A Small Intestinal Helminth Infection Alters Colonic Mucus and Shapes the Colonic Mucus Microbiome

Abstract

Infecting humans with controlled doses of small intestinal helminths, such as human hookworm, is proposed as a therapy for the colonic inflammatory disease ulcerative colitis. Strengthening the colonic mucus barrier is a potential mechanism by which small intestinal helminths could treat ulcerative colitis. In this study, we compare C57BL/6 mice infected with the small intestinal helminth Heligmosomoides polygyrus and uninfected controls to investigate changes in colonic mucus. Histology, gene expression, and immunofluorescent analysis demonstrate that this helminth induces goblet cell hyperplasia, and an upregulation of mucin sialylation, and goblet-cell-derived functional proteins resistin-like molecule-beta (RELM-β) and trefoil factors (TFFs), in the colon. Using IL-13 knockout mice, we reveal that these changes are predominantly IL-13-dependent. The assessment of the colonic mucus microbiome demonstrates that H. polygyrus infection increases the abundance of Ruminococcus gnavus, a commensal bacterium capable of utilising sialic acid as an energy source. This study also investigates a human cohort experimentally challenged with human hookworm. It demonstrates that TFF blood levels increase in individuals chronically infected with small intestinal helminths, highlighting a conserved mucus response between humans and mice. Overall, small intestinal helminths modify colonic mucus, highlighting this as a plausible mechanism by which human hookworm therapy could treat ulcerative colitis.

Keywords: helminth; hookworm; intestinal barrier; microbiome; mucus; trefoil factor.

Figure 1

H. polygyrus induces pan-intestinal goblet cell hyperplasia. (A) Percentage of PAS-stained area in colonic sections from uninfected mice (black) and mice infected with 200 H. polygyrus larvae (pink) for one and two weeks, n = 6–7 from two experiments. (B) Percentage of sections positively stained by PAS in the jejunum and colon of uninfected mice and mice infected with H. polygyrus for two weeks, n = 6–8 from two experiments. (C) Representative images from Periodic acid Schiff (PAS) stained jejunum and colon from uninfected and infected mice. Goblet cells-stained magenta. Scale bar = 100 μm. (D) Colonic mucus thickness measured in PAS/Alcian Blue stained sections in uninfected and infected mice, n = 7–8 from two experiments. (E) Relative differences in mRNA expression of gel-forming mucins (Muc2, Muc5ac, Muc5b, and Muc6) in the jejunum and colon between uninfected and infected mice. Rplp0 was used as a housekeeping gene. n = 8–10 from two experiments. Data were analysed with unpaired student t-test (A,D,E) or ANOVA with post-hoc Dunnett’s test (C). Mean ± SEM. ns, not significant; **, p < 0.01.

Figure 2

H. polygyrus induces pan-intestinal alterations in goblet cell-derived proteins and mucin sialylation. (A) Relative differences in mRNA expression of trefoil factors (Tff1, Tff2, and Tff3) and Relm-β in the jejunum and colon of uninfected mice and mice infected with 200 H. polygyrus larvae for two weeks, n = 8–10 from two experiments. (B) Representative immunofluorescence image staining for TFF1 (red) and TFF2 (red) in the jejunum and colon, respectively. DAPI = blue; green = autofluorescence. 40× magnification. (C) Relative differences in mRNA expression of sialyltransferases (St3gal3 and St6galnac1) and sulphotransferases (Chst-4 and Gal3st2) in the jejunum and colon of uninfected and infected mice, n = 8–10 from two experiments. Rplp0 or Gapdh was used as a housekeeping gene. Data were analysed with unpaired student t-test. Mean ± SEM. *, p < 0.05; **, p < 0.01.

Figure 3

Goblet cell hyperplasia, upregulation of goblet-cell derived proteins, and colonic mucin sialylation are sustained in the chronic phase of an H. polygyrus infection. (A) Percentage of sections positively stained by PAS in the jejunum and colon of uninfected mice (black) and mice infected with H. polygyrus (pink) for six weeks, n = 6–9 from two experiments. (B,C) Relative difference in mRNA expression in the jejunum (B) and colon (C) between uninfected and infected mice six weeks post infection. Only genes demonstrating a significant difference in the acute infection phase were assessed. n = 8–10 from two experiments. Rplp0 or Gapdh was used as a housekeeping gene. Data were analysed with unpaired student t-test. Mean ± SEM. *, p < 0.05.

Figure 4

H. polygyrus induces a sustained alteration in the composition of the colonic mucosa-associated microbiome. (A,B) No significant differences in β-diversity (PERMANOVA, p > 0.05) (A) or α-diversity (paired Wilcoxon test, p > 0.05) (B) were observed between uninfected and infected cage-mates in the acute or chronic infection phase. Black dots = uninfected mice, pink dots = infected mice, grey line = co-housed mice (C) Relative abundances at the Phylum level were unchanged following infection (ANOVA, p > 0.05). (D) An increase in the abundance of the taxa R. gnavus was observed in the acute and chronic infection phases in mice infected with H. polygyrus (DESeq2). Data are from 5 mice/group from a single experiment.

Figure 5

H. polygyrus induces goblet cell hyperplasia and upregulates goblet-cell-derived proteins in the lung. (A) Percentage of PAS-stained area in lung sections from uninfected mice (black) and mice infected with H. polygyrus (pink) for one, two, and six weeks, n = 6 per timepoint from two experiments. (B) Representative images from PAS-stained lung from mice one week post-infection, and uninfected mice. Goblet-cells-stained magenta. Scale bar = 100 μm. (C,D) Relative differences in mRNA expression in the lung between uninfected mice and mice infected with H. polygyrus for one week (acute infection phase), n = 6 from two experiments. (E) Relative differences in mRNA expression in the lungs of uninfected and mice infected with H. polygyrus for six weeks (chronic infection phase). Only genes demonstrating a significant difference in the acute infection phase were assessed. n = 6 from two experiments. Rplpo0 or Gapdh was used as a housekeeping gene. Data were analysed with unpaired student t-test (C–E) or ANOVA with post-hoc Dunnett’s test (A). Mean ± SEM. ns, not significant; *, p < 0.05. **, p < 0.01, ****, p < 0.0001.

Figure 6

IL-13 drives mucus responses at sites distant to the intestinal niche of H. polygyrus. (A) Intestinal helminth burdens in wild-type (WT) and IL-13 knockout (IL-13^−/−) mice two weeks after being infected with 200 H. polygyrus larvae, n = 8 mice/group from a single experiment. (B) Percentage of sections positively stained by PAS in the colon and lung of uninfected and infected IL-13^−/− mice in the acute infection phase (two weeks post-infection for colon and one week post-infection for lung), n = 5–7 from two experiments. (C) Representative images from PAS-stained colon and lung from uninfected and infected IL-13^−/− mice. Goblet cells are stained magenta. Scale bar = 100 μm. (D,E) Relative differences in mRNA expression in the colon (D) and lung (E) between uninfected and infected IL-13^−/− mice. Only genes demonstrating a significant difference in the acute infection phase in WT mice were assessed. RPLP0 or GAPDH was used as a housekeeping gene. N = 8–10 from two experiments. Data were analysed with unpaired student t-test. Mean ± SEM. ns, not significant; *, p < 0.05; **, p < 0.01.

Figure 7

Serum TFF1 increases in humans experimentally challenged with human hookworm. (A) Median faecal egg counts in 12 healthy participants experimentally challenged with the human hookworm, N. americanus, during the 48-week follow-up period. (B) Number of visualised intestinal hookworm using capsule endoscopy at 20 weeks post-infection, line = median. (C–F) ELISA results showing concentrations of human serum TFF1 (A), faecal TFF1 (B), serum TFF2 (C), and faecal TFF2 (D) at baseline and 4, 8, 12, 24, 36, and 48 weeks post-infection. n = 12. Single outlying participant removed from (C) to improve data visualisation (data including outlier displayed in Figure S1A. Data were analysed using Friedman test with Dunn’s post-hoc multiple comparisons test. Median and IQR. Significant changes compared to baseline levels are indicated with an asterisk. *, p < 0.05.