Zebrafish (Danio rerio) are becoming a widely-used vertebrate animal model for microbial colonization and pathogenesis. This protocol describes the use of the protozoan Paramecium caudatum as a vehicle for food-borne infection in zebrafish larvae. P. caudatum readily internalizes bacteria and get taken up by larval zebrafish through natural preying behavior.
Due to their transparency, genetic tractability, and ease of maintenance, zebrafish (Danio rerio) have become a widely-used vertebrate model for infectious diseases. Larval zebrafish naturally prey on the unicellular protozoan Paramecium caudatum. This protocol describes the use of P. caudatum as a vehicle for food-borne infection in larval zebrafish. P. caudatum internalize a wide range of bacteria and bacterial cells remain viable for several hours. Zebrafish then prey on P. caudatum, the bacterial load is released in the foregut upon digestion of the paramecium vehicle, and the bacteria colonize the intestinal tract. The protocol includes a detailed description of paramecia maintenance, loading with bacteria, determination of bacterial degradation and dose, as well as infection of zebrafish by feeding with paramecia. The advantage of using this method of food-borne infection is that it closely mimics the mode of infection observed in human disease, leads to more robust colonization compared to immersion protocols, and allows the study of a wide range of pathogens. Food-borne infection in the zebrafish model can be used to investigate bacterial gene expression within the host, host-pathogen interactions, and hallmarks of pathogenicity including bacterial burden, localization, dissemination and morbidity.
Zebrafish share morphologically and functionally conserved features with mammals, including granulocytic lineages (e.g., neutrophils), monocyte/macrophage-like cells, Toll-like receptors, pro-inflammatory cytokines, and antimicrobial peptides1. The intestinal tract in zebrafish is fully developed at 6 days post fertilization (dpf) and shows morphological and functional conservation with the mammalian gastrointestinal tract, such as conserved transcriptional regulation in intestinal epithelial cells2. This makes zebrafish an excellent model for intestinal microbial colonization and pathogenesis. A wide range of enteric microbes has been studied in the zebrafish model, including enterohemorrhagic Escherichia coli3, Vibrio cholerae4,5, Salmonella enterica6, the zebrafish microbiota7,8, and the role of probiotics in intestinal immunity9. A distinct advantage of the zebrafish model is that it is colonized by many microbes without disrupting the endogenous microbiota, which allows the investigation of microbial behavior in the context of mixed microbial populations3,6. Currently, most zebrafish models of gastrointestinal colonization and disease rely on the administration of microbes by bath immersion, where zebrafish are incubated in a bacterial suspension for a specific amount of time10. However, this makes it difficult to determine the exact dose of bacteria administered, and leads to limited colonization with some microbes, particularly with non-pathogenic bacteria. Alternatively, a bacterial suspension is administered to fish via oral gavage11, but this is technically challenging and limited to older larvae and adult fish.
This protocol describes the use of the unicellular protozoan Paramecium caudatum as a vehicle for food-borne delivery of microbes to the gastrointestinal tract of zebrafish larvae. Paramecia are easy and cheap to maintain and are capable of feeding on a wide variety of microbes, including algae, fungi, and bacteria, which they internalize through a ciliated oral groove12,13,14. Once internalized, bacteria are held in vacuoles, which eventually acidify and contents are degraded over a time frame of several hours15. Larval zebrafish capture paramecia as natural prey soon after hatching, around 3–4 dpf depending on temperature16, and take them up with high efficiency. The process of prey capture takes on average 1.2 s from detection to capture17, and captured paramecia are quickly digested in the zebrafish foregut, such that internalized viable bacteria are released into the intestinal tract3. As a result, paramecia can be used as a quick and easy method to deliver a high and consistent dose of bacteria into the gastrointestinal tract of zebrafish. The delivered bacteria can either be transformed to express a fluorescent protein, such as mCherry as described here, or, in the case of genetically intractable bacteria, they can be pre-stained with a fluorescent dye to allow visualization within the gastrointestinal tract.
This protocol describes the food-borne delivery of enteropathogenic E. coli (enterohemorrhagic E. coli [EHEC] and adherent invasive E. coli [AIEC]), and Salmonella enterica ssp. Typhimurium. Both pathogenic E. coli and S. typhimurium are transmitted via the fecal-oral route18,19, and can be acquired via contaminated food, such as meat, vegetables, and dairy. Using P. caudatum as a vehicle, E. coli and S. typhimurium successfully colonize the zebrafish larvae within 30–60 min of co-incubation with the paramecium vehicle. The achieved bacterial burden is robust enough to visualize colonization and determine burden by plating tissue homogenates.
Zebrafish care, breeding, and experiments described here are in accordance with the Guide for the Care and Use of Laboratory Animals, and have been approved by the Institutional Animal Welfare Committee of the University of Texas Health Science Center, protocol number AWC-16-0127.
1. Growth and Maintenance of Paramecia
2. Determination of Bacterial Dose Administered to Zebrafish
3. Food-borne Infection of Zebrafish
Paramecium caudatum readily internalizes a wide range of bacteria into storage vacuoles. The intracellular bacterial density depends on the densities of bacteria and paramecia in the co-culture, as well as the bacterial species used. Over time, the vacuoles acidify and bacterial degradation ensues. The rate of degradation has to be individually determined for all strains used. For pathogenic E. coli, the initial bacterial density is 790 bacteria/paramecium, and bacteria are degraded with a half-life of approximately 2.3 h (Figure 1).
Figure 1: Determination of bacterial half-life in paramecia. (A) Following 2 h of co-incubation with infectious E. coli, P. caudatum was washed and transferred to medium without bacteria. At the indicated time points, numbers of viable E. coli cells were determined by dilution plating on selective agar. Results are means ± standard error of the mean (SEM; n = 3). (B) Typical image of paramecium carrying internalized bacteria, with bright field (Bi), fluorescent bacteria (Bii), and merged channels (Biii). Scale bar = 20 μm. Please click here to view a larger version of this figure.
Further, the zebrafish preying rate, that is, the rate at which zebrafish internalize bacteria-loaded paramecia upon co-incubation, was studied. Larval zebrafish start to hunt and capture live prey from 5 dpf20, although it was found that, when raised at 30 °C, larval development is accelerated and animals display preying behavior from 4 dpf. Preying is accompanied by a characteristic striking behavior20 (Figure 2A), and the determination of the preying rate is based on the assumption that each strike leads to internalization of one paramecium, although this can only be regarded an approximation (see Discussion). Based on the herein described observations of preying zebrafish larvae, the rate of paramecia uptake is approximately 1,539 per h (Figure 2B).
Figure 2: Determination of zebrafish preying rate. (A) Still images from a preying video, showing a zebrafish larvae (5 dpf) preying on paramecia carrying fluorescent bacteria. Time in [seconds]. Arrow indicates the main axis of movement during striking. (B) Quantification of preying rate (paramecia intake per hour), based on n = 10 videos taken over the full 2 h exposure time. Please click here to view a larger version of this figure.
Following internalization of paramecia, the zebrafish efficiently degrades the prey in the foregut, releasing infectious bacteria into the digestive system. As described herein, paramecia degradation proceeds quickly, and free bacteria can be detected in the intestinal tract within 30 minutes of preying. Free bacteria then move from the foregut to the mid- and posterior intestine, where they are detected approximately 1–2 h after the beginning of preying (Figure 3). Bacterial persistence in the intestine depends on species and dose but ranges from several h to several days in the case of E. coli and S. enterica. S. enterica localizes primarily in the intestinal mucosae, with some epithelial invasion (Figure 3D), leading to infiltration of neutrophils into the epithelium (Figure 3C).
Figure 3: Colonization of zebrafish with bacteria. Zebrafish at 5 dpf were left uninfected (A) or colonized with mCherry expressing (B) E. coli or (C) S. enterica. Infection experiments may be performed in wild type (A and B) fish or transgenic lines (e.g., the line Tg(MPO::EGFP)i114 expressing green fluorescent neutrophils shown in (C). The rectal opening is marked by an arrow. (D) Higher magnification of intestinal section from whole-mount embedded larvae infected with Salmonella enterica infection. (Di) Blue = Hoechst marking nuclei, (Dii) Purple = phalloidin marking F-actin, (Diii) Red = Salmonella, (Div) merge. Scale bar = 5 μm. Please click here to view a larger version of this figure.
Movie 1: Video footage of the prey capture. Please click here to download this video.
The basic protocol described here has been optimized for pathogenic E. coli, and has been successfully adapted for other bacterial species, including Salmonella enterica and Vibrio cholerae. For some species that do not colonize the zebrafish gut following bath immersion, including some Salmonella enterica strains and some anaerobes, food-borne infection as described here can be used to successfully establish colonization. Compared to microgavage, which is also used to establish high bacterial burdens in the larval intestinal tract, food-borne infection is technically less challenging and requires less specialized equipment. However, critical parameters should be optimized for the bacterial species and strains to be used. Such factors include bacterial and paramecium density for the bacteria-paramecium co-culture step: If bacterial numbers within paramecia are low, this could be improved by increasing the bacterial density in the co-culture step. Some bacterial species may cause damage to the paramecium host, and this should be assessed by microscopy.
Another important factor in this protocol is prey capture by zebrafish. The preying rate as described here is based on the assumption that every prey capture strike results in the ingestion of one paramecium. High densities of paramecia per fish are used in the protocol to ensure high preying rates. However, prey capture is dependent on the density of paramecia in the system, and in very dilute paramecium cultures, preying rates may be as low as 13–15 paramecia per hour21,22. A limitation is that prey capture rates are also strongly dependent on lighting conditions and in the dark, capture rates are 80% lower than in light conditions21 and this should be taken into account when setting up experiments. If exposure times to prey have to be expanded to optimize colonization, consideration has to be given to secondary exposure to bacteria through feces. Under the conditions described above – 2 h of prey exposure – this exposure is negligible, since gut passage time of bacteria is more than 1 h and the concentration of bacteria in the vehicle is much higher than in feces. However, if prey exposure time is significantly increased, this may become a significant factor.
Appropriate controls should be included in this protocol, including colonization of zebrafish following feeding with paramecia containing non-pathogenic E. coli MG1655. If multiple bacterial strains are compared for their ability to colonize the zebrafish host, it is important to test whether their half-life within paramecia is comparable. Bacterial mutations, including those compromising cell wall integrity or acid sensing, may compromise bacterial stability within paramecia. In such cases, zebrafish feeding has to be adjusted to account for the differences in dosage.
The protocol described here can be used to investigate bacterial colonization and its consequences, including by imaging bacterial colonization of zebrafish as described above, as well as by determining CFU per zebrafish from tissue homogenate3, or investigating infection-associated morbidity and mortality. Ideally, for bacterial visualization, bacterial strains expressing fluorescent proteins such as mCherry or red fluorescent protein (RFP) should be used. This will allow the visualization of growing bacterial populations. If the bacterial strain is not genetically tractable or the use of fluorescent protein expression is precluded for other reasons, bacteria may be stained with a fluorescent dye, such as FM 4-64FX, prior to co-culture with paramecia. When using the protocol described here, co-culture with paramecia does not decrease the brightness of the dye and stained bacteria are clearly visible in the zebrafish intestine. However, the dye will be diluted over time should significant bacterial proliferation occur within the zebrafish host. In either case, red-fluorescent bacteria are preferable over green-fluorescent bacteria, since tissue autofluorescence can be higher in the green than in the red channel.
It has been found that this protocol can be adapted for aerobic and microaerophilic bacterial species. It may be possible to adapt it for the feeding of spores and fungal species, although this remains to be tested experimentally.
The authors have nothing to disclose.
We would like to thank members of the Krachler group for critical reading and comments on the manuscript. This work was funded by a UT Systems STAR award, the BBSRC, and the NIH (R01AI132354).
Paramecium caudatum, live | Carolina | 131554 | no not store growing cultures below room temperature |
0.4% Trypan Blue Solution | Sigma | T8154-20ML | liquid, sterile-filtered, suitable for cell culture; prepared in 0.81% sodium chloride and 0.06% potassium phosphate, dibasic |
Dimethyl sulfoxide (DMSO) | Sigma | 276855-100ML | store in a solvent safety cabinet |
Escherichia coli, MG1655 | ATCC | ATCC 700926 | can be replaced by any other non-pathogenic E. coli strain |
FM 4-64FX stain | Thermo Fisher | F34653 | aliquot and store frozen |
Formaldehyde | Sigma | F8775-4X25ML | |
LB Broth | Sigma | L3397-1KG | |
Phosphate buffered saline tablets | Thermo Fisher | 18912014 | |
Tetracycline | Sigma | 87128-25G | toxic, irritant |
Tricaine (Ethyl 3-aminobenzoate methanesulfonate) | Sigma | E10521-10G | |
Triton X-100 | Sigma | X100-100ML | |
Trypan Blue Solution, 0.4% | Sigma | 93595-50ML | |
UltraPure Low Melting Point Agarose | Thermo Fisher | 16520050 | |
hemocytometer or cell counter | any | ||
stereomicroscope | any | ||
table-top centrifuge | |||
microwave | |||
rotator wheel | |||
heated shaking incubator | |||
aquatics facilities | |||
breeding tanks |