ABSTRACT
Bacteria utilize type III secretion systems (T3SS) to deliver effectors directly into host cells. Hence, it is very important to identify the functions of bacterial (T3SS) effectors to understand host-pathogen interactions. Edwardsiella piscicida encodes a functional T3SS effector, EseK, which can be translocated into host cells and affect bacterial loads. Here, it was demonstrated that an eseK mutant (the ΔeseK mutant) significantly increased the phosphorylation levels of p38α, c-Jun NH2-terminal kinases (JNK), and extracellular signal-regulated protein kinases 1/2 (ERK1/2) in HeLa cells. Overexpression of EseK directly inhibited mitogen-activated protein kinase (MAPK) signaling pathways in HEK293T cells. The ΔeseK mutant consistently promoted the phosphorylation of MAPKs in zebrafish larva infection models. Further, it was shown that the ΔeseK mutant increased the expression of tumor necrosis factor alpha (TNF-α) in an MAPK-dependent manner. Importantly, the EseK-mediated inhibition of MAPKs in vivo attenuated bacterial clearance in larvae. Taken together, this work reveals that the E. piscicida T3SS effector EseK promotes bacterial infection by inhibiting MAPK activation, which provides insights into the molecular pathogenesis of E. piscicida in fish.
INTRODUCTION
A key feature of the virulence of many Gram-negative bacterial pathogens is the ability to deliver effector proteins into their cognate host cells via a dedicated type III secretion system (T3SS) (1). The T3SS is an essential virulence factor, and bacteria that possess a T3SS cause a wide range of diseases in plants, animals, and humans, highlighting the central importance of the effector proteins in disease (2, 3). These bacterial effectors are usually potent modulators of host cellular processes essential for counteracting bacterial infection. In general, T3SS effectors display subtle functions inside host cells, often causing restrained alterations in host cell physiology and usually being involved in ubiquitination (4), phosphorylation (5), reorganizing the actin cytoskeleton (6), disrupting the microtubule to promote pathogen internalization (7), and regulating the NF-κB and mitogen-activated protein kinase (MAPK) signaling pathways to modulate the host innate immune response (8).
The MAPK signaling pathways consist of a series of at least three kinases which convert extracellular signals into cellular responses (9): a MAPK kinase kinase (MAP3K) that activates a MAPK kinase (MAP2K), which in turn activates the MAPK by dual phosphorylation of the activation motif (10, 11). There are three well-defined MAPK signaling pathways: extracellular signal-regulated protein kinases 1/2 (ERK1/2), the p38 kinases, and c-Jun NH2-terminal kinases 1/2 (JNK1/2) (12). MAPK signaling pathways play important roles in activating host innate immune responses, such as inducing the expressions of tumor necrosis factor alpha (TNF-α), gamma interferon (IFN-γ), and interleukin-6 (IL-6) (13), all of which participate in host inflammatory responses (10). To date, several T3SS effectors of diverse pathogenic bacteria have been shown to target the MAPK signaling pathways. For example, YopJ (Yersinia spp.) targets the pheromone response MAPK signaling pathways in the Saccharomyces cerevisiae yeast model system. The T3SS effector protein of Vibrio cholerae, VopE, interferes with the functioning of the cell wall integrity (CWI)-MAPK pathways (14). The Shigella flexneri T3SS effector OspB is required for the full activation of ERK1/2 and p38 MAPKs and the cytosolic phospholipase A2 (cPLA2) (15).
Edwardsiella piscicida, previously named Edwardsiella tarda, has a broad host range, infecting not only freshwater and marine life (including fish, amphibians, reptiles, birds, and mammals) but also humans (16–18). E. piscicida was reported to possess a functional T3SS and a functional type VI secretion system (T6SS), which play an essential role in promoting bacterial internalization and replication intracellularly (19, 20); however, although the function of the effectors is not conserved, the bacteria utilize the multicomplexes of their effector weapons to target innate immune signaling and promote their infection, and the functions of these effectors still need to be further clarified. For example, EseG and EseJ were previously identified to be the T3SS effectors regulating microtubule structures (7, 21) and bacterial adhesion (22), respectively. EseH, a T3SS effector, was found to inhibit MAPKs and affect the bacterial burden via its phosphothreonine lyase activity (8). EvpP, a T6SS effector, was found to inhibit the NLRP3 inflammasome and limit bacterial infection (23). Very recently, we identified a new E. piscicida T3SS effector, EseK, which can be translocated into host cells with the aid of two chaperones, EscH and EscS. We found that EseK is required for bacterial virulence in zebrafish with unknown mechanisms (24).
In this study, we investigated the functions of a T3SS effector, EseK, in facilitating bacterial virulence using both mammalian and zebrafish larva infection models. Our results demonstrated that EseK could inhibit the phosphorylation of p38α, JNK, and ERK1/2 in both mammalian and ZF4 cells in vitro and zebrafish larvae in vivo. Furthermore, EseK significantly inhibited the expression of TNF-α in an MAPK-dependent manner. In vivo infection experiments revealed that EseK promoted E. piscicida loads in zebrafish larvae by inhibiting MAPK activation. Our results provide evidence that E. piscicida utilizes the T3SS effector EseK to attenuate MAPK-mediated immune responses and promote bacterial infection in vivo.
RESULTS
EseK inhibited the phosphorylation of MAPKs in mammalian cells.We have previously demonstrated that the E. piscicida protein EseK is injected into host cells in a T3SS-dependent manner (24). However, the functions of EseK during infection remain unknown. Previous studies showed that E. piscicida infection activates MAPK signaling pathways, which mediate antibacterial immune responses in adult zebrafish infection in vivo (8). Further, EseK contains a conserved catalytic triad (C101, H178, and D193), and its predicted protein secondary structure shares a certain degree of homology with that of the Shigella OspI effector, which belongs to the cysteine protease-like family (see Fig. S1A in the supplemental material). One of the functions of the cysteine protease-like family is to regulate MAPK signaling pathways (25). Thus, we assessed whether EseK has an inhibitory effect on MAPK signaling pathways during bacterial infection. The wild-type strain EIB202 and the eseK deletion (ΔeseK) mutant were used to infect HeLa cells, and the phosphorylation levels of p38α, JNK, and ERK1/2 were significantly increased in ΔeseK mutant-infected cells compared with that in EIB202-infected cells at 3 h postinfection. Simultaneously, complementation with pUEseK counteracted the activation of MAPKs in ΔeseK mutant-infected cells (Fig. 1A and S2). To examine whether EseK alone inhibits MAPK activation, we expressed hemagglutinin (HA)-tagged EseK in HEK293T cells. Compared with the phosphorylation levels in the controls, the decreased phosphorylation levels of the MAPKs induced by TNF-α were detected in EseK-HA-expressing cells at the 24-h and 16-h time points and the 36-h and 16-h time points (Fig. 1B), suggesting that EseK could inhibit TNF-α-activated MAPK signaling pathways. Currently, four MAP2Ks (MEK1/2, MKK3/6, MKK4, and MKK7) are known to phosphorylate and activate MAPKs (ERK1/2, p38, and JNK proteins) (10). To clarify whether the MAP2Ks participate in EseK-related regulation, we explored the effects of EseK on the phosphorylation of MAP2Ks induced by TNF-α in HEK293T cells. EseK did not affect the activation of MEK1/2, MKK3/6, MKK4, or MKK7 (Fig. 1C). Taken together, these data demonstrate that the E. piscicida T3SS effector EseK targets MAPKs and directly inhibits their phosphorylation in mammalian cells.
EseK inhibited the activation of MAPKs in mammalian cells. (A) HeLa cells were infected with EIB202, the ΔeseK mutant, or the ΔeseK/pUEseK strain for 3 h. (B) EseK prevented the activation of MAPKs upon TNF-α stimulation in HEK293T cells. HEK293T cells were transfected with pCEseK, pCDH, or no plasmid for 24 h or 36 h, and then the cells were serum starved for 16 h before TNF-α stimulation. (C) EseK prevented the activation of MAPKs upon TNF-α stimulation directly in HEK293T cells. HEK293T cells were transfected with pCEseK, pCDH, or no plasmid for 24 h, and then the cells were serum starved for 16 h before TNF-α stimulation. The numbers beneath the bands indicate the relative intensity of p-p38α, p-JNK, or p-ERK1/2, which were calculated by normalizing ΔeseK (panel A) or Blank (panels B and C) to a relative intensity of 100. Data are representative of those from three independent experiments.
EseK inhibited the phosphorylation of MAPKs in zebrafish models.Our previous experiments confirmed that EseK inhibits the MAPK signaling pathways in mammalian cells (Fig. 1A). Because E. piscicida is a common fish pathogen that causes systemic infections in a wide variety of marine and freshwater fish around the world (23), we explored EseK-mediated MAPK signaling pathway regulation in fish models. First, zebrafish fibroblasts (ZF4 cells) were infected with EIB202, the ΔeseK mutant, and the ΔeseK/pUEseK strain. Immunoblotting demonstrated that compared with EIB202, the ΔeseK mutant induced significantly increased phosphorylation of p38α, JNK, and ERK1/2 in ZF4 cells, which was impaired by complementation with pUEseK (Fig. 2A and S3A). Further, to investigate the role of EseK in regulating MAPKs in vivo, healthy zebrafish embryos (2 to 3 days postfertilization) were infected with E. piscicida by microinjection as described previously (26). The wild-type bacteria, the ΔeseK mutant, and the eseK deletion strain complemented with wild-type EseK were injected into the yolk sac of zebrafish larvae at 50 CFU per fish. At 18 h after infection, 5 zebrafish larvae were sampled from each group and probed with antibodies. Notably, on the basis of the comparable β-actin levels, significantly larger amounts of phosphorylated p38α (p-p38α), phosphorylated JNK (p-JNK), and phosphorylated ERK1/2 (p-ERK1/2) were detected in ΔeseK mutant-infected zebrafish larvae than in wild-type-infected zebrafish larvae (Fig. 2B and S3B). To further verify this phenomenon, the larvae were pretreated with specific MAPK inhibitors (SB202190, SP600125, and U0126-ethanol [EtOH]) and then microinjected with the ΔeseK mutant. It was found that SB202190, SP600125, and U0126-EtOH inhibited the p38α, JNK, and ERK1/2 phosphorylation induced by ΔeseK mutant infection, respectively (Fig. 2C and S3C). Together, these findings suggest that the T3SS effector EseK inhibits the phosphorylation of MAPKs but not that of the MAP2Ks in both ZF4 cells and zebrafish larvae in vivo.
EseK inhibited the phosphorylation of MAPKs in ZF4 cells and zebrafish larvae. (A) ZF4 cells were infected with EIB202, the ΔeseK mutant, or the ΔeseK/pUEseK strain for 3 h. (B) Zebrafish larvae were injected with EIB202, ΔeseK mutant, or the ΔeseK/pUEseK strain at 50 CFU per fish for 18 h. (C) Zebrafish larvae were pretreated with a p38α inhibitor (SB202190), JNK inhibitor (SP600125) or ERK1/2 inhibitor (U0126-EtOH) for 2 h and then were injected with the ΔeseK mutant at 50 CFU per fish for 18 h. The numbers beneath the bands indicate the relative intensity of p-p38α, p-JNK, or p-ERK1/2, which were calculated by normalizing ΔeseK to a relative intensity of 100. Data are representative of those from three independent experiments.
EseK inhibited the expression of TNF-α in an MAPK-dependent manner.Because MAPK signaling pathways are capable of regulating the transcriptional activation of a wide array of inflammatory cytokines and contribute to host innate immune defenses (11), we investigated whether EseK affects the expressions of MAPK-regulated inflammatory cytokines. Compared with EIB202, the ΔeseK mutant significantly enhanced the transcription of TNF-α in HeLa cells (Fig. 3A) and ZF4 cells (Fig. 3B) at 3 h postinfection, which was counteracted by complementation with pUEseK. In contrast, no significant differences in IL-6 and IFN-γ transcript levels were observed between EIB202- and ΔeseK mutant-infected cells (Fig. S4). To confirm the regulation of TNF-α transcription in vivo, we compared the levels of TNF-α transcription induced by the wild-type strain and the ΔeseK mutant in zebrafish larvae and found that the ΔeseK mutant significantly enhanced TNF-α transcription (Fig. 3C), verifying the importance of EseK in regulating the expression of TNF-α in vivo. Furthermore, zebrafish larva infection also showed no significant differences in IL-6 and IFN-γ transcript levels between EIB202- and ΔeseK mutant-injected fish (Fig. S4). In addition, to assess whether EseK limited TNF-α transcription by regulating MAPK activity, we pretreated zebrafish larvae with 50 μM a p38α inhibitor (SB202190), JNK inhibitor (SP600125), or ERK1/2 inhibitor (U0126-EtOH) for 2 h before microinjection. The RNA in infected zebrafish larvae was extracted at 18 h postinfection and evaluated by quantitative real-time PCR (RT-PCR). The MAPK inhibitor impaired ΔeseK mutant-enhanced TNF-α transcription in zebrafish larvae compared to the controls (Fig. 3D). Taken together, these findings suggest that E. piscicida T3SS effector EseK-mediated MAPK inhibition significantly decreased the transcription of TNF-α following bacterial infection.
The expression of TNF-α was inhibited by EseK in a MAPK-dependent manner. (A, B) HeLa cells (A) and ZF4 cells (B) were infected with EIB202, the ΔeseK mutant, or the ΔeseK/pUEseK strain for 3 h. (C) Zebrafish larvae were injected with EIB202, the ΔeseK mutant, or the ΔeseK/pUEseK strain at 50 CFU per fish for 18 h. (D) Zebrafish larvae were injected with the ΔeseK mutant at 50 CFU per fish in the presence or absence of competitive MAPK inhibitors. Then, the infected HeLa cells, infected ZF4 cells, or injected fish were lysed, RNA was extracted from the cells or fish, and the expression of TNF-α was determined using real-time PCR. PBS-treated cells or fish were used as a control. The expression of the gene was normalized to that of the β-actin transcript and expressed as the fold change relative to the expression seen in PBS-infected cells. The data are presented as the mean ± SD for triplicate samples per experimental condition (n = 3 per experimental condition with 3 technical replicates), and the results are representative of those from three separate experiments. Statistical analysis was performed using one-way ANOVA. *, P < 0.05 compared to ΔeseK mutant-infected cells; **, P < 0.01 compared to ΔeseK mutant-infected cells; ns, no significant difference compared to ΔeseK mutant-infected cells.
EseK inhibited MAPKs and promoted E. piscicida colonization in vivo.EseK is an E. piscicida T3SS effector that significantly affects bacterial colonization in adult zebrafish (24). Here, we infected zebrafish larvae with EIB202, the ΔeseK mutant, and the ΔeseK/pUEseK strain via microinjection at a dose of 50 CFU/larva. The ΔeseK mutant showed a trend of bacterial colonization in whole zebrafish larvae similar to that of the wild-type and complemented strains, and the difference was not statistically significant (Fig. 4A), confirming that EseK has a role promoting bacterial colonization in vivo. To assess whether EseK promoted bacterial infection by regulating MAPK activity, we pretreated the zebrafish larvae with the specific inhibitors of MAPK signaling, SB202190, SP600125, and U0126-EtOH, for 2 h before microinjection. By comparing the bacterial loads of the ΔeseK mutant in zebrafish larvae in the presence or absence of MAPK inhibitors, we found that MAPK inhibitor administration significantly decreased the loads of the ΔeseK mutant in zebrafish larvae at 36 h and 48 h after infection (Fig. 4B). Taken together, these results provide evidence that E. piscicida T3SS effector EseK-mediated MAPK inhibition promotes E. piscicida infection in vivo.
EseK increased the E. piscicida loads in zebrafish larvae in a MAPK-dependent manner. The bacterial loads in zebrafish larvae were measured at the indicated time points after injection with EIB202, the ΔeseK mutant, or the ΔeseK/pUEseK strain (A) and after injection of the ΔeseK mutant in the presence or absence of competitive MAPK inhibitors (B). Five zebrafish were used per group per time point. Fish were injected with 50 CFU per fish and then sacrificed at 6 h, 12 h, 24 h, 36 h, and 48 h after injection. The data are presented as the combined means from three separate experiments. Statistical analysis was performed using one-way ANOVA. *, P < 0.05 compared to ΔeseK mutant-injected larvae; **, P < 0.01 compared to ΔeseK mutant-injected larvae; ***, P < 0.001 compared to ΔeseK mutant-injected larvae; ns, no significant difference compared to ΔeseK mutant-injected larvae. hpi, hours postinfection.
DISCUSSION
The most extensively studied group of vertebrate MAPKs to date is p38, JNK, and ERK1/2, which are activated by dual phosphorylation at neighboring threonine and tyrosine residues in the activation loop (27). Many pathogens target host intracellular MAPK signaling pathways to inhibit immune responses (28). Shigella flexneri dephosphorylates the MAPKs p38 and ERK in infected epithelial cells and thereby dampens innate immunity by its type III secreted effector OspF (29). Leishmania spp. target the MAPK signaling pathway, which is responsible for regulating the production of cytokines in macrophages and dendritic cells and alters the formation of a Th1 response to Th2, leading to parasite prevalence in the host (30, 31). Bacillus anthracis utilizes a toxic protease to inhibit the activation of the p38 MAPK in macrophages, resulting in the induction of macrophage death, which promotes inflammasome activation and IL-1β production, thereby promoting antimicrobial immunity (32). In addition, MAPK signal transduction cascades are highly conserved in species ranging from yeast to mammals (33). Recent studies showed that MAPK signaling pathways are regulated in zebrafish. The JNK molecule, but not the ERK or p38 MAPK molecule, is a critical factor mediating primitive macrophage migration in vivo in response to acute injury (33). MAPK signaling is upstream of the intact microtubule organization in zebrafish (34) and is critical for normal macrophage chemotaxis toward laser-induced injury on the yolk sac (35). These evidences support the high degree of evolutionary conservation of MAPK signaling pathways in innate immunity between humans and zebrafish (36). As we all know, adult zebrafish are good for studying bacterial colonization from the infection site to different immune-related organs, for example, the kidney, spleen, and liver (24). However, during the early stages of development, zebrafish larvae possess only innate immunity (37); thus, we believe that they are a better model for studying the interaction between bacterial effectors and host innate immune responses. Considering the advantages of zebrafish larvae, in this study, we used them as the infection model and revealed that the E. piscicida T3SS effector EseK inhibits MAPK signaling pathways and promotes bacterial infection, which provides direct evidence that MAPK signaling pathways play important roles in fish antibacterial immune responses and therefore are an essential target disturbed by invading fish pathogens for immune evasion.
The immune response is one of several key functions regulated by MAPKs, with the output being the production of immunomodulatory cytokines, such as TNF-α, interleukin-1 (IL-1), IL-6, and IFN-γ, which are induced by the activation of the p38 MAPK, JNK, and ERK pathways (11, 38, 39). Such inflammatory cytokines mediate the immune response by killing intracellular pathogens and facilitating the clearance of invaders at the early infection stage in host cells (11, 40–43). Earlier studies showed that the cytokine TNF-α is crucial for the clearance of bacteria. Tumor necrosis factor receptor (TNFR)-mediated mechanisms are essential for the clearance of Yersinia infection (44). Multiple logistic regressions, adjusting for age, sex, and HIV infection status, showed a highly significant correlation between bacterial loads and TNF-α levels in sputum (43). Overexpression of signaling lymphocyte activation molecule 1 (SLAMF1) significantly increases the production of the inflammatory factor TNF-α by regulating the NF-κB signaling pathway, which contributes to bacterial clearance in infected RAW 264.7 cells and in the lungs of infected mice (45). The plant lectin ConBr induces the expression of TNF-α, which reduces the bacterial burden in macrophages infected with Salmonella enterica serovar Typhimurium (46). The lectin ScLL from Synadenium carinatum latex, used as an adjuvant with Leishmania amazonensis soluble antigens in murine models of vaccination, reduces the parasite load within macrophages, which is accompanied by TNF-α expression (47). In this study, we examined the levels of transcription of TNF-α, IL-6, and IFN-γ in zebrafish larvae infected with E. piscicida and found that the ΔeseK mutant promoted TNF-α transcription by upregulating the phosphorylation of MAPKs. Considering that MAPK activation decreased the bacterial burden in zebrafish larvae, whether zebrafish TNF-α participates in limiting E. piscicida infection in vivo needs further investigations.
In our previous work (24), we reported that EseK is a newly discovered T3SS effector protein in E. piscicida. We also found that EseK is injected into host cells with the aid of the chaperones EscH and EscS and is specifically associated with cell membrane fractions. Sequence analysis revealed that although it shows a low similarity in the primary protein sequence to the Shigella OspI effector, EseK shares a certain homology in the protein secondary structure to the Shigella OspI effector and also contains a conserved catalytic triad (C101, H178, and D193) (see Fig. S1A in the supplemental material). OspI shares structural homology with a cysteine protease-like family which contains many T3SS effectors to express virulence in host cells, and one of the features of the cysteine protease-like family is the presence of the conserved catalytic triad (C101, H178, and D193) (48). In this study, we found that EseK inhibits the phosphorylation of MAPKs rather than the phosphorylation of MAP2Ks, suggesting that the EseK might directly regulate MAPKs. However, the conserved amino acids (C101 and D193) have no role in the regulation of MAPKs (Fig. S1B). Because the protein EseK (H178A) could not be expressed in HEK293T cells in our study, we did not show data for the protein EseK (H178A). Thus, how the membrane-associated T3SS effector EseK regulates intracellular MAPK processes during bacterial infection still needs to be further clarified. Although our study revealed that the function of EseK is to inhibit the activation of MAPKs, the underlying mechanisms by which EseK regulates MAPK signaling pathways remain unknown, and subsequent studies could be focused on how EseK inhibits MAPK activation, such as the interaction between EseK and MAPKs.
MATERIALS AND METHODS
Cells and cell culture.HeLa cells (ATCC CCL-2) and HEK293T cells (ATCC CRL-11268) were cultured in Dulbecco minimal essential medium (DMEM; Gibco) supplemented with 10% fetal bovine serum (FBS; Gibco) in a 5% CO2 atmosphere. Zebrafish fibroblasts (ZF4 cells; ATCC CRL-2050) were cultured in DMEM–Ham's F-12 medium (Gibco) supplemented with 10% fetal bovine serum (FBS; Gibco) in a 5% CO2 atmosphere.
Bacterial strains, media, and culture conditions.For in vitro infection assays, the bacterial strain E. piscicida EIB202 (CCTCC number M208068), which was isolated from infected farmed turbots, and its derived strains were grown at 30°C in tryptic yeast broth (TYB). Escherichia coli strain DH5α was cultured at 37°C in Luria-Bertani broth (LB). The bacterial strains and plasmids used in this study are described in Table 1.
Edwardsiella piscicida strains and constructs used in this study
Construction of plasmids.For construction of the plasmids pUT-t-P0456-EseK-HA and pCDH-EseK-HA, the eseK gene was amplified from EIB202 and was cloned into pUT-t-P0456 and pCDH, and then these plasmids were transformed into Escherichia coli DH5α and then electrotransformed into EIB202 and its derived strains. The primers used in this study are described in Table 2.
Primers used in RT-PCR
Antibodies and reagents.Mouse polyclonal anti-ERK1/2 antibody and rabbit polyclonal anti-p38α, anti-JNK, anti-p-p38α, anti-p-JNK, anti-p-ERK1/2, anti-phosphorylated MEK1/2 (anti-p-MEK1/2), anti-phosphorylated MKK3/6 (anti-p-MKK3/6), anti-phosphorylated MKK4 (anti-p-MKK4), and anti-phosphorylated MKK7 (anti-p-MKK7) antibodies were used at a dilution of 1:1,000 (Cell Signaling Technology). The rabbit polyclonal anti-HA antibody (1:5,000; catalog number R120921; HuaAn Biotechnology) and the mouse monoclonal antibody anti-β-actin (1:5,000; catalog number M1210-2; HuaAn Biotechnology) were used as primary antibodies for immunoblotting. Goat anti-mouse IgG (catalog number A0216) and goat anti-rabbit IgG (catalog number A0208) secondary antibodies (at a dilution of 1:1.000) were purchased from Beyotime Biotechnology. The MAPK signaling pathway inhibitors SB202190 (catalog number S1077; Selleck), SP600125 (catalog number S1460; Selleck), and U0126-EtOH (catalog number S1102; Selleck) were purchased from Selleck. These compounds were prepared in dimethyl sulfoxide (DMSO) to make stock concentrations of 20 mM and stored at −20°C.
Cell infection.HeLa cells and ZF4 cells were used for the cell infection assays, in which cells were seeded in 24-well plates at 105 per well. E. piscicida was grown for 14 h in TYB at 30°C with shaking and then diluted 2:100 in TYB without shaking at 30°C until the optical density at 600 nm (OD600) reached 0.8. Harvested bacteria in fresh DMEM suspensions were added to HeLa cells or ZF4 cells at a multiplicity of infection (MOI) of 100:1. The plates were then centrifuged at 600 × g for 10 min at 30°C or 28°C (HeLa cells at 30°C and ZF4 cells at 28°C). At 3 h after incubation at 35°C or 28°C (HeLa cells at 35°C and ZF4 cells at 28°C) under 5% CO2, the supernatants were collected and the cells were lysed with cell lysis buffer on ice for 15 min. Then we added methanol and trichloromethane into the supernatants of the lysed cells and collected the precipitates by centrifugation at 13,000 × g for 5.5 min at 4°C. The precipitates were washed with methanol, followed by centrifugation. The resulting precipitates were resolved on 5× SDS sample buffer containing 5% β-mercaptoethanol and stored at −20°C before Western blotting.
Cell transfection.HEK293T cells were seeded in 24-well plates at 5 × 104 per well. After 24 h for culturing, HEK293T cells were transfected with a plasmid by the standard calcium phosphate method (double-distilled H2O, CaCl2, Hanks balanced salt solution [NaCl, KCl Na2HPO4, glucose, HEPES]) for 24 h or 36 h, and then the medium was changed to serum-free medium for 16 h (serum-free medium for 16 h can largely decrease the serum-induced activation of the MAPK signaling pathway) before TNF-α stimulation. Sample collection was the same as described above under “Cell infection.”
Infection of zebrafish larvae with E. piscicida.Three days postfertilized healthy zebrafish larvae were randomly infected with E. piscicida by microinjection. Microinjection in zebrafish larvae was performed to determine the contribution of EseK to pathogenesis. Bacteria inoculated from a fresh plate were grown for 16 h at 30°C with shaking in TYB and subcultured in TYB for 3 h at 30°C without shaking. The bacteria were then washed three times in phosphate-buffered saline (PBS), and the OD600 was adjusted to 1.0. In the absence of MAPK inhibitors, wild-type bacteria, the ΔeseK mutant, and the eseK deletion strain complemented with wild-type EseK were injected into the yolk sac at 50 CFU per fish. For pharmacological pretreatment, 50 μM the MAPK inhibitor SB202190 (p38α inhibitor), SP600125 (JNK inhibitor), or U0126-EtOH (ERK inhibitor) was used to pretreat the zebrafish larvae for 2 h before microinjection of the ΔeseK mutant. The inhibitor treatment continued throughout the incubation period. The zebrafish larvae were treated with vehicle only (0.1% DMSO) as a control. In the presence or absence of MAPK inhibitors, bacteria were injected into the yolk sac at 50 CFU per fish. At 6 h, 12 h, 24 h, 36 h, and 48 h postinoculation, five zebrafish larvae were harvested and homogenized, and a dilution series was spread onto deoxycholate hydrogen sulfide lactose (DHL) plates supplemented with colistin. We counted the colonies and then analyzed the data to make a diagram.
RNA extraction and quantitative real-time PCR.HeLa cells, ZF4 cells, or zebrafish larvae were infected as described above under “Cell infection” and “Infection of zebrafish larvae with E. piscicida.” RNA was extracted from all samples using an RNA isolation kit (Tiangen, Beijing, China). One microgram of each RNA sample was used for cDNA synthesis with a FastKing one-step RT-PCR kit (Tiangen) and quantitative real-time PCR (RT-PCR) was carried out on an FTC-200 detector (Funglyn Biotech, Shanghai, China) by using a SuperReal PreMix Plus (SYBR green) kit (Tiangen). Total RNA from whole HeLa cells was prepared as described above. The expression of TNF-α, IL-6, and IFN-γ was determined using real-time PCR (ABI StepOne quantitative PCR). Each primer pair (Table 2) was designed using the NCBI/Premier-BLAST program. The expression of each gene was normalized to that of the β-actin transcript and expressed as the fold change relative to the expression seen in PBS-injected zebrafish larvae. All quantitative PCRs were performed for three biological replicates, and the data for each sample were expressed relative to the expression level of the β-actin gene by using the 2−ΔΔCT threshold cycle (CT) method.
Statistical analysis.Statistical analyses were performed using the GraphPad Prism program (GraphPad Software). Data are presented as the mean ± standard deviation (SD) for triplicate samples per experimental condition unless noted otherwise. Representative results are shown in the figures. Statistical analyses were performed using one-way analysis of variance (ANOVA). Differences were considered significant at P values of <0.05, <0.01, and <0.001, as indicated in the figures.
ACKNOWLEDGMENTS
This work was supported by NSFC grants 31430090 to Y.Z., 31472308 to Q.L., and 31622059 to Q.L. D.Y. was supported by an NSFC grant (31602187), the Shanghai Pujiang Program (2016PJD020), and the Young Elite Scientists Sponsorship Program (CAST no. 2016QNRC001).
We declare that we have no conflicts of interest with the contents of this article.
FOOTNOTES
- Received 3 May 2018.
- Returned for modification 6 June 2018.
- Accepted 1 July 2018.
- Accepted manuscript posted online 9 July 2018.
Supplemental material for this article may be found at https://doi.org/10.1128/IAI.00233-18.
REFERENCES
- Copyright © 2018 American Society for Microbiology.
