Activated T cells of tilapia produce IFN-γ

The cloning of Nile tilapia IFN-γ cDNA has been reported in a previous study [37]. In the present study, we found that the IFN-γ gene is located on chromosome LG17 of Nile tilapia and its gene location and structure are evolutionarily conserved, as revealed by the synteny and exon–intron analyses (S1A and S1B Fig). As previously shown tilapia IFN-γ possesses a signature motif and nuclear localization signal (NLS), and it clusters well with IFN-γ homologues from other vertebrates (S1C and S1D Fig). These observations thus suggest that the Nile tilapia encodes a typical IFN-γ. As a multipotent cytokine, IFN-γ is principally produced by lymphocyte lineages such as Th1 cells, CD8⁺ T cells, and natural killer cells. Here, IFN-γ transcription was detected in all examined lymphoid-related tissues of tilapia, including head kidney, spleen, trunk kidney, liver, brain, skin, gill, intestine, and peripheral blood, with the highest expression level in peripheral blood (Fig 1A). To investigate the association of IFN-γ with lymphocytes in tilapia, we used a previously established protocol to isolate tilapia leukocytes. According to our previous report [36], lymphocytes constitute more than 90% of the isolated spleen leukocytes (Fig 1B). Thus, these spleen leukocytes containing a high proportion of lymphocytes were considered as lymphocytes in subsequent experiments.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 1. Tilapia T cells produce IFN-γ.

(A) The expression of IFN-γ in different tissues was examined by qPCR, n = 5. (B) Flow cytometry analyzed the FSC/SSC profiles of spleen leukocytes and the gate represents lymphocytes. (C-E) Spleen leukocytes were stimulated with PMA plus ionomycin (P+I) (C), PHA (D) or α-CD3ε/α-CD28 (E), and relative mRNA levels of IFN-γ were examined by qPCR, n = 4–6. (F) SDS-PAGE assay of recombinant IFN-γ protein. (G) Maldi-TOF-TOF analysis of recombinant IFN-γ protein. (H) Candidate IFN-γ mAbs were screened by ELISA. (I) Western blot assay showed the binding of mAbs to overexpressed tilapia IFN-γ in HEK 293T cells. (J-P) Spleen or head kidney leukocytes were stimulated with P+I for 4 h in the presence of GolgiPlug. The expression of IFN-γ in leukocytes was detected by western blot (J) or flow cytometry (K), and the expression of IFN-γ in CD3⁺ T cells from spleen (M) and head kidney (O) were detected by flow cytometry. Scatter plot figures showed the percentages of IFN-γ in gated lymphocytes (L) or CD3⁺ T cells (N, P), n = 6. These experiments were repeated for three independent times. *: p<0.05, **: p<0.01, ***: p<0.001, determined by a two-tailed Student’s t-test.

https://doi.org/10.1371/journal.ppat.1010913.g001

IFN-γ transcription in spleen lymphocytes was markedly induced by PMA plus ionomycin (P+I) stimulation (Fig 1C), indicating that activated lymphocytes potentially produce IFN-γ in tilapia. To further assess the expression of IFN-γ by T-cell lineages, spleen lymphocytes were stimulated by T-cell specific mitogen PHA or anti-tilapia CD3ε mAb plus anti-tilapia CD28 mAb. As predicted, T-cell activation enhanced IFN-γ expression (Fig 1D and 1E), thus providing transcriptional evidence that activated T cells of tilapia produce IFN-γ. However, evidence regarding IFN-γ-producing cells in early vertebrates is still lacking at the protein level owing to the scarcity of reliable antibodies. To this end, tilapia IFN-γ recombinant protein was prepared using a prokaryotic expression system (Fig 1F), confirmed by mass spectrometry (Fig 1G), and used to immunize mice for mAb development. After four rounds of immunizations, the splenocytes of mice were fused with SP2/0 myeloma cells. The supernatants from more than ten hybridoma clones, including 1E10, 2A3, 2G7, and 1A7, showed binding to recombinant tilapia IFN-γ (Fig 1H). The hybridoma 1A7 was then selected for sub-cloning by limiting dilution, and supernatants from sub-clone 1A7C8, 1A7C9, 1A7G10, and 1A7H9 were confirmed to specifically recognize overexpressed tilapia IFN-γ in HEK 293T cells (Fig 1I). Hybridoma cells obtained from the clone 1A7H9 were intraperitoneally injected into mice, and ascites were collected, followed by purification and biotin labelling of mAbs. Using this IFN-γ mAb, we revealed that spleen lymphocytes produced IFN-γ protein upon activation by P+I stimulation (Fig 1J). In addition, the proportion of IFN-γ⁺ lymphocytes in spleen and head kidney leukocytes was markedly increased after activation (Fig 1K and 1L). Furthermore, using tilapia CD3ε mAb, we found that the percentage of CD3⁺ T cells that produced IFN-γ was significantly higher after P+I stimulation in both spleen (Fig 1M and 1N) and head kidney (Fig 1O and 1P), suggesting that activated T cells of tilapia produce IFN-γ. Therefore, our findings at the transcriptional, protein, and cellular levels collectively support the conclusion that activated T cells of Nile tilapia produce the cytokine IFN-γ.

Th1 cells are present in Nile tilapia

IFN-γ is a signature cytokine of Th1 cells. Although sorted CD4-1⁺ lymphocytes of zebrafish have been reported to express IFN-γ at the mRNA level [32], evidence at the protein level and for CD3⁺CD4-1⁺ T cells is still lacking. Therefore, we sought to investigate the presence of CD3⁺CD4⁺IFN-γ⁺ cells in tilapia. In bony fish, the CD4-1 receptor with four Ig domains is considered as a homologue of mammalian CD4 [14]. Herein, we developed CD4-1 mAb to identify the CD4⁺ T-cell lineage in tilapia. The full-length coding region of CD4-1, which encodes four Ig domains and one transmembrane domain (Fig 2A), was cloned into the MIGR1 vector. Then, BOSC23 packaging cells were transfected with the resulting plasmid for retrovirus generation (Fig 2B). NIH/3T3 cells were infected with the CD4-1 retrovirus to obtain CD4-1-expresssing cells, which were used for mouse immunization (Fig 2C and 2D). After four immunizations, mouse splenocytes were fused with SP2/0 myeloma cells, and the hybridomas were screened by flow cytometry. The supernatant of the hybridoma 1C3G6 yielded a distinct positive population among gated tilapia spleen lymphocytes (Fig 2E and 2F). To determine whether this candidate mAb recognizes the tilapia CD4-1 receptor, we sorted the CD4-1⁺ or CD4-1^- spleen lymphocytes and analyzed their gene expression patterns. CD4-1 was expressed specifically in CD4-1⁺ lymphocytes, whereas CD8α, IgM, and CD20 were expressed only in CD4-1^- cells (Fig 2G). Additionally, CD4-1⁺ lymphocytes showed considerably higher transcript levels of CD3ε and TCRβ than CD4-1^- lymphocytes (Fig 2G). Furthermore, the immunofluorescence staining of spleen lymphocytes using anti-CD3ε and anti-CD4-1 mAbs showed that most CD4-1⁺ cells were CD3ε positive, whereas some CD3ε⁺ cells, likely CD8⁺ T cells, did not express CD4-1 (Fig 2H). These findings collectively confirmed the validity of this CD4-1 mAb, and the availability of reliable mAbs against CD3ε, CD4-1, and IFN-γ facilitates the study of Th1 cell response in tilapia. Flow cytometry revealed that, within the gated spleen lymphocyte population, few CD3⁺CD4-1⁺ T cells were IFN-γ⁺ in absence of stimulation (Fig 2I and 2J). However, activation of T cells by P+I markedly enhanced the capacity of CD3⁺CD4-1⁺ T cells to produce IFN-γ, as indicated by the higher proportion of CD3⁺CD4-1⁺IFN-γ⁺ T cells (Fig 2I and 2J). Similar results were found in the head kidney of Nile tilapia (Fig 2K and 2L). Overall, these results indicate for the first time that CD3⁺CD4-1⁺IFN-γ⁺ Th1 cells are present in bony fish.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 2. Th1 cells are present in tilapia.

(A) Domain prediction of tilapia CD4-1. The accession number of CD4-1 was listed in S1 Table. (B) BOSC23 cells were transfected with MIGR1-CD4-1 to produce retrovirus, and the fluorescence microscopy showed the BOSC23 cells expressing CD4-1. (C, D) Fluorescence microscopy (C) and flow cytometry (D) showed the NIH/3T3 cells expressing tilapia CD4-1. (E, F) Dot plots (E) and overlaid histogram (F) showed the tilapia spleen leukocytes stained with the CD4-1 mAb 1C3G6. (G) Gene expression profiles of CD4-1⁺ and CD4-1^- lymphocytes were detected by RT-PCR. (H) Immunofluorescence showed spleen leukocytes that stained with CD3ε (green) and CD4-1 (red) mAbs. (I-L) Spleen and head kidney leukocytes were stimulated with P+I for 4 h in the presence of GolgiPlug. Flow cytometry showed the expression of IFN-γ in CD3⁺CD4-1⁺ T cells from spleen (I) and head kidney (K). The percentages of IFN-γ in CD3⁺CD4-1⁺ T cells from spleen (J) and head kidney (L), n = 6. These experiments were repeated for three independent times. *: p<0.05, **: p<0.01, ***: p<0.001, determined by a two-tailed Student’s t-test.

https://doi.org/10.1371/journal.ppat.1010913.g002

Th1 cells are crucial for tilapia to resist intracellular bacterial infection

Next, we investigated whether Th1 cells of tilapia exert immunological functions. In mammals, upon encountering an intracellular pathogen infection, naïve CD4⁺ T cells preferentially differentiate into Th1 cells, which predominantly produce IFN-γ and lymphotoxin, leading to the eradication of intracellular pathogens [1]. However, the involvement of Th1 cells in the adaptive immune response and their detailed functions in bony fish are not well understood. In the present study, the Th1 cell response was investigated after tilapia was infected by the intracellular bacteria Edwardsiella piscicida. On 5 days post infection (DPI), the percentage and absolute number of CD3⁺ splenic T cells were markedly higher than those in the control fish (Fig 3A and 3B), indicating the robust initiation of a T-cell response during the primary immune response in tilapia. Moreover, CD3⁺CD4-1⁺ T cells were more abundant in tilapia infected with E. piscicida than in control one (Fig 3A and 3C). Notably, more CD3⁺CD4-1⁺ T cells were observed to produce IFN-γ in infected fish (Fig 3D and 3E), suggesting that intracellular bacterial infection triggers a robust Th1 cell response in Nile tilapia. To further verify the crucial role of CD4⁺ T cells or Th1 cells in anti-bacterial immune response, CD4-1⁺ cells were selectively depleted in tilapia using anti-CD4-1 mAb in combination with tilapia anti-mouse IgG1 serum (Fig 3F). On day 9 after antibody administration, approximately 60% of CD4-1⁺ T cells were depleted in the experimental group (Figs 3G and 3H and S2). Then, the animals were infected with E. piscicida. Compared with the uninfected tilapia (Fig 3A and 3B), nondepleted fish infected with E. piscicida showed a marked expansion of CD4-1⁺ T cells on 5 DPI (Fig 3I middle panel and 3J), but infected fish with CD4-1⁺ T-cell depletion exhibited a significant defect in CD4-1⁺ T-cell expansion during bacterial infection (Fig 3I and 3J). Notably, the ability of CD3⁺CD4-1⁺ T cells to produce IFN-γ (Fig 3K) and the percentage and absolute number of CD3⁺CD4-1⁺IFN-γ⁺ Th1 cells (Fig 3L) were markedly lower in CD4-1⁺ T-cell-depleted tilapia. The deficiency of CD3⁺CD4-1⁺ T cells and Th1 cells made it challenging to manage the infection (Fig 3M) and rendered the tilapia more vulnerable to E. piscicida (Fig 3N). Collectively, our findings suggest that Th1 cells are crucial for tilapia to resist intracellular bacterial infection.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 3. Tilapia Th1 cells are essential for resisting intracellular bacterial infection.

(A-E) Tilapia individuals that infected with E. piscicida or uninfected controls were i.p. injected with BFA 6 hours before sacrifice, and spleen lymphocytes were isolated for assays. (A) Flow cytometry showed the cells stained with CD3ε and CD4-1 mAbs. (B, C) Scatter plot figures showed the percentage and absolute numbers of CD3⁺ T cells (B) and CD3⁺CD4-1⁺ T cells (C), n = 5. (D, E) Representative dot plots (D), percentage and absolute numbers (E) of IFN-γ⁺ cells in gated CD3⁺CD4-1⁺ T cells, n = 5. (F) Strategy for CD4-1⁺ cells depletion. (G) Flow cytometry showed the percentage of CD3⁺CD4-1⁺ T cells in spleen lymphocytes of CD4-1-depleted or non-depleted tilapia on 9-day post depletion. (H) Scatter plot figures showed the percentages of CD4-1⁺ T cells among lymphocytes or T cells of isotype control and CD4-1-depleted tilapia, n = 3. (I-N) Tilapia individuals were infected with E. piscicida or not on day 9 after depletion, and the spleen leukocytes were harvested on 5 DPI for assay. Flow cytometry and scatter plot figures showed the percentage and absolute numbers of CD3⁺CD4-1⁺ T cells (I, J) and CD3⁺CD4-1⁺IFN-γ⁺ T cells (K, L), n = 4. (M) E. piscicida titers in spleen of CD4-1-depleted or nondepleted tilapia on 5 DPI, n = 5. (N) Kaplan-Meyer survival plot showed the survival percentage of CD4-1-depleted or nondepleted tilapia during E. piscicida infection, n = 25. These experiments were repeated for at least two independent times. *: p<0.05, **: p<0.01, ***: p<0.001, determined by a two-tailed Student’s t-test.

https://doi.org/10.1371/journal.ppat.1010913.g003

Th1 cells facilitate the anti-bacterial response of tilapia by enhancing macrophage activity

Because Th1 cells were involved in resisting the E. piscicida infection in tilapia, we further investigated the precise mechanism by which Th1 cells mediated this immune response. First, we found that recombinant IFN-γ activated tilapia macrophages in vitro, as evidenced by the induced expression of proinflammatory cytokines IL-1β, TNF-α, and IL-12 in these cells (Fig 4A), and consequently enhanced their phagocytic ability against E. piscicida (Fig 4B). E. piscicida infection caused severe mortality in tilapia within 8 days (Fig 4C), but IFN-γ supplementation rendered the animals less vulnerable to this pathogen (Fig 4C) and promoted elimination of the infection, as indicated by the lower bacterial burden in the spleen (Fig 4D). Moreover, in CD4-1⁺ cell-depleted tilapia, whose Th1 cells were markedly impaired during E. piscicida infection (Fig 3K and 3L), the phagocytic ability of macrophages was correspondingly weakened (Fig 4E and 4F), indicating the critical role of CD4⁺ T cells or Th1 cells in regulating macrophage activity in tilapia. The dependency of macrophage activity on activated T cells was also confirmed by an in vitro experiment, in which macrophages in CD3-mAb-activated leukocytes exhibited higher phagocytic activity toward E. piscicida (Fig 4G and 4H). However, the increase in phagocytic activity induced by activated T cells was markedly blocked by anti-IFN-γ mAb (Fig 4G and 4H), suggesting that T cells facilitate macrophage-mediated immune response by producing IFN-γ. To further confirm whether the IFN-γ secreted by Th1 cells enhances macrophage anti-bacterial activity in a cognate manner, we sorted macrophages and CD3⁺CD4-1⁺ T cells from a single tilapia individual. The macrophages co-cultured with activated CD4⁺ T cells exhibited robust phagocytic activity toward E. piscicida, and this response was significantly diminished by adding anti-IFN-γ mAb (Fig 4I and 4J), indicating the essential role of IFN-γ produced by CD3⁺CD4-1⁺ T cells in regulating macrophage activity. Thus, these data demonstrate that Th1 cells facilitate the anti-bacterial response of tilapia by enhancing macrophage activity.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 4. Th1 cells activate macrophages to facilitate anti-bacterial immune response in tilapia.

(A) Head kidney macrophages were stimulated with recombinant IFN-γ protein, and relative mRNA levels of indicated molecules were examined at 12 h post stimulation by qPCR, n = 4–6. (B) Dot plots showed the phagocytic rates of head kidney macrophages against FITC-labelled E. piscicida in the presence or absence of recombinant IFN-γ. (C, D) Tilapia individuals that infected with E. piscicida were injected with recombinant IFN-γ or PBS. (C) Kaplan-Meyer survival plot showed the survival percentage of tilapia, n = 25. (D) E. piscicida titers in spleen on 5 DPI, n = 6. (E, F) CD4-1⁺ cells were depleted as described in Fig 3F. Tilapia individuals were infected with E. piscicida on day 9 after depletion, and head kidney leukocytes were harvested on 5 DPI for assay. The flow cytometry (E) and scatter plot figures (F) showed the phagocytosis of macrophages from CD4-1-depleted or nondepleted tilapia against FITC-labelled E. piscicida at 12 h, n = 4–5. (G, H) The head kidney leukocytes that stimulated with CD3ε mAb or not in the presence of IFN-γ mAb or isotype antibody were incubated with FITC-labelled E. piscicida. The flow cytometry (G) and scatter plot figures (H) showed the phagocytosis of macrophages at 12 h, n = 4. (I, J) The head kidney macrophages and CD3⁺CD4-1⁺ T cells that sorted from the same tilapia individual were co-cultured with or without CD3ε mAb, IFN-γ mAb and FITC-labelled E. piscicida. The flow cytometry (I) and scatter plot figures (J) showed the phagocytosis of macrophages at 12 h, n = 4. These experiments were repeated for at least two independent times. *: p<0.05, **: p<0.01, ***: p<0.001, determined by a two-tailed Student’s t-test.

https://doi.org/10.1371/journal.ppat.1010913.g004

Tilapia utilize the conserved STAT1/T-bet axis to initiate Th1 cell differentiation

The IFN-γ production and Th1 cell differentiation are transcriptionally controlled by multiple transcription factors. Of these, T-bet, STAT1, and STAT4 are the most important factors responsible for Th1 cell development [38], and we found that all these transcription factors are encoded in the Nile tilapia genome (S3A Fig). Similar to its mouse homologue, tilapia T-bet possesses the TBOX domain, a conserved core domain for sequence-specific DNA binding (S3B Fig). Additionally, arrangement of the functional domains of tilapia STAT1 and STAT4 is similar to that in their respective mouse homologues (S3C and S3D Fig). Both STAT1 and STAT4 proteins contain a STAT_int domain at the N terminus, which is crucial for protein interaction, phosphorylation, nuclear translocation, and transcriptional activity [39,40]. The STAT_int domains are followed by a STAT_α domain, a STAT_bind domain, and an SH2 domain, which ensure the proper interaction of STAT1/4 with proteins, DNA, and the phosphorylated tail of other STATs, respectively [41–43]. A STAT1 TAZ2-binding domain, which enables a protein to induce or modulate the transcription of target genes [44], is located at the C-terminal end of STAT1 but not STAT4 (S3C and S3D Fig). Notably, the tertiary structure of tilapia T-bet, STAT1, and STAT4, was highly similar to that of their respective mouse homologues (S3E–S3G Fig). Together, these observations suggest that tilapia possesses a set of evolutionarily conserved transcription factors for Th1 cell differentiation.

Next, we examined the regulation of these transcription factors on IFN-γ transcription and Th1 cell differentiation in tilapia. The mRNA levels of T-bet, STAT1, and STAT4 were significantly higher in spleen lymphocytes activated by P+I (S4A Fig), PHA (Fig 5A), or CD3 plus CD28 mAbs (S4B Fig) than in control cells. Furthermore, the proportion of spleen lymphocytes in which STAT1 was phosphorylated markedly increased early after activation (Figs 5B and S4C). Moreover, overexpression of tilapia T-bet and STAT1 but not STAT4 elevated the transcription levels of IFN-γ in HEK 293T cells (Fig 5C), and the LUC activity driven by the IFN-γ promoter was enhanced by T-bet but not by STAT1 or STAT4 (Fig 5D), indicating that tilapia T-bet promotes IFN-γ expression by interacting with its promoter regions, whereas STAT1 is indirectly involved in this process. Additionally, STAT1 inhibition by the specific inhibitor Fludarabine [45] significantly reduced T-bet expression upon PHA- or CD3 mAb-induced T-cell activation (S4D Fig), suggesting that the T-bet-regulated IFN-γ transcription is STAT1 dependent. In this case, we knocked down STAT1 and T-bet to investigate whether the STAT1/T-bet axis is involved in Th1 cell differentiation in tilapia. Interference of STAT1 impaired the T-bet and IFN-γ expression induced by T-cell activation (Fig 5E), whereas T-bet interference compromised IFN-γ expression (Fig 5F); however, STAT1 or T-bet knockdown did not affect STAT4 expression (S4E and S4F Fig). Moreover, deficiency of STAT1 or T-bet diminished CD3⁺CD4⁺IFN-γ⁺ cells upon T-cell activation (Fig 5G). Notably, during E. piscicida infection, STAT1 or T-bet knockdown, which impaired the expression of key transcription factors and IFN-γ at the mRNA (S4G and S4H Fig) or protein level (Fig 5H), reduced the proportion (S4I and S4J Fig) and absolute number (Fig 5I–5L) of both CD3⁺CD4-1⁺ T cells and IFN-γ-producing CD3⁺CD4-1⁺ T cells. Thus, these findings highlight the central role of the STAT1/T-bet axis in regulating Th1 cell differentiation in tilapia. Not only gene knockdown but the blockade of STAT1 activity by Fludarabine also impaired the E. piscicida-induced IFN-γ expression and CD3⁺CD4⁺IFN-γ⁺ cells in spleen lymphocytes (Fig 5M and 5N), compromised bacterial clearance (S4K Fig), and eventually increased the mortality of tilapia (Fig 5O). Considering that IL-12 signaling and the transcription factor Runx3 are crucial for Th1 cell differentiation in mammals, we examined their expression during T-cell activation in tilapia. The mRNA expression of IL-12 receptors, but not Runx3, was induced upon PHA or TCR stimulation (S5 Fig), indicating that IL-12 signaling is associated with Th1 cell differentiation in tilapia as well. Together, our findings suggest that the conserved STAT1/T-bet axis is crucial for initiating Th1 cell differentiation in tilapia.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 5. STAT1/T-bet axis regulates the Th1 differentiation in tilapia.

(A) Spleen leukocytes were stimulated with PHA for 12 h. Relative mRNA levels of T-bet, STAT1 and STAT4 were examined by qPCR, n = 4. (B) Flow cytometry showed the phosphorylation level of STAT1 in lymphocytes that stimulated with PHA. (C) Transcriptional levels of IFN-γ in HEK 293T cells that transfected with tilapia T-bet, STAT1 and STAT4, n = 6. (D) HEK 293T cells were co-transfected with tilapia T-bet, STAT1 or STAT4, and pGL3-IFN-γ promoter. The LUC activities were assessed at 48 h post-transfection, n = 4. (E-G) Spleen leukocytes from tilapia that i.p. injected with T-bet-specific, STAT1-specific or control siRNA for 2 days were harvested and stimulated with CD3ε mAb for 12 h. Relative mRNA levels of STAT1, T-bet and IFN-γ (E, F, n = 4), and the percentage of CD3⁺CD4-1⁺IFN-γ⁺ T cells (G) were examined. (H-L) Tilapia i.p. injected with T-bet-specific, STAT1-specific or control siRNA were infected with E. piscicida. Tilapia was i.p. injected with BFA 6 h before sacrifice, and spleen leukocytes were harvest for assay. Western blot assay showed the expression of IFN-γ at 48 h post infection (H). Flow cytometry and scatter plot figures showed the percentage and absolute numbers of CD3⁺CD4-1⁺ T cells (I, K) and CD3⁺CD4-1⁺IFN-γ⁺ T cells (J, L) on 5 DPI, n = 4. (M-O) Tilapia individuals that infected with E. piscicida were injected with STAT1 inhibitor Fludarabine or PBS, and animals were i.p. injected with BFA 6 hours before sacrifice. (M) Western blot assay showed the protein levels of IFN-γ in spleen leukocytes on indicated days. (N) Flow cytometry showed the percentage of IFN-γ in spleen CD4-1⁺ T cells on day 7 post-infection. (O) Kaplan-Meyer survival plot showed the survival percentage of tilapia, n = 25. These experiments were repeated for at least two independent times. *: p<0.05, **: p<0.01, ***: p<0.001, determined by a two-tailed Student’s t-test. The accession numbers of selected sequences were listed in S1 Table.

https://doi.org/10.1371/journal.ppat.1010913.g005

IL-2 induces Th1 cell differentiation of tilapia via STAT5 and mTORC1 signaling

Differentiation of naïve CD4⁺ T cells into the Th1 lineage is tightly regulated by several cytokine signaling pathways. IL-2 initiates Th1 cell development by promoting the STAT5-dependent expression of IL-12Rβ2 and T-bet [46]. Although the IL-2 gene was not annotated in the Nile tilapia genome in NCBI GenBank, the genomic synteny analysis identified a gene located at the same locus as IL-2 homologues in other vertebrates (Fig 6A). The fact that this candidate gene exhibits a similar organizational structures, functional domains, tertiary structure, and phylogenic profile as the IL-2 homologues in other vertebrates (S6 Fig), suggests that this unannotated gene encodes tilapia IL-2. Upon PHA-induced T-cell activation, IL-2 expression in spleen lymphocytes was evidently upregulated (Fig 6B). Then, we developed an IL-2 recombinant protein to elucidate its potential role in triggering Th1 cell differentiation (Fig 6C). IL-2 stimulation elevated the transcription levels of T-bet, STAT1, and IL-12Rβ2 in spleen lymphocytes and eventually upregulated IFN-γ expression (Fig 6D). Mechanistically, the in vitro application or in vivo injection of IL-2 enhanced STAT5 phosphorylation (Fig 6E and 6F) and activated MAPK/ERK and mTOR pathways, as evidenced by the increased phosphorylation of ERK1/2 and AKT in spleen lymphocytes (Fig 6E and 6F). Because IL-2 initiated both STAT5 and mTORC1 signaling in tilapia, we further investigated the relationship between these pathways in regulating Th1 cell differentiation. Treatment of spleen lymphocytes with the STAT5 inhibitor AC-4-130 markedly decreased IL-2-induced STAT5 phosphorylation, which subsequently compromised the activation of mTORC1 signaling, as indicated by the diminished phosphorylation of mTOR, S6, and 4EBP1 (Fig 6G). Moreover, lymphocytes lacking STAT5 activity failed to elevate STAT1, T-bet, and IFN-γ upon IL-2 stimulation (Fig 6H). Remarkably, the blockade of mTORC1 signaling by rapamycin in turn reduced the IL-2-induced STAT5 phosphorylation (Fig 6I) and consequently impaired the expression of Th1 cell-related transcription factors and cytokines (Fig 6J). These observations indicate that downstream of IL-2 signaling, STAT5 and mTORC1 promotes Th1-cell differentiation via a bidirectional regulation and mutually reinforcing manner.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 6. IL-2 promotes Th1 cell differentiation of tilapia by activating STAT5 and mTORC1 signaling.

(A) Synteny analysis of tilapia IL-2 and other homology genes by comparative genomics. (B) Spleen leukocytes were stimulated with PHA, and mRNA levels of IL-2 were examined by qPCR, n = 6. (C) SDS-PAGE assay showed the recombination of tilapia IL-2 protein in E.coli. (D, E) Spleen leukocytes were treated with recombinant IL-2, and expression of indicated molecules were examined by qPCR (D, n = 6) or western blot (E) at indicated time points. (F) Tilapia individuals were i.p. injected with recombinant IL-2, the expression of indicated molecules were examined by western blot. (G, H) Spleen leukocytes were stimulated with recombinant IL-2 in the presence of STAT5 inhibitor AC-4-130 for 12 h. The expression levels of indicated molecules were examined by western blot (G) or qPCR (H, n = 4). (I, J) Tilapia was injected with mTORC1 inhibitor rapamycin for 2 days before spleen leukocytes were stimulated with recombinant IL-2 for 12 h. The expression levels of indicated molecules were examined by western blot (I) or qPCR (J, n = 6). These experiments were repeated for at least two independent times. *: p<0.05, **: p<0.01, ***: p<0.001, determined by a two-tailed Student’s t-test.

https://doi.org/10.1371/journal.ppat.1010913.g006

mTORC1 ensures Th1 cell development in tilapia during E. piscicida infection

The mTORC1 pathway helps naïve T cells to exit quiescence and regulates their differentiation into the Th1 lineage [47]. Our previous study revealing that mTORC1 plays a pivotal role in the T-cell response of tilapia [36], temps us to investigate whether and how this pathway contributes to Th1 cell development in fish. Spleen lymphocytes or CD3⁺CD4-1⁺ T cells exposed to TCR and costimulatory stimulation showed markedly higher S6 and 4E-BP1 phosphorylation (Fig 7A and 7B), indicating the potential involvement of mTORC1 signaling in the CD4⁺ T-cell response of tilapia. Then, the mTORC1 inhibitor rapamycin, which effectively blocked mTORC1 activity in activated T cells (Fig 7A) and CD4⁺ T cells (Fig 7B), was administered to tilapia during E. piscicida infection. Although mTORC1 inhibition did not affect the percentage of T-cell populations, it markedly reduced the absolute numbers of CD3⁺ T cells and CD3⁺CD4-1⁺ T cells compared with those in the untreated animals (Fig 7C–7E). Moreover, both the ability and capacity of CD3⁺CD4-1⁺ T cells to produce IFN-γ were markedly impaired in the absence of mTORC1 activity (Fig 7F and 7G). Consistent with the cellular evidence, the transcription level of IFN-γ in spleen lymphocytes was elevated upon E. piscicida infection, and this inducible expression was severely dampened by the rapamycin treatment (Fig 7H). Furthermore, lymphocytes lacking mTORC1 activity showed lower expression of IL-2 and IL-2R beta chain (Fig 7I), indicating an impairment of T-cell activation. Remarkably, mTORC1 deficiency hampered the proliferation of CD4-1⁺ T cells during E. piscicida infection, as indicated by the reduced BrdU incorporation in these cells (Fig 7J). Moreover, the addition of exogenous IL-2 slightly increased the proliferation of CD4-1⁺ T cells in both healthy and E. piscicida-infected tilapia, and this response was also found to be dependent on mTORC1 activity (Fig 7J). The induction of Th1 cell development by E. piscicida infection was supported by the upregulation of T-bet, STAT1, STAT4, and IL-12Rβ2 (Fig 7K), whereas mTORC1 deficiency markedly reduced the inducible expression of these crucial molecules (Fig 7K), suggesting that the transcriptional networks are regulated by the mTORC1 signaling pathway. Overall, our findings suggest that the mTORC1 signaling pathway regulates Th1 cell development in tilapia during E. piscicida infection by modulating cellular activation and proliferation and IFN-γ transcription.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 7. Tilapia requires mTORC1 signaling for Th1 cell development during E. piscicida infection.

(A, B) Spleen leukocytes from rapamycin treated or untreated tilapia were stimulated with CD3ε/CD28 mAbs for 12 h. (A) Western blot assay showed protein or phosphorylation levels of S6 and 4EBP1 in spleen leukocytes. (B) Overlaid histograms showed the phosphorylation levels of S6 in CD4-1⁺ T cells. (C-K) Tilapia individuals that infected with E. piscicida were treated with rapamycin or PBS. (C) Flow cytometry showed the spleen leukocytes stained with CD4-1 and CD3ε mAbs on 7 DPI. (D) Scatter plot figures showed the percentage and absolute numbers of T cells, n = 4. (E) Scatter plot figures showed the percentage and absolute numbers of CD4-1⁺ T cells, n = 4. (F, G) Spleen leukocytes harvested from the tilapia on 7 DPI were stimulated with P+I or not with the presence of Golgiplug in vitro for 4 h, and flow cytometry (F) and Scatter plot figures (G) showed the percentages of IFN-γ⁺CD4+ T cells, n = 4. (H, I, K) The mRNA levels of indicated molecules were examined by qPCR on 4 DPI, n = 6. (J) Tilapia individuals that infected with E. piscicida were treated with rapamycin or PBS in the presence or absence of recombinant IL-2. Tilapia was i.p. injected with BrdU 1 day before sacrifice, and flow cytometry showed the BrdU staining on gated CD4-1⁺ T cells on 5 DPI. These experiments were repeated for at least two independent times. *: p<0.05, **: p<0.01, ***: p<0.001, determined by a two-tailed Student’s t-test.

https://doi.org/10.1371/journal.ppat.1010913.g007

IFN-γ enhances Th1 cell polarization via STAT1/T-bet positive feedback

Downstream of IFN-γ signaling, two IFN-γ receptors, IFNγR1 and IFNγR2, were identified in Nile tilapia. Both receptor chains were high similar to their respective mouse homologues in tertiary structure (S7A Fig), and they clustered well with their counterparts from other vertebrates (S7B Fig), suggesting their evolutionarily conserved properties. The transcription levels of IFNγR1 and IFNγR2 were markedly elevated upon induction of T-cell activation by PHA or anti-CD3 plus anti-CD28 mAbs (Fig 8A and 8B), indicating their potential participation in the T-cell response of tilapia. Recombinant GST fusion proteins encoding the extracellular domains of IFNγR1 and IFNγR2, respectively, were prepared (Fig 8C), and the GST pull-down assay using GST resins confirmed that both IFNγR1 and IFNγR2 interacted with tilapia IFN-γ in vitro (Fig 8D). Moreover, the addition of recombinant IFN-γ enhanced STAT1 phosphorylation (Fig 8E) and induced the expression of the transcription factors T-bet, STAT1, and STAT4 (Fig 8F), which eventually elevated the mRNA level of IFN-γ in spleen lymphocytes (Fig 8G). In addition, the blockade of STAT1 with Fludarabine impaired the IFN-γ-induced T-bet expression (Fig 8H), suggesting that IFN-γ promotes Th1 cell commitment through the STAT1/T-bet axis. This conclusion was further substantiated by the results of RNAi experiments, in which STAT1 or T-bet knockdown markedly reduced the IFN-γ-induced T-bet and/or IFN-γ expression (Fig 8I and 8J) and consequently decreased the proportion of IFN-γ-producing CD3⁺CD4-1⁺ T cells (Fig 8K). Therefore, our findings support the notion that Th1 cells in tilapia utilize a sophisticated STAT1/T-bet positive feedback loop to promote their commitment.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 8. IFN-γ enhances tilapia Th1 cell commitment.

(A, B) Spleen leukocytes were stimulated with PHA or CD3ε/CD28 mAbs for 12 h, and mRNA levels of IFNγR1 and IFNγR2 were examined by qPCR, n = 6. (C) SDS-PAGE assay showed the recombination of tilapia IFNγR1 and IFNγR2 with GST-tag in E.coli. (D) GST pull-down assay showed the interaction of tilapia IFN-γ with IFNγR1 and IFNγR2. (E) Flow cytometry showed the phosphorylation level of STAT1 in lymphocytes that stimulated with recombinant IFN-γ. (F, G) Spleen leukocytes were stimulated with recombinant IFN-γ for 12 h, and mRNA levels of indicated molecules were examined by qPCR, n = 5. (H) Tilapia was injected with STAT1 inhibitor for 2 days before spleen leukocytes were stimulated with recombinant IFN-γ for 12 h. The expression levels of T-bet were examined by qPCR, n = 6. (I-K) Spleen leukocytes from tilapia that i.p. injected with T-bet-specific, STAT1-specific or control siRNA for 2 days were harvested and stimulated with recombinant IFN-γ for 12 h. Relative mRNA levels of STAT1, T-bet and IFN-γ (I, J, n = 5), and the percentage of IFN-γ⁺ cells in gated CD3⁺CD4-1⁺ T cells (K) were examined. These experiments were repeated for at least two independent times. *: p<0.05, **: p<0.01, ***: p<0.001, determined by a two-tailed Student’s t-test.

https://doi.org/10.1371/journal.ppat.1010913.g008

Ethics statement and animals

Nile tilapia larvae were purchased from an aquatic farm in Guangzhou, Guangdong Province, China, and maintained in a circulating water system with continuous aeration at 28°C until they reached to 10 cm, at the Biological Station of East China Normal University. Fish were fed twice one day with commercial pellets, and were randomly transferred to independent tanks before experiment. BALB/c mice were purchased and seeded in Minhang Laboratory Animal Center of East China Normal University. All animal care and experimental procedures were performed in accordance with the Guide for the Care and Use of Laboratory Animals of the Ministry of Science and Technology of China and were approved by the East China Normal University Experimental Animal Ethics Committee, with an approval number m20190108. All efforts were made to minimize the pain of animals.

Cell lines

HEK293T, BOSC23 and NIH/3T3 cells were purchased from the American Type Culture Collection (ATCC, Manassas, VA, USA). The mouse myeloma SP2/0 was kindly gifted by the Ocean University of Shanghai, China. HEK 293T, BOSC23, and NIH/3T3 cells were cultured at 37°C, 5% CO₂ in DMEM medium supplemented with 10% FBS and 1% penicillin/streptomycin. The SP2/0 cells were cultured at 37°C, 5% CO₂ in GIT medium.

Sequence, structure and phylogenic analysis

The cDNA and amino acid sequences of related genes were obtained from National Center for biotechnology information (NCBI). Chromosomal locations of IFN-γ genes from human, chicken, turtle, frog, zebrafish, and Nile tilapia were obtained from NCBI, and gene structure information of IFN-γ from other species was downloaded from NCBI. Multiple sequence alignments were generated by Clustalx and presented by The Sequence Manipulation Suite (http://www.bio-soft.net/sms/index.html). Phylogenetic tree was constructed in MEGA7 using neighbor-joining (NJ) method with 1000 bootstrap replications. The conserved domains were predicted using simple modular architecture research tool (http://smart.embl-heidelberg.de/), and the domain organization was displayed by DOG 2.0 software. The protein tertiary structure was predicted by SWISS-MODEL (https://swissmodel.expasy.org/interactive) and displayed by PyMOL software. Accession numbers for these sequences used were listed in S1 Table.

Recombinant protein preparation

The fragments encoding Nile tilapia IFN-γ and IL-2 (exclude signal peptide) were ligated into vector pET-28a (+), and fragments encoding the extracellular domains of IFNγR1 and IFNγR2 were ligated into vector pGEX-4T-3. The recombinant plasmids were transformed into transetta (DE3) chemically competent cells, and then induced by 0.5 mM IPTG at 37°C for 4 h. Recombinant protein IFN-γ and IL-2 with His-tag were purified by affinity chromatography using Ni-NTA resins. Subsequently, purified protein was re-folded by gradient dialysis at 4°C, and finally dialyzed in PBS thoroughly. Recombinant protein of IFNγR1 and IFNγR2 with GST-tag were purified by affinity chromatography using GST resins. The obtained proteins were concentrated with a microcon-10 kDa molecular mass centrifugal filters unit (Millipore), measured concentration with BCA Protein Assay Kit (Sangon Biotech), and stored at -80°C before use. Accession numbers for the sequences used were listed in S1 Table.

Retrovirus packaging and cell infection

The full length of coding cDNA sequence for Nile tilapia CD4-1 was cloned into MIGR1 plasmid. Subsequently, MIGR1-CD4-1 plasmid and helper plasmid were co-transfected into BOSC23 cells to produce retrovirus. The BOSC23 cells were seeded in 6 cm dish at 2×10⁶ cells overnight. After the medium was replaced by 3 mL fresh DMEM with 1‰ chloroquine, the cells were then co-transfected with 10 μg MIGR1-CD4-1 plasmid and 5 μg helper plasmid in the presence of 50 μL 2.5 M of CaCl₂ and 500 μL 2×HEBS. Eight hours after transfection, the culture medium was replaced by fresh DMEM, and the cell supernatant containing retrovirus were collected at 48 h after transfection. NIH/3T3 cells were seeded in 6 cm dish at 2×10⁵ cells overnight, and 120 μL of retrovirus was used to infect NIH/3T3 cells in the presence of 5 μg/mL polybrene. At 4 h after infection, 500 μL fresh DMEM medium was added to dilute the virus, and the cell medium was replaced by 4 mL fresh DMEM medium on the other day. The infected cells were collected at 48 h after infection, and were used for flow cytometry and mouse immunization.

Monoclonal antibodies (mAbs) preparation

Anti-tilapia CD3 and CD28 mAbs were previously developed by our lab. For IFN-γ mAb preparation, 100 μg purified IFN-γ protein was thoroughly emulsified with Freund’s complete adjuvant and immunized BALB/c mice by intraperitoneal (i.p.) injection. Two weeks later, another immunization with the same dose of antigen was i.p. injected using Freund’s incomplete adjuvant. For the last two immunizations, mice were immunized by intravenous injection of IFN-γ protein in PBS at 1-week intervals. To develop CD4-1 mAb, the NIH/3T3 cells expressing Nile tilapia CD4-1 were used to immunize BALB/c mice by i.p. injection for four times with the same intervals as above. Then, BALB/c mice were euthanized and splenocytes were fused with SP2/0 cells in the presence of polyethylene glycol (PEG). The fused cells were seeded in 96-well plates together with mouse thymocytes. After growing in the medium containing hypoxanthine, aminopterin and thymidine (HAT) over 10 days, the supernatants of hybridomas were screened for positive clones by ELISA and flow cytometry. The positive hybridomas were sub-cloned by limiting dilution method. The positive hybridomas were then i.p. injected into BALB/c mice to induce ascites, and mAbs were purified from ascites using protein G agarose (Invitrogen) and labelled with FITC or biotin.

ELISA assay

ELISA plate was coated with 5 μg recombinant IFN-γ at 4°C overnight. After blocked with PBS containing 1% BSA at 37°C for 1 h, 100 μL of hybridomas supernatant was added and incubated at 37°C for another 1 h. After washed with PBST for 3 times, the samples were stained with goat anti-mouse IgG AP-conjugate secondary antibody at 37°C for 1 h, followed by incubating with 0.1% (w/v) p-nitrophenyl phosphate (pNPP) for 30 min. Finally, the reaction was stopped by 2 M NaOH, and then the values of OD (405 nm) were measured with microplate reader.

Leukocyte isolation

The leukocytes were isolated from Nile tilapia spleen and head kidney according to our previous description [36]. Briefly, the spleen and head kidney were grinded in pre-cooled Leibovitz’s L-15 medium (Gibco), and filtered through nylon mesh. The cell suspension was added slowly onto a 52%/34% discontinuous density-gradient percoll (GE Healthcare). After centrifuged at 500 g, room temperature for 35 min with the lowest acceleration and deceleration, the cells between 34% and 52% percoll were collected for indicated assay.

Immunofluorescence assay

The spleen leukocytes were spun onto slide by Cytospin (Thermo) and fixed with methanol at room temperature for 5 min, and then blocked with PBS containing 1% BSA at 37°C for 1 h. After 3 times’ wash with PBST, the cells were incubated with mouse anti-tilapia CD4-1 mAb at 37°C for 1 h, and stained with PE-conjugated goat anti-mouse IgG secondary antibody for another 1 h. The samples were then washed, and followed by the staining with FITC-conjugated mouse anti-tilapia CD3ε mAb. Finally, the cells were mounted with antifade mounting medium containing Hoechst 33342. The images were acquired by fluorescence microscope and analyzed by photoshop CS6.

Leukocyte stimulation

The tilapia leukocytes that cultured at 28°C, 5% CO₂, in DMEM containing 10% FBS and 1% penicillin/streptomycin were stimulated with phorbol 12-myrustate 13-acetatae (PMA, 50 ng/mL) plus ionomycin (500 ng/mL) or 2 μg/mL of PHA or 2 μg/mL α-CD3ε mAb at 28°C for indicated time. Then, the cells were harvested for qPCR, flow cytometry and western blot assay. For α-CD3ε/α-CD28 stimulation, spleen leukocytes were activated with plate-bound α-CD3ε (2 μg/mL) plus soluble α-CD28 (2 μg/mL) at 28°C for indicated time, and the activated cells were harvested for assays.

Flow cytometry and cell sorting

To identify the CD3⁺ and CD4-1⁺ T-cell population, tilapia leukocytes were collected and incubated with CD4-1 mAb, followed by staining with PE-goat anti-mouse IgG secondary antibody. Subsequently, the cells were stained with FITC-conjugated CD3ε mAb and live/dead-Violet. Incubations of primary and secondary antibodies are on ice for 30 min, and two times of wash with FACS buffer (PBS with 2% FBS) were performed after each incubation. To identify IFN-γ producing cells, the leukocytes were stimulated with 50 ng/mL PMA plus 500 ng/mL ionomycin in the presence of 1 μg/mL of Golgiplug at 28°C for 4 h. After cell surface staining as above, cells were fixed with BD cytofix/cytoperm buffer on ice for 30 min, and washed twice with 1× BD perm/wash buffer. Subsequently, cells were stained with biotin-conjugated mouse anti-tilapia IFN-γ, and followed by APC-Streptavidin staining. For p-STAT1 staining, leukocytes were fixed with Foxp3 Fixation/Permeabilization working solution (eBioscience) on ice for 2 h, and then stained with Alexa Fluor 647 conjugated-p-STAT1 (Tyr701) Rabbit mAb (CST) on ice for 30 min. Stained cells were then washed twice with 1× permeabilization buffer. All the samples were analyzed by BD LSRFortessa flow cytometer, and data were analyzed using FlowJo software. For cell sorting, head kidney macrophages were gated according to our previous study [36], and head kidney leukocytes were stained with CD3 and CD4-1 as above. CD3⁺CD4-1⁺ T cells and macrophages were sorted using a BD FACSAria II flow cytometer.

Western blot analysis

The cells were lysed in NP40 lysis buffer containing 1mM PMSF, 1‰ protease inhibitor and 1% phosphorylase inhibitor on ice for 30 min. After centrifuged, the supernatant was acquired and separated by 12% SDS-PAGE, and then transferred to nitrocellulose membrane. The nitrocellulose membrane was blocked with 4% skim milk powder in PBST for 1 h, and incubated with anti-tilapia IFN-γ mAb or 1:1000 diluted primary antibody against S6, 4EBP1, phospho-S6 (Ser240/244), phospho-4EBP1 (Thr37/46), phosphor-STAT5 (Tyr694), phosphor-Erk1/2 (Thr202/Tyr204), phosphor-Akt (Thr308), phosphor-Akt (Ser473), and β-actin from Cell Signaling Technology at 4°C overnight, and further incubated with goat anti-rabbit or mouse IgG H&L 800 or 680 (Abcam). To detect the overexpressed IFN-γ in 293T cells, the nitrocellulose membrane was incubated with anti-tilapia IFN-γ mAb, followed by Ap-conjugated goat anti-mouse IgG secondary antibody (Southern Biotech), and visualized with freshly prepared substrate solution (100 mM NaCl, 100 mM Tris, and 5 mM MgCl₂; pH 9.5) containing nitroblue tetrazolium (Sigma) and 5-bromo-4-chloro-3-indolylphosphate (Sigma). Images were acquired by Odyssey CLx Image Studio.

Edwardsiella piscicida infection

Edwardsiella piscicida was kindly gifted by the East China University of Science and Technology, and was cultured in Tryptone Yeast extract Broth (TYB) at 37°C overnight. After centrifuged at 6000 g for 3 min, the bacteria were collected and washed with PBS, and then resuspended in PBS to the OD₆₀₀ value of 1.0. Nile tilapia were i.p. injected of 100 μL E. piscicida on the concentration of 2×10⁷ CFU/mL. The control group were injected with the same volume of PBS. Spleen leukocytes were isolated for assays on the indicated days.

Depletion of CD4-1⁺ cells

Depletion of CD4-1 cells were performed according to a previous report with modification [78]. Tilapia individuals were i.p. injected with 20 μg of mouse anti-tilapia CD4-1 or mouse IgG1 (Catalog #401402, BioLegend) as isotype control antibody. After 24 h, the tilapia was i.p. injected with 50 μL tilapia anti-mouse IgG1 serum or tilapia control serum, respectively. Nine days after mAbs injection, spleen leukocytes were obtained, and the CD3ε⁺CD4-1⁺ T cells were stained and examined by flow cytometry. The nondepleted and CD4-1-depleted fish were then infected with E. piscicida, and animals were sacrificed on indicated days for assay.

Inhibitor treatment

STAT1 inhibitor Fludarabine, STAT5 inhibitor Ac-4-130, mTORC1 inhibitor rapamycin and Brefeldin A were purchased from MedChemExpress. Nile tilapia was i.p. injected with 5 mg/kg Fludarabine or 1 mg/kg rapamycin every two days during E. piscicida infection. Nile tilapia was i.p. injected with 13 mg/kg Brefeldin A to block cytokine secreting at 6 h before sacrifice, and the spleen leukocytes were isolated for IFN-γ producing by flow cytometry or western blot on day 5 or 8 post infection. For the STAT5 inhibition, the isolated spleen leukocytes were treated with 5 mM Ac-4-130, and the cell were stimulated with or without recombinant IL-2 for 12 h.

siRNA interference

Three pairs of tilapia T-bet or STAT1 specific siRNA, and one pair of negative control siRNA were designed by and ordered from GenePharma company. The primer sequences were listed in S2 Table. Tilapia were i.p. injected with 3 μg specific siRNA or control siRNA in 100 μl 1:500 diluted Lipo6000. Two days later, the spleen leukocytes were isolated and stimulated with 2 μg/ml plate-bound α-CD3ε for 12 h, and the activated cells were harvested for assays. For in vivo experiments, tilapia infected with E. piscicida were i.p. injected with 3 μg T-bet specific siRNA or STAT1 specific siRNA on day 2 and 4 post infection. The spleen leukocytes were isolated on indicated days for assays.

Cytokines treatment

For in vitro treatment, the spleen leukocytes isolated from Nile tilapia were cultured in DMEM (10% FBS) and stimulated with 2 μg/ml recombinant IFN-γ or IL-2 for indicated times. Stimulated cells were collected for quantitative RT-PCR (qPCR) analysis, flow cytometry and western blot. For IFN-γ in vivo treatment, Nile tilapia were i.p. injected with 1 mg/kg recombinant IFN-γ each day during E. piscicida infection. At 5 days after infection, the bacterial titer in spleen was examined by coating the TYB solid medium and counting the number of colonies. For IL-2 in vivo treatment, Nile tilapia were i.p. injected with 1 mg/kg recombinant IL-2 for indicated time, and spleen leukocytes were isolated for western blot assay.

Transfection and Dual-luciferase assay

Nile tilapia IFN-γ promoter (from -2000 to +0 bp) was cloned and ligated into pGL3 vector. The coding regions of T-bet, STAT1 and STAT4 were cloned into pcDNA3.1 vector. For dual-luciferase assay, 10⁴ of HEK 293T cells were seeded in 24-well plates overnight, and co-transfected with TF expression plasmid (pcDNA3.1-T-bet, STAT1 or STAT4), reporter genes (pGL3-IFN-γ promoter), and pRL-TK plasmid in 100 μL opti-DMEM medium containing 1 μL lipofectamine 2000. Six hours after transfection, the medium was replaced by fresh DMEM, and the cells were collected at 48 h after transfection for assay. The Luc activities were measured according to the manual of Dual-Luciferase Reporter Assay System (Promega). Cells were harvested and lysed by 1×PLB, and Luc activities were measured using the luminometer after mixing LAR II, and mixed Stop & Glo Reagent, respectively.

Pull-down assay

The Recombinant protein GST tag-IFNγR1 and GST tag-IFNγR2 were incubated with anti-GST-Agarose beads at 4°C overnight, respectively. The beads were then washed 4 times with resin buffer, and then incubated with the recombinant His-tag IFN-γ at 4°C for 4 h. The beads were precipitated, washed, and resuspended in 5×SDS sample buffer, then boiled and separated by 12% SDS-PAGE, and stained with Coomassie brilliant blue. Images were acquired by Odyssey CLx Image Studio.

Quantitative RT-PCR (qPCR)

Total RNA was extracted with TRIzol reagent (Invitrogen) according to the manufacturer’s instructions. After treated with gDNA pure to remove genomic DNA, the total RNA was used as template to synthesize the first strand of cDNA using 2×NovoScript Plus 1st Strand cDNA Synthesis SuperMix (Novoprotein). Then, the qPCR was performed with NovoStart SYBR qPCR SuperMix Plus (Novoprotein) on a QuantStudio 5 Real-Time PCR Instrument (Applied Biosystems). β-actin was used as internal reference, and the expression of target genes were analyzed by 2^-△△Ct method. The gene information and primer sequences were listed in S2 Table.

BrdU incorporation

Nile tilapia was i.p. injected with 0.75 mg BrdU (Sigma) in 200 μL PBS, at 16 hours before the spleen leukocytes were isolated for assay. After the cells were stained with surface CD3ε and CD4-1 as above, they were fixed with BD Cytofix/Cytoperm Buffer on ice for 30 min, and then washed twice with BD Perm/Wash Buffer. After that, the cells were treated with BD Cytoperm Plus Buffer for 10 min and BD Cytofix/Cytoperm Buffer for 5 min on ice, respectively, and then digested with 300 μg/mL DNase at 37°C for 1 h. Samples were subsequently stained with FITC-anti-BrdU antibody (BD) at room temperature for 20 min and analyzed with flow cytometry.

Phagocytosis assay

FITC-labelled E. piscicida was incubated with leukocyte suspension, sorted macrophages, or sorted macrophage plus sorted CD4-1⁺ T cells containing 5 μg of recombinant IFN-γ or not at 28°C for 6 or 12 h. After that, 0.2% trypan blue was added to quench the fluorescence from attached bacteria. The macrophage population of Nile tilapia was gated by flow cytometry according to our previous study [36], and their phagocytosis ability was then examined.

IFN-γ mAb blocking

The head kidney leukocytes, or sorted head kidney macrophage plus CD4-1⁺ T cells were activated with 2 μg/mL plate-bound α-CD3ε, and 10 μg/ml IFN-γ mAb was added to block the IFN-γ. FITC-labelled E. piscicida were then added into the cells and incubated for 12 h. The macrophage population of Nile tilapia was gated and to measure their phagocytosis ability.

Statistical analysis

Prism version 8.0 software (GraphPad Software Inc., San Diego, CA, USA) was used for statistical analysis. The statistical significance of results was analyzed using two-tailed, unpaired, Student’s t-test. Animal survival data were analyzed by log-rank analysis. The significance was indicated by *, p < 0.05, **, p < 0.01, ***, p < 0.001.