Modulatory actions of Echinococcus granulosus antigen B on macrophage inflammatory activation

Abstract

Cestodes use own lipid-binding proteins to capture and transport hydrophobic ligands, including lipids that they cannot synthesise as fatty acids and cholesterol. In E. granulosus s.l., one of these lipoproteins is antigen B (EgAgB), codified by a multigenic and polymorphic family that gives rise to five gene products (EgAgB8/1-5 subunits) assembled as a 230 kDa macromolecule. EgAgB has a diagnostic value for cystic echinococcosis, but its putative role in the immunobiology of this infection is still poorly understood. Accumulating research suggests that EgAgB has immunomodulatory properties, but previous studies employed denatured antigen preparations that might exert different effects than the native form, thereby limiting data interpretation. This work analysed the modulatory actions on macrophages of native EgAgB (nEgAgB) and the recombinant form of EgAg8/1, which is the most abundant subunit in the larva and was expressed in insect S2 cells (rEgAgB8/1). Both EgAgB preparations were purified to homogeneity by immunoaffinity chromatography using a novel nanobody anti-EgAgB8/1. nEgAgB and rEgAgB8/1 exhibited differences in size and lipid composition. The rEgAgB8/1 generates mildly larger lipoproteins with a less diverse lipid composition than nEgAgB. Assays using human and murine macrophages showed that both nEgAgB and rEgAgB8/1 interfered with in vitro LPS-driven macrophage activation, decreasing cytokine (IL-1β, IL-6, IL-12p40, IFN-β) secretion and ·NO generation. Furthermore, nEgAgB and rEgAgB8/1 modulated in vivo LPS-induced cytokine production (IL-6, IL-10) and activation of large (measured as MHC-II level) and small (measured as CD86 and CD40 levels) macrophages in the peritoneum, although rEgAgB8/1 effects were less robust. Overall, this work reinforced the notion that EgAgB is an immunomodulatory component of E. granulosus s.l. Although nEgAgB lipid's effects cannot be ruled out, our data suggest that the EgAgB8/1 subunit contributes to EgAgB´s ability to regulate the inflammatory activation of macrophages.

Keywords: Echinococcus granulosus; HLBP family; antigen B; cestodes; immunomodulation; lipoprotein; macrophage.

Figure 1

Characterization of anti-EgAgB VHHs/VHs clones: binding to different EgAgB8 recombinant subunits and soluble expression. (A) The supernatants of anti-EgAgB VHH/VHs clones were assessed for binding to ELISA wells coated with drEgAgB8/1-3 (0.5 µg/well). Note that no reaction was detected for drEgAgB8/2 and drEgAgB8/3. (B) SDS-PAGE analysis of soluble (S) and insoluble (I) fractions of clones 1, 3 and 7. Samples were run on a 12.5% polyacrylamide gel and protein bands were stained using Coomassie Blue staining.

Figure 2

Immunoaffinity yielded pure nEgAgB and rEgAgB8/1 preparations. Analysis by SDS-PAGE (A) and Western Blot (B) of the fractions corresponding to the purification protocol of nEgAgB from HF: bound (QS+) and unbound (QS-) fractions to Q-Sepharose, the low-density fraction (Ld_f) obtained by ultracentrifugation of QS+ on a KBr density gradient, and the fractions obtained by immunoaffinity of Ld_f on anti-EgAgB clone1–Sepharose. Notice the typical EgAgB pattern with regularly spaced bands. Analysis by SDS-PAGE (C) and Western Blot (D) of the fractions corresponding to the purification protocol of rEgAgB8/1 from the supernatant collected at 7 dpi: the fraction bound to Strep-Tactin XT-agarose column (StrepTactin+) and those obtained by its subsequent fractionation on anti-EgAgB clone1–Sepharose. In all cases samples were run on a 15% polyacrylamide gel under reducing conditions (6 mM DTT) and protein bands were detected by Colloidal Coomassie Blue staining. Western blot was performed using the mAb EB7. FT, flow through; W_NaCl, washes with 1M NaCl; W_pH5, washes using a pH5.

Figure 3

rEgAgB8/1 is a complex lipoprotein slightly higher than nEgAgB in size. (A) Determination of the hydrodynamic radius (R_H) by DLS. Samples were analysed in triplicates and results are shown as the volume distribution (%) plotted against the hydrodynamic radius (R_H). Results are representative of four and two independent batches of nEgAgB and rEgAgB8/1, respectively. Both samples showed a monomodal size distribution of particles. The mean R_H value obtained is indicated. (B, C) Analysis of the protein and lipid content by native gel electrophoresis, using Coomassie Brilliant Blue or Sudan Black for visualization, respectively. nEgAgB and rEgAgB8/1 were analysed in parallel to HDL₂, HDL₃, LDL and murine and human plasma samples. High amounts of plasma samples were analysed in order to detect lipid components. Results showed that nEgAgB and rEgAgB8/1 are complex lipoproteins, with a principal component of around 7 and 12 nm in diameter, respectively. MW, Molecular weight standard.

Figure 4

The content of neutral and polar lipids of rEgAgB8/1 showed differences with that of nEgAgB. A quantitative lipid analysis was performed by HPLC-MS employing ACE C4 and ZICpHILIC columns for characterising the FAs and polar lipids, respectively. The 50 most abundant (A) neutral and (B) polar lipids are represented as heatmaps. For each heatmap, the relative content of each lipid species was normalised by the total amount of detected lipids. Data correspond to the analysis of nEgAgB prepared from seven hydatids and rEgAgB8/1 prepared from two independent productions, and are expressed as the mean of technical triplicates for neutral and duplicates for polar lipids.

Figure 5

nEgAgB and rEgAgB8/1 interfered with the LPS-induced production of cytokines and nitrite by macrophages. THP-1 and BMDM macrophages were stimulated with nEgAgB, rEgAgB8/1 or vehicle (PBS_EBAb) in the absence or presence of LPS. After stimulation for 24 hours cytokine secretion and nitrite (indicative of ·NO production) levels were determined in cell supernatants. The levels of (A) IL-1β, (B) IL-6, (C) IFN-β for THP-1 macrophages, and (D) IL-12p40, (E) IL-6 and (F) nitrite for BMDM are plotted as the mean ± SEM of three independent experiments with analytical triplicates. Data showing significant differences from LPS are indicated with * (two-way ANOVA and Tukey’s test, *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001).

Figure 6

nEgAgB and rEgAgB8/1 did not alter CD40, CD86 or MHC-II expression in BMDM, in the presence or absence of LPS. BMDM were stimulated with nEgAgB, rEgAgB8/1 or vehicle (PBS_EBAb) in the absence or presence of LPS. After 24 hours, MHC-II, CD86 and CD40 expression was measured by flow cytometry. (A-C) Representative histograms of the fluorescence intensity (FI) obtained for MHC-II, CD86 and CD40. (D-F) Bar graphs represent MHC-II, CD86 and CD40 surface expression (mean FI ± SEM) corresponding to two independent experiments with analytical triplicates (two-way ANOVA and test of Tukey, *p < 0.05, **p < 0.01).

Figure 7

nEgAgB and rEgAgB8/1 altered the cytokine production induced by LPS in the peritoneal cavity. nEgAgB or rEgAgB8/1 (50 µg/mouse) and PBS_EBAb (vehicle control) were i.p. injected in Balb/c mice in the absence or presence of LPS (15 µg/mouse). Graphs show the peritoneal levels of IL-6 (A, D), IL-12p40 (B, E) and IL-10 (C, F) after 4 hpi for 3 and 2 independent experiments with nEgAgB and rEgAgB8/1, respectively. Each point in the graphs represents an individual (n=5 per group). The median value for each independent experiment is shown as a thin and short horizontal line while the median value corresponding to the set of independent experiments is shown as a thick and long horizontal line. The asterisks denote differences between groups (Mack Skillings two-way non-parametric exact test, followed by the Conover post hoc multiple-comparison test with Benjamini and Hochberg correction (*p < 0.05, **p < 0.01, ***p < 0.001).

Figure 8

nEgAgB modulated the LPS-induced surface expression of MHC-II, but not of CD86 and CD40 in LPM; rEgAgB8/1 did not provoke alterations of any of these surface molecules. nEgAgB or rEgAgB8/1 (50 µg/mouse) and PBS_EBAb (vehicle control) were i.p. injected in Balb/c mice in the absence or presence of LPS (15 µg/mouse). Peritoneal cells were collected after 24 hpi and analysed by flow cytometry. Graphs show the surface expression of MHC-II (A, D), CD86 (B, E) and CD40 (C, F) in LPM (defined as CD19-Ly6C-F4/80++ cells) corresponding to 3 and 2 independent experiments using nEgAgB and rEgAgB8/1, respectively. Data is presented as the FI normalised to the vehicle control. Each point in the graphs represents an individual (n=5 per group). The median value for each independent experiment is shown as a thin and short horizontal line while the median value corresponding to the set of independent experiments is shown as a thick and long horizontal line. The asterisks denote differences between groups (Mack Skillings two-way non-parametric exact test, followed by the Conover post hoc multiple-comparison test with Benjamini and Hochberg correction (*p < 0.05, **p < 0.01, ***p < 0.001).

Figure 9

nEgAgB modulated the early induction of CD86 and CD40 promoted by LPS in SPM while rEgAgB8/1 only affected CD86 expression. nEgAgB or rEgAgB8/1 (50 µg/mouse) and PBS_EBAb (vehicle control) were i.p. injected in Balb/c mice in the absence or presence of LPS (15 µg/mouse). Peritoneal cells were collected after 4 hpi and analysed by flow cytometry. Graphs show the surface expression of MHC-II (A, D), CD86 (B, E) and CD40 (C, F) in SPM (defined as (CD19⁻F4/80^+/-SSC^lowLy6C^-MHC-II⁺⁺ cells) corresponding to 3 and 2 independent experiments using nEgAgB and rEgAgB8/1, respectively. Data is presented as the FI normalised to the vehicle control. Each point in the graphs represents an individual (n=5 per group). The median value for each independent experiment is shown as a thin and short horizontal line while the median value corresponding to the set of independent experiments is shown as a thick and long horizontal line. The asterisks denote differences between groups (Mack Skillings two-way non-parametric exact test, followed by the Conover post hoc multiple-comparison test with Benjamini and Hochberg correction (*p < 0.05, **p < 0.01, ***p < 0.001).

Figure 10

nEgAgB effects on LPS-induced CD86 and CD40 expression in SPM were sustained after 24 hpi. nEgAgB or rEgAgB8/1 (50 µg/mouse) and PBS_EBAb (vehicle control) were i.p. injected in Balb/c mice in the absence or presence of LPS (15 µg/mouse). Peritoneal cells were collected after 24 hpi and analysed by flow cytometry. Graphs show the surface expression of MHC-II (A, D), CD86 (B, E) and CD40 (C, F) in SPM (defined as (CD19⁻F4/80^+/-SSC^lowLy6C^-MHC-II⁺⁺ cells) corresponding to 3 and 2 independent experiments using nEgAgB and rEgAgB8/1, respectively. Data is presented as the FI normalised to the vehicle control. Each point in the graphs represents an individual (n=5 per group). The median value for each independent experiment is shown as a thin and short horizontal line while the median value corresponding to the set of independent experiments is shown as a thick and long horizontal line. The asterisks denote differences between groups (Mack Skillings two-way non-parametric exact test, followed by the Conover post hoc multiple-comparison test with Benjamini and Hochberg correction (*p < 0.05, **p < 0.01, ***p < 0.001).