Skip to main content
NCBI home page
Proceedings of the National Academy of Sciences of the United States of America logoLink to Proceedings of the National Academy of Sciences of the United States of America
. 2013 Mar 25;110(15):6151–6156. doi: 10.1073/pnas.1303446110

Single dopaminergic neurons that modulate aggression in Drosophila

Olga V Alekseyenko 1,1, Yick-Bun Chan 1,1, Ran Li 1, Edward A Kravitz 1,2
PMCID: PMC3625311  PMID: 23530210

Abstract

Monoamines, including dopamine (DA), have been linked to aggression in various species. However, the precise role or roles served by the amine in aggression have been difficult to define because dopaminergic systems influence many behaviors, and all can be altered by changing the function of dopaminergic neurons. In the fruit fly, with the powerful genetic tools available, small subsets of brain cells can be reliably manipulated, offering enormous advantages for exploration of how and where amine neurons fit into the circuits involved with aggression. By combining the GAL4/upstream activating sequence (UAS) binary system with the Flippase (FLP) recombination technique, we were able to restrict the numbers of targeted DA neurons down to a single-cell level. To explore the function of these individual dopaminergic neurons, we inactivated them with the tetanus toxin light chain, a genetically encoded inhibitor of neurotransmitter release, or activated them with dTrpA1, a temperature-sensitive cation channel. We found two sets of dopaminergic neurons that modulate aggression, one from the T1 cluster and another from the PPM3 cluster. Both activation and inactivation of these neurons resulted in an increase in aggression. We demonstrate that the presynaptic terminals of the identified T1 and PPM3 dopaminergic neurons project to different parts of the central complex, overlapping with the receptor fields of DD2R and DopR DA receptor subtypes, respectively. These data suggest that the two types of dopaminergic neurons may influence aggression through interactions in the central complex region of the brain involving two different DA receptor subtypes.

Aggression is an innate behavior commonly used to obtain resources such as territories, mates, or food. Although some features of aggression are species-specific, broad similarities exist across species in the behavioral patterns and neurochemical systems involved (1, 2). Monoamines such as dopamine (DA) and serotonin have been linked to aggression in many species (3). Elaboration of the neural pathways that control aggression, however, has proven difficult because monoaminergic neurons also regulate other behaviors (1, 2). The pharmacological and genetic manipulations usually used for altering amine neuron function influence most of the neurons within a given monoaminergic population and thereby cause many behavioral changes. This makes it difficult to pinpoint the relationship between any particular amine-induced behavioral phenotype and an associated neuronal pathway.

The fruit fly offers enormous advantages in these regards because existing powerful genetic methods permit manipulation of small subsets of brain cells for exploration of the neural mechanisms of behavior (4–6). In the Drosophila model of aggression (7, 8), males fight and form stable hierarchical relationships. Using this model, we have shown that acute shutdown of serotonergic neurons yielded flies that could initiate fights but showed a reduced ability to escalate aggression. Activation of serotonergic neurons, by contrast, had an opposite effect (9). Acute shutdown of DA neurons, on the other hand, produced hyperactive flies that did not engage in social interactions, making it difficult to ask whether DA served any specific role in aggression. As in other species, dopaminergic neuron systems in fruit flies regulate a wide range of behaviors, including arousal (10, 11), courtship (12), memory (13), and locomotion (14). To ask whether small subsets of dopaminergic neurons might be involved in the regulation of aggression without affecting other behaviors, we restricted the numbers of targeted DA neurons using an intersectional approach in which the GAL4/upstream activating sequence (UAS) binary system was combined with the Flippase (FLP) recombination technique. For this purpose, we first generated and used a collection of enhancer-trap FLP lines in combination with a DA-specific tyrosine hydroxylase (TH)-GAL4 driver (10), aiming to express a GFP reporter in small subsets of DA neurons in an inheritable and reproducible way. Then, to explore the function of those neurons, a genetically encoded inhibitor of neurotransmitter release, the tetanus toxin light chain (TNT) (15), and the temperature-sensitive neuronal activator, dTrpA1 (16), were expressed instead of GFP. Recently, a similar experimental approach was used to subdivide fruitless (FruM)-expressing neurons into distinct classes (17) and to identify the particular Fru neurons involved in the control of courtship song (16). However, specific sets of aminergic neurons that modulate aggression and the circuitry they are involved with are unknown.

Here we identify two pairs of DA neurons that modulate aggression but have no major impact on other behaviors. We also show that a different subset of DA neurons affects movement and sleep but has no direct effects on aggression. We report that each pair of aggression-modulating DA neurons projects to different parts of the central complex. Neurons from the first pair are members of the PPM3 cluster: They innervate the fan-shaped body and the noduli where DopR subtype of DA receptors is expressed. Neurons from the second pair are T1 neurons that project to the protocerebral bridge, where their presynaptic arbors overlap with antibody staining for the DD2R subtype of DA receptors. These observations support the notion that small numbers of DA neurons can exert profound effects on behavior, in this case having modulatory actions on aggression in Drosophila, most likely through endings within the central complex region of the brain.

Results

Enhancer-Trap FLP Screen.

To reduce the total populations of DA neurons down to individual anatomically distinguishable neurons, we used an intersectional genetics approach. First, we generated and screened a collection of enhancer-trap (et) FLP recombinase transgenic lines that expressed the enzyme in different sets of neurons. A combination of the et-FLP lines with various GAL4 drivers and a UAS>stop>mCD8::GFP reporter allowed the expression of GFP only in cells that contained both the GAL4 driver and the FLP recombinase enzyme. We selected 36 lines with a reliable GFP signal driven by a panneuronal elav-GAL4 driver in the brain. Then, using the TH-Gal4 (10) driver, the GFP expression pattern was restricted further to subsets of DA neurons. We confirmed the amine specificity of the GFP signal by double staining with anti-TH antibodies. Twenty-five different FLP lines combined with the TH-Gal4 driver resulted in GFP expression in some populations of dopamine-positive neurons (Fig. 1A). By dissecting multiple brains for each line, we checked the reproducibility of the results and narrowed the numbers down further to 15 “dopaminergic” specific, highly reproducible lines. Some of these lines resulted in major overlap between GFP-positive neurons and the total populations of DA neurons (for example, line FLP383; Fig. 1 B and C). Only a few FLP lines resulted in highly restricted numbers of neurons that were positive for both GFP and anti-TH staining. For most of the following experiments, we selected three DA-specific (Fig. 1 D–F, Table S1) lines that targeted distinctive individual pairs of neurons in at least 70% of all brain preparations. A few other cells occasionally targeted by these lines were not labeled consistently under normal experimental conditions. Instead, they showed varying numbers of neurons in the indicated range in different brain preparations (Table S1).

Fig. 1.

Fig. 1.

Open in a new tab

Dopaminergic neurons identified by the et-FLP screen. (A) A schematic representation of the intersectional approach to restrict the numbers of targeted DA neurons. (B) Example of dopaminergic neuron-specific TH immunostaining pattern in Drosophila brain (the full z stack is shown). (Scale bar, 100 μm.) (C) Example of a line that targets most of the DA neurons in the fly brain. The GFP signal is driven by a combination of FLP383, TH-Gal4, and UAS>stop>GFP transgenes (the full z stack is shown). (Scale bar, 100 μm.) (D–F) Individual DA neurons targeted by the use of different et-FLP lines. The GFP signals restricted by et-FLPs are shown in green, neuropil areas stained by an nc82 neuronal marker are shown in gray, and anti-TH immunostaining is shown in magenta. Dotted boxes indicate the magnification fields shown in lower panels. Different frontal z stacks through either anterior or posterior areas of the same triple-stained brains were created when needed to demonstrate either processes or cell bodies. The full frontal z stacks are shown in Fig. S1. (Scale bars, 50 μm.) (D) FLP243 line restricts GFP expression to a bilateral pair of neurons from the T1 cluster (green) that arborize in the protocerebral bridge (gray) (also see Fig. S1A). (E) FLP447 line restricts GFP expression to one or two bilateral neurons from the PPM3 cluster (green) that arborize in the fan-shaped body and the noduli (gray) (also see Fig. S1B). (F) FLP346 line restricts GFP expression to two bilateral neurons from the PPL1 cluster (green) that arborize in the heel, lower stalk, and junction regions of the mushroom bodies (gray) (also see Fig. S1C).

Anatomical Characterization of Individual Dopaminergic Neurons.

To reveal the identity and the arborization patterns of individual neurons targeted by the different et-FLP lines, we used a triple-labeling procedure. Anti-TH antibodies were used to visualize the biosynthetic enzyme for the neurotransmitter, anti-CD8 (cluster of differentiation 8 transmembrain glycoprotein) antibody was used to amplify a membrane-tethered mCD8::GFP, and the nc82 antibody that recognizes the Drosophila protein Bruchpilot was used to label neuropil areas of the brain.

Line FLP243 targeted a pair of neurons, previously named T1 (18), that are located in a medial position near the esophageal foramen. These neurons arborize in the tritocerebrum, and also project to the protocerebral bridge of the central complex (Fig. 1D; full-brain z stack in Fig. S1A). Line FLP447 targeted one or two pairs of neurons from the PPM3 cluster that project to other parts of the central complex—the fan-shaped body and the noduli (Fig. 1E; full-brain z stack in Fig. S1B). Finally, FLP346 reliably labeled two pairs of neurons from the PPL1 cluster (Fig. 1F; full-brain z stack in Fig. S1C). As previously described (13), arbors from these PPL1 cells project to regions of the mushroom body: One neuron innervates the lower stalk and junction region, whereas the second neuron projects to the heel region.

Manipulation of Individual Dopaminergic Neurons Has Selective Effects on Aggression.

To ask whether inactivation of any of the targeted dopaminergic neurons had effects on behavior, we chronically silenced those neurons by expressing TNT (16), which cleaves the synaptic vesicle-associated protein synaptobrevin. In these experiments, socially naïve males of the same genotype were paired in multiwell plate aggression chambers.

Movement deficits.

To examine whether chronic neuronal silencing triggered changes in locomotion and general activity, we first measured the numbers of midline crossings for both paired flies during the first 5 min of aggression assays. TNT inactivation of large numbers of DA neurons (FLP383; Fig. 2A) led to severe movement deficits compared with control males lacking the FLP recombinase. Similar phenotypes were also observed in other screened lines that targeted many DA neurons. When single pairs of neurons were targeted, a locomotion phenotype varied. A slight movement deficit was found in flies with inactivated neurons from the PPL1 (FLP346) cluster (Fig. 2A), which was later confirmed by other behavioral tests (see below). Males with inactivated T1 (FLP243) or PPM3 (FLP447) neurons were not different from controls (Fig. 2A). To rule out the possibility that the FLP transgene insertion sites may have unintended effects on behavior, we examined flies carrying one copy of each et-FLP transgene but without a Gal4 driver and a UAS effector. None of these FLP lines crossed to wild-type Canton-S replicated the locomotion (Fig. 2B) or aggression (Fig. 2C) phenotypes seen with the full complement of transgenes (see below).

Fig. 2.

Fig. 2.

Open in a new tab

TNT inactivation of DA neurons has selective effects on behavior. (A) Inactivation of large numbers of DA neurons (light gray bar) results in very low levels of locomotion, whereas inactivation of individual DA neurons (dark gray bars) has either a small or no effect on locomotion. Data are presented as means ± SEM; ***P < 0.001 vs. corresponding control (white bar), analyzed by nonparametric two-independent-sample Mann–Whitney U test. (B) Males carrying one copy of the recombinase transgene from different et-FLP lines (light and dark gray bars) have the same levels of locomotion as wild-type Canton-S males (white bar). Data are presented as means ± SEM. (C) Numbers of lunges between pairs of males carrying one copy of various et-FLP transgenes. None of the transgenic et-FLP lines crossed to the wild-type Canton-S strain replicated the aggression phenotypes observed with experimental flies (Fig. 3). Each dot represents the lunge count for an individual pair of flies. The data are presented as boxplots with a median line. The lower and upper parts of the boxes are 25th and 75th percentiles, respectively. *P < 0.05 vs. Canton-S control (white dots), analyzed by nonparametric two-independent-sample Mann–Whitney U test.

Inhibition and activation of the dopaminergic T1 and PPM3 neurons enhance aggression.

Dopamine has been implicated in control of arousal and excitability in Drosophila (10, 11, 19), and our earlier studies demonstrated that acute shutdown of DA neurons produced hyperactive flies that did not engage in social interactions (9). Here we show that chronic silencing of large populations of DA neurons (FLP383; Fig. 3A) yields unhealthy flies, which have difficulty climbing or landing on the food cup in the fight chamber and, as a consequence, do not fight. However, inactivation of individual pairs of T1 (FLP243) and PPM3 (FLP447) neurons that project to the central complex did not produce any obvious activity deficits but did significantly increase the numbers of lunges (Fig. 3A), without changing the latency to the first lunge or the time to establish dominance (Fig. 3B). To confirm the specificity of the observed aggression phenotype, we next acutely activated the same targeted DA neurons using the UAS>stop>dTrpA1Myc transgene (16). dTrpA1 is a temperature-sensitive cation channel that can be expressed in neurons to selectively activate them. To ensure that the dTrpA1Myc transgene was expressed in the neurons of interest, after the aggression assays experimental fly brains were processed for anti-Myc staining. Induced activation of the T1 and PPM3 DA neurons by small temperature shifts also significantly increased the numbers of lunges and the latency to the first lunge (Fig. 3 C and D; FLP243 and FLP447). At the same time, the presence of these et-FLP transgenes by themselves did not result in any significant change in the numbers of lunges (Fig. 2C; FLP243 and FLP447).

Fig. 3.

Fig. 3.

Open in a new tab

Manipulation of individual T1 and PPM3 DA neurons targeted by FLP243 and FLP447 increases aggression. (A) Total numbers of lunges performed by pairs of males with TNT-inactivated subsets of DA neurons (chronic inactivation). Also see Fig. S3A for statistical analysis without outliers. (B) Latency to the first lunge and to the establishment of dominance in flies with TNT-inactivated subsets of DA neurons (chronic inactivation). (C) Total numbers of lunges performed by pairs of males with dTrpA1-activated subsets of DA neurons (acute, short-term activation). Also see Fig. S3B for statistical analysis without outliers. (D) Latency to the first lunge and to the establishment of dominance in flies with dTrpA1-activated subsets of DA neurons (acute, short-term activation). Each dot in A and C represents the lunge count for an individual pair of flies. Data are presented as boxplots with a median line. The lower and upper parts of the boxes show 25th and 75th percentiles, respectively. Latencies in B and D are presented as means ± SEM; *P < 0.05, ***P < 0.001 vs. corresponding control (white bar or white dots), analyzed by nonparametric two-independent-sample Mann–Whitney U test.

To rule out possible development-related differences caused by chronic TNT inactivation (Fig. 3A) and acute dTrpA1 activation (Fig. 3C), we measured aggression in flies with continuously activated T1 or PPM3 neurons. We used the same dTrpA1Myc strategy, but now the flies were raised at a constant +27 °C (Methods). Interestingly, chronic activation of T1 and PPM3 neurons also resulted in increased numbers of lunges (Figs. S2 and S3C). Inactivation or activation of the PPL1 neurons (FLP346) that project to the mushroom bodies had no effects on aggression (Fig. 3).

Effects of Single-Neuron Inactivation on Other Behaviors.

To determine whether the phenotypes observed by manipulating single pairs of DA neurons were specific for aggression, we examined courtship behavior and found no deficits in any of the lines tested (Table S2), although line FLP447 did show a small increase in its courtship vigor index. Next, we examined locomotor activity and sleep behavior using TriKinetics Activity Monitors and found no major defects in the lines that produced aggression phenotypes (Fig. S4). Interestingly, TNT inactivation of the PPL1 neurons (FLP346) that had no effect on aggression generated flies that slept more than controls (Fig. S4 B and C). These flies also had a small but significant negative geotaxis deficiency (Fig. S5).

Targets of the Dopaminergic T1 and PPM3 Neurons.

To identify putative presynaptic endings and dendritic receiving areas of the two pairs of DA neurons that modulate aggression, we used neuronal synaptobrevin:GFP UAS>stop>nsyb::GFP and Drosophila Down syndrome cell adhesion molecule:GFP UAS>stop>DsCam::GFP constructs (16). Both types of DA neurons displayed extensive arbors of presynaptic endings in distinct and different regions of the central complex. In contrast, the brightest labeling of dendritic DsCam::GFP puncta was observed outside of central complex structures. T1 neurons had extensive presynaptic arborizations in the protocerebral bridge (Fig. 4A, nsyb:GFP) and distinctive dendritic labeling in puncta in the tritocerebrum (Fig. 4A, DsCam:GFP). PPM3 neurons showed strong presynaptic nsyb:GFP labeling in the third layer (20) of the fan-shaped body and in the noduli (Fig. 4C, nsyb:GFP) and strong labeling of dendritic puncta along neuronal processes outside of central complex structures (Fig. 4C, DsCam:GFP). To examine DA-receptor expression patterns in the regions of presynaptic endings of T1 and PPM3 neurons, we counterstained the nsyb:GFP-labeled projection patterns of both types of neurons with antibodies to the DD2R (21) and DopR (22) subtypes of DA receptors. Anti-DD2R staining overlapped with the presynaptic terminals of T1 neurons in the protocerebral bridge (FLP243; Fig. 4B and Movie S1). In contrast, anti-DopR staining partially overlapped with the presynaptic terminals of PPM3 neurons in the fan-shaped body and noduli (FLP447; Fig. 4D and Movie S2). No overlap was observed between anti-DopR staining and the presynaptic projections of T1 neurons (Fig. S6A) or between anti-DD2R staining and PPM3 axonal projections (Fig. S6B). These results suggest that the two types of DA neurons might modulate aggression differentially by acting in distinct anatomical areas of the central complex, through different subtypes of DA receptors.

Fig. 4.

Fig. 4.

Open in a new tab

Putative targets of aggression-modulating dopaminergic neurons in the central complex. (A) The polarity of the dopaminergic T1 neurons. (Left) Arborization pattern of the T1 neurons visualized by membrane-bound CD8::GFP. (Center) Presynaptic terminals of the T1 neurons revealed by the presynaptic marker nsyb::GFP. (Right) The dendritic arbors of the T1 neurons visualized by expression of the postsynaptic marker DsCam:GFP. (B) An overlap between presynaptic terminals of the T1 dopaminergic neurons (green) and anti-DD2R antibody staining (magenta) in the protocerebral bridge region of the central complex (60× objective). See z stack in Movie S1. (C) The polarity of the dopaminergic PPM3 neurons. (Left) Arborization pattern of the PPM3 neurons visualized by membrane-bound CD8::GFP. (Center) Presynaptic terminals of the PPM3 neurons revealed by the presynaptic marker nsyb::GFP. (Right) The dendritic arbors of the PPM3 neurons visualized by expression of the postsynaptic marker DsCam:GFP. (D) A partial overlap between presynaptic terminals of the PPM3 dopaminergic neurons (green) and anti-DopR antibody staining (magenta) in the fan-shaped body and the noduli of the central complex (60× objective). See z stack in Movie S2. (Scale bars, 20 μm.)

DopR-receptor staining also was abundant in the mushroom bodies, where it overlapped with the PPL1 neuron terminals (FLP346) in the areas of the heel and the lower stalk (Fig. S6C). This supports the notion that the localization of receptor subtypes might help to explain the anatomical divergence of different behavioral phenotypes controlled by the same neuromodulator.

Discussion

Aggression in Drosophila is an innate behavior whose core circuitry is likely to be wired in the nervous system before eclosion. Appropriate displays of aggression rely on the correct identification of a potential competitor, an evaluation of the environmental signals, and the physiological state of the animal. With fixed numbers of neurons and neuronal circuits available, further flexibility in nervous system utilization is added by neuromodulators that can efficiently and reversibly reconfigure the function of networks without changing their “hardwiring.”

Dopamine, among other neuromodulators, is released by interneurons and acts at multiple sites within circuitries to alter the output of systems (23). Aminergic neurons in the fly nervous system display arbors that branch widely and cover multiple neuropil areas, through which they affect virtually all aspects of fly behavior. An open question remains whether individual neurons have selective actions on specific behavioral pathways or generalized actions on multiple behaviors. In this paper, we use an intersectional genetics approach to alter the function of single neurons. This allows us to ask whether individual DA neurons are involved in the regulation of aggression, where that regulation is exerted, and whether this is a selective action on aggression or these neurons modulate other behaviors as well.

Our previous attempt to examine the role of DA in aggression in Drosophila by acute shutdown of dopaminergic neurotransmission was inconclusive, because the flies became hyperactive and failed to engage in social interactions (9). A large and complex literature suggests that DA is important for arousal in Drosophila (11, 24), just as it is important for arousal in other species (25). In flies, dopaminergic modulation of arousal has been reported at “endogenous” levels as in sleep/wake daily cycles, and at “exogenous” stimulus-evoked levels as in higher-order complex behaviors. In some cases, the effects of altered dopaminergic function appear to be simple and linear; in other cases, the responses are distinctly nonlinear (11). In a recent study (24), elimination of one subtype of DA receptor in flies had opposite effects on sleep/wake cycles and on air puff-evoked startle responses. These traits were separately rescued by receptor replacement in different brain areas, leading the authors to propose that a segregation of brain pathways of arousal was likely involved. In the studies reported here, we searched for “arousal” effects of dopaminergic neurons on the sleep/wake cycle, movement, courtship, and aggression, but now at a single-neuron level.

The results with chronic inactivation of isolated dopaminergic neurons showed a clear separation of their effects on tested behaviors. To illustrate, inactivation of the pair of DA neurons from the T1 cluster that project to the protocerebral bridge (FLP243) yielded more aggressive flies that were not different from controls in their courtship behavior, sleep/wake activity, and locomotion. Inactivation of a pair of PPM3 neurons that innervate the fan-shaped body and noduli (FLP447) also increased aggression but, in addition, had small effects on the courtship vigor index and the average waking activity of flies but did not change their sleep patterns. Another DA neuron from the PPM3 cluster innervated the ellipsoid but not the fan-shaped body, and was reported to promote ethanol-induced locomotion (22). These data suggest that even within a single cluster, dopaminergic neurons can differ morphologically and functionally from each other. Finally, inactivation of a small number of PPL1 neurons that innervate selective regions of the mushroom bodies yielded flies with no aggression phenotype but with increased sleep, decreased locomotion, and lowered negative geotaxis responses. An overlap between DopR-receptor (22) immunostaining and the arborizations of the PPL1 neurons within the mushroom bodies (Fig. S5C) suggests that the observed effects on sleep and activity might be mediated, at least in part, via this receptor subtype. However, another pair of PPL1 neurons project to the dorsal part of the fan-shaped body, and these have been suggested to promote wakefulness through DopR-receptor subtypes (26). Another receptor subtype, DopR2 (27), is highly expressed in the mushroom bodies as well, and might also mediate the arousal phenotype of PPL1 neurons (27).

It is interesting that acute and chronic activation of T1 and PPM3 DA neurons via the dTrpA1 channel yield the same enhanced aggression phenotype as does chronic inactivation of these neurons with TNT. This suggests that a “U-shaped” relationship governs the action of DA on the circuits in which the amine functions to influence aggression. In mammalian systems, a model has been suggested for the relationship between D1-receptor stimulation and working memory performance, in which both sub- and supraoptimal activation of DA receptors impairs working memory function (28). In Drosophila, a similar effect has been reported with octopaminergic neurons (octopamine is the invertebrate analog of the catecholamine norepinephrine) and courtship behavior. In that example, both lowering and enhancing the function of octopaminergic neurons resulted in increased male–male courtship (29).

The results reported here suggest that the modulation of aggression by identified DA neurons may be mediated via at least two subtypes of DA receptors, DopR and DD2R, located within different parts of the central complex of the brain. Drosophila DopR receptors reportedly correspond to the postsynaptic D1-receptor type in mammals (30) and mediate responses to environmental stressors (24) and ethanol-induced hyperactivity (22). These receptors are abundant on neuronal processes within the fan-shaped body, noduli, and ellipsoid body of the central complex, where they appear to be in close contact with TH-positive neurons (31). Our results also show close proximity between DopR immunostaining and the sites of presynaptic arbors of targeted PPM3 within the fan-shaped body and the noduli (Fig. 4 C and D). The D2R (DD2R) receptors correspond to the D2 family in mammals (32) that is found at both pre- and postsynaptic locations. These are expressed in only a few cell bodies in the Drosophila brain, but in many neurons in the ventral nerve cord (21). Neurons bearing these receptors have been implicated in the control of locomotion in previous studies (21). DD2R antibody staining overlaps with the presynaptic arborization of T1 neurons in the protocerebral bridge region of the central complex (Fig. 4 A and B). Despite dense immunostaining of both presynaptic GFP and DD2R, the two types of neuronal endings appear to intermingle but not colocalize (Movie S1), suggesting that the DD2R receptors are postsynaptic to the dopaminergic nerve terminals of T1 neurons or are on presynaptic terminals of other neurons in the region. Further functional and morphological evidence will be required, however, to determine whether the processes of neurons expressing DopR and DD2R receptors within the central complex represent key synaptic linkages in the pathway of regulation of aggression.

Thus, modulation of higher-level aggression seems to include two morphologically distinguishable dopaminergic neurons whose endings are found within different neuroanatomical segments of the central complex. The proximity of dopaminergic endings originating from different types of neurons within one neuroanatomical region offers possible sites where DA neurons might interact to modulate the ability to escalate aggression. The details of how, where, or whether these particular DA neurons interact to exert their behavioral effects, however, remain to be established. The intersectional genetics approach in combination with the other binary systems available [Q (5); LexA systems (33)], and the methods designed to define synaptic interactions [GRASP (34)], should allow expansion of the analysis of complex behavioral circuits in fruit flies to single identified neuron levels of resolution.

Methods

Fly Stocks and Crosses.

The following fly lines were used in this study: TH-Gal4 from Serge Birman (Developmental Biology Institute of Marseille, Marseille, France); w1118, Canton-S, elav-Gal4, and y−w−; Δ2–3 from the Bloomington Stock Center; and UAS>stop>CD8::GFP, UAS>stop>TNT, UAS>stop>dTrpA1Myc, UAS>stop>nsyb::GFP, and UAS>stop>DsCam::GFP from Barry Dickson (the Research Institute of Molecular Pathology, Vienna, Austria). An et-FLP library was generated as described below. To obtain experimental flies, females carrying TH-Gal4 in combination with the corresponding UAS>stop>effector were crossed to the males of one of the et-FLP lines. For the genetic controls, the same genotype females carrying TH-Gal4 in combination with the corresponding UAS>stop>effector were crossed to w1118 males. In a second set of control experiments, Canton-S females were crossed to males of different et-FLP lines.

Generation of Enhancer-Trap FLP Library.

An enhancer-trap FLP insertion on chromosome I (a gift from Liqun Luo, Stanford University, Stanford, CA) was mobilized onto the autosome by crossing it to transposase line y−w−; Δ2–3. A total of ∼350 et-FLP lines balanced to either CyO or TM3 were generated. Individual et-FLP lines were then crossed to elav-Gal4; UAS>stop>mCD8::GFP. One- to 3-d-old male progeny were dissected and immunostained to check for consistent GFP expression in the nervous system.

Immunohistochemistry.

Adult male brains were dissected, fixed, and stained as described previously (35). The following primary antibodies were used: mouse anti-GFP (1:1,000) (Invitrogen), rat anti-mouse CD8a (1:100) (Invitrogen), rabbit anti-TH (1:500) (Novus Biologicals), mouse nc82 (1:20) (Developmental Studies Hybridoma Bank), rabbit anti-DD2R (1:1,000) (a gift from Isabelle Draper, Tufts Medical Center, Boston, MA), rabbit anti-DopR1 (1:1,250) (a gift from Fred Wolf, Ernest Gallo Clinic and Research Center, Emeryville, CA), and rabbit anti-Myc (1:4,000) (Abcam). The secondary antibodies used included Alexa Fluor 488-, Alexa Fluor 594-, and Alexa Fluor 647-conjugated cross-adsorbed antibodies (Invitrogen). Confocal z stacks were acquired using an Olympus FluoView FV1000 confocal microscope with a UAPO 20× air or 60× oil-immersion objective. Images were processed with ImageJ software (National Institutes of Health).

Behavioral Assays.

Flies were reared on a standard cornmeal medium at +25 °C and 50% relative humidity on a 12-h light/dark cycle. Socially naïve male flies were subjected to aggression, locomotion, courtship, or negative geotaxis assays within the first 1–2 h after lights-on. The activity and sleep of individual flies were recorded for 3 consecutive days using TriKinetics Drosophila Activity Monitors. A detailed description of behavioral assays is provided in SI Methods.

For UAS>stop>dTrpA1Myc data, flies were reared at +19 °C and transferred to a +27 °C experimental room 15 min before the aggression assay. For chronic dTrpA1 activation experiments, flies were reared and the aggression assay was performed at a constant +27 °C. At the completion of the assays, flies were recaptured and individual fly brains were collected and processed for anti-Myc staining.

Statistical Analyses.

Data were analyzed using the SPSS 16.0 for Mac statistical software package. Because most of the datasets did not pass the Shapiro–Wilk normality test, the nonparametric two-independent-sample Mann–Whitney U test for pairwise comparisons was used. Two-tailed P values were determined with the significance level set at *P < 0.05, **P < 0.01, ***P < 0.001. In cases where outlier data points were detected (Fig. 3 A and C and Fig. S2A), the outliers were excluded and the data analysis was run again to confirm that observed significant differences were not due to the outliers (Fig. S3).

Supplementary Material

Supporting Information
supp_110_15_6151__index.html (7.1KB, html)

Acknowledgments

We thank Drs. Serge Birman, Liqun Luo, and Barry Dickson for fly lines; Drs. Isabelle Draper and Fred Wolf for sharing dopamine-receptor antibodies; Drs. Jill Penn, Maria de la Paz Fernandez, Adelaine Leung, Sarah Certel, and Joanne Yew for helpful discussions; Drs. Alex Keene and Maria de la Paz Fernandez for their help setting up the activity monitors and analyzing sleep data; and Dr. Ravi Allada for kindly sharing the “Sleep Counting Macro” for sleep analysis. This research was supported National Institute of General Medical Sciences Grants GM0067645, GM074675, and GM099883 (to E.A.K.) and by a departmental National Institutes of Health training grant (to O.V.A.).

Footnotes

The authors declare no conflict of interest.

This article contains supporting information online at www.pnas.org/lookup/suppl/doi:10.1073/pnas.1303446110/-/DCSupplemental.

References

  • 1.Nelson RJ, Trainor BC. Neural mechanisms of aggression. Nat Rev Neurosci. 2007;8(7):536–546. doi: 10.1038/nrn2174. [DOI] [PubMed] [Google Scholar]
  • 2.Huber RH, Kravitz EA. Aggression: Towards an integration of gene, brain and behavior. In: Székely T, Moore AJ, Komdeur J, editors. Social Behaviour: Genes, Ecology and Evolution. Cambridge, UK: Cambridge Univ Press; 2010. pp. 165–180. [Google Scholar]
  • 3.de Almeida RM, Ferrari PF, Parmigiani S, Miczek KA. Escalated aggressive behavior: Dopamine, serotonin and GABA. Eur J Pharmacol. 2005;526(1–3):51–64. doi: 10.1016/j.ejphar.2005.10.004. [DOI] [PubMed] [Google Scholar]
  • 4.Venken KJ, Simpson JH, Bellen HJ. Genetic manipulation of genes and cells in the nervous system of the fruit fly. Neuron. 2011;72(2):202–230. doi: 10.1016/j.neuron.2011.09.021. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 5.Potter CJ, Tasic B, Russler EV, Liang L, Luo L. The Q system: A repressible binary system for transgene expression, lineage tracing, and mosaic analysis. Cell. 2010;141(3):536–548. doi: 10.1016/j.cell.2010.02.025. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 6.Bohm RA, et al. A genetic mosaic approach for neural circuit mapping in Drosophila. Proc Natl Acad Sci USA. 2010;107(37):16378–16383. doi: 10.1073/pnas.1004669107. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 7.Hoffmann AA, Parsons PA. Inter- and intraspecific variation in the response of Drosophila melanogaster and D. simulans to larval cues. Behav Genet. 1986;16(2):295–306. doi: 10.1007/BF01070805. [DOI] [PubMed] [Google Scholar]
  • 8.Chen S, Lee AY, Bowens NM, Huber R, Kravitz EA. Fighting fruit flies: A model system for the study of aggression. Proc Natl Acad Sci USA. 2002;99(8):5664–5668. doi: 10.1073/pnas.082102599. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 9.Alekseyenko OV, Lee C, Kravitz EA. Targeted manipulation of serotonergic neurotransmission affects the escalation of aggression in adult male Drosophila melanogaster. PLoS One. 2010;5(5):e10806. doi: 10.1371/journal.pone.0010806. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 10.Friggi-Grelin F, et al. Targeted gene expression in Drosophila dopaminergic cells using regulatory sequences from tyrosine hydroxylase. J Neurobiol. 2003;54(4):618–627. doi: 10.1002/neu.10185. [DOI] [PubMed] [Google Scholar]
  • 11.van Swinderen B, Andretic R. Dopamine in Drosophila: Setting arousal thresholds in a miniature brain. Proc Biol Sci. 2011;278(1707):906–913. doi: 10.1098/rspb.2010.2564. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 12.Liu T, et al. Increased dopamine level enhances male-male courtship in Drosophila. J Neurosci. 2008;28(21):5539–5546. doi: 10.1523/JNEUROSCI.5290-07.2008. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 13.Waddell S. Dopamine reveals neural circuit mechanisms of fly memory. Trends Neurosci. 2010;33(10):457–464. doi: 10.1016/j.tins.2010.07.001. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 14.Riemensperger T, et al. Behavioral consequences of dopamine deficiency in the Drosophila central nervous system. Proc Natl Acad Sci USA. 2011;108(2):834–839. doi: 10.1073/pnas.1010930108. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 15.Stockinger P, Kvitsiani D, Rotkopf S, Tirián L, Dickson BJ. Neural circuitry that governs Drosophila male courtship behavior. Cell. 2005;121(5):795–807. doi: 10.1016/j.cell.2005.04.026. [DOI] [PubMed] [Google Scholar]
  • 16.von Philipsborn AC, et al. Neuronal control of Drosophila courtship song. Neuron. 2011;69(3):509–522. doi: 10.1016/j.neuron.2011.01.011. [DOI] [PubMed] [Google Scholar]
  • 17.Yu JY, Kanai MI, Demir E, Jefferis GS, Dickson BJ. Cellular organization of the neural circuit that drives Drosophila courtship behavior. Curr Biol. 2010;20(18):1602–1614. doi: 10.1016/j.cub.2010.08.025. [DOI] [PubMed] [Google Scholar]
  • 18.Nässel DR, Elekes K. Aminergic neurons in the brain of blowflies and Drosophila: Dopamine- and tyrosine hydroxylase-immunoreactive neurons and their relationship with putative histaminergic neurons. Cell Tissue Res. 1992;267(1):147–167. doi: 10.1007/BF00318701. [DOI] [PubMed] [Google Scholar]
  • 19.Andretic R, van Swinderen B, Greenspan RJ. Dopaminergic modulation of arousal in Drosophila. Curr Biol. 2005;15(13):1165–1175. doi: 10.1016/j.cub.2005.05.025. [DOI] [PubMed] [Google Scholar]
  • 20.Kahsai L, Winther AM. Chemical neuroanatomy of the Drosophila central complex: Distribution of multiple neuropeptides in relation to neurotransmitters. J Comp Neurol. 2011;519(2):290–315. doi: 10.1002/cne.22520. [DOI] [PubMed] [Google Scholar]
  • 21.Draper I, Kurshan PT, McBride E, Jackson FR, Kopin AS. Locomotor activity is regulated by D2-like receptors in Drosophila: An anatomic and functional analysis. Dev Neurobiol. 2007;67(3):378–393. doi: 10.1002/dneu.20355. [DOI] [PubMed] [Google Scholar]
  • 22.Kong EC, et al. A pair of dopamine neurons target the D1-like dopamine receptor DopR in the central complex to promote ethanol-stimulated locomotion in Drosophila. PLoS One. 2010;5(4):e9954. doi: 10.1371/journal.pone.0009954. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 23.Birmingham JT, Tauck DL. Neuromodulation in invertebrate sensory systems: From biophysics to behavior. J Exp Biol. 2003;206(Pt 20):3541–3546. doi: 10.1242/jeb.00601. [DOI] [PubMed] [Google Scholar]
  • 24.Lebestky T, et al. Two different forms of arousal in Drosophila are oppositely regulated by the dopamine D1 receptor ortholog DopR via distinct neural circuits. Neuron. 2009;64(4):522–536. doi: 10.1016/j.neuron.2009.09.031. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 25.Robbins TW. Arousal systems and attentional processes. Biol Psychol. 1997;45(1–3):57–71. doi: 10.1016/s0301-0511(96)05222-2. [DOI] [PubMed] [Google Scholar]
  • 26.Liu Q, Liu S, Kodama L, Driscoll MR, Wu MN. Two dopaminergic neurons signal to the dorsal fan-shaped body to promote wakefulness in Drosophila. Curr Biol. 2012;22(22):2114–2123. doi: 10.1016/j.cub.2012.09.008. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 27.Han KA, Millar NS, Grotewiel MS, Davis RL. DAMB, a novel dopamine receptor expressed specifically in Drosophila mushroom bodies. Neuron. 1996;16(6):1127–1135. doi: 10.1016/s0896-6273(00)80139-7. [DOI] [PubMed] [Google Scholar]
  • 28.Goldman-Rakic PS, Muly EC, III, Williams GV. D(1) receptors in prefrontal cells and circuits. Brain Res Brain Res Rev. 2000;31(2–3):295–301. doi: 10.1016/s0165-0173(99)00045-4. [DOI] [PubMed] [Google Scholar]
  • 29.Certel SJ, et al. Octopamine neuromodulatory effects on a social behavior decision-making network in Drosophila males. PLoS One. 2010;5(10):e13248. doi: 10.1371/journal.pone.0013248. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 30.Sugamori KS, Demchyshyn LL, McConkey F, Forte MA, Niznik HB. A primordial dopamine D1-like adenylyl cyclase-linked receptor from Drosophila melanogaster displaying poor affinity for benzazepines. FEBS Lett. 1995;362(2):131–138. doi: 10.1016/0014-5793(95)00224-w. [DOI] [PubMed] [Google Scholar]
  • 31.Kahsai L, Carlsson MA, Winther AM, Nässel DR. Distribution of metabotropic receptors of serotonin, dopamine, GABA, glutamate, and short neuropeptide F in the central complex of Drosophila. Neuroscience. 2012;208:11–26. doi: 10.1016/j.neuroscience.2012.02.007. [DOI] [PubMed] [Google Scholar]
  • 32.Hearn MG, et al. A Drosophila dopamine 2-like receptor: Molecular characterization and identification of multiple alternatively spliced variants. Proc Natl Acad Sci USA. 2002;99(22):14554–14559. doi: 10.1073/pnas.202498299. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 33.Lai SL, Lee T. Genetic mosaic with dual binary transcriptional systems in Drosophila. Nat Neurosci. 2006;9(5):703–709. doi: 10.1038/nn1681. [DOI] [PubMed] [Google Scholar]
  • 34.Feinberg EH, et al. GFP Reconstitution Across Synaptic Partners (GRASP) defines cell contacts and synapses in living nervous systems. Neuron. 2008;57(3):353–363. doi: 10.1016/j.neuron.2007.11.030. [DOI] [PubMed] [Google Scholar]
  • 35.Certel SJ, Thor S. Specification of Drosophila motoneuron identity by the combinatorial action of POU and LIM-HD factors. Development. 2004;131(21):5429–5439. doi: 10.1242/dev.01418. [DOI] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Supporting Information
supp_110_15_6151__index.html (7.1KB, html)
1303446110_pnas.201303446SI.pdf (2.1MB, pdf)
Download video file (10.7MB, avi)
Download video file (13MB, avi)

Articles from Proceedings of the National Academy of Sciences of the United States of America are provided here courtesy of National Academy of Sciences

RESOURCES