Adaptive tolerance to phenolic biocides in bacteria from organic foods: Effects on antimicrobial susceptibility and tolerance to physical stresses

Food Research International 85 (2016) 131–143

Contents lists available at ScienceDirect

Food Research International journal homepage: www.elsevier.com/locate/foodres

Adaptive tolerance to phenolic biocides in bacteria from organic foods: Effects on antimicrobial susceptibility and tolerance to physical stresses Rebeca Gadea, Miguel Ángel Fernández Fuentes, Rubén Pérez Pulido, Antonio Gálvez ⁎, Elena Ortega Área de Microbiología, Departamento de Ciencias de la Salud, Facultad de Ciencias Experimentales, Universidad de Jaén, 23071-Jaén, Spain

a r t i c l e

i n f o

Article history: Received 19 February 2016 Received in revised form 19 April 2016 Accepted 24 April 2016 Available online 26 April 2016 Keywords: Antibiotics Biocides Tolerance Organic foods Stress survival

a b s t r a c t The aim of the present study was to analyze the effects of step-wise exposure of biocide-sensitive bacteria from organic foods to phenolic biocides triclosan (TC) and hexachlorophene [2,2′-methylenebis(3,4,6trichlorophenol)] (CF). The analysis included changes in the tolerance to the biocide itself, the tolerance to other biocides, and cross-resistance to clinically important antibiotics. The involvement of efflux mechanisms was also studied as well as the possible implication of modifications in cytoplasmic membrane fluidity in the resistance mechanisms. The influence of biocide tolerance on growth capacity of the adapted strains and on subsequent resistance to other physical stresses has also been analyzed. Repeated exposure of bacteria from organic foods to phenolic biocides resulted in most cases in partially increased tolerance to the same biocide, to dissimilar biocides and other antimicrobial compounds. Nine TC-adapted strains and six CF-adapted strains were able to develop high levels of biocide tolerance, and these were stable in the absence of biocide selective pressure. Most strains adapted to TC and one CF-adapted strain showed significantly higher anisotropy values than their corresponding wildtype strains, suggesting that changes in membrane fluidity could be involved in biocide adaptation. Exposure to gradually increasing concentrations of CF induced a decrease in heat tolerance. Biocide adaptation had no significant effects of gastric acid or bile resistance, suggesting that biocide adaptation should not influence survival in the gastrointestinal tract. © 2016 Elsevier Ltd. All rights reserved.

1. Introduction Biocides are widely used in health and agricultural settings as well as in the food industry to prevent bacterial contamination. However, elimination of food-borne bacteria has proven difficult and it has become clear that the efficacy of biocides may be questionable in some circumstances (Davidson & Harrison, 2002). Under-dosing of applied disinfectants and insufficient cleaning before disinfection can also significantly reduce the efficacy of disinfectants. Under such conditions, bacteria are regularly exposed to sub-lethal concentrations of disinfectants, and this can lead to adaption and reduction of susceptibility to dissimilar disinfectants or antimicrobial compounds (Braoudaki & Hilton, 2005; Loughlin, Jones, & Lambert, 2002; Tattawasart, Maillard, Furr, & Russell, 1999; Thomas, Maillard, Lambert, & Russell, 2000). Moreover, it has been suggested that exposure to sub-lethal concentrations of biocides could potentially have an impact on the responses of bacteria to commonly used food processes, enabling microorganisms to survive challenges such as the concentrations of biocides currently permitted for use in food environments (Sheridan, Lenahan, Duffy, Fanning, & ⁎ Corresponding author at: Área de Microbiología, Departamento de Ciencias de la Salud, Facultad de Ciencias Experimentales, Edif. B3, Universidad de Jaén, Campus Las Lagunillas s/n, 23071-Jaén, Spain. E-mail address: [email protected] (A. Gálvez).

http://dx.doi.org/10.1016/j.foodres.2016.04.033 0963-9969/© 2016 Elsevier Ltd. All rights reserved.

Burgess, 2012) or different physical and chemical food treatments. However, at present, there is a limited understanding of the mechanisms which contribute to biocide tolerance. Phenols, cresols, and their chlorinated derivatives may not only act on several cellular targets, but also induce leakage of intracellular materials from bacteria (McDonnell & Russell, 1999). At low concentrations, triclosan inhibits the NADH-dependent enoyl-[acyl carrier protein] reductase (Schweizer, 2001) but at higher concentrations, it also has membranotropic effects (Villalain, Mateo, Aranda, Shapiro, & Micol, 2001). This is in contrast to antibiotics, which are considered to have only one major target site of activity (Russell, 2003) thereby, facilitating the development of resistance. However, recent evidence suggests that mechanisms providing tolerance to biocides may also provide crossprotection to the activity of antibiotics (Condell et al., 2012; Mavri & Smole Možina, 2013; Middleton & Salierno, 2013). It has been suggested that cross-resistance to antimicrobial compounds, following exposure and adaptation to a biocide, could occur in a limited number of situations. These can be summarized as follows: (i) when the biocide and an antimicrobial compound act on the same cellular target, (ii) when the biocide and the antimicrobial compound have the same transport mechanism, (iii) where a biocide and antibiotic can be accommodated by the same resistance mechanism (Gilbert & McBain, 2003), and finally, (iv) in situations where genes contributing toward biocide tolerance and antibiotic resistance are carried on the same mobile genetic element

132

R. Gadea et al. / Food Research International 85 (2016) 131–143

(Poole, 2004). Exposure to biocides in the environment may select for bacterial strains tolerant to these compounds and exhibiting increased resistance to antibiotics through co- or cross-resistance mechanisms (Chuanchuen et al., 2001; Maillard, 2007; Cottell, Denyer, Hanlon, Ochs, & Maillard, 2009). There is also a need to determine the effect of biocide-induced sublethal injury of bacteria on subsequent resistance to other stresses. Cross-protection against heat as a result of various environmental stresses, including exposure to starvation conditions, ethanol, acids, H2O2 and alkali has been previously described in different bacteria (Leyer & Johnson, 1993; Lou & Yousef, 1996; Ryu, Deng, & Beuchat, 1999). Gastrointestinal tract (GIT) stress factors also influence the ability of food-borne bacteria to survive and maintain their effects on the host. Biocide adaptation of bacterial cells could be responsible for subsequent increased tolerance to both physical treatments usually applied in food industry and physiological protection against food-borne pathogens along the host gastrointestinal tract. These aspects should be further investigated. The aim of the present study was to analyze the effects of step-wise exposure of biocide-sensitive bacteria from organic foods to phenolic biocides triclosan (TC) and hexachlorophene [2,2′-methylenebis(3,4,6trichlorophenol)] (CF). The analysis included changes in the susceptibility to the biocide itself, the susceptibility to other biocides, and crossresistance to clinically important antibiotics. The involvement of efflux mechanisms was studied on the basis of presence of multidrug efflux pumps genes and restored sensitivity in the presence of the efflux pump inhibitor (EPI) reserpine in Gram-positive strains. Anisotropy values of the fluorescence polarization of DPH (1,6-diphenyl-1,3,5hexatriene) in wildtype and biocide-adapted strains were analyzed in order to determine the influence of exposure to increasing sub-lethal concentrations of these biocides on membrane fluidity. The effect of biocide-induced sub-lethal injury of bacteria on growth capacity and on subsequent resistance to other physical stresses as heat, gastric acid and bile salts was also analyzed.

terminator cycle sequencing with Quick Start kit (Beckman Coulter, CA, USA) according to the manufacturer's instructions. The sequence data were analyzed with a CEQ DNA analysis system (version 4.0). The overlapping sequences obtained with SP3, SP4 and SP5 were merged using Lasergene program, version 5.05 (DNASTAR, Inc., Madison, WI, USA). A search for homology of the DNA sequence was done using the BLAST algorithm available at the National Centre for Biotechnology Information (NCBI, USA). 2.3. Adaptation to biocides The tolerance of sensitive strains was gradually increased by serially inoculating 100 μl from an overnight bacterial culture into 10 ml of Tryptic Soy Broth (TSB; Scharlab, Barcelona) containing a range of concentrations of the phenolic biocides triclosan (TC) and hexachlorophene [2,2′-methylenebis(3,4,6-trichlorophenol)] (CF) according to the technique described by To, Favrin, Romanova, and Griffiths (2002). The cultures were incubated at 37 °C for 24 to 48 h and additional subcultures were prepared from the tube containing the highest concentration of biocide that resulted in turbidity after incubation. Strains were subcultured in tubes containing the same concentration and the next higher concentration of biocides. This procedure was continued until no growth was observed after 72 h of incubation at 37 °C. The concentrations of the biocides were as follows: 0.01, 0.1, 1, 5, 10, 50, 100, 200, 500 μg/ml, 1, 2, 5 and 10 mg/ml, depending upon the growth of the adapted microorganism. Control strains were cultivated in biocide-free medium in parallel to adapted strains. The suspension in the last tube with recorded growth was seeded on a TSA plate and the bacterial growth was collected, resuspended in 1 ml BHI broth supplemented with 20% glycerol and stored at −80 °C. The stability of the biocide-tolerant phenotype was determined in each adapted strain by repeated subculture in biocide-free medium. Subcultures were performed every 24 h, for 20 days. The MICs were determined after 5, 10, 15 and 20 passages. The culture purities were analyzed at each stage in terms of cell morphology and Gram staining.

2. Materials and methods 2.1. Bacterial strains A total of 76 biocide-sensitive bacterial strains previously isolated from organic foods and classified as sensitive to biocides and antibiotics (Fernández-Fuentes, Ortega Morente, Abriouel, Pérez Pulido, & Gálvez, 2012) were selected for this study. Strains were isolated from samples of 39 commercial organic foods (including flours, fruits and vegetables, legumes, cereals, rice, pastes, sauces, cheeses and manufactured products). All studied foods were certificated as obtained by organic production according to Spanish regulations. Strains were stored at −80 °C in Brain Heart Infusion (BHI) broth (Scharlab, Barcelona, Spain) supplemented with 20% glycerol. For the preparation of inocula, strains were incubated for 15 h in BHI broth at 37 °C. 2.2. Strain identification In the present study, the selected strains were identified by conventional tests (Gram staining, catalase and oxidase tests) and 16S rDNA sequencing. DNA was extracted with a bacterial genomic DNA extraction kit (GenElute™, Sigma-Aldrich, Madrid) and 16S rDNA was amplified as described by Abriouel et al. (2005). PCR amplification products were purified using a GFX PCR DNA and Gel Band Purification Kit (GE-Healthcare, Spain), and then sequenced by using the primers Sp3 (5′-TACGCATTTCACCKCTACA-3′, position 684 reverse), Sp4 (5′-CTCGTTGCGGGACTTAAC-3′, position 1089 reverse) and Sp5 (5′-GNTACCTTGTTACGACTT-3′, position 1492 reverse) according to Weisburg, Barns, Pelletier, and Lane (1991) in a CEQ 2000 XL DNA Analysis System (Beckman Coulter, CA, USA). The DNA sequence of amplicons was determined by using CEQ 2000 dye

2.4. Determination of adaptive tolerance, sensitivity to biocides and antibiotics The wildtype strains and the corresponding strains adapted to TC and CF were tested for sensitivity to other biocides and to antibiotics. The minimum inhibitory concentrations (MICs) of biocides and antibiotics were determined by the broth microdilution method in 96-well microtiter plates. Briefly, serial dilutions of each substance (previously dissolved in absolute alcohol when necessary) were incubated with bacterial suspensions adjusted to 5 × 10 5 colony-forming units (CFU)/ml in Trypticase Soya Broth (TSB; Scharlab). Growth and sterility controls were included for each isolate, as well as the vehicle, as a negative control. Microtiter plates were incubated at 37 °C and readings were performed after 20 h of incubation by visual reading and optical density (OD 595 nm) determination in an iMark Microplate Reader (BioRad, Madrid). The MIC value was defined as the lowest compound concentration that prevented bacterial growth after incubation for 20 h. When turbidity of wells did not allow OD to be determined, counts of CFU after plating 100 μl from the wells onto nutrient agar plates were used to determine the minimum bactericidal concentration (MBC). Biocides employed for the assays – didecyldimethylammonium bromide (AB), cetrimide (CE), hexachlorophene [2,2′-methylenebis(3, 4,6trichlorophenol)] (CF), chlorhexidine (CH), hexadecylpyridinium chloride (HDP), benzalkonium chloride (BC) and triclosan (TC) – were obtained from Sigma-Aldrich. Antibiotics were selected as representatives of main groups widely used for human therapy. Ampicillin (AM) was obtained from Laboratorio Reig Jofré, Barcelona (Spain), cefotaxime (CTX) and ceftazidime (CAZ) from Laboratorios Normon, Madrid (Spain), ciprofloxacin (CIP) from Fluka, Madrid (Spain), imipenem

R. Gadea et al. / Food Research International 85 (2016) 131–143

(IPM) from MSD, Madrid (Spain), sulfamethoxazol (SXT) and the combination sulfamethoxazol/trimethoprim (TMP/SXT) from Laboratorios Almirall, Barcelona (Spain), tetracycline (TE), and nalidixic acid (NA) (for Salmonella) from Sigma-Aldrich. The twofold serial dilutions used included: for ampicillin (AM), concentrations of 32 and 64 μg/ml (excepting Enterococcus: 16 and 32 μg/ml); for cefotaxime (CTX), concentrations of 64 and 128 μg/ml; for ceftazidime (CAZ) and nalidixic acid (NA) (for Salmonella), concentrations of 32 and 64 μg/ml; for ciprofloxacin (CIP), concentrations of 4 and 8 μg/ml; for imipenem (IPM) and tetracycline (TE), concentrations of 16 and 32 μg/ml; for sulfamethoxazol (SXT) concentrations of 512 and 1024 μg/ml and for the combination sulfamethoxazol/ trimethoprim (TMP/SXT), concentrations of 4/76 and 8/152 μg/ml. These concentrations correspond to the MICs and twice the MICs interpretative standard of antibiotic-resistant strains defined by Clinical and Laboratory Standards Institute standards (CLSI, 2012). The assays were carried out in triplicate, to confirm the reproducibility of the MIC data. 2.5. PCR detection of resistance genes The presence of resistance determinants for the different efflux pumps was investigated by PCR amplification using the primers and specific PCR conditions previously described by Swick, Morgan-Linnell, Carlson, and Zechiedrich (2011) for acrB and mdfA; He et al. (2011) for sugE; Nishino and Yamaguchi (2002) for yhiUV and evgA; Jonas, Murray, and Weinstock (2001) for 455emeA; Smith, Gemmell, and Hunter (2008) for qacA/B, qacC, qacG, qacH and qacJ; Kazama, Hamashima, Sasatsu, and Arai (1998) for qacEΔ1; Lee, Huda, Kuroda, Mizushima, and Tsuchiya (2003) for efrA and efrB; Patel, Kosmidis, Seo, and Kaatz (2010) for mdeA, mepA, norA, norB, norC, sdrM and sepA. Primers previously described were used to PCR amplify genes involved in resistance to β-lactam antibiotics (bla, the β-lactamase gene and mecA, coding for a low affinity penicillin-binding protein) and chloramphenicol (cat, the chloramphenicol acetyltransferase gene) (Hummel, Hertel, Holzapfel, & Franz, 2007; Patel et al., 2010), aminoglycosides (aac(6_)-Ie-aph(2_)-Ia, aph(2_)-Ib, aph(2_)-Ic, aph(2_)-Id, aph(3_)-IIIa, and ant(4_)-Ia genes) (Vakulenko et al., 2003), macrolides (the ermA, ermB, ermC, msrA/B, ereA, ereB, mphA, and mefA genes) (Sutcliffe, Grebe, Tait-Kamradt, & Wondrack, 1996) and lincosamide and streptogramin A (lsa gene) (Singh, Weinstock, & Murray, 2002; Singh & Murray, 2005). Enterococcus faecalis ATCC 29212 and S. aureus strains CECT 4465 were used as positive controls. 2.6. Role of efflux pumps in biocide resistance The involvement of efflux mechanisms in biocide tolerance and antibiotic resistance of those Gram-positive strains showing biocide or antibiotic resistance phenotype and genetic resistance determinants was studied on the basis of restored sensitivity to both type of compounds in the presence of the efflux pump inhibitor reserpine (Sigma-Aldrich, Madrid) at 25 μg/ml, according to Vecchione, Alexander, and Sello (2009). 2.7. Growth capacity

133

final concentration of approx. 106 CFU/ml, and incubated during 5 min in a thermostatically controlled circulating water bath at 60, 65 and 70 °C according to Osaili et al. (2006). After heating, samples were cooled immediately in an ice-bath and serial dilutions were surfaceplated on TSA to determine the viable cell concentration in CFU/ml. 2.9. Resistance to simulated gastric fluid and bile salts Survival in the presence of simulated gastric acid fluid and tolerance to bile salts, were evaluated in wildtype and biocide-adapted cells. The experiments were performed in triplicate. Gastric acid fluid was simulated by addition of 0.5% NaCl and pepsin (3 g/l) to TSB medium (adjusted to pH 2.0 with 0.1 N HCl) (Kim et al., 2014). Bile salt tolerance was determined in TSB broth (pH 7.0) supplemented with 0.5% bile salt (Difco, USA) according to De Boever and Verstraete (1999), slightly modified. The control for each strain was a TSB tube with no bile supplements. The samples were inoculated with approx. 106 CFU/ml and incubated for 0, 90 and 180 min at 37 °C. The number of colony-forming units at time zero and the survival of the strains after incubations were determined by surface plating on TSA of serial dilutions of samples. 2.10. Fluorescence anisotropy measurements Membrane fluidity was determined by measuring fluorescence anisotropy in intact whole cells by using hydrophobic 1,6-diphenyl1,3,5-hexatriene (DPH) (Sigma-Aldrich, Madrid, Spain) as a probe, as described by Alonso-Hernando, Alonso-Calleja, and Capita (2010). After incubation of inoculated TSB at 37 °C for 24 h, stationary phase cells were harvested by centrifugation at 3000 × g for 10 min and washed once in 10 ml of a sterile phosphate-buffer-solution (PBS) (Sigma-Aldrich). The pellet was suspended in 9.9 ml of the same buffer. The concentration of bacterial suspensions, monitored by plating on TSA, was approximately 108 CFU/ml. The determinations of fluorescence anisotropy were optimized as follows: a total of 100 μl 5 mM DPH, previously diluted in tetrahydrofuran (Sigma-Aldrich), was added to 9.9 ml cell suspension (final probe concentration 50 μM) and the intensity of emission of polarized light from labeled whole cells was monitored after 10 min of incubation at room temperature in darkness. Fluorescence anisotropy was determined with a Cary Eclipse Fluorescence Spectrophotometer (Varian Inc., CA, USA). The samples were excited with vertically polarized light at 350 nm, and the fluorescence intensity of the light emitted at 428 nm was measured through a polarizer both vertically and horizontally relative to the excitation light. The cutoff filter was set at 418 nm for the light emitted. The slit width was 5 nm for both excitation and emission light. Ten readings were automatically obtained for each sample. The fluorescence intensities measured were corrected for background fluorescence and light scattering from the unlabeled samples. Fluorescence anisotropy was calculated as follows: R ¼ ðIvv−GIvhÞ=ðIvv þ 2GIvhÞ G ¼ Ihv=Ihh

Growth capacity of wildtype and biocide-adapted strains was determined by optical density (OD 595 nm) reading in an iMark Microplate Reader (BioRad, Madrid) of 96-well microtiter plates inoculated with approx. 106 CFU/ml of each tested strain in TSB medium after 2, 4, 6, 8, 10 and 24 h of incubation at 37 °C. Sterility controls were included for each isolate. Each strain was inoculated in triplicate.

Ivv and Ivh are the fluorescence intensities determined at vertical and horizontal orientations of the emission polarizer when the excitation polarizer is set in the vertical position. This applies similarly for Ihv and Ihh with the horizontal excitation polarizer. G is a correction factor for background fluorescence and light scattering.

2.8. Heat tolerance

2.11. Statistical analysis

For heat tolerance assessment, triplicate tubes with 2 ml of peptone water were inoculated with 20 μl of bacterial suspensions to achieve a

Significant differences between wildtype and biocide-adapted strains regarding growth capacity, heat tolerance and fluorescence

134

R. Gadea et al. / Food Research International 85 (2016) 131–143

Table 1 Minimum inhibitory concentrations of wildtype strains and changes of minimum inhibitory concentrations in adapted strains after step-wise exposure to triclosan. BIOCIDE Strain

TC

CF

BC

HDP

CE

CH

AB

Wildtype Adapted Wildtype Adapted Wildtype Adapted Wildtype Adapted Wildtype Adapted Wildtype Adapted Wildtype Adapted (MIC)* (fold)** (MIC)* (fold)** (MIC)* (fold)** (MIC)* (fold)** (MIC)* (fold)** (MIC)* (fold)** (MIC)* (fold)** Bacillus cereus UJA62p Bacillus cereus UJA67p Bacillus cereus UJA69q Bacillus cereus UJA72g Bacillus cereus UJA82t Bacillus licheniformis UJA55q Bacillus sp. UJA49u Bacillus sp. UJA49v Bacillus sp. UJA55u Bacillus sp. UJA65p Enterococcus casseliflavus UJA49b Enterococcus faecium UJA44a Enterococcus faecium UJA49e Enterococcus faecium UJA49t Enterococcus faecium UJA53c Enterococcus faecium UJA76h Enterococcus sp. UJA49h Enterococcus sp. UJA81m Staphylococcus saprophyticus UJA47j Staphylococcus saprophyticus UJA69g Staphylococcus saprophyticus UJA69h Staphylococcus xylosus UJA47g Staphylococcus sp. UJA69f Staphylococcus sp. UJA76i Enterobacter sp. UJA47m Enterobacter sp. UJA51m Enterobacter sp. UJA77x Enterobacter sp. UJA79p Enterobacter sp. UJA79t Pantoea agglomerans UJA48k Pantoea ananatis UJA59k Pantoea ananatis UJA79q Pantoea sp. UJA52o Pantoea sp. UJA80h Chryseobacterium sp. UJA48p Klebsiella oxytoca UJA49o Klebsiella oxytoca UJA56k

5

3

1

5

0.1

50

10

1

5

2

5

2

5

1

10

4

a

5

1

0.2

25

5

1

5

2

5

2

5

1

10

4a

5

1

5

4

5

1

5

1

5

2

10

1

5

3

1

5

0.1

100

5

1

10

1

5

2

5

1

5 5

3 3

1 1

20 5

30 0.2

1 25

0.1 10

300 1

5 5

10 2

5 5

16 2

5 15

4 1

5 5 5 5 5

2 2 2 2 2

1 1 1 5 1

1 1 1 1 5

0.2 0.5 0.5 1 0.1

25 10 10 10 50

10 10 10 5 0.5

1 1 1 1 10

10 10 10 5 5

1 1 1 3 2

5 5 5 5 5

16 18 10 2 4

5 5 5 5 5

2 2 2 1 1

10

2

1

5

0.1

100

5

1

10

1.5

5

1

5

1

10

2

5

1

0.5

10

0.5

10

5

1

5

1

5

1

10

4

a

1

5

0.5

10

0.5

10

10

1

5

2

5

1

10

3a

5

1

0.5

20

5

1

10

1

5

2

5

1

5

3

1

5

0.1

50

5

2

1

5

5

2

5

1

10

2

1

5

0.5

10

5

1

5

1

5

8

5

1

10

2

5

1

0.2

25

5

1

5

1

5

8

5

1

10

200 a

5

30

0.1

300

1

15

120

5

2

10

10

4a

5

1

0.1

50

5

1

5

2

5

2

5

3

5

3

1

5

5

1

5

1

1

10

5

6

5

2

5

5

1

1

0.1

1

5

1

1

5

5

1

1

5

5

2

5

1

5

1

30

1

30

1.6

5

1

5

1

150

1

5

5

1

10

5

2

5

2

1

5

1

10

1

5

5

1

10

1

20

1

5

2

5

1

10

2

5

6

10

2

30

1

50

1.2

5

18

15

10

2

10

4

20

50

1

5

12

10

9

5

1

10

10

3

10

1.5

30

1

20

1

10

8

15

1

15

10

1

5

1

60

1

10

1

5

3

5

1

150

1

5

0.5

10

10

2

30

1

5

3

5

3

1

200 a

1

30

1.5

20

10

2

60

1

1

10

5

3

1

5

1

5

1

30

5

6

10

4

5

4

5

3

5 10 10

3 2 2

1 5 1

1 1 5

0.5 5 10

30 1 1

30 5 5

1 1 1

4 30 15

5 5 5

2 2 1

5 5 5

4 2 1

10

2

10

3

0.5

40

30

1

10

4

5

18

5

4

5

3

1

5

1

5

1

10

1.5

5

2

5

2

0.1

0.1

1.5

10

0.5

0.5

5 0.5 1

1.5

1.3 6 1.3

R. Gadea et al. / Food Research International 85 (2016) 131–143

135

Table 1 (continued) BIOCIDE Strain

TC

CF

BC

HDP

CE

CH

AB

Wildtype Adapted Wildtype Adapted Wildtype Adapted Wildtype Adapted Wildtype Adapted Wildtype Adapted Wildtype Adapted (MIC)* (fold)** (MIC)* (fold)** (MIC)* (fold)** (MIC)* (fold)** (MIC)* (fold)** (MIC)* (fold)** (MIC)* (fold)** Salmonella UJA59l Salmonella UJA82k Salmonella UJA82l Enterobacteriaceae UJA 74q

10 15 10 100

2 200 a 200 a 2

1 5 5 10

40 2 6 2

0.5 2.5 30 10

40 12 1 2

5 30 40 20

4 1 1 1

20 5 80 30

3 12 1 1

5 5 5 10

18 10 2 9

5 15 15 10

3 1 1 1.5

* MIC of wildtype strains (in μg/ml). ** Fold increase in MIC for biocide-adapted strain. BC: benzalkonium chloride; HDP: hexadecylpyridinium chloride; CE: cetrimide; TC: triclosan; CH: chlorhexidine; CF: hexachlorophene [2,2′-methylenebis(3,4,6-trichlorophenol)]; AB: didecyldimethylammonium bromide. Significant alteration in susceptibility patterns (of at least fourfold increases) are displayed in bold. a Adaptive resistance was stable after 20 subcultures in biocide-free medium.

anisotropy measurements were analyzed with a Student t-test using the statistical software Statgraphics Centurion Version XVI.II (StatPoint Technologies, Inc., USA). Data on the increased tolerance to biocides after step-wise exposure to TC and CF were analyzed by principal component analysis (PCA) using MATLAB R2011b (7.13) version (MathWorks, USA). The Pearson correlation coefficient (r) was applied, and correlations were defined as very weak (0.00–0.19), weak (0.20–0.39), moderate (0.4–0.59), strong (0.60–0.79) or very strong (0.80–0.99), with a P significance of b0.05. 3. Results 3.1. Adaptation to TC Tolerance to triclosan was increased in 41 of the 76 strains (53.9%) exposed to gradually increasing concentrations of the biocide (Table 1). Strains with increased TC tolerance were identified as members of the genera Bacillus (10 strains), Enterococcus (8 strains), Staphylococcus (6 strains), Chryseobacterium (one strain), Enterobacter (5 strains), Pantoea (5 strains), Klebsiella (2 strains) and Salmonella (3 strains) as well as one non-identified Fam. Enterobacteriaceae member (Table 1). The MICs increased by 2-fold to 4-fold in most of cases, although an increase of 200-fold in the MIC compared to wildtype strains was observed in strains UJA47j (Staphylococcus saprophyticus), UJA59k (Pantoea ananatis), UJA82k and UJA82l (both identified as genus Salmonella), and an increase of 150-fold was observed in strains 76i (Staphylococcus sp.) and UJA48k (Pantoea agglomerans). Five strains showed 5 to 15-fold increases in their MICs after step-wise exposure to triclosan: UJA47g, (Staphylococcus xylosus), UJA47m, UJA79p and UJA79t (genus Enterobacter) and UJA79q (P. ananatis). Adaptive tolerance was stable after 20 subcultures in biocidefree medium only in nine of the TC-adapted strains, which were selected for subsequent studies: UJA67p and UJA69q (Bacillus cereus), UJA49t and UJA53c (Enterococcus faecium), UJA47j and UJA69g (S. saprophyticus), UJA 59k (P. ananatis), UJA 82k and UJA82l (both identified as genus Salmonella). This group also includes the four TC-adapted strains showing 200-fold increases in the MIC compared to wildtype strains. No differences were found in the MICs of control strains, cultivated in biocide-free medium in parallel to adapted strains, for the biocides tested. 3.2. Tolerance to other biocides in TC-adapted strains The results indicated that 87.8% of TC-adapted strains also showed increased tolerance to CH, 70.7% were more tolerant to BC, 63.4%

increased their tolerance to CF, 48.8% to CE and 41.5% to AB. The tolerance to HDP was increased in 29.2% of the TC-adapted strains as compared to wildtype strains. Remarkably, TC adaptation resulted in an increased tolerance to BC of between 50 and 300-fold in some strains (Table 1). Biocide tolerance profiles in TC-adapted strains showed two strains with increased tolerance to all six biocides tested compared to wildtype strains: Salmonella UJA59l and P. ananatis UJA79q. Nine TC-adapted strains showed increased tolerance to five of the biocides tested (two Enterococcus strains, two Staphylococcus, two Pantoea, one B. cereus, one Enterobacter and one Klebsiella). Seven TC-tolerant strains were more tolerant to four of the biocides tested: two Bacillus strains, two Staphylococcus, one Enterococcus, one Pantoea and one Salmonella. Remaining TC-adapted strains showed increased tolerance to 2 or 3 of the biocides tested and only strain Enterobacter sp. UJA79t, showed similar tolerance profile. Principal component analysis of increased tolerance to biocides in TC-adapted strains (Fig. 1A) revealed no correlation between TCadaptation and increased tolerance to HDP, CH or AB (r = 0.090 to −0.058). Weak correlations were found with BC- and CF-increased tolerance (r = 0.288 and 0.383, respectively) and moderate correlation was found when increased tolerance to CE was evaluated (r = 0.396). On the other hand, PCA analysis clearly revealed a strong correlation (r = 0.812) between increased tolerance to BC and CE in TC-adapted strains, indicating similar behavior against these biocides in TCadapted strains. PCA analysis also showed moderate correlations between CF and BC (r = 0.440) and between CF and CE (r = 0.429), indicating similar increases of tolerance to these couple of biocides in strains adapted to TC.

3.3. Resistance to antibiotics in TC-adapted strains Table 2 shows antibiotic resistance profiles in wildtype strains and in TC-adapted strains. Analysis of the antibiotic resistance profiles after exposure to TC shows an increase of resistance to at least one of the antibiotic tested in 26 of 41 TC-adapted strains (63.4%), requiring double the MIC of the corresponding wildtype strain for inhibition. Three strains (Staphylococcus sp. UJA69f, P. ananatis UJA79q and Salmonella UJA82k) showed increased resistance to four different antibiotics after exposure to TC and six strains (B. cereus UJA82t, S. saprophyticus UJA69h, P. agglomerans UJA48k, Enterobacteriaceae UJA74q, Pantoea sp. UJA80h and Salmonella UJA82l) showed increased resistance to three of the antibiotics tested. Twenty nine TC-adapted strains showed increased resistances to sulfamethoxazol (SXT), 14 to ceftazidime (CAZ), 12 to cefotaxime (CTX) and 7 to ampicillin (AM). Only four strains for the combination sulfamethoxazol/trimethoprim (TMP/SXT), two strains for nalidixic

136

R. Gadea et al. / Food Research International 85 (2016) 131–143

Enterobacteriaceae member. Thirteen of the strains had in common that they could also adapt to TC as previously shown in Table 1. Increases in MICs of adapted strains ranged from 3- to 80-fold compared to wildtype strains. Increases of 50 to 80-fold the MICs were detected in three strains: Enterococcus casseliflavus UJA53e, Enterobacter sp. UJA47m and Chryseobacterium sp. UJA48p. Adaptive biocide tolerance was stable after 20 subcultures in biocide-free medium in all these three strains. Increases of 10- to 40-fold were detected in eleven strains identified as Bacillus (one strain), Enterococcus (three strains), Pantoea (three strains), Enterobacter (two strains), Klebsiella and Salmonella (one strain each). Three of these strains showed stable resistance after 20 subcultures in biocide-free medium (Enterococcus sp. UJA53t, B. cereus UJA69q and Enterobacter ludwigii UJA79z). The remaining strains showed MIC increases of at least three-fold compared with wildtype strains. No differences were found in the MICs of biocides for control strains, cultivated in biocide-free medium in parallel to adapted strains. 3.5. Tolerance to other biocides in CF-adapted strains

Fig. 1. Biplot principal component analysis of increased tolerance to other biocides after step-wise exposure to TC (A) and CF (B). BC: benzalkonium chloride; HDP: hexadecylpyridinium chloride; CE: cetrimide; TC: triclosan; CH: chlorhexidine; CF: hexachlorophene [2,2′-methylenebis(3,4,6-trichlorophenol)]; AB: didecyldimethylammonium bromide.

acid (NA) and one strain for imipenem (IPM) showed an increased antibiotic resistance profile after exposure to gradually increasing concentrations of TC. None of the TC-adapted strains showed increased resistance to ciprofloxacin or tetracycline. On the other hand, 14 of the 41 TC-adapted strains showed a reduction in their antibiotic resistance (34.1%), although profiles of increased sensitivity were detected just for one of the antibiotics tested. Five strains showed identical antibiotic resistance profiles before and after the exposure to TC.

An 85.2% of CF-adapted strains showed increased tolerance to BC and 81.5% were more tolerant to CH. 55.5% increased their tolerance to CE or AB and 40.7% were more tolerant to HDP (Table 3). The tolerance to TC was also increased in 33.3% of the CF-adapted strains as compared to wildtype ones. On the contrary, 37% of CF-adapted strains showed increased sensitivity to TC, 22.2% to HDP and 14.8% and 11.1% also showed increased sensitivity to CE and AB respectively. Biocide tolerance profiles in CF-adapted strains showed one strain (Enterobacteriaceae UJA74q) with increased tolerance to all six biocides tested compared to its wildtype strain. Nine CF-adapted strains showed increased tolerance to five of the biocides tested (four Enterococcus isolates, one Bacillus, two Enterobacter, one Pantoea and one Klebsiella). Five CF-adapted strains were more tolerant to four of the biocides tested: two Pantoea isolates, one Enterococcus, one Chryseobacterium and one Salmonella. Remaining CF-adapted strains showed increased tolerance to 1 to 3 of the biocides tested. In contrast, decreased tolerance to one or two of the biocides tested was detected in 16 out of 27 CF-adapted strains. All these strains showed co-occurring increased tolerance to other biocides. Principal component analysis of increased tolerance to the rest of biocides after step-wise exposure to CF (Fig. 1B) revealed negative correlations between CF adaptation and increased tolerance to TC (r = − 0.108), to BC (r = − 0.030), and to CH (r = − 0.018). Weak correlations were found with CE- and AB-increased tolerance (r = 0.224 and 0.255, respectively) and moderate correlation was found when increased tolerance to HDP in CF-adapted strains was evaluated (r = 0.521). Changes in TC tolerance after step-wise exposure to CF showed a completely different pattern when compared with the rest of biocides, as shown by the negative correlations observed with all of them (r = − 0.099 to − 0.206). In fact, ten of the CF-adapted strains even showed decreased tolerance to TC (Table 3). On the contrary, moderate correlations (r = 0.425 to 0.567) were found between BC and CH, BC and AB, and HDP and AB, indicating similar changes in tolerance to these pairs of biocides after selective pressure with CF. 3.6. Resistance to antibiotics in CF-adapted strains

3.4. Adaptation to CF Tolerance to CF was increased in 27 of 76 strains (35.5%) after exposure to gradually increasing concentrations of the biocide (Table 3). CF-adapted strains were identified as members of genera Bacillus (3 strains), Enterococcus (7 strains), Chryseobacterium (one strain), Enterobacter (5 strains), Pantoea (7 strains), Klebsiella (one strain) and Salmonella (2 strains) as well as the non-identified Fam.

Table 4 shows antibiotic resistance profiles in wildtype strains and in CF-adapted strains. An increase of resistance to at least one of the antibiotic tested was observed in 13 out of 27 CF-adapted strains (48.1%). Only one strain (Enterococcus sp. UJA49h) showed increased resistance to three antibiotics after exposure to CF and remaining CF-adapted strains showed increased resistance to one or two of the antibiotics tested.

R. Gadea et al. / Food Research International 85 (2016) 131–143

137

Table 2 Antibiotic resistance profiles in wildtype strains and in strains after step-wise exposure to gradually increasing concentrations of TC. Antibiotic resistance profile

Antibiotic resistance profile

Wildtype strains

TC-adapted strains

Wildtype strains

TC-adapted strains

Bacillus cereus UJA62p Bacillus cereus UJA67p Bacillus cereus UJA69q Bacillus cereus UJA72g Bacillus cereus UJA82t Bacillus licheniformis UJA55q Bacillus sp. UJA49u Bacillus sp. UJA49v Bacillus sp. UJA55u Bacillus sp. UJA65p Enterococcus casseliflavus UJA49b Enterococcus faecium UJA44a Enterococcus faecium UJA49e Enterococcus faecium UJA49t Enterococcus faecium UJA53c Enterococcus faecium UJA76h Enterococcus sp. UJA49h Enterococcus sp. UJA81m

CTX, 2CAZ – 2CAZ, 2SX CTX, 2CAZ, 2SX 2CAZ CAZ CTX, 2CAZ 2CTX, 2CAZ 2CTX, 2CAZ, 2SX – 2CAZ, 2SX 2CTX, 2CAZ, 2SX CIP, 2CAZ, 2SX 2CAZ, 2SX 2CTX, 2CAZ, 2SX CAZ, 2SX 2CAZ, 2SX 2SX

2CAZ, 2SX 2CAZ, 2SX 2SX 2CAZ, 2SX 2AM, 2CTX, 2CAZ, 2SX 2CAZ, 2SX 2CTX, 2CAZ, 2SX 2CAZ 2CTX, 2CAZ 2CAZ, 2SX 2CTX, 2CAZ, 2SX 2CTX, 2CAZ, 2SX 2CAZ, 2SX CTX, 2CAZ, 2SX 2CTX, 2CAZ, 2SX 2CAZ, 2SX 2CAZ, 2SX 2CAZ, 2SX

Staphylococcus xylosus UJA47g Staphylococcus sp. UJA69f Staphylococcus sp. UJA76i Enterobacter sp. UJA47m Enterobacter sp. UJA51m Enterobacter sp. UJA77x Enterobacter sp. UJA79p Enterobacter sp. UJA79t Pantoea agglomerans UJA48k Pantoea ananatis UJA59k Pantoea ananatis UJA79q Pantoea sp. UJA52o Pantoea sp. UJA80h Chryseobacterium sp. UJA48p Klebsiella oxytoca UJA49o Klebsiella oxytoca UJA56k Salmonella UJA59l Salmonella UJA82k

– – – SX AM, 2CTX, 2CAZ, 2SX 2CAZ, 2SX 2AM 2AM, 2CTX, 2CAZ, 2SX – 2CAZ – 2CAZ, 2SX – 2CTX, 2CAZ, 2SX 2CTX, 2CAZ, 2SX 2CTX, 2CAZ, 2SX 2CAZ 2CAZ, 2SX, 2NA

Staphylococcus saprophyticus UJA47j Staphylococcus saprophyticus UJA69g Staphylococcus saprophyticus UJA69h

– 2CAZ, 2SX CAZ, SX

2SX 2CAZ, 2SX 2CTX, 2CAZ, 2SX

Salmonella UJA82l Enterobacteriaceae UJA74q

2AM, CTX, CAZ, 2SX, NA SX

2CAZ, 2SX 2AM, 2CTX, 2CAZ, 2SX CAZ, 2SX 2CAZ, 2SX 2AM, 2CAZ, 2SX 2AM, 2CTX, 2CAZ, 2SX 2AM, 2CAZ, 2SX 2CTX, 2CAZ, 2SX 2AM, 2CAZ, 2SX 2TMP/SXT, 2SX AM, CTX, 2TMP/SXT, 2SX 2CAZ, 2SX 2CTX, 2CAZ, 2SX 2CAZ, 2SX 2CAZ, 2SX 2CAZ, 2SX 2SX, 2NA 2AM, CTX, IPM, 2CAZ, 2TMP/SXT, 2SX, 2NA AM, 2CTX, 2TMP/SXT, 2SX, 2NA 2CTX, 2CAZ, 2SX

TC: triclosan; AM: ampicillin; CTX: cefotaxime; CIP: ciprofloxacin; IPM: imipenem; CAZ: ceftazidime; TE: tetraciclin, TMP/SXT: trimetroprim/sulfametoxazol; SX: sulfametoxazol; NA: nalidixic acid. Significant alteration in antibiotic susceptibility patterns are displayed in bold. Antibiotic concentrations correspond to the MICs and twice the MICs (2X) interpretative standard of antibiotic-resistant strains defined by CLSI standards: AM: 32 μg/ml (except for Enterococcus sp.: AM = 32 μg/ml); CTX: 64 μg/ml; CIP: 4 μg/ml; IPM: 16 μg/ml; CAZ: 32 μg/ml; TE:16 μg/ml; SX: 512 μg/ml; TMP/SXT:4/76 μg/ml; NA:32 μg/ml.

Seven strains increased their resistance to ampicillin, six strains to the combination sulfamethoxazol/trimethoprim and three to sulfamethoxazol. Only two CF-adapted strains showed increased resistance for tetracycline and one strain for ceftazidime and cefotaxime. None of the CF-adapted strains showed increased resistance to ciprofloxacin, imipenem or nalidixic acid. On the contrary, 20 out of the 27 CF-adapted strains showed a reduction in their resistance to at least one of the antibiotics tested (74.1%). Moreover, five strains (Enterococcus durans UJA79n, Enterobacter sp. UJA51m, Klebsiella oxytoca UJA56k, E. ludwigii UJA79z and Salmonella UJA82k) showed decreased resistance to three antibiotics. A total of 16 CF-adapted strains showed decreased resistance to ceftazidime, 9 to cefotaxime and 7 to sulfamethoxazol. Loss of phenotypic resistance profile was also detected in four strains for ampicillin and one strain for nalidixic acid. 3.7. Detection of genes encoding for efflux pumps and antibiotic resistance in biocide-adapted strains Resistance genes detected in biocide-adapted strains are shown in Table 5. TC-adapted strains exhibited a limited number of efflux pump genes: two genes were detected in strains E. faecium UJA49t and Salmonella UJA82k and only one efflux pump gene was shown in remaining TC-adapted strains (except for S. saprophyticus UJA47j which showed no presence of efflux pump genes). Genes encoding resistance to quaternary ammonium compounds qacH and qacJ were detected only in two and one strains respectively. The acrB pump gene of the AcrAB/TolC system was detected in three strains, and other efflux pump genes (norA, mepA, mdeA, and sepA) were detected in just one strain. The remaining efflux pump genes investigated (mdfA, sugE, norB, norC, sdrM, 455emeA, qacA/B, qacC, qacG, qacEΔ1, efrA, efrB, yhiUV and evgA) were not detected in any TCadapted strain. As to specific antibiotic resistance determinants they were found in six TC-adapted strains. The mecA gene, which codes for an alternative penicillin binding protein (PBP29 or PBP2a) was detected in three

TC-adapted strains and ermC gene, involved in macrolide resistance, was detected in TC-adapted B. cereus UJA69q strain. Phosphotransferase type I (mphA gene) involved in inactivation of macrolides and the aminoglycoside resistance determinant ant(4_)-Ia were detected in two TCadapted strains (B. cereus UJA67p and Salmonella UJA82k). TC-adapted UJA67p strain also showed the aph(2_)-Ic gene while aac(6_)1eaph(2_)-Ia gene was detected in Salmonella UJA82l strain. Other resistance determinants analyzed (bla, cat, aph(2_)-Ib, aph(2_)-Id, aph(3_)IIIa, ermA, ermB, mrsA/B, ereA, ereB, mefA and lsa) were not detected in any TC-adapted strain. Among the CF-adapted strains, (Table 5), three efflux pump genes were detected in two of them: B. cereus UJA69q (sugE, 455emeA and qacEΔ1) and E. ludwigii UJA79z (acrB, norB, qacH) while two genes were detected in strains E. casseliflavus UJA53e (acrB and 455emeA) and Enterobacter sp. UJA47m (sugE and efrA). acrB pump gene of the AcrAB/TolC system was also detected in CF-adapted strains Enterococcus sp. UJA53t and Chryseobacterium sp. UJA48p. As to specific antibiotic resistance determinants, aminoglycoside resistance determinants were found in CF-adapted Enterobacter sp. UJA47m strain (ant(4_)-Ia and aac(6_)1e-aph(2_)-Ia) and E. ludwigii UJA79z strain (aph(2_)-Ic and ant(4_)-Ia). The multi-component macrolide efflux pump mrsA/B was also found in CF-adapted UJA79z strain. The remaining resistance determinants analyzed (bla, cat, aph(2_)-Ib, aph(2_)-Id, aph(3_)-IIIa, ermA, ermB, ermC, ereA, ereB, mefA, mphA and lsa) were not detected in any CF-adapted strains. 3.8. Effect of reserpine on biocide and antibiotic resistance in biocide adapted strains Gram-positive strains adapted to TC and to CF and showing high biocide tolerance or antibiotic resistance phenotype as well as genetic resistance determinants were tested for sensitivity to the different biocides and antibiotics in the presence of the efflux pump inhibitor reserpine. For the TC-adapted strains, addition of reserpine only induced a twofold reduction in the MIC for chlorhexidine in B. cereus UJA67p and E. faecium UJA53c, but it had no effect on the MICs for

138

R. Gadea et al. / Food Research International 85 (2016) 131–143

Table 3 Minimum inhibitory concentrations of wildtype strains and changes of minimum inhibitory concentrations in adapted strains after exposure to hexachlorophene. MIC (μg/ml) CF Strain

Bacillus cereus UJA69q Bacillus licheniformis UJA55p Bacillus sp. UJA65p Enterococcus casseliflavus UJA53e Enterococcus durans UJA79n Enterococcus faecalis UJA69t Enterococcus faecium UJA43b Enterococcus faecium UJA43c Enterococcus sp. UJA49h Enterococcus sp. UJA53t Enterobacter ludwigii UJA79z Enterobacter sp. UJA47m Enterobacter sp. UJA51k Enterobacter sp. UJA51m Enterobacter sp. UJA53b Chryseobacterium sp. UJA48p Pantoea agglomerans UJA49m Pantoea agglomerans UJA52n Pantoea agglomerans UJA56l Pantoea agglomerans UJA71o Pantoea ananatis UJA59k Pantoea ananatis UJA79q Pantoea sp. UJA52o Klebsiella oxytoca UJA56k Salmonella UJA59l Salmonella UJA82k Enterobacteriaceae UJA74q

TC

BC

HDP

CE

CH

AB

Wildtype Adapted Wildtype Adapted Wildtype Adapted Wildtype Adapted Wildtype Adapted Wildtype Adapted Wildtype Adapted (MIC)* (fold)** (MIC)* (fold)** (MIC)* (fold)** (MIC)* (fold)** (MIC)* (fold)** (MIC)* (fold)** (MIC)* (fold)** 5

30a

10

0.1

0.1

50

5

1

5

2

5

1

5

1

5

8

10

0.5

0.5

60

30

1

50

2

5

6

15

2

5 1

8 50 a

5 10

0.2 1

1 0.5

10 20

5 5

2 3

5 10

4 5

5 5

3 4

5 5

2 2

5

8

10

1

0.5

20

5

3

1

40

5

3

5

2

1

20

10

4

0.5

10

5

0.2

0.5

10

5

8

5

1

1

10

10

1

0.5

10

5

1

10

1

5

1

5

2

5

8

10

0.1

0.2

50

5

4

10

8

5

8

5

3

1

5

10

1

0.5

10

5

1

5

1

5

2

5

1

5

12 a

10

0.5

0.5

60

5

8

10

8

5

6

5

6

10

10 a

10

0.1

5

6

10

3

10

2

5

2

5

4

a

1

1

5

4

10

15

20

3

5

3

5

4

1

60

5

8

10

0.1

5

2

50

0.5

50

2

20

2

15

1

5

6

10

0.1

10

1

30

0.5

50

1

5

4

15

1

1

40

10

1

10

1

30

1

30

2

10

2

10

1

10

1

10

1

5

6

1

20

5

2

5

2

1

0.5

10

50

1

100

0.2

5

1

10

0.5

1

5

1

10

0.5

a

1

80

5

3

5

5

3

0.1

100

0.5

10

500

0.1

100

1

15

0.1

300

0.5

10

50

0.1

10

0.5

5

2

5

5

4

0.1

10

1

5

50

0.5

100

0.4

5

1

10

1

20

1

10

1

10

10

1

60

1

1

10

5

2

1

5

1

10

0.5

10

5

1

10

0.5

5

2

5

1

1 1

40 40

5 5

2 2

0.5 7

60 1

30 5

1 3

5 10

16 4

5 5

6 2

5 5

6 2

1 5 10

40 5 4

10 15 100

0.5 2.5 10

20 4 3

5 30 20

2 0.3 2

20

1 4 3

5 5 10

2 4 3

5 15 10

2 1 3

0.1 0.3 2

5 30

1 0.5

* MIC of wildtype strains (in μg/ml). ** Fold increase in MIC for biocide-adapted strain. BC: benzalkonium chloride; HDP: hexadecylpyridinium chloride; CE: cetrimide; TC: triclosan; CH: chlorhexidine; CF: hexachlorophene; [2.2′-methylenebis(3,4,6-trichlorophenol)]; AB: didecyldimethylammonium bromide. Significant increases in resistance patterns (of at least fourfold) are displayed in bold. Susceptibility increases are underlined. a Adaptive resistance was stable after 20 subcultures in biocide-free medium. Strains highlighted in bold also had the capacity to adapt to triclosan (see Table 1).

the other biocides including TC (data not shown). Antibiotic sensitivity in presence and absence of reserpine showed similar results, so the level of antibiotic resistance was not significantly modified in presence of the efflux pump inhibitor in any of the isolates analyzed (data not shown). Addition of reserpine to CF-adapted strains induced a two-fold reduction in the MIC's in E. casseliflavus UJA53e for didecyldimethylammoniumbromide and in B. cereus UJA69q for benzalkonium chloride, but it did not have any effect on the levels

of tolerance to the remaining biocides or in antibiotic resistance levels (data not shown). 3.9. Growth capacity, physical and chemical resistance of biocide-adapted cells A selection of 9 TC-adapted strains and 6 CF-adapted strains chosen according to their stable adaptive resistance after 20 subcultures in biocide-free medium were characterized phenotypically.

R. Gadea et al. / Food Research International 85 (2016) 131–143

139

Table 4 Antibiotic resistance profiles in wildtype strains and in strains after step-wise exposure to gradually increasing concentrations of hexachlorophene. Antibiotic resistance profile

Bacillus cereus UJA69q Bacillus licheniformis UJA55p Bacillus sp. UJA65p Enterococcus casseliflavus UJA53e Enterococcus durans UJA79n Enterococcus faecalis UJA69t Enterococcus faecium UJA43b Enterococcus faecium UJA43c Enterococcus sp. UJA49h Enterococcus sp. UJA53t Enterobacter ludwigii UJA79z Enterobacter sp. UJA47m Enterobacter sp. UJA51k Enterobacter sp. UJA51m

Antibiotic resistance profile

Wildtype strains

CF-adapted strains

2CAZ, 2SX AM, 2CAZ, 2SX – 2CTX, 2CAZ, 2SX 2CTX, 2CAZ, 2SX 2AM, 2SX CTX, 2CAZ, 2SX 2CTX, 2CAZ, 2SX 2CAZ, 2SX 2CTX, 2CAZ, 2SX 2AM, 2CAZ, 2SX SX AM, 2SX AM, 2CTX, 2CAZ, 2SX

– AM, 2SX – 2AM, 2SX – 2CAZ, 2TMP/SXT, 2SX 2CAZ, 2SX 2AM, 2SX 2CTX, 2CAZ, TE, 2TMP/SXT, 2SX AM, 2SX TE, 2TMP/SXT, SX – 2TMP/SXT, 2SX 2SX

Enterobacter sp. UJA53b Pantoea agglomerans UJA49m Pantoea agglomerans UJA52n Pantoea agglomerans UJA56l Pantoea agglomerans UJA71o Pantoea ananatis UJA59k Pantoea ananatis UJA79q Pantoea sp. UJA52o Chryseobacterium sp. UJA48p Klebsiella oxytoca UJA56k Salmonella UJA59l Salmonella UJA82k Enterobacteriaceae UJA74q

Wildtype strains

CF-adapted strains

AM, 2CTX, 2CAZ, 2SX 2CAZ, 2SX – 2CAZ, 2SX 2SX 2CAZ – 2CAZ, 2SX 2CTX, 2CAZ, 2SX 2CTX, 2CAZ, 2SX 2CAZ 2CAZ, 2SX, 2NA SX

2AM, 2SX 2SX – – 2SX 2TMP/SXT, SX 2TMP/SXT, SX 2AM, 2CAZ, 2SX AM, 2SX – – – AM, 2SX

CF: hexachlorophene; AM: ampicillin; CTX: cefotaxime; CAZ: ceftazidime; TE: tetraciclin, TMP/SXT: trimetroprim/sulfametoxazol; SX: sulfametoxazol; NA: nalidixic acid. Significant alteration in antibiotic susceptibility patterns are displayed in bold. Antibiotic concentrations correspond to the MICs and twice the MICs (2X) interpretative standard of antibiotic-resistant strains defined by CLSI standards: AM: 32 μg/ml (except for Enterococcus sp.: AM = 32 μg/ml); CTX: 64 μg/ml; CAZ: 32 μg/ml; TE:16 μg/ml; SX: 512 μg/ml; TMP/SXT:4/76 μg/ml; NA:32 μg/ml.

3.9.1. Growth capacity of biocide-adapted strains Similar growth capacities were found for wildtype and TC-adapted strains in all incubation times recorded. Higher optical density values were detected in CF-adapted strains from 4 h to 10 h of incubation compared to wildtype ones, although differences were not significant and similar optical densities were recorded in both groups after 24 h incubation, as shown in supplementary Table 1. 3.9.2. Heat tolerance of biocide-adapted strains Heat tolerance of wildtype and TC-adapted strains is shown in supplementary Table 2. We only found significantly higher heat tolerance in strains B. cereus UJA69q and P. ananatis UJA59k as well as impaired heat tolerance in strains B. cereus UJA67p and Salmonella UJUA82l after step-wise exposure to gradually increasing concentrations of TC, compared to wildtype strains. Wildtype strains UJA69q and UJA59k were inactivated at all temperatures tested, while TCadapted strains were only inactivated when temperature was raised to 70 °C. On the contrary, wildtype strains UJA67p and UJA82l were

Table 5 Presence of efflux pumps and specific antibiotic resistance genes in biocide-adapted strains. Efflux pump genes

Antibiotic resistance genes

TC-adapted strains E. faecium UJA49t E. faecium UJA53c S. saprophyticus UJA47j S. saprophyticus UJA69g B. cereus UJA67p

qacH, mep A qacH – mdeA sepA

B. cereus UJA69q P. ananatis UJA59k Salmonella UJA82k Salmonella UJA82l

acrB norA acrB, qacJ acrB

– mec A mecA – mecA, aph(2_)-Ic, ant(4_)-Ia, mphA ermC – ant(4_)-Ia, mphA aac(6_)-Ie-aph(2_)-Ia

CF-adapted strains E. casseliflavus UJA53e Enterococcus sp. UJA53t B. cereus UJA69q Enterobacter sp. UJA47m

acrB, 455emeA acrB sugE, 455emeA, qacEΔ1 sugE, efrA

E. ludwigii UJA79z

acrB, norB, qacH

Chryseobacterium sp. UJA48p

acrB

CF: hexachlorophene; TC: triclosan.

– – – aac(6_)-Ie-aph(2_)-Ia, ant(4_)-Ia aph(2_)-Ic, ant(4_)-Ia, mrsA/B

able to survive at the lower temperature tested, but step-wise exposure to TC impaired their heat tolerance, being inactivated even at 60 °C. Both wildtype and TC-adapted strains UJA49t and UJA53c, both identified as E. faecium, were able to survive at all heat treatments applied (60, 65 and 70 °C). CF adaptation induced a decrease of heat tolerance in 3 of 7 strains analyzed (supplementary Table 2): UJA79z (identified as Enterobacter), UJA53e and UJA53t (Enterococcus). Wildtype strain UJA79z was able to survive all heat treatments applied, while CF-adapted strain was inactivated even at 60 °C. Wildtype strains UJA53e and UJA53t were only inactivated after incubation at 70 °C, while these strains showed decreased heat tolerance after step-wise exposure to CF. On the contrary, wildtype strain UJA47m (Enterobacter sp.) was inactivated at all temperatures tested, while CF-adapted strain was able to survive at 60 °C. Remaining strains showed similar heat tolerance previously and after exposure to the biocide-induced stress. 3.9.3. Resistance to simulated gastric fluid and bile salts Incubation of cells in presence of NaCl and pepsin (pH 2) allowed us to evaluate the tolerance to simulated gastric acid fluid in wildtype and biocide-adapted cells in order to determine the adaptation after biocide stress. Results are shown in supplementary Table 3. Most of analyzed strains (both wildtype as well as TC-adapted) showed initial counts between 105 and 106 CFU/ml, and were completely inactivated after 90 or 180 min of incubation at 37 °C in the simulated gastric fluid. However, wildtype strains B. cereus UJA69q, P. ananatis UJA59k and Salmonella UJA82k were completely inactivated immediately after contacting with the simulated gastric acid fluid, whereas TC-adapted strains showed similar counts to other strains (in the range of 105 to 106 UFC/ml) at time 0. Tolerance to gastric acid fluid in CF-adapted strains showed no significant differences when compared to wildtype cells, so all strains were immediately inactivated after contact with simulated gastric acid fluid (data not shown). Tolerance to bile salts in wildtype and TC-adapted strains is shown in supplementary Table 4. No statistically significant differences were detected in the number of viable cells at time 0, 90 or 180 min in control samples (incubated with no bile salts) compared to samples incubated with 0.5% bile salts. Only wildtype strain B. cereus UJA69q was inactivated by 0.5% bile salts after 90 and 180 min of incubation, as well as wildtype and TC-adapted strain B. cereus UJA67p after 180 min of incubation. Similar results were obtained when wildtype and CF-adapted strains were compared. Again, only wildtype B. cereus UJA69q strain was

140

R. Gadea et al. / Food Research International 85 (2016) 131–143

inhibited by 0.5% bile salts after 90 and 180 min of incubation (supplementary Table 4). 3.10. Influence of exposure to increasing sub-lethal concentrations of biocides on membrane fluidity Anisotropy values of the fluorescence polarization of DPH (1,6diphenyl-1,3,5-hexatriene) in the selected biocide-adapted strains showing a stable biocide tolerance phenotype and their corresponding wildtype strains are shown in Table 6. Higher values for DPH anisotropy were observed in all TC-adapted strains indicating higher membrane rigidity after growth in the presence of this biocide. Moreover, results were statistically significant in all strains tested excepting for S. saprophyticus UJA69g (increased anisotropy was also detected in this strain after exposure to TC, although it was not statistically significant). The anisotropy values in the membranes of CF-adapted cells were also higher than those of their corresponding wildtype cells, although significant differences (p b 0.01) were only found in the CF-adapted strain E. ludwigii UJA79z. 4. Discussion Results from the present study indicate that sequential exposure to increasing concentrations of the bisphenols TC and CF can induce biocide tolerance in a variety of gram-positive and gram-negative bacteria. Most important, the repeated exposure to TC and CF also induces increased resistance to other biocides and antibiotics. Several studies have demonstrated a relation between triclosan tolerance and resistance to antibiotics (Randall, Cooles, Piddock, & Woodward, 2004; Braoudaki & Hilton, 2004a, 2004b; Tkachenko et al., 2007; Middleton & Salierno, 2013), although some researchers have failed to demonstrate such cross-resistance (Lambert, 2004; Lear, Maillard, Dettmar, Goddard, & Russell, 2006; Stickler & Jones, 2008). A significant relationship between TC resistance and mar (multiple antibiotic resistance) phenotype was also described in triclosan-tolerant fecal coliforms isolated from surface waters when compared with TC-sensitive strains (Middleton & Salierno, 2013). On the contrary, increased susceptibility to specific aminoglycosides in TC-tolerant strains of E. coli was also described (Cottell et al., 2009). This was the case in some of our adapted strains, as we have also observed higher sensitivity to both biocides and antibiotics after step-wise exposure to TC or CF, compared with the wildtype strains (up to a 74% of CF-adapted strains showed a reduction in their resistance to at least one of the antibiotics tested). We have also found decreased resistance to one or two of the biocides tested in 16 out of 27 CF-adapted strains (mainly Gram-negatives), while all these strains showed co-occurring increased tolerance to other biocides. Similar results have been previously described in triclosan-tolerant mutants of E. coli (Cottell et al., 2009), in Campylobacter jejuni and

C. coli adapted to biocides (Mavri & Smole Možina, 2013) and in Pseudomonas aeruginosa following exposure to gradually increasing concentrations of CHA (Thomas et al., 2000). Benzalkonium chloride-tolerant strains have also been described as more sensitive to tobramycin, amikacin and gentamicin (Joynson, Forbes, & Lambert, 2002). The outer membrane of gram-negative bacteria acts as a permeability barrier to many antibacterial agents. This functional attribute is associated with the presence of structural components, lipopolysaccharide (LPS) and porins, and the precise molecular organization of these components (Vaara, 1992). Modification of these components causes an alteration of outer membrane permeability and leads to changes in antibacterial susceptibility. Hexachlorophene acts mainly by disrupting bacterial membranes, so step-wise exposure to CF may result in adaptations that facilitate diffusion of other antimicrobials across the outer membrane as previously proposed by Tattawasart, Maillard, Furr, and Russell (2000). Jones, Herd, and Chrisite (1989) also reported changes in fatty acid profiles consistent with outer membrane modifications in a strain of P. aeruginosa that had been adapted to an amphoteric surfactant and to a QAC. In addition, Guerin-Mechin, Dubois-Brissonnet, Heyd, and Leveau (1999) observed specific variations of fatty acid composition in P. aeruginosa adapted to grow in increasing concentrations of QACs. Modifications in cytoplasmic membrane fluidity have also been shown to be involved in the adaptation to environmental stresses (e.g. variations in temperature, pressure, ion concentrations, pH, nutrient availability, and xenobiotics) (Mykytczuk, Trevors, Leduc, & Ferroni, 2007; Badaoui Najjar, Chikindas, & Montville, 2007), so fluorescence anisotropy measurements, with DPH as a probe, were used to investigate the variations in membrane fluidity in a selection of TC- and CF-adapted strains showing a stable adaptation phenotype. Fluorescence anisotropy gives an inverse indication of membrane fluidity, i.e. a decreasing value of anisotropy indicates an increased disorder or fluidity (Shechter, 1997). Our results show significant increases of anisotropy in all TC-adapted strains, indicating higher membrane rigidity. The anisotropy values in the membranes of cells adapted to CF were also higher than those of wildtype cells, although they were less significant. These results suggest that adaptation to these compounds, especially to TC, is related to changes in membrane fluidity, as previously described for the acid decontaminants citric acid and peroxyacids (Alonso-Hernando et al., 2010). According to our results, adaptive tolerance to biocides after stepwise exposure to these compounds was stable after 20 subcultures in biocide-free medium only in nine of the TC-adapted strains and six of the CF-adapted strains. A rapid decrease in antimicrobial resistance in Pseudomonas aeruginosa following exposure to gradually increasing concentrations of CHA has also been reported (Thomas et al., 2000). The biocide-stress response of bacteria from organic foods seems to be transient, as has been shown for E. coli (Taglicht, Padan, Oppenheim, & Schuldiner, 1987). However, the duration of

Table 6 DPH (1, 6-diphenyl-1, 3, 5-hexatriene) anisotropy values in selected strains with stable biocide adaptation phenotypes to triclosan and hexachlorophene and their corresponding wildtype strains. TC-adapted strains

CF-adapted strains

Strain

Wildtype

Adapted

Strain

Wildtype

Adapted

E. faecium UJA49t E. faecium UJA53c S. saprophyticus UJA47j S. saprophyticus UJA69g B. cereus UJA67p B. cereus UJA69q P. ananatis UJA59k Salmonella UJA82k Salmonella UJA82l

0.139 ± 0.021 0.237 ± 0.041 0.168 ± 0.034 0.146 ± 0.028 0.243 ± 0.010 0.191 ± 0.041 0.224 ± 0.052 0.103 ± 0.017 0.259 ± 0.054

0.201 ± 0.031** 0.286 ± 0.051* 0.248 ± 0.033** 0.149 ± 0.037 0.289 ± 0.011* 0.274 ± 0.018** 0.411 ± 0.034** 0.219 ± 0.085** 0.324 ± 0.044**

E. casseliflavus UJA53e Enterococcus sp. UJA53t B. cereus UJA69q Enterobacter sp. UJA47m E. ludwigii UJA79z Chryseobacterium sp. UJA48p

0.142 ± 0.031 0.115 ± 0.029 0.136 ± 0.024 0.172 ± 0.085 0.177 ± 0.039 0.342 ± 0.096

0.150 ± 0.027 0.121 ± 0.021 0.141 ± 0.027 0.184 ± 0.050 0.233 ± 0.023** 0.404 ± 0.113

TC: triclosan; CF: hexachlorophene. ⁎ p b 0.05; ** p b 0.01.

R. Gadea et al. / Food Research International 85 (2016) 131–143

the biocide-stress response in situ and in food matrices needs to be determined. Cross-resistance between different biocides has been reported for P. aeruginosa by Lambert, Joynson, and Forbes (2001) and Murtough, Hiom, Palmer, and Russell (2001) and similar reports were found for E. coli O157 and Salmonella enterica (Braoudaki & Hilton, 2004b). This linkage might be due to a nonspecific reduction in cell permeability, as suggested by our results on higher membrane rigidity found in strains after growth in the presence of phenolic biocides, which might not allow chemically unrelated molecules into the resistant cells. Protection against biocides may also be achieved by a reduced uptake or alternatively and less commonly, by enzymatic biodegradation of the compound. Biocide inactivation is known but comparatively rare and specific to a few classes of biocides [e.g. organomercurials (McDonnell & Russell, 1999; Miller, 1999)]. More commonly, protection results from changes in cell envelope permeability or enhanced biocide efflux (McDonnell & Russell, 1999; Schweizer, 2001). Screening for multidrug efflux pumps and specific antibiotic resistance genes revealed a limited number of resistance determinants in our biocide-adapted strains. Although the presence of reserpine did not induce a loss of resistance in most of the biocide-adapted grampositive strains, efflux pumps seem to play an important role in acquisition of resistance in specific strains, as TC-adapted B. cereus UJA67p and E. casseliflavus UJA53c for chlorhexidine, CF-adapted E. casseliflavus UJA53e for didecyldimethylammoniumbromide, or B. cereus UJA69q for benzalkonium chloride. Furthermore, a fourfold reduction in amoxicillin MIC was observed in E. casseliflavus UJA53e in presence of reserpine. Reserpine is known as a competitive efflux pump inhibitor (EPI) that acts on the members of the ATP-binding cassette transporters, the major facilitator super-family (MFS) and the resistance nodulation division (RND) family (Braoudaki & Hilton, 2005; Marquez, 2005). The differences in the efficiency and the target preferences between the EPIs have been reported previously (Braoudaki & Hilton, 2005; Mavri & Smole Možina, 2012; Pannek et al., 2006). Thus the exclusion of one of these efflux systems can lead to increased activity of another efflux system, or even to the activation of other resistance mechanisms. Based in our results, efflux mechanisms may play an important role in the adaptation to biocides in our strains from organic foods as diverse resistance genes have been detected in biocide-adapted cells. This coincides with a previous finding on the interactive involvement of the CmeABC and CmeDEF efflux pumps in the extrusion of toxic compounds in C. jejuni, (Akiba, Lin, Barton, & Zhang, 2006). When the response to other stresses in biocide-adapted strains was evaluated, we found significantly higher heat tolerance in TC-adapted strains B. cereus UJA69q and P. ananatis UJA59k, compared to wildtype strains. Cross-protection against heat as a result of alkaline stress had also been previously documented for both Gram-positive (Flahaut, Hartke, Giard, & Auffray, 1997; Lou & Yousef, 1996) and Gram-negative bacteria (Humphrey, Richardson, Gawler, & Allen, 1991). Since induction of thermo-tolerance is known to occur in bacteria exposed to various environmental stresses, the potential for development of thermo-tolerance in biocide-adapted bacteria from foods is likely. Cross-contamination of organic foods with bacteria from environmental sources remains a critical issue in this food product manufacture. It is reasonable to propose that stressed cells may survive otherwise lethal heat treatments and proliferate on foods during subsequent storage. Post-processing contamination of food with stressed bacterial cells may permit survival of the pathogen during reheating just prior to consumption as previously described for Listeria monocytogenes (Taormina & Beuchat, 2001). Similar scenarios may compromise the safety of minimally processed, ready-to-eat, or heatand-serve foods in retail, food service, and home settings, where biocides are being used with increased frequency. Exposure to gradually increasing concentrations of CF, on the contrary, induced loss of heat tolerance in 3 of 7 strains analyzed, suggesting that these strains

141

would have lower chances of surviving mild heating processes in the food chain. We also evaluated the tolerance to gastric acid fluid and bile salts in wildtype and biocide-adapted cells. Only TC-adapted strains B. cereus UJA69q, P. ananatis UJA59k and Salmonella UJA82k showed increased immediate resistance to gastric acid fluid when compared to wildtype strains, while resistance after 90 or 180 min of incubation was similar in both groups. Tolerance to gastric acid fluid in CF-adapted strains showed no significant differences when compared to wildtype cells, as well as tolerance to bile salts in most of TC-adapted and CF-adapted strains, as the number of viable cells at time 0, 90 or 180 min in control samples (incubated with no bile salts) compared to samples incubated with 0.5% bile salts were similar. Only TC- and CF-adapted B. cereus UJA69q strain showed enhanced tolerance to bile salts after 90 and 180 min of incubation compared to wildtype UJA69q strain. All gastrointestinal tract (GIT) stress factors influence the ability of bacteria to maintain their effects to host macro-organism (Lebeer et al., 2008). The crossadaptation between acid and bile resistance may be present, considering some of the common adaptation mechanisms, although bile salt hydrolases are responsible for specific adaptation mechanisms of GIT bacteria to bile (De Boever & Verstraete, 1999). Our results show no significant effects on gastric acid or bile resistance after step-wise exposure to phenolic biocides in most of the strains analyzed, suggesting that biocide adaptation has no subsequent influence on the physiological protection against food-borne pathogens along the host gastrointestinal tract. 5. Conclusions In conclusion, repeated exposure of bacteria from organic foods to phenolic biocides resulted in most cases in partially increased tolerance to the same biocide, to dissimilar biocides and other antimicrobial compounds. However these resistances were only partially stable and we have also observed biocide-adapted strains with higher sensitivity to both biocides and antibiotics compared with the wildtype strains, especially when strains were exposed to CF. Nevertheless, a reduced percentage of strains were able to develop high levels of resistance, and these were stable in the absence of biocide selective pressure. Active efflux may be one of the mechanisms responsible for this adaptive resistance, wherein more than one type of efflux pumps is involved. Significant increases of anisotropy in all strains after step-wise exposure to TC suggests that adaptation to this compound is also probably related to changes in membrane fluidity. Exposure to gradually increasing concentrations of CF induced loss of heat tolerance in 3 of 7 strains analyzed so the use of this biocide could be recommended to prevent proliferation of these bacteria during subsequent storage. Our results show no significant effects on gastric acid or bile resistance after stepwise exposure to phenolic biocides so these compounds could also be used in food environments with no subsequent influence on the physiological protection against food-borne pathogens along the host gastrointestinal tract. Acknowledgments This work was supported by research group AGR230, grant P08AGR-4295, and the University of Jaen research support plan. Appendix A. Supplementary data Supplementary data to this article can be found online at http://dx. doi.org/10.1016/j.foodres.2016.04.033. References Abriouel, H., Lucas, R., Ben Omar, N., Valdivia, E., Maqueda, M., Martínez-Cañamero, M., & Galvez, A. (2005). Enterocin AS-48RJ: A variant of enterocin AS-48 chromosomally encoded by Enterococcus faecium RJ16 isolated from food. Systematic and Applied Microbiology, 28, 383–397. http://dx.doi.org/10.1016/j.syapm.2005.01.007.

142

R. Gadea et al. / Food Research International 85 (2016) 131–143

Akiba, M., Lin, J., Barton, Y. -W., & Zhang, Q. (2006). Interaction of CmeABC and CmeDEF in conferring antimicrobial resistance and maintaining cell viability in Campylobacter jejuni. Journal of Antimicrobial Chemotherapy, 57, 52–60. http://dx.doi.org/10.1093/ jac/dki419. Alonso-Hernando, A., Alonso-Calleja, C., & Capita, R. (2010). Effects of exposure to poultry chemical decontaminants on the membrane fluidity of Listeria monocytogenes and Salmonella enterica strains. International Journal of Food Microbiology, 137, 130–136. http://dx.doi.org/10.1016/j.ijfoodmicro.2009.11.022. Badaoui Najjar, M., Chikindas, M., & Montville, T. J. (2007). Changes in Listeria monocytogenes membrane fluidity in response to temperature stress. Applied and Environmental Microbiology, 73, 6429–6435. http://dx.doi.org/10.1128/AEM.00980-07. Braoudaki, M., & Hilton, A. C. (2004a). Adaptive resistance to biocides in Salmonella and Escherichia coli O157 and cross-resistance to antimicrobial agents. Journal of Clinical Microbiology, 42, 73–78. http://dx.doi.org/10.1128/JCM.42.1.73-78.2004. Braoudaki, M., & Hilton, A. C. (2004b). Low level of cross-resistance between triclosan and antibiotics in Escherichia coli K-12 and E. coli O55 compared to E. coli O157. FEMS Microbiology Letters, 253, 305–309. http://dx.doi.org/10.1111/j.1574-6968.2004. tb09603.x. Braoudaki, M., & Hilton, A. C. (2005). Mechanisms of resistance in Salmonella enterica adapted to erythromycin, benzalkonium chloride and triclosan. International Journal of Antimicrobial Agents, 25, 31–37. http://dx.doi.org/10.1016/j.ijantimicag.2004.07. 016. Chuanchuen, R., Beinlich, K., Hoang, T. T., Becher, A., Karkhoff-Schweizer, R. R., & Schweizer, H. P. (2001). Cross-resistance between triclosan and antibiotics in Pseudomonas aeruginosa is mediated by multidrug efflux pumps: exposure of a susceptible mutant strain to triclosan selects nfxB mutants overexpressing MexCD-Opr. Antimicrobial Agents and Chemotherapy, 45, 428–432. http://dx.doi.org/10.1128/AAC. 45.2.428-432.2001. Clinical and Laboratory Standards Institute (2012). Methods for dilution antimicrobial susceptibility tests for bacteria that grow aerobically (9th ed.). Wayne, PA: Clinical and Laboratory Standards Institute; CLSI publication M7-A9. Condell, O., Iversen, C., Cooney, S., Power, K. A., Walsh, C., Burgess, C., & Fanning, S. (2012). Efficacy of biocides used in the modern food industry to control Salmonella enterica, and links between biocide tolerance and resistance to clinically relevant antimicrobial compounds. Applied and Environmental Microbiology, 78, 3087–3097. http://dx.doi. org/10.1128/AEM.07534-11. Cottell, A., Denyer, S. P., Hanlon, G. W., Ochs, D., & Maillard, J. -Y. (2009). Triclosan-tolerant bacteria: Changes in susceptibility to antibiotics. Journal of Hospital Infection, 72, 71–76. http://dx.doi.org/10.1016/j.jhin.2009.01.014. Davidson, P. M., & Harrison, M. A. (2002). Resistance and adaptation to food antimicrobials, sanitizers, and other process controls. Food Technology, 56, 69–78 (ISSN 0015-6639). De Boever, P., & Verstraete, W. (1999). Bile salt deconjugation by Lactobacillus plantarum 80 and its implication for bacterial toxicity. Journal of Applied Microbiology, 87, 345–352. http://dx.doi.org/10.1046/j.1365-2672.1999.00019.x. Fernández-Fuentes, M. A., Ortega Morente, E., Abriouel, H., Pérez Pulido, R., & Gálvez, A. (2012). Isolation and identification of bacteria from organic foods: Sensitivity to biocides and antibiotics. Food Control, 26, 73–78. http://dx.doi.org/10.1016/j.foodcont. 2012.01.017. Flahaut, S., Hartke, A., Giard, J. -C., & Auffray, Y. (1997). Alkaline stress response in Enterococcus faecalis: Adaptation, cross-protection, and changes in protein synthesis. Applied and Environmental Microbiology, 63, 812–814 (PMC168376). Gilbert, P., & McBain, A. J. (2003). Potential impact of increased use of biocides in consumer products on prevalence of antibiotic resistance. Clinical Microbiology Reviews, 16, 189–208. http://dx.doi.org/10.1128/CMR.16.2.189-208.2003. Guerin-Mechin, L., Dubois-Brissonnet, F., Heyd, B., & Leveau, J. Y. (1999). Specific variations of fatty acid composition of Pseudomonas aeruginosa ATCC 15442 induced by quaternary ammonium compounds and relation with resistance to bactericidal activity. Journal of Applied Microbiology, 87, 735–742. http://dx.doi.org/10.1046/j.13652672.1999.00919.x. He, G. X., Zhang, C., Crow, R. R., Thorpe, C., Chen, H., Kumar, S., ... Varela, M. F. (2011). SugE, a new member of the SMR family of transporters, contributes to antimicrobial resistance in Enterobacter cloacae. Antimicrobial Agents and Chemotherapy, 55, 3954–3957. http://dx.doi.org/10.1128/AAC.00094-11. Hummel, A. S., Hertel, C., Holzapfel, W. H., & Franz, C. M. (2007). Antibiotic resistances of starter and probiotic strains of lactic acid bacteria. Applied and Environmental Microbiology, 73, 730–739. http://dx.doi.org/10.1128/AEM.02105-06. Humphrey, T. J., Richardson, N. P., Gawler, A. H. L., & Allen, M. J. (1991). Heat resistance of Salmonella enteritidis PT4: the influence of prior exposure to alkaline conditions. Letters in Applied Microbiology, 12, 258–260. http://dx.doi.org/10.1111/j.1472-765X. 1991.tb00553.x. Jonas, B. M., Murray, B. E., & Weinstock, G. M. (2001). Characterization of emeA, a NorA homolog and multidrug resistance efflux pump, in Enterococcus faecalis. Antimicrobial Agents and Chemotherapy, 45, 3574–3579. http://dx.doi.org/10.1128/ AAC.45.12.3574-3579.2001. Jones, M. V., Herd, T. M., & Chrisite, H. J. (1989). Resistance of Pseudomonas aeruginosa to amphoteric and quaternary ammonium biocides. Microbiology, 58, 49–61 (P MID: 2500580). Joynson, J. A., Forbes, B., & Lambert, R. J. (2002). Adaptive resistance to benzalkonium chloride, amikacin and tobramycin: The effect on susceptibility to other antimicrobials. Journal of Applied Microbiology, 93, 96–107. http://dx.doi.org/10.1046/j.13652672.2002.01667.x. Kazama, H., Hamashima, H., Sasatsu, M., & Arai, T. (1998). Distribution of the antisepticresistance genes qacE and qacE⍰1 in gram-negative bacteria. FEMS Microbiology Letters, 159, 173–178. http://dx.doi.org/10.1111/j.1574-6968.1998.tb12857.x. Kim, Y. W., Jeong, Y. J., Kim, A. Y., Son, H. H., Lee, J. A., Jung, C. H., ... Kim, J. (2014). Lactobacillus brevis strains from fermented aloe vera survive gastroduodenal environment

and suppress common food borne enteropathogens. PloS One, 9(3), e90866. http:// dx.doi.org/10.1371/journal.pone.0090866. Lambert, R. J. W., Joynson, J., & Forbes, B. (2001). The relationships and susceptibilities of some industrial, laboratory and clinical isolates of Pseudomonas aeruginosa to some antibiotics and biocides. Journal of Applied Microbiology, 91, 972–984. http://dx.doi. org/10.1046/j.1365-2672.2001.01460.x. Lambert, R. J. (2004). Comparative analysis of antibiotic and antimicrobial biocide susceptibility data in clinical isolates of methicillin-sensitive Staphylococcus aureus, methicillin-resistant Staphylococcus aureus and Pseudomonas aeruginosa between 1989 and 2000. Journal of Applied Microbiology, 97, 699–711. http://dx.doi.org/10. 1111/j.1365-2672.2004.02345.x. Lear, J. C., Maillard, J. -Y., Dettmar, P. W., Goddard, P. A., & Russell, A. D. (2006). Chloroxylenol- and triclosan-tolerant bacteria from industrial sources—susceptibility to antibiotics and other biocides. International Biodeterioration & Biodegradation, 57, 51–56. http://dx.doi.org/10.1016/j.ibiod.2005.11.002. Lebeer, S., Claes, I. J., Verhoeven, T. L., Shen, C., Lambrichts, I., Ceuppens, J. L., ... De Keersmaecker, S. C. J. (2008). Impact of luxS and suppressor mutations on the gastrointestinal transit of Lactobacillus rhamnosus GG. Applied and Environmental Microbiology, 74, 4711–4718. http://dx.doi.org/10.1128/AEM.00133-08. Lee, E. W., Huda, M. N., Kuroda, T., Mizushima, T., & Tsuchiya, T. (2003). EfrAB, an ABCmultidrug efflux pump in Enterococcus faecalis. Antimicrobial Agents and Chemotherapy, 47, 3733–3738. http://dx.doi.org/10.1128/AAC.47.12.3733-3738.2003. Leyer, G. J., & Johnson, E. A. (1993). Acid adaptation induces cross-protection against environmental stresses in Salmonella typhimurium. Applied and Environmental Microbiology, 59, 1842–1847 (PMC182170). Lou, A., & Yousef, A. E. (1996). Resistance of Listeria monocytogenes to heat after adaptation to environmental stresses. Journal of Food Protection, 59, 465–471. Loughlin, M. F., Jones, M. V., & Lambert, P. A. (2002). Pseudomonas aeruginosa cells adapted to benzalkonium chloride show resistance to other membrane-active agents but not to clinically relevant antibiotics. Journal of Antimicrobial Chemotherapy, 49, 631–639. http://dx.doi.org/10.1093/jac/49.4.631. Maillard, J. Y. (2007). Bacterial resistance to biocides in the healthcare environment: Should it be of genuine concern? Journal of Hospital Infection, 65, 60–72. http://dx. doi.org/10.1016/S0195-6701(07)60018-8. Marquez, B. (2005). Bacterial efflux systems and efflux pumps inhibitors. Biochimie, 87, 1137–1147 (1016/j.biochi.2005.04.012). Mavri, A., & Smole Možina, S. (2012). Involvement of efflux mechanisms in biocide resistance of Campylobacter jejuni and Campylobacter coli. Journal of Medical Microbiology, 61, 800–808. http://dx.doi.org/10.1099/jmm.0.041467-0. Mavri, A., & Smole Možina, S. (2013). Development of antimicrobial resistance in Campylobacter jejuni and Campylobacter coli adapted to biocides. International Journal of Food Microbiology, 160, 304–312. http://dx.doi.org/10.1016/j.ijfoodmicro. 2012.11.006. McDonnell, G., & Russell, A. D. (1999). Antiseptics and disinfectants: Activity, action and resistance. Clinical Microbiology Reviews, 12, 147–179 (PMC88971). Middleton, J. H., & Salierno, J. D. (2013). Antibiotic resistance in triclosan tolerant fecal coliforms isolated from surface waters near wastewater treatment plant outflows (Morris County, NJ, USA). Ecotoxicology and Environmental Safety, 88, 79–88. http:// dx.doi.org/10.1016/j.ecoenv.2012.10.025. Miller, S. M. (1999). Bacterial detoxification of Hg (II) and organomercurials. Essays in Biochemistry, 34, 17–30. http://dx.doi.org/10.1042/bse0340017. Murtough, S. M., Hiom, S. J., Palmer, M., & Russell, A. D. (2001). Biocide rotation in the healthcare setting: Is there a case for policy implementation? Journal of Hospital Infection, 48, 1–6. http://dx.doi.org/10.1053/jhin.2001.0950. Mykytczuk, N. C. S., Trevors, J. T., Leduc, L. G., & Ferroni, G. D. (2007). Fluorescence polarization in studies of bacterial cytoplasmic membrane fluidity under environmental stress. Progress in Biophysics and Molecular Biology, 95, 60–82. http://dx.doi.org/10. 1016/j.pbiomolbio.2007.05.001. Nishino, K., & Yamaguchi, A. (2002). EvgA of the two-component signal transduction system modulates production of the YhiUV multidrug transporter in Escherichia coli. Journal of Bacteriology, 184, 2319–2323. http://dx.doi.org/10.1128/JB.184.8.2319-2323.2002. Osaili, T., Griffis, C. L., Martin, E. M., Beard, B. L., Keener, A., & Marcy, J. A. (2006). Thermal inactivation studies of Escherichia coli O157:H7, Salmonella, and Listeria monocytogenes in ready-to-eat chicken-fried beef patties. Journal of Food Protection, 69, 1080–1086 (PMID: s(16715808). Pannek, S., Higgins, P. G., Steinke, P., Jonas, D., Akova, M., Bohnert, J. A., ... Kern, W. V. (2006). Multidrug efflux inhibition in Acinetobacter baumannii: comparison between 1-(1- naphthylmethyl)-piperizine and phenyl-arginine-β-naphthylamide. Journal of Antimicrobial Chemotherapy, 57, 970–974. http://dx.doi.org/10.1093/jac/dkl081. Patel, D., Kosmidis, C., Seo, S. M., & Kaatz, G. W. (2010). Ethidium bromide MIC screening for enhanced efflux pump gene expression or efflux activity in Staphylococcus aureus. Antimicrobial Agents and Chemotherapy, 54, 5070–5073. http://dx.doi.org/10.1128/ AAC.01058-10. Poole, K. (2004). Acquired resistance. In A. Fraise, P. A. Lambert, & J. -Y. Maillard (Eds.), Russell, Hugo & Ayliffe's principles and practice of disinfection, preservation and sterilization (pp. 154–169). Malden, Massachusetts: Blackwell Publishing Ltd. Randall, L. P., Cooles, S. W., Piddock, L. J. V., & Woodward, M. J. (2004). Effect of triclosan or a phenolic farm disinfectant on the selection of antibiotic-resistant Salmonella enterica. Journal of Antimicrobial Chemotherapy, 54, 621–627. http://dx.doi.org/10. 1093/jac/dkh376. Russell, A. D. (2003). Similarities and differences in the responses of microorganisms to biocides. Journal of Antimicrobial Chemotherapy, 52, 750–763. http://dx.doi.org/10. 1093/jac/dkg422. Ryu, J. -H., Deng, Y., & Beuchat, L. R. (1999). Behavior of acid-adapted and unadapted Escherichia coli O157:H7 when exposed to reduced pH achieved with various organic acids. Journal of Food Protection, 62, 451–455 (PMID:10340663).

R. Gadea et al. / Food Research International 85 (2016) 131–143 Schweizer, H. P. (2001). Triclosan: A widely used biocide and its link to antibiotics. FEMS Microbiology Letters, 202, 1–7. http://dx.doi.org/10.1111/j.1574-6968.2001.tb10772.x. Shechter, E. (1997). Fluidité Membranaire. In E. Shechter (Ed.), Biochimie et Biophysique des Membranes. Aspects Structuraux et Fonctionnels (pp. 95–110). Paris: Masson. Sheridan, A., Lenahan, M., Duffy, G., Fanning, S., & Burgess, C. (2012). The potential for biocide tolerance in Escherichia coli and its impact on the response to food processing stresses. Food Control, 26, 98–106. http://dx.doi.org/10.1016/j.foodcont.2012.01.018. Singh, K. V., & Murray, B. E. (2005). Differences in the Enterococcus faecalis lsa locus that influence susceptibility to quinupristin–dalfopristin and clindamycin. Antimicrobial Agents and Chemotherapy, 49, 32–39. http://dx.doi.org/10.1128/AAC.49.1.32-39.2005. Singh, K. V., Weinstock, G. M., & Murray, B. E. (2002). An Enterococcus faecalis ABC homologue (Lsa) is required for the resistance of this species to clindamycin and quinupristin–dalfopristin. Antimicrobial Agents and Chemotherapy, 46, 1845–1850. http://dx.doi.org/10.1128/AAC.46.6.1845-1850.2002. Smith, K., Gemmell, C. G., & Hunter, I. S. (2008). The association between biocide tolerance and the presence or absence of qac genes among hospital-acquired and community acquired MRSA isolates. Journal of Antimicrobial Chemotherapy, 61, 78–84. http://dx. doi.org/10.1093/jac/dkm395. Stickler, D. J., & Jones, G. L. (2008). Reduced susceptibility of Proteus mirabilis to triclosan. Antimicrobial Agents and Chemotherapy, 52, 991–994. http://dx.doi.org/10.1128/AAC. 01094-07. Sutcliffe, J., Grebe, T., Tait-Kamradt, A., & Wondrack, L. (1996). Detection of erythromycin resistant determinants by PCR. Antimicrobial Agents and Chemotherapy, 40, 2562–2566 (PMC163576). Swick, M. C., Morgan-Linnell, S. K., Carlson, K. M., & Zechiedrich, L. (2011). Expression of multidrug efflux pump genes acrAB-tolC, mdfA, and norE in Escherichia coli clinical isolates as a function of fluoroquinolone and multidrug resistance. Antimicrobial Agents and Chemotherapy, 55, 921–924. http://dx.doi.org/10.1128/AAC.00996-10. Taglicht, D., Padan, E., Oppenheim, A. B., & Schuldiner, S. (1987). An alkaline shift induces the heat shock response in Escherichia coli. Journal of Bacteriology, 169, 885–887 (PMC211862). Taormina, P. J., & Beuchat, I. R. (2001). Survival and heat resistance of Listeria monocytogenes after exposure to alkali and chlorine. Applied and Environmental Microbiology, 67, 2555–2563. http://dx.doi.org/10.1128/AEM.67.6.2555-2563.2001. Tattawasart, U., Maillard, J. -Y., Furr, J. R., & Russell, A. D. (1999). Development of resistance to chlorhexidine diacetate and cetylpyridinium chloride in Pseudomonas

143

stutzeri and changes in antibiotic susceptibility. Journal of Hospital Infection, 42, 219–229. http://dx.doi.org/10.1053/jhin.1999.0591. Tattawasart, U., Maillard, J. -Y., Furr, J. R., & Russell, A. D. (2000). Outer membrane changes in Pseudomonas to chlorhexidine diacetate and cetylpyridinium chloride. International Journal of Antimicrobial Agents, 16, 233–238. http://dx.doi.org/10.1016/ S0924-8579(00)00206-5. Thomas, L., Maillard, J. -Y., Lambert, R. J. W., & Russell, A. D. (2000). Development of resistance to chlorhexidine diacetate in Pseudomonas aeruginosa and the effect of a “residual” concentration. Journal of Hospital Infection, 46, 297–303. http://dx.doi.org/10. 1053/jhin.2000.0851. Tkachenko, O., Shepard, J., Aris, V. M., Joy, A., Bello, A., Londono, I., ... Peteroy-Kelly, M. A. (2007). A triclosan–ciprofloxacin cross-resistant mutant strain of Staphylococcus aureus displays an alteration in the expression of several cell membrane structural and functional genes. Research in Microbiology, 158, 651–658. http://dx.doi.org/10. 1016/j.resmic.2007.09.003. To, M. S., Favrin, S., Romanova, N., & Griffiths, M. W. (2002). Postadaptational resistance to benzalkonium chloride and subsequent physicochemical modifications of Listeria monocytogenes. Applied and Environmental Microbiology, 68, 5258–5264. http://dx. doi.org/10.1128/AEM.68.11.5258-5264.2002. Vaara, M. (1992). Agents that increase the permeability of the outer membrane. Microbiological Reviews, 56, 395–411 (PMC372877). Vakulenko, S. B., Donabedian, S. M., Voskresenskiy, A. M., Zervos, M. J., Lerner, S. A., & Chow, J. W. (2003). Multiplex PCR for detection of aminoglycoside resistance genes in enterococci. Antimicrobial Agents and Chemotherapy, 47, 1423–1426. http://dx. doi.org/10.1128/AAC.47.4.1423-1426.2003. Vecchione, J. J., Alexander, B., Jr., & Sello, J. K. (2009). Two distinct major facilitator superfamily drug efflux pumps mediate chloramphenicol resistance in Streptomyces coelicolor. Antimicrobial Agents and Chemotherapy, 53, 4673–4677. http://dx.doi.org/ 10.1128/AAC.00853-09. Villalain, J., Mateo, C. R., Aranda, F. J., Shapiro, S., & Micol, V. (2001). Membranotropic effects of the antibacterial agent triclosan. Archives of Biochemistry and Biophysics, 390, 128–136. http://dx.doi.org/10.1006/abbi.2001.2356. Weisburg, W. G., Barns, S. M., Pelletier, D. A., & Lane, D. J. (1991). 16S ribosomal DNA amplification for phylogenetic study. Journal of Bacteriology, 173, 697–703 (PMC207061).