Epithelioma papulosum cyprinid (EPC) cells were cultured in an M199 (HyClone, USA) culture medium with 10% fetal bovine serum (FBS) (Lonsa, USA). Wild-type E. ictaluri (PPD130/91 strain) were cultured in tryptic soy broth (TSB, BD Biosciences) at 28°C.

The specific primers (bid-F1: CCGGAATTCATGGACTTCAACAGGAAC and bid-R1: were designed according to the zebrafish Bid EST sequence (GenBank accession No: XM_005159109.4) and used for RT-PCR to clone the full open reading frame (ORF) of zebrafish Bid with the cDNA of the wild type of zebrafish as templates. The PCR products and p3xFLAG-CMV™-14 were digested using EcoR I and Xba I restriction enzymes and ligated using T4 DNA ligase (TaKaRa, USA). All plasmids were confirmed by sequencing.

To investigate the expression changes of zebrafish Bid during early development, wild-type zebrafish embryos or larvae were collected at 6 h, 12 h post-fertilization (hpf), 1 day (d), 3 days, 5 days, and 10 days post-fertilization (dpf). For each time point, 30–50 embryos or larvae were collected and used for RNA extraction.

To study the inducible expression pattern of Bid, zebrafish larvae at 4 dpf were infected with 2 × 10⁸ CFU/ml E. ictaluri in a total volume of 5 ml. After immersion in the bacterial suspension for 6 h, zebrafish larvae were maintained in 60-mm sterile disposable Petri dishes with 25 ml water. Larvae were collected at 12, 24, and 48 h post-infection (hpi). All samples were snap-frozen in liquid nitrogen and stored at −80°C.

To determine the role of Bid in apoptosis, EPC cells seeded overnight in six-well plates at 1 × 10⁶ cells per well were transiently transfected using Lipofectamine 2000 (Invitrogen, USA) with 2 μg p3xFLAG or Bid-FLAG plasmid. At 24 h post-transfection, the cells were infected with various concentrations of E. ictaluri with an MOI of 1°C, 5°C, and 10 at 28°C. After infection for 1 h, the medium was removed and replaced with MEM medium containing gentamicin (50 µg/ml) to kill the extracellular bacteria. At 6 hpi, the culture medium was discarded and trypsin without EDTA was added for 2 min. The cell suspension was collected, centrifuged at 4°C and 500 g, and washed twice with precooled PBS solution. Apoptosis was detected using an Annexin V-FITC/PI Apoptosis Detection Kit (Vazyme, China) on a CytoFLEX LX flow cytometer (Beckman, USA).

For in vitro bacterial invasion assays, EPC cells seeded overnight in 24-well plates at 3 × 10⁶ cells per well were transiently transfected using Lipofectamine 2000 (Invitrogen, USA) with 500 ng p3×FLAG or Bid-FLAG plasmid. Twenty-four hours post-transfection, the cells were infected with E. ictaluri at MOI 10. After infection for 1.5 h, the medium was removed and replaced with MEM medium containing gentamicin (50 µg/ml) to kill the extracellular bacteria. At 3 and 6 hpi, the cells were washed and lysed in PBS containing 1% (v/v) Triton X-100 at room temperature for 20 min. The lysates were diluted with TBS and then plated onto TSA to calculate the intracellular bacterial colony-forming units (CFU). The invasion assays were performed in triplicate.

For in vivo bacterial invasion assays, the p3×FLAG empty plasmid or Bid-FLAG plasmid was diluted to the desired concentration of 200 ng/μl and microinjected into fertilized eggs at the one-cell or two-cell stage with a injected volume of 2 nl. The injected embryos were raised to 4 dpf at 28°C and then were divided into two groups (three replicates per group, with 10 larvae per replicate) used for E. ictaluri infection. Briefly, the larvae were infected with E. ictaluri at an MOI of 10 for 6 h, then 20 ml aquaculture water was added. At 48 hpi, 10 larvae per group were collected, rinsed with 1 ml PBS, and ground via a glass homogenizer to obtain homogenates. Serial dilutions of the homogenates were plated onto TSB agar, and CFU were enumerated after 24 h of incubation at 28°C.

The p3×FLAG empty plasmid or Bid-FLAG plasmid was diluted to the desired concentration of 200 ng/μl and microinjected into fertilized eggs at the one-cell or two-cell stage with an injected volume of 2 nl. The injected embryos were raised to 4 dpf at 28°C, and the apoptosis-related genes, including birc5, epidermal growth factor receptor (egfr), caspase 8, caspase 3a, caspase 3b, caspase 9, programmed cell death protein 8 (PDCD8), bax, and bid, were examined using quantitative real-time PCR (qPCR). The expression of these genes at 12 h post-E. ictaluri infection was also examined using qPCR. The bacterial infection was completed as mentioned above. All primers used for gene expressions were listed in Table S1 .

To further analyze the roles of Bid and p53 for fish survival and antibacterial activity, the fertilized eggs at the one-cell or two-cell stage were randomly divided into three groups (three replicates per group, with 30 eggs per replicate): the control group, which was microinjected with p3×FLAG empty plasmid; the p53 group, which was injected with p53-FLAG; and the p53-Bid group, which was co-injected with Bid-FLAG and p53-FLAG. The injected embryos were raised to 4 dpf at 28°C and then were infected with E. ictaluri at an MOI of 10 for 6 days. The numbers of surviving larvae were counted daily, and the survival curves were generated by GraphPad Prism 7. The numbers of E. ictaluri in larvae were examined at 24 hpi as the in vivo bacterial invasion assays.

The fertilized eggs at the one-cell or two-cell stage were pretreated with 50 μM Z-IETD-FMK (caspase-8 inhibitor) (15) for 8 h, then infected with 2 × 10⁸ CFU/ml E. ictaluri. The survival rates of larvae were counted for 6 days. The larvae were also collected at 6, 24, and 48 hpi, and the number of E. ictaluri in larvae was examined as an in vivo bacterial invasion assay. In addition, the 4-dpf larvae, developed from the one-cell or two-cell stage fertilized eggs that were microinjected with p3×FLAG empty plasmid or Bid-FLAG plasmid, were treated with 50 μM Z-IETD-FMK for 8 h and then infected with 2 × 10⁸ CFU/ml E. ictaluri. The numbers of E. ictaluri in larvae at 24 and 48 hpi were calculated as mentioned above.

Total RNA was extracted from 30–50 zebrafish embryos or larvae using TRIzol reagent (Invitrogen, USA). The cDNA was reverse transcribed using a RevertAid™ First Strand cDNA Synthesis Kit (Thermo Fisher, USA). The qPCR was performed on a Thermo Fisher Scientific QuantStudio™ 3 Real-Time PCR Instrument (96-well 0.2 ml Block) under the following conditions: 3 min at 95°C, followed by 50 cycles of 15 s at 94°C, 15 s at 52°C–60°C, and 30 s at 72°C. All reactions were performed in triplicate in a 96-well plate, and the mean values were recorded. The specificity of primers was confirmed by sequencing and melting curves. The relative expression of target genes was normalized to the expression of housekeeping genes gapdh and EF-1α and expressed as fold changes relative to the corresponding control group. The primers used for qRT-PCR are listed in Table S1 .

Statistical analysis and graphs were performed and produced using GraphPad Prism 7.0 software. Expression data of qPCR are presented as means and standard error of the mean (SEM). Two-tailed Student’s t-test or ANOVA was used to compare means and SEM between groups. All data are representative of two or three biologic replications. The level of significance is shown as follows: *p < 0.05, **p < 0.01. Significance testing in the cumulative survival analysis used the log-rank test.

The ORF length of zebrafish bid is 564 bp (GenBank accession No. XM_005159109), encoding 187 amino acids. Zebrafish Bid contains a conserved BH3 domain and several conserved cleavage sites (9), including the caspase-8 cleavage site (LETD), the granzyme B cleavage site (HELQ), and the calpain cleave site ( Figure 1 ). Zebrafish Bid shares 70.4%, 66.5%, 34.5%, 16.3%, and 19.7% overall sequence identities with the Bid of common carp, grass carp, rainbow trout, mouse, and human, respectively ( Figure 1 ). To further understand the evolutionary history of vertebrates’ Bid, a phylogenetic tree was constructed using Mega X software with the neighbor-joining (N-J) method (16). The tree is divided into four main clades (clades A to D), supported by high bootstrap values (≥ 99%) ( Figure 2 ). Clade A is composed of Bid of fish. In clade A, zebrafish Bid clustered with the Bid of Cyprinidae. Clade B is composed of Bid from amphibians, and clade C is composed of Bid from birds. Clade D includes the Bid of mammals.

First, we analyzed the expression of zebrafish bid by qPCR. Results showed that zebrafish bid was expressed ubiquitously across all early developmental stages examined, showing a tendency first to decrease and then increase slowly ( Figure 3A ). We also assessed the expression changes of zebrafish bid in larvae infected with E. ictaluri. Results showed that zebrafish Bid was significantly upregulated at 48 hpi following E. ictaluri infection ( Figure 3B ).

To test whether Bid can induce apoptosis, we first transfected the EPC cells with Bid-FLAG or p3×FLAG empty plasmid. Then, the cells were infected with E. ictaluri, and the apoptosis rates were examined. At the early apoptosis stage, the apoptosis rates for cells transfected with Bid-FLAG were significantly higher than that transfected with empty plasmid when the MOIs of E. ictaluri were 5 and 10 (p < 0.05) ( Figure 4A ). At the late apoptosis stage, higher apoptosis rates for cells transfected with Bid-FLAG were also observed when the MOI of E. ictaluri was 10, compared with those for cells transfected with empty plasmid ( Figure 4B ).

The antibacterial activities of zebrafish Bid were assessed using bacterial invasion assays in vitro and in vivo. Results showed that overexpression of zebrafish Bid could significantly inhibit the proliferation of E. ictaluri in EPC cells at 3 and 6 hpi ( Figure 5A ). In vivo bacterial invasion assays showed that the numbers of E. ictaluri in fish microinjected with Bid-FLAG were significantly lower than those in fish microinjected with empty plasmid at 48 hpi ( Figure 5B ).

To further understand the antibacterial mechanism of zebrafish Bid, we first examined the expression of bicr5, egfr, p53, caspase8, caspase3a, caspase3b, caspase9, pdcd8, and bax in larvae microinjected with bid-FLAG. Results showed that without E. ictaluri infection, the expression of p53 was significantly inhibited in larvae microinjected with Bid. No significant changes were observed for other examined genes ( Figures 6A–C ). The larvae microinjected with Bid-FLAG were then infected with E. ictaluri, and the expressions of the above genes were examined at 12 hpi. Results showed that the expression of p53 was significantly downregulated and that birc5 and egfr were significantly upregulated ( Figures 6D–F ), when compared with those in larvae microinjected with the p3×FLAG empty plasmid.

The finding that overexpression of Bid could significantly inhibit the transcript of p53 with or without E. ictaluri infection ( Figure 6 ) indicated that Bid exerts its antibacterial activity by regulating p53. Thus, we first analyzed the effects of p53 on the survival rates of fish infected with E. ictaluri. The survival rate of larvae microinjected with p53-Flag was significantly lower than that of fish microinjected with the p3×FLAG empty plasmid (p = 0.0002) ( Figure 7A ). We also found that the survival rates of larvae co-microinjected with Bid-FLAG and p53-FLAG were significantly increased (p < 0.01) compared with those of larvae microinjected with p53-Flag ( Figure 7A ). Furthermore, the in vivo bacterial invasion assays showed that the number of bacteria in larvae transfected with p53 was significantly higher than that of cells transfected with p3×FLAG empty plasmid (p < 0.01) and cells co-transfected with Bid-Flag and p53-FLAG (p < 0.01) ( Figure 7B ). Further, the number of bacteria in cells co-transfected with Bid-FLAG and p53-FLAG was significantly lower than that of cells transfected with p3×FLAG empty plasmid (p < 0.01) ( Figure 7B ).

Z-IETD-FMK is the inhibitor of the caspase-8 inhibitor (15). We found that the larval survival at 6 dpi was decreased significantly when the larvae were pretreated with Z-IETD-FMK ( Figure 8A ). The in vivo bacterial invasion assays revealed that the numbers of E. ictaluri in larvae treated with Z-IETD-FMK at 24 and 48 hpi were higher than those of larvae without Z-IETD-FMK treatment ( Figure 8B ). Furthermore, we found that the number of E. ictaluri in larvae microinjected with bid-Flag decreased significantly, even if the larvae were treated in advance with Z-IETD-FMK ( Figure 8C ).