(A) NMDS analysis of the rarefied operational taxonomic unit (OTU) table from 16S amplicon pyrosequencing comparing cecal content from control naïve mice (n = 5), mice with low-dose T. muris infection (n = 5), and T. muris internal microbiota (n = 15; three individual T. muris from each infected mouse) at day 0 (d0) and day 41 (d41). Data were collated from randomized, cohoused, and nonrandomized repeat experiments. The boxed section in the left panel is enlarged on the right. All samples are significantly different by permutational multivariate analysis of variance (PERMANOVA) tests (P_adjusted < 0.01). Axis represents scale for Euclidian distance between samples centered on zero, and stress indicates the quality of fit of data (<0.2 is a good fit). (B) Cross section of an adult T. muris intestinal tract (bright field) hybridized with a 16S universal Cy3 probe (red) for 2 hours and the merged image (100×). (C) The number of bacteria per worm was estimated by qPCR using 16S rRNA gene universal primers on DNA extracted from day 42 (n = 5) and day 91 (n = 5) T. muris. Error bars are ±SEM. (D) Live T. muris adults (bright field) were incubated overnight with GFP-expressing E. coli (green), washed in RPMI 1640 supplemented with penicillin/streptomycin solution, and processed for sectioning and FISH with a 16S universal Cy3 probe (red; 60×). Scale bars, 10 μm.

(A) Comparisons of proportions at the phylum level between samples. Phyla representing less than 0.5% in all treatments were grouped together into “Other.” (B) Community abundance differences were compared at all taxonomic levels to identify significant shifts between groups. N0 (naïve day 0), N41 (naïve day 41), I41 (infected day 41), Tm (T. muris day 41). Error bars are SEM. The asterisk denotes samples that are significantly different to all other samples from corrected post hoc Dunn test after false discovery rate (FDR) correction on Kruskal-Wallis test results (see table S2). (C) Venn diagram of OTUs present in all samples of a group and those shared between groups. See table S1 for list of shared OTUs. (D) Taxonomic identification of conserved OTUs found in all T. muris microbiotas. Data were collated from randomized, cohoused, and nonrandomized repeat experiments.

(A) Experimental design for repeated T. muris infection. C57BL/6 mice were infected with a low dose (~20) of T. muris eggs at day 0 and day 41 p.i. so that mice had single or repeat infections. (B) Worm burdens from low-dose infections (~20 eggs) in C57BL/6 mice comparing single and repeat infections where eggs hatch in different host intestinal microbiota environments. (C) Cecal contents were taken from naïve and infected mice and incubated with ~130 embryonated T. muris eggs. Hatching rates are reported in comparison with naïve results. Egg batches, A and B, are from two separate egg harvests. (D) Worm burdens from low-dose infections into a naïve drug-treated host microbiota (X – Meb – LD) or a drug-treated infected microbiota void of worms (LD – Meb – LD). (E) Worm burdens from low-dose infections in SCID mice comparing single and repeat infections. *P < 0.002, using unpaired t test. Error bars are ±SEM. Mice were cohoused and randomized.

(A) Adult T. muris worms were incubated for 5 days with an antibiotic treatment to deplete their microbiota. Medium only was used as a negative control. Motility scoring was performed every 24 hours [3, normal motility; 2, low motility (less than controls); 1, very low motility/just at one end; 0, no motility/dead]. *P < 0.0001, using analysis of variance (ANOVA). (B) Treatment of sterile L1 T. muris larvae with antibiotics or medium only as a negative control over 5 days. The number of larvae at each time point was counted, and the percentage was calculated. (C) Worm burdens from day 14 p.i. of GF (n = 8) and WT control mice (n = 8) infected with a high dose (~500) of sterile T. muris L1 larvae or a high dose of sterile eggs. **P < 0.008. (D) Worm burdens from low-dose infections at day 14 p.i. and day 42 p.i. of either GF animals that had been seeded with Bt strain VPI-5482 bacteria or GF animals that had been seeded with bacteria from a naïve C57BL/6 animal (FS, fecal slurry). C57BL/6 mice were used as infection controls (WT). Error bars are ±SEM. Experimental design precluded cohousing.