Modeling Fetal Alcohol Spectrum Disorders in Zebrafish to Characterize the Impact of an Adverse Embryonic Environment on Adult Social Behavior

Abstract

Fetal alcohol spectrum disorders (FASD) describe all alcohol-induced birth defects. Birth defects such as growth deficiencies, craniofacial, behavioral, and cognitive abnormalities are associated with FASD. Social difficulties are common behavioral abnormalities associated with FASD and often result in serious health issues. Animal models are critical to understanding the mechanisms responsible for ethanol-induced social defects. Zebrafish are social vertebrates that produce externally fertilized transparent eggs; these characteristics provide researchers with a precise yet simple procedure for creating the FASD phenotype and an innate behavior that can be leveraged to model the social deficits associated with FASD. Thus, zebrafish are ideal for characterizing the social deficits of FASD. The goal of the current protocol is to provide the user with a simple behavioral assay that can be used to characterize the consequences of a negative environment early during development and the effects it can have on social behavior in adulthood. The protocol can be used to characterize the effect mutations or teratogens have on adult social behavior. The protocol outlined here demonstrates how to characterize the social behavior of individual fish during a 20-min social assay. Furthermore, the data obtained using the current protocol provides evidence that the protocol can be used to characterize the effects of embryonic ethanol-induced social defects in adult zebrafish.

Introduction

Prenatal alcohol exposure can lead to a variety of birth defects collectively known as fetal alcohol spectrum disorders (FASD)¹. Impaired behavior, such as social difficulties, are common birth defects associated with FASD^2,3. Unfortunately, social difficulties frequently result in serious mental health issues⁴, which can adversely affect the quality of life for individuals with FASD. Thus, understanding the mechanisms responsible for ethanol-induced social defects is paramount.

Zebrafish have biological and behavioral characteristics which make them well suited to advancing our understanding of the mechanisms responsible for ethanol-induced social defects. For instance, zebrafish produce large quantities of transparent externally fertilized eggs; these biological characteristics allow researchers to easily create precise and replicable FASD phenotypes⁵. To expose embryos to ethanol at 24 h postfertilization (hpf), one simply has to use a dissecting microscope to examine the transparent egg and stage the embryo based on previously published work such as Kimmel et al.⁶, then place the egg in the desired ethanol concentration for the desired duration. Since the chorion is a weak barrier to alcohol⁷, the ethanol readily bathes the embryo. To stop the exposure, one simply has to remove the eggs from the ethanol solution. Besides providing researchers with a simple yet accurate method for creating FASD phenotypes, zebrafish also allow researchers to make genetic comparisons to humans because 70% of human genes have a zebrafish orthologue, thus they are a valuable tool for understanding human diseases-related genes⁸. Additionally, unlike other animal models zebrafish form social groups⁹ called shoals¹⁰. Shoaling behavior can be used to characterize the effects embryonic ethanol exposure has on social behavior¹¹. Furthermore, in zebrafish a social response can be elicited by using computer controlled social stimuli¹² or a live social stimulus¹³.

Previous works have characterized the social response of adult zebrafish in groups¹⁴, however a limitation of this approach is the inability to correlate the behavior of an individual fish with a specific measure such as changes in neurotransmitter levels¹¹. The following protocol will give users the ability to characterize the social behavior of an individual adult zebrafish. Since social behavior is acquired for individual fish, users of the protocol can now correlate the acquired behavioral profile of each fish with a dependent outcome. For example, previous work has shown that embryonic ethanol exposure impairs the dopaminergic response to a social stimulus¹¹. While the data shown here has used embryonic ethanol exposure as the independent variable, protocol users can characterize the effects other pharmacological treatments or genetic mutations have on social behavior. Furthermore, protocol users are not limited to examine how embryonic treatments alter behavior but can also determine how acute pharmacological treatments in adult zebrafish impact social behavior¹⁵.

Protocol

All methods described here have been approved by the Institutional Animal Care and Use Committee (IACUC) of the University of South Dakota.

1. Zebrafish housing, care, and embryonic ethanol exposure

1. Raise and breed zebrafish as described¹⁶.
2. Ethanol exposure
   1. Choose the appropriate developmental stage at which to conduct the ethanol exposure. In this protocol embryos were exposed to ethanol at 24 hpf.
   2. Place eggs in 1.0% ethanol volume/volume for 2 h¹⁵. A ratio of 1 egg per mL of EM is good practice. Specifically, place embryos in 50 mL of embryo medium (EM; please see Westerfield¹⁶ for the EM recipe) then remove 500 µL of EM and replace with 500 µL of ethanol.
   3. After ethanol exposure, raise embryos as described¹⁷. Assay social behavior when fish are 16 weeks old.

2. Randomization and tank setup

1. Using an online random sequence generator, randomize all trials a priori to conducting the behavioral assays. Ensure that the stimulus side and the treatment groups vary randomly.
2. To avoid any confounding factors such as time of day or day of testing, start and end the behavioral assay at the same time every day and conduct the behavioral testing on consecutive days until all experimental fish have been tested.
3. For this assay (Figure 1), use a 37 L tank (50 cm x 25 cm x 30 cm, L x W x H) with 1.4 L tanks placed outside along the width of the tank as previously described¹⁷.
4. Line the back and the bottom of the 37 L tank with white corrugated plastic to increase the contrast between the experimental fish and the background to improve video tracking.
5. Place the corrugated plastic on the outer wall of the 1.4 L tanks to increase the contrast of the social stimulus for the experimental fish. Finally, place white corrugated plastic between the 1.4 L tanks and 37 L tanks; this opaque barrier is used to prevent the experimental fish from viewing the social stimulus during habituation.
6. Place the camera at a distance that is far enough to capture the entire length of the 37 L tank plus half of the 1.4 L tanks and accurately track the adult experimental fish.
   NOTE: Seeing which side holds the stimulus ensures that researchers have labelled the tracking zones correctly and provides redundancy as a backup.
7. If no infrared tracking is being used make sure to illuminate the 37 L tank. Use any commercially available aquarium hood lights with a 15 W T8 full spectrum lamp.
   ​NOTE: If multiple arenas are being set up use identical aquarium hoods with identical lamps.

3. Conducting the social assay

1. Begin by filling the 37 L tank used for the behavioral assay with water that is identical to the water used in the housing rack. Ensure that the water temperature is within 2 °C of the housing rack.
2. At the end of the day empty the 37 L tank. Begin each testing day with fresh water. Ensure the water level in the 37 L tank and the water level in the 1.4 L tanks are identical. If the temperature of the room does not keep the water in the 37 L tank at 28.5 °C 2 °C, then replace the water with warm water at 28.5 °C.
3. Next set up the zones of interest. Consult user's manual tracking software of choice to create zones. In this protocol along the length of the 37 L tank on the top and the bottom a tape measure was used to mark 5 cm increments. Given that the testing tank is 50 cm, 5 cm increments lead to 10 zones that are 5 cm each.
4. Using the marks made on the 37 L tank as a reference, use the software to construct the zones by connecting the 5 cm point on the top to the corresponding 5 cm point on the bottom. Additionally, create a zone along the bottom and along the stimulus. Customize zones of interest.
5. Use the tracking software in this protocol to measure the distance from a zone and the duration spent in the zone. In this protocol 12 zones were used.
6. Quantify the time spent in zones 1 to 10 and the distance from stimulus and the bottom as Zone 1, as the zone closest to the stimulus while Zone 10 is the furthest away from the stimulus.
7. Next, select the two males and two females that will be used for the social stimulus. Best practice would be to use males and females from the same cohort as the experimental fish. If that is not possible try to find fish that match the strain, age, and size of the experimental fish.
8. Transfer the experimental fish from the housing tank to the testing arena (37 L tank). Use a net to catch the fish in the housing tank. Place the net with the fish in it, into a container with fish water.
   NOTE: Using this approach will reduce the stress on experimental fish while moving between tanks.
9. Place the experimental fish in the center of the testing arena.
10. Once the detection settings are satisfactory based on user's software of choice (see user's manual), begin the 20 min trial. During the first 10 min, leave the opaque barrier between the 37 L tank and 1.4 L tanks in place. This will prevent the experimental fish from seeing the social stimulus, consisting of two male and two female zebrafish, and allow the fish to acclimatize to the testing arena^13,17.
11. After 10 min carefully remove the opaque barriers by pulling them from behind the tank; this will allow the experimental fish to see the social stimulus.
12. Use tracking software to track and quantify the behavior of the experimental fish ^{11, 12, 17} based on the user's preferences and software manual. In the current protocol, quantify the distance from the social stimulus and the bottom of the tank as well as the time spent in all zones.
13. Analyze data using traditional data analysis tools.

Representative Results

Figure 2 has been modified from Fernandes et al.¹⁷ and shows that embryonic ethanol exposure blunts the shoaling response by examining the distance from the stimulus. The data in Figure 2 represents the distance from the social stimulus during the 20 min trial. The Y-axis shows the distance in centimeters while the X-axis shows the 20 min trial broken down into 1 min intervals. The black bar along the X-axis represents the time when the opaque barriers have been removed and the social stimulus is visible to the test fish. Across all groups initially there is a rapid decrease towards the social stimulus once it is made visible, which can be determined by the steep decrease in distance between minutes 9 and 10, however, while control fish remain very close to the social stimulus alcohol treated are not; this suggests that embryonic ethanol exposure affects adult social behavior^12,18.

Figure 3 has been modified from Fernandes et al.¹⁷ and uses data gathered from measuring the distance to the stimulus to further show the effect embryonic ethanol exposure has on the shoaling response. The data in Figure 3 represents the reduction in the distance toward the social stimulus once it is visible. To calculate the average reduction in the distance to stimulus; the average distance to the stimulus during the stimulus period (when the stimulus is visible) was subtracted from the average distance to the stimulus during habituation (when the social stimulus is not visible), thus a larger negative value represents a stronger social response.

Figure 4 shows the time spent in the zones versus the distance to the stimulus. Figure 4 has been modified from Fernandes et al.¹⁷. Figure 4A shows the amount of time (Y-axis) fish spend in each zone (X-axis) while the stimulus is visible. While fish from all groups appear to show a preference for the zone closest to the social stimulus control fish spend almost twice the time in zone 1 compared to alcohol treated fish (Figure 4B). Figure 4C shows that embryonic ethanol exposure did not impair mobility, since there was no difference between groups across the zones. Finally, Figure 4D shows that in the absence of a social stimulus fish do not spend time in the zone closest to the social stimulus. Thus, the results show that control fish approach and stay very close to the social stimulus when visible, while ethanol treated fish do not. Furthermore, the data shown in Figure 2 and Figure 4C suggest that embryonic ethanol exposure does not impair the ability of fish to see the social stimulus (Figure 2) or move (Figure 4C), therefore providing strong evidence that embryonic ethanol exposure impairs social behavior in adult zebrafish.

Figure 1: Schematic of the behavioral apparatus. A 37 L tank was used to assay the social behavior of adult zebrafish (named ZT140) with and without embryonic ethanol exposure. This figure has been modified from Fernandes et al.¹⁷. Please click here to view a larger version of this figure.

Figure 2: Embryonic ethanol exposure blunts the social behavior in adult zebrafish. Average distance between the adult experimental fish and the live shoal plotted for 1 min intervals of the 20 min behavioral session. Mean ± SEM are shown. Control (n = 12); 1% ethanol (EtOH; n = 11). The horizontal bar above the X-axis, from 10 to 20 min, depicts a timeline during which the live shoal is visible to the experimental subjects. The alcohol concentration is shown above the graphs. This figure has been modified from Fernandes et al.¹⁷. Please click here to view a larger version of this figure.

Figure 3: Alcohol treated fish are significantly further way from the live shoal. Bars represent the difference between the distance of fish from the live shoal before and after the live shoal is visible. Larger negative values suggest a stronger response to the conspecifics. Mean ± SEM are shown. Mean ± SEM are shown. Control (n = 12); 1% ethanol (EtOH; n = 11). This figure has been modified from Fernandes et al.¹⁷. Please click here to view a larger version of this figure.

Figure 4: Embryonic ethanol exposure alters the duration of time spent in zone 1 only when the live shoal is visible. (A) Bars represent the time spent in all 10 zones during stimulus presentation. (B) Bars represent the time spent in zone 1 during the stimulus presentation. Zone 1 is the zone closest to the live shoal, while zone 10 is the furthest away from the live shoal. Note the significant difference in the amount of time fish control fish spend in zone. (C) Bars represent the average percentage of time spent in all zones during habituation, the first 10 min. (D) Bars represent the time spent in zone 1 during habituation. Mean ± SEM are shown. Sample sizes are as follows: Control (n = 12); ethanol (EtOH; n = 11). This figure has been modified from Fernandes et al.¹⁷. Please click here to view a larger version of this figure.

Discussion

Zebrafish have a number of biological and behavioral characteristics making them a highly attractive organism for research involving genes, the environment, and behavior^5,19. This protocol gives the end user a relatively simple guide to assay social behavior, multiple ways to quantity the social behavior, and has the potential to link the behavioral responses of individual fish with treatments such as embryonic ethanol exposure, genetic mutations, or other pharmacological substances.

To assay social behavior in adult zebrafish carefully follow this protocol in conjunction with the user's manual of the tracking software of choice. To build the testing arena, one simply needs a 37 L tank with a lid and light, corrugated plastic, two smaller tanks for the social stimulus, and a camera; most of these items can be purchased at pet stores and any big box retailer. When a single zebrafish is presented with a group of zebrafish the natural behavior of the single zebrafish is to reduce the distance to the group while increasing the time spent in close proximity to the group¹⁹, this behavior is called a shoaling response¹⁹. The shoaling response is a social behavior that can be characterized in two ways: first, when the social stimulus is visible measure the distance between the social stimulus and the single fish¹² and second measure the time spent in the zone closest to the social stimulus when it is visible. Having the ability to quantify the time zebrafish spend in a 5 cm zone when the social stimulus is visible provides strong evidence of the social response given that these fish on average are approximately 4 cm in length.

Dividing the 50 cm tank into 10, 5 cm areas aids in statistical analysis because the probability of the test fish being in any zone is the same. At first glance, having two measures for the same outcome appears to be redundant. However, redundancy can provide validation. Additionally, the distance and duration measures during habituation can be used to determine whether a treatment affects vision or movement. This protocol used commercially available automated tracking software^13,18 but it is amenable to other tracking systems or even manual tracking by a trained observer.

Regardless of the tracking software used there are critical steps to be taken to optimize tracking. First, select a background that provides a significant contrast compared to the fish being recorded. Wild-type zebrafish have horizontal stripes that are golden and blue²⁰; they also have mosaics of yellow xanthophores, silvery or blue iridophores, and black melanophores²⁰. Thus, when characterizing the behavior of AB wild-type fish use a white background ^12,13,17, on the other when characterizing the social behavior of a casper zebrafish that lack pigment use a black background¹⁷. Second, make sure that there is sufficient lighting to track the experimental fish. If there are multiple tracking arenas, ensure that the lighting is identical between the arenas. Another critical step is to make sure the water level in the 37 L tank matches the water level in the 1.4 L tanks, to ensure that the test fish cannot swim in areas that do not have the social stimulus. Additionally, ensure that the 1.4 L tank are in frame, when recording the videos; doing this will provide a backup in case manual coding of the data is required or verification of the stimulus side is needed.

Typical issues within the protocol are the subject not being tracked correctly, an insufficient habituation period and a brief time lag when pulling the opaque barriers when multiple arenas are set up. To avoid the subject not being tracked correctly ensure that there is sufficient contrast between the fish and background as mentioned earlier and the software's guidelines are followed. Additionally, if the user's software permits infrared tracking, this can alleviate issues associated with contrast. While 10 min may seem to be a long time for a habituation period, our unpublished work suggests that reducing the habituation time negatively impacts the protocol. Finally, if the opaque barrier between the 37 L and 1.4 L tanks are pulled manually and if multiple arenas are set up all the barriers cannot be pulled at exactly the same time, unless multiple people are pulling the barriers. Alternatively, if only one person is conducting the assay, then counterbalancing which arena barriers get pulled first should be implemented.

Though the protocol is straight forward, the cost of tracking software and the extra time need to characterize fish individually are potential pitfalls. This protocol used commercially available software to track the behavior of experimental fish which avoids observer bias and increases throughput by setting up multiple arenas and thus recording multiple fish. Even though commercial software was used in this protocol, other tracking software is available, including free software. Using this protocol with the end user's tracking of choice will reproduce the social assay. In this protocol the behavior of an individual fish was characterized; others have focused on examining groups of zebrafish¹⁴ which is one way to address the potential pitfall of characterizing one fish at a time. Alternatively, as mentioned earlier, using multiple testing arenas at the same time can increase throughput while maintaining the ability to assign a behavioral profile with an individual fish. Maintaining the ability to correlate a behavioral profile of an individual fish is important because it gives the researcher the opportunity to look for variability within a treatment group. Although embryonic ethanol exposure was used as independent variable in the representative data, future applications of the technique outlined in the current protocol are not limited to simply characterizing the effects of embryonic ethanol exposure on social behavior. For example, one can determine the effect genetic mutations²¹ or other pharmacological treatments²² have on social behavior. Furthermore, besides embryonic ethanol exposure future works can characterize the effect other teratogens have on social behavior. Thus, the current protocol is useful to any researcher interested in understanding how genes and/or the environment affect social behavior.

Materials

Name	Company	Catalog Number	Comments
1.4-l ZT140 Aquaneering tanks	Aquaneering	ZT140	Tanks for social stimulus
Aqueon 20" Deluxe Fluorescent Full Hood aquarium light	https://www.petco.com/shop/en/petcostore/product/aqueon-aquarium-black-24-fluorescent-deluxe-full-hood-215740		Light for the 37-I tank
Aqueon Standard Open-Glass Glass Aquarium Tank, 10 Gallon	https://www.petco.com/shop/en/petcostore/product/aga-10g-20x10x12bk-tank-170917		37-l tank for the social assay
Ethanol	Fisher Scienticfic	BP28184
Ethovision XT tracking system	https://www.noldus.com/ethovision-xt
R-Capable Color Basler GigE Camera	https://www.noldus.com/ethovision-xt
White corrugated plastic	https://www.homedepot.com/p/Coroplast-48-in-x-96-in-x-0-157-in-4mm-White-Corrugated-Twinwall-Plastic-Sheet-CP4896S/205351385		Plastic to line the back and the bottom of the 37-I tank and back of the tanks used for the social stimulus