the applied stimulus (Meza et al. 2012). This suggests that the actual concentration of K+ reaching the brain would be about 10 mM, a KCl concentration that induces a mild depolarization in fly brain neurons (Campusano et al. 2007). In these conditions, KCl (300 mM) induces a detectable and reproducible ΔBAefflux from the fly brain of 115.1 ± 35.8 nM of BAs (data not shown). These data show that a non-specific depolarizing agent induces the release of neuroactive molecules from the fly brain. We decided to stimulate the tissue with nicotine, the prototypic non-selective agonist of nAChRs, which are the main mediators of fast excitatory neurotransmission in fly brain (Gundelfinger and Hess 1992). As preliminary experiments demonstrated no exhaustion of this preparation after several consecutive stimuli with a single nicotine concentration (Fig. 1d), we carried out several pharmacological manipulations in a single brain. Our results obtained from multiple brains stimulated with a range of drug concentrations, show that nicotine induces a dose-dependent increase in ΔBAefflux (n = 10, Fig. 2) with an apparent EC50 of 1.1 ± 0.2 mM and a maximal response of 1.3 ± 0.2 μM.

Figure 2.

Dose-dependent efflux of biogenic amines (BAs) induced by fast application of increasing concentrations of nicotine (in black) and PNU 282927 (in gray) upon the isolated fly brain. Each point represents the mean ± SEM of at least five independent recordings. Inset, representative traces obtained from a single fly brain exposed to increasing concentrations of nicotine up to 5 mM.

Several reports show that most of the nAChRs expressed in the brain of insects share similar properties with the vertebrate homomeric α-bungarotoxin-sensitive nAChRs (Goldberg et al. 1999; Campusano et al. 2007). We therefore evaluated the effect of PNU, a selective agonist of these homomeric receptors in vertebrate preparations (Wonnacott and Barik 2007) on BA efflux in our fly brain preparation. PNU mimics the dose-dependent effects of nicotine with an apparent EC50 of 1.6 ± 0.3 mM and a maximal response of 4.4 ± 0.5 μM (n = 10, Fig. 2).

In agree with these results, α-bungarotoxin (10 nM), an antagonist of calcium-permeable nAChRs in vertebrate and invertebrate neurons (Campusano et al. 2007), almost completely blocked the responses induced by nicotine (Fig. 3a and b) and PNU (ΔBAefflux in presence of PNU = 1.0 ± 0.3 μM, n = 4; ΔBAefflux in PNU plus α-bungarotoxin = 0.03 ± 0.02 μM, n = 4; p < 0.05, data not shown). On the other hand, erysodine (1 μM), a nAChR antagonist structurally related to dihydro-β-erythroidine, which show higher affinity for heteromeric nAChRs in vertebrate preparations, did not affect nicotine-induced responses (Fig. 3c). These results show that the activation of α-bungarotoxin-sensitive nAChRs induces ΔBAefflux in the fly brain.

Figure 3.

Nicotine-induced release of endogenous biogenic amines (BAs) depends on the activation of α-bungarotoxin-sensitive nAChRs. (a) Representative trace of an experiment in which a fly brain was exposed to nicotine (5 mM) in absence or presence of the calcium-permeable nAChR antagonist α-bungarotoxin (BGT, 10 nM). The effect of the antagonist was reversible (Recovery). (b) Nicotine (5 mM) induced effect on endogenous BA release is blocked by BGT (*p < 0.05, n = 6). (c) Nicotine (5 mM) induced effect is not modified by the vertebrate heteromeric nAChR antagonist erysodine (ERY, 1 μM; p > 0.05, n = 5).

We then decided to evaluate the identity of the BA(s) released by nAChRs activation. We were expecting that the differences in redox ratio recorded for the different amines would let us distinguish the amines being released from the fly brain preparation. However, this was not possible as redox ratios are very similar for DA and serotonin or Oct and Tyr, as shown in Fig. 1a. Thus, we decided to use genetic tools to try to identify the amine whose release was modified by nAChR activation.

We assessed the nicotine-induced ΔBAefflux in fly brains obtained from Tdc2-Gal4/+;UAS-RNAiTβH/+ animals. TβH is the enzyme that converts tyramine to Oct (Roeder 2005). As we directed the expression of the RNAi for this enzyme to tyraminergic/octopaminergic neurons, it is expected that Oct synthesis is impaired in these neurons, similar to what happens in animals expressing a mutation for TβH (Monastirioti et al. 1996). In these flies the nicotine-induced ΔBAefflux is dramatically reduced (Fig. 4a). Moreover, the drug-induced BA efflux remains unaffected in DDC mutants, which are affected in DA and serotonin synthesis (data not shown). Therefore, these data support the idea that the BA whose release is being modulated by nAChRs is Oct.

Figure 4.

Nicotine stimulation selectively induces the efflux of Oct from fly brain. (a) Brains obtained from flies expressing an RNAi directed to tyramine beta hydroxylase (TβH) mRNA in tyrosine decarboxylase 2 (Tdc2) neurons (Tdc2/RNAiTβH) show a reduced nicotine-induced ∆BAefflux compared with its genetic control, the undriven RNAi (Cs/RNAiTβH; n = 5, **p < 0.001). (b) Full blockade of the nicotine-induced BA efflux in brains obtained from flies expressing the temperature-sensitive shibire mutation in Tdc2 neurons (Tdc2/Shits) is observed at the restrictive temperature (32°C) when compared with the response at the permissive temperature (20°C) (n = 6, *p < 0.05). (c) Increasing the temperature to 32°C does not affect the nicotine-induced efflux of BAs in flies expressing the temperature-sensitive shibire mutation in DA decarboxylase (DDC) neurons (DDC/Shits) (n = 6, p > 0.05).

To confirm this, we studied the nicotine-induced ΔBAefflux in brains obtained from Tdc2-Gal4/+;UAS-shits1/+ animals at the permissive (20°C) and non-permissive (32°C) temperatures. Nicotine-induced ΔBAefflux is substantially reduced at the non-permissive temperature compared with the BA efflux recorded at 20°C (Fig. 4b). Furthermore, nicotine-induced BA efflux is not affected in Drosophila brains from DDC-Gal4/+;UAS-shits1/+ animals (Fig. 4c), a genetic manipulation that affects DA and serotonin release. Overall, these results demonstrate that activation of α-bungarotoxin-sensitive nAChRs induces Oct efflux in the fly brain.

We finally asked whether the nicotine-induced Oct release has any physiological consequence in Drosophila behavior. It has been previously shown that flies exposed to volatilized nicotine show a range of behaviors that depends on the dose of the drug used. For instance, about 50% of flies exposed to volatilized nicotine (6 μg) show reduced startle response to a mechanical stimulus. As the nicotine pharmacological target is nAChRs, and these receptors are only expressed in the fly CNS (Gundelfinger and Hess 1992), it is thought that the effect of nicotine is centrally mediated (Bainton et al. 2000).

Our chronoamperometry results suggest that the nicotine-induced activation of nAChRs modulates Oct release in fly brain, which could be responsible for a modification of motor behavior and ultimately the effect of nicotine on the startle response. In the original report from Bainton et al. (2000), it was reported that DA was at least partially responsible for the nicotine-induced effect, something that we also corroborated: flies expressing an RNAi for DDC driven by TH-Gal4 (TH-Gal4/UAS-RNAiDDC) showed no reduction in startle response when exposed to volatilized nicotine, compared with the effect observed in Cs/UAS-RNAiDDC, the undrivenRNAiDDC genetic control animals, after exposure to this drug (110 ± 14.5 and 58 ± 7%, in flies expressing the RNAiDDCand in genetic controls, respectively, p < 0.05). The 100% setpoint corresponds to the fly ability to climb up after exposure to volatilized saline (5.4 ± 0.5 and 5.56 ± 0.6 mm, in Cs/UAS-RNAiDDC genetic control animals and TH-Gal4/UAS-RNAiDDC animals, respectively).

We decided to further evaluate whether flies deficient in Oct signaling would also show impairment in the nicotine-induced effect (Fig. 5). First, flies deficient in Oct synthesis by expression of an RNAi for TβH in octopaminergic/tyraminergic neurons were exposed to volatilized nicotine (6 μg) and their startle response was recorded. Results obtained show that the reduced startle response observed in genetic control flies exposed to nicotine, disappears in flies deficient in Oct synthesis (Fig. 5a). Consistent with this data, flies expressing a mutation in the TβH enzyme were exposed to volatilized nicotine. It was not possible to observe a reduction in the mechanically induced startle response in the mutant flies (Fig. 5a). Therefore, these data obtained using two different genetic approaches suggest that Oct mediates the reduced startle response observed when flies are exposed to nicotine.

Figure 5.

Flies deficient in Oct synthesis or release do not show the nicotine-induced effect on startle response. Flies were exposed to volatilized nicotine (6 μg) as described in Bainton et al. (2000). (a) Genetic control flies show a reduced startle response after nicotine exposure (Cs/RNAiTβH). This nicotine-induced effect is not observed in flies expressing an RNAi for tyramine beta hydroxylase (TβH) (Tdc2/RNAiTβH), the enzyme responsible for Oct synthesis. This lack of effect of nicotine stimulation in the startle response is also observed in flies expressing a mutation for the TβH enzyme (tβHmut). Genetic control animals for the RNAi manipulation and the mutant, climbed up to 4.2 ± 0.6 and 3.2 ± 0.3, respectively, in presence of volatilized saline (100% setpoint). (b) Flies expressing tetanus toxin in octopaminergic neurons do not show the nicotine-induced effect (Tdc2/Tetx). Control flies for this genetic manipulation (Cs/Tetx) show a reduced startle response when exposed to volatilized nicotine. These control animals, climbed up to 4.2 ± 1.4 in presence of volatilized saline (100% setpoint). Data represent mean + SEM of at least three different experiments, each one consisting of at least 10 male flies. * and ** indicate statistical significance with experimental control, p < 0.05 and p < 0.01, t-test and anova.

In a final series of experiments, by using the Gal4/UAS technique we expressed tetanus toxin in tyraminergic/octopaminergic neurons. This is genetic tool that impairs neurochemical transmission in the neurons of interest by blockade of synaptic vesicle docking (Sweeney et al. 1995). Exposure to volatilized nicotine in the tetanus toxin-expressing flies did not produce the nicotine-induced effect on the startle response, different from genetic controls where the nicotine effect was observed (Fig. 5b). Altogether, these data suggest that Oct mediates the nicotine-induced response.


By means of pharmacological and genetic tools we show for the first time that activation of α-bungarotoxin-sensitive nAChRs selectively induces the efflux of endogenous BAs from Drosophila brain, using our novel in vitro preparation. We further show that the BA whose release is stimulated by nAChR activation is Oct. Finally, we show that the nAChR regulation of Oct release is involved in the nicotine impairment of the mechanically induced startle response in Drosophila.

Previous studies by Jill Venton's group have shown that it is possible to measure the release of endogenous DA (Vickrey et al. 2009) and serotonin (Borue et al. 2009) using fast-scan cyclic voltammetry, in a fly larval ventral nerve cord preparation stimulated by optogenetic tools. Here, we evaluated the proposition that nAChR activation modulates the release of endogenous BAs in Drosophila adult brain as it does in vertebrate systems, using high-speed chronoamperometry. This technique allows measurements on the milisecond to second scale, making it appropriate to detect fast ΔBAefflux, as demonstrated in other systems elsewhere (e.g. Sabeti et al. 2002; Meza et al. 2012).

As we have the potential to detect different amines in our setup, it is interesting to note that in the mid-ventral side of the fly brain, the BA whose efflux is modulated by nAChR activation is Oct. As explained above, we positioned the electrode dorsal to the antennal lobe, on top of the brain region where the ellipsoid body is located. This is an area of the fly brain that belongs to the central complex, a structure associated to motor control in Drosophila (Ilius et al. 2007). Although it has been shown the occurrence of some DA processes and cell bodies in this fly brain area (Drobysheva et al. 2008), there is a strong octopaminergic innervation to this area (Busch et al. 2009), which suggests that Oct play a role in motor behavior regulation, proposition that has been demonstrated in mutants for the octopaminergic/tyraminergic neuronal pathways (Brembs et al. 2007). Thus, it is possible to propose that we placed the electrode on top of an Oct-enriched area, as opposed to DA-enriched areas that could be elsewhere in the fly brain. Therefore, these results show that the activation of nAChRs would be responsible for the modulation of BA efflux, in particular Oct.

Our results show a maximal Oct efflux of 4.4 ± 0.3 μM after PNU exposure. These values correspond to Oct overflow from the brain preparation and it should be expected that the actual concentration of Oct being released by nAChR activation in fly brain synapses would be several times higher. Several factors prevent us from getting an accurate estimation of the real amine concentration in this preparation (diffusion distance from the site of BA release to the electrode and BA re-uptake rate, among others). Nevertheless, our results agree with previous reports in vertebrate systems showing that an electrical pulse in the mesolimbic pathway can increase DA levels ranging from 0.54 ± 0.9 to 1.0 ± 0.6 μM in the nucleus accumbens of freely moving mice (Budygin et al. 2002).

One of the most active lines of research in the field is to try to characterize the nAChRs involved in the nicotine-induced effects, aiming at finding potential new therapies for nicotine abuse (Exley and Cragg 2008; Changeux 2010). Studies carried out in mammalian systems suggest that heteromeric α4-containing nAChRs and homomeric α7 receptors would be the main mediators of the behavioral effects induced by nicotine, explained in part by their localization in different brain neuronal types in specific brain regions and particular pharmacological properties (Albuquerque et al. 2009). For instance, it has been suggested that α4-containing nAChRs directly modulate the firing rate of midbrain dopaminergic neurons, while the homomeric, highly calcium-permeable α7 nAChRs indirectly induce the activation of these neurons by modulating excitatory glutamatergic inputs to these dopaminergic neurons (Albuquerque et al. 2009).

Little is known about the properties of nAChRs in Drosophila brain. Ten Drosophila nAChR subunits have been cloned: Seven α-type (Dα1/ALS, Dα2/SAD, Dα3, Dα4, Dα5, Dα6 and Dα7), and three β-type subunits (Dβ1/ARD, Dβ2/SBD and Dβ3) (Jones and Sattelle 2010). No combination of these 10 subunits has been shown to form a functional nAChR or even a high-affinity nicotine-binding site in heterologous expression systems. However, co-expression of Drosophila α with vertebrate β subunits generates a functional nAChR (Bertrand et al. 1994; Schulz et al. 1998; Lansdell and Millar 2000). These results are controversial as it is not known how vertebrate partners influence the properties of the invertebrate subunits. Moreover, isolation of native nAChRs by α-bungarotoxin column affinity, have shown receptor configurations containing both α and β type subunits (Chamaon et al. 2000, 2002; Lansdell and Millar 2002). In addition, at the cell level, it has been shown that most if not all the insect nAChRs would be calcium permeable and could mediate cellular plastic events (Goldberg et al. 1999; Campusano et al. 2007).

Our results show that the nicotine-induced response is blocked by α-bungarotoxin while is not affected by erysodine, antagonists of homo and heteromeric nAChRs in vertebrate systems, respectively. These data are consistent with previous reports suggesting that insect nAChRs show pharmacological properties resembling vertebrate homomeric receptors. Furthermore, PNU, a highly selective agonist of homomeric calcium-permeable nAChRs in vertebrate systems, also induces the release of BAs from fly brain, an effect blocked by α-bungarotoxin. The higher efficacy of the response induced by PNU compared with nicotine supports the notion that insect nAChRs behave as vertebrate homomeric nAChRs. Further experiments are required to obtain a better characterization of the functional properties of native fly nAChRs responsible for the regulation of Oct release.

With regard to the amine whose release is regulated by nAChR activation, Oct is considered the invertebrate homolog of vertebrate noradrenaline and is linked to the modulation of peripheral organs, sensory inputs and several complex behaviors including aggression and learning and memory (Roeder 2005). In this regard, different reports demonstrate that the activation of nAChRs induces the release of several BAs in vertebrate systems including noradrenaline (Albuquerque et al. 2009). In fact, it has been suggested that nicotine-induced noradrenaline release in the hippocampal region would be responsible for the mnemonic properties of this alkaloid (Azam and McIntosh 2006). Thus, our findings that nAChR activation modulates the release of Oct in fly brain are consistent with previous reports from other systems and further argue that Drosophila is a good system to evaluate the cellular and/or molecular mechanisms mediating this phenomenon and its potential physiological consequences.

A final interesting observation regarding the identification of Oct as the amine being released from the fly brain is the fact that it was shown previously that the mutation in Tβh causes an elevation in tyramine content in the fly (Monastirioti et al. 1996). Tyramine is the precursor for Oct synthesis and acts also as a neuroactive molecule in the fly brain. Therefore, it could have been predicted that in octopaminergic neurons depleted of Oct, tyramine would be released instead of octopamine. However, this does not seem to occur. There are several possible explanations for this observation. For instance, it is possible that Oct and tyramine are stored in different cell types or vesicle subpopulations and released under different conditions. Two splice variants for the fly vMAT gene have been reported in Drosophila. DA, serotonin and at least some octopaminergic neurons were associated to one of these isoforms, vMAT-A, while no association was evident between vMAT-A-positive vesicles and tyramine (Greer et al. 2005). This is consistent with previous data suggesting that Oct and tyramine are expressed in different neuronal populations and serve different functions (Nagaya et al. 2002; Lange 2009).

In addition to this, Devoto et al. (2012) recently reported that disulfiram, a drug that prevents DA conversion to noradrenaline by inhibiting the Dopamine-β-hydroxylase enzyme, potentiates DA release in the prefrontal cortex after a cocaine exposure. This would be consistent with the idea that when conversion to noradrenaline is impaired, the precursor in the synthesis, DA, is released instead. However, the authors studied the effects of these pharmacological manipulations in other brain regions and surprisingly, discovered that in the Nucleus Accumbens, the cocaine-induced DA release is not affected, even though the content of DA in this brain region is increased (Devoto et al. 2012). Therefore, these reports suggest that an increase in the tissue content of a neuroactive molecule is not sufficient to determine an increased amine release, but there are other factors that also play a role.

The demonstration that activation of nAChRs, the main responsible for fast excitatory neurotransmission in insects, is responsible for the release of Oct, a BA which has been associated to important events in fly physiology suggest that at least part of the behavioral effects induced by nicotine in flies would depend on octopaminergic signaling. This is a proposition that has not been assessed previously as far as we know and we decided to evaluate here. Our results show that the reduced startle response observed when flies are exposed to a specific nicotine concentration is not detected in animals deficient in either Oct synthesis or release. As it was previously reported that DA systems partially contributes to the behavioral effects induced by nicotine (Bainton et al. 2000), it is very likely that these two amines play a role in the modification of the motor programs associated to this behavioral response. This would be consistent with the literature in mammals showing that when vertebrates are exposed to drugs of abuse, different neurochemical systems are activated and would be responsible for the entire set of behaviors observed in these animals (Wonnacott 1997).

In summary, data presented here demonstrate that α-bungarotoxin-sensitive nAChRs modulate the release of Oct in fly brain, an effect that could be involved in the modulation of a specific nicotine-induced behavior in the fly. Because of the similarities between Drosophila and mammalian systems, we expect the knowledge acquired in this system, and particularly in this new preparation, could be useful in trying to understand the specific contribution of nAChRs to the behavioral effects in which these receptors are involved.


We thank all members of the Varas and Campusano labs for comments and suggestions. We also acknowledge the help of Mr Juan Pablo Soto in the generation of the graphical abstract. Erysodine was a kind gift of Dr. Patricio Iturriaga-Vasquez.

This study was supported by Fondo Nacional de DesarrolloCientífico y Tecnológico (FONDECYT) Grant Nº 1100965 (JMC), MSI 10-063-F (RV and JMC) and 10-035-F (RV) Grants, and VRI-PUC Grant (RV).

Authors declare no conflicting interests.