Distinct members of the C. elegans CeMbio reference microbiota exert cryptic virulence and infection protection

Abstract

Microbiotas are complex microbial communities that colonize specific niches in the host and provide essential organismal functions that are important in health and disease. A key aspect is the ability of each distinct community member to promote or impair host health, alone or in the context of the community, in hosts with varied levels of immune competence. Understanding such interactions is limited by the complexity and experimental accessibility of current systems and models. Recently, a reference twelve-member microbiota for the model organism C. elegans, known as CeMbio, was defined to aid the dissection of conserved host-microbiota interactions. Understanding the physiological impact of the CeMbio bacteria on C. elegans is in its infancy. Here, we show the differential ability of each CeMbio bacterial species to activate innate immunity through the conserved PMK-1/p38 MAPK, ACh/WNT, and HLH-30/TFEB pathways. Using immunodeficient animals, we uncovered several examples of bacterial 'cryptic' virulence, or virulence that was masked by the host defense response. The ability to activate the PMK-1/p38 pathway did not correlate with bacterial virulence in wild type or immunodeficient animals. In contrast, ten out of twelve species activated HLH-30/TFEB, and most showed virulence towards hlh-30-deficient animals. In addition, we identified Pseudomonas lurida as a pathogen in wild type animals, and Acinetobacter guillouiae as avirulent despite activating all three pathways. Moreover, short pre-exposure to A. guillouiae promoted host survival of infection with P. lurida, which was dependent on PMK-1/p38 MAPK and HLH-30/TFEB. These results suggest that the microbiota of C. elegans is rife with "opportunistic" pathogens, and that HLH-30/TFEB is a fundamental and key host protective factor. Furthermore, they support the idea that bacteria like A. guillouiae evolved the ability to induce host innate immunity to improve host fitness when confronted with pathogens, providing new insights into how colonization order impacts host health.

Figure 1.. Growth and morphological characteristics of CeMbio bacteria.

A. Representative images of individual CeMbio colonies grown on TSA at 25 °C for 24–48 h. Scale bars = 1 mm. B. Gram staining of individual CeMbio cultures grown overnight in TSB at 25 °C for 24 h. Scale bars = 5 μm.

Figure 2.. C. elegans microbiota bacteria differentially activate Pt24b8.5::gfp expression along the intestinal epithelium.

A - D. Representative brightfield and epifluorescence micrographs of Pt24b8.5::gfp animals fed E. coli, S. aureus, P. nemavictus, and C. scophthalmum for 24 h at 25 °C. Animals were straightened using FIJI. Scale bars = 200 μm. E - G. Pt24b8.5::gfp mean fluorescent intensity (MFI) along the intestine of animals exposed to the indicated bacteria for 24 h at 25 °C. Representative of 3 biological replicates, n = 15 – 25 animals per biological replicate. H. Area under the curve (AUC) quantitative analysis of Pt24b8.5::gfp expression in the intestine, normalized to E. coli control. Animals fed on the indicated bacteria for 24 h at 25 °C. Three biological replicates, n = 15 – 25 animals per biological replicate. ****p ≤ 0.0001; *p ≤ 0.05, ordinary one-way ANOVA, Dunnett’s multiple comparisons test, compared to E. coli controls.

Figure 3.. PMK-1/p38 MAPK is differentially required for C. elegans survival on distinct CeMbio bacteria.

A and B. Survival of wild type (A) or pmk-1(km25) (B) animals after transfer to full lawns of CeMbio bacterial monoculture on TSA at 25 °C with the indicated antibiotics to prevent E. coli contamination. Data are representative of two biological replicates. Error bars are ± SEM. n = 50 – 100 animals per condition per replicate. C – F. Survival of wild type and pmk-1(km25) animals after transfer to full lawns of CeMbio bacterial monoculture on TSA at 25 °C on days 3 (C and D) and 7 (E and F). Trials were performed in two internally controlled batches as shown. Data points represent median survival from individual biological replicates; at least 2 biological replicates for each treatment. Error bars are ± SEM. n = 50 – 100 animals per condition per replicate. ****p ≤ 0.0001, ***p ≤ 0.001, **p ≤ 0.01, *p ≤ 0.05, ordinary two-way ANOVA, Dunnett’s multiple comparisons test.

Figure 4.. C. elegans microbiota bacteria differentially activate Pclec-60::gfp expression along the intestinal epithelium.

A - D. Representative brightfield and epifluorescence micrographs of Pclec-60::gfp animals fed E. coli, S. aureus, L. amnigena, and C. scophthalmum for 24 h at 25 °C. Scale bars = 200 μm. E - G. Pclec-60::gfp mean fluorescent intensity (MFI) along the intestine of animals exposed to the indicated bacteria for 24 h at 25 °C. Representative of 3 biological replicates, n = 15 – 25 animals per biological replicate. H. Area under the curve (AUC) quantitative analysis of Pclec-60::gfp expression in the intestine, normalized to E. coli control. Animals fed on the indicated bacteria for 24 h at 25 °C. Three biological replicates, n = 15 – 25 animals per biological replicate. ****p ≤ 0.0001, ***p ≤ 0.001, **p ≤ 0.01, *p ≤ 0.05, ordinary one-way ANOVA, Dunnett’s multiple comparisons test, compared to E. coli controls.

Figure 5.. C. elegans microbiota bacteria differentially drive HLH-30::GFP nuclear localization.

A - D. Representative brightfield and epifluorescence micrographs of Phlh-30::hlh-30a::gfp (HLH-30::GFP) animals fed E. coli, S. aureus, S. molluscorum, and C. scophthalmum for 30 min at 25 °C. Scale bars = 200 μm. E and F) Representative epifluorescence micrographs of HLH-30::GFP animals posterior intestine fed E. coli or S. aureus for 30 min at 25 °C. Scale bars = 50 μm H and I) Representative epifluorescence micrographs of HLH-30::GFP animals head fed E. coli or S. aureus for 30 min at 25 °C. Scale bars = 20 μm G and H. Quantitative analysis of HLH-30::GFP nuclear localization in the intestinal epithelium (G) or head (H) of animals fed the indicated bacteria for 30 min. Data are means ± SEM for three biological replicates; n = 15 – 25 animals per biological replicate. ****p ≤ 0.0001, ***p ≤ 0.001, **p ≤ 0.01, *p ≤ 0.05, ordinary one-way ANOVA, Dunnett’s multiple comparisons test, compared to E. coli controls.

Figure 6.. HLH-30 is differentially required for C. elegans survival on distinct CeMbio bacteria.

A – D. Survival of wild type and hlh-30(tm1978) animals after transfer to full lawns of CeMbio bacterial monoculture on TSA at 25 °C on days 3 (A and B) and 7 (C and D). Trials were performed in two internally controlled batches as shown. Data points represent median survival from individual biological replicates; at least 2 biological replicates for each treatment. Error bars are ± SEM. n = 50 – 100 animals per condition per replicate. ****p ≤ 0.0001, ***p ≤ 0.001, **p ≤ 0.01, *p ≤ 0.05, ordinary two-way ANOVA, Dunnett’s multiple comparisons test.

Figure 7.. A. guillouiae exerts immune-mediated host protection against P. lurida infection.

A. Schematic of the experimental approach. Animals were transferred from E. coli to A. guillouiae on TSA for 8 h at 25 °C. Subsequently, they were transferred to full lawns of P. lurida on TSA at 25 °C and scored for survival. In parallel, controls animals were transferred to E. coli instead of A. guillouiae and treated identically. B. Survival of P. lurida infection of wild type animals pre-exposed to E. coli or A. guillouiae for 8 h. Representative of 3 biological replicates, n = 50–100 animals. Error bars are ± SEM from 3 technical replicates. ****p ≤ 0.0001, ***p ≤ 0.001, **p ≤ 0.01, *p ≤ 0.05, Kaplan-Meier Log-rank (Mantel-Cox) test. C. Survival of P. lurida infection of pre-exposed wild type and pmk-1(−) animals. Representative of 3 biological replicates, n = 50–100 animals. Error bars are ± SEM from 3 technical replicates. ****p ≤ 0.0001, ***p ≤ 0.001, **p ≤ 0.01, *p ≤ 0.05, Kaplan-Meier Log-rank (Mantel-Cox) test. D. Survival of P. lurida infection of pre-exposed wild type and hlh-30(−) animals. Representative of 3 biological replicates, n = 50–100 animals. Error bars are SEM from 3 technical replicates. ****p ≤ 0.0001, ***p ≤ 0.001, **p ≤ 0.01, *p ≤ 0.05, Kaplan-Meier Log-rank (Mantel-Cox) test.

Figure 8.. Summary of pathway analysis data.

Reporter activity and survival at days 3 and 7 of animals exposed to the indicated CeMbio bacteria. Represented data are from Fig. 2 – 6. In the survival heat maps, the WT data are represented twice to facilitate comparison with the pmk-1(−) and hlh-30(−) mutants.