HDAC6/aggresome processing pathway importance for inflammasome formation is context-dependent

Abstract

The inflammasome is a large multiprotein complex that assembles in the cell cytoplasm in response to stress or pathogenic infection. Its primary function is to defend the cell and promote the secretion of pro-inflammatory cytokines, including IL-1β and IL-18. Previous research has shown that in immortalized bone marrow-derived macrophages (iBMDMs) inflammasome assembly is dependent on the deacetylase HDAC6 and the aggresome processing pathway (APP), a cellular pathway involved in the disposal of misfolded proteins. Here we used primary BMDMs from mice in which HDAC6 is ablated or impaired and found that inflammasome activation was largely normal. We also used human peripheral blood mononuclear cells and monocyte cell lines expressing a synthetic protein blocking the HDAC6-ubiquitin interaction and impairing the APP and found that inflammasome activation was moderately affected. Finally, we used a novel HDAC6 degrader and showed that inflammasome activation was partially impaired in human macrophage cell lines with depleted HDAC6. Our results therefore show that HDAC6 importance in inflammasome activation is context-dependent.

Keywords: DARPin; HDAC6; degrader; inflammasome; inhibitors; interleukin.

Figure 1

The NLRP3 inflammasome is activated normally in HDAC6 KO or ZnF^mprimary mouse BMDMs.A, workflow for evaluating NLRP3 inflammasome activation in BMDMs. Bone marrow was isolated from mice strains with different mutations in HDAC6 as indicated and BMDMs were expanded for 7 days in the presence of M-CSF1. Subsequently, BMDMs were primed for 4 h with LPS (1 μg/ml) and NLRP3 inflammasome formation was induced by nigericin (20 μM) or MSU (200 μg/ml) addition. Release of IL-1β was monitored by ELISA and activation/cleavage of Caspase-1 was evaluated by immunoblotting. B, normal NLRP3 activation in WT, HDAC6 KO and HDAC6 ZnF^m BMDMs. Immunoblot monitoring HDAC6, Capase-1 (p20) cleavage and α-tubulin as loading control. Nigericin and MSU were used with LPS primed BMDMs (n = 3). C, normal mIL-1β release in the supernatant from WT, HDAC6 KO and ZnF^m BMDMs. IL-1β from supernatant in (B) was measured by ELISA and statistical analysis was done with one-way ANOVA. ns, no significant difference. D, no difference in cytotoxicity following inflammasome induction in WT, HDAC6 KO and ZnF^m BMDMs. Lactate dehydrogenase (LDH) was measured in the culture supernatant from (B) and statistical analysis was done by one-way ANOVA. ns, no significant difference.

Figure 2

A DARPin blocking the HDAC6-Ub interaction has a modest effect on IL-1β release from THP-1 cells.A, structure of the DARPin F10 - HDAC6 ZnF domain complex (PDB: 7ZYU) showing blockade of Ub recruitment. The surface representation shows HDAC6 ZnF (yellow) and the C-terminal LRGG of Ub (light pink) (overlayed with PDB:3GV4), the ribbon representation depicts DARPIN F10 (cyan). B, establishment of a THP-1 F10-FKBP cell line with degradable DARPin F10. Immunoblotting with lysates of THP-1 F10-FKBP cells treated at the indicated concentration with dTAG (for 6 h). The leftmost lane (WT) shows the parental THP-1 cells. The membrane was probed with antibodies against actin as loading control and F10 (HA-F10-FKBP ^F36V, detected with anti-HA). C, IL-1β release under various treatments in THP-1 F10-FKBP cells at 3 h post-nigericin stimulation. Cells were treated as indicated in the scheme at the top. The graph at the bottom shows the measurement of IL-1β in cell supernatants, based on three independent experiments. MCC950 and CGP084892 are NLRP3 and Caspase-1 inhibitors, respectively. One-way ANOVA test was applied for statistical analysis. ∗∗∗∗p <0.0001. Cells were first treated with PMA (0.5 μM) for 3 h, then medium was exchanged and dTAG was added (1 μM); MCC950 (10 μM) and CGP084892 were added 30 min before nigericin (15 μM) treatment. D, IL-1β release from 0 h to 3.5 h after nigericin treatment in THP-1 or THP-1-F10-FKBP cells. Experimental set-up was as in (B) and IL-1β concentration was monitored at timepoint 0, 0.5, 1.5, and 3.5 h post nigericin treatment. Two-way ANOVA tests were applied, and comparison was performed within each timepoint between with and without dTAG group. ∗∗p< 0.01. Experimental setup as in (C).

Figure 3

Effect of XY-07 to 35 mediated HDAC6 degradation on NLRP3 inflammasome activation in PBMCs.A, schematic of the mode of action of an HDAC6 degrader. The PROTAC engages both HDAC6 and the E3 ligase cereblon (CRBN), which then leads to ubiquitination and degradation of HDAC6. B, chemical structure of the HDAC6 degrader XY-07-35 and of the negative control XY-07-191 which no longer recruits CRBN. C, in vitro enzymatic assay of compound XY-07-35 on the activity of different HDACs. IC₅₀ for each enzyme is shown. D, XY-07-35 mediated HDAC6 degradation in PBMCs. PBMCs (without or with LPS priming for 4 h) were treated with the indicated compounds and HDAC6 levels were detected by immunoblotting with an anti-HDAC6 antibody. Detection of actin was used as a control for loading. CGP, CGP084892 (Caspase-1 inhibitor); AFN700 is an inhibitor of the NF-kB pathway (IKK inhibitor). E, HDAC6 degradation impact on IL-1β release in PBMCs following MSU 200 μg/ml (5 h), nigericin 2 μM (2 h) or ATP 3 mM (0.5 h) treatment. Cells were first primed with 0.2 ng/ml LPS for 4 h at 37 °C, then treated with MSU, nigericin, and ATP as indicated. The graphs below show the results of three independent experiments using the inducers as indicated. NP3-948 is an inhibitor of the NLRP3 inflammasome.

Figure 4

Effect of HDAC6 degradation on pyrin inflammasome activation in U937 cells.A, XY-07-35 mediated HDAC6 degradation in pyrin-expressing U937 cells. Cells were treated with the indicated compounds and HDAC6 was detected by immunoblotting with an anti-HDAC6 antibody. Detection of actin was used as a control for protein loading. B, no impact of HDAC6 degradation on IL-18 release in U937-pyrin cells treated with the indicated compounds. Pyrin inflammasome was activated by BAA473 (200 μM) for 4 h at 37 °C. The scheme at the top depicts the experimental outline. The graph at the bottom presents IL-18 release based on four independent experiments. CGP, CGP084892 (200 μM). ns, no significant difference.