Tolerance of Listeria monocytogenes to Cell Envelope-Acting Antimicrobial Agents Is Dependent on SigB

Mutation of sigB impairs the ability of Listeria monocytogenesto grow in sublethal levels, and to survive in lethal concentrations,of the bacteriocins nisin and lacticin 3147 and the antibioticsampicillin and penicillin G. SigB may therefore represent anattractive target for the development of new control and treatmentstrategies for this important pathogen.

INTRODUCTION

Top

Introduction

Listeria monocytogenes is the etiological agent of listeriosis,an opportunistic infection that affects primarily pregnant andimmunocompromised individuals. Food is the major source of infection,and those foods most frequently implicated include soft cheeses,dairy products, salads, and refrigerated ready-to-eat products(6). The ß-lactams ampicillin and penicillin G arethe antibiotics of choice in the treatment of listeriosis (12);however, despite their therapeutic use, up to one-third of patientsdie (16). As a result, listeriosis is a significant cause ofmortality due to food-borne disease; it is estimated to be responsiblefor approximately 27.6% of food-related deaths in the UnitedStates annually (16). L. monocytogenes is recognized as a seriousrisk to public health and food safety, and the bacteriumwas responsible for 71% of all recalls of food products dueto bacterial contamination in the United States between 1993and 1998 (25).

L. monocytogenes must overcome the numerous environmental extremesencountered during food processing, handling, and storage andin vivo following consumption. One important mediator of thebacterium's stress responses is the alternative sigma factorSigB. It has been shown to assist the in vitro survival of cellsunder a variety of environmental insults, including low pH,high osmolarity, and elevated bile concentrations, and duringoxidative stress and carbon starvation (1-3, 7, 8). It is alsobecoming increasingly evident that SigB regulates stress lociimportant for intrahost survival, such as bile salt hydrolase(bsh) (3, 22), and also assists in the regulation of dedicatedvirulence factors such as the principal virulence regulatorPrfA (19, 20).

Bioinformatic analysis of the SigB regulon.

Top

Bioinformatic analysis of the...

The entire L. monocytogenes SigB regulon has not yet been defined;however, initial efforts using a 208-gene microarray identified55 genes positively regulated by SigB (i.e., their expressionis significantly reduced in a sigB mutant) (13). Analysis ofthe data produced by that study and examination of the sequencedgenome of L. monocytogenes strain EGDe with the SigB consensuspromoter sequence that was generated (GTTTN_13-17GGGWAT was enteredas a "Pattern search" in the genome web server ListiList [http://genolist.pasteur.fr/ListiList/],and the search was restricted to patterns located within 350bp upstream of a predicted open reading frame and showing onemismatch to the consensus) revealed a number of loci which,based on homology searches, may contribute to tolerance of antimicrobialcompounds. These loci encode putative efflux pumps, penicillinbinding proteins, autolysins, or proteins involved in the modificationof the cell envelope (Table 1). It was therefore decided toinvestigate whether the SigB regulon contributes to the toleranceof bacteriocins (peptide antimicrobials) and antibiotics (nonpeptideantimicrobials).

View this table:
[in this window]
[in a new window]
TABLE 1. Genetic loci that are regulated/putatively regulated by SigB and that may contribute to the tolerance of antimicrobial agents

SigB contributes to bacteriocin tolerance.

Top

Sigb contributes to bacteriocin...

The growth rate of a sigB mutant (a nonpolar internal deletionmutant) (24) was comparable to that of the wild type when grownin brain heart infusion (BHI) broth at 37°C, indicatingthat SigB is not required for growth under normal physiologicalconditions (data not shown). Addition of sublethal levels ofnisin (a bacteriocin currently used as a biopreservative) (5)or lacticin 3147 (a bacteriocin that shows potential as a biopreservative)(11) had noticeable effects on growth (Fig. 1A and B). Survivalassays performed with lethal levels of each bacteriocin revealedmore-dramatic differences at every concentration tested. Althoughthere was an initial kill of the parent strain at the levelsshown in Fig. 1, it was detected at high numbers over the 6-hexperiment, whereas the mutant was not detected after 1 h inthe broth supplemented with nisin or after 4 h in the brothsupplemented with lacticin. At bacteriocin levels higher thanthese, the parent was rapidly and completely inactivated. Interestingly,agar well diffusion assays (not shown), overlay assays (Fig.1E and F), and MIC determinations by the broth dilution method(using twofold serial dilutions of the bacteriocins in BHI broth)(data not shown) did not reveal any differences between thestrains. However, these experiments are analyzed at a singletime point after 24-h incubation periods, after which the numbersof wild-type and mutant bacteria were identical in our experiments(data not shown). Overall, our data strongly suggest an importantrole for SigB in bacteriocin tolerance, which may be importantfor the survival of L. monocytogenes in foods. The results alsohighlight the importance of assay selection in the assessmentof the contribution of a specific locus to bacteriocin tolerance.For example, Moorehead and Dykes (17) concluded that SigB didnot play a role in L. monocytogenes tolerance of the bacteriocinsnisin and sakacin A as determined by overlay assays.

View larger version (29K):
[in this window]
[in a new window]
FIG. 1. (A to D) Growth and survival of wild-type L. monocytogenes 10403S ( ${blacksquare}$ ) and the sigB mutant ( ${circ}$ ) in broth supplemented with bacteriocins. Overnight cultures were inoculated (3%) into BHI broth supplemented with either $~$ 45 AU/ml lacticin 3147 (A), 50 µg/ml nisin (B), $~$ 100 AU/ml lacticin 3147 (C), or 300 µg/ml nisin (D). (Lacticin was prepared as described in reference 14, and nisin powder was obtained from Sigma.) Cultures were incubated with shaking at 37°C. Cell growth was measured spectrophotometrically by determining the optical density at 600 nm. Viable cell counts were performed by serial dilution in one-quarter-strength Ringer's solution and enumeration on BHI agar. Error bars, standard deviations from three independent experiments. (E and F) Overlay assays of wild-type (wt) and sigB mutant ( ${Delta}$ sigB) L. monocytogenes with bacteriocin-producing strains. Ten microliters of the lacticin 3147 overproducer Lactococcus lactis subsp. cremoris MG1363 plus pMRCO1 and pOM02 (E) or the nisin producer Lactococcus lactis subsp. cremoris pNZ9700 (F) was spotted onto GM17 agar. After overnight incubation at 30°C, spots were UV treated for 10 min and then overlaid with wild-type L. monocytogenes 10403S or the ${Delta}$ sigB strain in BHI semisolid agar. Plates were incubated overnight at 37°C.

SigB contributes to antibiotic tolerance.

Top

Sigb contributes to antibiotic...

The abilities of the wild type and the sigB mutant to withstandexposure to penicillin G and ampicillin were compared, becausethese are the antibiotics of choice in the treatment of listeriosis(12). Initial agar diffusion experiments with antibiotic disks(Oxoid) revealed that the diameters of the zones of bacterialgrowth inhibition surrounding the filter disks were similarfor the two strains (Fig. 2E and F). In addition, there wasno difference in MICs as determined by the broth dilution method(using twofold serial dilution of the antibiotics in BHI broth)(data not shown). However, detailed growth curves (Fig. 2A and B)and survival assays (Fig. 2C and D) revealed that the sigB mutantwas significantly impaired in growth in sublethal levels ofeach antibiotic and was killed more rapidly at lethal levels.It is therefore possible that SigB contributes significantlyto the survival of L. monocytogenes in clinical settings. Designingfuture therapies to target SigB may improve the treatment oflisteriosis, which is presently inefficient.

View larger version (28K):
[in this window]
[in a new window]
FIG. 2. (A to D) Growth and survival of wild-type L. monocytogenes 10403S ( ${blacksquare}$ ) and the sigB mutant ( ${circ}$ ) in broth supplemented with antibiotics. Overnight cultures were inoculated (3%) into BHI broth supplemented with either 0.02 µg/ml penicillin G (A), 0.05 µg/ml ampicillin (B), 32 µg/ml penicillin G (C), or 32 µg/ml ampicillin (D). (Antibiotic powders were purchased from Sigma and solubilized in water.) Cultures were incubated with shaking at 37°C. Cell growth was measured spectrophotometrically by determining the optical density at 600 nm. Viable cell counts were performed by serial dilution in one-quarter-strength Ringer's solution and enumeration on BHI agar. Error bars, standard deviations from three independent experiments. (E and F) Antibiotic disk diffusion assays. Overnight wild-type and ${Delta}$ sigB cultures were diluted 1:10 in Ringer's solution and swabbed onto BHI agar plates (20 ml). Commercially purchased disks (diameter, 6 mm; Oxoid) containing penicillin G (E) or ampicillin (F) were placed on the surface using a sterile forceps. Plates were incubated at 37°C overnight, after which the diameters of the zones of bacterial growth inhibition surrounding the filter disks were measured.

Role of SigB in the tolerance of antimicrobial agents.

Top

Role of sigb in...

The exact role of SigB in L. monocytogenes tolerance of antimicrobialagents has yet to be determined. It is likely that SigB playsa role in controlling membrane characteristics (e.g., chargeor lipid composition) and that altering these properties significantlyaffects the cell's ability to tolerate antimicrobial compounds.It is noteworthy that bacteriocins and antibiotics act on bacterialcell walls and that L. monocytogenes sigB mutants have beenshown to be significantly more sensitive than parent cells tostresses that exert their effects on the cell wall, such asbile (3). In addition, cell surface alterations have previouslybeen shown to be important in the tolerance of several bacteriocinsand antibiotics (4, 9, 10, 18, 23). SigB may also regulate generalstress proteins or proteins involved in extrusion of antimicrobialsout of the cell. Indeed, htrA (lmo0292), which encodes a putativemolecular chaperone that has been shown to be involved in toleranceof penicillin G (21), and mdrL (lmo1409), which encodes an antibioticefflux pump (15), both possess consensus SigB binding sites.

In conclusion, we report our novel observation that SigB contributespositively to L. monocytogenes tolerance of the bacteriocinsnisin and lacticin 3147 and of the antibiotics ampicillin andpenicillin G. Because SigB may contribute to the survival ofL. monocytogenes both in the food-processing environment andin vivo during infection, it may represent an attractive targetfor the development of new control and treatment strategiesfor this important pathogen.

ACKNOWLEDGMENTS

We acknowledge the funding received from the Irish Governmentunder the National Development Plan 2000—2006 and throughfunding of the Alimentary Pharmabiotic Centre by the ScienceFoundation of Ireland Centres for Science Engineering and Technology(CSET) scheme.

We also thank Martin Wiedmann (Cornell University) and PaulCotter (University College Cork) for supplying L. monocytogenesand Lactococcus lactis strains, respectively.

FOOTNOTES

* Corresponding author. Mailing address: Department of Microbiology and Alimentary Pharmabiotic Centre, University College Cork, Cork, Ireland. Phone: 353-21-4903173. Fax: 353-21-4901373. E-mail: c.hill{at}ucc.ie.

REFERENCES

Top