Surface Attachment of Listeria monocytogenes Is Induced by Sublethal Concentrations of Alcohol at Low Temperatures

Sublethal concentrations of ethanol or isopropanol increasedattachment of Listeria monocytogenes at 10, 20, or 30°C;no induction occurred at 37°C. The alcohol induction phenotypewas retained in sigB and cesRK mutants; however, the degreeof induction was affected. These results suggest that alcoholmay contribute to the persistence of L. monocytogenes.

INTRODUCTION

Top

Introduction

The persistence of pathogenic bacteria, such as Listeria monocytogenes,in food production plants presents a serious problem and hasbeen suggested to relate to surface attachment. Ethanol andisopropanol mixtures are often used as disinfectants and cleaningagents. We investigated the influence of sublethal concentrationsof alcohol on the surface attachment of Listeria monocytogenesat various temperatures, the role of possible regulators, andstrain variation in the response.

Alcohol-induced attachment at various temperatures.

Top

Alcohol-induced attachment at...

The influence of sublethal concentrations of ethanol or isopropanolon attachment at 10, 20, 30, or 37°C was determined forthree wild-type L. monocytogenes strains, EGD (2), LO28 (12),and 412 (6). Surface attachment was quantified using a semiautomaticmicrotiter plate assay which was developed from manual assaysdescribed previously (4, 11). Overnight cultures in brain heartinfusion broth (BHI) at 30°C without shaking were dilutedin BHI or BHI with 2.5% (vol/vol) ethanol or isopropanol, and150-µl portions were loaded in polystyrene flat-bottomed96-well microtiter plates (Greiner Bio-One GmbH, Frickenhausen,Germany) in 8 replicate wells. The plates were sealed with adhesivesealing tape (Nunc A/S, Roskilde, Denmark) to minimize evaporationand incubated at 10, 20, 30, and 37°C for 8, 2, 1, and 1days, respectively. The adhered cells were quantified usinga Biomek 2000 laboratory automation workstation (Beckman Coulter,Fullerton, California). The pipetting robot removed the culture,washed the wells twice with 160 µl 0.9% NaCl, added 160µl 0.1% (wt/vol) crystal violet, and after 10 min stainingremoved the crystal violet, performed two washes with 170 µl0.9% NaCl, and added 180 µl 96% ethanol. After 1/2 hourof destaining, the eluted crystal violet was measured as A₅₉₀(absorbance at 590 nm) using a Wallac Victor2 1420 microplatereader (Perkin Elmer, Wellesley, Massachusetts). Two independentexperiments were made for strains EGD and 412, and one experimentwas made for strain LO28. The significance of differences betweendata was determined by pairwise tests with the general linearmodel procedure of the SAS system statistical software, release8.02 TS level 02MO, Windows version 5.1.2600 (SAS InstituteInc., Cary, NC), using the statement LSMEANS and the optionPDIFF.

Despite its use in cleaning and disinfection, alcohol significantlyincreased the attachment of all three strains at 10 and 20°C(P $≤$ 0.0013 and P < 0.0001, respectively), compared to thatof the same strain grown in BHI at the same temperature (Fig.1). Attachment of strains EGD and 412 was also induced by alcoholat 30°C (P < 0.0001). There was no significant increasedue to the presence of alcohol at 37°C. The alcohol-inducedattachment in general reached the highest levels at 10°C.The induction by isopropanol was larger than the induction byethanol. This may reflect a greater growth-inhibitory effect(results not shown) rather than the effect of specific alcoholper se. Sublethal levels of ethanol, N-propanol, and isopropanolhave also been reported to enhance the attachment of Staphylococcusepidermidis (9).

View larger version (17K):
[in this window]
[in a new window]
FIG. 1. Attachment of cultures of L. monocytogenes 412 (A), EGD (B), and LO28 (C) , determined as eluted crystal violet in a microtiter plate assay following growth in BHI (white), BHI with 2.5% ethanol (light grey), or BHI with 2.5% isopropanol (dark grey) at 10, 20, 30, and 37°C for 8, 2, 1, and 1 days, respectively. The error bars show the standard deviations of the average values.

Possible regulators of alcohol-induced attachment.

Top

Possible regulators of alcohol...

In L. monocytogenes, it has previously been observed that exposureto 2% ethanol increased ${sigma}$ ^B activity (1) and that the two-componentsignal transduction system CesRK was involved in the abilityto tolerate ethanol (8). In order to investigate whether thesigB and cesRK gene products were involved in alcohol inductionof attachment, a ${Delta}$ sigB in-frame deletion mutant of strain EGD(3) and ${Delta}$ cesR and ${Delta}$ cesK in-frame deletion mutants of strain LO28(8) were tested in the microplate assay and compared to therespective wild-type strains. Two independent experiments weremade for strain EGD and the sigB mutant, and one experimentwas made for strain LO28 and the cesR and cesK mutants.

All three mutants retained the alcohol induction phenotype at10, 20, and 30°C; however, the degree of induction was insome cases affected. At 10°C (Fig. 2A), the levels of destainedcrystal violet were significantly lower for the sigB mutantexposed to isopropanol than for strain EGD in the same medium(P = 0.0019) and significantly higher for the cesK mutant exposedto ethanol than for strain LO28 (P = 0.0347). At 20°C (Fig.2B), the sigB mutant exposed to both ethanol and isopropanolsimilarly had lower attachment than strain EGD in the same medium(P = 0.0003 and P < 0.0001, respectively), while isopropanolconferred higher attachment of the cesR and cesK mutants thanof strain LO28 (P = 0.0005 and P < 0.0001, respectively).At 30°C, there was no significant difference between attachmentof deletion mutants and that of the corresponding wild-typestrains in the same medium (results not shown).

View larger version (24K):
[in this window]
[in a new window]
FIG. 2. Attachment of a sigB deletion mutant of strain EGD and cesR and cesK mutants of strain LO28, determined as eluted crystal violet in a microtiter plate assay following growth at 10°C for 8 days (A) or 20°C for 2 days (B) in BHI (white), BHI with 2.5% ethanol (light grey), or BHI with 2.5% isopropanol (dark grey). Values for the mutants that are marked ***, **, and * differ significantly (P < 0.001, P < 0.01, and P < 0.05, respectively) from the values of the corresponding wild-type strains in the same medium and at the same temperature (the compared values are shown by dashed lines). The error bars show the standard deviations of the average values.

In S. epidermidis, alcohol induction of attachment was reportedto be independent of ${sigma}$ ^B (10). Our observation that the inductionwas reduced in the sigB mutant and enhanced in the cesRK mutantsunder several of the tested conditions suggests that both ${sigma}$ ^Band CesRK influence the regulation of attachment of L. monocytogenesby alcohol by a fine-tuning or cross-regulatory function, e.g.,which would possibly differ from the regulatory cascade in S.epidermidis.

Strain variation in alcohol induction.

Top

Strain variation in alcohol...

To further characterize the strain variation in the alcoholinduction phenotype, we tested five persistent L. monocytogenesisolates from food production plants in the United Kingdom (strains64587/35, 64587/146, 42222/60, and 42222/373) (7) and Denmark(strain DMRICC 3653; T. Jacobsen, Danish Meat Research Institute,personal communication). Inocula were prepared by incubationsat 20°C for 2.5 days and diluted in either BHI or BHI with3% ethanol. The microtiter plates were incubated at 10°Cfor 7 days; three independent experiments were made.

Biofilm development at 10°C was significantly enhanced (P< 0.0001) by the presence of 3% ethanol in strains 42222/373,DMRICC 3653, and 42222/60 (Fig. 3). Strain 64587/146 had verylow attachment in both media. For strain 64587/35, there wasno significant effect of ethanol on attachment. In a parallelset of microtiter plates, adhered cells were stained with 0.1%(wt/vol) ruthenium red, a nonspecific polysaccharide stain.After 45 min of staining, the solution was removed manually,and the amount of remaining stain was evaluated by visual inspection.The alcohol-induced attachment in strains 42222/373, DMRICC3653, and 42222/60 was associated with increased ruthenium redstaining (results not shown), suggesting an enhanced exopolysaccharideproduction.

View larger version (13K):
[in this window]
[in a new window]
FIG. 3. Attachment of persistent L. monocytogenes strains following growth at 10°C for 7 days in BHI (white) or BHI with 3% ethanol (light grey), determined as eluted crystal violet in a microtiter plate assay. The error bars show the standard deviations of the average values.

Similar to our observations, observations of alcohol-inducedattachment were made in most, but not all, strains of S. epidermidis(9). In this organism, alcohol induction occurs through ${sigma}$ ^B-independenticaR repression, giving enhanced icaADBC transcription and exopolysaccharideproduction (10). The genetic basis of exopolysaccharide productionin L. monocytogenes is currently completely unknown; no genesin the sequenced genome (5) are annotated to have such a function.Further work is in progress to characterize the genetic basisof exopolysaccharide synthesis and alcohol-induced surface attachmentin L. monocytogenes.

ACKNOWLEDGMENTS

We thank Hanne Mordhorst and Lene Gertman for technical assistance,Andreas Reisner for advice on use of the Biomek robot, TinaBeck Hansen for advice on the statistical testing, and JohnHolah and Tomas Jacobsen for supplying strains.

This work was supported by the Danish Research Agency (grant5011-00-0027). C.L. was supported through the Leonardo da Vincitrainee program.

FOOTNOTES

* Corresponding author. Present address: Danisco A/S, Edwin Rahrs Vej 38, DK-8220 Braband, Aarhus, Denmark. Phone: 45 8943 5000. Fax: 45 8625 1077. E-mail: Anne.Gravesen{at}danisco.com.

REFERENCES

Top