T cell responses in repeated controlled human schistosome infection compared to natural exposure

Abstract

In Schistosoma-endemic regions a lack of natural sterilizing immunity means individuals are repeatedly infected, treated and reinfected. Due to difficulties in tracking natural infection, kinetics of host immune response during these reinfections have not been elucidated. Here, we use repeated (3x) controlled-human-Schistosoma mansoni infection (CHI) to study how antigen-specific T cells develop during reinfection (NCT05085470 study). We compared these responses to naturally infected endemic Ugandan individuals (HALLMARK study). A mixed Th1/Th2/regulatory CD4⁺ T cell response develops in repeated CHI. Adult-worm-specific responses after repeated CHI were similar to endemic-natural infection. However, endemic participants showed differential responses to egg- and cercariae-antigens. Repeated CHI with sequential exposure to cercariae of different sexes (male-female-male) revealed an elevated CD4⁺ T cell cytokine response to adult-worm and egg-antigens. Our findings demonstrate that single-sex schistosome infection elicits adult-worm-specific T cell cytokine responses that reflect endemic-natural infection. This study advances our understanding of the immunology of schistosome (re)infection in the human host.

Fig. 1. Subtle alterations to T cell phenotype during repeat schistosome infection.

A Overview of trial design. Volunteers were randomised to reinfection (n = 12) or infection control (n = 12) groups. Reinfection volunteers (green) underwent three exposures (weeks 0, 9, and 18) and treatment (weeks 8, 17, and 30) cycles, whereas the infection control group (pink) underwent a single exposure (at week 18). At the second infection, n = 5 volunteers were unintentionally exposed to female cercariae (f) with groups referred to from then onwards as reinfection male-female-male (m-f-m, yellow) or reinfection male-male-male (m-m-m, blue). B Stacked bar chart of T cell subset frequencies. Each bar section represents the mean frequency. C Stacked bar chart of CD4 T cell memory frequencies. Each bar section represents the mean frequency. D–F Ribbon plots depicting mean (lines) frequency and standard error (shaded area) of the mean of specified populations, coloured by infection group. In panels (B–F), a two-sided linear mixed model was performed to (separately) assess changes in the reinfection and infection control group compared to the week 0 baseline, with the volunteer as a random effect. False discovery rate (FDR) corrected p values are displayed in black in figure C, and green (reinfection) and pink (infection control) in figure (D–F). *FDR <0.05, **FDR <0.01, ***FDR <0.001, ****FDR <0.0001. To compare between reinfection (all) and infection control groups, Welch’s T-tests were performed at each timepoint, with significant p values (uncorrected) displayed with a grey line *p < 0.05. G Ribbon plots depicting mean (lines) frequency and standard error (shaded area) of the mean of CD38and/or HLADR+ effector memory (EM) CD4 T cells in the reinfection (m-m-m) and reinfection (m-f-m) groups. To compare between reinfection (m-m-m) and reinfection (m-f-m) groups, two-sided Mann–Whitney tests were performed at each timepoint with significant p values (uncorrected) displayed with a grey line *p < 0.05, **p < 0.01. H Scatter plots comparing CD38 and/or HLADR+ EM CD4 T cell frequency to blood eosinophilia at week 26. Spearman’s rank correlation analysis was performed with the correlation coefficient rho and p value reported.

Fig. 2. Robust worm (AWA) but not cercariae (CA) - specific CD4 T cell responses during repeat schistosome infection.

A Heatmap displaying frequencies of cytokine-positive CD4+ ­T cells post AWA stimulation of PBMCs from reinfection volunteers, coloured by frequency. B–G Ribbon plots depicting mean (lines) frequency and standard error (shaded area) of the mean of cytokine production in AWA-stimulated CD4 T cells, coloured by infection group, comparing reinfection (all) and infection control. H Heatmap displaying frequencies of cytokine-positive CD4⁺ T cells post CA-stimulation of PBMCs from reinfection volunteers, coloured by frequency. I, J Ribbon plots depicting mean (lines) frequency and standard error (shaded area) of the mean of cytokine production in CA-stimulated CD4 T cells, coloured by infection group, comparing reinfection (all) and infection control. In panels (A–J), a two-sided linear mixed model was performed to (separately) assess changes in the reinfection and infection control group compared to the week 0 baseline, with the volunteer as a random effect. FDR values are displayed in black (panels A, B) or green (panels C–J) for reinfection (all) or red (infection control) *FDR <0.05, **FDR <0.01, ***FDR <0.001. To compare between reinfection (all, n = 11–12) and infection control groups (n = 11) two-sided Welch’s T-tests were performed at each timepoint with significant p values (uncorrected) displayed with a grey line *p < 0.05, **p < 0.01. **p < 0.001. K–N. Ribbon plots depicting mean (lines) frequency and standard error (shaded area) of the mean of specified stimuli and populations, coloured by infection group, comparing reinfection (m-f-m) and reinfection (m-m-m). To compare between reinfection (m-m-m, n = 7) and reinfection (m-f-m, n = 4–5) groups, two-sided Mann–Whitney tests were performed at each timepoint with significant p values (uncorrected) displayed with a grey line *p < 0.05. All frequencies shown are post medium subtraction.

Fig. 3. Enhanced inflammation at week 26 in the reinfection (m-f-m) group tends to reduce by week 30.

A Boxplot comparing CD38^and/orHLADR⁺ EM CD4 T cell frequency between reinfection (m-m-m) and reinfection (m-f-m) at week 26 and 30. B Boxplot comparing serum cytokines between reinfection (m-m-m) and reinfection (m-f-m) at weeks 26 and 30. C–E Boxplots comparing AWA-specific (C), SEA-specific (D), and CA-specific (E) CD4 T cell cytokine frequency between reinfection (m-m-m) and reinfection (m-f-m) at weeks 26 and 30. All frequencies shown are post medium subtraction. Differing shapes in reinfection (m-f-m) volunteers denote different individuals and are consistent between panels. Sample sizes are n = 6 for m-m-m at week 26, n = 7 at week 30, and n = 4 for m-f-m at both time points. Boxplots display the central line (median), and hinges (25th and 75th percentile) with whiskers extending from the hinge to the largest/smallest value or 1.5x the interquartile range. P values (uncorrected) are displayed, derived from two-sided Wilcoxon signed-rank tests when comparing between time points, or two-sided Mann–Whitney test to compare reinfection (m-m-m) and reinfection (m-f-m) groups. For all comparisons: *p < 0.05, **p < 0.01, ***p < 0.001.

Fig. 4. Comparing T cell phenotype and serum cytokines in controlled human infection and endemic infection.

A–-D Comparisons of week 30 reinfection male-male-male (m-m-m, n = 7), reinfection male-female-male (m-f-m, n = 4) and endemic schistosome infected (n = 9–12) and uninfected individuals (n = 13–16). Samples are coloured by group. A Principal component analysis (PCA) was applied to individual samples on the basis of frequencies of T cell populations. B Boxplots comparing T cell phenotype. C Boxplots comparing serum cytokine levels. D Boxplot comparing serum IL-10:TNF ratio. E–H Comparisons of schistosome uninfected and infected endemic (n = 9) individuals. Samples are coloured by group. E PCA applied to individual samples on the basis of frequencies of T cell populations. F Boxplots comparing T cell phenotype. G Boxplots comparing serum cytokine levels. H) Boxplot comparing serum IL-10:TNF ratio. Boxplots display the central line (median), and hinges (25th and 75th percentiles) with whiskers extending from the hinge to the largest/smallest value or 1.5x the interquartile range. P values are displayed, these are derived from two-sided Dunn’s test for pairwise multiple comparisons in (B–D) and from a two-sided Mann–Whitney test in (F–H). *p < 0.05, **p < 0.01, ***p < 0.001.

Fig. 5. Responses to different schistosome stages (AWA, SEA, CA) in controlled human infection and endemic infection.

A, B, D, E, G, H Comparisons of reinfection male-male-male (m-m-m, n = 6), reinfection male-female-male (m-f-m, n = 4), and endemic schistosome-infected (n = 9). Samples are coloured by group. C, F, I Comparisons of uninfected (n = 13) and schistosome-infected (n = 9) endemic individuals. Samples are coloured by group. A PCA applied to individual samples based on AWA-specific cytokine frequencies. B, C Boxplots comparing AWA-specific CD4 T cell cytokine frequencies. D PCA applied to individual samples based on SEA-specific cytokine frequencies. E, F Boxplots comparing SEA-specific CD4 T cell cytokine frequencies. G PCA applied to individual samples based on CA-specific cytokine frequencies. H, I Boxplots comparing CA-specific CD4 T cell cytokine frequencies. All boxplots display the central line (median), and hinges (25th and 75th percentile) with whiskers extending from the hinge to the largest/smallest value or 1.5x the interquartile range. All frequencies shown are post medium subtraction. P values are displayed, these are derived from two-sided Dunn’s test for pairwise multiple comparisons in (B, E, H), and from a two-sided Mann–Whitney test in (C, F, I). *p < 0.05, **p < 0.01, ***p < 0.001.