The intestinal life cycle of Bacillus subtilis and close relatives

Abstract

Bacillus subtilis is considered a soil organism for which endospore formation provides a means to ensure long-term survival in the environment. We have addressed here the question of what happens to a spore when ingested. Spores displaying on their surface a heterologous antigen, tetanus toxin fragment C (TTFC), were shown to generate anti-TTFC responses not to the antigen contained in the primary oral inoculum but to those displayed on spores that had germinated and then resporulated. We then used reverse transcription-PCR to determine expression of vegetative genes and sporulation-specific genes in the mouse gut following oral dosing with spores. Significant levels of germination and sporulation were documented. Using natural isolates of B. subtilis that could form biofilms, we showed that these strains could persist in the mouse gut for significantly longer than the laboratory strain. Moreover, these isolates could grow and sporulate anaerobically and exhibited a novel phenomenon of being able to form spores in almost half the time required for the laboratory isolate. This suggests that spores are not transient passengers of the gastrointestinal tract but have adapted to carry out their entire life cycle within this environment. This is the first report showing an intestinal life cycle of B. subtilis and suggests that other Bacillus species could also be members of the gut microflora.

FIG. 1.

RT-PCR primers. Physical map of the cotB-tetC and rrnO-tetC chimeric genes. The primer pairs FcotB1 and RtetC5 and FrrnO4 and RtetC5 were used to amplify a 559-bp segment across the cotB-tetC fusion junction and a 522-bp segment across the rrnO-tetC fusion junction. FcotB2 and RtetC5 and FrrnO5 and RtetC5 were, respectively, used to amplify 466- and 354-base competitor templates. The hatched boxes indicate the sequence included in the 5′ end of the FcotB2 primer, which is recognized by the FcotB1 primer, and the sequence included in the 5′ end of the FrrnO5 primer, which is recognized by the FrrnO4 primer.

FIG. 2.

Serum anti-TTFC-specific IgG responses. (A) Groups of 7 C57BL/6 mice were dosed orally with 2 × 10¹⁰ spores expressing a C. tetani antigen, TTFC, as follows: ▪, spores expressing TTFC on the spore surface (strain RH103 amyE::cotB-tetC); ▵, spores expressing TTFC only when the spore germinates (strain DL237, amyE::rrnO-tetC). Control groups were groups immunized with nonrecombinant spores of strain PY79 (•; spo⁺) and a naive group consisting of nonimmunized animals (○). The dosing strategy comprised three sets of three separate doses indicated by arrows. Serum samples were tested for anti-TTFC-specific IgG. (B) Groups of 7 C57BL/6 mice were dosed orally with 2 × 10¹⁰ spores as follows: ▪, RH103 spores that express TTFC on the spore surface (amyE::cotB-tetC); ×, UL12 spores (amyE::cotB-tetC gerD-cwlBΔ::neo) that express CotB-TTFC on the spore surface but are unable to germinate; □, spores of RH103 that had been stripped of CotB-TTFC by treatment with SGF plus SIF. Control groups were mice immunized with nonrecombinant PY79 spores PY79 (•; spo⁺) and a naive group (○). The dosing strategy comprised three sets of three separate doses indicated by arrows.

FIG. 3.

RT-PCR analysis of germination and sporulation in the mouse gut. RT-PCR analysis of germination and sporulation genes in vivo. Groups of mice (4) were orally dosed with suspensions of spores (2 × 10¹⁰) of DL291 (PY79 amyE::cotB-tetC thrC::rrnO-tetC), BT17 (HU58 amyE::rrnO-tetC), and BT47 (HU58 amyE::cotB-tetC). Mice were sacrificed at the indicated times, sections of the GIT were dissected, and total RNA was recovered. To detect germination gene expression, two primers (FrrnO4 and RtetC5) were used to amplify a 522-bp cDNA (Fig. 1). To detect sporulation gene expression, primers FcotB1 and RtetC5 were employed to amplify a 549-bp cDNA (Fig. 1). The marker (M) is a 100-bp ladder. The negative control (−) was from naive mice, and the positive control (+) was amplification from chromosomal DNA of either RH103 (cotB-tetC) or DL237 (rrnO-tetC). Reactions from one mouse in each group are shown, but all mice per group behaved identically. D, duodenum; J, jejunum; I, ileum; L, large intestine.

FIG. 4.

Biofilm formation. Pellicle formation in two undomesticated strains (HU58 and HU78) of B. subtilis versus no pellicle formation in the laboratory strain PY79 grown on CMK medium.

FIG. 5.

Behavior of domesticated and undomesticated strains. (A) Counts of spores excreted in the feces of mice following a single fixed oral dose (1 × 10⁹) of spores of PY79 and two undomesticated strains (HU58 and HU78). Heat-resistant (65°C, 1 h) counts were determined from fresh feces collected from individual mice as described previously (7). The detection limit was 1 × 10³ and is indicated by the dashed line. Counts shown are means of results from 6 animals. (B) Development of heat-resistant spores during sporulation in PY79, HU58, and HU78. Cells were induced to form spores by the exhaustion method, and the number of heat-resistant spores was determined (80°C, 20 min) at time points following the initiation of spore formation (T₀).