Abstract
OBJECTIVE
To evaluate if opioid-induced behavioral effects, such as sedation, can be detected using a shuttle box experimental apparatus and whether thermal preference following noxious stimulation using mustard oil is reversed by morphine administration in fish.
METHODS
5 goldfish (Carassius auratus) underwent 2 randomized blinded experimental trials, with a crossover study design. First, opioid effects were tested in a shuttle box without painful stimulus. Fish were injected 5 days apart with butorphanol at 0.4 or 10 mg/kg, morphine at 5 or 10 mg/kg, or saline IM. After 30 minutes, each fish was placed in a shuttle box for 2 hours with a temperature gradient of 26 to 28 °C. Temperature preference, time spent immobile, and swimming velocity were assessed. The second trial consisted of cutaneous noxious stimulation using mustard oil immersion for 5 minutes followed by an assessment of thermal preference for 4 minutes in the shuttle box after either morphine at 10 mg/kg or saline IM injections. Linear mixed models were used to compare treatment groups.
RESULTS
Before noxious stimulation, a low dose of morphine caused sedation compared with control group and high-dose morphine and butorphanol treatments. Immersion in mustard oil caused fish to spend more time in the cold area in the control group. Morphine administration reversed this pattern.
CONCLUSIONS
The sedative and analgesic effects of opioids were detected through this model.
CLINICAL RELEVANCE
The shuttle box model could be used to assess the analgesic effects of other opioids in goldfish while reducing biases associated with the sedative and stimulatory effects of these drugs.
Fishes are the third most frequent companion animals in the US1 and the second most common group of animals used in Canadian research laboratories after rodents.2 Given this popularity as both laboratory and companion animals, ethical questions regarding appropriate welfare practices3–7 and studies exploring their clinical care are increasing.8–10 This includes investigations of nociception, the processing of noxious stimuli by an organism,4 and the dampening of this sensory pathway through pharmaceuticals.10 Although pain has been studied in fishes, both in clinical and laboratory contexts,4,6 few studies have investigated the effects of analgesic drugs in fishes compared to the vast body of literature available in rodents. For comparison, a keyword search in scientific search engines conducted in May 2024 for the words “rodent” and “antinociceptive” retrieved 14,156 published articles. The same search with the words “fish” and “antinociceptive” returned only 83 articles. Clearly, more research on fish analgesia is needed to keep up with the growing presence of fish in laboratory, domestic, and clinical settings.
Experiments testing the effectiveness of opioids as antinociceptive agents in zebrafish (Danio rerio) larvae typically administer the drugs via immersion in water.11 Drug administration via immersion is advantageous in aquatic medicine due to its practicality, but water hardness and pH can influence pharmacokinetics and pharmacodynamics of drugs in fishes,12,13 as documented for morphine in goldfish (Carassius auratus).12–14 Therefore, for clinical applications in fish medicine, it would be advantageous to test the antinociceptive effects of injectable drugs both to limit the opioid residues in effluent water and facilitate administration to specific individuals. Despite these benefits, injectable compounds add another difficulty to the testing process due to the pharmacokinetics of the drugs associated with fluctuating plasma concentrations of active metabolites. In addition, it is currently unknown if the model using a shuttle box developed in zebrafish larvae could be extrapolated to larger teleosts. Individual specifics, such as water salinity, temperature preference, size, and species, pose some challenges for extrapolation of pharmacologic knowledge.4 Developing a model that would allow for the testing of the effect of opioids in fishes with less bias than previous surgical models is critically important to improve fish welfare.10,15 Previous nociceptive models studied in fish have assessed postoperative appetite, respiratory rate, velocity, traveling distance, or position in the water column, which are parameters intrinsically affected by sedation.10,11,15,16 Other studies17 have found no adverse effects associated with opioids in fishes. Due to the lack of experimental models in larger teleosts and experimental trials of opioid effects, a novel model was created and tested.
The shuttle box is an experimental setup that has successfully been used in zebrafish to screen the effects of potential analgesic drugs intended for use in humans18 and for behavioral assessment19 in various fish species.20 A shuttle box consists of 2 tanks each representing distinct temperature zones, a warmer and a cooler zone, connected by a channel through which a fish can freely pass. Water temperature is constantly monitored in each zone by temperature sensors, and water temperature is maintained by a heating/cooling system located outside of the shuttle box. Water pumps supplying the water used to maintain temperature are set up in opposite directions to minimize water mixing. Above the enclosure, a camera connected to a movement tracking system (Shuttlesoft; Loligo Systems) is installed. The shuttle box can be set in either static or dynamic temperature modes. In static mode, temperatures in each zone are determined before experimentation and remain stable in each zone for the duration of the experiment. In dynamic mode, the temperature difference between the zones is maintained, but the temperature in both zones changes depending on the location of the tested fish. Fish location is tracked by the tracking software and is used to calculate fish velocity and incremental preferred temperature.
The shuttle box model has been used in zebrafish alone or in combination with variable compounds intended to sensitize zebrafish to warm water temperature (allyl isothiocyanate)18 or to enhance hypoactivity (acetic acid) or hyperactivity (citric acid).11 Allyl isothiocyanate, also called mustard oil, is a skin vesicant causing cold water preference in zebrafish.18,21 In a previous study,11 immersion in morphine reversed hypoactivity associated with acetic acid in zebrafish. This finding was interpreted as a confirmation of an antinociceptive effect of morphine. However, the hyperactivity potentially due to morphine was not assessed in the same trial, which could bias the results.11 As opioids may cause hyperactivity in fish,10 since it is not systematic, it is relevant to evaluate this effect in the same experiment through positive control groups that receive opioid drugs without a nociceptive stimulus. In a case where opioid exposure induces hyperactivity in the tested fish, immersion of the fish in allyl isothiocyanate before introduction in the shuttle box may be a better option than using acetic or citric acid to determine opioid effects and assess opioid antinociceptive effects.
The objective of the present experiment was, first, to assess the behavioral effects (the thermal preference, swimming velocity, and time spent immobile) of opioids in goldfish using a shuttle box experimental apparatus. Next, we assessed the antinociceptive effects of the opioids following allyl isothiocyanate vesicant exposure before behavioral tests in the shuttle box. Goldfish were selected as the tested species as they are widely available and have been previously used in experimental trials about opioid drugs and pain in fishes.22–26 The opioid drugs selected for this trial were butorphanol and morphine due to the availability of scientific literature on these opioid drugs in fishes,8,10,27,28 as well as their variable affinity for opioid receptors. Butorphanol is a κ-agonist and μ-agonist, while morphine is a pure μ-agonist.28 The presence of μ- and κ-opioid receptors in the central nervous system is documented in goldfish.29 We hypothesized that the shuttle box set in static mode would allow the detection of hyperactivity or sedation induced by opioids in goldfish. We also hypothesized that the addition of mustard oil (ie, allyl isothiocyanate) by immersion before exposure to the shuttle box would sensitize the skin of the test subject and enhance the detection of analgesia induced by opioids without causing adverse effects in goldfish.
Methods
This project was approved by an IACUC at the Université de Montréal (23-Rech-2265).
Animals
Five healthy adult goldfish of unknown sex (total length ranging from 10.8 to 17.5 cm; body weight ranging from 16 to 63 g) were used in trials 1 and 2 of this project (Table 1). Fish were acquired from a pet shop specializing in aquatic animals (SubAquatique). Fish were housed collectively in a 70-L glass aquarium during the trials and during the washout period of 21 days between trials. The life support system consisted of an external canister filter (Fluval 207; 60- to 220-L canister filter; Hagen Inc). Water quality, including pH, ammonia, nitrites, and nitrates, was monitored using a colorimetric test kit (Fluval master test kit for aquarium; Hagen Inc) daily for 2 weeks and then weekly during the rest of the study. The holding tank contained a submersible heating probe (Marina; 100 W; Hagen Inc) with a temperature set at 24 °C and was equipped with polyvinyl chloride tubing and ornaments for shelter and environmental enrichment. Water pH ranged between 7.2 and 7.6, total ammonia ranged between 0 and 1.2 mg/L, nitrite ranged between 0 and 0.1 mg/L, and nitrate ranged between 0 and 5 mg/L throughout the experimentation period. Fish were kept in a temperature-controlled room set at 23 °C. Photoperiod consisted of a 12:12 light:dark cycle. Fish were fed 6 times a week ad libitum using a pelleted diet (Fluval Bug Bites Goldfish formula; Hagen Inc) and monitored daily.
Experimentation fish identification and anatomical details.
| Fish number | Weight (g) | Total length (cm) |
|---|---|---|
| 1 | 20 | 11.2 |
| 2 | 16 | 10.8 |
| 3 | 19 | 11.0 |
| 4 | 63 | 17.0 |
| 5 | 36 | 17.5 |
The fish underwent a 2-week acclimation period followed by a complete physical examination under general anesthesia with tricaine methanesulfonate (MS-222 Syncaine; 1,000 mg/g; Syndel) at 140 mg/L buffered 1:1 with sodium bicarbonate (baking soda; Arm & Hammer). The same buffering technique was used throughout the experiment. Skin scrapes and gill clips were performed, and no parasites were detected. All individuals then received a prophylactic antiparasitic treatment with 2 mg/L praziquantel for 3 days (1,000 mg/g; Fishman Chemical). Limit points of the study were set using a point-based scale that evaluated the skin condition (discoloration, erythema, and swelling), systemic abnormalities (lethargy, hyperactivity, and buoyancy disorders), and level of respiratory depression. Points scores ranged from 0 being normal, 1 being light signs, 2 being moderate signs, and 3 being severe signs.30,31 Naloxone and naltrexone were kept available in case of severe opioid side effects needing reversal, and dosing was based on previous studies.32,33
Seven days after the end of the trials, the fish were adopted out after an unremarkable physical examination conducted by a veterinarian. All animals remained healthy 2 months after adoption at the time of manuscript submission.
Experimentation
Trial 1
Trial 1 was a controlled randomized blinded crossover study with each fish receiving a different treatment at each session. Of note, opioids are not approved by the FDA for use in food animal species, including fish; extralabel drug use has to comply with provisions of the Animal Medicinal Drug Use Clarification Act. If used in fish kept in zoological institutions or as companion animals, veterinarians should contact the Canadian Global Food Animal Residue Avoidance Database to determine residue for each case in Canada, and a disclaimer should be added to the medical record. Two doses of morphine (5 and 10 mg/kg) and butorphanol (0.4 and 10 mg/kg) were administered IM as treatments, while saline was administered IM as a control. Five different injectable solutions were prepared and labeled as A, B, C, D, and E to keep the experimentations blinded. These solutions were prepared weekly, and fish received a different injectable drug at each session, with a minimum interval of 5 days between injections. This washout period was determined based on the known half-life of morphine at 16 °C, which is approximately 15 hours,23 resulting in the elimination of the drug from goldfish plasma after 3 days. A duration of 5 days was chosen to be very conservative, as morphine elimination in goldfish is expected to be quicker at a temperature of 24 °C and higher, similar to other drugs.34
The experimental 40 X 40-cm shuttle box (Loligo Systems) was placed on a separate table surrounded by opaque curtains to obscure the fish’s visibility of exterior stimuli (Figure 1). The shuttle box was set up and calibrated before each trial including temperature probe calibration, pixel calibration, experiment area mask, and temperature zone delimitation. The warmer tank was set at 28 °C, and the cooler tank was set at 26 °C. Temperature selection was based on published temperature tolerance for goldfish, with the goal of remaining well below the mean critical thermal maximum at 30.8 °C due to animal welfare.35 Minimal current in the tanks was confirmed using dye, and the temperature was set before each experiment. Once the shuttle box was operating at the desired parameters, a selected fish was transferred to an anesthesia tank maintained at 25 °C and containing buffered MS-222 at 140 mg/L. The order of fish in each run of the experiment, as well as the treatment administered to each fish during a given experimental trial, was determined randomly by drawing a number from a bag. Once the fish was immobilized, it was administered the preset injectable solution IM in the cranial epaxial muscles by a blinded operator (Table 2).8,27,28,36–38 Injection sites were alternated on a weekly basis. The fish was then transferred to a recovery tank kept at 25.5 to 26 °C and monitored for 30 minutes for it to reach the estimated time of maximum drug concentration in the blood based on a previous study23 about IM morphine pharmacokinetics in goldfish.
Images representing the shuttle box setup (A), the temperature controllers (B), and visualization of the test zones (C) in the Loligo Shuttlesoft software. A—Arrows point to the experimentation tank, the circles represent the reservoir tanks, and the squares represent the reheating and cooling reservoirs. Blue color represents cool water and red color represents warm water. B—Data acquisition system (DAQ; yellow asterisk) that controls the 4 water pumps to set the heating-cooling direction of water. The red and blue arrows point toward the temperature analyzing system connected to the submersible temperature probes. C—Top view of the shuttle box tank as seen on the software. The red asterisk is on the warm water side, and the blue asterisk is on the cool water side. Both tanks are connected by a transition channel represented with a green asterisk. The arrows demonstrate the flow of water in each tank. The green plus sign and line represent the software tracing of the fish within the experimentation tank.
Citation: American Journal of Veterinary Research 85, 11; 10.2460/ajvr.24.06.0172
Drugs administered to adult goldfish (Carassius auratus) during the first trial, using a crossover study design.
| Medication | Dose (mg/kg) | Concentration | Solution | Manufacturer |
|---|---|---|---|---|
| Low-dose butorphanol | 0.4 | 10 mg/mL | Injectable | Dolorex; Merck Animal Health |
| High-dose butorphanol | 10 | 10 mg/mL | Injectable | Dolorex; Merck Animal Health |
| Low-dose morphine | 5 | 10 mg/mL | Injectable | Morphine sulfate injection, USP; Sandoz |
| High-dose morphine | 10 | 10 mg/mL | Injectable | Morphine sulfate injection, USP; Sandoz |
| Sterile NaCl | Same volume as morphine and butorphanol high-dose groups | 0.9% | Injectable | 0.9% sodium chloride injection, USP; Baxter |
Each individual was placed in the cold tank of the shuttle box at 30 minutes postinjection. Fish were tracked in the shuttle box for the following 120 minutes. Fish were supervised continuously by a silent experimenter using a video camera (USB 20 SE; IDS Imaging Development Systems GmbH). The fish was allowed to swim freely between the cold and warm tanks. After 120 minutes, the fish was placed in a postexperimentation tank kept at 25 to 26 °C and monitored continuously for 30 minutes. A partial water change was performed in the shuttle box before the introduction of another fish, whenever debris or feces were observed. Data were recorded every second automatically by the software (Shuttlesoft; Loligo Systems) and included fish velocity, fish position in the zones, temperature of each zone, total time spent in the cold and warm zone since the beginning of the session, and incremental preferred temperature. Four hours following testing, the fish were reevaluated to detect any adverse effects.
Trial 2
After a washout period of 21 days, trial 2 was a controlled randomized blinded crossover study with 2 to 3 fish receiving the same treatment at each session. Morphine or saline was administered IM to the fish successively with an interval of 7 days between trials. These solutions were prepared each week.
Mustard oil was acquired from a commercial source (allyl isothiocyanate; Sigma-Aldrich, Merck) and diluted to a concentration of 4 μmol/L on the day of each experiment as previously described.18 The shuttle box was prepared the same way as in trial 1. The fish was placed in an anesthesia tank as in trial 1. Once immobilized, the fish received the same volume of either saline or 10 mg/kg morphine, IM. Twenty minutes after the injection, fish were exposed to the mustard oil vesicant solution for 5 minutes. Initially, an immersion for 10 minutes was planned, but the first tested fish displayed apnea and buoyancy disorders after only 5 minutes of immersion. Therefore, the duration of immersion in mustard oil solution was set to 5 minutes in all tested fish. Adverse effects were reversed immediately when the fish was removed from the mustard oil bath. Each individual, including the first fish, was then placed in the shuttle box tank with a temperature set at 26 °C in the cold tank and 28 °C in the warm tank. Fish were filmed for 4 minutes using a camera (Hero8 and Hero11; GoPro). After experimentation, each fish was placed in its housing tank and monitored for 2 hours and then once a day for 2 days. Any clinically detectable adverse effects were recorded. The percentage of time spent in the warmer zone was recorded for each fish.
Data analysis
Raw data for each fish trial were imported to a spreadsheet, and data visualization and analysis software (Excel; Microsoft) selected results that were collected and organized. To assess fish activity, 2 parameters were evaluated: fish maximum velocity over 2 hours and percentage of time spent immobile (velocity of 0 cm/s) over 2 hours, as previously described in zebrafish.11,18 In addition, to assess thermal preference in trial 1, incremental preferred temperatures after 10 minutes in the shuttle box and at the end of the trial were recorded for each trial, as well as the percentage of time spent in the warm zone. To assess thermal preference in trial 2, the percentage of time spent in the warm zone was evaluated.
Data were analyzed using the R statistical software (R version 4.2.1; R Foundation for Statistical Computing) with the lme4 package. Linear mixed models were constructed with each parameter for activity and thermal preference as response variables, drug administered as fixed factors, and fish individual and session order as random factors. Likelihood ratio tests were used to evaluate each parameter. The normality of model residuals was confirmed using a Shapiro-Wilk test.
In addition, to test the effect of the trial on the percentage of time fish spent immobile, a linear mixed model was created with the percentage of immobility as independent variable, the trial number as dependent variable, and the individual fish as random variable. A likelihood ratio test was then conducted.
The significance level was set at P < .05. Post hoc analysis was performed, and a Benjamini-Hochberg correction was applied to the P values for pairwise comparisons.
Results
All fish recovered from the trials. No adverse effects were detected during the first trial, while signs of discomfort were noted when the mustard oil bath was used. After 2 minutes in the mustard oil bath, 2 fish displayed buoyancy disorders, became apneic, and stopped moving until they were stimulated or removed from the immersion bath at 5 minutes. After exposure to mustard oil, fish appeared lethargic for 1 hour after being placed in their holding tank. All fish ate normally the evening of the assay. One day after immersion in the mustard oil solution, 1 fish, which had not displayed buoyancy disorders during the mustard oil bath, developed mild discolored lesions of the pelvic fins. These lesions healed spontaneously within a week and no change in appetite or general demeanor was noted in this fish.
During one of the sessions of trial 1, the software disconnected for a few minutes, which resulted in the loss of about 5 minutes of data. Consequently, analyses about incremental temperatures were conducted on 24 rather than 25 data points.
In trial 1, a significant difference was noted among treatments regarding the percentage of time spent immobile (P = .04; Figure 2). Fish receiving 5 mg/kg of morphine spent more time immobile than the control group (P = .0005). Fish receiving 5 mg/kg of morphine spent more time immobile than fish receiving butorphanol at 0.4 mg/kg (P = .0008) or at 10 mg/kg (P = .008) or morphine at 10 mg/kg (P = .007). Fish receiving morphine at 10 mg/kg spent significantly more time immobile than fish receiving 10 mg/kg of butorphanol (P = .04). Finally, fish receiving 0.4 mg/kg of butorphanol had a tendency toward a lower percentage of time spent immobile than the control group, but this difference was not statistically significant (P = .10). No significant difference was detected among trials in the percentage of time fish spent immobile (P = .10).
Percentage of time 6 goldfish (Carassius auratus) spent immobile between 30 minutes and 2.5 hours after receiving cranial intramuscular injections of either saline, morphine at 5 mg/kg, morphine at 10 mg/kg, butorphanol at 0.4 mg/kg, or butorphanol at 10 mg/kg. Goldfish were kept between 26 and 28 °C in a shuttle box arena. *P < .05, statistically significant differences among groups.
Citation: American Journal of Veterinary Research 85, 11; 10.2460/ajvr.24.06.0172
No significant difference among treatment groups was detected regarding overall fish incremental temperature or in the percentage of time spent in the warm zone, meaning that fish spent a similar amount of time in each of the temperature zones (median percentage of time in the warm zone was 62% for the morphine low-dose group, 67% for the morphine high-dose group, 62% for the butorphanol low-dose group, 88% for the butorphanol high-dose group, and 76% for the control group, P = .25). No significant difference among groups was noted in the fish incremental temperature when considering the position of the fish during the first 10 minutes of the trial (P = .29). No significant difference among treatment groups was detected regarding fish maximum velocity during the 2 hours of trial (P = .57).
In trial 2, all fish receiving saline before exposure to mustard oil spent less than 50% of their time in the warm area of the tank. Fish receiving morphine at 10 mg/kg, IM, spent significantly more time in the warm zone (median of 72% of the time, ie, 2 minutes and 52 seconds) than the control group (median of 39% of the time, ie, 1 minute and 34 seconds) exposed to mustard oil after saline injection (P = .036; Figure 3).
Percentage of time 6 goldfish (Carassius auratus) spent in warm water following exposure to mustard oil and injection with morphine or saline control. Increased tolerance of thermonociception was seen with morphine when compared to saline control (P = .036).
Citation: American Journal of Veterinary Research 85, 11; 10.2460/ajvr.24.06.0172
Discussion
In this study, behavioral effects, including sedation following morphine injection, were detected using the shuttle box system in goldfish. In addition, contrary to the control group, fish administered morphine did not strongly avoid the warm zone following mustard oil exposure, suggesting that injectable morphine acts as an analgesic drug in goldfish. In teleost fish, polymodal nociceptors are present in the skin, including a majority of myelinated Aδ fibers and a minority of unmyelinated C fibers.39 Thus, thermal aversion involves the same neurologic pathways as nociception,40 making it a relevant model to study pain in fishes. Evaluating noxious models in fishes before testing novel drugs is critical to practicing evidence-based medicine, as extrapolating models used in mammals may not be relevant due to the differences in fish physiology. Fishes display several types of somatosensory receptors, which include polymodal nociceptors, mechanothermal nociceptors, mechanochemical nociceptors, and mechanical receptors.41 Comparative studies41 have shown that nociceptors of cold-water fishes such as trout are not activated by temperatures below 4 °C, which could lead to different pain perceptions associated with cold when compared to other vertebrates.41–43 In addition, polymodal Aδ fibers are more common in fishes and display a similar function to C fibers in mammals.41,44 These nociceptors also display lower mechanical and heat thresholds in trout than those of cutaneous nociceptors in mammals, and these thresholds are similar to those of mammalian cornea.41 Thus, various models may lead to various results in terms of drug efficacy, and models derived from mammals may not be directly extrapolated to fishes. In the present study, the shuttle box model enabled the detection of analgesic effects associated with morphine in goldfish, as previously reported in this species.40
In the first trial, morphine at 5 mg/kg, IM, produced sedation, as shown by the increased percentage of time spent immobile compared to the control group during a period of 2.5 hours after injection. While a low dose of morphine led to sedative effects in the present study, no such effects were detected after administration of a higher 10-mg/kg dose in the first trial, as no difference was detected in the time fish spent immobile. Similarly, in rats, a dose of 15 mg/kg of morphine causes an initial sedation phase for 1.5 to 2 hours, while a higher dose of 30 mg/kg causes hyperactivity, which is mediated by dopamine receptors.45 In goldfish, it is unknown if hyperactivity caused by opioids is also mediated by dopamine receptors. Since the trial sequence was randomized and the effects of the trial sequence on fish immobility were insignificant, it is very unlikely that habituation to the shuttle box contributed to the results observed in the present study. Thus, hyperactivity was attributed to morphine at high dose rather than habituation to the study device. Another potential cause of this increased activity in the high-dose morphine group would be hepatic enzyme induction increasing the metabolism of opioids in later trials. While cytochrome P450 enzymes have been documented in fishes,46 their half-life has not been evaluated in goldfish, making changes in enzyme activity difficult to measure. Regardless, enzyme induction typically requires multiple administrations at high doses,47 which was not the case in the present study.
A high interindividual variability was observed in the control group, which could be due to individual fish traits, as suggested in a previous study.48 Another explanation would be that these 5 experimental fish had very variable ages, given their body weight range. As age has been shown to influence how fish react to chemicals,49,50 this effect of age could partially explain the variability observed within treatment groups. Finally, the crossover study design implied that 1 fish was in the control group in each trial. As fish were getting more accustomed to the shuttle box, it is possible that they explored the setup more, but the model testing for the effect of the trial order showed this effect was not significant. Interindividual variability could have contributed to a type II statistical error in the present study, due to the dispersion of the data in the control group. A nonstatistically significant tendency toward hyperactivity was detected after injection of butorphanol at 0.4 mg/kg. This was not unexpected as previous studies10,38,45,51 have shown hyperactivity associated with opioids in koi (Cyprinus carpio), another Cyprinidae.
The maximal velocity of fish did not vary across treatments. This was surprising as it was expected that sedation would be associated with decreased swimming velocity. In fact, rather than continuously swimming around the tank as control fish did, sedated fish stayed immobile and displayed intermittent bouts of acceleration with a similar velocity as nonsedated control fish. This was not attributed to a remnant effect of MS-222 anesthesia as a previous study52 has shown that MS-222 does not affect fish behavior nor swimming velocity evaluated 30 minutes after anesthesia. Of note, higher swimming velocity has been associated with thermal stress in some fish species,18 while others display lower swimming velocity after air exposure causing stress.53 Regardless, a change in swimming velocity was not observed in the present study. Conversely, the percentage of time spent by the fish in the warm zone during trial 1 was unaffected by opioid drugs, showing that sedation did not affect this parameter. This result was determinant in the selection of the nociceptive stimulus used in the second trial of this study. Citric acid and acetic acid could not be used in goldfish since the drug tested had an effect on the activity level of fish.
In trial 2, mustard oil proved to be an efficient solution for cutaneous sensitization as goldfish showed signs of discomfort. Interindividual variability was lower in the group receiving saline than in the group receiving morphine, with all control fish spending less than 50% of their time in the warm area of the shuttle box. As expected, morphine reversed the aversion of fish to warm water. A similar pattern has been observed in zebrafish larvae administered buprenorphine by immersion.18 However, in previous studies, the shuttle box was set in a dynamic mode, meaning that the temperature kept increasing if the fish stayed in the warm area. This setup was not elected in the present study for increased fish safety, and the static mode proved sufficiently sensitive to detect antinociceptive effects of morphine in goldfish.
The present study confirmed that experiments using a shuttle box apparatus were appropriate for detecting the antinociceptive effects of morphine at 10 mg/kg, IM, in goldfish, independent of its sedative effect. Indeed, the percentage of time spent in the warm zone was a reliable parameter to assess thermal analgesia. The percentage of time spent in the warm zone was also unaffected by the behavioral effects of morphine in the first trial, contrary to the activity level of fish. This is similar to what has been previously observed in zebrafish.18 However, the difference with the present model is that an injectable route was used, which is a clinically relevant administration method. This required the trial to be performed during the expected peak of action of the drug, implying an available pharmacokinetic study at the same temperature in the target species. Such a study was only available for morphine at 40 mg/kg, IM, in goldfish at 16 °C, precluding the evaluation of butorphanol at this point. Since drug pharmacokinetics vary according to environmental temperature in ectotherms,34 future studies should evaluate the pharmacokinetic profile of opioid drugs in fish at specific temperatures used during experimentation before pharmacodynamic testing in the shuttle box.
Adverse effects observed after exposure to the mustard oil solution were not described previously in zebrafish even after immersion for 10 minutes.18 This may be due to a higher sensitivity of goldfish to allyl isothiocyanate. Alternatively, apnea may be more difficult to monitor in zebrafish larvae due to their small size. Regardless, the respiratory and positional adverse effects observed in goldfish were temporary, and all fish recovered uneventfully after removal from the immersion bath. In zebrafish, the vesicant effect of mustard oil is expected to last for about 60 minutes,18 which is similar to the duration of lethargy observed in the goldfish of the present study.
Certain precautions should be observed when using this model to avoid confounding factors. In particular, any visual, auditory, olfactive, or pheromonal cue affecting fish preference for one side or the other of the shuttle box should be avoided. Therefore, manipulators should remain hidden and silent when performing a trial with the shuttle box. Conspecifics should not be visible. Finally, the acclimation temperature is of critical importance when setting the shuttle box temperature gradient as acclimation temperature has been shown to condition fish thermal preference.35
While a thermal model may not be directly extrapolated to a clinical situation, due to the different visceral receptors stimulated during surgery, this model could be a first step to screen opioid efficacy at various doses in goldfish. Future studies should aim to evaluate the analgesic effects of different analgesic drugs in goldfish using this promising model.
Acknowledgments
The authors thank the animal care staff at the FANI Experimental Unit of the Faculté de Médecine Vétérinaire and Ysanne Michaud-Simard for their care of the fish during the study period. The authors also thank the Montréal Aquarium for contributing to this study. Drs. Vergneau-Grosset and Binning are members of the Ressources Aquatiques Québec network.
Disclosures
The authors have nothing to disclose. No AI-assisted technologies were used in the generation of this manuscript.
Funding
Funding for this project was provided by Fonds du Centenaire Program of the Faculté de Médecine Vétérinaire, Université de Montréal, the MITACS Acceleration Program, the FRQNT relève professorale program, and Eco-recreo Inc.
References
- 1.↑
US pet ownership and demographics sourcebook. AVMA. Accessed May 23, 2024. https://ebusiness.avma.org/files/ProductDownloads/eco-pet-demographic-report-22-low-res.pdf
- 2.↑
CCAC animal data report 2022. Council on Animal Care Care. Accessed May 23, 2024. https://ccac.ca/Documents/AUD/CCAC_Animal_Data_Report_2022.pdf
- 3.↑
Zahl IH, Samuelsen O, Kiessling A. Anaesthesia of farmed fish: implications for welfare. Fish Physiol Biochem. 2012;38(1):201–218. doi:10.1007/s10695-011-9565-1
- 5.
Sloman KA, Bouyoucos IA, Brooks EJ, Sneddon LU. Ethical considerations in fish research. J Fish Biol. 2019;94(4):556–577. doi:10.1111/jfb.13946
- 6.↑
Mason GJ, Lavery JM. What is it like to be a bass? Red herrings, fish pain and the study of animal sentience. Front Vet Sci. 2022;9(4):948567. doi:10.3389/fvets.2022.948567
- 7.↑
Chatigny F. The controversy on fish pain: a veterinarian’s perspective. J Appl Anim Welf Sci. 2019;22(4):400–410. doi:10.1080/10888705.2018.1530596
- 8.↑
Gräns A, Sandblom E, Kiessling A, Axelsson M. Post-surgical analgesia in rainbow trout: is reduced cardioventilatory activity a sign of improved animal welfare or the adverse effects of an opioid drug? PLoS One. 2014;9(4):e95283. doi:10.1371/journal.pone.0095283
- 9.
Crivelaro RM, Thiesen R, Aldrovani M, Silva PES, Barros Sobrinho AAF, Moraes PC. Behavioural and physiological effects of methadone in the perioperative period on the Nile tilapia Oreochromis niloticus. J Fish Biol. 2019;94(5):823–827. doi:10.1111/jfb.13959
- 10.↑
Baker TR, Baker BB, Johnson SM, Sladky KK. Comparative analgesic efficacy of morphine sulfate and butorphanol tartrate in koi (Cyprinus carpio) undergoing unilateral gonadectomy. J Am Vet Med Assoc. 2013;243(6):882–890. doi:10.2460/javma.243.6.882
- 11.↑
Lopez-Luna J, Al-Jubouri Q, Al-Nuaimy W, Sneddon LU. Reduction in activity by noxious chemical stimulation is ameliorated by immersion in analgesic drugs in zebrafish. J Exp Biol. 2017;220(8):1451–1458. doi:10.1242/jeb.146969
- 12.↑
Vorbach BS, Chandasana H, Derendorf H, Yanong RPE. Pharmacokinetics of oxytetracycline in the giant danio (Devario aequipinnatus) following bath immersion. Aquaculture. 2019;498:12–16. doi:10.1016/j.aquaculture.2018.08.02
- 13.↑
Vorbach BS, Zhou J, Lang Y, Bulitta JB, Yanong RPE. Population pharmacokinetics of enrofloxacin and florfenicol in the giant danio (Devario aequipinnatus) following oral administration of both antibiotics and bath administration of enrofloxacin. Aquaculture 2024;579:1–10. doi:10.1016/j.aquaculture.2023.740222
- 14.↑
Newby NC, Wilkie MP, Stevens ED. Morphine uptake, disposition, and analgesic efficacy in the common goldfish (Carassius auratus). Can J of Zool. 2009;87(5):388–399. doi:10.1139/Z09-023
- 15.↑
Harms CA, Lewbart GA, Swanson CR, Kishimori JM, Boylan SM. Behavioral and clinical pathology changes in koi carp (Cyprinus carpio) subjected to anesthesia and surgery with and without intra-operative analgesics. Comp Med. 2005;55(3):221–226.
- 16.↑
Kenney JW, Scott IC, Josselyn SA, Frankland PW. Contextual fear conditioning in zebrafish. Learn Mem. 2017;24(10):516–523. doi:10.1101/lm.045690.117
- 17.↑
Sneddon LU, Schroeder P, Roque A, et al. Pain management in zebrafish: report from a FELASA Working Group. Lab Anim. 2024;58(3):261–276. doi:10.1177/00236772231198733
- 18.↑
Curtright A, Rosser M, Goh S, et al. Modeling nociception in zebrafish: a way forward for unbiased analgesic discovery. PLoS One. 2015;10(1):e0116766. doi:10.1371/journal.pone.0116766
- 19.↑
Pather S, Gerlai R. Shuttle box learning in zebrafish (Danio rerio). Behav Brain Res. 2009;196(2):323–327. doi:10.1016/j.bbr.2008.09.013
- 20.↑
Christensen EAF, Andersen LEJ, Bergsson H, Steffensen JF, Killen SS. Shuttle-box systems for studying preferred environmental ranges by aquatic animals. Conserv Physiol. 2021;9(1):coab028. doi:10.1093/conphys/coab028
- 21.↑
Taylor JC, Dewberry LS, Totsch SK, et al. A novel zebrafish-based model of nociception. Physiol Behav. 2017;174:83–88. doi:10.1016/j.physbeh.2017.03.009
- 22.↑
Ward JL, McCartney SP, Chinnadurai SK, et al. Development of a minimum-anesthetic-concentration depression model to study the effects of various analgesics in goldfish (Carassius auratus). J Zoo Wildl Med. 2012;43:214–222. doi:10.1638/2010-0088.1
- 23.↑
Nordgreen J, Kolsrud HH, Ranheim B, et al. Pharmacokinetics of morphine after intramuscular injection in common goldfish Carassius auratus and Atlantic salmon Salmo salar. Dis Aquat Organ. 2010;88(1):55–63. doi:10.3354/dao02147.
- 24.
Gallon R. Effects of shock intensity on shuttlebox avoidance conditioning in goldfish. Psychol Rep. 1972;31(3):855–858. doi:10.2466/pr0.1972.31.3.855
- 25.
Dunlop R, Millsopp S, Laming P. Avoidance learning in goldfish (Carassius auratus) and trout (Oncorhynchus mykiss) and implications for pain perception. Appl Anim Behav Sci. 2006;97(2–4):255–271. doi:10.1016/j.applanim.2005.06.018
- 26.↑
Cordova MS, Braun CB. The use of anesthesia during evoked potential audiometry in goldfish (Carassius auratus). Brain Res. 2007;1153:78–83. doi:10.1016/j.brainres.2007.03.055
- 27.↑
Martins T, Valentim AM, Pereira N, Antunes LM. Anaesthesia and analgesia in laboratory adult zebrafish: a question of refinement. Lab Anim. 2016;50(6):476–488. doi:10.1177/0023677216670686
- 28.↑
Martins T, Valentim A, Pereira N, Antunes LM. Anaesthetics and analgesics used in adult fish for research: a review. Lab Anim. 2019;53(4):325–341. doi:10.1177/0023677218815199
- 29.↑
Brooks AI, Standifer KM, Cheng J, Ciszewska G, Pasternak GW. Opioid binding in giant toad and goldfish brain. Receptor. 1994;4(1):55–62.
- 30.↑
Tzaneva V, Perry SF. The control of breathing in goldfish (Carassius auratus) experiencing thermally induced gill remodelling. J Exp Biol. 2010;213(21):3666–3675. doi:10.1242/jeb.047431
- 31.↑
Chalon J, Martin P, Roberts C, Ramanathan S, Katz R, Turndorf H. Anaesthetic uptake by the goldfish: effect of respiratory rate. Acta Anaesthesiol Scand. 1983;27(5):361–365. doi:10.1111/j.1399-6576.1983.tb01968.x
- 32.↑
Wolkers CPB, Barbosa Junior A, Menescal-de-Oliveira L, Hoffmann A. Stress-induced antinociception in fish reversed by naloxone. PLoS One. 2013;8(7):e71175. doi:10.1371/journal.pone.0071175
- 33.↑
Strobel MM, Baker KC, Berliner AL, Tang KN, Hepps Keeney CM, Lewbart GA. Naltrexone as a promising treatment for clinical signs of lateral line depigmentation in palette surgeonfish (Paracanthurus hepatus). J Zoo Wildl Med. 2023;54(1):137–142. doi:10.1638/2022-0022
- 34.↑
Sohlberg S, Aulie A, Søli NE. Temperature-dependent absorption and elimination of flumequine in rainbow trout (Oncorhynchus mykiss walbaum) in fresh water. Aquaculture. 1994;119(1):1–10. doi:10.1016/0044-8486(94)90439-1
- 35.↑
Ford T, Beitinger TL. Temperature tolerance in the goldfish, Carassius auratus. J Therm Biol. 2005;30(2):147–152. doi:10.1016/j.jtherbio.2004.09.004
- 36.↑
Oulton JJ, McCrohan LJ, Sneddon CR. The efficacy of three types of analgesic drugs in reducing pain in the rainbow trout, Oncorhynchus mykiss. J Appl Anim Behav Sci. 2011;133(3):265–274. doi:10.1016/j.applanim.2011.06.009
- 37.
Neiffer DL, Stamper AM. Fish sedation, anesthesia, analgesia, and euthanasia: considerations, methods, and types of drugs. ILAR J. 2009;50(4):343–360. doi:10.1093/ilar.50.4.343
- 38.↑
Chatigny F, Creighton CM, Stevens DE. Updated review of fish analgesia. J Am Assoc Lab Anim Sci. 2018;57(1):5–12.
- 39.↑
Sneddon LU. Trigeminal somatosensory innervation of the head of a teleost fish with particular reference to nociception. Brain Res. 2003;972(1–2):44–52. doi:10.1016/s0006-8993(03)02483-1
- 40.↑
Nordgreen J, Garner JP, Janczak AM, Ranheim B, Muir WM, Horsberg TE. Thermonociception in fish: effects of two different doses of morphine on thermal threshold and post-test behaviour in goldfish (Carassius auratus). Appl Anim Behav Sci. 2009;119(1–2):101–107. doi:10.1016/j.applanim.2009.03.015
- 41.↑
Ashley PJ, Sneddon LU, McCrohan CR. Nociception in fish: stimulus-response properties of receptors on the head of trout Oncorhynchus mykiss. Brain Res. 2007;1166:47–54. doi:10.1016/j.brainres.2007.07.011
- 42.
Sneddon LU. Evolution of nociception and pain: evidence from fish models. Philos Trans R Soc Lond B Biol Sci. 2019;374(1785):20190290. doi:10.1098/rstb.2019.0290
- 43.↑
Sneddon LU. Evolution of nociception in vertebrates: comparative analysis of lower vertebrates. Brain Res Brain Res Rev. 2004;46(2):123–130. doi:10.1016/j.brainresrev.2004.07.007
- 44.↑
Sneddon LU. Comparative physiology of nociception and pain. Physiology. 2018;33(1):63–73. doi:10.1152/physiol.00022.2017
- 45.↑
Magnus-Ellenbroek B, Havemann-Reinecke U. Morphine-induced hyperactivity in rats-a rebound effect? Naunyn Schmiedebergs Arch Pharmacol. 1993;347(6):635–642. doi:10.1007/BF00166947
- 46.↑
Celander MC, Goldstone JV, Brun NR, et al. Resistance to Cyp3a induction by polychlorinated biphenyls, including non-dioxin-like PCB153, in gills of killifish (Fundulus heteroclitus) from New Bedford harbor. Environ Toxicol Pharmacol. 2021;83:103580. doi:10.1016/j.etap.2020.103580
- 47.↑
Sweeney BP, Bromilow J. Liver enzyme induction and inhibition: implications for anaesthesia. Anaesthesia. 2006;61:159–177. doi:10.1111/j.1365-2044.2005.04462.x
- 48.↑
Killen SS, Marras S, Metcalfe NB, McKenzie DJ, Domenici P. Environmental stressors alter relationships between physiology and behaviour. Trends Ecol Evol. 2013;28(11):651–658. doi:10.1016/j.tree.2013.05.005
- 49.↑
Vergneau-Grosset C, Cruz Benedetti IC. Fish sedation and anesthesia. Vet Clin North Am Exot Anim Pract. 2022;25(1):13–29. doi:10.1016/j.cvex.2021.08.001
- 50.↑
Perret-Thiry C, Raulic J, Vergneau-Grosset C. Evaluation of prolonged immersion in tricaine methanesulfonate for juvenile goldfish (Carassius auratus) euthanasia. J Am Vet Med Assoc. 2022;260(8):911–915. doi:10.2460/javma.21.09.0416
- 51.↑
Bao W, Volgin AD, Alpyshov ET, et al. Opioid neurobiology, neurogenetics and neuropharmacology in zebrafish. Neuroscience. 2019;404:218–232. doi:10.1016/j.neuroscience.2019.01.045
- 52.↑
Nordgreen J, Tahamtani FM, Janczak AM, Horsberg TE. Behavioural effects of the commonly used fish anaesthetic tricaine methanesulfonate (MS-222) on zebrafish (Danio rerio) and its relevance for the acetic acid pain test. PLoS One. 2014;21;9(3):e92116. doi:10.1371/journal.pone.0092116
- 53.↑
Lopez-Luna J, Al-Jubouri Q, Al-Nuaimy W, Sneddon LU. Impact of stress, fear and anxiety on the nociceptive responses of larval zebrafish. PLoS One. 2017;12(8):e0181010. doi:10.1371/journal.pone.0181010
