Caenorhabditis elegans Is a Model Host for Listeria monocytogenes

Here we report that Caenorhabditis elegans nematodes fed Listeriamonocytogenes die over the course of several days, as a consequenceof an accumulation of bacteria in the worm intestine. Mutantstrains previously shown to be important for virulence in mammalianmodels were also found to be attenuated in their virulence inC. elegans. However, ActA, which is required for actin-basedintracellular motility, appears to be dispensable during infectionof C. elegans, indicating that L. monocytogenes remains extracellularin C. elegans.

INTRODUCTION

Top

Introduction

There is a continuing need for the development of simple animalmodels for the study of host-pathogen interactions. The implementationof nonmammalian model organisms such as Drosophila melanogasterand Caenorhabditis elegans has recently proven successful, therebyreducing both cost and the ethical constraints associated withvirulence studies using mammalian hosts. The realization thatinnate immunity plays a significant role in resistance to bacterialinfections has further supported the use of nonmammalian modelorganisms which display an innate immune response but lack theadaptive immune system (6).

Listeria monocytogenes kills C. elegans.

Top

Listeria monocytogenes kills c....

Listeria monocytogenes is a gram-positive facultative intracellularbacterial pathogen that gives rise to serious invasive infectionsin humans and animals. With the aim of investigating whetherListeria is also a pathogen of C. elegans, we examined the killingof the nematode when it was fed various L. monocytogenes strains.While the wild-type N2 worms were feeding on Listeria, locomotionwas progressively decreased and the number of eggs laid wasdecreased when compared to OP50 (data not shown). In addition,we observed that the worms became laden with eggs, which insome cases hatched internally, a phenotype called "bag of worms"(data not shown). Bagging is suggested to be caused by a weakeningof the infected worm, rendering it unable to lay eggs normally(12). Since killing assays extend over the generation time ofC. elegans, new progeny may interfere with the counting of survivingnematodes. Therefore, to avoid new progeny and the bagging phenotype,we used the temperature-sensitive sterile C. elegans pha-1(e2123ts),which grows normally at 15°C but at 25°C is embryoniclethal (11). Mutation in the pha-1 gene does not affect theimmune function of C. elegans (data not shown). Individual bacterialstrains were grown overnight in either Luria broth (LB) or brainheart infusion broth at 37°C. Ten microliters of overnightculture was spread on 3.5-cm LB agar plates and grown for 16h at 37°C. Thirty L4 hermaphrodites were transferred fromnematode growth medium plates seeded with OP50 to each 3.5-cmLB agar plate seeded with the various strains, and plates wereincubated at 25°C and scored for live or dead worms every24 h. Each assay was carried out in triplicate and repeatedthree times. We examined the ability of several well-characterizedL. monocytogenes strains to kill C. elegans. The wild-type strainsEGDe, EGD (BUG600), and 10403S are known to be virulent in amouse model (2, 3, 8), and we found that they also kill C. elegansmuch faster compared to the avirulent Escherichia coli OP50and Listeria innocua (Fig. 1). Furthermore, when nematodes werefed heat-killed L. monocytogenes EGDe or BUG600, no killingwas observed and we saw a life span comparable to that of wormsfed OP50 (results not shown), indicating the need for directassociation between viable L. monocytogenes and C. elegans forkilling.

View larger version (16K):
[in this window]
[in a new window]
FIG. 1. Survival of C. elegans pha-1(e2123ts) fed on E. coli and Listeria. (A) Worms fed E. coli OP50 (dotted lines, diamonds), Listeria innocua (triangles), L. monocytogenes 10403S (solid squares), or L. monocytogenes 10403S ${Delta}$ actA (dashed lines, open squares). (B) Worms fed E. coli OP50 (dotted lines, diamonds), L. monocytogenes EGDe (dashed lines, solid circles), L. monocytogenes EGDe ${Delta}$ prfA (dashed lines, open circles), L. monocytogenes EGD (BUG600) (solid squares), or L. monocytogenes EGD (BUG600) ${Delta}$ degU (open squares).

Worms fed L. monocytogenes accumulate bacteria within their digestive tracts.

Top

Worms fed l. monocytogenes...

Two distinct modes of bacterium-mediated killing of C. eleganshave been identified. Certain Pseudomonas aeruginosa strainscan produce low-weight molecular toxins that can kill the nematodewithin hours, known as "fast killing" (13). In contrast, whenthe killing occurs over the course of several days, it is referredto as "slow killing" and correlates with the accumulation ofbacteria in the nematode gut (13). To follow the fate of L.monocytogenes, we fed C. elegans with L. monocytogenes EGDethat expresses green fluorescent protein (GFP) (5), and, after72 h, we observed a distension of the intestinal lumen witha large number of intact bacteria (data not shown). This distensionwas not observed when worms were fed OP50 expressing GFP (datanot shown). Similar results have previously been found whenC. elegans was fed GFP-expressing Salmonella enterica serovarTyphimurium for 72 h (1).

C. elegans can be used to identify L. monocytogenes virulence factors.

Top

C. elegans can be...

Several bacterial pathogens, both gram positive and gram negative,kill C. elegans when supplied as a food source, and a varietyof bacterial virulence factors, identified in mammalian models,have also been shown to be required for nematode pathogenesis(9, 12, 13). To evaluate whether the C. elegans model will beuseful for studying and identifying Listeria virulence factors,mutant strains that exhibit a reduced virulence in mammals wereexamined. PrfA, a regulator of virulence gene expression, andDegU, a two-component regulator involved in motility, are bothimportant for virulence in mice (3, 8). We investigated theeffect of prfA and degU mutants and found both mutants to havereduced nematocidal activity (Fig. 1B), demonstrating that thevirulence factors PrfA and DegU are required for L. monocytogenesvirulence in C. elegans as well as in vertebrate hosts. In additionto PrfA L. monocytogenes utilizes several factors during itsintracellular life. One of these is ActA, which mediates actin-basedmotility in the host cytosol and is required for the bacteriumto spread to adjacent cells (4). actA mutants are known to behighly attenuated in a murine model (2). However, we found thatan actA mutant had the same ability to kill the nematode asthe wild type (Fig. 1A), indicating that intracellular survivaland spread are not important for killing C. elegans. This observationis consistent with our observation that Listeria expressingGFP was restricted to the luminal space (data not shown). Inaddition, extracellular replication of L. monocytogenes is notrestricted to C. elegans but has also recently been demonstratedin the murine model (7).

In conclusion, we report here the establishment of the nematodeCaenorhabditis elegans as a model for analyzing the virulenceof L. monocytogenes. Our results show that several genes importantfor full virulence in vertebrates are also required for killingC. elegans, suggesting that the nematode could be an attractivemodel for identifying new virulence factors in L. monocytogenes.In D. melanogaster, another invertebrate model used for L. monocytogenes(10), an actA mutant is able to replicate within the insectcells but is unable to spread to adjacent cells (10). However,ActA appears to be dispensable during the infection of the worm,and future work will reveal the genetic repertoire requiredfor infection of C. elegans by L. monocytogenes.

ACKNOWLEDGMENTS

We thank D. A. Portnoy, G. Knudsen, L. Dons, N. Fortinea, andW. Goebel for generous gifts of strains and plasmids. The E.coli and C. elegans strains used in this work were providedby the Caenorhabditis Genetics Center (University of Minnesota,Minneapolis).

This work was supported by a grant from the Danish TechnicalResearch Council and a fellowship from the Howard Hughes MedicalInstitute.

FOOTNOTES

* Corresponding author. Mailing address: Department of Veterinary Pathobiology, The Royal Veterinary and Agricultural University (KVL), Stigbøjlen 4, DK-1870 Frederiksberg C, Denmark. Phone: 45 35282773. Fax: 45 35282755. E-mail: hi{at}kvl.dk.

REFERENCES

Top