|Home | About | Journals | Submit | Contact Us | Français|
The embryonic zebrafish is an excellent research model to examine the neural networks that coordinate locomotive behavior. It demonstrates robust locomotive behavior early in development, its nervous system is relatively simple and accessible compared to mammalian systems, and there are mutants available with specific molecular and motor deficits. We have developed a series of four exercises that provide students with a basic understanding of locomotive behavior development, nervous system organization, development of neurotransmitter responsiveness, and genetics. The first two exercises can be performed in one 3-h laboratory period, and the third and fourth exercises, which build on the first two, can be completed in one or two subsequent periods. In the first exercise, students observe and quantify two distinct behaviors that characterize different developmental stages, spontaneous movement, and touch-evoked tail coiling. In the second, the students use a pharmacological approach to determine if the neurotransmitter glycine is required for the embryo to perform each behavior. In the third, they use simple lesions to assess whether the brain is required for each type of behavior. In the fourth, the students examine bandoneon, a zebrafish motility mutant that has a glycine receptor defect, by observing its behavior during spontaneous movement and touch-evoked tail coiling, performing lesions, and applying pharmacological drugs. These exercises are readily adaptable, such that portions can be omitted or expanded to examine other neurotransmitter systems or later stages of locomotive behavior development.
Locomotive behavior in vertebrates, such as walking or swimming, relies upon neural networks in the brain and spinal cord. The embryonic zebrafish has several features that have made it an increasingly popular research system to examine how these networks develop and function.1–3 Zebrafish embryos demonstrate robust locomotive behavior, develop rapidly and external to the mother, and are amenable to pharmacological drug application. Further, there are mutants available with specific locomotor behavior deficits. In the series of laboratory exercises that we describe here, these characteristics are leveraged to illustrate how the nervous system develops and controls locomotive behavior.
The development of locomotion in vertebrates is composed of a characteristic sequence of behaviors.4 Zebrafish embryos progress rapidly through these behaviors (Fig. 1). The first behavior observed, beginning at 17 hours postfertilization (hpf), is spontaneous movement, during which embryos repeatedly perform slow, alternating tail coils. Beginning at 21 hpf, embryos first react to touch and demonstrate rapid tail coils in response to touch stimuli. Finally, at 27 hpf, the embryo begins to exhibit swimming behavior, in which it can move itself forward by at least one body length. In subsequent stages of development, embryos move faster and acquire different swimming behaviors, such as burst swimming and beat-and-glide swimming.5,6 These different behaviors reflect the development of locomotor networks in the brain and spinal cord. Using simple lesions, pharmacological drug application, and a locomotor mutant, this lab investigates the role of the brain and the neurotransmitter glycine in the first two behaviors, spontaneous movement and touch-evoked tail coiling.
Compared to mammalian systems, the embryonic zebrafish hindbrain and spinal cord is relatively simple, with fewer cells and fewer axon tracts than mammalian systems.7,8 The hindbrain, composed of seven segments or rhombomeres, is known to integrate sensory information and initiate locomotive behavior.9,10 A major component of the zebrafish hindbrain is the roughly 50 reticulospinal neurons, whose axons project into the spinal cord to regulate spinal cord network activity. These axons begin to grow into the spinal cord at 22 to 24 hpf, well after spontaneous movement and touch-evoked tail coiling has begun.11 In the embryonic zebrafish spinal cord, systematic efforts to identify neuronal classes revealed only motor neurons, seven types of interneurons, and Rohon-Beard sensory neurons.12–14 One study examined if the hindbrain is required to perform spontaneous movement, touch-evoked tail coiling, and swimming, by performing simple lesions and observing the behavior of tails from which brain input has been removed.15 Using these “spinalized” preparations, it was found that spinal cord networks can mediate each of these behaviors independent of the hindbrain, although swimming behavior is more difficult to elicit. The lesion component of this lab exercise will recapitulate aspects of this study to demonstrate these findings.
Glycine, classically thought of as an inhibitory neurotransmitter, plays a central role in the locomotor networks in the hindbrain and spinal cord.16–18 In several different vertebrate preparations, it has been shown to provide inhibition within the hindbrain to control reticulospinal neuron output to the spinal cord. Within the spinal cord, glycine has been found to coordinate the alternating left and right rhythms that are essential for locomotor behavior4,19 (Fig. 2). Glycine exerts these effects through its receptor, a pentameric ligand-gated ion channel that selectively fluxes chloride ions (Fig. 3). Since glycine is important in both hindbrain and spinal cord networks, interrupting glycine receptor function disrupts motor behavior. Pharmacological blockade of glycine receptors with the competitive antagonist strychnine causes severe muscle convulsions, and mutation of glycine receptor subunits can result in hereditary motor disorders in humans, horses, cattle, mice, and zebrafish.16,20,21
In zebrafish embryos, bath application of pharmacological drugs that target the glycine receptor alters motor behavior differently depending on the stage of development. Application of strychnine during touch-evoked tail coiling disrupts the alternating muscle contractions that constitute the alternating tail coils. Instead, the embryos compress along the rostral-caudal axis, like an accordion, due to simultaneous muscle contraction of the left and right sides.15,22 Application of glycine at these stages results in paralysis, caused by global activation of glycine receptors. Exposing embryos to these drugs earlier in development, during spontaneous movement, generates different behavior. Instead of causing accordion-like behavior, strychnine decreases the rate of spontaneous tail coils, while glycine application at this stage also leads to paralysis. The cellular and molecular mechanisms behind this finding are still unclear, but the different responses to strychnine at 19 and 24 hpf likely reflect the different roles of glycine during these two phases of locomotor network development.
Examination of a locomotor mutant reinforces the observations obtained using pharmacological drugs. bandoneon (beo) is a mutant recovered from large-scale mutagenesis screens to identify genes essential for locomotor behavior.22 In subsequent studies, it was found to harbor a mutation in glrb2, a glycine receptor subunit required for its proper subcellular localization.23 Consistent with a disruption in glycinergic neurotransmission, beo mutants perform accordion-like compressions indistinguishable from wild-type embryos treated with strychnine during the touch-evoked tail coiling phase of development.22 Moreover, beo mutants are resistant to glycine application, and they continue to move at glycine concentrations that paralyze wild-type embryos. beo mutants do not show a motor defect during the spontaneous movement phase of locomotor behavior, indicating that the glrb2 subunit is not required at this stage of development.
We have developed a series of laboratory exercises that use a combination of lesion analysis, pharmacological drug application, and beo mutants to illustrate how the spinal cord and the neurotransmitter glycine play essential roles in locomotive behavior development. These exercises are modular, and different combinations can be used that fit within one or more 3-h time periods. These exercises can also be tailored to individual interests and can be expanded to examine additional neurotransmitter pathways or later stages of development. The main learning objectives of this lab are
Information about zebrafish husbandry, breeding, staging, egg collection, and common chemical solutions can be found in The Zebrafish Book.24 The Zebrafish Book and a wealth of information about zebrafish can be obtained from the zebrafish community website, http://zfin.org. Additional information can be found on the zebrafish education website, http://zfic.org.
For this laboratory exercise, zebrafish adults should be maintained and bred in a nearby facility. Virtually any wild-type strain can be used to obtain wild-type embryos. The beo line is available from ZIRC, the Zebrafish International Resource Center (http://zebrafish.org). Heterozygous beo adults should be acquired and bred to obtain homozygous mutant embryos.
Embryos should be collected, and the fertilized eggs sorted and transferred into large 100-mm Petri dishes filled with E3. Roughly 160 embryos should be collected for each group of students the day before the lab. The embryos should be staged and sorted so that each student group has approximately 80 embryos at 19 hpf and 80 embryos at 26 hpf for the lab period. These two time points were chosen because 19 hpf is peak for the spontaneous tail coiling, and by 26 hpf the response to touch is evident with little spontaneous movement. However, embryos at any age between 17 and 20 hpf can be used to demonstrate spontaneous movement, and embryos between 26 and 30 hpf can be used to demonstrate response to touch. Embryo development can be slowed by placing them at room temperature, instead of 28.5°C. The following calculation can be used to estimate the developmental stage according to the temperature in which it is raised:
where HT is the hours of development at temperature T, and h = the hours of development to reach the stage at 28.5°C.25
Small (60×15 mm) and large (100×15 mm) Petri dishes (many scientific suppliers, e.g., VWR [West Chester, PA] cat. nos. 25384-090 and 25384-070)
Dumont forceps #5 (Fine Science Tools [Foster City, CA] cat. no. 11250-20)
Razor blades (e.g., VWR cat. no. 55411-050) or scalpel blades (e.g., Fine Science Tools cat. no. 10010-00)
Surgical blade holder (optional) (e.g., Fine Science Tools cat. no. 10060-13)
Dissecting stereomicroscope (many scientific suppliers)
Straight Probe (Fine Science Tools cat. no. 10140-01)
Plastic transfer pipettes (Fisher [Pittsburg, PA] cat. no. 13711-7)
E3 (5mM NaCl, 0.17mM KCl, 0.33mM CaCl2, 0.33mM MgSO4)
1× Ringer's solution (116 mM NaCl, 2.9 mM KCl, 1.8 mM CaCl2, 5mM HEPES)
50 mM Glycine (EMD BioSciences [Gibbstown, NJ] cat. no. EM-4810) in Ringer's solution
20 mM Strychnine (Sigma-Aldrich [St. Louis, MO] cat. no. 50532-256) in Ringer's solution
4% Tricaine (Ethyl 3-aminobenzoate methanesulfonate salt, pH to 7.4; Sigma-Aldrich cat. no. A5040)
bandoneon (beo) embryos
Depending on the level of the student, there are several scientific articles directly relevant to this lab exercise. Granato et al. (1996) outline the screen in which the locomotive mutants were characterized.22 Brustein et al. (2003) give a nice overview on development of motor behavior in the zebrafish.1 Hirata et al. (2005) characterize beo as a glycine receptor mutation.23 Lastly, Downes and Granato (2006) examine supraspinal input and its role in glycine-mediated behavior.15
The lab is divided into the following four exercises. In Exercise 1, students are responsible for dechorionating and staging the embryos. The students are then introduced to spontaneous tail coiling at 19 hpf and touch-evoked tail coiling at 26 hpf. The following exercises build upon these two different stages of motor behavior. Exercise 2 focuses on the use of pharmacologic drugs, glycine and strychnine, to elucidate the role of glycine in generating these two different motor behaviors. In Exercise 3, the students remove the head and analyze spinalized preparations to examine the distinct roles of the brain and spinal cord during 19 and 26 hpf (Fig. 4). Finally, Exercise 4 uses beo, a mutant known to have a mutation in the glycine receptor, as another tool to examine the role of glycine in these two motor behaviors. As previously stated, these are modular exercises, each of which builds on previous one(s). Exercise 1 can be performed alone or in combination with 2, and as long as both 1 and 2 have been completed, either Exercise 3 or 4 (or both) can be done. Exercises 1 and 2 can be performed in one 3-h laboratory period, while Exercises 3 and 4 can each be done in one or more additional laboratory periods. We recommend that students work in groups of two to four. See Appendix for a step-by-step approach to each of the exercises.
We typically have students tabulate all data using Excel, either in the lab or later, entering the data that were recorded in their laboratory notebook. Using Excel, they can calculate the average, standard deviation, and standard error of the mean from five embryos from each group. These data can also be graphed. The primary goals of each exercise are to compare differences in spontaneous activity and touch-evoked tail-coiling between the different stages of development (Exercise 1); compare wild-type, drug-treated embryos with wild-type untreated embryos (Exercise 2); compare spinalized preparations with nonlesioned embryos (Exercise 3); and compare beo mutants with wild-type embryos (Exercise 4). For 19 hpf embryo data, the students should tabulate the frequency of tail coiling for control wild-type embryos (C), wild-type treated with strychnine (S), wild-type treated with glycine (G), and beo mutants. To examine the response to touch, a table can be made that indicates a positive or negative response as well as any comments on the qualitative nature of the touch response (e.g., rostral-caudal compressions) at 19 and 26 hpf for all conditions (C, S, G, spinalized for 26 hpf, and beo). The students can then include tables and graphs in a formal lab report. We typically recommend that the students write lab reports in the style of a research article, including an introduction summarizing motor behavior development in zebrafish; a detailed methods section; a results section with figures, tables, and graphs; a discussion of their data in the context of the neural circuitry of zebrafish embryos and previously published results; and references. Statistical analysis could also be performed on the data if the students have a background in statistics. ANOVA or a Student's t-test can be used for comparison with wild type. The instructor or the students may also pool data from the whole class as a basis for comparison with individual group data to demonstrate the power of large sample sizes. The students could also be prompted to use the following questions as guidelines in writing up their lab reports:
How do spontaneous movement and touch-evoked motor behavior change through development?
What are the defining features of zebrafish locomotive behaviors at 19 and 26 hpf?
How do strychnine treatment and glycine treatment affect behavior?
Are these results consistent with each other?
Relate your findings to the zebrafish neural circuitry.
What were your controls and what do they control for?
What do lesions tell us about spontaneous and touch-evoked movement?
How does the beo mutant behavior compare to the behavior of strychnine-treated embryos?
What is the result of treating beo mutants with glycine? Explain.
What were your controls and what do they control for?
In Exercise 1, students should observe spontaneous tail coiling from embryos at 19 hpf, which decreases in frequency after 21 hpf. The response to touch is evident only after 21 hpf. We recommend using embryos 26 hpf or later since much of the spontaneous coiling has stopped, which makes it easier to distinguish the touch response. In Exercise 2, application of strychnine at ~ 19 hpf results in a decrease in the number of tail coils per minute compared to the control. Application of glycine leads to paralysis at both 19 and 26 hpf. However, the response to strychnine at 26 hpf differs from that at 19 hpf. The strychnine-treated embryos exhibit the accordion phenotype, a rostral-caudal compression along the length of the embryo, in response to touch. In Exercise 3, spinalized embryos at both time points exhibit a similar phenotype to the nonlesioned embryos, indicating that supraspinal input is not required for these two behaviors. However, the reliability of the response to touch at 26 hpf may be slightly diminished. In Exercise 4, the beo mutant exhibits an accordion-like phenotype similar to that of strychnine-treated embryos after 24 hpf. Application of glycine has little effect on the beo mutant, reflecting the defect in the glycine receptor subunit.
The overarching goal of this series of laboratory exercises is to teach students to think about the neural basis of simple behaviors in zebrafish embryos, with an emphasis on the role of the neurotransmitter glycine in mediating these behaviors. Exercises 1 and 2, which can be done together in a single 3-h laboratory period, can achieve these goals. These exercises require the students to, at minimum, make and record observations, quantify behavior, and interpret the results of pharmacological manipulations. Depending on the level of sophistication required by the instructor in analysis and interpretation, the students may also perform statistical analysis of their quantitative data. Exercises 3 and 4 teach simple surgical techniques and an appreciation for the power of genetic mutants. Several simple models have been used to describe the neural basis of locomotor behavior in the zebrafish,1,15 and students at many levels should be able to analyze their results in the context of these models.
The authors have collectively performed these experiments several times, with and without labs full of students. While we have not performed any sort of assessment of their impacts on student learning, our feedback has been uniformly positive. In our experience this is one of the most straightforward, enjoyable, and successful teaching labs we have administered.
While these exercises give students an opportunity to think critically, perform quantitative analysis, and compare their results with the published models of locomotor circuitry, it is becoming increasingly clear that to really engage our students and teach them how to ask the right questions (and then seek out the answers to those questions) it is critical that we provide them with inquiry-based learning opportunities.26 An important goal of this lab is to give students the skills that they need to design experiments of their own, such as studying swimming, a more complex behavior that is observed beginning around 27 hpf, or examining the roles of other neurotransmitters at these and other stages of development.
Instructor Note: This may also be done before lab if time is limited.
Instructor Note: Three embryos instead of five can be used if short on time. Each group's data can be pooled at the end to increase the sample size.
Instructor Note: The intent is to recognize that evoked movement is not detected in embryos earlier than 24 hpf.
Instructor Note: Strychnine is a poison. Reinforce laboratory safety when performing experiments with drugs. Students should be sure to wear gloves and safety goggles when handling the pharmacological drugs. After use, strychnine should be disposed of with hazardous waste.
Instructor Note: This should be done with embryos in a large Petri dish using a dissecting stereomicroscope. Ensure only the tip of the tail is cut off (Fig. 4). We do not typically use tricaine for head removal; however, it may be desirable in some cases. The effects of tricaine are rapidly reversible, so it is not likely to affect the outcome of the experiment, but we recommend testing this before doing the experiment in a classroom setting.
Instructor Note: Lesioning should be performed after embryo staging, but at least 20 min before analysis.
Instructor Note: The final concentration of tricaine should be between 0.0016% and 0.04%. It is important to add enough tricaine to ensure that the embryos do not move, but too much tricaine will lead to death.
Instructor Note: The diagram depicting where the lesion should be made may be helpful (Fig. 4).
Instructor Note: This portion of the lab is a good place to review genetics. Also note that the beo phenotype is not exhibited until approximately 24 hpf.
Instructor Note: The students should appreciate the concept that both the neurotransmitter and its receptor are necessary to generate a functional neural circuit.
We thank members of the Downes Lab and Dr. John Nambu for their critical reading of the manuscript. We also thank Dr. Michael Granato for his contribution in establishing these laboratory exercises.
No competing financial interests exist.