Face validity describes how much a model “looks like” the disorder being modeled. In our case, face validity is the degree to which Drosophila recapitulate alcohol-induced behaviors seen in humans. Accordingly, Drosophila show behavioral and neurobiological responses to alcohol that are very consistent with humans and other mammalian species. These include locomotion changes, development of tolerance, learned preference, withdrawal symptoms, and effects on social behavior (Devineni and Heberlein, 2013). One feature of alcohol's neurobiological activity is the biphasic behavioral response: a period of nervous system stimulation followed by a period of nervous system depression. In the stimulatory phase, blood alcohol content rises, and an individual may experience disinhibition, euphoria, and hyperactivity (Fadda and Rossetti, 1998). Later, during the sedative phase, as blood alcohol content peaks, an individual becomes less active, experiencing motor and cognition impairment, and eventually coma and death (Fadda and Rossetti, 1998; Hendler et al., 2013). This biphasic action likely contributes to the development of alcohol dependence. The association of rising blood alcohol content with elevated mood during the stimulatory phase may positively reinforce alcohol drinking (Addicott et al., 2007). The biphasic alcohol response (see Figure 1) is also noted in Drosophila (Bainton et al., 2000; Singh and Heberlein, 2000). Upon exposure to ethanol vapor in a video tracking assay, flies show an initial peak of hyperactivity in response to the vapor lasting less than a minute, due to a sensory startle response to ethanol's odor. Following habituation to the odor, flies' locomotion level lessens compared to the startle response. As flies begin to absorb the ethanol and experience intoxication, their locomotion starts to increase again as a consequence of ethanol's pharmacodynamic action on the brain. With further exposure, locomotor activity begins to decline, and sedation takes effect, indicating that flies experience a biphasic ethanol response similar to mammals (Wolf et al., 2002).

Drosophila mirror other characteristics of the mammalian alcohol response, including tolerance, withdrawal, and reinforcing properties leading to learned preference for alcohol. Functional tolerance involves adaptations in neuronal activity following exposure to a psychoactive substance, rather than metabolic tolerance, which depends on changes to enzymatic metabolism of ethanol. As functional tolerance develops, a person (or fly) requires increasing amounts of alcohol to become intoxicated in the future (see Figure 2). Drosophila demonstrate functional tolerance in as little as 2 h after an initial alcohol exposure (Scholz et al., 2000). Flies and humans also have similarities regarding withdrawal from alcohol. Withdrawal causes a variety of psychological and physiological symptoms, and because drinking alcohol alleviates these symptoms, such attempts to curb withdrawal may contribute to the persistence of AUD (Schuckit, 2009). In Drosophila larvae and adults, alcohol withdrawal is associated with neuronal hyperexcitability, which also occurs in humans (Bayard et al., 2004; Cowmeadow et al., 2006; Ghezzi et al., 2014).

Additionally, like humans, flies can develop a learned preference for alcohol (see Figure 3). Similar to mammals, Drosophila do not have an innate preference for alcohol. Upon a first offer of ethanol for consumption, flies are either indifferent or avoidant, depending on the exact presentation parameters (Devineni and Heberlein, 2009; Peru y Colón de Portugal et al., 2014). However, flies develop persistent, experience-dependent preference following exposure to alcohol (Peru y Colón de Portugal et al., 2014). Flies' acquisition of preference for alcohol is a critical component of their usefulness as an animal model. Humans similarly develop alcohol preference that can drive problematic drinking behavior and lead to AUD (Fadda and Rossetti, 1998). In mammals, preference frequently becomes attached to specific contexts or patterns, a phenomenon examined in a conditioned place preference (CPP) assay, wherein a particular environmental context gains attractive qualities after repeated pairing with a drug (Cunningham et al., 2006). Similar behavioral reinforcement occurs in flies, when they acquire preference for an innocuous odor that has been paired with alcohol vapor (Kaun et al., 2011).

Mechanistic validity refers to the consistency of neurobiological mechanisms and molecules underlying alcohol response between Drosophila and humans. Historically, much of the complexity of studying alcohol lies in its widespread effects throughout the brain. While some drugs, like cocaine, act primarily through a single mechanism (blocking monoamine reuptake into the presynaptic terminal, in cocaine's case) (Hummel and Unterwald, 2002), alcohol's effects on neurotransmission occur in a dose-dependent manner that involves diverse effects across neurotransmitters. Ethanol easily crosses the blood-brain barrier and acts much more globally than other drugs, meaning its mechanisms of action occur quickly and efficiently, contributing to alcohol's propensity for abuse. These diverse mechanisms of alcohol action are consistent from flies to humans.

Many of the genes implicated in mammalian alcohol reactions and human AUD have conserved functions in Drosophila (Grotewiel and Bettinger, 2015; Lathen et al., 2020). Some of these involve common molecular pathways for alcohol response, such as cyclic adenosine monophosphate (cAMP) (Moore et al., 1998) and neuropeptide Y (neuropeptide F in flies) (Wen et al., 2005). There is also a high level of conservation in neurotransmitter systems between Drosophila and vertebrate species, including humans. While behavioral outcomes may differ, vertebrates and invertebrates share signaling by glutamate, gamma-aminobutyric acid (GABA), acetylcholine (ACh), glycine, dopamine (DA), serotonin (5HT), and numerous neuropeptides (Deng et al., 2019). Although in flies there is no evidence of adrenergic signaling, octopamine (OA), and tyramine (TA) fulfill behavioral roles similar to norepinephrine (NE) and epinephrine in mammals.

In mammals, several neural circuits are involved in behavioral responses to alcohol and the development of AUD [see Abrahao et al. (2017) for a recent review]. For example, in the ventral tegmental area of the mammalian brain, the mesolimbic dopamine pathway is involved in mediating reinforcement and communicating with the nucleus accumbens to drive reward signaling. The fly brain's neuroanatomical structure differs from mammals', but both have circuits implicated in specific behaviors. There are several anatomical regions of interest in the discussion of alcohol-related behaviors. These include the mushroom bodies (MB), a center for associative learning, the antennal lobe (AL), the primary olfactory processing center, and the central complex, which houses the fan-shaped body (FSB), a center of higher integration, and ellipsoid body (EB), a pre-motor structure. Like in mammalian neural circuits, inputs to these brain regions by specific neurotransmitters mediate various alcohol-induced behavioral responses. These are most well-studied for dopamine.

In Drosophila, there are many tools available for genetic manipulations, including forward genetics (going from a phenotype to a causative gene), reverse genetics (going from a targeted mutated gene, using a system like CRISPR/Cas9 mutagenesis, to a phenotype), and genomic approaches (Griffiths et al., 2000). One reverse genetics approach is the GAL4/UAS system (Brand and Perrimon, 1993). This approach involves one transgene carrying GAL4 (a transcriptional activator in yeast), which is under the control of a specific promoter determining the spatial and temporal expression of GAL4. This is combined with a second effector transgene under the control of the upstream activating sequence (UAS), where GAL4 binds. A plethora of distinct GAL4 lines exist, including thousands that drive GAL4 expression in different subsets of neurons (Jenett et al., 2012). The effector transgenes include cDNAs for overexpression, RNAi for gene knockdown, or tools to activate and silence neurons under experimenter control. In conjunction with these genetic tools, Drosophila's relatively low cost, ease of maintenance, and short generation time make it amenable to a variety of experimental manipulations.

Much of our knowledge about the effects of alcohol on neurotransmitters comes from studies in mammalian models, particularly rodents. Briefly, we will discuss these findings as a point of comparison with Drosophila. As mentioned in the introduction, alcohol exerts action by taking advantage of existing biological pathways, necessitating the study of alcohol in the context of known impacts on these pathways and subsequent behavioral alterations. Alcohol's effects on neurotransmission occur in a dose-dependent manner, differentially impacting neurotransmitter systems (Hummel and Unterwald, 2002). Ethanol acts quickly, efficiently, and globally. As discussed in relation to flies' face validity, alcohol's effects are biphasic: initial low doses produce euphoria and hyperactivity, while over time, higher doses depress activity and eventually lead to sedation (Carlsson et al., 1972; Pohorecky, 1977).

The two phases of the alcohol response involve different neurotransmitter systems Figure 1B. At low doses, alcohol acts as a stimulant, causing disinhibition, euphoria, and hyperactivity as blood alcohol content rises (Fadda and Rossetti, 1998). Shortly after ingesting alcohol, mice show a sharp increase in locomotion, attributed to DAergic activation (Carlsson and Lindqvist, 1973). Specifically, these behaviors arise from increased release of DA in the brain's reward system, a mechanism demonstrated in rodents (Yim et al., 1998) as well as humans (Boileau et al., 2003). Due to its involvement in reward processing, DA contributes to both the development and persistence of alcohol dependence (Di Chiara, 1995). In rats, even very small amounts of alcohol administered intravenously increase DA levels in the brain's reward centers and contribute to sustained alcohol self-administration (Lyness and Smith, 1992). Rewarding stimuli are processed via DAergic signaling in the ventral tegmental area (VTA) and nucleus accumbens (NAc). Low doses of alcohol cause dose-dependent activation of DAergic neurons in the rat VTA (Gessa et al., 1985), and alcohol acutely increases synaptic DA levels throughout the reward system, but particularly in the NAc (Di Chiara and Imperato, 1988). Alcohol reduces the activity of GABA in the VTA, thereby disinhibiting DAergic neurons and increasing DAergic activity (Kohl et al., 1998). 5HT is also involved in behavioral regulation, including in brain regions responsible for reward processing, which are implicated in AUD. In humans, 5HT metabolites are more plentiful in blood and urine after drinking alcohol, indicating increased serotonergic transmission, and alcohol consumption increases brain levels of 5HT in animal models (LeMarquand et al., 1994a,b). Additionally, 5HT_1B (5HT receptor) knockout mice show less ethanol-induced locomotor impairment, indicative of intoxication, across 11 days of ethanol feeding and testing. Therefore, 5HT may have a role in exacerbating the effects of alcohol and in determining alcohol sensitivity (Crabbe et al., 1996). Serotonergic signaling has particular clinical significance due to the comorbidity of AUD with anxiety and mood disorders, which are often treated with selective serotonin reuptake inhibitor (SSRI) drugs (Gimeno et al., 2017).

As alcohol consumption continues, blood alcohol content peaks, and behaviors associated with CNS depression occur. In the sedative phase, alcohol primarily exerts depressant effects by suppressing excitatory neurotransmission and heightening inhibitory neurotransmission. Alcohol activation of GABA_A receptors produces cell hyperpolarization via an influx of chloride ions. Co-administration of ethanol and GABA-mimetic drugs, such as baclofen, enhances the sedative effects of alcohol. Similar experiments with GABA antagonists, such as picrotoxin, reduce alcohol-related incoordination (Martz et al., 1983).

Along with enhancing inhibition, alcohol also suppresses excitation. Beginning in the 1980s, researchers investigated the impact of alcohol on glutamate receptors, showing that even small amounts of alcohol could suppress ion flow through N-methyl-D-aspartate (NMDA) receptors in cultured rat neurons (Lovinger et al., 1989). Alcohol limits the NMDA-mediated release of neurotransmitters like DA, NE, and ACh, further impairing communication between neurons (Göthert and Fink, 1989; Woodward and Gonzales, 1990). These findings provide a starting point for understanding neurotransmitters' involvement in behavioral responses to alcohol. However, genetic manipulations necessary for greater mechanistic insight are more limited in mammalian models than invertebrates. Therefore, Drosophila are an appealing candidate for probing this relationship in greater detail.

Drosophila are a useful organism for the study of neurotransmitters because, as described, neurotransmitters are well-conserved from flies to mammals, and they often exert similar effects on behavior. These behavioral effects are particularly useful when considering the effects of alcohol since there is no unique neurobiological pathway for alcohol. However, alcohol has known effects on neurotransmitters that are associated with changes in behavior. See Table 2 for a summary of neurotransmitter roles in alcohol-related behaviors. Additionally, flies have over 40 neuropeptides and signaling hormones, many of which are shared with vertebrate species (Hewes and Taghert, 2001). The best-studied in the context of alcohol is Neuropeptide F (NPF). NPF has a role in Drosophila alcohol-related behaviors such as consumption, conditioned preference (Shohat-Ophir et al., 2012; Bozler et al., 2019), and preference for egg-laying in alcohol-containing food (Kacsoh et al., 2013). However, here we focus on small molecule neurotransmitters.

DA is produced by the metabolism of essential amino acid phenylalanine or its metabolite, non-essential amino acid tyrosine. Dietary ingestion is the primary source of both phenylalanine and tyrosine. Tyrosine is converted by tyrosine hydroxylase (TH) to L-DOPA, which is then converted to DA by dopamine decarboxylase (Cole et al., 2005). There are two classes of DA receptors, classified due to their similarities to mammalian DA receptors: D1-like and D2-like. There are two D1-like receptors, Dop1R1, which signals via Gαs to stimulate cAMP production, and Dop1R2, which couples to Gαq to increase cytosolic calcium (Handler et al., 2019). The D2-like receptor, Dop2R, inhibits the adenylyl cyclase/cAMP pathway (Scholz-Kornehl and Schwärzel, 2016). Both classes are G protein-coupled receptors (GPCRs) with seven transmembrane domains (Karam et al., 2020). Flies also have DopEcR, a G protein-coupled DA/ecdysteroid receptor that can be activated by either DA or the insect hormone ecdysone (Srivastava et al., 2005). Much of the structure of these receptors is conserved between vertebrates and Drosophila (Karam et al., 2020). After the presynaptic neuron releases DA into the synapse and DAergic action occurs, the dopamine transporter (DAT) takes DA back up into the neuron.

DA has numerous important functions in Drosophila behavior, in part due to the widespread projection of DAergic neurons throughout the brain. Circuits responsible for the alcohol response involve anatomical regions such as the MB and central complex, specifically the EB. There is an extensive body of research on DA's roles in fly behavior, so we will focus here on the behaviors and neural circuitry most relevant for alcohol, namely locomotion, which involves the central complex, and learning and memory, punishment, and reward, which involve the MB. DAergic inputs to the central complex mediate motor activity and sleep (Strausfeld and Hirth, 2013), multisensory processing (Wolff and Rubin, 2018), and social behaviors like aggression (Alekseyenko et al., 2013). See Table 1 for a summary of DA roles in fly behavior.

Like in mammals, alcohol impacts the Drosophila DAergic system (Bainton et al., 2000), and many DA-related behaviors are linked to and affected by alcohol. Alcohol affects several DA-mediated behaviors in Drosophila, such as locomotion, sedation, and reward. Additionally, DA has an important role in flies' preference for laying eggs in ethanol-containing food. Subsets of competing DA neurons enhance or inhibit this preference (Azanchi et al., 2013), possibly suggesting a DAergic role for flies' innate attraction to alcohol's odor at low concentrations (Ogueta et al., 2010). Significantly, alcohol potentiates the release of DA in the fly brain, which may explain the noted enhancement of locomotion and reinforcing behavioral effects following ethanol exposure (Ojelade et al., 2019).

Like DA, TA and OA are produced by the metabolism of essential amino acid phenylalanine or its metabolite, non-essential amino acid tyrosine, both found in food. Tyrosine is converted to TA by tyrosine decarboxylase (Tdc), and TA is converted to OA by tyramine beta-hydroxylase (Tbh). Manipulating Tbh concentration alters the OA/TA equilibrium: Tbh-null flies have increased TA levels and decreased OA levels (Monastirioti et al., 1996). There are 4 GPCRs for OA and 3 for TA. All seven types show high expression in the brain (El-Kholy et al., 2015). Based on parallels with the vertebrate adrenergic system, fly OA receptor classifications include α-adrenergic-like, β-adrenergic-like, and OA/TA or TA receptors (Evans and Maqueira, 2005). These receptors exert a variety of effects. Activation of the α-adrenergic-like receptor leads to elevation of calcium ions, and activation of the three β-adrenergic-like receptors increases intracellular cAMP levels (Balfanz et al., 2005; Maqueira et al., 2005). Interestingly, though researchers have described a specific OA transporter in many insects, one has not been identified in Drosophila. OA reuptake in flies may occur via DAT, although this requires further investigation (Arancibia et al., 2019).

TA and OA influence a wide variety of behaviors. Both neurotransmitters were initially of interest to researchers because they are critical for insect physiological processes like modulation of organs and muscles, and since vertebrates lack receptors for both, they provided a potential target for insecticides (Roeder, 2005). Although TA was historically thought to function primarily as a precursor of OA and exert few of its own effects, the presence of TA-activated GPCRs suggests that it may function independently as a neurotransmitter (Borowsky et al., 2001), and TA has been independently implicated in some behaviors. See Table 1 for a summary of OA and TA roles in fly behavior.

Although research has historically focused more on DA, OA, and TA are also involved in behavioral responses to alcohol. OA and TA are implicated in ethanol-related behaviors such as locomotion, sensitivity, tolerance, preference, and olfactory attraction. Investigations often explore TA and OA in conjunction due to their common precursor, and it is not clear whether there are distinct TAergic and OAergic neurons.

Unlike the previously discussed neurotransmitters, 5HT does not originate from the amino acid tyrosine, but tryptophan. The precursor tryptophan, absorbed in the diet, is converted to 5-hydroxytryptophan (5-HTP) by tryptophan hydroxylase. Then, aromatic amino acid decarboxylase converts 5-HTP to 5-hydroxytryptamine (otherwise known as serotonin or 5HT) (Coleman and Neckameyer, 2005). Flies have five different G protein-coupled 5HT receptors. 5-HT1A, 5-HT1B, and 5-HT7 are all coupled to the cAMP signaling cascade, while 5-HT2A and 5-HT2B activation lead to Ca²⁺ signaling (Blenau et al., 2017). Some of these receptor subtypes are involved in specific outcomes, like the role of the 5-HT2B receptor in minimizing anxiety-like behaviors (Mohammad et al., 2016). 5HT is removed from the synapse via reuptake by the Drosophila serotonin transporter (SerT) (Demchyshyn et al., 1994). SerT colocalizes with 5HT neurons throughout the brain (Giang et al., 2011), and its mutations provide a useful tool for investigating phenotypic outcomes of serotonergic signaling.

5HT affects numerous behaviors in Drosophila, and manipulation of the fly serotonergic system has recapitulated symptoms of neuropsychiatric disorders like depression and anxiety (Ries et al., 2017). Importantly for consideration of alcohol-related behaviors, 5HT is critical for memory formation in Drosophila (Sitaraman et al., 2008). Memory performance worsened by genetically blocking serotonergic neurotransmission during a task for learned avoidance of high temperatures. A similar result was noted upon the pharmacological blockage of 5HT (Sitaraman et al., 2008). See Table 1 for a summary of 5HT roles in fly behavior.

Although historically not the subject of intense research efforts, new evidence increasingly supports a role for 5HT in Drosophila's ethanol-related behaviors. These behaviors include olfactory attraction, preference, and sensitivity.

In flies, GABA is synthesized by glutamic acid decarboxylase (GAD) enzymes, including Gad1 (expressed exclusively in the nervous system) and Gad2 (expressed exclusively in glia) (Manev and Dzitoyeva, 2010). GAD is implicated in the formation of synapses at the neuromuscular junction (NMJ) and may also be involved in local regulation of glutamate at NMJ synapses (Featherstone et al., 2000). Researchers have also mapped the expression of Gad1 and Gad2 within the fly brain and found that while only a few neurons release GABA, most of the neurons in the antennal lobe receive inhibitory signals (Okada et al., 2009). GABA exerts action on both ionotropic receptors and GPCRs: ligand-gated GABA_A-type receptors and metabotropic GABA_B-type receptors (Hosie et al., 1997). Flies have subtypes of both of these receptors, which are experimentally useful in their sensitivities to different pharmacological manipulations. For example, RDL receptors (GABA_A-type), named for resistance to the insecticide dieldrin (RDL), are highly distributed in the insect CNS and are therefore the target of numerous insecticides (McGonigle and Lummis, 2009). Importantly, fly GABA receptors do not respond to pharmacological agents the same way that mammalian GABA receptors do, so this will be important to consider when evaluating the relationship between alcohol and the fly GABAergic system. GABA action terminates in the synapse through a variety of mechanisms, such as changes in the density of GABA receptors or GABA uptake by astrocytic GABA transporters (GATs) (Muthukumar et al., 2014). The vesicular GABA transporter (VGAT), located pre-synaptically in GABAergic neurons, packages GABA into synaptic vesicles for later release (Enell et al., 2007).

In vertebrates and invertebrates, GABA activity impacts numerous behaviors since it is highly expressed in different types of neurons throughout the brain. In Drosophila, these behaviors include locomotion (Leal and Neckameyer, 2002), olfactory learning (Liu et al., 2007), and sleep regulation (Agosto et al., 2008). See Table 1 for a summary of GABA roles in fly behavior.

GABA is ubiquitously expressed and involved in regulating numerous behaviors, making it a good target for alcohol, which acts in a widespread, non-selective manner throughout the brain. Alcohol increases GABA release in vertebrates, suggesting a possible role in flies (Kelm et al., 2011). Researchers have particularly focused on the role of metabotropic GABA_B receptors in alcohol-related behaviors such as sensitivity, tolerance, and locomotion.

ACh is derived from choline, which flies ingest through the diet. Choline acetyltransferase (ChAT) catalyzes ACh biosynthesis, and acetylcholinesterase (Ace) breaks down ACh. There are two categories of ACh receptors: ionotropic (nicotinic) and metabotropic (muscarinic). Nicotinic ACh receptors (nAChRs) in Drosophila mediate fast, excitatory synaptic currents (Su and O'Dowd, 2003). Muscarinic ACh receptors (mAChRs) are not as well-understood, although researchers have identified three types (A, B, and C) that signal via activation of different G-protein subunits to initiate various downstream intracellular processes (Collin et al., 2013; Ren et al., 2015). Once synthesized presynaptically, ACh must be loaded into vesicles by the vesicular ACh transporter (VAChT) (Kitamoto et al., 1998). While AChE normally terminates ACh action in the synaptic cleft, some drugs prevent this process. Inhibiting AChE is lethal, so irreversible AChE inhibitor compounds are extremely toxic and often used as insecticides (Menozzi et al., 2004). Irreversible AChE inhibitors are also lethal to humans, although reversible AChE inhibitors have some therapeutic applications, like as pharmacological treatments for neurodegenerative diseases, including Alzheimer's (Colović et al., 2013).

Although we know relatively little about cholinergic effects on Drosophila behavior, olfactory associative learning is one area of investigation. Silva et al. examined the role of mAChR-A in aversive learning. Researchers generated fly lines to visualize mAChR-A with GFP, and they confirmed mAChR-A expression in the MBs (Silva et al., 2015). By disrupting mAChR-A pharmacologically or genetically, they significantly impaired the formation of aversive olfactory memory in Drosophila larvae and adult flies (Silva et al., 2015). However, when flies received a more intense shock during training (90V compared to 50V), learning was not impacted. Thus, mAChR-A may only reinforce moderate aversion (Bielopolski et al., 2019). These effects were localized to mAChR-A activity in the adult gamma Kenyon cells, showing that aversive olfactory learning and short-term memory require mAChRs (Bielopolski et al., 2019).

Investigators also examined olfactory learning in ionotropic AChRs. Mutant flies with disrupted nAChRs in the MBONs showed a reversal in odor driven behavior. They approached an aversive odor, establishing a role for nAChR subunits in the MBONs in olfactory behaviors (Barnstedt et al., 2016). In another experiment on naïve avoidance utilizing the same aversive odor, knockdown of mAChR-A did not impact naïve avoidance (Bielopolski et al., 2019), suggesting that ionotropic receptors specifically mediate this behavior.

Although little research exists regarding ACh's role in alcohol-related behaviors in Drosophila, other drugs of abuse have been examined, such as nicotine. Although nicotine's mechanisms of action and behavioral outcomes differ from those of alcohol, both are associated with a period of elevated mood and increased activity at low doses and aversive effects at higher doses (Little, 2000). Additionally, in humans, use of both substances may arise for similar reasons and share similar patterns of use and abuse (Little, 2000). Nicotine has known effects on neurotransmitter systems. When flies are exposed during development, there are fewer TH-positive neurons in the PPM3 cluster of the adult brain, suggesting dopaminergic impacts (Morris et al., 2018). Nicotine exerts direct effects on the cholinergic system by activating nAChRs and producing fly behavioral responses such as hyperactivity (Ren et al., 2012), disrupted geotaxis (King et al., 2011), and loss of startle response (Fuenzalida–Uribe et al., 2013). These behavioral alterations are similar to those we have discussed in reference to alcohol. Also, in probing these effects at various developmental stages for Drosophila, research shows that fly responses to nicotine are comparable to mammals throughout development (Velazquez-Ulloa, 2017). Therefore, nicotine research in Drosophila may provide a useful point of reference for future explorations of alcohol and ACh.

Because it is the primary excitatory neurotransmitter, ACh activation drives the downstream release of neuromodulators like DA and OA. nAChR activation with pharmacological agents led to a rapid, dose-dependent release of OA (Fuenzalida–Uribe et al., 2013). In a startle-induced negative geotaxis assay, flies exposed to nicotine do not rapidly climb up the vial following a mechanical disruption that taps them to the bottom of the tube. However, disrupting OA transmission abolished the nicotine-induced impairment of flies' startle response, returning negative geotaxis to normal. Therefore, the behavioral response to drugs like nicotine involves AChR-induced OA release (Fuenzalida–Uribe et al., 2013).

Glutamate is an amino acid that is produced by neuron-glia interactions in the glutamate-glutamine cycle, which involves the enzymes glutamate dehydrogenase (Gdh) and glutamine synthetase (GS) (Vernizzi et al., 2019). Gdh converts glutamate to alpha-ketoglutarate and ammonia (Plaitakis et al., 2017), and GS is a cytosolic enzyme that produces glutamine (Spodenkiewicz et al., 2016). Cytosolic glutamate is a precursor in the synthesis of GABA (Daniels et al., 2008). The fly genome contains 30 iGluR subunits (Littleton and Ganetzky, 2000), one metabotropic receptor (Mitri et al., 2004), and one glutamate-gated chloride channel (Cully et al., 1996), suggesting that glutamate can exert numerous effects in the fly brain. The vesicular glutamate transporter (VGlut) fills synaptic vesicles with glutamate. There is a single Drosophila VGlut found on synaptic vesicles at the NMJ at synapses on motoneurons and interneurons throughout the CNS (Daniels et al., 2004).

Because glutamate in isolation has historically been challenging to study outside the NMJ, it is not well-understood what behavioral effects glutamate is uniquely involved in regulating. Glutamatergic activity is better characterized in vertebrates, and these findings provide foundations for studying the role of glutamate in behavior for Drosophila as well. In vertebrates, much of our understanding of glutamatergic activity comes from understanding the action of the three iGluRs: AMPA, kainate, and NMDA receptors (Traynelis et al., 2010). Although Drosophila iGluRs share sequence similarity with the vertebrate receptors, subsequent Investigations have been somewhat limited because invertebrate neurons are small and challenging to access, complicating further characterization of these receptors' functions (Li et al., 2016a).

In mammals, glutamatergic activation likely has circadian fluctuations (Prosser, 2001), which is also true for flies (Zimmerman et al., 2017). Reducing glutamatergic release from glutamatergic neurons decreased wakefulness, and increased glutamatergic activity promoted wakefulness (Zimmerman et al., 2017). These results suggest that glutamate is wake-active in Drosophila (Zimmerman et al., 2017). Optogenetic studies indicate that the dorsal population of circadian clock neurons use glutamate as an inhibitory transmitter to promote sleep. This signaling may play a particularly significant role in daytime sleep (Guo et al., 2016). See Table 1 for a summary of glutamate roles in fly behavior.

Glutamate likely plays an important role in the antennal lobe. The AL is known to contain both ionotropic and metabotropic glutamate receptors, including the fly homolog of the NMDA receptor, Nmdar, which is thought to have a conserved function in synaptic plasticity and, therefore, olfactory learning and memory (Xia et al., 2005). In the fly olfactory circuit, glutamate is a neurotransmitter with inhibitory effects, primarily influencing the response that projection neurons in the AL have to olfactory stimuli (Liu and Wilson, 2013). Interrupting glutamatergic transmission by expressing an RNAi transgene for one of the NMDA receptor subunits, Nmdar1, reduces Nmdar1 receptor subunit levels. Reducing Nmdar1 activity in a subset of AL projection neurons responsive to a specific odor blocked short and long-term olfactory habituation to that specified odor without impacting habituation to other odors (Das et al., 2011). Additionally, RNAi knockdown of Gad1 or VGlut in AL local interneurons blocks short and long-term olfactory habituation, suggesting that both GABAergic and glutamatergic activity in local interneurons are important for habituation (Das et al., 2011).

There is little research focused specifically on glutamate's role in ethanol-related behaviors in Drosophila. In olfactory receptor neurons of the AL, a single alcohol exposure induces excitotoxic cell death via glycogen synthase kinase-3 beta and NMDA receptors, suggesting a glutamatergic role for alcohol-induced neural deficits in the fly (French and Heberlein, 2009). Some recent evidence suggests that a circuit involving DAergic modulation of glutamatergic MBONs is involved in consolidation and expression for alcohol-associated memories (Scaplen et al., 2020). Two DAergic projections are to glutamatergic MBONs implicated in arousal (Sitaraman et al., 2015; Scaplen et al., 2020). DAergic activity may inhibit these MBONs, permitting consolidation of alcohol preference. These effects provide insight into neural mechanisms for the association of alcohol with context cues and memories, which is a critical feature of the persistence of addiction behaviors (Scaplen et al., 2020). As genetic and behavioral tools become increasingly advanced, this area of research should continue to expand.

Glutamate is unique in that it has a multi-faceted role in the fly brain, exerting both excitatory and inhibitory effects (Jan and Jan, 1976; Liu and Wilson, 2013). Furthermore, the complex action of poorly understood neurotransmitters like glutamate and ACh is complicated by phenomena such as dual neurotransmission. Dual neurotransmission overrides the classical view of “one neuron, one transmitter” and shows that neurons often release two or more neurotransmitters (Vaaga et al., 2014). The majority of OA neurons in the Drosophila brain are also glutamatergic, and dual neurotransmission is involved in behaviors like aggression and courtship (Sherer et al., 2020). These discoveries have been made possible via new tools enabling detailed glutamatergic manipulations like RNAi knockout of glutamate in OA neurons (Sherer et al., 2020) and circuit tracing and labeling with trans-Tango (Talay et al., 2017). Other developments include the electrophysiological characterization of neurons (Liu and Wilson, 2013), GAL4/UAS inhibition of specific glutamatergic neurons (Liu and Wilson, 2013), and post-synaptic knockdown of glutamate receptors (Das et al., 2011). Using these glutamatergic advancements as a model, outcomes for other complex and overlapping neurotransmitter systems will continue to be clarified.