Helminth extracellular vesicles co-opt host monocytes to drive T cell anergy

Abstract

Parasitic helminths secrete extracellular vesicles (EVs) into their host tissues to modulate immune responses, but the underlying mechanisms are poorly understood. We demonstrate that Ascaris EVs are efficiently internalised by monocytes in human peripheral blood mononuclear cells and increase the percentage of classical monocytes. Furthermore, EV treatment of monocytes induced a novel anti-inflammatory phenotype characterised by CD14⁺, CD16^-, CC chemokine receptor 2 (CCR2^-) and programmed death-ligand 1 (PD-L1)⁺ cells. In addition, Ascaris EVs induced T cell anergy in a monocyte-dependent mechanism. Targeting professional phagocytes to induce both direct and indirect pathways of immune modulation presents a highly novel and efficient mechanism of EV-mediated host-parasite communication. Intra-peritoneal administration of EVs induced protection against gut inflammation in the dextran sodium sulphate model of colitis in mice. Ascaris EVs were shown to affect circulating immune cells and protect against gut inflammation; this highlights their potential as a subject for further investigation in inflammatory conditions driven by dysregulated immune responses. However, their clinical translation would require further studies and careful consideration of ethical implications.

Keywords: Ascaris; colitis; extracellular vesicles; helminth; host‐parasite interaction; immune modulation; inflammatory bowel disease; monocytes.

FIGURE 1

Characterization of Ascaris suum EVs. (a). CryoTEM images of purified Ascaris suum EVs (40,000× magnification) showing EVs of various sizes with the presence of a visible corona. All scale bars are 100 nm. (b). Size distribution profiles of unlabelled EVs and DOPE‐Rho‐labelled A. suum EVs acquired with NTA. The graph displays the average concentration (solid lines) and size distribution of particles (Blue: unlabelled, and red: DOPE‐Rho labelled). The standard error of the mean is expressed as shaded areas around the curves. EV, extracellular vesicles; CryoTEM, cryogenic transmission electron microscopy; NTA, nanoparticle tracking analysis.

FIGURE 2

Ascaris suum EVs are taken up and internalised by monocytes. (a). Human PBMCs were stimulated with DOPE‐Rho labelled EVs (30,000 particles/cell) and analysed by flow cytometry. UT cells and cells stimulated with unlabelled EVs were included as negative controls. The bar chart shows the percentage of EV uptake by cell type. The groups were compared using one‐way ANOVA followed by a Tukey test. ****p < 0.0001. Error bars: Mean ± SD. n = 3 donors. (b, c and d). PBMCs incubated with DOPE‐Rho labelled EVs (30,000 particles/cell) were analysed by imaging flow cytometry. Cell surface markers for T cells (CD3⁺), B‐cells (CD19⁺) and monocytes (CD14⁺) were detected to analyse EV uptake (DOPE‐Rho⁺) in PBMCs. Monocytes (CD14⁺ cells) were significantly more positive for DOPE‐Rho compared to the other cell types and to the UT and unlabelled EVs. BF = Brightfield. (e). Representative images of high and low internalisation scores. Images of one cell are shown horizontally in the brightfield channel without and with the AdaptiveErode (M01, Ch01, 75), CD14 APC, DOPE‐Rho channel without and or with the mask displayed and an image with CD14APC and DOPE‐Rho merged. The internalisation score is shown as yellow numbers on the merged images. (f). Accumulated internalisation score for all CD14⁺, DOPE‐Rho⁺ cells plotted against the normalised frequency. The internalisation of DOPE‐Rho EVs in CD14⁺ monocytes was determined using the mask shown in e. Created with IDEAS Software. EV, extracellular vesicles; PBMCs, peripheral blood mononuclear cells; UT, untreated.

FIGURE 3

Ascaris suum EVs reduced the percentage of CD14+, CD16+ cells, increased CD14+, CD16− cells, and modulated the expression of PD‐L1 and CCR2. (a) and (b). Human PBMCs stimulated with EVs (A) or EV‐depleted ESP (B), and the subpopulations of monocytes (classical: CD14⁺, CD16⁻; intermediate: CD14⁺, CD16⁺; non‐classical monocytes: CD14⁻, CD16⁺) were analysed by flow cytometry. A group containing all MNCs represents the combined subpopulations. Exposure of monocytes to increasing numbers of particles/cell was undertaken; the EV‐depleted ESP was normalised by protein concentration to the equivalent number of EVs for comparative purposes. The (UT) cells were incubated with culture media alone. The groups were compared using two‐way ANOVA followed by a Tukey test. *p < 0.05, **p < 0.01, ***p < 0.001. Error bars: Mean ± SD. n = 3 donors. (c) and (d). The MFI of PD‐L1 (c) and CCR2 (d) expression on all MNCs was analysed by flow cytometry on human PBMCs stimulated with either EVs or EV‐depleted ESP. The groups were compared using two‐way ANOVA followed by a Tukey test. *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001. Error bars: Mean ± SD. n = 3 donors. (e). Dotplot of PD‐L1 and CCR2 co‐expression in one representative donor untreated, stimulated with EVs or EV‐depleted ESP. ESP, excretory‐secretory products; EV, extracellular vesicles; MFI, median fluorescence intensity; MNCs, mononuclear cells; PBMC, peripheral blood mononuclear cells; PD‐L1, programmed death‐ligand 1.

FIGURE 4

Removal of N‐linked glycans on the surface of Ascaris suum EVs did not alter their effect on human PBMCs. (a) and (b). AFM micrographs of untreated (a) and 400 U/mL PNGase F treated (b) Ascaris suum EVs. Untreated EVs appear to be decorated by a diffuse corona of filaments, which mostly disappears after PNGase treatment. the process does not compromise the structural integrity of EVs. Further details and quantitative measurements can be found in Figure S‐10. (c). Distribution of the glycoprotein corona thickness as measured via AFM on individual untreated (blue) and PNGase‐treated A. suum EVs (green). (d) and (e). The effect of PNGase and untreated EVs on the monocyte subgroups (Classical: CD14⁺, CD16⁻, Intermediate: CD14⁺, CD16⁺, Non‐classical: CD14⁻, CD16⁺) or the MFI of PD‐L1 (C) or CCR2 (D) were similar to untreated EVs. EVs were given in particles/cell doses (1000, 3500, or 10,000 particles/cell). The UT was culture media alone. Error bars: Mean ± SD. n = 3 donors. AFM, atomic force microscopy; EV, extracellular vesicles; MFI, median fluorescence intensity; UT, untreated cells.

FIGURE 5

Proteolytic inhibition partially recovered EV‐dependent loss of CD16 expression. (a), (b) and (c). Human PBMCs were stimulated with Ascaris suum EVs (30,000 per cell) and two different concentrations of the metalloprotease inhibitor TAPI‐0 (5 µM and 20 µM) and analysed using flow cytometry. UT cells and EV‐depleted ESP‐treated cells were included as controls. The percentage of CD14+/CD16+ cells (a) or the MFI of CD16 (b) after stimulation with the different treatments was assessed. The MFI of CCR2 expression after stimulation with the different treatments (c). The groups were compared using one‐way ANOVA followed by a Tukey test. *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001. Error bars: Mean ± SD. n = 3 donors. ESP, excretory‐secretory products; EV, extracellular vesicles; MFI, median fluorescence intensity; PBMCs, peripheral blood mononuclear cells; UT untreated.

FIGURE 6

Ascaris suum EVs reduce T cell activity in a monocyte‐dependent manner. (a) and (b). Human PBMCs were treated with anti‐CD3/anti‐CD28 after 16 h of incubation with EVs (1,000, 3,500, and 10,000 particles/cell) or the equivalent protein concentration of EV‐depleted ESP. Cytokines were analysed either 24 h post‐stimulation (HPS) (a) or 48 HPS (b). UT cells were included as a control. Before the statistical analysis, the data were normalised to the UT group due to the large donor‐to‐donor variation. Afterwards, the groups were compared using Two‐way ANOVA followed by a Tukey test. *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001, † = Significant different from the corresponding EV‐depleted dose. Error bars: Mean ± SD. n = 3 donors. (c) and (d). PBMCs were depleted of CD14⁺ cells before being stimulated with EVs (3500 and 10,000 particles/cell) and subsequently activated with anti‐CD3/anti‐CD28. Cytokines were analysed either 24 HPS (c) or 48 HPS (d). UT cells and non‐depleted PBMCs from the same donor were included as a control. Before the statistical analysis, the data were normalised to the UT group due to the large donor‐to‐donor variation. Afterwards, the groups were compared using two‐way ANOVA followed by a Tukey test. *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001. Error bars: Mean ± SD. n = 3 donors. ESP, excretory‐secretory products; EV, extracellular vesicles; MFI, median fluorescence intensity; PBMCs, peripheral blood mononuclear cells; UT untreated.

FIGURE 7

Mitigating effects of Ascaris suum EVs on DSS colitis. (a). Mice received three i.p. injections on days 0, 2, and 5 (20 µg per injection for the Ascaris suum EV treatment group). Starting from day 0, mice were subjected to 2. 5% (w/v) DSS in their drinking water until euthanised on day 8, with continuous recording of body weight and clinical scores. (b). Relative weight gain/loss for mice during the study. Statistical comparisons between the treatment and PBS vehicle groups were conducted using the Mann‐Whitney test; * p < 0.1; (c). Cumulative clinical scores encompassing piloerection, faeces consistency, and rectal thickening/injury on days 0, 6, 7 and 8 of the study. Statistical comparisons between the treatment and PBS vehicle groups were conducted using the Kruskal‐Wallis test; * p < 0.05; ** p < 0.01. EV, extracellular vesicles i.p, intraperitoneali.