Report. B. subtilis GS67 Protects C. elegans from Gram-Positive Pathogens via Fengycin-Mediated Microbial Antagonism

Transcription

1 Current Biology 24, , November 17, 2014 ª2014 Elsevier Ltd All rights reserved B. subtilis GS67 Protects C. elegans from Gram-Positive Pathogens via Fengycin-Mediated Microbial Antagonism Report Igor Iatsenko, 1,3 Joshua J. Yim, 1,2 Frank C. Schroeder, 2 and Ralf J. Sommer 1, * 1 Department of Evolutionary Biology, Max Planck Institute for Developmental Biology, Spemannstraße 37, Tübingen, Germany 2 Boyce Thompson Institute and Department of Chemistry and Chemical Biology, Cornell University, Ithaca, NY 14853, USA Summary Studies on Caenorhabditis elegans have provided detailed insight into host-pathogen interactions [1 7]. Usually, the E. coli strain OP50 is used as food source for laboratory studies, but recent work has shown that a variety of bacteria have dramatic effects on C. elegans physiology, including immune responses [8 18]. However, the mechanisms by which different bacteria impact worm resistance to pathogens are poorly understood. Although pathogen-specific immune priming is often discussed as a mechanism underlying such observations [13, 14], interspecies microbial antagonism might represent an alternative mode of action. Here, we use several natural Bacillus strains to study their effects on nematode survival upon pathogen challenge. We show that B. subtilis GS67 persists in the C. elegans intestine and increases worm resistance to Gram-positive pathogens, suggesting that direct inhibition of pathogens might be the primary protective mechanism. Indeed, chemical and genetic analyses identified the lipopeptide fengycin as the major inhibitory molecule produced by B. subtilis GS67. Specifically, a fengycin-defective mutant of B. subtilis GS67 lost inhibitory activity against pathogens and was unable to protect C. elegans from infections. Furthermore, we found that purified fengycin cures infected worms in a dose-dependent manner, indicating that it acts as an antibiotic. Our results reveal a molecular mechanism for commensal-mediated C. elegans protection and highlight the importance of interspecies microbial antagonism for the outcome of animalpathogen interactions. Furthermore, our work strengthens C. elegans as an in vivo model to reveal protective mechanisms of commensal bacteria, including those relevant to mammalian hosts. Results and Discussion A previous systematic analysis of Bacillus strains isolated from different sources revealed that the majority of strains were benign to the two model nematodes C. elegans and Pristionchus pacificus [19]. However, we also discovered B. thuringiensis DB27, a highly virulent pathogen that kills C. elegans in less than 16 hr [20]. Subsequently, we have shown the C. elegans defense mechanisms against this strain to involve dcr-1/dicer [21] and have identified two new pore-forming toxins as 3 Present address: Global Health Institute, School of Life Sciences, École Polytechnique Fédérale de Lausanne (EPFL), 1015 Lausanne, Switzerland *Correspondence: ralf.sommer@tuebingen.mpg.de bacterial virulence factors [20, 22], revealing additional important components of extensively studied C. elegans response to B. thuringiensis pore-forming toxins [23 29]. A number of studies have shown that different bacteria have dramatic effects on C. elegans immune responses [8 18], with pathogen-specific immune priming often being discussed as an underlying mechanism [13, 14]. To test whether naturally isolated bacteria have an effect on worm resistance to B. thuringiensis DB27, we selected from our library [19] several nonpathogenic strains of different Bacillus species that are often found in the natural habitat of C. elegans [9, 14] and tested whether feeding on those strains has an effect on C. elegans survival (Figure 1A). Indeed, we found that preincubation with nine strains of Bacillus resulted in significant levels (20% 50%) of worm protection (p % 0.05) compared to E. coli OP50 (Figure 1B). Out of all tested strains, B. subtilis GS67 showed the strongest effect on C. elegans, with up to 50% survival after 16 hr postinfection and with an average survival of up to 3 days (Figure 1C). These results demonstrate that natural Bacillus isolates have an effect on C. elegans response to pathogens. To investigate the underlying molecular mechanisms, we concentrated on B. subtilis GS67. First, we asked whether the observed effect of B. subtilis GS67 is specific to B. thuringiensis DB27 or whether the nematodes acquire resistance to other pathogens as well. Upon exposure of worms precultured on B. subtilis GS67 to another Gram-positive pathogen, Staphylococcus aureus, we again observed significant (p % 0.001) increase of survival as compared to worms precultured on E. coli OP50 (Figure 1D). In contrast, when we tested survival on two Gram-negative pathogens S. marcescens (Figure 1E) and P. aeruginosa (Figure 1F), we found no differences in survival between worms precultured on either B. subtilis GS67 or E. coli OP50. Thus, B. subtilis GS67 increases C. elegans resistance to a group of Gram-positive pathogens but has no effect on resistance to the tested Gram-negative bacteria. We noticed that C. elegans grown on B. subtilis GS67 accumulated large amounts of Bacillus spores and cells in the intestine (Figure 2A). To estimate the period of persistence of B. subtilis GS67 in the C. elegans gut, we transferred GS67- grown worms to OP50 and determined colony-forming units (cfu) of B. subtilis GS67 after defined time intervals. Even 24 hr after transfer, we could still isolate GS67 from nematodes (Figure 2B), indicating that B. subtilis GS67 can persist for long times in the C. elegans gut. We therefore hypothesized that B. subtilis GS67 might provide protection to C. elegans via direct inhibition of B. thuringiensis DB27 and other Gram-positive pathogens. To test this hypothesis, we employed a simple disc-diffusion inhibition assay. Indeed, we found that supernatant of B. subtilis GS67 strongly inhibits growth of B. thuringiensis DB27 (Figure 2C). We further confirmed the antagonistic properties of GS67 in coculture experiments. When B. subtilis GS67 and B. thuringiensis DB27 were coinoculated at equal doses, B. thuringiensis DB27 was completely eliminated after 24 hr of culturing (Figure 2D). Even when B. thuringiensis DB27 was inoculated with a dose twice as high as B. subtilis GS67, B. thuringiensis DB27 was still

2 B. subtilis Protects C. elegans from Infection 2721 Figure 1. C. elegans Grown on Nonpathogenic Bacillus Is More Resistant to Gram-Positive Pathogens (A) Scheme of C. elegans feeding assay with Bacillus. Eggs obtained by the bleaching of worms grown on OP50 were put on plates with naturally isolated Bacillus and allowed to develop until adulthood. Adults were transferred to plates with pathogenic B. thuringiensis DB27, and survival was scored 16 hr later. (B) C. elegans survival on a pathogenic B. thuringiensis DB27 after feeding on different Bacillus strains. Survival was scored 16 hr after infection. Error bars represent 6SEM. Statistically significant values (*p < 0.05) compared to E.coli OP50 were determined with one-way ANOVA and Tukey s post hoc test. (C) Worms grown on B. subtilis GS67 survive on B. thuringiensis DB27 up to three times longer compared to OP50-grown counterparts (p < ). (D F) B. subtilis GS67 sighificantly (p < 0.001) increases C. elegans survival on S. aureus, but not on Gram-negative pathogens S. marcescens (E) and P. aeruginosa (F). Error bars represent 6SEM. outcompeted (Figure 2D). This result is not due to differences in growth rates because in the absence of antagonists, both cultures reach the same cell density (Figure 2D). When we investigated the spectrum of B. subtilis GS67 antagonistic activity, exposing several Gram-positive and Gram-negative bacteria to B. subtilis GS67 supernatant, we found that several Gram-positive bacteria, including different Bacillus species and S. aureus, were strongly inhibited by B. subtilis GS67 (Figure 2E). In contrast, all tested Gram-negative species were resistant to B. subtilis GS67 inhibition. Thus, B. subtilis GS67 specifically inhibits Gram-positive bacteria, a finding that is in line with the protective effect of B. subtilis GS67 against Gram-positive pathogens on C. elegans. To provide direct evidence that the bacterial antagonism observed in vitro contributes to C. elegans protection in vivo, we first estimated colonization of the C. elegans intestine by B. thuringiensis DB27. For this, worms were grown on OP50 (control), on B. subtilis GS67 (antagonist of DB27), or on B. subtilis 1A699 (a strain without antagonistic activity) and were subsequently infected with B. thuringiensis DB27. Worms grown on B. subtilis GS67 showed significant reduced pathogen accumulation in their intestines, as determined by

3 Current Biology Vol 24 No Figure 2. B. subtilis GS67 Persists in C. elegans Intestine and Inhibits B. thuringiensis DB27 (A) Nematodes grown on B. subtilis GS67 accumulate a large amount of bacterial cells and spores (white arrow) in the intestine. The scale bar represents 20 mm. (B) Worms transferred from B. subtilis GS67 to OP50 carry viable B. subtilis GS67 cells in the intestine up to 24 hr. Ten worms in three replicates were analyzed for each time point. Error bars represent 6SEM. (C) Representative image of disc-diffusion assay showing that cell-free supernatant of B. subtilis GS67 inhibits B. thuringiensis DB27 in contrast to control, where LB medium was used. Size of the inhibition zone is approximately 17 mm. (D) B. subtilis GS67 inhibits B. thuringiensis DB27 in coculture experiment. B. thuringiensis DB27 is completely eliminated after 24 hr of coculture with B. subtilis GS67, when inoculated at equal (1:1) or two times higher (1:2) dose. GS67 and DB27 alone exhibit similar (p > 0.05) growth rates. Growth of GS67 is significantly (p = 0.011) reduced only when DB27 is present at two times higher (1:2) dose, but not at equal (1:1) dose (p > 0.05). Error bars represent 6SEM. (E) Spectrum of B. subtilis GS67antagonistic activity determined by disc-diffusion assay. Multiple strains of each species were tested. Results for representative strain of each species are shown. Differences between Bacillus species are not significant (p > 0.05, ANOVA, and Tukey s post hoc test). Error bars represent 6SEM. (F) B. thuringiensis DB27 colonization of C. elegans intestine. Worms grown on OP50, GS67 (antagonist of DB27), and 1A699 (nonantagonist of DB27) were infected with B. thuringiensis DB27 for 4 hr, and DB27 cfu were determined for each case. Ten worms in three replicates were used for each tested group. GS67 significantly (p < ) reduces number of DB27 cfu compared to OP50 or 1A699. Error bars represent 6SEM. (G) Curing of DB27-infected nematodes. Worms (n = 50 in three replicates for each treatment) infected with DB27 for 4 hr were transferred to OP50 plates pretreated with supernatants of GS67 or 1A699 and scored for survival over time. GS67 curing effect is significant (p < ) compared to 1A699 or no-treatment control. Shown are representative results of at least two independent experiments. cfu counts (Figure 2F). In contrast, worms grown on B. subtilis 1A699 accumulated high amounts of B. thuringiensis DB27, similar to OP50-grown worms (Figure 2F). These results suggest that the B. subtilis GS67 antagonistic properties can reduce colonization of C. elegans intestine by the pathogen.

4 B. subtilis Protects C. elegans from Infection 2723 Figure 3. Identification of Inhibitory Molecule (A) Cell-free supernatant of B. subtilis GS67 was subjected to heat (100 C for 1 h) or enzymatic treatment (final enzyme concentration: 1 mg/ml, for 1 hr at 37 C) and tested for the remaining activity in the disc-diffusion assay. Proteinase K (p < 0.01), trypsin, protease, and lipase (p < ) significantly reduce antimicrobial activity. (B) Structures of fengycins A and B. Characteristic fragmentation patterns in ESI + -MS are shown. [M + H] + values are shown in black, and [M + Na] + values are shown in gray. See also Figure S1. Next, we studied whether B. subtilis GS67 can cure worms from an already existing B. thuringiensis DB27 infection. Worms grown on OP50 were infected with B. thuringiensis DB27 and then treated with B. subtilis supernatants. Strikingly, nearly 80% of the worms treated with B. subtilis GS67 supernatant remained alive after 24 hr (Figure 2G) and then exhibited normal lifespan, as if they were never infected (Figure 2G), indicating that B. subtilis GS67 can completely rescue infected worms. In contrast, untreated worms and worms treated with B. subtilis 1A699 supernatant mostly died 24 hr posttreatment (Figure 2G). These results suggest that B. subtilis GS67 produces a factor, for example, a small molecule, with specific activity against B. thuringiensis DB27 and possibly other Gram-positive bacteria. To identify the biochemical nature of the putative inhibitory molecule, cell-free supernatant of B. subtilis GS67 was treated with different enzymes and tested for remaining activity. Treatment with trypsin, protease, and lipase completely eliminated the antagonistic activity, suggesting that a lipopeptide was involved in the inhibition (Figure 3A). Heat stability (Figure 3A) and activity against closely related bacteria (Figure 2E) suggest that this lipopeptide acts similarly to classical bacteriocins, known proteinaceous compounds produced by bacteria to kill closely related species [30]. To identify the inhibitory molecule, we fractionated metabolite extract from 6 l of B. subtilis GS67 culture, using the disc-diffusion inhibition assay to monitor activity. This led to isolation of 12 mg of active material that appeared to represent a mixture of lipopeptides, based on initial nuclear magnetic resonance-spectroscopic analysis. Detailed analysis via high-performance liquid chromatography-mass spectrometry using electrospray ionization mass spectrometry (ESI + -MS) [31] revealed three intense peaks in the base peak chromatogram, with m/z = 1,435.8, 1,449.8, and 1,477.8, along with corresponding doubly charged species (Figure S1 available online). MS/MS analysis of the doubly charged species corresponding to the parent ion of 1,435.8 and 1,449.8 showed fragmentation ions of m/z = 988.4, 1,080.5, and 1,102.5 (Figure S1). Similarly, MS/MS analysis of the parent ion of 1,477.8 showed fragmentation ions of m/z = 1,016.4, 1,108.6, and 1,130.5 (Figure S1). These fragmentation patterns are very characteristic of the well-known lipopeptides fengycin A and fengycin B, which produce fragment ions representing the lactone ring and the lactone ring with an attached ornithine residue fragment (Figure 3B) [32].The mass spectrometric results further indicated the presence of both carbon-14 (C-14) and carbon-15 (C-15) lipid chains in these lipopeptides. Taken together, the purified antagonistic material contained a mixture of C-14 fengycin A, C-15 fengycin A, and C-15 fengycin B, among which C-15 fengycin A was most abundant. To obtain direct evidence for the role of fengycin for the observed phenotype, we generated a mutant defective in fengycin production by disrupting one of the five open reading frames of the fengycin (plipastatin) operon via a single recombination event. The resulting ppsa mutant lost many characteristics typically associated with lipopeptide production (Figure S2). To test whether fengycin is involved in B. thuringiensis DB27 inhibition, we tested the ppsa mutant in bacterial inhibition assays. We found that the fengycindefective mutant exhibits significantly reduced antagonistic activity against B. thuringiensis DB27 in the disc-diffusion assay (Figure 4A) and confirmed this phenotype in coculture experiments. Whereas the wild-type B. subtilis GS67 strain completely eliminated B. thuringiensis DB27 after 24 hr of culture, the ppsa mutant was outcompeted by B. thuringiensis DB27 (Figure 4B). These results indicate that the ppsa mutant has lost the antagonistic properties against B. thuringiensis DB27 and that the fengycins are the major inhibitory molecule produced by B. subtilis GS67. Next, we tested whether the fengycins are also required for the protection of C. elegans against B. thuringiensis DB27. We compared colonization of C. elegans intestines by B. thuringiensis DB27 in worms grown on the ppsa mutant and wild-type GS67. Indeed, worms grown on ppsa mutants showed a much higher pathogen load in their guts compared to GS67-grown worms (Figure 4C), which was not due to the reduced ability of ppsa mutant to persist in C. elegans intestine (Figure S2D). Also, we found that supernatant of the ppsa mutant lost the ability to cure worms from B. thuringiensis DB27 infection because ppsa-treated worms exhibited survival similar to untreated controls (Figure 4D). Finally, we grew worms on ppsa mutants and measured their survival on the B. thuringiensis DB27 pathogen. Worms grown on fengycin-defective mutants showed significantly reduced survival on DB27 (Figure 4E) and S. aureus (Figure 4F) compared to worms grown on wild-type B. subtilis GS67. Thus, fengycin-mediated inhibition of pathogens is the mechanism of B. subtilis GS67-based C. elegans protection. We

5 Current Biology Vol 24 No Figure 4. Fengycin Mediates Protective Effects of B. subtilis GS67 (A) Representative image of disc-diffusion assay showing that ppsa mutant is strongly impaired in antagonism against DB27, as represented by a very small zone of inhibition. (legend continued on next page)

6 B. subtilis Protects C. elegans from Infection 2725 also noticed a strong correlation between bacterial antagonism and nematode protection. Bacillus strains with strong antagonistic activity against DB27 were also able to increase C. elegans resistance to DB27, whereas none of the nonantagonistic strains showed the protective effect (Figure S3). Thus, the strong correlation between antagonism and protection suggests that this antagonism is the crucial mechanism behind commensal-mediated C. elegans protection. In the next set of experiments, we quantified the effect of fengycin and tested the purified compound mixture for its protective effect on C. elegans. Specifically, we determined the minimal inhibitory concentration (MIC) for the purified mixture of fengycins using the broth dilution method. We found that 2.5 mg/ml was sufficient to inhibit DB27 growth. To assess the viability of fengycin-treated DB27 cells, we applied LIVE/ DEAD staining and measured growth of DB27 in the presence of fengycin. In contrast to viable control cells that stain green (Figure 4G), most of the fengycin-treated cells stain red (Figure 4H), indicating they are not viable and have compromised membranes. Consistent with this, fengycin exhibits a dosedependent lytic effect on DB27, as indicated by decline in optical density (OD) (Figure 4I) and cfu (Figure S4A) values over time. Finally, we prepared nematode growth medium (NGM) plates with different concentrations of the purified mixture of fengycins and used them to treat DB27-infected worms. Indeed, fengycins rescue infected worms in a dose-dependent manner (Figure 4J). Whereas plates with the MIC showed only a mild increase in survival, we found that at concentrations of 15 mg/ml, nearly 80% of worms survived (Figure 4J), indicating that purified fengycins can cure infected worms directly. Together, these results provide direct evidence for the role of fengycin in C. elegans protection. To test whether fengycin could also protect worms indirectly via stimulation of an immune response, we first assessed by quantitative PCR (qpcr) the expression of ten DB27-responsive genes in worms treated with fengycin. DB27-responsive genes were identified in a previous study analyzing the transcriptional response of C. elegans after exposure to different bacteria [5]. Five of these ten genes did not change their expression, whereas the other five were even repressed upon fengycin treatment (Figure S4B). Finally, immunecompromised pmk-1 mutants still exhibited increased resistance to DB27 after GS67 treatment (Figure S4C), suggesting that GS67 protects worms independently of an immune stimulation. We have shown that replacement of E. coli OP50 with natural Bacillus strains can strongly increase C. elegans resistance to pathogens. The strongest effect was observed for B. subtilis GS67, which promoted C. elegans survival on B. thuringiensis DB27 up to 3-fold but had no effect on resistance to Gramnegative pathogens. Our study provides direct evidence for a commensal-mediated mechanism of protection by antagonism of pathogens via antibiotic production, a mechanism previously discussed in vertebrates [33, 34]. Combining chemical and genetic analysis, we identified the lipopeptide fengycin, a nonribosomally synthesized lipopeptide primarily known to act against fungi [35, 36], to be responsible for the observed antagonism. Fengycin-producing B. subtilis strains have been applied in biological control of plant pathogenic fungi [37], but the spectrum of activity and the exact mechanism of action are not fully understood. Bacteriocins and related lipopeptides have protective properties in other animal models as well. For example, it was shown that a probiotic strain of Lactobacillus salivarius produces a bacteriocin in vivo, which can protect mice against Listeria monocytogenes infection [38]. Similarly, plipastatin (fengycin) was identified as the major B. subtilis molecule responsible for S. aureus inhibition, and administration of fengycin was shown to restrict growth of S. aureus on murine skin [39]. Therefore, bacteriocins and lipopeptides might be effective alternatives to classical antibiotics, given the growing problem of antibiotic resistance. In addition, our results indicate that C. elegans can be used as an in vivo model to reveal protective mechanisms of commensal bacteria. In conclusion, our work suggests that the commensal-mediated B. subtilis GS67 protective effect on C. elegans is primarily mediated by its direct activity against the pathogen. This finding highlights the importance of interspecies commensal-pathogen interplay for the outcome of C. elegans host-pathogen interactions. Thus, the protective mechanisms of commensal bacteria against C. elegans pathogens are more complex than previously recognized and have to be taken into account in studies addressing the defensive role of commensals. Experimental Procedures Bacillus Feeding Assay Each Bacillus strain tested was grown in a shaking incubator at 30 C overnight in lysogeny broth (LB), and 40 ml was spread on 6 cm NGM plates and incubated overnight. C. elegans eggs obtained by bleaching were placed in these NGM plates, with either Bacillus (experiment) or E. coli OP50 (control, normal conditions). Plates were incubated at 25 C until worms reached L4/ young adult stage and used in a pathogen survival assays. Detection of Antagonistic Activity The disc-diffusion assay was used for the test of antagonistic activities of B. subtilis GS67 against other bacteria. Briefly, 50 ml of overnight cultures (B) ppsa mutant does not inhibit DB27 in coculture. When inoculated at equal doses, wild-type GS67 eliminates DB27 after 24 hr, whereas ppsa mutant is overgrown by DB27. Growth of ppsa mutant alone is comparable (p > 0.05) to the growth of GS67 and DB27. DB27 exhibits significantly (p < , DB27 versus ppsa:db27) reduced growth when cocultured with ppsa mutant. Error bars represent 6SEM. (C) Colonization of C. elegans intestine is not affected by ppsa mutant (p > 0.05 OP50 versus ppsa) in contrast to wild-type GS67 (p < OP50 versus GS67). Ten worms in three replicates were used for each tested group. Error bars represent 6SEM. See also Figure S2. (D) ppsa mutant does not cure worms that have the DB27 infection. Survival of DB27-infected worms treated with the supernatant of ppsa mutant is comparable (p > 0.05) to untreated control worms. Shown are representative results of at least two independent experiments (n = 50 in three replicates for each treatment). (E and F) ppsa mutant lost protective properties against DB27 (E) and S. aureus (F) infections. Worms grown on ppsa mutant exhibited survival on DB27 (E) and S. aureus (F) comparable (p > 0.05) to control worms grown on OP50. Error bars represent 6SEM. (G and H) Representative images of DB27 cells either untreated (G) or treated (H) with fengycin (15 mg/ml, 3 hr) and stained with BacLight bacterial viability kit. Viable bacteria stain green; nonviable cells with compromised membranes stain red. Scale bars represent 20 mm. (I) Bactericidal effect of fengycin. Aliquots of DB27 culture (OD = 0.4) supplemented with defined doses of fengycin were incubated without shaking at 30 C. OD 600 values were used to measure the growth dynamic over time. See also Figure S4. (J) Purified fengycin cures DB27-infected worms. Worms infected (n = 20 in three replicates per treatment) with DB27 for 4 hr were transferred to NGM plates with defined concentration of fengycin. Surviving worms, free of DB27 infection, were scored 24 hr posttransfer. Curing effect of fengycin is significant (p < 0.05) compared to untreated control at 5 mg/ml, 10 mg/ml, and 15 mg/ml. Error bars represent 6SEM.

7 Current Biology Vol 24 No of test bacteria with adjusted cell densities (OD = 1.0) was spread on 10 cm LB plates. Paper discs (Whatman, 6 mm) were placed on top of the agar, and 30 ml of the cell-free, filtered supernatant of B. subtilis GS67 grown for 48 hr in LB at 30 C was added to paper discs; LB was used as a negative control. The plates were incubated for 24 hr at 25 C and examined for clear inhibition zone around the discs. The experiment was done in triplicates and repeated at least three times. Mutant Strain Construction An integrative plasmid was used to disrupt the production of fengycin. To construct the plasmid pmut-ppsa, we PCR amplified a 1 kb fragment of ppsa gene from B. subtilis GS67 and ligated it into BamHI site of pmutin vector. To generate ppsa mutant, we transformed the wild-type strain B. subtilis GS67 by electroporation with the pmut-ppsa plasmid. Positive transformants were selected based on erythromycin resistance and were molecularly confirmed using PCR and sequencing. Supplemental Information Supplemental Information includes Supplemental Experimental Procedures and four figures and can be found with this article online at /j.cub Author Contributions I.I. and R.J.S. conceived the study and designed the experiments. J.J.Y. purified antagonistic substance from GS67 culture and performed fractionations and ESI-MS. F.C.S. provided equipment and reagents. J.J.Y. and F.C.S. performed the ESI-MS analysis. I.I. performed all other experiments. I.I., R.J.S., and J.J.Y. wrote the manuscript. Acknowledgments We thank the Caenorhabditis Genetic Center, which is funded by NIH Office of Research Infrastructure Programs (P40 OD010440), and the Bacillus Genetic Stock Center for providing strains. We also thank D. Wistuba (University of Tübingen) for help with mass spectrometry. This work was funded by the Max Planck Society and by DFG RTG1708 (to R.J.S.). Received: July 29, 2014 Revised: September 4, 2014 Accepted: September 23, 2014 Published: November 6, 2014 References 1. Irazoqui, J.E., Urbach, J.M., and Ausubel, F.M. (2010). Evolution of host innate defence: insights from Caenorhabditis elegans and primitive invertebrates. Nat. Rev. Immunol. 10, Nicholas, H.R., and Hodgkin, J. (2002). Innate immunity: the worm fights back. Curr. Biol. 12, R731 R Hodgkin, J., Félix, M.-A., Clark, L.C., Stroud, D., and Gravato-Nobre, M.J. (2013). Two Leucobacter strains exert complementary virulence on Caenorhabditis including death by worm-star formation. Curr. Biol. 23, Tan, M.-W., and Shapira, M. (2011). Genetic and molecular analysis of nematode-microbe interactions. Cell. Microbiol. 13, Sinha, A., Rae, R., Iatsenko, I., and Sommer, R.J. (2012). System wide analysis of the evolution of innate immunity in the nematode model species Caenorhabditis elegans and Pristionchus pacificus. PLoS ONE 7, e Schulenburg, H., Kurz, C.L., and Ewbank, J.J. (2004). Evolution of the innate immune system: the worm perspective. Immunol. Rev. 198, Pukkila-Worley, R., and Ausubel, F.M. (2012). Immune defense mechanisms in the Caenorhabditis elegans intestinal epithelium. Curr. Opin. Immunol. 24, Watson, E., MacNeil, L.T., Ritter, A.D., Yilmaz, L.S., Rosebrock, A.P., Caudy, A.A., and Walhout, A.J.M. (2014). Interspecies systems biology uncovers metabolites affecting C. elegans gene expression and life history traits. Cell 156, Gusarov, I., Gautier, L., Smolentseva, O., Shamovsky, I., Eremina, S., Mironov, A., and Nudler, E. (2013). Bacterial nitric oxide extends the lifespan of C. elegans. Cell 152, Cabreiro, F., and Gems, D. (2013). Worms need microbes too: microbiota, health and aging in Caenorhabditis elegans. EMBO Mol Med 5, Clark, L.C., and Hodgkin, J. (2014). Commensals, probiotics and pathogens in the Caenorhabditis elegans model. Cell. Microbiol. 16, Kim, D.H. (2013). Bacteria and the aging and longevity of Caenorhabditis elegans. Annu. Rev. Genet. 47, Kim, Y., and Mylonakis, E. (2012). Caenorhabditis elegans immune conditioning with the probiotic bacterium Lactobacillus acidophilus strain NCFM enhances gram-positive immune responses. Infect. Immun. 80, Montalvo-Katz, S., Huang, H., Appel, M.D., Berg, M., and Shapira, M. (2013). Association with soil bacteria enhances p38-dependent infection resistance in Caenorhabditis elegans. Infect. Immun. 81, Mahajan-Miklos, S., Tan, M.W., Rahme, L.G., and Ausubel, F.M. (1999). Molecular mechanisms of bacterial virulence elucidated using a Pseudomonas aeruginosa-caenorhabditis elegans pathogenesis model. Cell 96, Hodgkin, J., Kuwabara, P.E., and Corneliussen, B. (2000). A novel bacterial pathogen, Microbacterium nematophilum, induces morphological change in the nematode C. elegans. Curr. Biol. 10, Aballay, A., Yorgey, P., and Ausubel, F.M. (2000). Salmonella typhimurium proliferates and establishes a persistent infection in the intestine of Caenorhabditis elegans. Curr. Biol. 10, Labrousse, A., Chauvet, S., Couillault, C., Kurz, C.L., and Ewbank, J.J. (2000). Caenorhabditis elegans is a model host for Salmonella typhimurium. Curr. Biol. 10, Rae, R., Iatsenko, I., Witte, H., and Sommer, R.J. (2010). A subset of naturally isolated Bacillus strains show extreme virulence to the freeliving nematodes Caenorhabditis elegans and Pristionchus pacificus. Environ. Microbiol. 12, Iatsenko, I., Boichenko, I., and Sommer, R.J. (2014). Bacillus thuringiensis DB27 produces two novel protoxins, Cry21Fa1 and Cry21Ha1, which act synergistically against nematodes. Appl. Environ. Microbiol. 80, Iatsenko, I., Sinha, A., Rödelsperger, C., and Sommer, R.J. (2013). New role for DCR-1/dicer in Caenorhabditis elegans innate immunity against the highly virulent bacterium Bacillus thuringiensis DB27. Infect. Immun. 81, Iatsenko, I., Corton, C., Pickard, D.J., Dougan, G., and Sommer, R.J. (2014). Draft genome sequence of highly nematicidal Bacillus thuringiensis DB27. Genome Announc 2, e00101 e Hui, F., Scheib, U., Hu, Y., Sommer, R.J., Aroian, R.V., and Ghosh, P. (2012). Structure and glycolipid binding properties of the nematicidal protein Cry5B. Biochemistry 51, Bischof, L.J., Kao, C.-Y., Los, F.C.O., Gonzalez, M.R., Shen, Z., Briggs, S.P., van der Goot, F.G., and Aroian, R.V. (2008). Activation of the unfolded protein response is required for defenses against bacterial pore-forming toxin in vivo. PLoS Pathog. 4, e Bellier, A., Chen, C.-S., Kao, C.-Y., Cinar, H.N., and Aroian, R.V. (2009). Hypoxia and the hypoxic response pathway protect against pore-forming toxins in C. elegans. PLoS Pathog. 5, e Griffitts, J.S., Haslam, S.M., Yang, T., Garczynski, S.F., Mulloy, B., Morris, H., Cremer, P.S., Dell, A., Adang, M.J., and Aroian, R.V. (2005). Glycolipids as receptors for Bacillus thuringiensis crystal toxin. Science 307, Wei, J.-Z., Hale, K., Carta, L., Platzer, E., Wong, C., Fang, S.-C., and Aroian, R.V. (2003). Bacillus thuringiensis crystal proteins that target nematodes. Proc. Natl. Acad. Sci. USA 100, Kao, C.-Y., Los, F.C.O., Huffman, D.L., Wachi, S., Kloft, N., Husmann, M., Karabrahimi, V., Schwartz, J.-L., Bellier, A., Ha, C., et al. (2011). Global functional analyses of cellular responses to pore-forming toxins. PLoS Pathog. 7, e Los, F.C.O., Kao, C.-Y., Smitham, J., McDonald, K.L., Ha, C., Peixoto, C.A., and Aroian, R.V. (2011). RAB-5- and RAB-11-dependent vesicletrafficking pathways are required for plasma membrane repair after attack by bacterial pore-forming toxin. Cell Host Microbe 9, Klaenhammer, T.R. (1993). Genetics of bacteriocins produced by lactic acid bacteria. FEMS Microbiol. Rev. 12, Ma, Z., Hu, J., Wang, X., and Wang, S. (2014). NMR spectroscopic and MS/MS spectrometric characterization of a new lipopeptide antibiotic bacillopeptin B1 produced by a marine sediment-derived Bacillus amyloliquefaciens SH-B74. J. Antibiot. 67,

8 B. subtilis Protects C. elegans from Infection Wang, J., Liu, J., Wang, X., Yao, J., and Yu, Z. (2004). Application of electrospray ionization mass spectrometry in rapid typing of fengycin homologues produced by Bacillus subtilis. Lett. Appl. Microbiol. 39, Buffie, C.G., and Pamer, E.G. (2013). Microbiota-mediated colonization resistance against intestinal pathogens. Nat. Rev. Immunol. 13, Kamada, N., Chen, G.Y., Inohara, N., and Núñez, G. (2013). Control of pathogens and pathobionts by the gut microbiota. Nat. Immunol. 14, Steller, S., Vollenbroich, D., Leenders, F., Stein, T., Conrad, B., Hofemeister, J., Jacques, P., Thonart, P., and Vater, J. (1999). Structural and functional organization of the fengycin synthetase multienzyme system from Bacillus subtilis b213 and A1/3. Chem. Biol. 6, Romero, D., de Vicente, A., Rakotoaly, R.H., Dufour, S.E., Veening, J.- W., Arrebola, E., Cazorla, F.M., Kuipers, O.P., Paquot, M., and Pérez- García, A. (2007). The iturin and fengycin families of lipopeptides are key factors in antagonism of Bacillus subtilis toward Podosphaera fusca. Mol. Plant Microbe Interact. 20, Ongena, M., and Jacques, P. (2008). Bacillus lipopeptides: versatile weapons for plant disease biocontrol. Trends Microbiol. 16, Corr, S.C., Li, Y., Riedel, C.U., O Toole, P.W., Hill, C., and Gahan, C.G.M. (2007). Bacteriocin production as a mechanism for the antiinfective activity of Lactobacillus salivarius UCC118. Proc. Natl. Acad. Sci. USA 104, Gonzalez, D.J., Haste, N.M., Hollands, A., Fleming, T.C., Hamby, M., Pogliano, K., Nizet, V., and Dorrestein, P.C. (2011). Microbial competition between Bacillus subtilis and Staphylococcus aureus monitored by imaging mass spectrometry. Microbiology 157,