Pharmacological targeting of the GABAB receptor alters Drosophila's behavioural responses to alcohol

Abstract When exposed to ethanol, Drosophila melanogaster display a variety of addiction‐like behaviours similar to those observed in mammals. Sensitivity to ethanol can be quantified by measuring the time at which 50% of the flies are sedated by ethanol exposure (ST50); an increase of ST50 following multiple ethanol exposures is widely interpreted as development of tolerance to ethanol. Sensitivity and tolerance to ethanol were measured after administration of the gamma‐aminobutyric acid receptor B (GABAB) agonist (SKF 97541) and antagonist (CGP 54626), when compared with flies treated with ethanol alone. Dose‐dependent increases and decreases in sensitivity to ethanol were observed for both the agonist and antagonist respectively. Tolerance was recorded in the presence of GABAB drugs, but the rate of tolerance development was increased by SKF 97451 and unaltered in presence of CGP 54626. This indicates that the GABAB receptor contributes to both the sensitivity to ethanol and mechanisms by which tolerance develops. The data also reinforce the usefulness of Drosophila as a model for identifying the molecular components of addictive behaviours and for testing drugs that could potentially be used for the treatment of alcohol use disorder (AUD).

promising role in altering the sensitivity and preference to alcohol when pharmacologically targeted. [9][10][11][12][13] The GABA B receptor has been investigated and pharmacologically challenged with different GABA B receptor ligands in different animal models for its potential role in modulating alcohol-associated behaviours in AUD. Zaleski et al., 14 Cousins et al., 9 Dzitoyeva et al., 10 and Hwa et al., 11 have all reported on the effects of administering a GABA B receptor ligand to overcome the effects of alcohol addictionlike behaviours in both mammalian and invertebrate models. Likewise, the pharmacological activation of the GABA B receptor has been shown to reduce the behavioural effects of addiction to both cocaine and nicotine. 9,15,16 While these previous studies support the notion that GABA B receptors play a role in addiction pathways, further investigations using GABA B receptor drugs have the potential to elucidate the specific addiction mechanisms that are affected by GABA B receptor activation or inhibition.
Among the wide variety of animal models employed in addiction research, Drosophila melanogaster has been identified as being a strong model for studying the acute and chronic effects of alcohol abuse. [17][18][19] The underlying fundamental when using any animal model to replicate a human disease is to ensure simplicity, reproducibility, and mechanistic validity with respect to human behaviours, and Drosophila, in this instance, are a highly valuable model when used to study alcohol addiction-like behaviours 18,19 .
In this work, GABA B receptor agonist, SKF 97541, and GABA B receptor antagonist, CGP 54626, (previously used in Drosophila by Dzitoyeva et al., Dacks et al., and Root et al., 10,21,22 ) were used to target the GABA B receptor to demonstrate that this receptor plays a role in alcohol sensitivity and tolerance. This novel research supports the possibility of targeting the GABA B receptor as a strategy for the treatment of AUD.

| Drosophila husbandry
All flies used were Canton S wild-type male Drosophila (#64349) obtained from Bloomington Drosophila Stock Centre, Indiana, USA.
Flies were kept at 25°C with an average humidity of 60% to 80% on standard ready mixed dried food (Phillip Harris, UK). All flies tested were aged 1 to 5-day old (1-day old on first day of testing) and were collected the morning before testing via light CO 2 anaesthesia to allow sufficient recovery after sedation.

| Tolerance assay
The tolerance assay was conducted by following the methodology described by Maples and Rothenfluh. 23 Male flies were sorted into groups of eight and were placed into vials containing standard food 24 hours prior to testing. This was completed so that the group of test flies could be transferred to the tolerance chamber with no CO 2 anaesthesia. Five hundred microliters of 100% ethanol were added to the centre of a cotton vial plug, which was used to plug the tolerance chamber with the ethanol soaked side facing inwards.
Flies were left to be exposed to the ethanol vapour until 50% of the total number of flies were sedated; this time point was defined as ST50 (time taken for 50% of flies in the vial to become sedated).
Vials were tapped three times onto the lab bench once per minute to disorientate and startle the flies and to knock them to the bottom of the vial. Each vial was observed for 10 seconds, and the number of stationary flies was recorded every minute. The term "sedation" was determined by making clear distinguishable rules to follow; after flies had been knocked to the bottom of the vial, any fly not able to upright themselves within a 10 second observation time was determined as being sedated. Any leg movement or vibrating of the wings was ignored, and if the flies were still upon their backs after being startled to the bottom of the vial with the inability to self-right, the fly was also recorded as being sedated.
Each assay was conducted at room temperature (with a control sample of eight 1-to 5-day-old male flies every time to ensure no bias was created with a change in room temperature) until the ST50 for each vial was achieved. Once sedated, flies were placed into an ethanol vapour free vial to recover at room temperature. After the flies had recovered and normal motility had resumed, they were housed back upon standard dry mix food at 25°C, and 24 hours later, the assay was repeated.
Tolerance was measured over a period of 4 days, and the ST50 was recorded for each vial. Tolerance to ethanol is identified as being an increase in ST50 over consecutive days following repeated ethanol exposures. This results in flies having a decreased sensitivity to ethanol, consequently resulting in an increased ST50 over time when compared with the first ethanol exposure on day one. [23][24][25][26] Data from the tolerance assay were collected by testing one vial of each condition on eight consecutive weeks. Each vial contained eight 1 to 5-day-old male flies (1-day old on first day of testing) per vial. Mean values are N = 8 ± standard error of the mean (SEM).

| Drug selection and delivery
SKF 97541 is a potent and selective GABA B receptor agonist, and CGP 54626 is a potent and selective GABA B receptor antagonist (Tocris Biotechne, UK). SKF 97451 was dissolved in water at a stock concentration of 25 mM and CGP 54626 was dissolved in DMSO (Tocris Biotechne, UK) at a stock concentration of 25 mM. Flies were exposed to the drugs at the desired concentrations in the liquid food (5% yeast extract, 5% sucrose, 1% red coloured dye [Sigma Aldrich, UK]). Drug and control solutions were loaded into four 5 μL glass capillaries (Jaytec, UK) within a CAFÉ assay chamber as described by Ja et al. 27 Flies were allowed to consume the liquid food via the fourglass capillaries for 24-hour ad libitum. Flies' abdomens were observed to see if they were red prior to testing; this was used as a positive indicator as to whether the flies had consumed/recently consumed the liquid food containing the red food colouring. The average consumption of liquid food was estimated from measuring the capillary meniscus over 24 hours, and it was estimated to be 0.42 μL ± 0.046 SEM with no significant difference in the consumption of different drug containing foods ( Figure S1). Vials with flies that had less than four out of eight red abdomens were discarded. Control flies were fed the same liquid food with no added drug. A vehicle control group was treated with 0.002% DMSO (Tocris Biotechne, UK) which corresponds to the dose of DMSO delivered with CGP 54626. Liquid foods with/without the drug were given to Drosophila with different protocols: (a) for 24 hours prior and during the 4 days of tolerance testing, (b) only for 24 hours prior to tolerance testing, (c) only for 24 hours after the second ethanol exposure, prior to the day 3 ethanol exposure ( Figure 1).

| Locomotor activity assay
The locomotor climbing assay was constructed by wrapping 2-cm graph paper with horizontal lines around the rear of a standard 9.5 × 2.5 cm vial (height × diameter). A vial was positioned in close proximity to a timer, and a camera was positioned to capture both the climbing assay vial and the timer within one frame. The camera was set to record once the flies were transferred within the assay vial and cameras field of view. All flies were then tapped to the bottom of the vial and at the time of doing so, the timer was manually started.
The frames of video were visualised through QuickTime Player Version 10.4 (Apple Inc, USA) by scanning through frame by frame until all flies had crossed the 8 cm mark on the vial or 15 seconds had elapsed, whichever was sooner. The time at which all flies had crossed the 8-cm mark was then recorded.

| Ethanol concentration measurement
Ethanol absorption was measured with a spectrophotometric assay as described previously by Moore et al. 28 Thirty 1 to 5-day-old male flies were exposed to 500 μL of 100% ethanol in a tolerance chamber (as described above). After 30 minutes, flies were frozen in dry ice and were homogenised in 300 μL of ice-cold Tris-HCl (pH 7.5). The homogenate was centrifuged at 4500 revolutions per minute for 20 minutes at 4°C. Five microlitres of supernatant was added to 495 μL of ethanol assay reagent (Sigma, UK). The manufacturers protocol and calculation instructions were followed. To calculate the ethanol concentration in individual flies, the volume of one fly was estimated to be 2 μL. 28 The estimated ethanol concentration in one individual fly after 30 minutes was calculated to be 6.5 mM. A control group (no ethanol vapour) was also assayed, and this result was negative ( Figure S2).

| Statistical analysis
All graphs and analysis were made and executed using the Prism for Mac OS X Version 7.0 (GraphPad Software Inc, USA).
The change in sensitivity (in minutes) was calculated by subtracting the ST50 value of the treated flies by the ST50 value for the control flies of the same day of treatment. The percentage change in sensitivity from day 2 to day 4 (tolerance) was calculated by the formula ([('Day X' ST50/Day 1 ST50] −1) × 100. ANOVA analysis was carried out to determine statistical significance.

| Development of tolerance to ethanol
Repeated exposures to ethanol in wild type flies leads to a reduced sensitivity to its sedative effect. Tolerance development is defined as a change in the time (ST50) needed for half of a set of flies exposed to ethanol vapour to become sedated as compared with the first ethanol exposure. ST50 was measured in sets of eight male flies and was shown to increase over a period of 5 days with daily ethanol exposures ( Figure 2). These findings are consistent with previous reports confirming that D. melanogaster are a suitable model to study the development of alcohol tolerance as an addiction like behaviour, similar to that seen in humans. 20,23,24,26,29 FIGURE 1 Administration schedule of ethanol and gammaaminobutyric acid receptor B (GABA B ) drugs. All flies were exposed to ethanol vapour four times and their ST50 recorded (ST50 assay 1-4). GABA B drugs were administered to the fly via liquid food presented in capillary tubes to which the flies had access ad libitum. Three different GABA B drug administration schedules were used: A, continuous administration: flies were fed with the drug for 24 hours before the first tolerance assay (when flies were naive to ethanol) and throughout the period in between consequent ST50 assays, B, preadministration: drugs were fed only for 24 hours before the first ST50 assay (when flies were naive to ethanol), C, mid administration: drugs were fed for 24 hours only between the second and third ST50 assay. Each arrow represents drug administration for a 24-hour period

| Tolerance development with continuous administration of agonist SKF 97541
The GABA B receptor agonist SKF 97541 was dissolved in liquid food and was administered via glass capillaries ad libitum for 24 hours prior to the first ST50 assay and during the subsequent 3 days of testing ( Figure 1A

| Tolerance development with continuous administration of antagonist CGP 54626
To further explore the role of the GABA B receptor in response to ethanol exposure, the GABA B antagonist CGP 54626 was administered to the flies with the same protocol used for the agonist ( Figure 1A, Figure 4B and Figure 5).       increasing ST50 values when compared with the control for that respective day ( Figure 7B). The effect of the drug was diminished to zero over the next 2 days of testing, and this indicates that the drug takes more than 24 hours to reach full efficacy before starting to lose its effect ( Figure 7C-D).

| CGP 54626 antagonist administration prior to and during tolerance development
When CGP 54626 was administered after 2 days of ethanol tolerance, the ST50 values were increased to highly significant values when compared with the control for day 3 at all the CGP 54246 concentrations used, before slowly losing efficacy on day 4 (2 days/ 48 hours after initial administration) and there being an evident decrease in significance when compared with the control for day 4 in a dose dependent manner ( Figure 7C,D).

| DISCUSSION
Within this study, we have used a tolerance assay as described by Maples and Rothenfluh 23 to show that D. melanogaster can develop tolerance with repeated exposures over consecutive days, mimicking an alcohol-related behaviour that is seen with AUDs in humans ( Figure 2). The Drosophila model allows pharmacological interventions that can aid in dissecting the mechanisms of tolerance development. 23,24,26 In this work, we have used GABA B receptor ligands to explore tolerance development. The range of concentrations of drug used were determined by toxicity studies where flies survived for more than 96 hours at the doses used (data not shown) and by negative geotaxis assays to determine that the drugs did not cause substantial effect on locomotor activity ( Figure S3). The concentration of 1.25 mM SKF 97451 did affect locomotor activity; however, the results were retained in this study because it was of interest to determine what effect this higher concentration had when the files were exposed only for 24 hours pre or mid tolerance development ( Figure 6). Additionally, the fact that tolerance developed even after continuous administration of this high concentration (Figure 3) indicates that the effect on locomotor activity and ST50 measurement is not necessarily correlated.
Whilst the idea that the GABA B receptor could be used as a potential drug target to combat alcohol addiction like behaviours is not novel, [9][10][11]16,21 this study provides direct evidence of an agonist and an antagonist ligand that are capable of increasing and decreasing sensitivity to ethanol vapour respectively after different numbers of alcohol exposures (Figures 3, 5-7).
The data presented indicate that the GABA B drugs affect the sensitivity to alcohol by reducing/increasing the time to sedation, while permitting tolerance development to occur ( Figure 4A,B). It should also be noted that the agonist SKF 97451 decreased the ST50 and altered the rate at which tolerance (defined as an increase in ST50) developed after four ethanol sensitivity assays. This novel data reinforce the concept that the development of tolerance is a multifactorial process where GABA B receptors (like other receptor systems) contribute to, but do not fully determine the development of tolerance. Whilst it is known that ethanol increases GABA release 30 and that the function of GABA A receptors is affected by ethanol, 3,8 it is important to also fully characterise the role of GABA B receptors in ethanol-induced processes. The novel contribution of this work is that the modulation of the GABA B receptors alters the sensitivity to ethanol as well as affecting the rate of tolerance development. There are potentially different mechanisms by which modulation of the GABA B

FIGURE 6
Effect of SKF 97541 pre-administration and mid-administration on sedation time (ST50). Flies were fed with SKF 97451 only for 24 hours prior to the 4-day tolerance assay (pre-administration, white bars) or for 24 hours in between the second and third day of the 4-day tolerance assay (mid-administration, shaded bars). Horizontal bars represent changes in ST50 in minutes in SKF 97451-treated flies as compared with control untreated flies for the same day of the tolerance assay. N = 8 (eight vials with eight flies per vial). The ST50 of the control flies is indicated in minutes ± SEM at the zero-difference point for each consecutive day. Averages ± SEM. 2-way ANOVA with Tukey vs control of that day. *** = <0.001, **** = <0.0001

FIGURE 7
Effect of CGP 54626 pre-administration and mid-administration on sedation time (ST50). Flies were fed with CGP 54626 only for 24 hours prior to the 4-day tolerance assay (pre-administration, white bars) or for 24 hours in between the second and third day of the 4-day tolerance assay (mid-administration, shaded bars). Horizontal bars represent changes in ST50 in minutes in CGP 54626-treated flies as compared with control-untreated flies for the same day of the tolerance assay. N = 8 (eight vials with eight flies per vial). The ST50 of the control flies is indicated in minutes ± SEM at the zero-difference point for each consecutive day. Averages ± SEM. 2-way ANOVA with Tukey vs control of that day. * = <0.05, ** = <0.01, *** = <0.001, **** = <0.0001 receptors could affect the change in tolerance development, which include modulation of GABA or other neurotransmitter release in the brain, and alcohol metabolism within the brain and other organs.
Prior to the successful cloning of the GABA B receptor subunits by Mezler, Müller and Raming, 31 Drosophila was thought to lack GABA B receptors, as the GABA B receptor agonist Baclofen has been reported to have no effect in Drosophila. 13 However, the two pharmacological probes (agonist SKF 97541 and antagonist CGP 54626) used here have been reported to work as a specific GABA B receptor agonist and antagonist in Drosophila. 10,21,22 It still remains to be established however, as to why baclofen does not appear to be functional in Drosophila.
The ultimate goal of alcohol addiction research is to resolve addiction-like behaviours in individuals with AUD. This work has strengthened the possibility that the GABA B receptor could be pharmacologically targeted to improve AUD related behaviour.
However, further investigations on the role of GABA B receptors are required. The Drosophila model and the vast genetic tools that have been made available by the Drosophila research community have the potential to identify pharmacological strategies that could then be tested and applied in human subjects.