Paradigms for Behavioral Assessment in Drosophila Model of Autism Spectrum Disorder

Summary

Autism Spectrum Disorder (ASD) is associated with impaired social and communicative behavior and the emergence of repetitive behavior. For studying the interrelation between ASD genes and behavioral deficits in the Drosophila model, five behavioral paradigms are described in this paper for assaying social spacing, aggression, courtship, grooming, and habituation behavior.

Abstract

Autism Spectrum Disorder (ASD) encompasses a heterogeneous group of neurodevelopmental disorders with common behavioral symptoms including deficits in social interaction and ability for communication, enhanced restricted or repetitive behaviors, and also, in some cases, learning disability and motor deficit. Drosophila has served as an unparalleled model organism for modeling a great number of human diseases. As many genes have been implicated in ASD, fruit flies have emerged as a powerful and efficient way to test the genes putatively involved with the disorder. As hundreds of genes, with varied functional roles, are implicated in ASD, a single genetic fly model of ASD is unfeasible; instead, individual genetic mutants, gene knockdowns, or overexpression-based studies of the fly homologs of ASD-associated genes are the common means for gaining insight regarding molecular pathways underlying these gene products. A host of behavioral techniques are available in Drosophila which provide easy readout of deficits in specific behavioral components. Social space assay and aggression and courtship assays in flies have been shown to be useful in assessing defects in social interaction or communication. Grooming behavior in flies is an excellent readout of repetitive behavior. Habituation assay is used in flies to estimate the ability for habituation learning, which is found to be affected in some ASD patients. A combination of these behavioral paradigms can be utilized to make a thorough assessment of the human ASD-like disease state in flies. Using Fmr1 mutant flies, recapitulating Fragile-X syndrome in humans, and POGZ-homolog row knockdown in fly neurons, we have shown quantifiable deficits in social spacing, aggression, courtship behavior, grooming behavior, and habituation. These behavioral paradigms are demonstrated here in their simplest and straightforward forms with an assumption that it would facilitate their widespread use for research on ASD and other neurodevelopmental disorders in fly models.

Introduction

Autism Spectrum Disorder (ASD) encompasses a heterogeneous group of neurological disorders. It includes a range of complex neuro-developmental disorders characterized by multi-contextual and persistent deficits in social communication and social interaction and the presence of restricted, repetitive behavioral and activity patterns and interests¹. According to World Health Organization (WHO), 1 in 100 children is diagnosed with ASD worldwide with a male-to-female ratio of 4.2². The disease becomes evident in the second or third year of life. ASD children show a lack of interest in social-emotional reciprocity, non-verbal communication, and relationship skills. They exhibit repetitive behaviors like stereotyped motor movement, inflexible and ritualized routine following, and intense focus on restricted interests. ASD children show a high degree of response towards touch, smell, sound, and taste whereas pain and temperature response is comparatively low¹. The penetrance of this disorder is also different among different patients suffering from ASD and hence, the variability increases.

Current clinical diagnosis of ASD is based on behavioral assessment of the individuals as there is no confirmatory biomarker-based or common genetic test that covers all forms of ASD³. Deciphering the genetic and neurophysiological bases would be helpful in targeting treatment strategies. In the last decade, a large body of research has resulted in the identification of hundreds of genes that are either deleted or mutated or whose expression levels are altered in ASD patients. Ongoing research emphasizes the validation of the contribution of these candidate genes using model organisms like the mouse or fruit-fly, in which, these genes are knocked out or knocked down followed by tests for ASD-like behavioral deficits and elucidation of underlying genetic and molecular pathways causing the anomalies. A mouse model recapitulating Copy Number Variations (CNVs) in the human chromosomal loci 16p11.2 shows some of the ASD behavioral defects^4,5,6. Prenatal exposure to a teratogenic drug valproic acid (VPA) is another mouse model depicting traits resembling human ASD^7,8. In addition, there exists a range of mouse models that exhibit genetic syndrome-associated autism, for example, single-gene syndromic models caused by mutations in Fmr1, Pten, Mecp2, Cacna1c, and single-gene non-syndromic models caused by mutations in genes like Cntnap2, Shank, Neurexin, or Neuroligin genes⁵.

Fruit-fly (Drosophila melanogaster) is another prominent model organism for studying the cellular, molecular, and genetic bases of a plethora of human disorders⁹, including ASD. Drosophila and humans share highly conserved biological processes at the molecular, cellular, and synaptic levels. Fruit-flies have been used successfully in ASD studies^10,11,12 to characterize genes linked to ASDs and decipher their exact role in synaptogenesis, synaptic function and plasticity, neural circuit assembly, and maturation; fly homologs of ASD-associated genes were found to have roles in the regulation of social and/or repetitive behavior^{11,13,14,15,16,17,18,19,20,21}. The fruit-fly has also worked as a model for the screening of ASD genes and their variants^15,22,23. The biggest challenge in ASD research in flies is that, unlike other disease models, there is no single ASD fly model. To understand the impact of mutations or knocking down of a specific ASD gene, a researcher needs to validate whether the behavioral phenotypes sufficiently mimic the symptoms of ASD patients and then, proceed towards understanding the molecular or physiological underpinnings of the phenotypes.

Hence, the detection of ASD-like phenotypes is vital to ASD research in the fly model. A handful of behavioral techniques have emerged over the years that enable us to detect abnormalities like deficits in social behavior/interaction, communication, repetitive behaviors, and responsiveness to stimuli. In addition, several modifications and upgrades of these behavioral techniques have been made in different labs to suit specific requirements such as upscaling, automation of assays, readouts, quantification, and comparison methods. In this video article, the most basic versions of five behavioral paradigms are demonstrated, which, in combination, can be used to detect ASD-like behavioral outcomes in the easiest way.

Aggression is an evolutionarily conserved innate behavior affecting survival and reproduction²⁴. Aggressive behavior towards conspecifics is influenced by 'motivation for socialization'^25,26 as well as 'communication'²⁷, both being compromised in ASD-affected individuals. Aggressive behavior is well described in Drosophila and its quantifiability through the robust aggression assay^28,29,30 and a well-understood genetic and neurobiological basis³¹ makes it a suitable behavioral paradigm³² for assessing the ASD phenotype in a fly model. Aggression is affected by social isolation away from a social environment, which leads to enhanced aggression; the same has been observed when male flies are housed in isolation for a few days^33,34. Another behavioral assay that quantifies sociability in flies is the Social Space Assay³⁵, which measures distances between nearest neighbors and interfly distances in a small group of flies, making it perfectly suited for testing the roles of ASD gene orthologs in fly^12,21,36,37 as well as environmentally induced ASD fly models^38,39.

The Drosophila courtship assay is another behavioral paradigm frequently used to assay for alteration in social and communication skills upon circuit or genetic manipulation, including Autism related genes^18,19,21,40. Repetitive patterns of behavior are prevalent in ASD patients, which is recapitulated in flies by grooming behavior-a series of distinct, stereotyped actions performed for cleaning and other purpose. It has been successfully used to assay for the impact of ASD gene mutations in flies^21,41 as well as exposure to chemicals^38,39. Multiple advancements and automation in the assay have been described before^16,41,42,43; here, we are demonstrating the most basic assay pattern, which is easy to adopt and quantify.

ASD is known to impact the ability for habituation, learning, and memory in some patients^{44,45,46,47,48,49,50}, ASD model organisms^51,52 and also causes deficits in different olfactory behaviors⁵⁰. Drosophila light-off jump habituation has been used previously to screen for ASD genes²³. Habituation can be assayed by a simple method of olfactory habituation assay^53,54,55. We describe the method to induce olfactory habituation and assay the outcome using a classic Y-maze-based binary odor-choice assay⁵⁶ that can be used to detect defects in habituation in ASD gene mutant or gene knockdown condition. To assess whether the impact of a mutation (or gene knock-down) or a pharmacological treatment on the behavior of a fly amounts to an ASD-like phenotype, one can use a combination of these 5 assays described here.

Protocol

See the Table of Materials for details related to all materials and reagents used in this protocol.

1. Aggression assay

1. Preparing the aggression assay arena
   1. Take a standard 24-well plate (Figure 1A) and use each well of the plate as a single 'arena' (Figure 1B) for fly aggression. Fill half of each well with regular fly food and allow it to dry overnight.
      NOTE: Optionally, a focal point for aggression may be included in the arena, such as a dot of yeast paste or a decapitated female to ensure male aggression.
   2. Take any transparent plastic/acrylic sheet, enough to cover the surface of about three to four wells. Perforate the sheet to make a small aperture of ~2.5 mm diameter for inserting individual flies into the wells through it.
2. Preparing the fly aspirator
   1. Take a 30-50 cm long rubber/silicone tube (Figure 1C1).
   2. Take a 200 µL (or 1,000 µL) pipette tip and chop off ~1 cm from the narrow tip. Insert the cut end at one end of the rubber tube; this tip will be the 'mouth end', used for mouth-based aspiration.
   3. Take another pipette tip and cut the narrow end of this tip to make an opening sufficient for a fly to enter through it. Insert the base of this pipette tip into the tail end of the tube; this will be the 'fly end' of the aspirator, from where the fly will get aspirated (Figure 1C2).
   4. Insert a piece of mesh cloth or a thin layer of cotton on the 'fly end' of the tube-at the junction between the tube and tip. Ensure that a fly aspirated into the pipette tip remains trapped in the tip, in front of the mesh.
   5. Seal the junctions between the pipe and the pipette tips using parafilm.
3. Preparing single-housing tubes and group-housing vials
   1. Preparation of single-housing tubes
      1. Add nearly 500 µL of freshly prepared fly food to a 2 mL microfuge tube (Figure 1D). Use a mini centrifuge to pull down the food to the bottom part of the tube.
      2. Set the food to solidify at room temperature by keeping the lid open overnight. Cover the tubes with a fine piece of cloth to prevent stray flies from entering the tubes.
      3. Perforate the lids of the microfuge tubes with needles for air circulation, and close the lids after the food is solidified.
   2. Preparation of group-housing vials
      1. Take regular fly vials and fill a bare minimum of the vials with freshly prepared food (Figure 1D).
4. Preparing the flies before the experiment
   1. Maintain fly lines of the desired genotypes in regular glass/plastic bottles containing standard fly food in an incubator at 25 °C on a 12 h light/dark cycle.
   2. Collect newly eclosed (0 to 24 h) flies of the desired genotype and sex-separate them under carbon dioxide anesthesia using a stereomicroscope.
   3. Insert half the male flies individually into 'single-housing tubes'. Keep the other half of the male flies in a group of 10 with the female flies in regular 'group-housing vials' creating a social condition. Store all tubes and vials at 25 °C for 5 days.
      NOTE: Maintain a temperature of 24-25 °C and ~50% humidity in the behavior room. The fly strains used were: w¹¹¹⁸ and FMR trans-heterozygote mutant flies (Fmr1Δ50M/Fmr1Δ113M)⁵⁷. Fmr1 flies, which were kept in isolation for 5 days, show significantly decreased aggression bouts towards another male (Figure 1E) .
5. Performing the aggression behavior experiment
   1. Perform the aggression experiment during the ZT0-ZT3 time window as the flies show peak activity during this time of the day.
      NOTE: ZT=Zeitgeber time in a standard 12 h light/dark cycle; ZT0= lights on, that is, the start of light phase, ZT3= 3 h after the start of the light phase, and so on. ZT12= lights off.
   2. Transfer two male flies from either the single- or group-housing chambers by the mouth-aspiration method to the aggression arena through the hole in the lid; move the hole away immediately to ensure that the flies cannot escape.
   3. Allow the flies to acclimate within the arena for 1-2 min. Start a timer for 20 min. Video record the flies for the entire 20 min by placing a camera or a mobile phone exactly vertically above the arena using a stand. Ensure that the types of aggressive bouts are visible in the video.
      NOTE: Ensure clear illumination of the arena and avoid any glare/reflection from the lid directly falling onto the lens of the camera.
   4. Repeat the experiment for 15-20 pairs of flies for each genotype and every type of housing condition.
      ​NOTE: Unconscious bias can be negated by double blinding of the control and test flies as well as video files belonging to multiple genotypes.
6. Analyzing the aggression assay data
   1. Transfer the video files to a computer with a sufficiently large screen so that the aggressive bouts are visible.
   2. Play the video and count the total number of aggressive bouts in a time span of 20 min^58,59.
   3. Compile and organize data from each genotype and each housing pattern in a spreadsheet and perform data analysis using any statistical software. Perform a two-tailed t-test and plot the data as box and whiskers.

2. Social space assay

NOTE: The assay protocol, arena, and analysis described here have been described previously^60,61.

1. Preparing the social space assay (SSA) arena
   1. Place both the triangular acrylic spacers (height = 8.9 cm, base = 6.7 cm, and thickness = 0.3 cm) flat on top of a rectangular glass pane (13.5 cm x 10.4 cm, thickness 0.3 cm) such that the right angles of the acrylic spacers are aligned with the corners of the glass pane (Figure 2A).
   2. Place two rectangular acrylic spacers (6.7 cm x 1.5 cm, thickness = 0.3 cm) flat on the glass pane, aligned with the bases of the two triangular spacers. After this arrangement, confirm that the four acrylic spacers surround a triangular arena (~2.16 sq. cm⁶¹) on the rectangular glass pane that is not covered by spacers (Figure 2A,B). Place a sticker of a ruler (Figure 2C) on one of the triangular spacers and ensure that it is visible from the top.
   3. Now place a second rectangular glass pane (13.5 cm x 10.4 cm, thickness 0.3 cm) on top of the acrylic spacers such that it aligns with the glass pane at the bottom; acrylic spacers would end up sandwiched between the glass space, leaving a triangular space in the middle between two glass panes. Use four small binder clips to hold the panes and spacers.
2. Preparation of the flies for SSA
   1. Collect newly eclosed (0 to 24 h) flies of the desired genotype and sex-separate them under cold anesthesia using a stereomicroscope.
      NOTE: Chill the cold anesthesia apparatus (a Petri dish may be used) in a -20 °C incubator. Place the flies into empty vials, submerge these vials into ice (in an ice bucket) till the flies become immobile, place them on the cold Petri dish, and start sorting.
   2. Store the male and female flies separately for 24 h in food vials in a 25 ˚C incubator with a 12 h light/dark cycle.
      NOTE: For the experiments demonstrated here, the fly strains used were w¹¹¹⁸ and FMR trans-heterozygote mutant flies (Fmr1Δ50M/Fmr1Δ113M).
3. Performing the SSA experiment
   1. Maintain the same experimental conditions as the aggression assay (temperature: 24-25˚C, humidity ~ 50%) and perform the experiment during the ZT0-ZT3 time window.
   2. Remove the bottom right binder clip and slightly shift that rectangular spacer outwards so that a gap (~0.5 cm) is created between the two rectangular spacers.
   3. Transfer flies (collected on the previous day) from the food vial into an empty vial. Transfer them into the social space arena (the triangular area) by gentle aspiration through the space created between the rectangular spacers. Immediately close the space by sliding the rectangular spacer and binder clip back in their positions and ensure that no space is left for the flies to escape.
   4. By holding the chamber in the upright position, gently pound the chamber 3x onto a soft pad to ensure that all the flies are at the base of the chamber at the onset of the experiment.
   5. Clamp the SSA chamber in the upright position and start the timer. Take a clear photo of the arena after 20 min when the flies settle at their positions in the arena and show minimum movement. Avoid glare and irregular illumination.
   6. Repeat the experiment 3x for the same population of flies by pounding the flies and repeating the steps from steps 2.3.5 to 2.3.7 (internal replicates). Repeat 3x with a different population of flies of the same genotype/condition (independent repeats).
      ​NOTE: Freeze the assay chamber to discard the flies from the chamber. Wipe the glass and plastic surfaces with ethanol to remove all odors after one round of experiment is done with one group of flies.
4. Analyzing the social space assay data
   1. Analyze the images in ImageJ⁶² software and list the inter-fly nearest-neighbor distances as previously described⁶¹.
   2. Perform statistical analysis and plot graphs using statistical software.
      NOTE: The fly strains used for the SSA assay described here were w¹¹¹⁸ and Fmr1 trans-heterozygote mutant flies (Fmr1Δ50M/Fmr1Δ113M)⁵⁷. Fmr1 flies show significantly increased distance from the nearest neighbors (Figure 2D).

3. Courtship assay

1. The courtship chamber
   1. Assemble all four pieces of the courtship chamber (Figure 3A) together in the order shown in Figure 3B. Each perforation of the central disk, sandwiched between the lid and the base pieces, will work as the 'courtship chamber' (Figure 3C).
2. Obtaining premated females
   1. Start and maintain multiple culture bottles of CS (Canton S, wild type) flies with ~60-100 male and female flies in each food bottle.
   2. When adults emerge, discard all adult flies and transfer the flies emerging from these bottles every 2-3 h to new food vials supplemented with a tiny amount of yeast paste.
      ​NOTE: Avoid overcrowding of the vials. Be careful not to transfer old flies, larvae, or pupae to the mating vial.
   3. Incubate these "mating vials" for 4 days to ensure that all females have mated.
3. Preparation of single-housing tubes for males
   1. Add nearly 500 µL of fly food to each 2 mL microfuge tube. Use a minicentrifuge machine to pull down the food to the bottom part of the tube.
   2. Allow overnight solidification of the food, cut the lid of the microfuge tube, cover it with parafilm, and perforate the parafilm with a needle for air circulation.
4. Collection and isolation of virgin males
   1. Set up crosses with 10-15 males and 20-30 virgin females of the desired genotypes in separate food vials.
   2. Collect virgin males of the desired genotypes from the progeny and keep them individually in 'single-housing tubes' using the aspirator. Keep collecting newly eclosed males every 5-6 h (up to 40-50 males for each genotype) and isolate them in individual single-housing tubes.
   3. Re-seal the lid of the tube with the parafilm.
5. Performing the courtship assay experiment
   1. After 5 days of isolation of the test males, perform the courtship assay during the ZT0-ZT3 time window.
   2. Set up the video recording devices well in advance focusing on the assay workspace.
   3. With the help of an aspirator, collect one premated female (6-10 days old) from the mating vials and insert into a courtship chamber.
   4. Using the aspirator, gently transfer a male (control/test) fly from the single-housing tube to the courtship chamber containing the single premated female through the transfer hole. Rotate the lid quickly to close the chamber.
   5. Video record the behavior of the flies for 15 min.
6. Video data analysis and statistics
   1. Transfer the video files to a computer; note the duration of time spent by the male in courtship during 15 min by manually going through the video.
   2. Calculate the courtship index (CI) for each male fly, which is the fraction (or percentage) of time spent by the male in courting the female during 15 min.
   3. Count the total number of copulation attempts in 15 min.
   4. Note the duration of time lag shown by the male before its first attempt to court as the courtship latency (CL).
      NOTE: It is recommended to analyze 40-60 males per condition/genotype to achieve statistical power and determine the consistency of the CI results.

4. Grooming behavior assay

1. Grooming behavior assay arena
   1. Use a small circular chamber with a volume of ~0.4 cm³ as an arena for recording the grooming behavior.
      NOTE: The same courtship arena described in section 3 may be used for grooming behavior assay.
   2. As grooming behavior involves recording of movement of finer organs like legs, use high-resolution or high-contrast video recording. To follow this protocol, use a diffused glass-covered LED panel as a uniformly illuminated surface of dimensions 20 cm x 20 cm. Place the courtship chamber on top of the panel to ensure light passes from the bottom through the chamber.
2. Preparing the flies before the experiment
   1. Maintain experimental genotype flies in food bottles at 25 °C with a 12 h light/dark cycle.
   2. Collect newly eclosed (0 to 24 h) flies of the desired genotype and sex-separate them under cold anesthesia using a stereomicroscope as described in the social space assay.
   3. Isolate the male flies in single-housing tubes (as used in the aggression assay) for 24 h.
      NOTE: For the experiment demonstrated here, the fly strains used were: W¹¹¹⁸ and FMR trans-heterozygote mutant flies (Fmr1Δ50M/Fmr1Δ113M).
3. Performing the grooming behavior assay
   1. Maintain a temperature of 24-25 °C and humidity ~ 50%; perform the experiment during the ZT0-ZT3 time-window¹⁶.
   2. Transfer a single-housed male fly from a single-housing tube into the grooming arena by using an aspirator. Immediately slide away the hole in the lid to ensure that the flies cannot escape.
   3. Place the grooming arena on a diffused LED panel and allow the fly to acclimate within the arena for 1 min. Video record the fly for 10 min by placing the camera exactly vertically above the arena mounted on a stand. Ensure that the types of grooming bouts are visible and quantifiable in the video.
      NOTE: The grooming behavior includes rubbing of the head, eyes, antennae, proboscis thorax, abdomen, genitalia, and wings with legs (first pair: T1, second pair: T2, or third pair of legs: T3). Grooming behavioral parameters that have been taken into account in this study are as described in Andrew et al. ⁴¹ and demonstrated in Figure 4.
   4. Repeat the experiment for each genotype.
   5. Analyzing the grooming behavioral data
      1. Analyze the videos and calculatethe following four parameters: grooming Index (percentage of time spent in grooming); grooming latency (time until first grooming bout); grooming-bout number; and mean grooming-bout duration (total bout duration/bout number).
      2. Mark a single grooming bout as finished when a fly either stops showing any of the parameters and remains motionless for 2 s or stops the bout and walks at least 4 steps.

5. Assay for olfactory habituation

NOTE: As shown in Figure 5, the final assembly needs to be done on the day of Y-maze assay^54,56.

1. Preparing flies for habituation
   1. Raise flies of the desired genotypes for all the experiments in an incubator with an ambient temperature of 25 ^oC and 70% humidity under standard 12 h light: 12h h dark cycle (LD).
   2. Collect 0-12 h old, newly eclosed flies and transfer ~30-40 flies in a fresh medium bottle with a tightened cotton stopper.
   3. Assign codes to each bottle to keep the experimenter blind about the genotypes under experimentation.
2. Induction of olfactory habituation
   1. Place 1 mL of the preferred odorant diluted with paraffin liquid (light) in a 1.5 mL microfuge tube. For the control, just use 1 mL of paraffin liquid light in a 1.5 mL microfuge tube. Vortex the contents in the tube for 10 min to ensure uniform mixing and then cover it with evenly perforated plastic wrap.
      NOTE: In the video, 20% ethyl butyrate will be used.
   2. Use wire to suspend the tubes containing the diluted odorant or only the diluent liquid in separate media bottles containing flies.
   3. Cover the bottles well with cotton and then wrap with kraft paper to prevent diffusion of the odorant vapor. Label the control and odor-containing bottles as naïve and odor-exposed (habituated), respectively. Maintain these induction bottles for 3 days in an incubator with the above-mentioned conditions.
3. Preparation of flies and the Y-maze apparatus
   1. Transfer the flies from the induction bottles to vials containing only water-soaked filter paper.
   2. Starve them for approximately 16-18 h at room temperature prior to the experiment to increase motivation.
   3. Ensure that the components of the Y-maze apparatus are clean and odor-free; assemble the four parts in a vertical fashion. Attach the climbing chambers to the Y-maze, which is then connected to the top of the adaptor. Firmly attach the bottom of the Y-maze to the entry vial that serves as the central stem at the bottom as shown in Figure 5.
   4. Connect the tapering end of each climbing chamber with the reagent bottles containing odorant (10^-3 dilution with distilled water) using odor-free silicone tubes.
   5. Pump the odorant at an equal flow rate (~120 mL/min) from the gas bottles to the two arms of the Y with the help of a vacuum pump and allow it to saturate for 15 min for consistency.
4. Performing the classical Y-maze assay
   1. Gently introduce the starved flies of each vial into the entry vial of the Y-maze setup.
   2. Allow them to acclimate for a brief period; connect the entry vial with the Y-maze adapter to begin the test; and let the flies climb the Y-maze arms and get trapped in the two collection chambers. Set the time duration of each test to 1 min.
   3. Upon completion, tap back the flies to the bottom of the entry vial and switch the position of the arms of the Y-maze to avoid side bias. Take four readings for the same set of flies.
   4. Record the number of flies climbing each of the two arms of Y-maze within the time duration.
   5. Repeat the assay for at least 8 batches to obtain a representative dataset.
5. Analysis and interpretation of collected data
   1. Quantify the results by calculating the Performance Index (PI), which can be represented as the difference between the number of flies choosing the air arm (A) and the number of flies choosing the odor arm (O) as a fraction of the total number of participant flies.
   2. Use statistical tests (unpaired Student's t-test) to check if the PI of naïve versus odor-exposed flies are significant.

Results

Aggression assay
As a fly ASD model, Fmr1 mutant flies have been used^63,64. w¹¹¹⁸ males were used as control and Fmr1 trans-heterozygote Fmr1Δ113M/Fmr1Δ50M⁵⁷ male flies as experimental flies; adult males were housed in isolation tubes for 5 days. Homotypic males (same genotype, same housing conditions) were introduced in the aggression arena and their behavior was recorded for 20 min. The total number of aggressive bouts in 20 min were counted for each genotype. The number of aggressive bouts during 20 min was found to be significantly reduced in the case of mutant males than in the control males (Figure 1E). This reduced aggressiveness in Fmr1 flies could be due to reduced interest in social interaction with another fly. In contrast, group/socially housed Fmr1 mutant male flies show comparable aggression bout counts as that of socially housed w¹¹¹⁸ flies (data not shown), probably indicating a positive role of the social environment in the induction of social behavior.

Social Space assay
The social space assay was performed in w¹¹¹⁸ and ASD model (Fmr1Δ113M/Fmr1Δ50M) flies. The distance to the closest neighbor was calculated for each fly and 10 biological replicates were performed for each genotype. The mutant flies show significantly higher nearest-neighbor distances than the control counterparts (Figure 2D), indicating a preference for greater distancing from other flies.

Courtship assay
The courtship assay of single-housed male flies was performed to quantify the innate courtship behavior of two separate ASD model flies. First, the courtship behavior of Fmr1 transheterozygous flies was compared with that of w¹¹¹⁸; Fmr1 flies showed a significantly lower courtship index as well as a reduced number of attempts to copulation (Figure 3E1). In a second experiment, row, a Drosophila ortholog of the human POGZ gene (a highly prevalent ASD risk gene²⁰, was knocked down by expressing a short-hairpin microRNA against row in mushroom body neurons using the MB247-GAL4 driver line. These row-knockdown ASD model flies were tested for any defect in the innate courtship pattern; the courtship index and attempted copulation number were calculated from 15 min videos. The row knockdown flies showed a significantly reduced courtship index as well as counts of attempted copulation (Figure 3E2), indicating their defective communication towards females.

Grooming behavior
Grooming behavior is used as a readout for repetitive behavior in Drosophila^16,42,43. Grooming behavior was quantified in Fmr1 transheterozygote flies and compared with that of w¹¹¹⁸. Latency to first grooming is significantly decreased in Fmr1 flies whereas the mean grooming bout duration and grooming index are significantly increased in Fmr1 flies (Figure 4C), indicating enhanced grooming (or repetitive) behavior in ASD model flies. In contrast, the total number of grooming bouts in the mutant flies was not found to be altered significantly compared to that of control flies.

Olfactory Habituation assay
Habituation is found to be defective in a large number of autistic individuals and it can be used as an assay for the same in Drosophila²³. Here, olfactory habituation was tested in the ASD model by knocking down row in olfactory local interneurons driven by LN1-GAL4. An aversive odorant, 20% ethyl butyrate, has been used in these experiments to induce habituation in these flies. The results show that row-knockdown flies did not get habituated after a 3-day odorant exposure, (the last two bars in the graph in Figure 5E do not show a significant difference between naïve vs habituated flies) compared to wild type, flies which show typical habituation after exposure to ethyl butyrate for 3 days (exposed flies show significant drop in performance index compared to naïve flies). This indicates that row-knockdown flies do not get habituated after exposure to ethyl butyrate for 3 days and are repelled by the repulsive odorant, proving these ASD flies to be habituation-deficient.

Figure 1: Aggression arena and representative behavioral patterns. (A) The aggression arena setup in a 24-well plate with a perforated lid for fly entry. Inset: Dimensions of a single arena with fly food in it. (B) Photograph of a single-housing tube containing a single, isolated male fly and a group-housing vial containing male and female flies, recapitulating a social environment. (C) An aspirator for mouth aspiration-based transfer of flies. Items required for assembling a fly aspirator indicating the sides of the tube for fly entry (fly end) and the side that would be inserted into the mouth (mouth end) (C1) and an assembled fly aspirator (C2). (D) Representative male-male aggressive behavioral patterns: approach, wing-threat, fencing, holding, lunging, boxing, and tussling. (E) Fmr1 mutant male flies show reduced aggressive behavior towards other males compared to controls. Box plot showing the number of aggressive bouts for socially isolated males of w¹¹¹⁸ (control) versus Fmr1 trans-heterozygous flies (Fmr1Δ113M/Fmr1Δ50M). Please click here to view a larger version of this figure.

Figure 2: Social space assay arena setup. (A) Diagram showing dimensions and arrangement of glass panes and acrylic spacers. (B) The final arrangement of the arena components where the acrylic spacers are sandwiched between the glass pieces, leaving a small gap for fly entry at the bottom. The triangular space created between glass pieces and spacers is the SSA arena for the flies, here flies are only able to move in two dimensions. (C) Photograph of the final SSA arena setup with flies inside the chamber. Abbreviation: SSA = social space array. (D) ASD model (Fmr1 transheterozygous) flies show increased distancing from each other in the social space assay compared to w¹¹¹⁸. Please click here to view a larger version of this figure.

Figure 3: Courtship arena and representative behavioral patterns. Diagrams showing (A) design and (B) arrangement of the disks in a courtship assay arena. (C) Photo of the courtship arena after final arrangement. (D) Representative images showing courtship behavioral patterns of a naïve male towards a mated female: chasing, orientation, wing flickering, licking, attempt to copulation, and copulation. (E) Plots showing courtship index and number of attempted copulations in two ASD models: (E1) Fmr1 mutant flies showing significant reduction in courtship index and attempt for copulation compared to w¹¹¹⁸ control. (E2) row knockdown flies were used as another ASD model where row-shRNA was expressed in mushroom body neurons (MB247-GAL4); a significant reduction in courtship behavior was observed upon row knockdown. Please click here to view a larger version of this figure.

Figure 4: Arena setup and representative behavioral patterns of grooming behavior. (A) The disk or wheel-shaped arena (same courtship arena shown in Figure 3) is set up on an LED panel with a diffuser for uniform illumination from the bottom and video was captured from the top. This ensures a higher contrast required for identifying fine movement of appendages during grooming. (B) Photographs of grooming behavior patterns observed in adult male flies. T1 = 1^st pair of legs, T2 = 2^nd pair of legs, T3 = third pair of legs. Intuitive abbreviated nomenclature is used for grooming patterns in B: T1-head= head rubbed with first legs; T1-T1 = first pair of legs rubbed together, and so on. Abbreviation: LED = light-emitting diode. (C) Graphs showing grooming assay results performed in Fmr1 trans-heterozygous flies and compared with w¹¹¹⁸. Fmr1 mutant flies showed significantly reduced latency for the first grooming bout indicating the early start of grooming compared with the control. Moreover, mean grooming bout duration and grooming index were significantly increased in Fmr1 mutants, indicative of enhanced repetitive behavior in ASD model flies. In contrast, the total number of grooming bouts was not altered significantly in the mutant flies compared to the control. Please click here to view a larger version of this figure.

Figure 5: Induction and assay setup for olfactory habituation and representative outcome after the assay. (A) Components of the binary odor choice assay setup: (A1) the Y-maze, (A2) the adapter, (A3) the arrangement of these two, (A4) a pair of collection tubes for attaching at the top of the Y-maze arms, (A5) the glass bottle and glass tubes for keeping the odorant solution or water through which air will bubble before passing on to the Y-maze. (B) Fly food bottle setup for induction of olfactory habituation; cartoon diagram (B1) shows how the odorant-containing microfuge tube is hung from a copper wire midway inside the bottle; (B2) finally, the opening of the bottle is covered by cotton and sealed by brown paper envelops. (C) Photo of the total arrangement of the Y-maze setup. (D) Photographs showing the distribution of flies in collection chambers after the end of an experiment. (D1) Naïve flies (exposed to odorless paraffin only, control) show repulsion toward the aversive odorant (20% ethyl butyrate used for induction of habituation in this study) and choose in the arm of the Y-maze filled with regular air whereas (D2) exposed (habituated) flies distribute evenly in both the arms. Abbreviation: ID = internal diameter. (E) Olfactory habituation assay results of row-knockdown flies showed habituation defect. Wild type (Canton S) flies, after 3-day exposure to ethyl butyrate, showed a significant reduction in performance index compared to naïve flies, indicative of habituation. When progenies of LN1-GAL4 x row-shRNA flies were exposed for 3 days, the performance index was not found to be significantly different from that of naïve flies; this demonstrates that these row-knockdown flies did not get habituated upon 3-day exposure to odorant, recapitulating a human ASD-like symptom. Abbreviations: ASD = autism spectrum disorder; shRNA = short hairpin RNA. Please click here to view a larger version of this figure.

Discussion

Drosophila is used as a fine model organism for research in human neurological disorders due to a high degree of conservation of gene sequences between fly and human disease genes⁹. Numerous robust behavioral paradigms make it an attractive model for studying phenotypes manifested in mutants recapitulating human diseases. As hundreds of genes are implicated in autism spectrum disorder (ASD), no common ASD model exists in any model organism. Hence, for each mutant, researchers must first establish ASD-like phenotypes in the fly. In this paper, we have tried to streamline the process by proposing the use of five fly behavioral paradigms in concert for such assessment. These paradigms are employed in flies to assess major and commonly occurring behavioral deficits of human ASD patients: social and communication deficits, enhancement of repetitive behaviors, and a habituation defect. If a fly gene mutation shows defects in all or most of these five paradigms, it could be a strong indication of an ASD-like scenario.

Most of the behavioral paradigms described here have been used multiple times in previous literature for solving diverse research questions. It is not uncommon to find variations in the techniques among different studies. The tweaks in the techniques were made to match the research need and to suit the demand of the specific question being solved. For example, in the courtship assay, innate as well as learned courtship behavior can be quantified towards a tester fly which can be an immature male⁶⁵, a virgin female^66,67, a premated female^68,69,70, a freeze-killed virgin female⁶⁵, or even a decapitated female¹⁹. In the method described in this paper, wild type pre-mated females have been used as a tester to check the courtship pattern of a male ASD model fly. The sexual receptivity of a premated female is relatively lower than that of a virgin fly⁷¹. As a result, the number of male courtship bouts and their duration are more towards a mated female than towards a virgin; similarly, the latency to copulation is also longer than those of virgin tester females in which case early copulation leads to reduced courtship index, reducing the scope for effective comparison between genotypes. Hence, the use of premated female as tester provides a better comparative analysis of courtship behavior of an ASD fly vs a wild type fly. On the other hand, the use of virgin females may provide additional information that may be masked when mated females are used. The use of mated females may also influence and alter male courtship behavior. In sum, it is suggested that researchers choose their experimental strategy and detail depending on the necessity of the specific experiment.

One needs to be cautious here before applying the technique in any or all fly homologs of ASD genes. All these behavioral paradigms are regulated by a specific and small number of neurons. Expression of a certain gene must be established first before proceeding to a specific behavioral paradigm. For example, if the fly homolog of an ASD-associated gene is not expressed in mushroom body neurons or glutamatergic neurons, then courtship and grooming behavior may not be affected in its mutant.

Another point researchers need to be careful about is that the lack of a defect does not always mean that the mutation has no role. For example, in the grooming behavior experiments with Fmr1 mutants, the total grooming bout count is not significantly lower in mutants than in controls, whereas other parameters like grooming index, etc. demonstrate a defect in grooming behavior. This may be explained by the prolongation of each grooming bout in Fmr1 mutants, resulting in a low count of the total bout number despite the total bout duration being much higher. This stresses the need for quantification of multiple parameters for each behavioral pattern. Moreover, it must be remembered that the term ASD covers a broad range of diseases caused by variations in any one of the hundreds of genes, resulting in heterogeneous phenotypes with various degrees of penetrance among the affected individuals. For example, habituation deficit is observed in some patients but not in all. The same would be expected in fly models of the disease.

For the analysis of aggressive behavior, different types of aggressive parameters were taken into account in this study: lunging, wing threat, tussling, etc. (Figure 1)³⁰. Each bout needs to be identified and counted either as an individual bout type or as a total bout count. It is important to first get acquainted with the aggression patterns of male flies, which are described in^59,72 and shown in Figure 1. While counting aggressive bout numbers, the start and end points of a single aggression event need to be clearly identified; multiple types of aggression patterns might be shown within a single bout, followed by the end of the bout. An aggression bout is considered over when a fly either stops showing any of the aggressive behaviors or movement for 2 s or stops the bout and starts walking away.

Most of these behavioral paradigms have multiple variations and advancements described by multiple laboratories. Each of the specific modifications had been created to meet the needs of a specific purpose. Similarly, automated and algorithm-based quantification or analysis techniques have freed these techniques from the chances of manual errors. In this paper, the most basic techniques have been demonstrated with the least amount of material requirements, in the hope that these techniques can be easily adopted and used by the maximum number of researchers.

Materials

List of materials used in this article

Name	Company	Catalog Number	Comments
Aggression arena:
Standard 24-well plate made of transparent polystyrene			12 cm x 8 cm x 2 cm. Diameter of a single well= 18 cm. Sigma-aldrich #Z707791; depth = 1 cm
Transparent plastic/acrylic sheet			Alternative: a perforated lid of a cell culture plate
Social Space Assay:
Binder clips			19 mm
Glass sheets and acrylic sheets of customized sizes			Thickness = 5 mm
Courtship assay:
Nut and bolt with threading
Perspex sheets of customized shapes			i) Lid: A custom-made round transparent Perspex disk (2-3 mm thickness, 70 mm diameter) with one loading hole at the peripheral region and another screw hole at the center (diameter ~ 3 mm for each); ii) A second transparent thicker Perspex disk (3-4 mm thickness, 70 mm diameter), with 6-8 perforations of diameter 15 mm, equidistant from the center; iii) Base: Same as lid except without the loading hole
Grooming assay:
Diffused glass-covered LED panel			10–15-Watt ceiling mountable LED panel
Habituation and Y-maze assay
Climbing chambers			x2, Borosilicate glass
Adapter for connecting Y-maze with entry vial			Perspex, custom made, measurements in Figure 5A
Clear reagent bottles			Borosil #1500017
Gas washing stopper			Borosil #1761021
Glass vial			OD= 25 mm x Height= 85 mm; Borosilicate Glass
Odorant (Ethyl Butyrate)			Merck #E15701
Paraffin wax (liquid) light			SRL #18211
Roller clamps			Polymed #14098
Silicone tubes			OD = 0.6 cm, ID = 0.3 cm; roller clamps for flow control
Vacuum pump			Hana #HN-648 (Any aquarium pump with flow direction reversed manually)
Y-maze			Borosilicate glass
Y-shaped glass tube (borosilicate glass)			Custom made, measurements in Figure 5A
Common items:
Any software for video playback (eg.- VLC media player)			https://www.videolan.org/vlc/
Computer for video data analysis
Fly bottles			OD= 60 mm x Height= 140 mm; glass/polypropylene
Fly vials			OD= 25 mm x Height= 85 mm; Borosilicate Glass
Graph-pad Prism software			https://www.graphpad.com/scientific-software/prism/www.graphpad.com/scientific-software/prism/
ImageJ software			https://imagej.net/downloads
Timer
Video camera with video recording set up			Camcorder or a mobile phone camera will work
For Fly Aspirator:
Cotton			Absorbent, autoclaved
Parafilm			Sigma-aldrich #P7793
Pipette tips			200 µL or 1000 µL, choose depeding on outer diameter of the silicone tube
Silicone/rubber tube			length= 30-50 cm. The tube should be odorless
Composition of Fly food:
Ingredients (amount for 1 L of food)
Agar (8 g)			SRL # 19661 (CAS : 9002-18-0)
Cornflour (80 g)			Organic, locally procured
D-Glucose (20 g)			SRL # 51758 (CAS: 50-99-7)
Propionic acid (4 g)			SRL # 43883 (CAS: 79-09-4)
Sucrose (40 g)			SRL # 90701 (CAS: 57-50-1)
Tego (Methyl para hydroxy benzoate) (1.25 g)			SRL # 60905 (CAS: 5026-62-0)
Yeast Powder (10 g)			HIMEDIA # RM027
Fly lines used in the experiments in this study:
Wild type (Canton S or CS)			BDSC # 64349
w1118			BDSC # 3605
w[1118]; Fmr1[Δ50M]/TM6B, Tb[+]			BDSC # 6930
w[*]; Fmr1[Δ113M]/TM6B, Tb[1]			BDSC # 67403
MB247-GAL4 (Gaurav Das, NCCS Pune, India)			BDSC # 50742
LN1-GAL4			NP1227, NP consortium, Japan
row-shRNA			BDSC # 25971