Electric organ discharge from electric eel facilitates DNA transformation into teleost larvae in laboratory conditions

Animal experiments

This study was approved by the Ethics Review Board for Animal Experiments at Nagoya University (approval number A210748-002). We conducted the experiments using live animals, ensuring that they were under anesthesia with 0.01% tricaine, in strict accordance with the institutional guidelines. In cases where the animals experienced severe health deterioration during the experiment, we euthanized them under ice anesthesia with 0.01% tricaine.

Fish maintenance

The electric eel (Electrophorus sp.) was purchased from Meito Suien Co., Ltd (Nagakute, Japan). The fish was housed in a freshwater tank (90 × 45 × 45 cm) at 27 °C under a 14 h:10 h light: dark photoperiod cycle. To regulate the pH of the tank water within the range of 7.0–8.0, approximately 1.5 g of sodium hydrogen carbonate was added to the fish tank once a week. They were fed a pellet diet (Meito Suien, Namazu-Gohan M) or live goldfish or Japanese loach.

The Riken-Mic zebrafish strain was utilized as the recipient for electroporation. Adult zebrafish were housed in a circulation water tank designed for small fishes (Meito system) and fed a powder diet or live Artemia. Fertilized eggs and embryos were maintained in 0.3% sea salt water.

Plasmids

The donor DNA used for electroporation was an 8,888-bp plasmid containing the GFP coding sequence driven by the Oryzias latipes β-actin promoter (accession number LC775392). Plasmids were amplified in DH-5α competent cells (TaKaRa, #9057) and subsequently extracted and purified using the Plasmid DNA Extraction Midi Kit (FAVORGEN, #FAPDE 002). The DNA solutions were stored at −20 °C until use.

Devices to record and analyze electric pulses

The electric eel’s discharge waveform was recorded using the following method. Recording and reference electrodes were constructed by connecting a 99.9% carbon rod (diameter: 10 mm, length: 100 mm) to a lead wire. One of these electrodes was positioned in an experimental tank (60 × 30 × 36 cm) containing a single electric eel, and it served as the reference electrode while being grounded. Additionally, one of the two electrodes was placed on the electric eel’s head as the positive pole, and the other carbon electrode was positioned on its tail as the negative pole. The voltage between the two electrodes was attenuated to 1/500 of the original signal using a High Voltage Differential Probe (Micsig Technology, #DP10013) and then input into a USB Data Acquisition Module (Data Translation, #DT9812-10V), which was connected to a Windows PC for display and recording. The sampling frequency was set to 20 kHz. The waveform recording and subsequent analysis were conducted using the DataView software (https://www.st-andrews.ac.uk/~wjh/dataview/).

Electroporation

The electric eel was relocated to the experimental tank, which was equipped with electrodes, a grounding system, and a computer containing the USB Data Acquisition Module (as depicted in Fig. 1A). The tank was filled with approximately 50 liters of freshwater maintained at 27 °C. The 6-day post-fertilization zebrafish larvae were incubated overnight in a 1 ng/µl plasmid solution diluted in 0.3% sea salt water. Subsequently, the larval samples were transferred to Electroporation Cuvette (#5510-11: Thermo Fisher Scientific, Waltham, MA, USA) containing a fresh 1 ng/µl plasmid DNA solution. Each cuvette accommodated a maximum of ten larvae, and the requisite number of cuvettes for each trial was attached to the electrode. This approach enabled exposure of more than ten larvae to an equivalent electric discharge in a single trial. The cuvettes were submerged in the experimental tank, and the electric eel initiated EOD during prey capture by feeding on an anesthetized goldfish. The goldfish was secured in the electrode using clips and a 6 cm thread, positioning the bait fish within 6 cm of the electric eel’s head, as illustrated in Fig. 1B. The experiment duration, from submerging the cuvettes to the final retrieval, was maintained at approximately 30 s for each trial. In most of the trials, the electrical discharge during prey capture was completed within this 30 s timeframe. Control groups were established without EOD, DNA, or both components. Even under the no electrical stimulation condition, the larvae were placed in the cuvettes and kept for the same duration as the EOD manipulation.

Fluorescent imaging

For fluorescent microscopy, zebrafish larvae that were 1-day post-electroporation were placed on 1.0% agar substrate within Petri dishes. Images were acquired using a Leica M165FC microscope (Leica Microsystems) equipped with a suitable filter set (GFP, #10447408; RFP, #10447417). Both brightfield and fluorescent images were captured using a Leica K3C digital color camera, and the imaging process was facilitated with Leica Application Suite (LAS) software. Following the microscopy, the zebrafish larvae that had survived were euthanized using ice anesthesia with 0.01% tricaine solution.

Statistical calculation

The data is presented as the mean ± standard error of the mean. Statistical analyses were conducted using the freely available and open-source software, Jeffreys’s Amazing Statistics Program (JASP), which can be accessed at https://jasp-stats.org/. A p-value below 0.05 was regarded as statistically significance.

Electric eel EOD exposure to zebrafish larvae

In this study, we employed an electric eel (Electrophorus sp.; gender unknown; approximately 60 cm in total length) as the electric source for our investigation (Fig. 1A). To set up the system, we arranged a test tank with electrodes and a computer. The electric eel was fed a goldfish to elicit the EOD, and high-voltage pulses were exposed to the zebrafish embryo in an electroporation cuvette field with a DNA solution at close range (Fig. 1B). We used a GFP expression plasmid driven by the O. latipes β-actin (actb) promoter as a transgenic indicator (Fig. 1C). Prior to the experiment, we confirmed that the actb:gfp cassette induced strong GFP fluorescence in zebrafish larvae (Fig. 1D). The electric eel emitted EOD waveforms when it received an anesthetized goldfish as prey (Fig. 1E, Movie S1). This waveform was characterized by a sequence of distinct high-voltage pulses synchronized with the eel’s bite on the prey (Fig. 2A). During the bite, the EOD waveform comprised a series of several to approximately 10 high-frequency pulses (200–400 Hz), followed by low-frequency pulses (about 50 Hz) (Fig. 2B). This waveform pattern was repeated several to about 10 times with individual bites. The recorded individual EOD pulse waveform exhibited a monopolar (DC), head-positive bell shape (Fig. 2C) during a single bite, which is different from the square pulses produced by machine electroporators typically used in electroporation into organisms (Figs. 2D and 2E). The amplitude of the pulses measured 184.6 ± 1.9 V (Fig. 2F), with a half-width of 0.501 ± 0.001 ms (n = 447; Fig. 2G). The intervals between the EOD pulses varied, with a median of 13.8 ms (mean ± se = 19.4 ± 2.1 ms; Fig. 2H). Within approximately 10 s, the electric eel generated around 500 pulses during a single bite. This pattern was repeated approximately 2-3 times during a single feeding behavior. As a result, in electroporation using EOD, on average, bell-shaped pulses were applied 1,000–1,500 times within 30 s. Although the number of pulses per bite, amplitude, and intervals differed in each trial or eel’s behavior, the half-width remained consistent and reproducible.

GFP fluorescence in zebrafish larvae after EOD exposure

At the outset, we attempted to employ fertilized eggs at the cleavage stages or 1- to 2-day post-fertilization (dpf) embryos for electroporation driven by electric eel discharges. Regrettably, these embryos did not survive, primarily due to yolk sac rupture during electric exposure or other manipulations. Consequently, we opted to utilize 7-dpf embryos, which had advanced beyond the yolk sac stages and exhibited a more robust body structure. In each trial, we subjected 20–49 larvae to the plasmid DNA solution using distinct electric eel discharges, conducting a total of fifteen repetitions (Table S1). Following these experiments, all specimens were subjected to examination under a stereomicroscope, and their survival rates and the extent of green fluorescence expression were assessed by a blinded evaluator. In our experimental conditions, no apparent damage was observed in the surviving specimens after the EOD exposure. The body shapes of these specimens closely resembled those of non-EOD-exposed specimens. Upon visual inspection, weak fluorescence was noted in the ventral yolk and dorsal skin of most zebrafish larvae, with more pronounced fluorescence detected in some of the samples (Fig. 3A). Occasionally, even in the absence of DNA, isolated single cells exhibited green fluorescence. Such signals were categorized as autofluorescence or false positives and were excluded from being considered as positive in all conditions. Even when multiple instances of green fluorescence were identified, signals overlapping with red fluorescence were not included in the count (Fig. 3B). To distinguish these background signals or false positives from the expected GFP signal, our focus was on clusters of multiple cells displaying intense green fluorescence while lacking red fluorescence. These clusters of GFP-positive cells were primarily observed on the body’s surface, including the trunk skin, yolk surface, eyes, and other cephalic tissues (Fig. 3C). Based on this criterion, the survival rate and GFP-positive rate were compared among the four conditions, EOD-/DNA-, EOD-/DNA+, EOD+/DNA- and EOD+/DNA+. Despite some trials in the EOD+/DNA- condition appearing to exhibit a decreased survival rate, the differences observed were not statistically significant according to the Dunnett’s test (mean ± sd = 98.4 ± 2.1% for EOD-/DNA- (n = 12) vs. 96.0 ± 3.1% for EOD-/DNA+ (n = 9), P = 0.898; EOD-/DNA- vs. 87.2 ± 14.8% for EOD+/DNA- (n = 9), P = 0.029; EOD-/DNA- vs. 89.6 ± 11.4% for EOD+/DNA+ (n = 15), P = 0.057; Fig. 4A). Importantly, 88% of the samples survived until 1 day after the manipulation, even in the EOD+/DNA+ condition (468/532; Table S1). The frequency of larvae exhibiting GFP fluorescence in the EOD+/DNA+ condition was significantly different from the EOD-/DNA- condition according to the Dunnett’s test (mean ± sd = 0.0 ± 0.0% for EOD-/DNA- (n = 12) vs. 0.5 ± 1.1% for EOD-/DNA+ (n = 9), P = 0.986; EOD-/DNA- vs. 2.3 ± 5.5% for EOD+/DNA- (n = 9), P = 0.538; EOD-/DNA- vs. 4.8 ± 6.4% for EOD+/DNA+ (n = 15), P = 0.025; Fig. 4B). In total, 5.3% of the survived EOD+/DNA+ larvae exhibited GFP-positive cells (25/468; Table S1).