Abstract
Visceral toxicosis of catfish (VTC), a sporadic disease of cultured channel catfish (Ictalurus punctatus) often with high mortality, is caused by botulinum neurotoxin serotype E (BoNT/E). Presumptive diagnosis of VTC is based on characteristic clinical signs and lesions, and the production of these signs and mortality after sera from affected fish is administered to sentinel catfish. The diagnosis is confirmed if the toxicity is neutralized with BoNT/E antitoxin. Because small catfish are often unavailable, the utility of adult zebrafish (Danio rerio) was evaluated in BoNT/E and VTC bioassays. Channel catfish and zebrafish susceptibilities were compared using trypsin-activated BoNT/E in a 96-hr trial by intracoelomically administering 0, 1.87, 3.7, 7.5, 15, or 30 pg of toxin per gram of body weight (g-bw) of fish. All of the zebrafish died at the 7.5 pg/g-bw and higher, while the catfish died at the 15 pg/g-bw dose and higher. To test the bioassay, sera from VTC-affected fish or control sera were intracoelomically injected at a dose of 10 µl per zebrafish and 20 µl/g-bw for channel catfish. At 96 hr post-injection, 78% of the zebrafish and 50% of the catfish receiving VTC sera died, while no control fish died. When the VTC sera were preincubated with BoNT/E antitoxin, they became nontoxic to zebrafish. Histology of zebrafish injected with either VTC serum or BoNT/E demonstrated renal necrosis. Normal catfish serum was toxic to larval zebrafish in immersion exposures, abrogating their utility in VTC bioassays. The results demonstrate bioassays using adult zebrafish for detecting BoNT/E and VTC are sensitive and practical.
Introduction
Visceral toxicosis of catfish (VTC), an important disease in channel catfish (Ictalurus punctatus) aquaculture, is caused by botulinum neurotoxin serotype E (BoNT/E).4,6,10 Outbreaks of VTC cause substantial losses to the catfish industry, accounting for as much as 37.8% of total loss in some years (National Animal Health Monitoring System: 2011, Catfish 2010 Part III: changes in catfish health and production practices in the United States, 2002–09. Available at: http://www.aphis.usda.gov/animal_health/nahms/aquaculture/). Visceral toxicosis of catfish occurs in the spring and fall when water temperatures are 18–21°C, and is characterized by a rapid onset with high losses, often occurring in market-sized and brood stock catfish. Clinical signs commonly observed with VTC include erratic swimming and progressive muscular weakness leading to paralysis, lethargy, and death.6,10 Internal lesions included chylous or clear fluid (ascites) in the coelomic cavity, intussusception of the intestinal tract, reticular pattern to the liver, congested spleen, and eversion of the stomach into the oral cavity.4,6,10
Botulinum neurotoxins are produced by Clostridium botulinum, a Gram-positive, obligate anaerobic, spore-forming, rod-shaped bacterium. 8 Botulinum intoxication occurs in human beings and various domestic species including cattle, horses, dogs, cats, birds, and cultured fish.2,17–19 Toxicosis caused by BoNT/E occurs in aquatic birds after they consume BoNT-laden fish. 13
There are several diagnostic methods to detect botulinum toxin. The assays most commonly used are the mouse bioassay, enzyme-linked immunosorbent assay (ELISA) for botulinum neurotoxins, and polymerase chain reaction (PCR) to detect DNA from the bacterium C. botulinum.6,7 However these assays are not sensitive enough to reliably detect the low concentrations of BoNT/E present in the serum of VTC-affected fish. 6
A presumptive diagnosis of VTC is made from clinical signs and lesions. 10 A definitive diagnosis is made using a catfish bioassay; small channel catfish fingerlings are injected intracoelomically with serum from affected catfish at a rate of 20 µl/g body weight (g-bw). A positive bioassay results in mortality with characteristic VTC signs and lesions within 96 hr. 6 Confirmation requires mixing the test serum with a BoNT/E neutralizing antiserum with a resulting loss of toxicity. The catfish bioassay is a practical answer to an immediate diagnostic challenge. However, it lacks standardization, and many diagnostic laboratories do not have facilities for holding channel catfish. Furthermore, the blood collection yields are often low from moribund or freshly dead catfish that have been submitted as a diagnostic case. Therefore, bioassay fish must be small enough to allow for the 20 µl/g-bw dose in 3–5 replicates. This is complicated by the fact that VTC outbreaks are generally in the fall and early spring when the availability of small channel catfish fingerlings is limited.
In comparison, zebrafish (Danio rerio; approximately 0.3–0.4 g) are much smaller than catfish fingerlings (approximately 5–20 g). If zebrafish were sensitive to BoNT/E, they could be used in the bioassay and considerably decrease the amount of VTC-affected catfish sera needed for diagnostic investigations. Additionally, zebrafish adults are relatively uniform in size and have defined genetics, allowing the assays to be standardized. Furthermore, zebrafish are much more economical and convenient to use than channel catfish because they are available year round and use much less tank space. The objective of the current study was to evaluate the zebrafish as a bioassay organism for BoNT/E and VTC by comparing sensitivity of zebrafish and channel catfish and to evaluate a serum neutralization bioassay as a definitive test. Also, because zebrafish larvae are easy to reliably produce, a preliminary study was performed using them in VTC bioassays.
Materials and methods
Fish maintenance
Zebrafish (Tübingen strain), 0.33 g mean weight, and channel catfish, 20 g mean weight, were produced and housed in the College of Veterinary Medicine, Mississippi State University (CVM-MSU) specific pathogen–free fish hatchery. Production of larvae and husbandry conditions were as previously described.9,12 Experimentally challenged fish were maintained in 15-liter aerated free-flowing tanks receiving charcoal-filtered dechlorinated municipal water at a rate of 0.5 l/min at 25°C. All production and use of fish were under the oversight and approval of the Mississippi State University Institutional Animal Care and Use Committee and the Institutional Biosafety Committee.
Botulinum neurotoxin
Stock BoNT/E a (1 µg/µl concentration with specific toxicity of activated [nicked] BoNT of 6 × 107 mouse lethal dose50/mg) was diluted to 10 ng/µl in sterile gel phosphate buffer (2 g/l of gelatin, 4 g/l of Na2HPO4; pH 6.2). For challenges, this BoNT/E stock was nicked by digestion with 5% trypsin b (1:250; 5 µl of 10 ng/µl BoNT/E, 100 µl of trypsin, and 895 µl of gel phosphate buffer) at 22–23°C for 30 min.
VTC sera
Visceral toxicosis of catfish sera (designated serum 3, 8, 10, and 12) were collected from moribund channel catfish from a confirmed natural VTC outbreak. Control sera were collected from healthy channel catfish, and stored at 4°C.
Bioassays
In the BoNT/E injection trial, 5 catfish and 5 zebrafish were injected with control or each nominal BoNT/E dose. The activated toxin stock (50 pg/µl) was diluted with gel phosphate buffer to working stocks of 1.0, 0.5, 0.25, 0.125, and 0.0625 pg/µl for zebrafish and 3.0, 1.5, 0.75, 0.375, and 0.187 pg/µl for catfish. Zebrafish were intracoelomically (IC) injected with 10 µl and catfish were IC injected with 10 µl/g-bw of the respective working concentrations to provide nominal doses of 30, 15, 7.5, 3.75, and 1.87 pg/g-bw.
In VTC sera bioassays, each serum sample (4 VTC samples and 1 control sample) was injected into 7 zebrafish and 3 channel catfish (numbers of catfish were reduced because of the limited amount of each VTC serum sample). Channel catfish were injected with 20 µl/g-bw, zebrafish were injected with 10 µl per fish of VTC sera or control sera. Fish were observed 3 times a day for BoNT/E clinical signs: hyperactivity, abnormal swimming pattern, fin in-coordination, lack of tail movement, lethargy, exophthalmia, and mortality.6,19 Moribund fish (fish that were unable to swim and settled to the bottom of the tank or were unable to maintain an upright position) were euthanized by immersion in water containing 300 mg/l tricaine methane sulfonate c for 10 min past the loss of opercular movement and considered dead at that time point. All dead fish were necropsied and the brain and posterior kidney cultured for bacteria to rule out bacterial infection as a cause of death. This consisted of plating the samples on tryptic soy agar with 5% sheep blood and incubating the cultures aerobically at 22°C. After 96 hr, fish that survived were euthanized as previously described.
Neutralization assay
The VTC serum samples were incubated with monospecific BoNT/E d antitoxin (provided at a concentration of 10 international units/ml) at a 4:1 ratio (catfish serum:antitoxin) for 30 min at 22–25°C. 9 The respective no-antibody control samples was produced by incubating the VTC serum with phosphate buffered saline (PBS) at the 4:1 ratio. Five zebrafish were IC injected for each antitoxin-treated serum sample and for each PBS-treated serum sample.
Histology
In the histology experiment, fish were injected with 15 pg/g-bw of BoNT/E, VTC serum, or control serum as described above. The zebrafish coelom is much smaller than that of the catfish, and an IC injection could produce iatrogenic pathology to the viscera. To assess the effect of IC injections, 10 µl of PBS (pH 7.4)-injected zebrafish and noninjected zebrafish were also processed and examined for histology. At least 5 fish were treated for each of the noninjected, BoNT/E-, control serum–, and PBS-injected zebrafish and control serum–injected catfish groups. Smaller numbers (n = 2) of catfish were injected with VTC sera because of limited availability. Also, 2 zebrafish injected with serum 3 and 1 injected with serum 10 died before sampling so they were removed from the study.
When fish became moribund, they were euthanized, cultured for bacteria, and processed for histology. Control serum– and PBS-injected fish were sampled at 72 hr postinjection. Zebrafish were fixed in 10% buffered formalin after making an incision the length of the abdomen. The whole body was embedded in paraffin, serially sagittally sectioned, and stained with hematoxylin and eosin. The channel catfish tissues were excised, then fixed and processed similarly.
Exposing zebrafish larvae
Three zebrafish larvae were distributed to each well of 6-well plates containing 2 ml of water per well, and 2 replicate wells were evaluated per treatment group. In the initial trial, larvae were exposed to serum from VTC-affected catfish, healthy catfish, or no serum. The dosage was 60 µl serum/2 ml. In the second trial, to assess whether components of catfish serum were toxic, zebrafish larvae were exposed to either 60 or 20 µl of serum/2 ml of either heat-inactivated or non–heat-inactivated serum. Heat inactivation involved incubating the serum at 56°C for 30 min.
Experiments with larvae were initiated on the day of hatch. For each replicate, 3 larvae were placed in a well of a 6-well cell culture plate with 2 ml of water, and a nominal amount of VTC or control serum was added to the well. After 3 hr, the water was replaced, and the larvae were observed 3 times a day. After 96 hr, surviving larvae were euthanized in tricaine methane sulfonate. b
Results
BoNT/E injections
All zebrafish and catfish that were exposed to the 2 highest dose levels of purified toxin, 30 and 15 pg/g-bw, died (Table 1). At the 7.5 pg/g-bw dose, all zebrafish died but no catfish died. These catfish displayed signs compatible with intoxication: erratic swimming, fin in-coordination, followed by resting on the bottom of the tank. No fish exposed to the lower doses or the negative control died. The zebrafish exposed to the 3.75 pg/g-bw dose displayed the same signs of BoNT/E or VTC intoxications seen in the 7.5 pg/g-bw dose catfish. All bacterial cultures taken from moribund zebrafish and catfish showed no growth.
Number of zebrafish and channel catfish fingerlings that died within 96 hr following an injection with purified botulinum neurotoxin serotype E (BoNT/E) activated with trypsin at a dose of 30, 15, 7.5, 3.7, 1.87, or 0 pg/gram-body weight (g-bw).*
All treatment groups consisted of 5 fish.
Majority of the live fish showed typical clinical signs of visceral toxicosis of catfish.
VTC sera injections
Mortality was observed in catfish and zebrafish injected with VTC sera (Table 2). Overall, the 4 samples of VTC-affected sera killed 78.5% of the zebrafish and 50% of the catfish. There were no mortalities observed in channel catfish or zebrafish that were injected with control catfish serum. All fish that were injected with VTC sera displayed lethargy, fin in-coordination, and paralysis, which are all signs typical of VTC. All bacterial cultures taken from moribund fish showed no growth.
Number of fish that died following an intracoelomic injection of visceral toxicosis of catfish sera or negative control (healthy) catfish serum.
All treatment groups consisted of 7 fish.
All treatment groups consisted of 3 fish.
Majority of live fish showed typical clinical signs of visceral toxicosis of catfish.
Neutralization assay
All zebrafish injected with VTC sera showed the signs of VTC within 12 hr post-injection but those injected with VTC sera that were preincubated with BoNT/E-specific antitoxin d displayed none of these clinical signs. The VTC-positive sera killed 80% of the zebrafish, while control sera and VTC sera that were pretreated with BoNT/E-specific antitoxin d caused no mortality (Table 3).
Number of zebrafish that died after an intracoelomic injection of visceral toxicosis of catfish sera, negative control (healthy) catfish serum, or the respective serum pretreated with botulinum neurotoxin serotype E neutralizing antibody.*
All treatment groups consisted of 5 zebrafish.
Histology
There were varying degrees of histologic changes in zebrafish and channel catfish fingerlings exposed to the VTC serum and purified toxin. The severity of lesions is summarized in Table 4. Noninjected control zebrafish demonstrated mildly congested gills, and 1 control fish demonstrated mild cloudy swelling and hydropic degeneration of renal tubular epithelial cells. All other tissues appeared normal. Of 5 zebrafish that were injected with PBS (negative control), all showed a small amount of focal necrosis at the injection site and mild branchial congestion. One of these fish demonstrated mild cloudy swelling of hepatocytes.
Summary of lesion severity in channel catfish and zebrafish following injection with purified botulinum neurotoxin serotype E, sera from channel catfish showing clinical signs of visceral toxicosis of catfish (VTC), control serum from a healthy channel catfish, phosphate buffered saline (PBS), or not injected.
PKid = posterior kidney; AKid = anterior kidney.
Of the 6 zebrafish injected with channel catfish control serum (healthy catfish serum), 4 demonstrated mild hepatopathy that included diffuse hepatic hydropic and ballooning degeneration, distorted architecture, granular cytoplasm, and loss of cellular definition. All 6 displayed mild renal pathology including cloudy swelling and hydropic degeneration of tubular epithelial cells, and varying changes of the Bowman space.
Three of the 10 zebrafish injected with sera from VTC-affected fish died soon after the injection and were not used for histology: 2 injected with serum 3 and 1 injected with serum 10. All remaining VTC serum–injected zebrafish were moribund at 12 hr postinjection, and sampled. In zebrafish injected with serum 3, liver tissues demonstrated hydropic and ballooning degeneration, distorted architecture, cloudy swelling, granular cytoplasm, mild focal necrosis, and intercellular edema. Pancreatic tissue demonstrated diffuse necrosis. Kidney tissue demonstrated severe cloudy swelling and hydropic degeneration of tubular epithelial cells, marked diffuse tubular necrosis, necrosis of the interstitial tissue, and intertubular edema. Zebrafish injected with serum 10 demonstrated hepatic congestion with moderate lesions, but exhibited more severe kidney lesions compared to those injected with serum 3. Kidney lesions included marked congestion, hemorrhage of the interstitial tissue, severe cloudy swelling and hydropic degeneration of tubular epithelial cells, marked diffuse tubular necrosis, necrosis of the interstitial tissue, and intertubular edema. One fish demonstrated increased Bowman space (multifocal), and mild inflammation of the interstitial tissue.
Five zebrafish were injected with 15 pg/g-bw purified BoNT/E and sampled at 12 hr postinjection. In all fish, renal tubular epithelia demonstrated hydropic degeneration, karyorrhexis, and necrosis. In 2 fish, hydropic degeneration of hepatocytes was seen.
Of the channel catfish injected with control catfish serum, 1 of 3 demonstrated mild cloudy swelling of hepatocytes; all other tissues appeared normal. Both catfish injected with VTC serum 3 demonstrated moderate hepatic vessel dilation, hepatic perivascular edema, hydropic degeneration, distorted architecture, and cloudy swelling in the hepatocytes. Marked changes in the kidney interstitial tissue compared to the control serum–injected fish included glomerular necrosis, edema, and tubular necrosis. The channel catfish injected with VTC serum 10 demonstrated hepatic vessel dilation and mild hydropic degeneration of hepatocytes. Kidney lesions were mild compared to those of fish injected with serum 3 and consisted of mild glomerular and tubular necrosis. Channel catfish injected with purified BoNT/E demonstrated mild to moderate hypertrophy of renal tubular epithelia and mild leukocytic infiltration of the renal interstitial tissue, congested spleens, and moderate hydropic degeneration of hepatocytes with granular cytoplasm.
Exposing zebrafish larvae
Within 24 hr, all larvae exposed to either control catfish serum or VTC serum died, whereas no fish died in the non–serum-exposed wells. Preexposure of the serum to heat (56°C for 30 min) was then evaluated to determine if the toxic component was temperature sensitive. All larvae exposed to 60 or 20 µl serum/2 ml died within 24 hr, whereas all larvae exposed to heat-inactivated serum survived.
Discussion
The results of the current study support the use of zebrafish as a bioassay organism to detect VTC and BoNT/E intoxication. The zebrafish and channel catfish bioassays using purified BoNT/E were similar in sensitivity. Bioassays using VTC sera demonstrated similar clinical signs in both species and the utility of both species for serotyping using neutralizing antibodies. Because adult zebrafish are readily available throughout the year, and they are relatively uniform in size and much smaller (0.3–0.4 g) than channel catfish, the zebrafish bioassay has more practical utility. Additionally, the defined genetics of inbred zebrafish strains likely reduces genetic-based variation in susceptibility, which should reduce standard error, allow for better between laboratory comparisons, and allow for more precise analytical research. Furthermore, holding facilities and care costs are much lower than that needed for a comparable number of channel catfish or mice. 9
The behavioral signs (hyperactivity, abnormal swimming pattern, fin in-coordination, lack of tail movement, lethargy, exophthalmia, and settling at bottom of the tank) of zebrafish and channel catfish were similar in VTC sera–injected fish and purified BoNT/E-injected fish. All the zebrafish died at 7.5 pg/g and higher doses and showed VTC behavioral signs at 3.7 pg/g. This sensitivity suggests that, with optimization, the zebrafish could be used in titration assays to quantify BoNT/E activity in tissues and serum. The alternative technology, endopep mass spectrometry, is a sensitive, quantitative method for detecting active BoNT/E, but it is expensive and time consuming. 6 The mortality and clinical signs were similar in BoNT/E- and VTC-challenged channel catfish and zebrafish. However, histology of zebrafish and catfish injected with control serum suggested some differences. There were mild hepatic and renal changes in the zebrafish that were injected with control serum, whereas changes in the catfish injected with control serum or in zebrafish that were injected with PBS were minimal (focal necrosis was seen at the site of injection but this was thought to be due to mechanical damage that occurred by manually injecting the small zebrafish). Also, larval zebrafish were very sensitive to a heat labile component in the catfish serum in immersion exposures. This suggests that the control adult zebrafish may have been affected by mild serum toxicity. These mild lesion caused by catfish serum should be considered in any future studies that involve cross-species serum injections in fish. Although the current study was not designed for comparative histology, some of the lesions seen in the VTC serum–injected catfish and zebrafish were different from those described in channel catfish with VTC. 15 The current study showed only mild intestinal changes and no marked lymphoid depletion. The differences may be a result of the route and duration of exposure to the toxin.
The preliminary study using zebrafish larvae in immersion exposures was unsuccessful because the serum of channel catfish was toxic to the larvae. The toxic component in the serum was heat labile. One possible explanation for the serum being toxic is heterotypic activation of the complement system. The complement proteins in channel catfish sera may have been activated by zebrafish proteins, and the membrane attack complex or other components from the resulting complement cascade could have killed the larval zebrafish. Cross-species activation of complement has been documented as the cause of hyperacute rejection of xenotransplantations 16 and has been implicated in the acute toxicity of human serum to mice. 14 Botulinum neurotoxins are heat labile, therefore heat inactivation of the serum is not an option, but zebrafish larvae may be useful as a bioassay for detecting heat-stable toxins in serum samples.
Zebrafish are an attractive and widely used model organism for biomedical studies, genetic research, developmental biology, immunology, and toxicology.3,5,11 Zebrafish are an oviparous cyprinid with year-round fecundity. 12 Previously, zebrafish have been used to study enteric septicemia of catfish 12 and the effects of cholera toxin and Clostridium difficile toxin.11,15 The zebrafish genome is fully sequenced; the availability of various recombinant zebrafish and antibodies to specific markers further aid in research. The zebrafish assay described in the current study is a simple, sensitive method to detect BoNT/E in VTC sera, which can likely be applied to other biological samples. The mouse bioassay is the only method approved by the Centers for Disease Control and Prevention and the most widely accepted definitive diagnostic test for presence of BoNT.1,13 However, the mouse bioassay was not sensitive enough to detect the presence of BoNT/E in VTC sera. Endopep mass spectroscopy can detect the low levels of BoNT/E presence in the VTC sera but it is time consuming and expensive. Therefore, the zebrafish bioassay described in the current study is a reliable, practical alternative to rapidly detect BoNT/E in sera from VTC-affected catfish. Furthermore, it is a convenient alternative to the previously established catfish fingerling bioassay.4,6
Footnotes
Declaration of conflicting interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
This work was supported by research grant USDA NIFA 2011-67015-30174 from the U.S. Department of Agriculture, the Mississippi Agricultural and Forestry Experiment Station, and the College of Veterinary Medicine, Mississippi State University.
a.
Metabiologics, Madison, WI.
b.
215240, Trypsin 1:250, 100 g; Difco, BD, Franklin Lakes, NJ.
c.
MS-222, Tricaine-S; Western Chemical Inc., Ferndale, WA.
d.
Catalog no. BS3179, Centers for Disease Control and Prevention, Atlanta, GA.