By Capturing Inflammatory Lipids Released from Dying Cells, the Receptor CD14 Induces Inflammasome-Dependent Phagocyte Hyperactivation

Abstract

A heterogeneous mixture of lipids called oxPAPC, derived from dying cells, can hyperactivate dendritic cells (DCs) but not macrophages. Hyperactive DCs are defined by their ability to release interleukin-1 (IL-1) while maintaining cell viability, endowing these cells with potent aptitude to stimulate adaptive immunity. Herein, we found that the bacterial lipopolysaccharide receptor CD14 captured extracellular oxPAPC and delivered these lipids into the cell to promote inflammasome-dependent DC hyperactivation. Notably, we identified two specific components within the oxPAPC mixture that hyperactivated macrophages, allowing these cells to release IL-1 for several days, by a CD14-dependent process. In murine models of sepsis, conditions that promoted cell hyperactivation resulted in inflammation but not lethality. Thus, multiple phagocytes are capable of hyperactivation in response to oxPAPC, with CD14 acting as the earliest regulator in this process, serving to capture and transport these lipids to promote inflammatory cell fate decisions.

Keywords: SMOC; Toll-like Receptor; dendritic cells; hyperactivation; inflammasome; innate immunity; interleukin-1; macrophages; oxPAPC.

Figure 1. oxPAPC-Induced CD14 Endocytosis Prevents LPS Signaling

(A) Indicated immortal macrophage lines were treated with LPS (1 μg/mL) or oxPAPC (50 μM) for the indicated times. Surface levels of CD14 and TLR4 were measured by flow cytometry. Line graphs represent means and standard deviations of two biological repeats. (B) Primary DCs and macrophages were treated with oxPAPC (50 μM) for the indicated times. Surface levels of CD14 and TLR4 and TLR4 dimerization were measured by flow cytometry. Line graphs represent means and standard deviations of two independent experiments. (C) Primary DCs were treated, or not, with LPS (1 μg/mL) or oxPAPC (50 μM) for 15 min in the presence or absence of the PLCγ inhibitor U7322 (5 μM) or the Syk inhibitor Piceatannol (75 μM). Surface levels of CD14 were measured by flow cytometry. Bar graphs represent means and standard deviations of two independent experiments. (D) Immortal macrophages were treated with LPS or oxPAPC for times indicated. Activation of the type I IFN (p-STAT1 and p-IRF3), NF-κB (p-p65), and MAP kinase (p-p38 and p-ERK1/2) pathways were determined by western analysis. (E) Immortal macrophages were treated with oxPAPC (at the indicated concentration). Surface levels of TLR4 (left) and TLR4 dimerization (right) were measured by flow cytometry. Line graphs represent means and standard deviations of biological duplicates from one experiment representative of three. (F) Immortal macrophages were pretreated with indicated concentrations of oxPAPC for 30 min followed by LPS stimulation (10 ng/mL). 4 hr after LPS stimulation, cells were lysed and the activation of the type I IFN pathway were determined by western analysis for p-STAT1 and Viperin. See also Figure S1.

Figure 2. The Same Amino Acids within CD14 Detect oxPAPC and LPS

(A) Schematic of the four regions (R1–R4) used by LPS to induce CD14 internalization. (B) oxPAPC binding capacity of the CD14 mutants indicated was determined by biotinylated oxPAPC pull down assay. Lysates of 293T cells expressing the indicated CD14 mutants were incubated with biotinylated oxPAPC (10 μg). CD14-oxPAPC complexes were captured using neutravidin beads. The amount of CD14 retained by oxPAPC was determined by western analysis. The CD14 mutant with 26DEES₂₉ mutagenized into 26AAAA₂₉ was designated as CD14 1R. The CD14 mutant with 26DEES₂₉ and 37PKPD₄₀ mutagenized into 26AAAA₂₉ and 37AAAA₄₀ was designated as CD14 2R. The CD14 mutant with 26DEES₂₉, 37PKPD₄₀, 52DVE₅₄, and 74DLGQ₇₇ mutagenized into 26AAAA₂₉, 37AAAA₄₀, 52AAA₅₄, and 74AAAA₇₇ was designated as CD14 4R. (C) The 4R CD14 mutant is not internalized in response to oxPAPC or LPS treatments. Indicated macrophage lines were treated with LPS (1 μg/mL) or oxPAPC (50 μM) for the times indicated. Surface levels of CD14 were measured by flow cytometry. Line graph represents the average and error bars represent the standard deviation of biological duplicates from one representative experiment of three. Statistical significance of WT versus CD14 4R cells is shown for each treatment. (D) Lysates of 293T cells expressing CD14 were pretreated with indicated concentrations of oxPAPC (top) or LPS (bottom) for 30 min and then incubated with biotinylated LPS (1 μg/mL) (top) or biotinylated oxPAPC (20 μg/mL) (bottom). CD14-LPS (top) or CD14-oxPAPC (bottom) complexes were captured using neutravidin beads. The amounts of CD14 retained by biotinylated LPS (top) or biotinylated oxPAPC (bottom) were determined by western analysis. Representative blots from at least two independent experiments were shown. See also Figure S2.

Figure 3. CD14 Promotes Caspase-11-Mediated IL-1b and IL-18 Release from DCs

(A and B) WT DCs and CD14-deficient DCs were treated with LPS (1 μg/mL), Pam3CSK (P3C) (1 μg/mL), or oxPAPC (120 μM) or were primed with LPS or P3C for 3 hr and then treated with oxPAPC. 18 hr after LPS (A) or P3C (B) administration, IL-1β and TNFα secretion was measured by ELISA. Means and standard deviations of two independent experiments are shown. (B) Spleen-derived WT DCs, CD14-deficient DCs, and Caspase-11-deficient DCs were treated with LPS (1 μg/mL), ATP (1.5 mM), or oxPAPC (120 μM) or were primed with LPS for 3 hr and then treated with oxPAPC or ATP. 18 hr after LPS administration, IL-1β and TNFα secretion was measured by ELISA. Means and standard deviations of two replicates of one representative experiment of three are shown. (C) WT DCs, CD14-deficient DCs, and Caspase-11-deficient DCs were treated with LPS (1 μg/mL), oxPAPC (120 μM), or ATP (1 mM) or were primed with LPS for 3 hr and then treated with oxPAPC or ATP. IL-18 was measured by ELISA in the supernatant 18 hr later. (D) Spleen-derived WT DCs, CD14-deficient DCs, and Caspase-11-deficient DCs were treated with LPS (1 μg/mL), ATP (1.5 mM), or oxPAPC (120 μM) or were primed with LPS for 3 hr and then treated with oxPAPC or ATP. 18 hr after LPS administration, IL-1β (left) and TNFα (right) secretion was measured by ELISA. Means and standard deviations of two replicates of one representative experiment of three are shown. (E and F) WT and CD14-deficient DCs were either primed with LPS (1 μg/mL) for 3 hr and then treated with ATP (1 mM) or were electroporated with LPS (1 μg/mL) or MOCK electroporated. IL-1 (E) and LDH (F) release was monitored 18 hr later. See also Figure S3.

Figure 4. CD14 Transports oxPAPC into the Cell to Promote Inflammasome Activities

(A) WT DCs and CD14-deficient DCs were treated with TopFluor-PGPE (5 μg/mL, green) and Alexa568-labeled dextran (200 μg/mL, red) for 1 hr. Cell were imaged by confocal microscopy. Fluorescent images and quantification are representative of three independent experiments. Scale bar: 10 μm. (B and C) WT DCs and CD14-deficient DCs were treated with LPS (1 μg/mL), P3C (1 μg/mL), or oxPAPC (120 μM) or were primed with P3C for 3 hr and then treated with oxPAPC. As indicated, LPS or oxPAPC were complexed with DOTAP before addition to the cell culture. Where indicated, cells where treated with the pan-Caspase inhibitor zVAD 30 min before addition of the DOTAP/oxPAPC complex to the culture. 18 hr after stimuli administration, secreted (B) or cell-associated (C) IL-1β were measured by ELISA. Means and standard deviations of two independent experiments are shown. See also Figure S4.

Figure 5. Essential and Possibly Redundant Regulators of oxPAPC-Dependent Inflammasome Activities in DCs

(A) DCs derived from the indicated backgrounds were primed with LPS (1 μg/mL) and stimulated with oxPAPC (120 μM). Their capacity to secrete IL-1β was compared with WT DCs stimulated in the same manner. Percentages of remaining IL-1β compared to WT DCs are shown. N: number of independent measurement analyzed for each genetic background. Means and standard errors are shown. (B) WT DCs and CD14-deficient DCs were primed or not with P3C and activated or not with oxPAPC (120 μM). Supernatant (Sup) of non-treated (NT) or P3C-primed (P3C) cells were incubated with a biotinylated antibody directed against IL-1β (Bio-αIL-1β). IL-1β associated with the biotinylated antibody was captured by streptavidin beads and revealed by western analysis. Cell lysates (Lysate) were probed with the indicated antibodies. Shown is a representative blot out of three independent experiments. (C) FITC-Dextran uptake by WT DCs and CD14-deficient DCs was measured over time by cytofluorimetry. Line graphs represent the average and error bars represent the standard deviation of six independent experiments. See also Figure S5.

Figure 6. CD14 Promotes DC and Macrophage Hyperactivation in Response to oxPAPC Components

(A and B) WT or CD14-deficient DCs (A) or macrophages (B) were activated (or not) for 3 hr with LPS and stimulated, or not, with oxPAPC (120 μM), POVPC (120 μM), PGPC (120 μM). 18 hr after LPS administration, IL-1β (top) and TNFα (bottom) secretion was measured by ELISA. Means and standard deviations of two replicates of one representative experiment of three are shown. (C and D) WT or CD14-deficient DCs (A) or macrophages (B) were activated (or not) for 3 hr with LPS and stimulated, or not, with oxPAPC (120 μM), POVPC (120 μM), PGPC (120 μM). 18 hr after LPS administration, processed IL-1β in the tissue culture supernatants from DCs (C) or macrophages (D) was captured by a biotinylated IL-1β antibody followed by western analysis. Expression levels of cell associated pro-IL-1 β and NLRP3 in DCs (C) or macrophages (D) were determined by western analysis. (E) Lysates of 293T cells expressing CD14 were pretreated with indicated concentrations of POVPC or PGPC for 30 min and then incubated with biotinylated LPS (1 μg). CD14-LPS complex was captured using neutravidin beads. The amount of CD14 retained by LPS in the presence of POVPC or PGPC was determined by western analysis. Representative blots from at least two independent experiments were shown. (F) WT DCs or macrophages were treated with oxPAPC (120 μM), POVPC (120 μM), and PGPC (120 μM) for indicated time periods. Surface levels of CD14 from DCs or macrophages were measured by flow cytometry. Line graphs with standard deviations represent the average of duplicate readings from one representative experiment out of three. (G) WT, Caspase1/11 double-deficient, and NLRP3-deficient macrophages were treated with LPS or not for 3 hr, and then treated, or not, with oxPAPC (120 μM), POVPC (120 μM), PGPC (120 μM), and DMPC (120 μM). IL-1β secretion was measured by ELISA. Means and standard deviations of two replicates of one representative experiment out of three are shown. See also Figure S6.

Figure 7. Conditions that Hyperactivate Phagocytes Induce Inflammatory Sepsis in Mice, but Not Death

(A) Schematic representation of the in vivo experimental setting. PLs, phospholipids. (B–D) ELISA of serum cytokines from mice treated as in (A) 2 hr after challenge. (E) Survival curves of mice treated as in (A). (F and G) ELISA of serum cytokines from mice treated as in (A) 30 hr after challenge.