Bacterial contact induces polar plug disintegration to mediate whipworm egg hatching

Abstract

The bacterial microbiota promotes the life cycle of the intestine-dwelling whipworm Trichuris by mediating hatching of parasite eggs ingested by the mammalian host. Despite the enormous disease burden associated with Trichuris colonization, the mechanisms underlying this transkingdom interaction have been obscure. Here, we used a multiscale microscopy approach to define the structural events associated with bacteria-mediated hatching of eggs for the murine model parasite Trichuris muris . Through the combination of scanning electron microscopy (SEM) and serial block face SEM (SBFSEM), we visualized the outer surface morphology of the shell and generated 3D structures of the egg and larva during the hatching process. These images revealed that exposure to hatching-inducing bacteria catalyzed asymmetric degradation of the polar plugs prior to exit by the larva. Although unrelated bacteria induced similar loss of electron density and dissolution of the structural integrity of the plugs, egg hatching was most efficient in the presence of bacteria that bound poles with high density such as Staphylococcus aureus . Consistent with the ability of taxonomically distant bacteria to induce hatching, additional results suggest chitinase released from larva within the eggs degrade the plugs from the inside instead of enzymes produced by bacteria in the external environment. These findings define at ultrastructure resolution the evolutionary adaptation of a parasite for the microbe-rich environment of the mammalian gut.

Fig 1.. Bacteria display species-dependent effects on T. muris egg hatching.

(A) Representative light microscopy image of T. muris eggs induced to hatch after incubation with S. aureus at 37°C for 45 mins. White arrowhead denotes unhatched egg and black arrowhead denotes hatched egg. (B) Percent of T. muris eggs hatched after incubation in aerobic conditions with overnight cultures of E. coli and S. aureus, compared with their respective broth controls determined by light microscopy at indicated time points. Colony forming units (CFUs) of each bacterial species added to the eggs are indicated on the graphs. (C) Percent of T. muris eggs hatched after incubation in aerobic conditions with overnight cultures of S. epidermidis, B. subtilis, E. faecalis and E. faecium compared with their respective broth controls determined by light microscopy at indicated time points. Colony forming units (CFUs) of each bacteria added to the eggs are indicated on the graphs. (D) Percent of T. muris eggs hatched after incubation with S. aureus grown overnight in tryptic soy broth (TSB), Luria Bertani (LB) broth and Brain Heart Infusion (BHI) broth. (E) Percent of T. muris eggs hatched at 1 hour after 37°C incubation with a maximum number of E. coli, S. aureus and E. faecalis. (F) Percent of T. muris eggs hatched at 1 hour after 37°C incubation with an equal number (~10⁸ CFU) of E. coli and E. faecalis. (G) Percent of T. muris eggs hatched after incubation with S. aureus and E. faecalis alone or together compared with broth controls. For the S. aureus + E. faecalis condition, bacterial cultures were grown separately overnight, and then were mixed the day of the experiment. (H) Percent of T. muris eggs hatched after incubation in anaerobic conditions with overnight cultures of E. coli and S. aureus compared with their respective broth controls determined by light microscopy at indicated time points. (I) Number of worms harvested from the caecum per mouse (n = 9–12 mice per group). (J) Proportion of germ-free and S. aureus monocolonized mice that had harbored adult T. muris worms after double-dose infection. Data points and error bars represent mean and SEM from 3 independent repeats for (B), (C), (D), (G), and (H). Dots represent a single well and bars show means and SEM from 3 independent repeats for (E) and (F). Dots represent a single mouse and bars show means and SEM from 3 independent repeats for (I). Welch’s t test was used to compare area under the curve between each condition and its respective media control for (B), (C), and (H). Ordinary one-way analysis of variance (ANOVA) test followed by a Turkey’s multiple comparisons test was used to compare AUC of hatching induced by different conditions to each other for (D) and (G). Kruskal-Wallis test followed by a Dunn’s multiple comparisons test was used for (E). Mann Whitney test was used for (F) and (I). Fisher’s exact test was used to determine whether there was a significant association between gut microbial composition of mice and the presence of adult worms in the cecum for (J).

Fig 2.. Physical contact between the bacterial cell and egg is essential for S. aureus mediated hatching.

(A) Percent of T. muris eggs hatched after incubation with total overnight S. aureus culture, culture pellet resuspended in fresh media, culture supernatant filtered with a 0.22μm filter, or media control. (B) Experimental approach for determining whether a soluble hatching inducing factor is produced by S. aureus in response to exposure to eggs. Filtered supernatant from S. aureus grown with or without T. muris eggs for 4 h were transferred to a dish containing T. muris eggs. The ability of the supernatant to mediate egg hatching was evaluated over 4 hours. (C) Percent of T. muris eggs hatched after incubation with total overnight S. aureus culture or filtered supernatant obtained from S. aureus incubated with or without eggs as in (B) compared with media controls. (D) Percent of T. muris eggs hatched when placed in a 0.4μm transwell separated from bacteria or control media in the outer well compared with eggs incubated bacteria without transwell separation. Data points and error bars represent the mean and SEM from 3 independent experiments for (A) and (C). Dots represent a single well and bars show means and SEM from 3 independent experiments for (D). Ordinary one-way ANOVA test followed by a Turkey’s multiple comparisons test was used to compare resuspended pellet and supernatant conditions with total O/N culture for (A) and (C). Two-way ANOVA followed by a Turkey’s multiple comparisons test was used for (D).

Fig 3.. Collapse of the polar plug precedes hatching mediated by bacteria.

(A, B, C, D) Representative low (left) and high magnification (right) SEM images of eggs (clear arrowhead) that were exposed to S. aureus for 1 hour (A), E. coli for 1.5 hours (B) and E. faecalis for 1 hour (C) or untreated (D). White arrowheads correspond to bacteria on polar plug regions of the eggs denoted by black arrowheads. Yellow arrow in right panel of (A) and (B) indicates woolly substance present among bacteria. Debris on untreated eggs is denoted by the white arrow (D). (E) Representative low (left) and high magnification (right) SEM images of hatching eggs (clear arrowhead) that were exposed to S. aureus for 1 hour (top) and E. coli for 1.5 hours (bottom). White arrowheads correspond to bacteria on polar plug regions of the eggs. Emerging larvae are denoted by white diamonds. (F) Number of bacterial cells visible on polar plugs of eggs incubated with E. coli, S. aureus, or E. faecalis. Bars showing mean from 2 eggs per condition. (G) Width of collar openings on eggs that were treated with either E. coli or S. aureus and were either unhatched or in the process of hatching. For low magnification images, scale bar represents 2μm for unhatched egg in (A), (B), (C) and (D) and 10μm for hatched egg in (A) and (B). For high magnification images, scale bar represents 1μm for (A), (B) and (D) and 2μm for (C). Dots represent a single plug and bars show mean and SEM of collar sizes from 4–7 eggs per condition for (G). Two-way ANOVA followed by a Turkey’s multiple comparisons test was used for (G).

Fig 4.. Bacteria induce asymmetric disintegration of the polar plug.

(A) Representative electron micrograph of a longitudinal cross-section of a T. muris egg exposed to S. aureus for 1 hour. Image shows granules containing lipids (LG) within the larva, eggshell (ES) and larval cuticle (C). Distances measured are also indicated. The width of the collar opening (i), width of the widest part of the plug (ii), the height of the plug (distance from the top of the collar to the top of the plug) (iii) and the eggshell thickness (iv) were measured for all conditions. Insets show the regions boxed in black dotted line. (B, C) Representative electron micrographs and 3D reconstructions from SBFSEM data of a T. muris egg exposed to S. aureus for 1 hour (B) and E. coli for 1.5 hours (C). For the longitudinal section, the original micrograph (left), the micrograph with color overlays indicating segmented egg components and bacteria (middle) and 3D reconstruction of the egg (right) are shown. For the complete structure, two different angles are shown (left and right). S. aureus (red), E. coli (blue), polar plugs (green), eggshell (yellow) and larvae (purple) are all shown. Scale bars represent 10μm. (D) High magnification images of polar plugs from (B) and (C) on eggs exposed to S. aureus and E. coli compared with equivalent regions from an egg untreated with bacteria. Outer vitelline layer is denoted by white arrowheads and eggshell is denoted by black arrowheads. (E) 3D reconstruction of polar plugs on eggs exposed to E. coli. Multiple angles are shown (left to right).

Fig 5.. Degree of bacterial binding is directly proportional to the rate of T. muris egg hatching.

(A) Representative confocal microscopy image of eggs incubated with S. aureus GFP for 30 minutes. Regions defined as poles and sides are indicated. Scale bar represents 34μm. (B) Percent of GFP+ puncta present on the poles versus the sides of the eggs. (C) Percent of T. muris eggs hatched after incubation with 10-fold dilutions of overnight S. aureus culture ranging from approximately 10⁵ – 10⁸ CFU (n = 3). (D) Percent of eggs associated with GFP+ puncta after incubation with 10-fold dilutions of overnight S. aureus-GFP culture ranging from approximately 10⁵ – 10⁸ CFU. (E) Correlation analysis comparing log₁₀ CFUs of bacteria used and percent of eggs with GFP+ puncta present. (F) Number of GFP+ puncta bound per egg from (D). (G) Percent of GFP+ puncta present on the poles of the eggs versus the sides of the eggs from (D). Bars and error bars show means and SEM from 3 independent repeats for (B), (D), (F) and (G). Data points and error bars represent mean and SEM from 3 independent repeats for (C). Dots represent percentage of 40 eggs that were GFP+ from a single experiment for (D) and (E) and number of GFP+ puncta found on a single egg for (F). Mann-Whitney test was used for (B). Kruskal-Wallis test followed by a Dunn’s multiple comparisons test was used for (D) and (F). Simple linear regression was performed for (E). Two-way ANOVA followed by a Sidak’s multiple comparisons test was used for (G).

Fig 6.. Bacterial metabolic activity is essential for S. aureus mediated hatching.

(A) Percent of T. muris eggs hatched after incubation with 10⁸ CFU overnight S. aureus culture at time points indicated. (B) Percent of eggs that had GFP+ puncta bound after incubation with 10⁸ CFU of overnight S. aureus culture for different lengths of time. (C) Number of GFP+ puncta bound per egg after incubation with 10⁸ CFU of overnight S. aureus culture for different lengths of time. (D) Percent of GFP+ puncta present on the poles of the eggs versus the sides of the eggs after incubation with 10⁸ CFU of overnight S. aureus culture for different lengths of time. (E) Percent of T. muris eggs hatched after incubation with UV killed overnight S. aureus culture. (F) Percent of T. muris eggs hatched after incubation with untreated S. aureus, chloramphenicol (CAM)-treated S. aureus (100μg/ml), and CAM-treated S. aureus together with CAM-resistant S. aureus GFP for 4hrs. S. aureus GFP was spiked-in at the 2hr time point. Bars and error bars show means and SEM from 3 independent repeats for (B), (C), and (D). Data points and error bars represent mean and SEM from 3 independent repeats for (A), (E) and (F). Dots represent mean percentage of 40 eggs that were GFP+ from a single experiment for (B) and number of GFP+ puncta found on a single egg for (C). Kruskal-Wallis test followed by a Dunn’s multiple comparisons test was used for (B) and (C). Two-way ANOVA followed by a Sidak’s multiple comparisons test was used for (D). Ordinary one-way ANOVA followed by Turkey’s multiple comparisons test was used to compare AUC of different conditions tested for (F).

Fig 7.. Trichuris eggs harbor chitinase activity.

(A-C) Percent of T. muris eggs hatched after incubation with overnight S. aureus WT, S. aureus Δcrp (A), S. aureus ΔaurΔsspABΔscpAspl∷erm (B) and S. aureus ΔgehAΔgehBΔgehE (C) culture. (D, E) Amount of fluorescence detected in wells containing eggs hatched in response to S. aureus (D) and E. coli (E). AU = arbitrary unit. 5mU of stock chitinase were used as a positive control. (F) Amount of fluorescence detected in wells containing crushed eggs that were exposed to S. aureus for 20 minutes. 0.02mU of stock chitinase were used as a positive control. (G) Amount of fluorescence detected in wells containing crushed T. muris and T. trichiura eggs. 0.02mU of stock chitinase were used as a positive control. (H) Percent of T. muris eggs hatched after incubation with fluid from hatched eggs (purple open circles) and bacteria resuspended in fluid from hatched eggs (purple filled circles). Fluid from unhatched eggs was used as a control (yellow filled and open circles). (I) Graphical depiction of proposed mechanism. Data points and error bars represent means and SEM from 3 independent repeats for (A), (B), (C) and (H). Dots represent fluorescence detected in a single well and bars and error bars show means and SEM from 2–3 independent repeats for (D), (E), (F) and (G). Welch’s t test was used to compare area under the curve between mutant S. aureus strains and WT S. aureus for (A), (B) and (C). Kruskal-Wallis test followed by a Dunn’s multiple comparisons test was used for (D), (E), (F) and (G). Ordinary one-way ANOVA followed by Turkey’s multiple comparisons test was used to compare AUC of different conditions tested for (H).