DDX3X Links NLRP11 to the Regulation of Type I Interferon Responses and NLRP3 Inflammasome Activation

Abstract

Tight regulation of inflammatory cytokine and interferon (IFN) production in innate immunity is pivotal for optimal control of pathogens and avoidance of immunopathology. The human Nod-like receptor (NLR) NLRP11 has been shown to regulate type I IFN and pro-inflammatory cytokine responses. Here, we identified the ATP-dependent RNA helicase DDX3X as a novel binding partner of NLRP11, using co-immunoprecipitation and LC-MS/MS. DDX3X is known to enhance type I IFN responses and NLRP3 inflammasome activation. We demonstrate that NLRP11 can abolish IKKϵ-mediated phosphorylation of DDX3X, resulting in lower type I IFN induction upon viral infection. These effects were dependent on the LRR domain of NLRP11 that we mapped as the interaction domain for DDX3X. In addition, NLRP11 also suppressed NLRP3-mediated caspase-1 activation in an LRR domain-dependent manner, suggesting that NLRP11 might sequester DDX3X and prevent it from promoting NLRP3-induced inflammasome activation. Taken together, our data revealed DDX3X as a central target of NLRP11, which can mediate the effects of NLRP11 on type I IFN induction as well as NLRP3 inflammasome activation. This expands our knowledge of the molecular mechanisms underlying NLRP11 function in innate immunity and suggests that both NLRP11 and DDX3X might be promising targets for modulation of innate immune responses.

Keywords: DEAD-box helicase; IL-1; anti-viral; inflammasome; innate immunity; nod-like receptors; type I interferon.

Figure 1

NLRP11 interacts with DDX3X. (A) ifnb luciferase reporter assay in HEK293T cells overexpressing FLAG-IKKϵ (5 ng) and either empty vector (ev), myc-NLRP11, or NLRP11-eGFP. Means of three independent experiments conducted in triplicates, relative to the mean of 0 ng NLRP11, ± SEM are shown. **p < 0.01; Welch’s two-sided t-test. (B) Immunoblot of protein lysates of HEK293- and HeLa-eGFP and -NLRP11-eGFP cells, induced overnight with doxycycline. Probing for GFP and actin, or GAPDH as loading control is shown. Monomeric NLRP11-eGFP is marked by an arrow and aggregated NLRP11 by an asterisk. (C) Immunofluorescence micrographs of HeLa-NLRP11-eGFP cells. Cells were induced with doxycycline overnight, fixed and nuclei were stained. 3D-deconvolution of z-stacks of the signal of DNA (lower images) and eGFP (upper images) are shown. Stack size = 0.2 µm. Scale bar = 10 µm. (D) Immunoblots from anti-GFP immunoprecipitations (IP) from HEK293T-eGFP and HEK293-NLRP11-eGFP cells induced over night with doxycycline. IPs were probed for DDX3X and GFP, whole cell lysates (WCL) were probed for DDX3X and GAPDH as loading control. Representative blots for two independent experiments are shown. (E) Immunoblots from anti-GFP IPs from HEK293T cells expressing empty vector, or eGFP-DDX3X and the indicated myc-NLRP11 construct. IP lysates were probed for myc and GFP, whole cell lysates were probed for myc, GFP and actin as a loading control. Representative blots of two independent experiments are shown. *, NLRP11 aggregate; ∆, unspecific bands.

Figure 2

Interaction and co-localization of NLRP11 and DDX3X upon infection. (A) Co-immunoprecipitation of NLRP11-eGFP from HEK293-eGFP and HEK293-NLRP11-eGFP cells after induction with doxycycline overnight and infection with SeV for the indicated time. Immunoblots of the IPs and whole cell lysates (WCL) were probed for DDX3X and GFP. The blots are representative of two independent experiments. (B–F) Indirect immunofluorescence micrographs of HEK293 (B, E), or HeLa- (C, D, F), eGFP or NLRP11-eGFP cells induced with doxycycline overnight and infected with SeV for the indicated time. 3D deconvolution of DDX3X (B–D), or AIF (E, F) staining (red), together with the eGFP signal (green) are shown. Nuclei are stained with Hoechst (blue). Scale bars = 10 µm.

Figure 3

NLRP11 inhibits the phosphorylation of DDX3X by IKKϵ. (A, B) Immunoblot of whole cell lysates from HEK293T cells expressing FLAG-IKKϵ, myc-DDX3X and either myc-NLRP11 (A), or myc-NLRP11 deletion constructs (B) as indicated. Blots were probed for myc, FLAG, pIRF3 and actin as loading control. Representative blots of at least two independent experiments are shown. *, myc-NLRP11 ΔPYD (C) IFNβ release from macrophage-like differentiated shDDX3 THP1 cells. DDX3X targeting shRNA was expressed by induction with doxycycline (Dox) for 48 h and NLRP11 was targeted by siRNA transfection for 72 h, as indicated. Cells were infected with SeV for 16 h. IFNβ quantification by ELISA (left graph) and bioassay (right graph) are shown as means ± SEM (IFNβ: n = 5, SEAP: n = 3, ± SEM). *p < 0.05; **p < 0.01 Welch’s, two-sided t-test. Knock-down efficiency of NLRP11 and DDX3X was validated by endpoint PCR, and protein levels of DDX3X by immunoblot (upper panels). Immunoblots were probed for DDX3X and actin as loading control. n.s., not stimulated.

Figure 4

NLRP11 suppresses NLRP3 inflammasome activation by DDX3X. (A, B) iGLuc caspase-1 reporter assay in HEK293T cells expressing myc-NLRP3 (15 ng), HA-ASC, and caspase-1 (10 ng each) together with the indicated proteins. Cells were treated with 15 µM nigericin for 3 h. Means of three independent experiments (each conducted in triplicates) ± SEM are shown. *p < 0.05; **p < 0.01 Welch’s, two-sided t-test. (B), values relative to the mean of three independent experiments conducted in triplicates of cells, transfected with the reporter plasmid, NLRP3, ASC and caspase-1, not treated with nigericin, are shown. (C) Indirect immunofluorescent micrographs of HeLa-eGFP or HeLa-NLRP11-eGFP cells, induced with doxycycline overnight and transfected with HA-ASC and myc-NLRP3. Staining for HA and myc, as well as eGFP-signal is shown. Images representative of three independent experiments are shown. Scale bar = 50 µm. HA-ASC expressing cells without speck-formation are indicated with white arrow heads. Right panels: Quantification of cells with myc-NLRP3 specks and quantification of cells with HA-ASC staining (blinded counting of 150 cells per condition from n = 3). (D) iGLuc caspase-1 reporter assay in HEK293T cells expressing myc-NLRP3 (15 ng), HA-ASC, and caspase-1 (10 ng each) together with the indicated proteins. Cells were treated with 15 µM nigericin for 3 h. Means of three independent experiments (each conducted in triplicates) ± SEM are shown. (E) IL-1β release from macrophage-like differentiated THP1 cells after 72 h of siRNA-mediated knock-down. Cells were primed for 4 h with 100 ng/ml LPS followed by stimulation with 10 µM nigericin for 2 h. Means of five independent experiments ± SEM are shown. Inlay: endpoint PCR for NLRP11 and GAPDH of a representative experiment for validation of knock-down efficiency. n.s., not stimulated.

Figure 5

Schematic overview of the functional interplay of DDX3X and NLRP11 in innate immunity. The role of NLRP11 in attenuating RIG-I-induced type I interferon responses (left side) and in the activation of the NLRP3 inflammasome (right side) is depicted. NLRP11 interacts with DDX3X via its LRR domain. NLRP11 reduces type I interferon responses downstream of RIG-I activation by negative regulation of the activity of DDX3X in type I interferon induction. NLRP11 can also interfere with DDX3X-mediated NLRP3 inflammasome activation, leading to reduced IL-1β processing.