Mincle-mediated translational regulation is required for strong nitric oxide production and inflammation resolution

Abstract

In response to persistent mycobacteria infection, the host induces a granuloma, which often fails to eradicate bacteria and results in tissue damage. Diverse host receptors are required to control the formation and resolution of granuloma, but little is known concerning their regulatory interactions. Here we show that Mincle, the inducible receptor for mycobacterial cord factor, is the key switch for the transition of macrophages from cytokine expression to high nitric oxide production. In addition to its stimulatory role on TLR-mediated transcription, Mincle enhanced the translation of key genes required for nitric oxide synthesis through p38 and eIF5A hypusination, leading to granuloma resolution. Thus, Mincle has dual functions in the promotion and subsequent resolution of inflammation during anti-mycobacterial defence using both transcriptional and translational controls.

Figure 1. TDM stimulation diminishes NLRP3-dependent inflammasome activation and IL-1β production.

(a,b) WT and Mincle^−/− BMDMs were stimulated with LPS or co-stimulated with LPS and TDM for 12 h or the indicated times and then treated with ATP for 1 h (L: LPS; T: TDM; A: ATP; UT: untreated). (a) Left: immunoblot analysis of proIL-1β, mature IL-1β (p17), procaspase-1 (proCasp-1) and cleaved caspase-1 (p10) in cell culture supernatants (SN) and whole-cell lysates (XT). Right: kinetic quantitative analysis of proIL-1β and mature IL-1β (p17) from the immunoblots. (b) Release of lactate dehydrogenase (LDH) (assessing cell death) from cells. (c) ELISA of released IL-1β from LPS-treated WT and Mincle^−/− BMDMs, stimulated with TDM for 12 h and then treated with MSU, nigericin or poly(dA:dT) for 3 h. *P<0.05, **P<0.01, ***P<0.001 (two-tailed unpaired t-test). Data are representative of at least three independent experiments. (b,c: mean and s.d.).

Figure 2. TDM-induced nitric oxide upregulation inhibits activation of the NLRP3 inflammasome.

(a) WT and Mincle^−/− BMDMs were stimulated with LPS (L+A) or co-stimulated with LPS and TDM (L+T+A) for 12 h in the presence of the indicated chemical inhibitors, and then treated with ATP for 1 h. ELISA of released IL-1β. (b) Immunoblot analysis of total S-nitrosylated proteins (biotin), NLRP3 or caspase-1 in BMDMs treated with LPS or stimulated with TDM for 12 h. Below (lysate), immunoblot analysis of total lysate fractions. (c) Immunoblot analysis of IL-1β and caspase-1 from LPS-treated WT and iNOS^−/− BMDMs, stimulated with TDM for 12 h, and then treated with ATP for 1 h or nigericin for 3 h. Data are representative of one (a) or two (b,c) independent experiments.

Figure 3. Mincle signalling results in efficient iNOS mRNA translation.

(a) WT and Mincle^−/− BMDMs were stimulated with Pam3 or co-stimulated with Pam3 and TDM for the indicated times. Left: qRT-PCR analysis of iNOS mRNA levels from each type of stimulated macrophage; right: immunoblot analysis of iNOS protein expression. *P<0.05 (Student's t-test). (b) The half-life of the iNOS protein was measured from Supplementary Fig. 7. The ratio of iNOS signals from the actinomycin D- and cycloheximide-treated samples to those from samples treated only with LPS or co-treated with LPS and TDM for 6 h were calculated and plotted as percentages (initial value set at 100%). (c) Immunoblot analysis of 4EBP-1 and specific phosphor-4EBP-1 (Thr 37/46) from WT and Mincle^−/− BMDMs treated with Torin1, Pam3 (P), TDB (T) or co-treated with Pam3 and TDB (PT) for 24 h. UT, untreated. (d) Polysome profiles and immunoblot analysis of the S6 (small) and L7a (large) ribosomal subunit components and β-actin (nonribosomal protein) in lysates of WT macrophages stimulated with Pam3 or co-stimulated with Pam3 and TDM for 12 h. qRT-PCR analysis of iNOS (left) and β-actin (right) mRNA in polysome fractions from WT macrophages stimulated with Pam3 or co-stimulated with Pam3 and TDM, presented for each fraction relative to the sum of all 14 fractions. Data are representative of at least three independent experiments (a,b,d: mean and s.d.).

Figure 4. Mincle-induced p38 activation is required for iNOS translation.

(a,b) WT BMDMs were stimulated with TDM or Pam3 or co-stimulated with Pam3 and TDM in the presence of the indicated chemical inhibitors, for 12 h. (a) Nitric oxide production in culture supernatants. ***P<0.001 (Student's t-test). (b) left: qRT-PCR analysis of iNOS mRNA levels from stimulated macrophages; right: immunoblot analysis of iNOS protein expression. (c) Immunoblot analysis of phospho-p38 from WT and Mincle^−/− BMDMs treated with Pam3, TDM or co-stimulated with Pam3 and TDM for 1 or 12 h. Data are representative of at least three independent experiments.

Figure 5. eIF5A hypusination is required for Mincle-mediated iNOS translation.

(a,b) WT BMDMs were stimulated with TDM, Pam3 or co-stimulated with Pam3 and TDM in the presence of GC7 for 12 h. (a) Nitric oxide release in culture supernatants. (b) qRT-PCR (left) and immunoblot (right) analysis of iNOS mRNA or protein expression. (c) Fluorographic analysis of hypusinated eIF5A from WT BMDMs treated with Pam3, TDM or co-treated with Pam3 and TDM in the presence of control (con) or GC7. (d) Fluorographic analysis of hypusinated eIF5A from WT and Mincle^−/− (KO) BMDMs treated with Pam3, TDM or co-treated with Pam3 and TDM. (e) RIP analysis of iNOS and Actb mRNAs from NIH-3T3 cells co-transfected with expression vector for Flag alone (Flag-Mock) or Flag-tagged eFI5A WT or K50A mutant, plus an iNOS mRNA expression vector. Agarose gel electrophoresis (top) and qRT-PCR analysis (bottom) for the indicated genes. Data are expressed as per cent recovery relative to input RNA. *P<0.05, ***P<0.001 (Student's t-test). Data are representative of three (a,b,e) or two (c,d) independent experiments (a,b,e: mean and s.d.).

Figure 6. Specific genes selected from the polysome profiling assay are regulated by Mincle-eIF5A-dependent translation.

(a) Polysome fraction mRNAs from WT BMDMs co-treated with Pam3, TDM, with or without GC7 were analysed. qRT-PCR analysis of the indicated mRNAs in cells treated as indicated, presented for each fraction relative to the sum of all fourteen polysome fractions. (b) RIP analysis of the indicated gene transcripts from iBMDM cells stably expressing Flag-eIF5A, stimulated as indicated. Immunoprecipitation with anti-Flag antibody and qRT-PCR analysis of the indicated genes. (c) WT BMDMs were stimulated with Pam3, TDM or co-stimulated with Pam3 and TDM, in the presence of GC7 or control vehicle for 12 h. Immunoblot analysis of the indicated protein expression levels. Data are representative of three independent experiments (a–c: mean and s.d.).

Figure 7. iNOS deficiency and inhibition of eIF5A hypusination aggravate granuloma formation by TDM.

(a–e) WT and iNOS^−/− mice were injected intravenously with an oil-in-water emulsion of TDM (n>6 mice/group). (a) Histology of lungs at 7 and 14 days after injection (scale bars, 100μm). (b) TDM-induced lung swelling. On days 7 and 14 after injection of TDM, lung swelling was evaluated by LWI. (c) Identification of leucocyte subsets in lung granulomas on day 14 after TDM injection by flow cytometry. The number of neutrophils (PMN, CD11b⁺ Ly6G⁺), monocytes (Mono, CD11b⁺ Ly6G⁻), T cells (CD3⁺) and B cells (CD19⁺) are indicated. (d) ELISA of IL-1β in lung lysates on day 7 or 14 after TDM injection. (e) Immunoblot analysis of active caspase-1 (Casp1 p10) and iNOS in lysates of lungs from WT and iNOS^−/− mice injected with TDM; β-actin serves as a loading control (each lane represents an individual mouse). (f–j) Mice were administered with PBS, GC7 or CPX via intraperitoneal injection, and with oil-in-water emulsion of TDM or emulsion alone, via intravenous injection. On day 7 after the indicated injection, lungs were harvested and experiments were performed. (f) Lung histology was examined by H&E staining (scale bars, 100 μm.) (g) TDM-induced lung swelling. At day 7 after the indicated injection, lung swelling was evaluated by LWI. (h) ELISA of IL-1β in lung lysates on day 7 TDM injection. (i) Immunoblot analysis of active caspase-1 (Casp1 p10) and iNOS in lysates of lungs from mice injected with TDM with/without GC7 or CPX, assessed on day 7 after injection; β-actin serves as a loading control (each lane represents an individual mouse). (j) Lethal systemic inflammation by TDM (n=five mice/group, P-value calculated by Mantel–Cox test). *P<0.05, **P<0.01 (two-tailed unpaired Student's t-test). Data are representative of two experiments (b–d,g–i: mean and s.d. of three to five mice per group).

Figure 8. Model explaining Mincle-mediated translational regulation for nitric oxide production and inflammation resolution.

In this model, Mincle is essential not only for transcription of proinflammatory genes, but also for translation of the key NO synthesis genes. Mincle promotes strong NO release by enhancing iNOS translation via a mechanism dependent on p38-mediated hypusination of eIF5A, inhibiting NLRP3 inflammasome and caspase-1-dependent IL-1β production. ASC, Apoptosis-associated speck-like protein containing a caspase recruitment domain; Hyp, hypusination; SNO, S-nitrosylation.