Quantification of Drosophila melanogaster Grooming Behavior for Evaluation of Excessive Grooming Phenotypes

Summary

The method presented here involves the manual annotation of footage acquired of Drosophila melanogaster for specific grooming behaviors. It allows for quantification of both the number of grooming bouts and overall time spent grooming to evaluate for atypical self-grooming phenotypes.

Abstract

Observable changes in stereotyped grooming are applied translationally in model organisms. These changes are representative of pathologies that elicit similar deviations in human behavior; for example, excessive grooming acts as a proxy for obsessive and compulsive behaviors present in conditions like Tourette Syndrome or obsessive-compulsive disorder. The grooming assay presented allows for the evaluation of abnormal self-grooming phenotypes in Drosophila melanogaster. Flies are recorded for a period of 10 min, and these recordings are observed and annotated blind for previously defined grooming behaviors. Quantitative measures of both grooming bout frequency and the time spent engaging in self-grooming can be obtained by manually annotating the footage. The assay is relatively inexpensive, requires few materials not already available in laboratory environments, and is easily adaptable to fit the specific needs of any given study aiming to observe grooming. Additionally, the low level of skill needed to perform the assay, as compared to computer science-heavy automated methods, makes the protocol well-suited for small labs and students. We discuss in detail the steps required to perform this assay and its present limitations.

Introduction

Drosophila melanogaster is a well-established model organism in behavioral and neurobiological studies, providing insight into mechanisms driving analogous human behaviors. Self-grooming in this organism is a highly regulated and well-defined behavior, following stereotypical patterns that are easily distinguished from one another¹. The separate grooming behaviors exhibited by the fly can generally be classified by anatomical region², most easily being defined as posterior or anterior. Drosophila grooming will initially focus on the anterior region and subsequently transition to the posterior end³. Under typical conditions, flies exhibit grooming behaviors to maintain cleanliness (e.g., by removing dust) and occur in response to exposure to potentially harmful external stimuli like pathogenic microbes⁴.

Abnormalities in grooming behavior, specifically spontaneous obsessive grooming, have been used in various model systems as an indicator of obsessive and/or compulsive behavior. Translational findings observing obsessive grooming behaviors in organisms such as rodents, birds, and canines have given insight into conditions eliciting similar compulsive behavior in humans⁵. These include conditions such as trichotillomania, obsessive compulsive disorder, and Tourette Syndrome⁶. Excessive grooming behavior has also been used as a benchmark in evaluating behavioral phenotypes in models of similar neurodevelopmental conditions in Drosophila melanogaster. Obsessive grooming behaviors have been observed in fly models of Fragile-X Syndrome (FSX) and associated autism spectrum disorder (ASD). Excess spontaneous grooming occurs under mutations of dfmr1, the ortholog to the ASD and FSX associated gene FMR1⁷. There is additionally a notable change in grooming distribution between posterior and anterior ends in these mutants⁸. These changes are interpreted as reflective of obsessive and compulsive body-focused behaviors displayed by some patients with these conditions. In using the grooming assay described here, we observed grooming behaviors in flies after an RNAi-mediated knockdown of the Drosophila gene Atg8a produced by commercially available GAL4 drivers and UAS-RNAi lines⁹.

This method involves the manual annotation of footage taken of flies for specific grooming behaviors. Previous studies aiming to evaluate grooming behavior, such as those using indirect methods like dyes, while effective at quantifying the efficacy of grooming, do not allow for measurement of grooming duration or frequency¹⁰. This assay, however, allows for quantification of Drosophila grooming frequency and duration, both generally and by anatomical region. The method detailed here presents some advantages over current automated methods, as it is easily modified and can be conducted by individuals lacking a computational background. With the equipment required easily available in most laboratories we present a cost-efficient manner to evaluate for the presence of an excessive self-grooming phenotype (see Table of Materials). This makes the method readily accessible to primarily undergraduate institutions and easily adaptable to training environments or teaching laboratories.

Protocol

NOTE: An overview of the protocol is presented in Figure 1.

1. Preparation for filming

1. Place four three-well dishes side by side on a white surface, covering each with a glass slide.
2. Position a camera (here, a Raspberry Pi system, but any camera capable of high-definition video is sufficient) above the dishes.
   1. Check to ensure that all dishes are in frame and focus.
   2. Check to ensure that there is no glare on the slides that will limit the visibility of flies in the dish. Eliminate the glare by adjusting the plate position or dimming the lights in the recording space.
   3. Use high resolution (minimum 1920 x 1080 pixels) to accurately visualize fly limbs.
3. Mark the position of the dishes and camera on the bench where the recording is taking place.
   NOTE: This will limit time spent on step 2 in future use of the apparatus. Performing the assays at the same time and in the same space will limit variability in behavior based on factors beyond intended manipulations.
4. Lightly anesthetize the 4-9 day-old flies with cold on an ice block covered with parafilm and a paper towel (CO₂ anesthetization will have a significant impact on behavior).
   1. Tap the flies onto the block, and once anesthetized, move the flies to observation chambers as quickly as possible using a paintbrush.
      NOTE: This transfer should happen as quickly as possible, as excessive exposure to anesthetization of any kind can negatively impact flies.
5. Place a single fly in each well.
6. Allow files to acclimate to their new environment for 30 min post-anesthetization. Do not touch or disturb the wells containing flies after this point, as added stressors will impact behavior.

2. Recording the flies

1. After the 30-min acclimation, record the flies using a camera.
2. Use a Raspberry Pi camera run through PiSpy¹¹ for behavioral data collection.
   NOTE: The software is free and available on GitHub (https://github.com/gpask/PiSpy), and a Pi system can easily be set up for under $300. Other cameras, if available already, are also sufficient for recording footage
3. Start a recording with Pi with the following command flow.
   1. Open the terminal and type cd PiSpy. Then, type python3 PiSpy.py, enter the desired recording length, and enter the desired frame rate. Select the desired resolution and click Quick Capture.
   2. Use the Preview Camera function to ensure the flies are in focus and sufficiently visible.
4. Record the flies for 10 min at a recommended frame rate of 24 fps and save the file (PiSpy automatically saves to the videos folder).

3. Video analysis (Figure 2)

1. To analyze the video, import the file into the Elan 6.8 software¹².
   NOTE: This software, while originally developed for psycholinguistic tallows for detailed annotation of videos and in-depth analysis of the grooming behaviors
2. Assign each row as illustrated below to annotate the footage of individuals manually.
   1. Do this with the following command flow: Tier > Add New Tier > Type Location in the dish on Tier Name > Add.
3. Depending on the needs of a study, note and quantify various specific behaviors (anterior and posterior grooming, walking, sleeping, and standing).
   1. Make annotations following the steps below.
      1. Double-click the tier to be annotated. Right-click and drag the cursor along the selected tier for a period in which a single grooming bout is exhibited. Left-click, select New Annotation and double-click the newly highlighted time period. Type the abbreviation for the behavior exhibited at that time point.
   2. Observe the videos slowly, scrub through them with a mouse, and annotate. Zoom into a single fly well during annotation to prevent false identification of behaviors. These could include tapping on the edge of the glass by the fly or shifting the legs, which might initially look like grooming.
4. Once the video is completely annotated, obtain a breakdown of each behavior. Ensure that the same abbreviation or annotation is used throughout the videos to use this feature.
   NOTE: Elan 6.8 will provide the number of times a behavior occurs, the total duration of behaviors, and the mean time spent on each behavior.
   1. Define grooming bouts as 2 s of uninterrupted grooming. Identify each behavior observed as follows. Define the parameters for anatomical regions prior to the conduction of the assay.
      1. Anterior grooming: Identify grooming performed at the anterior region of the fly body as anterior grooming.
         NOTE: This behavior will almost always involve a rubbing motion on an anterior part, like the head or proboscis, with the front two legs of the fly.
      2. Posterior grooming: Identify grooming performed at the anterior-posterior end of the fly body as posterior grooming.
         NOTE: Additionally, wings were included in the posterior region for simplicity. This behavior will almost always involve a rubbing motion with the central and most posterior set of legs on the fly. A good indicator that posterior grooming is occurring will be that, at first glance, the fly may look as if it is missing its most posterior legs, as depicted in Figure 2D.

Results

This assay produces quantitative data measuring the time and frequency of grooming behaviors from annotated footage. A representative image of the setup and how behaviors are defined is outlined in Figure 2. Given the subjectivity introduced by video analysis, all annotations of videos should be blinded to the researcher performing the analysis.

This method was used to evaluate the role the Drosophila gene Atg8a has in the development of grooming...

Discussion

Overall, this easily executable and cost-effective assay allows for robust characterization of Drosophila melanogaster grooming behavior. This technique gives insight into the frequency, time spent engaging in, and anatomical distribution of a large number of previously identified grooming behaviors. A good indicator that grooming will occur or is underway is changes in leg placement, specifically raising any of the 6 legs. Taking note specifically of behaviors after leg movement can assist in identifying groomi...

Materials

List of materials used in this article

Name	Company	Catalog Number	Comments
16 GB Micro SD card	Adafruit	1294	For use in introducing the OS
Display	Adafruit	3578	Any display able to be connected by HDMI or flex cable is sufficient
Drosophila Atg8a-RNAi UAS line	Bloomington Stock Center	34340	Line used in representative data
Drosophila Ok6 Driver	N/A	N/A	Line used in representative data
Glass slides	Fisher Scientific	12-550-A3	Keeps flies in containment
Heatsinks	Adafruit		Prevents overheating of the computer
Lens	Adafruit	4563	Used with HD camera only
PiCamera	Adafruit	4561	The HD version was used here but a standard camera can be used under budget constraints
Raspberry pi 4	Adafruit	4292	Computer PiSpy is run on
Ribbion Cable	Adafruit	1648	For use in connecting camera and display
SB components Case	Adafruit	4301	Protects the computer
Spot Plate	Fisher Scientific	S99406	Pyrex versions were already avaliable for our use but cell culture dishes or plastic spot plates are sufficient as well
Tripod	Best Buy	6355959	For suspension and positioning of camera, any apparatus capable of this is sufficient
USB C Power Supply	Adafruit	1995	Supplies power to the computer