The Salmonella Effector Protein SopA Modulates Innate Immune Responses by Targeting TRIM E3 Ligase Family Members

Abstract

Salmonella Typhimurium stimulates inflammatory responses in the intestinal epithelium, which are essential for its ability to replicate within the intestinal tract. Stimulation of these responses is strictly dependent on the activity of a type III secretion system encoded within its pathogenicity island 1, which through the delivery of effector proteins, triggers signaling pathways leading to inflammation. One of these effectors is SopA, a HECT-type E3 ligase, which is required for the efficient stimulation of inflammation in an animal model of Salmonella Typhimurium infection. We show here that SopA contributes to the stimulation of innate immune responses by targeting two host E3 ubiquitin ligases, TRIM56 and TRIM65. We also found that TRIM65 interacts with the innate immune receptor MDA5 enhancing its ability to stimulate interferon-β signaling. Therefore, by targeting TRIM56 and TRIM65, SopA can stimulate signaling through two innate immune receptors, RIG-I and MDA5. These findings describe a Salmonella mechanism to modulate inflammatory responses by directly targeting innate immune signaling mechanisms.

Fig 1. SopA interacts with TRIM56 and TRIM65.

(A) FLAG-epitope-tagged SopA or its catalytic mutant SopA^C753S were transiently co-expressed in HEK293T cells with M45-epitope-tagged TRIM56 or TRIM65 and their interaction probed by immunoprecipitation and Western immunoblotting. This experiment was repeated 3 independent times with similar results. (B) HEK293T cells were transiently transfected with plasmids expressing M45-epitope tagged TRIM56 or TRIM65 and 18 h after transfection, cells were infected with a S. Typhimurium strain expressing FLAG-epitope tagged SopA^C753S or the equally tagged T3SS-effector protein PipA as a specificity control. Five hours after infection, cell lysates were analyzed by immunoprecipitation with anti-FLAG and Western immunoblotting with anti-M45 monoclonal antibodies. This experiments was repeated 2 independent times with similar results. (C) FLAG-epitope-tagged SopA^C753S was transiently co-expressed in HEK293T cells with M45-epitope-tagged TRIM25, TRIM56, TRIM65, TRIM5, TRIM39 or TRIM62. Cell lysates were then analyzed by immunoprecipitation with anti-FLAG and western immunoblotting with anti-M45 antibodies. This experiment was repeated 3 independent times with similar results.

Fig 2. SopA interacts with the RING finger domain of TRIM65 and TRIM56.

FLAG-epitope-tagged SopA^C753S was transiently co-expressed in HEK293T cells with M45-epitope-tagged TRIM56, TRIM65 or their RING-finger domain mutants TRIM56^C24A and TRIM65^C15A, and their interaction was analyzed by immunoprecipitation and Western immunoblotting. This experiment was repeated 3 independent times with similar results.

Fig 3. SopA enhances TRIM65 ubiquitination.

(A) Purified FLAG-epitope-tagged TRIM65 was incubated with SopA or its catalytic mutant SopA^C753S in the presence of ubiquitin, ATP, E1 (UBE1) and E2 (UbcH5b) and TRIM65 ubiquitination was detected by its mobility shift in Western blot analysis with anti-FLAG antibody. The chromatographic profiles of the SopA preparations used in the assays are shown. (B) HEK293T cells were co-transfected with plasmids encoding HA-epitope-tagged ubiquitin, and either FLAG-epitope-tagged TRIM65, or the mutants TRIM65^C15A (catalytically deficient) and TRIM65^T24K (autoubiquitination deficient) along with plasmids encoding either SopA or its catalytic mutant SopA^C753S. Cell lysates were evaluated for the levels of protein expression and TRIM65 ubiquitination by immunoprecipitation with anti-FLAG and Western immunoblot analysis with anti-FLAG, anti-HA, and anti-M45 antibodies, respectively. This experiment was repeated 2 independent times with similar results. (C) HEK293T cells were co-transfected with plasmids encoding HA-epitope-tagged ubiquitin and FLAG-epitope tagged TRIM65 and 18 hours after transfection, cells were infected with an S. Typhimurium strain expressing either wild type SopA or its catalytic mutant SopA^C753S. Five hours after infection, cell lysates were evaluated for the levels of protein expression and TRIM65 ubiquitination by immunoprecipitation and Western immunoblot analysis with anti-FLAG, anti-HA, and anti-M45 antibodies, respectively. This experiment was repeated 3 independent times with similar results.

Fig 4. TRIM65 interacts with MDA5 and enhances MDA5-stimulated interferon-β expression.

(A) FLAG-epitope-tagged RIG-I (2CARD) (the two CARD domains of RIG-I), RIG-I (full length), MDA5 (2CARD) (the two CARD domains of MDA5), or MDA5 (full length) were transiently expressed in HEK 293T cells together with M45-epitope-tagged TRIM65. Protein interactions were analyzed by immunoprecipitation with anti-FLAG and immunoblotting with anti-M45 and anti-FLAG antibodies. This experiment was repeated 3 independent times with similar results. (B-E) HEK293T cells were co-transfected with plasmids expressing TRIM65 or the catalytically-deficient TRIM65^C15A mutant along with plasmids expressing RIG-I (20 ng), MDA5 (50 ng), MDA5(2CARD) (10 ng) or MAVS (50 ng) as indicated, and the reporter plasmids expressing firefly luciferase under the control of interferon-β promoter and renilla luciferase (to standardize the transfection efficiency). The activation of the interferon-β promoter was analyzed 18 h after transfection by Dual Luciferase Reporter Assay System (Promega). Values represent the mean +/- standard deviation of the relative levels of normalized firefly luciferase from 3 independent experiments. The normalized firefly luciferase activity relative to mock-transfected cells was 10.5 ±2.9 for MDA5 only transfected cells (C), 11.1 ± 4.8 for RIG-I only transfected cells (D), 30.6 ± 17.8 for MAVS only transfected cells (E), and 15.4 ± 3.3 for MDA5(2CARD) only transfected cell (F). *: indicates statistically significant difference (p < 0.05) vs. mock transfected cells (B), vs. MDA5-transfected cells (C), vs. RIG-I-transfected cells (D), vs. MAVS-transfected cells (E) and vs. MDA5(2CARD)-transfected cells (F).

Fig 5. SopA enhances the ability of TRIM56 and TRIM65 to stimulate interferon-β expression.

(A-B) HEK293T cells were co-transfected with plasmids expressing either wild type SopA or its catalytic mutant SopA^C753S along with plasmids expressing TRIM65 (25 ng), TRIM56 (10 ng), MDA5 (50ng) or RIG-I(2CARD) (4ng) as indicated, and reporter plasmids expressing firefly luciferase under the control of interferon-β promoter and renilla luciferase (to standardize the transfection efficiency). The activity of the interferon-β promoter was analyzed 18 hs after transfection with the Dual Luciferase Reporter Assay System (Promega). Values represent the mean +/- standard deviation of the relative levels of normalized firefly luciferase from 3 independent experiments. The normalized firefly luciferase activity relative to mock-transfected cells was 7.3 ± 3.9 for MDA5 only transfected cell (B) and 17.8 ± 9.7 for RIG-I(2CARD) only transfected cells. *: indicates statistically significant difference (p < 0.05) (C) Primary MEFs were infected with wild type S. Typhimurium, the ∆sopA or the effectorless (∆sipA ∆sptP ∆avrA ∆sopE ∆sopE2 ∆sopA ∆sopB sopD ∆sopD2 ∆slrP) isogenic mutant strain and the levels of interferon-β gene expression were measured by qRT-PCR 4 and 8 hours after infection. Data represent the mean ± standard deviation of the n-fold expression of interferon-β mRNA over GAPDH relative to non-infected cells. This experiment was repeated 3 times with equivalent results. (D) Primary MEFs were infected with the effectorless S. Typhimurium strain (∆sipA ∆sptP ∆avrA ∆sopE ∆sopE2 ∆sopA ∆sopB sopD ∆sopD2 ∆slrP) or the effectorless strain expressing plasmid-born wild type SopA, its catalytic mutant SopA^C753S, or an irrelevant effector. The levels of interferon-β gene expression were measured by qRT-PCR 8 hs after infection. Data represent the mean ± standard deviation of the n-fold expression of interferon-β mRNA over GAPDH relative to non-infected cells. The results of a representative experiment (out of 5 independent experiments) is shown. The p value for the difference in interferon-β mRNA levels between MEFs infected with the Δeffectors (pSopA^wt) strain or Δeffectors (pSopA^C753S) was p = 0.002 (n = 5). (E) Immortalized (MEF wt) and TRIM56-deficient (MEF TRIM56 KO) murine embryonic fibroblasts were infected with wild type S. Typhimurium or the effectorless (∆sipA ∆sptP ∆avrA ∆sopE ∆sopE2 ∆sopA ∆sopB sopD ∆sopD2 ∆slrP) isogenic mutant strain expressing plasmid-born wild type SopA or its catalytic mutant SopA^C753S. The levels of interferon-β gene expression were measured by qRT-PCR 8 hours after infection. Data represent the mean ± standard deviation of the n-fold expression of interferon-β mRNA over GAPDH relative to non-infected cells. A representative experiment out of 7 independent experiments is shown. The p values for difference between the interferon-β mRNA levels in wild type MEFs infected with the Δeffectors (pSopA^wt) or Δeffectors (pSopA^C753S) S Typhimurium strains was p = 0.035 while the difference in TRIM56 KO cells was p = 0.639 (n = 7 in each category). (F) Effect of SopA in S. Typhimurium stimulation of pro-inflammatory cytokine expression in the mouse intestine. C57/BL6 nramp ^+/+ mice were orally infected with the effectorless S. Typhimurium strain (ΔsipA ΔsptP ΔavrA ΔsopE ΔsopE2 ΔsopA ΔsopB sopD ΔsopD2 ΔslrP) (n = 6) or the effectorless strain expressing plasmid-born wild type SopA (n = 6). Seventy-two hours after infection the relative levels of the indicated cytokines in the intestine were measured by qRT-PCR. Data were normalized to the levels of GAPDH and represent the mean ± standard error relative to uninfected control animals (n = 3). *: indicates statistically significant difference (p < 0.05).

Fig 6. Proposed model for SopA action.

After its delivery by the S. Typhimurium type III secretion system, SopA targets TRIM56 and/or TRIM65 and through ubiquitination, enhances their ability to modulate RIG-I and/or MDA-5-dependendant signal transduction pathways leading to pro-inflammatory cytokine production. The modulatory activity of SopA may require the prior stimulation of RIG-I and/or MDA-5 by other agonists such as nucleic acids.