lable at ScienceDirect

Food Microbiology 39 (2014) 81e88

Contents lists avai

Food Microbiology

journal homepage: www.elsevier .com/locate/ fm

Control of Listeria monocytogenes contamination in an Iberian porkprocessing plant and selection of benzalkonium chloride-resistantstrains

Sagrario Ortiz a, Victoria López b, Joaquín V. Martínez-Suárez a,*

aDepartamento de Tecnología de Alimentos, Instituto Nacional de Investigación y Tecnología Agraria y Alimentaria (INIA), Madrid, SpainbDepartamento de Bioinformática y Salud Pública, Instituto de Salud Carlos III, Majadahonda, Madrid, Spain

a r t i c l e i n f o

Article history:Received 22 April 2013Received in revised form24 September 2013Accepted 12 November 2013Available online 21 November 2013

Keywords:Listeria monocytogenesPulsed-field gel electrophoresis typesIberian porkProlonged survivalLow virulenceDisinfectant resistance

Abbreviations: BAC, benzalkonium chloride; EtBminimal inhibitory concentration; PFGE, pulsed-fieldpremature stop codon; QAC, quaternary ammonium c* Corresponding author. Tel.: þ34 91 3474027; fax:

E-mail address: joaquin@inia.es (J.V. Martínez-Suá

0740-0020/$ e see front matter � 2013 Elsevier Ltd.http://dx.doi.org/10.1016/j.fm.2013.11.007

a b s t r a c t

The aims of this study were to characterize the different strains of Listeria monocytogenes collected at anIberian pork processing plant and to investigate whether their specific characteristics were associatedwith prolonged survival in the plant. Using pulsed-field gel electrophoresis (PFGE), 29 PFGE types werepreviously identified during a three-year period. Eight of these PFGE types persisted in the plant duringthat period. In the present study, a subset of 29 PFGE type strains, which represented the 29 differentPFGE types, was further characterized by assessing the potential virulence, and using motility, surfaceattachment, and antimicrobial susceptibility tests. After changing the disinfection procedures in theplant, the isolation rate of L. monocytogenes decreased, and only four of the 29 PFGE types, includingthree of the eight persistent PFGE types, were found the following year. These four “surviving” PFGEtypes included three from PCR serogroup IIa that were characterized by their low virulence mutationsand low-level resistance to benzalkonium chloride (BAC). Furthermore, these PFGE types comprised theonly BAC-resistant isolates found in the study, and they appear to have been selected due to the controlof Listeria contamination. The resistance to increased sublethal concentrations of disinfectants may leadto prolonged survival of L. monocytogenes in food plants.

� 2013 Elsevier Ltd. All rights reserved.

1. Introduction

Listeria monocytogenes is a foodborne human pathogenresponsible for invasive infections presenting overall a high mor-tality. The species is subdivided into at least four evolutionary lin-eages (I, II, III, and IV) with varying virulence (Orsi et al., 2011). Theattenuated virulence displayed by a large number of lineage IIisolates may be due to mutations that give rise to premature stopcodons (PMSCs) in the inlA gene (Jacquet et al., 2004; Ragon et al.,2008; Kovacevic et al., 2013) and mutations in the prfA gene(Lindbäck et al., 2011; Roche et al., 2012). The variability in thevirulence of L. monocytogenes strains has important implications forthe detection of Listeria strains and risk analysis (van Stelten et al.,2010; López et al., 2013).

r, ethidium bromide; MIC,gel electrophoresis; PMSC,

ompound.þ34 91 3572293.rez).

All rights reserved.

A typical feature of Listeria contamination of food processingenvironments is prolonged plant contamination (Kornacki andGurtler, 2007). A limited number of genetically similar strains ofL. monocytogenes can be specifically found in a single food pro-cessing plant, where they can persist over several months or years,and these persistent strains are likely to contaminate food products(Keto-Timonen et al., 2007; López et al., 2008; Møretrø andLangsrud, 2004; Wulff et al., 2006). Several authors claim that theisolates of persistent subtypes possess specific characteristics thatfacilitate persistence (Fox et al., 2011; Verghese et al., 2011),whereas other researchers have found no evidence of such specificcharacteristics (Ferreira et al., 2011; Holch et al., 2013).

Given that biofilms can protect bacteria against disinfection(Møretrø and Langsrud, 2004), the initial surface attachment andsubsequent biofilm-forming ability are commonly thought to beamong the main factors affecting bacterial survival and persistence.However, many different factors, including strain-specific proper-ties (Nilsson et al., 2011), can affect bacterial adhesion and biofilmformation (Di Bonaventura et al., 2008; Renier et al., 2011). Motility,for example, is a variable property within L. monocytogenes thatinfluences surface attachment and detachment and is required for

Table 1Characteristics of the 29 PFGE types of Listeria monocytogenes isolated in years 1e3.

PFGE typea Serogroupb No. of isolates per year Persistencec

Year 1 Year 2 Year 3

S1 IIa 107 82 55 þS2 IIa 32 51 6 þS3 IIa 2 2 e

S4-1 IIb 32 20 10 þS4-2 IIb 14 14 e þS4-3 IIb 4 10 2 þS5 IIc 5 19 3 þS6 IIa 1 e e

S7-1 IVb 4 2 e

S7-2 IVb 1 4 e

S7-3 IVb e 1 e

S7-4 IVb e 1 e

S8 IIc 1 1 e

S9 IVb e 2 e

S10-1 IIa e 3 20 þS10-2 IIa e 1 e

S11 IIa e 1 e

S12 (S12-1) IIc e 7 4S13 IIb e 2 e

S14 IIa e 2 e

S15 IIa e 1 e

S16 IIa e 1 e

S17-1 IIb e 1 3 þS17-2 IIb e 1 e

S18 IIb e 1 e

S19 IIa e e 2S20-1 (S20) IIb e e 1S20-2 (S4-4) IIb 1 e e

S21 (S12-2) IIc e 1 e

a The pulsed-field gel electrophoresis (PFGE) types that are shown in bold weredetermined in this study using the BioNumerics software; the other data representthe previously identified PFGE types (Ortiz et al., 2010) that have been changed (inparentheses).

b Multiplex PCR serogroups (Doumith et al., 2004; Leclercq et al., 2011).c þ, PFGE type isolated three times or more in the environment or equipment

over a minimum of three months (Keto-Timonen et al., 2007; Ortiz et al., 2010).

S. Ortiz et al. / Food Microbiology 39 (2014) 81e8882

biofilm formation (Renier et al., 2011). Differences in motility andsurface attachment might influence the outcome of surface clean-ing and disinfection; thus such differences may affect the ability ofdifferent L. monocytogenes strains to persist in the environment.

Resistance to cadmium (Cd) has been reported to occur morefrequently in persistent compared with non-persistent strains ofL. monocytogenes (Harvey and Gilmour, 2001). Logically, resistanceto disinfectants may be one of the most important factors thatcontributes to the extended survival and persistence ofL. monocytogenes (Fox et al., 2011; Lundén et al., 2003), althoughseveral authors have not been able to demonstrate this association(Earnshaw and Lawrence, 1998; Heir et al., 2004; Holah et al., 2002;Kastbjerg and Gram, 2009). Differences in resistance betweenpersistent and non-persistent L. monocytogenes strains, if they exist,can be difficult to detect (Carpentier and Cerf, 2011; Lundén et al.,2003). Nonetheless, disinfectants can affect the persistence ofstrains with an enhanced response to antimicrobial stress. Wepreviously observed that biofilms of L. monocytogenes showed abetter ability to recover after chitosan damage when they wereformed by persistent strains compared to non-persistent ones(Orgaz et al., 2013).

The disinfectants that are commonly used in the food industriesinclude surface-active agents (surfactants), alcohols, aldehydes,peracetic acid, hypochlorite, and organic chlorine-releasing com-pounds (McDonnell and Russell, 1999; Holah et al., 2002). Surface-active agents include amphoteric, cationic (quaternary ammoniumcompounds [QACs] such as benzalkonium chloride [BAC]), andbiguanide/diguanide compounds.

Developed microbial resistance is a drawback of QACscompared with other disinfectants (Hegstad et al., 2010). The se-lection pressure due to the use/misuse of QACs may play a majorrole in BAC-resistance because bacteria isolated after disinfectionhave been shown to be more resistant than laboratory strains(Hegstad et al., 2010; Sidhu et al., 2002). BAC-resistance amongL. monocytogenes strains isolated from food sources can vary from10% (Aase et al., 2000) to more than 40% (Mereghetti et al., 2000;Soumet et al., 2005) and may be especially high among strainsfrom serotype 1/2a (or 3a) (60%) (Mullapudi et al., 2008). Resis-tance to QACs has been shown to be higher among selectedpersistent compared with non-persistent strains, and this differ-ence is most likely due to changes in the bacterial cell surfaces(Fox et al., 2011; Mereghetti et al., 2000). A second mechanism ofresistance to QACs is the overexpression or acquisition of effluxpumps (Aase et al., 2000; Soumet et al., 2005). Certain effluxpumps confer cross-resistance to various compounds because theyexpel QACs and other antimicrobial agents from cells (Gilbert andMcBain, 2003; Hegstad et al., 2010; Rakic-Martinez et al., 2011).Methods to identify isolates with a multidrug-resistant (MDR)efflux phenotype include minimal inhibitory concentration (MIC)determinations in the presence and absence of efflux pump in-hibitors, such as reserpine, and/or the use of alternative well-known substrates for MDR efflux pumps, such as ethidium bro-mide (EtBr) (Romanova et al., 2006).

In a previous three-year study, sources of L. monocytogenescontamination in an Iberian pig abattoir and processing plant wereanalyzed (Ortiz et al., 2010). The pulsed-field gel electrophoresis(PFGE) typing of 541 L. monocytogenes isolates resulted in theidentification of 29 different PFGE types. After the disinfectionprocedures were changed in the Iberian pig plant, the occurrence ofL. monocytogenes declined, and the isolates belonging to only alimited number of the original 29 PFGE types were recovered again.The aim of the present study was to investigate whether specificcharacteristics were associated with the sanitation survival andextended persistence of L. monocytogenes. In particular, isolatesbelonging to the 29 PFGE types were further characterized by

assessing the virulence potential, and using motility, surfaceattachment, and antimicrobial susceptibility tests.

2. Materials and methods

2.1. Food-related strains identified during years 1e3

All of the L. monocytogenes isolates recovered over a period ofthree years were subtyped by PCR-based serogrouping (Doumithet al., 2004; Leclercq et al., 2011) and PFGE (Halpin et al., 2010),as previously described (Ortiz et al., 2010). A total of 29 unique PFGEtypes were identified in the characterized isolates. Eight PFGE typeswere considered to be persistent in our previous report (Table 1;Ortiz et al., 2010) as they were isolated three or more times in theenvironment or equipment over aminimumof threemonths (Keto-Timonen et al., 2007).

Initially, the visual PFGE analysis yielded 29 different PFGE types(Ortiz et al., 2010). Based on the results generated in the presentstudy, four PFGE types were reassigned to new PFGE types (Table 1).This new analysis of the PFGE results was performed using theBioNumerics software (Version 4.5, Applied Maths, Kortrijk,Belgium). The similarity clustering was performed according to theBioNumerics PulseNet manual (http://www.pulsenetinternational.org/protocols/Pages/bionumerics.aspx).

The first isolate with a unique PFGE type was considered thePFGE type strain of each PFGE type. In the present study, the 29PFGE type strains were compared through potential virulence,

S. Ortiz et al. / Food Microbiology 39 (2014) 81e88 83

motility, and surface attachment analyses and disinfectant sus-ceptibility tests (as described in 2.4e2.6).

2.2. Food-related strains identified during year 4

After the three-year study that was previously reported (Ortizet al., 2010), 24 additional isolates were recovered from the sameprocessing plant. In the present study, these additional isolateswere also subtyped by PCR-based serogrouping and PFGE (asdescribed in 2.1), and subjected to disinfectant susceptibilitytesting (as described in 2.6). These 24 isolates were detected afterthe change in the disinfection procedures andwere found to belongto four of the PFGE types which were detected previously, and aretherefore, described in this report as “surviving” PFGE types (see3.1).

2.3. Standard culture methods

The strains were stored in tryptic soy yeast extract broth (TSYEB)containing 15% glycerol at �20 �C prior to use. Before each test, thebacteria were subcultured on two consecutive days on tryptic soyyeast extract agar (TSYEA) and subsequently grown for 18 h at 37 �Cin TSYEB. Both TSYEB and TSYEA were obtained from Biolife(Milano, Italy). Different media were used for the specific testsdescribed below.

2.4. Potential virulence

The reference virulent strain L. monocytogenes Scott A (ATCC49594) (Briers et al., 2011; Jensen et al., 2008), was included as acontrol in all of the phenotypic tests that were conducted. Theactivity of the virulence-associated proteins listeriolysin O (LLO),phosphatidylcholineephospholipase C (PCePLC), and phosphati-dylinositolephospholipase C (PIePLC) was determined qualita-tively based on the width of the halos formed by strains inoculatedinto the relevant differential solid media. These were representedby sheep blood agar (bioMérieux, Inc., Marcy l’Etoile, France) forLLO detection, Brilliance Listeria agar (Oxoid, Ltd., Basingstoke, UK)for PCePLC and CHROMagar Listeria (CHROMagar, Paris, France) forPIePLC. All inoculated media were incubated for 48 h at 37 �C. Theresults of the qualitative assays were recorded as negative (withouthalo), positive (width of the halo is similar to that of the Scott Astrain), or hyper-positive (width of the halo larger than that ob-tainedwith the Scott A strain). LLOwas also assessed quantitatively,by growing L. monocytogenes strains in Brain Heart Infusion broth(BHI; bioMérieux) at 37 �C to an optical density at 600 nm (OD600)of 1.0, and determining the hemolytic activity in the supernatant bymeasuring the amount of released hemoglobin, as previouslydescribed (Mertins et al., 2007).

The virulence genes inlA and prfAwere also sequenced to screenfor the presence of mutations associated with attenuated virulence(Roche et al., 2012). DNA sequencing of the complete inlA gene or afragment of prfA were conducted and compared to the respectivesequences of L. monocytogenes EGD-e (GenBank accession numberNC_003210.1, http://www.ncbi.nlm.nih.gov/genbank/). Primer se-quences used for PCR amplification and/or DNA sequencing as wellas PCR conditions were those used in Ragon et al. (2008) for inlA,and Chen et al. (2005) for prfA. PCR products were purified, quan-tified, and sequenced, and nucleotide sequences were assembledand analyzed with Bioedit 7.2.1. (Ibis Biosciences). PMSCs in inlAsequences were found using ORF Finder (Open Reading FrameFinder, http://www.ncbi.nlm.nih.gov/projects/gorf/). DNA se-quences were submitted to the GenBank database under accessionnumbers HQ111527 through HQ111555 (inlA), and JQ345449through JQ345477 (prfA).

2.5. Motility and surface attachment

Swimming motility was tested on semi-solid BHI agar (0.4%)plates that were inoculated with 2 mL of exponential bacterialcultures (OD600 w 0.7) grown in BHI at 37 �C. The plates weresealed and incubated for 48 h at 30 �C in complete darkness.Motility was quantified as the diameter of the swimming colony (DiBonaventura et al., 2008). Surface attachment (i.e., initial biofilmformation) of L. monocytogenes was analyzed by using microplateassays with crystal violet staining (Di Bonaventura et al., 2008),after incubation for 48 h at 37 �C in TSYEB supplemented with 1%glucose and 2% sodium chloride (sTSYEB) (Pan et al., 2010).

2.6. Susceptibility testing to antimicrobial agents

The susceptibility to cadmium chloride (CdCl2, abbreviated asCd) and BAC (both from SigmaeAldrich, Saint Louis, MO, USA) wasdetermined through an agar dilution assay on Mueller Hinton agar(MHA, Biolife) plates containing two-fold serial dilutions of theactive compounds (CLSI, 2008). The plates were incubated for 24 hat 37 �C. To determine the efflux pump activity, the inhibitorreserpine (at a final concentration of 10 mg L�1; SigmaeAldrich)was added to each MHA plate containing BAC. The changes to theMICs of BAC were observed both in the presence and in the absenceof reserpine (Romanova et al., 2006). The susceptibility to EtBr(SigmaeAldrich) was determined using the broth microdilutionmethod (Romanova et al., 2006). The growth at 37 �C was recordedvisually after 24 h of exposure.

For the three antimicrobial agents (Cd, BAC, and EtBr), the termresistance is used in the current study to describe the situation inwhich the MIC of a compound is clearly higher than that of thereference strains from the culture collections, i.e., L. monocytogenesScottA, which was tested in parallel (Gilbert and McBain, 2003).For BAC, resistance is also defined as stable and low-level(Kastbjerg and Gram, 2012), which means that resistance is nota reversible characteristic and that resistant strains only display atwo- to eight-fold increase in the MIC compared to the rest of thestrains (see 3.4).

2.7. Statistical analysis

The swimming motility, surface attachment, and susceptibilityagainst Cd, BAC, and EtBr were determined in triplicate, and theseexperiments were repeated on at least two different occasions.The motility and surface attachment results were expressed aspercentage of the values obtained for L. monocytogenes Scott A.The motility and attachment assay data satisfied the assumption ofnormality. Pearson’s correlation coefficient was used to evaluatethe interdependency between motility and surface attachment. Tocompare the differences in the normalized motility or surfaceattachment among the 29 PFGE type strains, one-way analysis ofvariance (ANOVA) followed by Tukey’s multiple comparison testwas performed. In addition, ANOVA was used to determinewhether there were statistically significant associations betweenthe strain motility or surface attachment and (i) the PCRserogroup, (ii) lineage, (iii) presence of a mutated prfA gene, (iv)presence of a PMSC in the inlA gene, (v) Cd resistance, or (vi) BAC-resistance. The association between all categorical variables andthe “surviving” PFGE types was tested using Fisher’s exact test.Differences with values of p < 0.05 were considered to be statis-tically significant. All of the calculations were performed usingMicrosoft Excel 2007 (Microsoft Corporation, Redmond, WA, USA)and the Minitab 16 statistical software (Minitab Inc., State College,PA, USA).

Table 2Characteristics of the 29 PFGE type strains of Listeria monocytogenes used in thisstudy.

PFGEtypestrain

LLOa inlAPMSCb

Motilityc Attachmentc Cd-Rd BAC-Re

S1 � 5 73.33 � 3.51 A,B 24.00 � 8.66 C þ þS2 þþ 6 72.67 � 9.50 A,B 53.67 � 9.24 A,B,C þ �S3 þþ 6 69.67 � 4.93 A,B 50.00 � 10.82 A,B,C þ �S4-1 þ 0 72.00 � 1.73 A,B 49.33 � 9.07 A,B,C � �S4-2 þ 0 74.00 � 1.73 A,B 68.00 � 9.85 A,B,C þ �S4-3 þ 0 72.67 � 0.58 A,B 53.67 � 11.02 A,B,C � e

S5 þ 12 77.67 � 4.51 A,B 43.67 � 3.06 A,B,C þ �S6 � 5 83.00 � 1.73 A 25.67 � 2.89 C þ �S7-1 þ 0 78.67 � 11.50 A,B 61.30 � 23.60 A,B,C � �S7-2 þ 0 76.00 � 3.61 A,B 64.30 � 20.50 A,B,C � �S7-3 þ 0 73.67 � 7.57 A,B 54.70 � 30.20 A,B,C � �S7-4 þ 0 70.67 � 9.07 A,B 51.00 � 33.20 A,B,C � �S8 þ 11 64.67 � 6.51 A,B 84.00 � 9.17 A þ �S9 þ 0 82.67 � 15.37 A 67.00 � 28.70 A,B,C � �S10-1 þ 6 81.00 � 12.00 A 79.70 � 28.00 A,B þ þS10-2 þ 6 50.33 � 9.87 B 41.30 � 21.20 A,B,C þ �S11 þ 0 65.67 � 2.08 A,B 57.30 � 21.50 A,B,C � �S12 þ 12 66.67 � 12.50 A,B 31.00 � 3.46 B,C � �S13 þ 0 81.33 � 7.77 A 61.00 � 14.11 A,B,C þ �S14 þ 0 76.67 � 2.08 A,B 39.67 � 4.16 A,B,C � �S15 þ 0 78.00 � 9.85 A,B 45.33 � 3.79 A,B,C � �S16 þ 5 74.00 � 2.65 A,B 49.67 � 8.33 A,B,C þ �S17-1 þ 0 79.67 � 10.97 A,B 40.33 � 7.02 A,B,C þ �S17-2 þ 0 62.67 � 15.95 A,B 27.67 � 11.93 C þ �S18 þ 0 81.33 � 10.12 A 54.67 � 3.79 A,B,C � �S19 þ 0 78.67 � 13.20 A,B 39.67 � 4.16 A,B,C þ �S20-1 þ 0 83.30 � 21.00 A 67.33 � 8.50 A,B,C � �S20-2 þ 0 71.00 � 3.46 A,B 53.33 � 10.12 A,B,C � �S21 þ 12 87.67 � 11.02 A 39.67 � 9.45 A,B,C � �a Qualitative detection of listeriolysin O (LLO) activity on blood agar plates (�,

non-hemolytic; þ, hemolytic; þþ, hyperhemolytic).b Identification of the types of Listeria monocytogenes inlA premature stop codon

(PMSC) mutations according to van Stelten et al. (2010). 0, full-length inlA sequence.c The swimming motility and surface attachment results were expressed as

percentage of the values obtained for L. monocytogenes Scott A control strain. Thevalues represent the average results of three independent replicates plus/minus thestandard deviations. The values in the same column with different upper case letter(A, B, C) are significantly different from each other (p <0.05).

d þ, MIC of cadmium chloride�40 mg L�1.e þ, MIC of benzalkonium chloride�10 mg L�1.

S. Ortiz et al. / Food Microbiology 39 (2014) 81e8884

3. Results and discussion

3.1. PFGE types

Based on the results obtained using the BioNumerics software,the final number of 29 PFGE types that had been obtained duringthe previous three-year study (Ortiz et al., 2010) was not modified.However, the greater precision of the new analysis resulted in thegeneration of four new denominations of four PFGE types due tochanges of AscI or ApaI profiles for two PFGE types. The changeswere limited to PCR serogroups IIb (S4-4 changed to S20-2) or IIc(S12-2 changed to S21) (Table 1) and clustering within the otherserogroups remained the same.

Reassignment of PFGE types is not uncommon. Outbreak in-vestigations by PFGE typing have revealed that the comparison ofvery similar patterns relies on a degree of individual subjectivitywhich change as databases get larger and different people areinvolved. While very small differences in patterns may be reflectiveof differences in genotype, they may also in some cases reflectmethodological differences that can arise, particularly when iso-lates are analysed over many years (PulseNet International, http://www.pulsenetinternational.org/).

3.2. Potential virulence of PFGE type strains

Three different hemolytic phenotypes were found among the 29PFGE type strains (Table 2). There were two non-hemolytic strains(PFGE type strains S1 and S6) and two hyperhemolytic strains(PFGE type strains S2 and S3), and 25 strains showed hemolyticactivity on blood agar plates similar to that obtained with theL. monocytogenes control strain Scott A. The LLO titration confirmedthe qualitative assay results: the S1 and S6 strains were bothslightly hemolytic, even though their hemolytic activities weresubstantially inferior (<10%) to that of the Scott A strain (Fig. 1A).The PCePLC and PIePLC activities of the S1 and S6 strains were alsolower than the rest of the L. monocytogenes strains albeit wereclearly detected at 48 h (Fig.1B); in fact, thesewere isolated on a PI-PLC differential medium [Agar Listeria according to Ottaviani andAgosti (ALOA)] (Ortiz et al., 2010).

To further characterize the 29 PFGE type strains, two virulencegenes were partially (prfA) or fully (inlA) sequenced. The S1 and S6PFGE type strains contained a previously described insertion ofseven nucleotides (CAGGAGT) at codon 171 of prfA, which resultedin a PMSC at position 178 and a deduced truncated PrfA protein(Roche et al., 2012). Both strains belonged to PCR serogroup IIa andwere closely related according to PFGE cluster analysis (Ortiz et al.,2010). A truncated PrfA protein would result in a potentially avir-ulent phenotype due to the lack of activation of the mainL. monocytogenes virulence genes (Roche et al., 2012).

Four different PMSC mutations in the inlA sequence (van Steltenet al., 2010) were found in 38% (n ¼ 11) of analyzed strains. PMSCswere found in all four PFGE type strains of PCR serogroup IIc (S5, S8,S12, S21) and in seven of those belonging to PCR serogroup IIa (S1,S2, S3, S6, S10-1, S10-2, S16) (Table 2).

Although mutations leading to PMSCs in virulence genes andattenuated virulence are common in serotype 1/2a, these are rare inserotype 4b (Roche et al., 2012; Kovacevic et al., 2013). In the pre-sent report, the identification of inlA PMSC mutations in sevenstrains (64%) of PCR serogroup IIa was consistent with the well-known fact that there is a subgroup of serotype 1/2a strains,mainly of food origin, whose virulence may be attenuated (Jacquetet al., 2004; Roche et al., 2012; Kovacevic et al., 2013; López et al.,2013). Serotypes 1/2a and 1/2c have been commonly associatedwith pork products (Thévenot et al., 2006), and the subtyping ofthis group of 541 isolates of L. monocytogenes (Table 1) has shown

that PCR serogroup IIa represents 68% of the isolates (Ortiz et al.,2010). The marked genetic diversity of serotype 1/2a strains andtheir frequent presence in food could facilitate the selection of lowvirulence mutations that may be advantageous for survival in foodenvironments (Orsi et al., 2011). Interestingly, four PFGE typestrains within the 11 PFGE type strains associated with PMSCs(Table 2), namely strain S1 (PMSC type 5), strains S2 and S10-1(PMSC type 6), and strain S5 (PMSC type 12), represent PFGEtypes that show evidence of persistence in the production chain ofthe studied pork plant (Table 1; Ortiz et al., 2010). However, theselow virulence mutants belonging to persistent PFGE types did notperform better in abiotic surface adherence or biofilm formationassays that those belonging to non-persistent PFGE types (Orgazet al., 2013).

3.3. Motility and surface attachment of PFGE type strains

The group of 29 food-related strains analyzed herein exhibited arange of motility between 50.33% and 87.67% of that obtained forthe L. monocytogenes strain Scott A (Table 2). Multiple comparisontests showed significant differences in motility among several PFGEtype strains (Table 2). One strain (S10-2) showed reduced motility,i.e., 50.33% of the Scott A motility, and seven strains (S6, S9, S10-1,S13, S18, S20-1, and S21) demonstrated significantly greater

Fig. 1. A. Titration of hemolytic activity in the supernatants of L. monocytogenes strains with different potential virulence phenotypes. Strains S1 and S6 (non-hemolytic), S2 and S3(hyperhemolytic), and S5 (representative of the 25 strains displaying standard hemolytic activity on blood agar plates) were compared to the L. monocytogenes control strain Scott A(100% hemolysis value). The hemolytic activity was determined in three independently performed experiments; error bars indicate the measured maximum and minimum values.B. Different halos on CHROMagar Listeria plates after 48-h incubation of three “S” strains, which represent three different phenotypes, and the Scott A control strain.

S. Ortiz et al. / Food Microbiology 39 (2014) 81e88 85

motility than S10-2 (Table 2). These differences, however, were notrelated to either persistence (Table 1) and/or “survival” of the cor-responding PFGE type after the three-year study (see 3.5 andTable 3) (p > 0.05).

All L. monocytogenes PFGE type strains adhered to polystyreneafter incubation for 48 h in sTSYEB, although the surface attach-ment values ranged from 24% to 84% relative to that obtained forScott A (Table 2). There were significant differences in surfaceattachment among several PFGE type strains. For example, S1, S6,and S17-2 showed a statistically significant lower level of surfaceattachment compared with S10-1 and S8 (Table 2).

The motility and surface attachment were not positively corre-lated (Pearson r¼ 0.12; p> 0.05), and neither of these two variableswere significantly associated with the PCR serogroup, lineage, Cdresistance, or BAC-resistance (p> 0.05). Motility was not correlatedwith low potential virulence mutations, but there was a statisticallysignificant association between surface attachment and the

Table 3Characteristics of the four PFGE types of Listeria monocytogenes isolated in year 4.

PFGE type No. of isolates MIC (mg L�1)

BACa BAC þ reserpineb EtBrc

S1 13 10 5 160S4-1 2 2.5 1.25 20S6d 1 10 5 160S10-1 8 20 20 40

a BAC, benzalkonium chloride.b Reserpine is an inhibitor of efflux pumps.c EtBr, ethidium bromide.d The isolate detected in year 1 (Table 1) was BAC-susceptible (Table 2), and

displayed the same MICs to BAC, BAC þ reserpine, and EtBr as S4-1.

presence of certain low potential virulence mutations. The two prfAmutant strains (PFGE type strains S1 and S6) showed lower levels ofattachment, whereas the strain with the inlA PMSC 11 (PFGE typestrain S8) demonstrated a significantly higher level of attachmentcompared with those strains with inlA PMSC 5 (PFGE type strainsS1, S6, and S16), or inlA PMSC 12 (PFGE type strains S5, S12, and S21)(Table 2).

Regarding the four “surviving” PFGE types (see 3.5 and Table 3),major differences were observed among their PFGE type strains. Forexample, two of these PFGE type strains displayed the lowest (S1,24%) and the second highest (S10-1, 79.7%) overall surface attach-ment values (Table 2). The low surface attachment values of theprfA mutants S1 and S6 were consistent with the results obtainedby Lemon et al. (2010), who found that PrfA positively regulatesbiofilm formation. However, our finding does not coincide withthose researchers’ conclusions that “the requirement of PrfA foroptimal biofilm formation may provide selective pressure tomaintain this virulence regulator when L. monocytogenes is outsidehost cells in the environment” (Lemon et al., 2010), because the prfAmutant S1 was the most abundant PFGE type in this processingplant. S1 represented 45% of the isolates from the three-year study(Table 1) and represented 54% of the isolates collected during thefourth year (Table 3).

Mutations in the prfA gene can be common in specific food-handling environments (Lindbäck et al., 2011; López et al.,2013; Roche et al., 2012). The PFGE profiles obtained with twodifferent restriction enzymes (Halpin et al., 2010) of the prfAmutant S1 were apparently very similar to those of the isolateNVH295 of L. monocytogenes studied in Norway (Lindbäck et al.,2011), which also exhibited the same mutation in the prfA gene.In the study conducted by Lindbäck et al. (2011), approximately50% of the samples collected from a fish processing plant were

S. Ortiz et al. / Food Microbiology 39 (2014) 81e8886

positive for the non-hemolytic strain NVH295, which is coinci-dent with the isolation results of the PFGE type S1. Thus, similarlow virulence subtypes of L. monocytogenes can predominate andpersist in very distantly located food processing plants (Jensenet al., 2008).

3.4. Susceptibility testing to antimicrobial agents of PFGE typestrains

Fifteen PFGE type strains were classified as sensitive to Cd withMICs� 5 mg L�1, whereas 14 PFGE type strains were resistant to Cd(MIC � 40 mg L�1). Resistance to Cd was less common amonglineage I strains compared with lineage II strains (Table 2).

Sensitivity to BAC was observed in 27 PFGE type strains(MIC � 2.5 mg L�1), and only two PFGE type strains were resistantto BAC, which included the persistent (Table 1) and “surviving”PFGE types S1 (MIC ¼ 10 mg L�1) and S10-1 (MIC ¼ 20 mg L�1) (see3.5 and Table 3).

L. monocytogenes isolates from lineage II contain more plasmidsthan those isolates of lineage I. These plasmids often confer resis-tance to heavy metals and likely other toxic compounds found inthe environment, such as disinfectants (Elhanafi et al., 2010; Orsiet al., 2011). Among the 29 analyzed L. monocytogenes strains,most of the Cd-resistant PFGE type strains and the two BAC-resistant PFGE type strains are from lineage II (PCR serogroups IIaand IIc; Tables 1 and 2).

3.5. “Surviving” PFGE types

In the considered pork processing plant, which was initiallystudied over a period of three years (Ortiz et al., 2010), thenumber of different PFGE types increased from 12 to 25 duringthe first two years (Table 1) due to a number of short-lived orsporadic PFGE types. This number decreased to 10 after thesecond year due to the implementation of different strategies tocontrol L. monocytogenes, which reduced the number of isolatesfrom most PFGE types (Ortiz et al., 2010). This decrease in theisolation of L. monocytogenes was accompanied by the recovery ofonly four of the 29 previous PFGE types (S1, S4-1, S6, and S10-1)during the fourth year of samplings in the same plant (Table 3).The four PFGE types were detected in the previous study, whichsuggests that these are long-term “surviving” PFGE types. Threeof these PFGE types (S1, S4-1, and S10-1) were considered to bepersistent in our previous report (Table 1; Ortiz et al., 2010), andtwo of them (S1 and S4-1) were arbitrarily considered predom-inant (present in over 10% of all of the isolates) (Ortiz et al.,2010). The three persistent and “surviving” PFGE types wererecovered from the processing plant over three (S10-1) or fouryears (S1 and S4-1) (Tables 1and 3). However, prolonged “sur-vival” was also observed for another PFGE type (S6) that wasdetected only once before the fourth year (Tables 1 and 3).

Of the PFGE type strains of PFGE types S1, S6, S4-1, and S10-1,three (PFGE type strains S1, S6, and S10-1) were characterized bytheir low potential virulence mutations (Table 2). The PFGE types ofthe two prfA mutants (PFGE type strains S1 and S6) constituted asignificant proportion of the “surviving” PFGE types (Table 3, 50%;p ¼ 0.01, Fisher’s exact test). Fisher’s exact test also showed thatthere was no significant association between the “survival” of aPFGE type and most of the characteristics of the correspondingPFGE type strains [PCR serogroup, lineage, Cd resistance, or pres-ence of most of the PMSC types in the inlA gene (p > 0.05)]; how-ever a significant association was found with inlA PMSC type 5(p ¼ 0.04). The PFGE type strains S1 and S6 showed statisticallysignificant lower levels of surface attachment (24% and 25.67%,respectively) compared to PFGE type strain S10-1 (p < 0.001;

Table 2). The attachment value of the PFGE type strain S4-1 was notsignificantly different (49.33%) than that obtained for the PFGE typestrains S1 and S6 (Table 2).

The test of the susceptibility to BAC of the 24 additional isolatesdetected in year 4 (Table 3) revealed that isolates of the PFGE typesS1, S6, and S10-1 were resistant (MIC � 10 mg L�1). Thus,depending on the isolation data, PFGE type S6 contained bothresistant and susceptible isolates (Tables 1 and 3). For the BAC-resistant isolates of PFGE types S1 and S6, there was a correspon-dence between the resistance to BAC and the resistance to EtBr(MIC of EtBr ¼ 160 mg L�1), but this was not the case for the BAC-resistant isolates of PFGE type S10-1 (MIC of EtBr ¼ 40 mg L�1,which was similar to the EtBr MICs for BAC-susceptible PFGE types,such as S4-1) (Table 3). The addition of reserpine, which is an effluxpump inhibitor, to the BAC-resistant isolates of PFGE types S1 andS6 resulted in a 50% decrease in the MIC for BAC, whereas no suchdecrease was observed for the BAC-resistant isolates of PFGE typeS10-1 (Table 3). Therefore, efflux pumps may be at least partiallyresponsible for the BAC-resistance of the isolates of PFGE types S1and S6. The resistance phenotype of S1 and S6 resembles thephenotype of naturally sensitive strains after in vitro adaptation toBAC (Romanova et al., 2006) and some naturally resistant strains(Soumet et al., 2005). A different mechanism is responsible for theBAC-resistance of PFGE type S10-1, i.e., cell wall modifications (Foxet al., 2011; Mereghetti et al., 2000).

In general, the PFGE types S1 and S10-1, together with the fourthyear isolate of PFGE type S6, were the only resistant strains found(Tables 2 and 3). Fisher’s exact tests showed that BAC-resistancewas one of only three variables (the other two being prfA muta-tion and inlA PMSC type 5) that were significantly associated withprolonged survival (75%; p¼ 0.001). The three BAC-resistant strains(PFGE types S1 and S10-1, and the fourth year isolate of PFGE typeS6) have the same PCR serogroup (IIa), whereas the other survivingPFGE type (S4-1) belonged to PCR serogroup IIb and was BAC-susceptible (Tables 1 and 3).

Nevertheless, values of the MIC of BAC-resistant strains(w10 mg L�1) are substantially lower than the concentrations atwhich QACs are used in food production facilities (500e1000 mg L�1 or higher) (Aase et al., 2000; Hegstad et al., 2010;Romanova et al., 2006; Soumet et al., 2005). The in vitro exposure toQACs can cause only stable low-level resistance, which is typicallycharacterized by a two- to four-fold increase in the MIC (Kastbjergand Gram, 2012). This would suggest that the QACs that arecommonly used in food environments should be effective againstBAC-resistant isolates of L. monocytogenes if used in accordancewith the manufacturers’ instructions.

The PFGE type strains of persistent PFGE types S1 and S10-1showed higher MIC values for BAC than the other PFGE typestrains isolated in years 1e2 (Table 2), and both PFGE types weredetected during the fourth year. Differences in the MIC values fordisinfectants between L. monocytogenes strains have been specu-lated to have an effect on the survival of the strains (Aase et al.,2000; Hegstad et al., 2010; Lundén et al., 2003). Low-level resis-tance may influence the survival of cells in situations in which theconcentration of the disinfectant is sublethal, which would provideisolates a selective advantage for adaptation to residual concen-trations of QACs (Hegstad et al., 2010). Sublethal concentrationsmay occur when disinfectants are applied to dirty or wet surfacesor in niches or reservoirs that are both difficult to clean/disinfectand difficult to rinse, i.e., some disinfectant may still be presentafter rinsing (Carpentier and Cerf, 2011; Lundén et al., 2003). Forexample, BAC can be found at concentrations of 0.05e6.03 mg L�1

in effluent water (Kümmerer et al., 1997). The low-level resistancemay also have a synergistic effect with disinfectant resistance inbiofilm cells (Aase et al., 2000).

S. Ortiz et al. / Food Microbiology 39 (2014) 81e88 87

The frequency of BAC-resistance among the isolates from year 4(91%) (Table 3) was considerably higher than the frequency foundin most studies (Aase et al., 2000; Mereghetti et al., 2000;Mullapudi et al., 2008; Soumet et al., 2005). In general, sources ofL. monocytogenes contaminationwere analyzed during four years inthe considered pork production facility, where QACs were usedmore frequently after the second year (personal communicationfrom the Company). Thus, the selection of L. monocytogenes isolatesof PCR serogroup IIa that are resistant to increased sublethal con-centrations of BAC may be associated with an increase in the use ofQACs. This finding was particularly evident in the case of the S10-1PFGE type, the resistance of which would explain the increase inthe number of isolates from the second to the third year, which waswhen most of the other PFGE types decreased (Table 1).

4. Conclusions

In conclusion, our findings suggest a relationship betweenresistance to increased sublethal concentrations of BAC and pro-longed survival of specific strains of L. monocytogenes in a foodprocessing environment. Conditions are likely to be present in foodfactories that may give rise to the development of persistingL. monocytogenes strains. The nature of this long-term persistence,however, is not thought to be due to the observed low-leveldisinfectant resistance but may be related to the type of disinfec-tant used and the frequency of cleaning and disinfection. Never-theless, our studies show that the current cleaning and disinfectionprogram, when applied correctly to equipment and environmentsthat are suitably hygienically designed, effectively control thepresence of most strains of L. monocytogenes in food factories.

Acknowledgments

The authors thank very much Pilar López and Natalia Monterofor technical assistance. This work was supported by ResearchProject grants RTA2008-00080-C02 and RTA2011-00098-C02 (INIAFEDER) from the Spanish Ministry of Economy and Competitive-ness. The support of the company Embutidos Fermín (La Alberca,Salamanca, Spain) is also gratefully acknowledged.

References

Aase, B., Sundheim, G., Langsrud, S., Rørvik, L.M., 2000. Occurrence of and a possiblemechanism for resistance to a quaternary ammonium compound in Listeriamonocytogenes. Int. J. Food Microbiol. 62, 57e63.

Briers, Y., Klumpp, J., Schuppler, M., Loessner, M.J., 2011. Genome sequence of Lis-teria monocytogenes Scott A, a clinical isolate from a food-borne listeriosisoutbreak. J. Bacteriol. 193, 4284e4285.

Carpentier, B., Cerf, O., 2011. Review. Persistence of Listeria monocytogenes in foodindustry equipment and premises. Int. J. Food Microbiol. 145, 1e8.

Chen, Y., Zhang, W., Knabel, S.J., 2005. Multi-virulence-locus sequence typingclarifies epidemiology of recent listeriosis outbreaks in the United States. J. Clin.Microbiol. 43, 5291e5294.

CLSI, 2008. Performance Standards for Antimicrobial Disk and Dilution Suscep-tibility Tests for Bacteria Isolated from Animals. Clinical and LaboratoryStandards Institute, Approved Standard, third ed. CLSI, Wayne, PA. CLSIdocument M31-A3.

Di Bonaventura, G., Piccolomini, R., Paludi, D., D’Orio, V., Vergara, A., Conter, M.,Ianieri, A., 2008. Influence of temperature on biofilm formation by Listeriamonocytogenes on various food-contact surfaces: relationship with motility andcell surface hydrophobicity. J. Appl. Microbiol. 104, 1552e1561.

Doumith, M., Buchrieser, C., Glaser, P., Jacquet, C., Martin, P., 2004. Differentiation ofthe major Listeria monocytogenes serovars by multiplex PCR. J. Clin. Microbiol.42, 3819e3822.

Earnshaw, A.M., Lawrence, L.M., 1998. Sensitivity to commercial disinfectants, andthe occurrence of plasmids within various Listeria monocytogenes genotypesisolated from poultry products and the poultry processing environment. J. Appl.Microbiol. 84, 642e648.

Elhanafi, D., Dutta, V., Kathariou, S., 2010. Genetic characterization of plasmid-associated benzalkonium chloride resistance determinants in a Listeria mono-cytogenes strain from the 1998e1999 outbreak. Appl. Environ. Microbiol. 76,8231e8238.

Ferreira, V., Barbosa, J., Stasiewicz, M., Vongkamjan, K., Moreno Switt, A.,Hogg, T., Gibbs, P., Teixeira, P., Wiedmann, M., 2011. Diverse geno- andphenotypes of persistent Listeria monocytogenes isolates from fermentedmeat sausage production facilities in Portugal. Appl. Environ. Microbiol. 77,2701e2715.

Fox, E.M., Leonard, N., Jordan, K., 2011. Physiological and transcriptional charac-terization of persistent and nonpersistent Listeria monocytogenes isolates. Appl.Environ. Microbiol. 77, 6559e6569.

Gilbert, P., McBain, A.J., 2003. Potential impact of increased use of biocides inconsumer products on prevalence of antibiotic resistance. Clin. Microbiol. Rev.16, 189e208.

Halpin, J.L., Garrett, N.M., Ribot, E.M., Graves, L.M., Cooper, K.L., 2010. Re-evaluation,optimization, and multilaboratory validation of the PulseNet-standardizedpulsed-field gel electrophoresis protocol for Listeria monocytogenes. Food-borne Pathog. Dis. 7, 293e298.

Harvey, J., Gilmour, A., 2001. Characterization of recurrent and sporadic Listeriamonocytogenes isolates from raw milk and nondairy foods by pulsed-field gelelectrophoresis, monocin typing, plasmid profiling, and cadmium and antibioticresistance determination. Appl. Environ. Microbiol. 67, 840e847.

Hegstad, K., Langsrud, S., Lunestad, B.T., Scheie, A.A., Sunde, M., Yazdankhah, S.P.,2010. Does the wide use of quaternary ammonium compounds enhance theselection and spread of antimicrobial resistance and thus threaten our health?Microb. Drug Resist. 16, 91e104.

Heir, E., Lindstedt, B.A., Røtterud, O.J., Vardund, T., Kapperud, G., Nesbakken, T.,2004. Molecular epidemiology and disinfectant susceptibility of Listeria mon-ocytogenes from meat processing plants and human infections. Int. J. FoodMicrobiol. 96, 85e96.

Holah, J.T., Taylor, J.H., Dawson, D.J., Hall, K.E., 2002. Biocide use in the food industryand the disinfectant resistance of persistent strains of Listeria monocytogenesand Escherichia coli. J. Appl. Microbiol. 92 (Suppl.), 111Se120S.

Holch, A., Webb, K., Lukjancenko, O., Ussery, D., Rosenthal, B.M., Gram, L., 2013.Genome sequencing identifies two nearly unchanged strains of persistent Lis-teria monocytogenes isolated at two different fish processing plants sampled 6years apart. Appl. Environ. Microbiol. 79, 2944e2951.

Jacquet, C., Doumith, M., Gordon, J.I., Martin, P.M., Cossart, P., Lecuit, M., 2004.A molecular marker for evaluating the pathogenic potential of foodborne Lis-teria monocytogenes. J. Infect. Dis. 189, 2094e2100.

Jensen, A., Thomsen, L.E., Jørgensen, R.L., Larsen, M.H., Roldgaard, B.B.,Christensen, B.B., Vogel, B.F., Gram, L., Ingmer, H., 2008. Processing plantpersistent strains of Listeria monocytogenes appear to have a lower virulencepotential than clinical strains in selected virulence models. Int. J. Food Micro-biol. 123, 254e261.

Kastbjerg, V.G., Gram, L., 2012. Industrial disinfectants do not select for resistance inListeria monocytogenes following long term exposure. Int. J. Food Microbiol. 160,11e15.

Kastbjerg, V., Gram, L., 2009. Model systems allowing quantification of sensitivity todisinfectants and comparison of disinfectant susceptibility of persistent andpresumed nonpersistent Listeria monocytogenes. J. Appl. Microbiol. 106, 1667e1681.

Keto-Timonen, R., Tolvanen, R., Lundén, J., Korkeala, H., 2007. An 8-year surveillanceof the diversity and persistence of Listeria monocytogenes in a chilled foodprocessing plant analyzed by amplified fragment length polymorphism. J. FoodProt. 70, 1866e1873.

Kornacki, J.L., Gurtler, J.B., 2007. Incidence and control of Listeria in food processingfacilities. In: Ryser, E.T., Marth, E.H. (Eds.), Listeria, Listeriosis and Food Safety.CRC Press, Boca Raton, FL, pp. 681e766.

Kovacevic, J., Arguedas-Villa, C., Wozniak, A., Tasara, T., Allen, K.J., 2013. Examinationof food chain-derived Listeria monocytogenes strains of different serotypes re-veals considerable diversity in inlA genotypes, mutability, and adaptation tocold temperatures. Appl. Environ. Microbiol. 79, 1915e1922.

Kümmerer, K., Eitel, A., Braun, U., Hubner, P., Daschner, F., Mascart, G., Milandri, M.,Reinthaler, F., Verhoef, J., 1997. Analysis of benzalkonium chloride in the effluentfrom European hospitals by solid-phase extraction and high-performance liquidchromatography with post-column ion-pairing and fluorescence detection.J. Chromatogr. A 774, 281e286.

Leclercq, A., Chenal-Francisque, V., Dieye, H., Cantinelli, T., Drali, R., Brisse, S.,Lecuit, M., 2011. Characterization of the novel Listeria monocytogenes PCRserogrouping profile IVb-v1. Int. J. Food Microbiol. 147, 74e77.

Lemon, K.P., Freitag, N.E., Kolter, R., 2010. The virulence regulator PrfA promotesbiofilm formation by Listeria monocytogenes. J. Bacteriol. 192, 3969e3976.

Lindbäck, T., Secic, I., Rørvik, L.M., 2011. A contingency locus in prfA in a Listeriamonocytogenes subgroup allows reactivation of the PrfA virulence regulatorduring infection in mice. Appl. Environ. Microbiol. 77, 3478e3483.

López, V., Navas, J., Martínez-Suárez, J.V., 2013. Low potential virulence associatedwith mutations in the inlA and prfA genes in Listeria monocytogenes isolatedfrom raw retail poultry meat. J. Food Prot. 76, 129e132.

López, V., Ortiz, S., Corujo, A., López, P., Poza, D., Navas, J., Moreno, R., Martínez-Suárez, J.V., 2008. Different contamination patterns of lineage I and II strainsof Listeria monocytogenes in a Spanish broiler abattoir. Poult. Sci. 87, 1874e1882.

Lundén, J., Autio, T., Markkula, A., Hellström, S., Korkeala, H., 2003. Adaptive andcross-adaptive responses of persistent and non-persistent Listeria mono-cytogenes strains to disinfectants. Int. J. Food Microbiol. 82, 265e272.

McDonnell, G., Russell, A.D., 1999. Antiseptics and disinfectants: activity, action, andresistance. Clin. Microbiol. Rev. 12, 147e179.

S. Ortiz et al. / Food Microbiology 39 (2014) 81e8888

Mereghetti, L., Quentin, R., Marquet-Van Der Mee, N., Audurier, A., 2000. Lowsensitivity of Listeria monocytogenes to quaternary ammonium compounds.Appl. Environ. Microbiol. 66, 5083e5086.

Mertins, S., Joseph, B., Goetz, M., Ecke, R., Seidel, G., Sprehe, M., Hillen, W.,Goebel, W., Müller-Altrock, S., 2007. Interference of components of the phos-phoenolpyruvate phosphotransferase system with the central virulence generegulator PrfA of Listeria monocytogenes. J. Bacteriol. 189, 473e490.

Møretrø, T., Langsrud, S., 2004. Listeria monocytogenes: biofilm formation andpersistence in food-processing environments. Biofilms 1, 107e121.

Mullapudi, S., Siletzky, R.M., Kathariou, S., 2008. Heavy-metal and benzalkoniumchloride resistance of Listeria monocytogenes isolates from the environment ofturkey-processing plants. Appl. Environ. Microbiol. 74, 1464e1468.

Nilsson, R.E., Ross, T., Bowman, J.P., 2011. Variability in biofilm production by Listeriamonocytogenes correlated to strain origin and growth conditions. Int. J. FoodMicrobiol. 150, 14e24.

Orgaz, B., Puga, C.H., Martínez-Suárez, J.V., SanJose, C., 2013. Biofilm recovery fromchitosan action: a possible clue to understand Listeria monocytogenes persis-tence in food plants. Food Control 32, 484e489.

Orsi, R.H., den Bakker, H.C., Wiedmann, M., 2011. Listeria monocytogenes lineages:genomics, evolution, ecology, and phenotypic characteristics. Int. J. Med.Microbiol. 301, 79e96.

Ortiz, S., López, V., Villatoro, D., López, P., Dávila, J.C., Martínez-Suárez, J.V., 2010.A 3-year surveillance of the genetic diversity and persistence of Listeria mon-ocytogenes in an Iberian pig slaughterhouse and processing plant. FoodbornePathog. Dis. 7, 1177e1184.

Pan, Y., Breidt Jr., F., Gorski, L., 2010. Synergistic effects of sodium chloride, glucose,and temperature on biofilm formation by Listeria monocytogenes serotype 1/2aand 4b strains. Appl. Environ. Microbiol. 76, 1433e1441.

Ragon, M., Wirth, T., Hollandt, F., Lavenir, R., Lecuit, M., Le Monnier, A., Brisse, S.,2008. A new perspective on Listeria monocytogenes evolution. PLOS Pathog. 4,e1000146.

Rakic-Martinez, M., Drevets, D.A., Dutta, V., Katic, V., Kathariou, S., 2011. Listeriamonocytogenes strains selected on ciprofloxacin or the disinfectant benzalko-nium chloride exhibit reduced susceptibility to ciprofloxacin, gentamicin,

benzalkonium chloride, and other toxic compounds. Appl. Environ. Microbiol.77, 8714e8721.

Renier, S., Hébraud, M., Desvaux, M., 2011. Molecular biology of surface colonizationby Listeria monocytogenes: an additional facet of an opportunistic Gram-positivefoodborne pathogen. Environ. Microbiol. 13, 835e850.

Roche, S.M., Grépinet, O., Kerouanton, A., Ragon, M., Leclercq, A., Témoin, S.,Schaeffer, B., Skorski, G., Mereghetti, L., Le Monnier, A., Velge, P., 2012. Poly-phasic characterization and genetic relatedness of low-virulence and virulentListeria monocytogenes isolates. BMC Microbiol. 12, 304.

Romanova, N.A., Wolffs, P.F., Brovko, L.Y., Griffiths, M.W., 2006. Role of efflux pumpsin adaptation and resistance of Listeria monocytogenes to benzalkonium chlo-ride. Appl. Environ. Microbiol. 72, 3498e3503.

Sidhu, M.S., Sørum, H., Holck, A., 2002. Resistance to quaternary ammoniumcompounds in food-related bacteria. Microb. Drug Resist. 8, 393e399.

Soumet, C., Ragimbeau, C., Maris, P., 2005. Screening of benzalkonium chlorideresistance in Listeria monocytogenes strains isolated during cold smoked fishproduction. Lett. Appl. Microbiol. 41, 291e296.

Thévenot, D., Dernburg, A., Vernozy-Rozand, C., 2006. An updated review of Listeriamonocytogenes in the pork meat industry and its products. J. Appl. Microbiol.101, 7e17.

van Stelten, A., Simpson, J.M., Ward, T.J., Nightingale, K.K., 2010. Revelation by singlenucleotide polymorphism genotyping that mutations leading to a prematurestop codon in inlA are common among Listeria monocytogenes isolates fromready-to-eat foods but not human listeriosis cases. Appl. Environ. Microbiol. 76,2783e2790.

Verghese, B., Lok, M., Wen, J., Alessandria, V., Chen, Y., Kathariou, S., Knabel, S., 2011.comK prophage junction fragments as markers for Listeria monocytogenes ge-notypes unique to individual meat and poultry processing plants and a modelfor rapid niche-specific adaptation, biofilm formation, and persistence. Appl.Environ. Microbiol. 77, 3279e3292.

Wulff, G., Gram, L., Ahrens, P., Vogel, B.F., 2006. One group of genetically similarListeria monocytogenes strains frequently dominates and persists in several fishslaughter- and smokehouses. Appl. Environ. Microbiol. 72, 4313e4322.