Ubiquitination is required for MHC class II degradation by F. tularensis infection supernatants. IFN-γ-pretreated BMDM were pulsed with infection supernatants for 6 h and chased with or without proteasome inhibitor MG-132 (5 μM) for an additional 6 h. MHC class II levels were analyzed by Western blotting. Densitometry results are normalized to the “+IFN-γ” medium control and expressed as the mean of three independent experiments ± the SEM. **, P ≤ 0.01 (statistical difference from the LVS supernatant plus MG-132 treatment).

F. tularensis infection supernatants induce ubiquitination of MHC class II. BMDM were pretreated or not pretreated with 100 U of IFN-γ/ml for 24 h and then treated with infection supernatant with or without 50 mM NH₄Cl as described in the text. Ubiquitinated MHC class II from cell lysates was precipitated with P62-UBA-ubiquitin-binding agarose beads. Ubiquitinated and total MHC class II was analyzed by Western blotting. Exposure of BMDM to infection supernatant leads to increased ubiquitination of MHC class II, observable only with the addition of NH₄Cl to prevent rapid class II degradation. Shown are representative results from one of two independent experiments.

Production of F. tularensis-elicited factor independent of macrophage activation and F. tularensis viability. (A) Producer BMDM activated with 100 U of IFN-γ/ml or left untreated prior to infection with F. tularensis LVS were used to generate infection supernatants. Reporter BMDM were incubated with infection supernatants for 15 h and analyzed for MHC class II expression by flow cytometry. IFN-γ activation of producer macrophages failed to block production of the factor responsible for class II downregulation. (B) Infection supernatants were generated from producer BMDM infected with live, UV-killed, or formaldehyde-fixed F. tularensis LVS and exposed to IFN-γ-activated reporter BMDM for analysis of class II expression by flow cytometry. Inactivation of F. tularensis LVS had no effect on the elicitation of the factor responsible for class II downregulation. The data are expressed as the means of three independent experiments ± the SEM.

PGE₂ indirectly responsible for MHC class II degradation. (A) BMDM were mock infected, infected with F. tularensis LVS, or treated with the indicated concentrations of PGE₂. After a 15-h incubation, supernatants were collected and added to IFN-γ-activated BMDM. BMDM were lysed, and the MHC class II levels were analyzed by Western blotting. Exposure of macrophages to LVS or PGE₂ supernatants leads to a loss of total MHC class II, suggesting that PGE₂ can induce the production of the factor responsible for class II degradation. Shown are representative results from one of three independent experiments. (B) Proposed mechanism of Francisella-induced macrophage class II downregulation. Macrophage invasion by F. tularensis results in the production of PGE₂, which acts in an autocrine/paracrine fashion to elicit the production of a high-molecular-weight (MW) protease-resistant factor that elicits the ubiquitin-dependent degradation of MHC class II molecules in acidic intracellular compartments (L, lysosomes) of non-Francisella-infected “reporter” macrophages.

Factor responsible for class II degradation is heat sensitive but resistant to PK treatment. (A) Infection supernatants were boiled for the indicated times or incubated with 10, 50, or 100 mg of PK-beads/ml for 30 min at 37°C prior to exposure to IFN-γ-activated BMDM. MHC class II levels were determined by Western blotting. Infection supernatants boiled at 100°C failed to downregulate macrophage class II, whereas PK-treated infection supernatants retained the ability to induce class II degradation. A representative Western blot from one of three independent experiments is shown. (B) To confirm the PK activity, infection supernatants were treated with PK-beads as described above, and the presence of protein was detected by SDS-PAGE and silver staining. A silver-stained gel representative of one of two experiments is shown.

Infection supernatant-mediated MHC class II degradation occurs in acidic vesicles. (A) BMDM pretreated or not pretreated with 100 U of IFN-γ/ml for 24 h were exposed to F. tularensis LVS infection supernatants for the indicated time points and analyzed by Western blotting for MHC class II expression. Statistically significant class II degradation was observed at 10 h postexposure to infection supernatants. Densitometry values are normalized to the “+IFN-γ” medium control. The results are representative of one of three independent experiments ± the SEM. **, P ≤ 0.01; ***, ≤ 0.001 (statistical difference from the control treatment). (B) BMDM pretreated or not pretreated with 100 U of IFN-γ/ml for 24 h were pulsed with infection supernatant for 6 h and cocultured with 10 or 50 mM ammonium chloride (NH₄Cl) for an additional 10 h before being lysed and analyzed by Western blotting. Blocking of macrophage lysosomal activity prevents infection supernatant-induced degradation of MHC class II in a dose-dependent manner, demonstrating the degradation of internalized class II in lysosomes. Shown are representative results from one of three independent experiments.

The factor responsible for MHC class II downregulation is distinct from PGE₂ but requires macrophage COX activity. (A) BMDM were exposed to infection supernatants or the indicated concentrations of PGE₂ for 15 h without IFN-γ treatment or after treatment with 100 U of IFN-γ/ml and then analyzed by flow cytometry. Concentrations of PGE₂ 10-fold greater than that observed during infection failed to elicit the same level of MHC class II inhibition as that induced by infection supernatants, suggesting that a non-PGE₂ factor is responsible for the majority of class II downregulation. The results represent the means of three independent experiments ± the SEM. *, Statistical difference (P ≤ 0.05) from the mock treatment. (B) F. tularensis infection supernatant was fractionated using a 10-kDa filter prior to addition to IFN-γ-activated BMDM. MHC class II levels were determined by Western blotting. Molecular mass filtration revealed that the factor responsible for MHC class II degradation is >10 kDa, suggesting the involvement of a nonprostaglandin factor. Shown are representative results from one of three independent experiments. (C) IFN-γ-activated reporter BMDM exposed to infection supernatants generated in the presence of 10 μM indomethacin or ethanol control. To account for the potential carryover effects of cyclooxygenase inhibition of reporter cells, supernatants generated in the absence of drug were “spiked” with indomethacin during incubation with reporter macrophages. Supernatants generated by infected BMDM with blocked COX1/2 activity failed to downregulate MHC class II levels, demonstrating the dependence of this enzyme in soluble-factor generation. A Western blot representative of one of three independent experiments with the relative densitometry plotted ± the SEM is shown. *, P ≤ 0.05; **, P ≤ 0.01 (statistical difference from the mock treatment).

Downregulation of MHC class II in IFN-γ-activated macrophages by F. tularensis infection supernatants. (A) BMDM were infected with F. tularensis LVS at an MOI of 100:1 or treated with medium alone (mock) for 2 h. Cells were treated with gentamicin for 1 h, washed, and incubated overnight to generate supernatants. Fresh reporter BMDM (pretreated or not pretreated 100 U of IFN-γ/ml for 24 h) were exposed to cleared infection supernatants for 15 h and analyzed for cell surface protein expression by staining for flow cytometry. Exposure of reporter BMDM to LVS infection supernatants leads to reduced surface expression of MHC class II and CD86, while FcγR and CD11b expression remains high. A lack of propidium iodide (PI) uptake indicates no decrease in cell viability. A vertical line represents the mean fluorescence intensity of the “+IFN” control for each surface marker. (B) After exposure to LVS or SchuS4 infection supernatants, the BMDM were lysed, and the MHC class II levels were analyzed by Western blotting. Exposure of macrophages to LVS or SchuS4 supernatants leads to a loss of total MHC class II, suggesting degradation after internalization. Shown are representative results from one of three independent experiments. (C) IFN-γ-pretreated BMDM were cultured with 10 μM HEL and F. tularensis LVS infection supernatant for 15 h. The BMDM were fixed with 1% paraformaldehyde, washed, and cocultured with the HEL peptide-specific h4Ly50.5 T-cell hybridoma line for 15 h. IL-2 levels were determined by cytometric bead array. Treatment of macrophages with LVS infection supernatants leads to a >90% loss of antigen presentation and resultant T-cell activation. The results are representative of one of three independent experiments ± the standard error of the mean (SEM). ***, P ≤ 0.001 (statistical difference from mock supernatant treatment).