Preparation of Conspecific Alarm Cue From Fathead Minnows

Summary

The goal of this protocol is to describe a procedure for a standard preparation of olfactory alarm cue via mechanical homogenization of the epithelial tissue of euthanized fathead minnows (Pimepheles promelas).

Abstract

Chemical alarm cues are powerful tools in behavioral research, particularly in studies of fish within the superorder Ostariophysi. In nature, injured fish release these cues into the water, where they are innately recognized by conspecifics and trigger fear responses such as freezing and hiding. Despite their widespread use, methods for the preparation of alarm cues lack consistency. Even within studies on the same species, preparation techniques and concentration often differ, making it difficult to generalize results across experiments. Here, a standardized protocol is described for preparing and administering alarm cues in a widely used model species, the fathead minnow (Pimephales promelas). To acquire the alarm cue, epithelial tissue samples were dissected from euthanized P. promelas females and homogenized, and then the resulting cue was diluted and filtered. Data are presented from a behavioral experiment comparing refugial use by 14 days post hatch (dph) larvae in the presence versus absence of these olfactory cues to confirm that fish detect and recognize the cues. This procedure is intended to promote consistency in chemical ecology research and facilitate more robust comparisons across studies of alarm cue-mediated behavior.

Introduction

Chemical alarm signaling in fish was first demonstrated by Karl von Frisch in 1938 when he showed that tissue injury in a freshwater minnow (Phoxinus phoxinus) triggered a fear response in nearby conspecifics¹. Upon detecting the alarm cue, the minnows exhibited avoidance behaviors, including swimming away from the injured fish and increasing clustering behavior². These results and those from subsequent studies identified club cells in the epidermis as the source of chemical alarm cues capable of inducing changes in behavior and alertness in conspecifics, with the cues being released into the local environment following the rupture of cell membranes^3,4. Although initially thought to be exclusive to Ostariophysi fishes, chemical alarm signaling has since been observed in other fish orders, such as Perciformes⁵, as well as in other aquatic taxa⁶. Because these responses are innate and thus reliably elicited, chemical alarm cues have become a robust tool for behavioral researchers to investigate antipredator behavior and chemical signaling in fish⁷. However, despite the widespread occurrence of chemical alarm cues across taxa and extensive study of their behavioral effects, the chemical composition of these cues is unknown for most species, which limits researchers’ ability to chemically quantify and standardize cue concentrations⁸.

Fathead minnows (Pimephales promelas, Cyprinidae) are an emerging model species for studies of chemical communication in fish due to their well-characterized behavior and ease of maintenance and breeding in laboratory conditions⁷. Minnows exhibit a range of antipredator behaviors in response to conspecific chemical alarm cues, including dashing, freezing, increased shoaling, and increased shelter use^9,10. These fish also demonstrate associative learning, linking chemical¹¹ and visual¹² predator cues to alarm cues, and extend this learning ability to non-biological stimuli like light and sound^13,14. However, different studies use varying procedures to create alarm cues^{3,9,10,15,16,17},  which makes it difficult to compare results. For example, whereas early studies used both male and female minnows to generate alarm cue¹⁰, male minnows have been reported to have fewer club cells¹⁵, and thus more recent studies have used only mature females (6+ months old)^16,17. Cue concentrations also vary considerably in the literature according to the amount of sample and dilution factor used^8,18,19. Last, laboratory procedures for tissue homogenization and sample filtration (e.g., via glass wool, cotton, or mechanical filtration) differ^10,16,20, which could cause discrepancies in the potencies of preparations made with comparable amounts of tissue.

This article describes a practical and scalable standard procedure for producing adult conspecific chemical alarm cues using mature (6+ months) female fathead minnows with an approximate concentration of 0.05 g of tissue per mL. Data from a larval behavioral experiment are also included to demonstrate that minnows recognize and respond to the preparation at an aqueous concentration of approximately 5 mL of cue per L. The protocol reduces inconsistencies in experimental technique by standardizing the source of chemical alarm cues, tissue preparation and homogenization procedures, and handling and storage conditions, to minimize variation in cue concentration and degradation. By improving reproducibility across preparations, this protocol is intended to facilitate more reliable comparisons among studies and provide greater consistency in chemical ecology research in this model species.

Protocol

This protocol adheres to the guidelines set forth by the Institutional Animal Care and Use Committee (IACUC) at Ball State University and is approved under IACUC protocol number (Approval: 1142896-17). All procedures ensure the humane treatment of animals and comply with institutional and federal regulations for the use of live vertebrate animals in research. The reagents and the equipment used are listed in the Table of Materials.

1. Euthanasia of fathead minnows

1. Put on nitrile gloves and select 15 live, adult female fathead minnows (≥6 months of age). Females are used to standardize cue composition and minimize cue heterogeneity associated with sex¹⁵.
2. Individually or collectively euthanize fish in accordance with approved institutional animal care and use protocols (e.g., rapid chilling or immersion in ≥250 mg/L MS-222 buffered with sodium bicarbonate to neutral pH).
   NOTE: Euthanasia using MS-222 does not alter subsequent recognition of, or behavioral responses to, conspecific alarm cue^14,15.
3. Monitor fish until 30 min after cessation of opercular movement and loss of reflexes.
   NOTE: A secondary method of euthanasia, such as cervical transection, may be applied to ensure death.
4. Dispose of MS-222 solution (if used) in a designated chemical waste container following institutional environmental safety regulations.
5. Transfer euthanized fish to a sterile net and rinse each fish three times with deionized or distilled water.

2. Dissection and isolation of epithelia

1. Place a rinsed fish on a clean, disinfected dissection surface. Using a sterile scalpel, sever the head and tail fin by making deep (approximately 0.5 cm to 1 cm) transverse incisions at the locations of the opercular flap and caudal peduncle. If the head and tail are not completely separated from the body with the first incision, repeat the cut. Remove the head and tail fin and discard these tissues in a biohazard waste container according to approved institutional protocols for the disposal of biological material.
2.  Measure the length of the remaining torso from the posterior edge of the operculum to the caudal peduncle using digital calipers with a resolution of 0.01 mm.
   NOTE: Only retain fish with torso lengths of 3.75 cm ± 0.5 cm to minimize variation in epithelial tissue mass among preparations.
3. Lay the fish on its lateral side with the ventral surface of the fish facing the researcher. While holding the fish in place with the non-dominant hand, make a straight, shallow (approximately 1 mm deep) ventral incision from the anus toward the anterior end of the fish using a scalpel or dissecting scissors.
   NOTE: The cut should be made with sufficient pressure to completely and cleanly separate the epithelial tissue. Do not tear or fray the edges of the skin at the edges of the incision because alarm cue is released from ruptured epithelial cells⁴.
4. Using a small metal spatula or spoon, separate the internal organs from the body cavity and remove them. Place the organs in a biohazard waste container and discard according to approved institutional protocols for the disposal of biological material.
5.  Blot excess moisture from the remaining body tissue with a lint-free laboratory wipe.
6. Weigh the retained body torso using an analytical or top-loading balance.
   NOTE: Only retain fish with post-dissection tissue weights of 1.25 g ± 0.25 g to maintain consistency in cue concentration across preparations.
7. Place the fish body on ice or in a cooler until homogenization.
8. Repeat steps 2.1–2.6 for all fish, retaining and pooling only those individuals that meet both length and mass criteria.

3. Homogenization of tissue

1. Place the dissected epithelial torso tissue from all 15 fish into a 1.42 L sterile blender jar and secure the jar to the blender base.
2. Add 50 mL of deionized or distilled water to the blender jar to facilitate initial tissue breakdown.
3. Homogenize the tissue for a total duration of 5 min at a speed of 22,000 revolutions per minute (RPM) using the pulse mode, which reduces heat generation and preserves cue integrity. Hold the pulse button down for 5 s and then release for the next 5 s. Repeat for the remainder of the blending period.
   NOTE: The slurry may appear to be non-homogeneous at the end of the blending period, and tissue fragments may be visible.
4. Add an additional 350 mL of deionized or distilled water to the blender jar.
5. Continue blending at 22,000 RPM for an additional 2 min using the pulse function as described in 3.3. The slurry should be smooth and uniform at the end of the second blending period, with no large tissue fragments visible.
6. Filter the slurry through a stainless steel 150-mesh sieve (105 µm) positioned over a 500 mL sterile collection vessel. Stir the slurry in the sieve using a glass rod, metal spoon, or metal spatula to facilitate the passage of the water through the sieve without forcing tissue through the mesh.
7. After all of the water has passed through the sieve into the collection vessel, rinse the sieve to remove the solid tissue material and pour the collected filtrate from the vessel through the sieve a second time into another 500 mL sterile vessel to ensure removal of most residual particulate matter.
   NOTE: The final cue solution in the collection vessel should appear homogeneous and free of visible solids, although it may be cloudy due to the presence of microscopic tissue particles (smaller than 105 µm).

4. Aliquoting and Storage

1. Mix the filtered cue solution with a metal spatula, spoon, or glass rod to ensure mixing.
2. Pipette 1 mL volumes of the cue solution into 1.5 mL sterile microcentrifuge tubes.
3. Label tubes clearly with date of preparation, species, and concentration.
4. Freeze aliquots immediately at −20 °C until use.
   NOTE: Microscopic tissue particles (smaller than 105 µm) that passed through the sieve may collect at the bottom of the microcentrifuge tube before the solution freezes. If present after thawing, this collected tissue matter can be removed from the aliquot with a metal spoon or spatula and discarded.

5. Preparation for use

1. Immediately prior to use, remove a single aliquot of frozen alarm cue from the freezer.
2. Thaw the aliquot by placing the tube in a warm water bath (approximately 45 °C to 55 °C) for 5 min, inverting the tube 5–7 times to mix once thawed.
3. Prepare the experimental alarm cue solution by adding 1 mL of thawed cue stock to 200 mL of non-conditioned, aged, and aerated water for use with live fish (i.e., aerated with an air stone for at least 24 h), mixing with a metal spoon, spatula, or glass rod to ensure even distribution.
   NOTE: Volumes may be scaled proportionally depending on need.
4. Use the prepared solution immediately in behavioral assays involving fish. Discard any unused aqueous alarm cue solution at the end of each testing session.
   NOTE: Alarm cue activity degrades within several hours of dilution in water²¹. Do not refreeze thawed aliquots or reuse cue-amended water.

Results

The general experimental schematic is shown in Figure 1. The mean (± SD) weight of the tissues used in this study was 1.12 g ± 0.23 g, which generated a stock solution of conspecific alarm cue with an approximate concentration of 0.05 g of tissue per mL. Concentration may also be expressed as fish-equivalents (by mass) per volume⁸, and this protocol yields a concentration of approximately 0.045 fish-equivalents per mL. The mean length of the fish used in this study was 3.6 cm ± 0.29 cm. Using fish only within this size range further promotes consistency in the amount of epithelial tissue used.

Whether this standard preparation was sufficient for perception and recognition by minnows was tested in a refugium assay involving larval P. promelas^22,23,24. The details of the experimental setup and assay procedure are given in Steinberg (2025)²⁴ and were modified slightly for this study. Briefly, clutches of P. promelas eggs were reared under control conditions in clean water until 14 days post hatch (dph). At 14 dph, larvae were individually tested in a 5-cm diameter arena. Approximately one-third of the arena consisted of a refugium that had an opaque black top, black walls, and a black bottom. A removable door separated the refugium chamber from the remaining arena, which had transparent walls and floor and an uncovered top (Figure 2A). The arena was filled with 10 mL of water amended with alarm cue (alarm cue treatment made from 200 mL aged, aerated water mixed with 1 mL of alarm cue), or clean water (control). The control treatment was included to confirm that the refugium itself did not influence larval behavior (i.e., by making it more or less likely that an animal would emerge). Dedicated arenas were used for trials involving alarm cue vs. control to prevent residual cue contamination. At the start of a trial, one larva was placed in a refugium and allowed to acclimate for 5 min. The door was then removed, and the fish was recorded with a GigE camera for 5 min. Emergence of a larva from the refugium during the trial was noted (binary variable: emergence = 1; no emergence = 0). For each condition (alarm cue and control), the proportion of larvae that emerged was compared to the null expectation of equal emergence (0.50) using exact binomial tests. Each larva was only used once.

Significantly fewer larvae emerged from the refugium in the presence of alarm cue (12 of 35; binomial exact test P = 0.04) than would be expected by chance, indicating that the alarm cue preparation is perceived and recognized by larvae. In the absence of an alarm cue, the number of larvae that emerged from the refugium did not differ from chance (21 of 39 larvae, P = 0.73) (Figure 2B).

Figure 1: Protocol illustration. A schematic outlining a standard procedure for generating conspecific alarm cue from adult fathead minnows, Pimephales promelas. Female fish are euthanized and dissected to isolate the epithelial tissues, which include the epidermal club cells that contain the cue. After dissection, the tissues are homogenized to disrupt the cell membrane and release the cue compound. The resulting slurry is filtered twice using a fine mesh and frozen in 1 mL aliquots until use. Please click here to view a larger version of this figure.

Figure 2: Refugium assay. (A) Schematic design of the experimental setup and procedure. Emergence of larvae from a darkened, sheltered refugium into a bright open space was evaluated in the presence and absence of the alarm cue. (B) Proportions (±95% binomial confidence intervals) of larvae that emerged from the refugium in the presence or absence of alarm cue. Asterisks indicate a significant difference from the null expectation of 0.5 in binomial tests (n = 35 and 39 larvae in the alarm cue and control treatments, respectively). Panel A was created in BioRender. Please click here to view a larger version of this figure.

Discussion

Only tissues from female P. promelas are used for this procedure because previous work has suggested that males have fewer alarm cue-producing epidermal club cells^10,15,25. Following euthanasia, epithelial tissues should be thoroughly rinsed before dissection to remove contaminants. During dissection, care should be taken to ensure that the internal organs are removed as thoroughly and cleanly as possible, leaving only the epidermal covering of the fish's body. However, it should be noted that a limitation of the method is that small amounts of non-epidermal tissue may be included in preparations, with the potential to contribute additional biochemical components.

A second limitation is that filtration through the 105-µm sieve may not remove all particulate matter. After homogenization, particles smaller than 105 µm that pass through the sieve may begin to settle during pipetting of the solution into aliquots. To ensure consistent alarm cue concentration across aliquots, the solution may be periodically remixed during this process. One issue that may be encountered because of this limitation is that a small pellet of tissue material may be present in the microcentrifuge tube after the aliquot is thawed. After gently inverting the tube to ensure thorough mixing, this remaining solid material can be removed and discarded. Importantly, alarm cue efficacy is sensitive to both heat and time, and prolonged exposure to either can reduce or eliminate cue effectiveness^21,26. Thus, to minimize the potential for thermal degradation prior to processing, euthanized fish should be kept on ice or stored in a cooler before and after dissection. Heating may also occur during tissue homogenization; for this reason, the pulse function of the blender should be used to limit heat buildup. To minimize degradation associated with time, alarm cue solutions should be frozen at –20 °C immediately after aliquoting and thawed only immediately prior to use. Although the results of the current study, as well as prior research^16,17,27 indicate that alarm cues of cyprinid fishes continue to elicit reliable responses from animals after being frozen at –20 °C for 3–6 months⁸, related studies²¹ indicate that alarm cue activity degrades in aqueous solution after approximately 3 h. Therefore, fresh solutions should be prepared if behavioral testing extends beyond this period.

This protocol is intended to standardize procedures for the generation of P. promelas alarm cues, but biological variability may exist that influences the potency of cues and/or organismal responses to them. This protocol standardizes donor sex and size, but the diet, physiological state, or epidermal club cell density of the donor fish could affect the concentration of alarm cue in the preparation^28,29,30. To minimize the potential for such variability, alarm cues could be generated from fish originating from the same population, bred in the lab, and maintained under standardized laboratory conditions²⁹. In addition, factors related to how the cues are administered may influence how animals respond to them. For some experiments, potential confounds may be avoided by filling the experimental arena or tank with an aqueous solution of water and alarm cue before introducing the focal fish, as described in this study. However, if the experiment requires the administration of the cue after the fish is in the arena, researchers should consider the potential effects of disturbance associated with cue delivery (e.g., water movement, handling, or visual cues), which should be minimized or appropriately controlled. In such cases, researchers should also consider the time required for alarm cues to diffuse throughout the experimental environment. Careful attention to these factors will improve interpretability and comparability of results across studies⁸.

Experimental use of chemical alarm cues has facilitated research across diverse contexts, including predator–prey interactions¹⁰, learning³¹, development³², social behaviour¹⁷, and conservation biology³³. However, despite their widespread and growing use, methods for generating chemical alarm cues continue to vary considerably across studies^16,25,34,35, and even among the studies of the same species, conducted by the same lab^16,17,24. The protocol presented here provides a validated and standardized approach for generating alarm cue from fathead minnows that is intended to promote greater consistency, comparability, and reproducibility across studies involving this model species.

Materials

List of materials used in this article

Name	Company	Catalog Number	Comments
150 mesh sieve (105 micron opening)	Amazon	B0CC2JT5Q2	Used to filter debris from slurry
Analytical balance	JM Science	JM-708	Used to measure weight of fish
Beaker (1000 mL)	Amazon	B082L53V4J	Used for liquid containment
Beaker (250 mL)	Amazon	B006VYXZ4M	Used for liquid containment
Beaker (300 mL)	Amazon	B01JM3B20C	Used for liquid containment
Blender	Amazon	B0758JHZM3	Used to homogenize tissues
Containment tray	Amazon	B0C4M2174M	Used to contain ice and fish bodies
Lab spoons	Fisher Scientific	14-375-10	Used to transfer materials
Lab wipes	Fisher Scientific	06-666A	Used to blot excess moisure from tissues
Microcentrifuge tubes (1.5 mL)	Fisher Scientific	14-666-318	Used to store frozen alarm cue
MS-222	PlayItKoy	TRIC-M-GC-0100	Used for euthanization
Nitrile gloves	Fisher Scientific	19-149-863C	Used for personal protection
Ruler	Fisher Scientific	S43612	Used for measurement of fish bodies
Scalpels	Fisher Scientific	S17800	Used for dissection of fish bodies
Small Dissecting Scissors	Fisher Scientific	08-940	Used for dissection of fish bodies