Fig 1.
Identification of Trxlp as a novel virulence effector.
(A) Supernatants (Sup.) and pellets (Pel.) from Trxs-HA fusion-expressing EIB202 were analyzed by immunoblotting. The bacteria were cultured in DMEM for 16 h and fractioned into pellets and supernatants. Anti-HA antibody was used to probe the Trxs-HA fusion protein, and anti-RNAP antibody was used as a bacterial cytosolic marker. (B) Supernatants (Sup.) and pellets (Pel.) from Trxlp-HA fusion-expressing EIB202, ΔT3SS, ΔT6SS E. piscicida, or HA-expression EIB202 were analyzed by immunoblotting. The bacteria were cultured in DMEM for 16 h and divided into pellets and supernatants. Anti-HA was used for Trxs-HA fusion protein probing, and anti-RNAP was used as a bacterial cytosolic marker. (C) Supernatants (Sup.) with/without Outer membrane vesicles (OMVs), OMVs, and pellets (Pel.) from Trxlp-HA fusion-expressing EIB202 were analyzed by immunoblotting. Anti-HA and anti-OmpA were used to probe Trxlp-HA fusion protein outer membrane protein vesicles, respectively, and anti-RNAP was used as a bacterial cytosolic marker. (D) Assays of intracellular translocation of Trxlp by immunoblotting. HeLa cells were infected with Trxs-HA fusion-expressing EIB202 at a MOI of 100 for 2 hours; cells were subjected to differential centrifugation to separate subcellular fractions. These fractions were analyzed by immunoblotting, as indicated. Calnexin, a marker of the host cell membrane; β-tubulin, a marker of cytosolic proteins; RNAP, a bacterial cytosolic protein. (E) Assay of intracellular translocation of indicated Trxs by fusing with β-lactamase. HeLa cells were infected with EIB202, ΔT3SS or ΔT6SS E. piscicida expressing effector-TEM fusion protein at a MOI of 100. Eight hours after infection, cells were loaded with CCF4-AM. Translocation of effector-TEM into the cell cytosol results in the cleavage of CCF4-AM, causing the emission of blue fluorescence. Uncleaved CCF4-AM emits green fluorescence. Scale bar = 50 μm. TEM, TEM-1-β-lactamase. (A-E) Data are representative of at least 3 experiments.
Fig 2.
Crystal structure analysis of Trxlp.
(A) Two orthogonal views of the overall structure of Trxlp. (B) Trxlp is crystallized in the reduced form. The electron density (2Fo-Fc) map, contoured at 1δ, is shown for the CXXC motif. (C)Superimposition of thioredoxin from various species. Trxlp (PDB: 5ZF2), green; human thioredoxin 1 (PDB: 1ERT), yellow; human thioredoxin 2 (PDB: 1UVZ), cyan; Escherichia coli thioredoxin 1 (PDB:2TRX), orange; Thermus thermophilus thioredoxin 1 (PDB: 2YZU), red. (D) The multiple sequence alignments of thioredoxin antioxidant system-related amino acid sequence with other known Trx sequences. The sequences used to generate the multiple sequence alignments are as follows: EIB202 Trxlp and TRX1 (GenBank accession no. ACY85021.1, ACY82946.1), E. coli TRX1 (GenBank accession no. AFG42725.1), Homo sapiens TRX1 (GenBank accession no. AFH41799.1), EIB202 TRX2 (GenBank accession no. ACY83406.1), Homo sapiens TRX2 (GenBank accession no. AAF86467.1), and T. thermophilus HB8 TRX (GenBank accession no. BAD71304.1). Residues that form the conserved redox catalytic motif are indicated in the red box. Secondary structure assignments based on Trxlp structure are shown as cylinders (α-helices) and arrows (β-strands).
Fig 3.
Trxlp interacts with ASK1-TBD residues to abrogate ASK1 activation.
(A) Immunoprecipitation assay of Trxlp and ASK1-TBD. HEK293T cells were cotransfected with indicated Trxlp-GFP and ASK1-TBD-HA expression vectors and cultured for 36 h. The cells were then lysed in lysis buffer and immunoprecipitated with HA beads for 4 h. Anti-HA and anti-GFP antibodies were used for immunoblotting analysis. (B-D) Pulldown assays of interaction between purified ASK1-TBD-His and Trxlp-MBP. Immunoblotting to detect ASK1-TBD-His and Trxlp-MBP is shown. (E) Trxlp and human Trx1 inhibit homophilic interaction of ASK1 through N-terminal region. HEK293T cells were transfected with the indicated combinations of ASK1-N-HA, ASK1-N-flag, Trxlp-HA and human TRX1-HA plasmids. Cell lysates were immunoprecipitated with anti-flag antibody. Immunoprecipitates and aliquots of each lysate were subjected to SDS-PAGE followed by immunoblotting with indicated antibodies. (F and G) Trxlp inhibits the phosphorylation of endogenous ASK1. HEK293T cells transfected with Trxlp-HA, human TRX1-HA or mutant Trxlp (FSXXS), respectively, and pretreated with 100 ng/ml TNF-α for 30 min after 36 h. The phosphorylation of endogenous ASK1 and ASK1 in 293T cells was detected by immunoblotting using anti-phospho-ASK1 (Thr845) and anti-ASK1 antibodies. Immunoblotting results for HA and β-actin are also shown. (A-G) The signal intensities were quantitatively analyzed using Quantity one software. Data (Means ± SD) are representative of at least 3 experiments.
Fig 4.
Trxlp suppresses the phosphorylation of Erk1/2 and p38 dependent on ASK1 during E. piscicida infection.
(A-C) Immunoblotting assays of ASK1, MAP2Ks and MAPKs activation during E. piscicida infection in HeLa cells. Wild-type HeLa cells and ASK1-KO HeLa cells were infected with EIB202, Δtrxlp, or trxlp-complemented E. piscicida at a MOI of 100 for 2 h. Cell lysates were collected and probed for anti-phospho-ASK1 and anti-ASK1 antibodies (A), or probed for anti-phospho-Erk1/2 and anti-Erk1/2, and anti-phospho-MEK1/2 and MEK1/2 antibodies (B), or probed for anti-phospho-p38α and anti-p38α, and anti-phospho-MKK3/6 and MKK3/6 antibodies (C). (D) Immunoblotting assays of ASK1 and MAPKs activation during E. piscicida infection in Zebrafish fibroblasts (ZF4). ZF4 cells were infected with EIB202, Δtrxlp, or trxlp-complemented E. piscicida for 2 h at a MOI of 10. Cell lysates were probed with anti-phospho-ASK1 and anti-ASK1, anti-phospho-Erk1/2 and anti-Erk1/2, anti-phospho-p38 and anti-p38, anti-phospho-JNK and anti-JNK antibodies. (A-D) β-Actin is shown as a loading control. The signal intensities were quantitatively analyzed using Quantity one software. Data (Means ± SD) are representative of at least 3 experiments.
Fig 5.
Trxlp promotes bacterial colonization in vivo.
(A) Schematic representation of the zebrafish larvae microinjection infection model. Zebrafish larvae were maintained in E3 medium for up to 8 days post-fertilization (dpf). Bacteria were microinjected into the yolk sac of zebrafish larvae at 3 dpf, and the survival, bacterial burden, and cytokine expression were analyzed. (B) Wild-type (WT) or ask1-MO zebrafish larvae were infected with EIB202, Δtrxlp, trxlp-complemented E. piscicida (50 CFUs/larvae), or PBS as a control, respectively. The survival of zebrafish larvae was monitored for 5 days. n = 50 fish per group. Data shown are from 3 representative experiments. ** p < 0.01. (C) The zebrafish larvae were infected as in Fig 5B and collected at the indicated post-infection time points, and homogenates were plated to determine the bacterial CFUs per larvae. n = 5 fish per group at each time point. Data are representative of at least 3 experiments. (D) The zebrafish larvae were infected as in Fig 5B, and mRNA levels of TNF-α and IL-10 in indicated zebrafish larvae infected with EIB202, Δtrxlp, or trxlp-complemented E. piscicida at indicated time points were determined by qRT-PCR. PBS-treated zebrafish was used as the control. Data (means ± SD) are representation of 3 experiments. * p < 0.05. (E) Diagram of Trxlp function during E. piscicida infection process. The novel virulence effector Trxlp targets and inhibits the phosphorylation of ASK1, thereby suppressing Erk1/2- and p38-MAPKs and inhibiting the expression of TNF-α and IL-10 during infection.
Fig 6.
Proposed mechanism for E. piscicida thioredoxin-like protein.
The novel bacterial virulence effector, thioredoxin-like protein (Trxlp), could be translocated into cytosol of host cells, mimicking the endogenous TRX1 to target ASK1-MAPK signaling to restrict the innate immune response in vivo.