Antibiotic and antimicrobial peptide combinations: synergistic inhibition of Pseudomonas fluorescens and antibiotic-resistant variants

Research in Microbiology 163 (2012) 101e108 www.elsevier.com/locate/resmic

Antibiotic and antimicrobial peptide combinations: synergistic inhibition of Pseudomonas fluorescens and antibiotic-resistant variants Karim Naghmouchi a,c,**, Christophe Le Lay b, John Baah a, Djamel Drider d,* a Lethbridge Research Center, Agricultureg and Agri-Food Canada, Lethbridge, AB, Canada STELA Dairy Research Center, Nutraceuticals and Functional Foods Institute, Universite´ Laval, G1K 7P4 Que´bec, QC, Canada c Laboratoire des Microorganismes et Biomole´cules Actives, Faculte´ des Sciences de Tunis, El Manar, Tunisia d Laboratoire des Proce´de´s Biologiques, Ge´nie Enzymatique et Microbien (ProBioGEM), UPRES-EA 1026, Polytech’Lille/IUTA, Universite´ Lille Nord de France, Avenue Paul Langevin, 59655 Villeneuve d’Ascq Cedex, France b

Received 24 October 2011; accepted 4 November 2011 Available online 27 November 2011

Abstract Variants resistant to penicillin G (RvP), streptomycin (RvS), lincomycin (RvL) and rifampicin (RvR) were developed from a colistinsensitive isolate of Pseudomonas fluorescens LRC-R73 (P. fluorescens). Cell fatty acid composition, Kþ efflux and sensitivity to antimicrobial peptides (nisin Z, pediocin PA-1/AcH and colistin) alone or combined with antibiotics were determined. P. fluorescens was highly sensitive to kanamycin, tetracycline and chloramphenicol at minimal inhibitory concentrations of 0.366, 0.305 and 0.732 mg/ml respectively. P. fluorescens, RvP, RvS, RvL and RvR were resistant to nisin Z and pediocin PA-1/AcH at concentrations 100 mg/ml but sensitive to colistin at 0.076, 0.043, 0.344, 0.344 and 0.258 mg/ml respectively. A synergistic inhibitory effect (FICI 0.5) was observed when resistant variants were treated with peptide/antibiotic combinations. No significant effect on Kþ efflux from the resistant variants in the presence of antibiotics or peptides alone or combined was observed. The proportion of C16:0 was significantly higher in antibiotic-resistant variants than in the parent strain, accounting for 32.3%, 46.49%, 43.3%, 40.1% and 44.1% of the total fatty acids in P. fluorescens, RvP, RvS, RvL and RvR respectively. Combination of antibiotics with antimicrobial peptides could allow reduced use of antibiotics in medical applications and could help slow the emergence of bacteria resistant to antibiotics. Ó 2011 Institut Pasteur. Published by Elsevier Masson SAS. All rights reserved. Keywords: Pseudomonas fluorescens; Pediocin PA-1/AcH; Nisin Z; Colistin; Antibiotics; Susceptibility

1. Introduction The evolution of resistance to antibiotics among pathogens must be viewed as a menace to public health. Understanding the epidemiology of resistance will allow development of preventative strategies to limit the spread of existing resistant pathogens and the emergence of strains with new resistance capabilities. By limiting the evolution of resistant mutants, * Corresponding author. ** Corresponding author. Laboratoire des Microorganismes et Biomole´cules Actives, Faculte´ des Sciences de Tunis, El Manar, Tunisia. E-mail addresses: [email protected] (K. Naghmouchi), [email protected] (D. Drider).

both new and existing antibacterial agents will have a prolonged lifespan (Olofsson and Cars, 2007). Pseudomonas fluorescens (P. fluorescens) is a Gramnegative psychrophilic bacterium. This highly heterogeneous species includes virulent strains and sub-clinical strains involved in opportunistic nosocomial infections (Rossignol et al., 2009). P. fluorescens is also a spoilage organism present in a variety of food-related environments. In the dairy industry, it is one of the most commonly isolated psychrotrophic bacteria and predominant in the microbial population in raw and pasteurized milk at the onset of spoilage (Dogan and Boor, 2003). Spoilage by this species is due largely to production of heat-stable extracellular lipases, proteases and lecithinases that survive thermal processing (Sillankorva et al.,

0923-2508/$ - see front matter Ó 2011 Institut Pasteur. Published by Elsevier Masson SAS. All rights reserved. doi:10.1016/j.resmic.2011.11.002

102

K. Naghmouchi et al. / Research in Microbiology 163 (2012) 101e108

2008). Some strains of Pseudomonas have been shown also to increase colonization of inert surfaces by Listeria monocytogenes and/or protect this pathogen against disinfectants (Carpentier and Chassaing, 2004). Although clinical infection by P. fluorescens is infrequent, treatment with antibiotics may be ineffective and, in rare cases, the outcome may be fatal (Scott et al., 1988). Donnarumma et al. (2010) warns that although P. fluorescens is viewed as an opportunistic pathogen, there is an urgent need for better understanding of its potential virulence. Prolonged intense antibiotic treatment of clinical infections by common pathogens has been shown to select multi-resistant strains of Pseudomonas (Brockhurst et al., 2007). Bacteriocins are antimicrobial peptides of proteinaceous nature, ribosomally synthesized and active against species related or not to the producer strain (Belguesmia et al., 2011). Nisin is the first bacteriocin to be approved for use as an inhibitor of bacteria in foods. To date, there for variants of nisin designed A, Z, Q produced by Lactococcus lactis, while variant U is produced by Streptococcus uberis. Overall, nisin has a potent activity against Gram-positive pathogens and structurally characterized by unusual amino acids such as methyllanthionine, dehydroalanine and dehydrobutyrine, produced by post-translational modifications. Pediocin-like bacteriocins are other model of peptides with interesting activities. They are small heat-stable peptides (37e53 residues) and stable over a wide range of pH and temperatures (Ennahar et al., 2000). Colistin is a polymyxin antibiotic obtained from species of Paenibacillus and used for the treatment of Gram-negative infections. After several years of clinical use, colistin was shown to be associated with significant nephrotoxicity and neurotoxicity (Lim et al., 2010) and its use then logically diminished. Recently studies pointed out the comeback of colistins for treatment of infections caused by multi-resistant organisms such as Pseudomonas aeruginosa, Acinetobacter baumannii and Klebsiella pneumoniae (Falagas and Kasiakou, 2005). In the present study, we first developed antibiotic-resistant variants of a colistin-sensitive isolate of P. fluorescens and then compared their fatty acid composition, Kþ permeability and sensitivity to antibiotics alone or combined with antimicrobial peptides such as nisin Z, pediocin PA-1/AcH and colistin. The aim of this research was to determine the synergistically effect of bacteriocins and antibiotics on sensitive and resistant variants of strains. 2. Materials and methods 2.1. Bacterial strains and culture conditions Nisin Z producer L. lactis subsp. lactis biovar diacetylactis UL719 (L. diacetylactis) (Meghrous et al., 1997) and pediocin PA-1/AcH producer Pediococcus acidilactici UL5 (P. acidilactici) (Daba et al., 1994) were obtained from the STELA dairy research center culture collection (Universite´ Laval, Quebec, PQ, Canada). Escherichia coli RR1 (E. coli) and P. fluorescens LRC-R73 (P. fluorescens) were obtained from the

Lethbridge Research Center culture collection. Listeria innocua HPB13 (L. innocua) was provided by Health Protection Branch (Health and Welfare Canada, Ottawa, ON, Canada). L. diacetylactis and P. acidilactici were grown aerobically at 30  C for 18 h in de Man, Rogosa and Sharpe broth (de Man et al., 1960) obtained from Rosell Institute (Montreal, PQ, Canada). P. fluorescens was grown aerobically for 16 h at 30  C in tryptic soy broth (TSB, Difco Laboratories, USA). All strains were reactivated from 20% glycerol stock at 80  C and sub-cultured at least three times at 24-h intervals before use in experiments. 2.2. Antibiotics The lysozyme and following antibiotics ampicillin, chloramphenicol, penicillin G, streptomycin, lincomycin, tetracycline, vancomycin, rifampicin and kanamycin were obtained from SigmaeAldrich (St. Louis, MO, USA). 2.3. Antimicrobial peptide purification Nisin Z and pediocin PA-1/AcH were purified from culture supernatants of their corresponding producer organisms using the methods described by Daoudi et al. (2001) and Gaussier et al. (2002). Colistin was purchased from SigmaeAldrich (Oakville, ON, Canada). Inhibitory activities of these antimicrobial peptides were confirmed qualitatively by the agar well diffusion method Wolf and Gibbons (1996). The indicator strains were P. acidilactici for nisin Z activity, L. innocua for pediocin PA-1/AcH and E. coli for colistin. 2.4. Development of antibiotic-resistant variants Antibiotic-resistant variants of P. fluorescens were developed by step-wise culture at 30  C in TSB with selected antibiotics at two, four, six, eight and ten times the MIC previously determined in this study (see Table 1). Resistant variants obtained were then routinely cultured in TSB containing the respective antibiotic at 10 MIC for 24 h at 30  C from 1% volume transfer. The stability of the acquired resistance was tested by at least 50 generations of exponential growth in antibiotic-free TSB. 2.5. Determination of protein concentration The amount of pure antimicrobial peptides was determined in triplicate using the Lowry method (Lowry et al., 1951). In this experiment, the bovine serum albumin (BSA) was used as internal standard. 2.6. Minimal inhibitory concentrations The sensitivity of antibiotic-resistant variants of P. fluorescens was expressed as the minimal inhibitory concentration (MIC), which was determined by the micro-dilution assay previously described by Naghmouchi et al. (2006).

K. Naghmouchi et al. / Research in Microbiology 163 (2012) 101e108

103

Table 1 Minimal inhibitory concentrations (MIC) of antibiotics and antimicrobial peptides for Pseudomonas fluorescens LRC-R73 and its antibiotic-resistant variants. Antimicrobial agent

MIC (mg/ml) P. fluorescens

Penicillin-resistant

Streptomycin-resistant

Lincomycin-resistant

Rifampicin-resistant

Antibiotics Penicillin G Ampicillin Vancomycin Lysozyme Kanamycin Tetracycline Streptomycin Lincomycin Chloramphenicol Rifampicin

50 50 50 50 0.366 0.305 4.395 50 0.732 1.22

500 3.906 0.976 0.122 0.0610 0.488 2.19e4.395 15.625 0.97 10.3

0.122 3.906 0.976 0.244 0.122 0.244 400 7.812 0.97 10.3

0.122 5.86 0.244 0.244 0.122 0.488 4.395 500 0.485e0.97 10.3

0.122 5.86 0.244 0.244 0.122 0.488 4.395 500 0.485e0.97 2.57e5.15

Antimicrobial peptides Pediocin PA-1/AcH Nisin Z Colistin

NA NA 0.076

NA NA

NA NA 0.344

NA NA 0.344

NA NA 0.258

0.043

MIC value is the median of two independent repetitions. NA: no activity at concentrations up to 100 mg/ml.

2.7. Fatty acid methyl ester analysis

2.9. Statistical analysis

Bacteria were grown at 30  C for 24 h on solid medium (30 g trypticase soy broth and 15 g of agar per liter of water), harvested, saponified, methylated, extracted and washed using the four reagents in compliance with the MIDI Labs protocol (Sasser, 1990). These were 45 g sodium hydroxide in 150 ml methanol and 150 ml distilled water (reagent 1); 325 ml 6.0 N hydrochloric acid plus 275 ml methyl alcohol (reagent 2); 200 ml hexane plus 200 ml methyl tert-butyl ether (reagent 3); 10.8 g sodium hydroxide dissolved in 900 ml distilled water (reagent 4). Gas chromatography was used to identify the fatty acid derivatives present in the prepared samples. The samples were analyzed in duplicate using the Sherlock Microbial Identification system. Bacterial acid methyl ester (BAME) standards were purchased from SigmaeAldrich (Oakville, ON, Canada) and run in each experimental set.

Each experiment was performed at least three times. Experimental values are given as means  SD. The statistical significance of the differences between the control and test values was evaluated using a one-way ANOVA t-test. The data were analyzed using the SAS version 9.2 statistical package (SAS Institute Inc., Cary, NC, USA).

2.8. Determination of potassium efflux from bacterial cells P. fluorescens and its antibiotic-resistant variants were grown at 30  C until the exponential phase in 10 ml of TSB, harvested by centrifugation at 10,000  g at 4  C for 10 min, washed twice with 50 mM phosphate saline buffer (PBS) at pH 7.0 and re-suspended in 10 ml of PBS plus 0.2% (wt/vol) glucose. Antimicrobial peptides (nisin Z, pediocin PA-1/AcH and colistin) were added alone or in combination with antibiotics (listed below). The suspensions were incubated at 30  C for 30 min and then micro-filtered. The potassium ion content of the filtrate was measured using an atomic absorption spectrophotometer (model 3300, PerkineElmer, Ueberlingen, Germany). Potassium ion efflux was expressed as the measured concentration in mM.

3. Results 3.1. Antibiotic and antimicrobial sensitivity The sensitivity of P. fluorescens to antimicrobial peptides and antibiotics is presented in Table 1. As shown this bacterium is resistant to nisin Z and pediocin PA-1/AcH at concentrations up to 100 mg/mL, but remains highly sensitive to colistin, with a MIC of 0.076 mg/ml. The bacterium (P. fluorescens) appeared highly sensitive to kanamycin, tetracycline and chloramphenicol (with MIC of about 0.366, 0.305 and 0.732 mg/ml, respectively), moderately sensitive to streptomycin, rifampicin and lysozyme (1.2, 4.4, 50 mg/ml, respectively) and least sensitive to penicillin G, ampicillin, vancomycin and lincomycin (MIC up to 50 mg/ml). A synergistic effect on P. fluorescens was apparent when nisin Z or pediocin PA-1/AcH was used in combination with antibiotics (Table 2), except for the case of tetracycline/pediocin (additive effect FICI: 0.75) and ampicillin/nisin Z (indifferent effect FICI: 1.3). When colistin was combined with tetracycline, vancomycin or lysozyme, an additive effect appeared, with fractional inhibitory concentration indexes (FICI) of respectively 0.8, 0.91 and 0.70. Combinations of antimicrobial peptides with penicillin G, streptomycin; lincomycin appeared to be highly synergistic. Penicillin G, streptomycin and lincomycin were used to

104

K. Naghmouchi et al. / Research in Microbiology 163 (2012) 101e108

Table 2 Minimal inhibitory concentrations obtained with antibiotic, antimicrobial peptides and their combinations against Pseudomonas fluorescens LRC-R73 (A), variant penicillin-resistant (B), variant streptomycin-resistant (C), variant lincomycin-resistant (D) and variant rifampicin-resistant (E). Antibiotic

MIC of antibiotic/antimicrobial peptide combination (mg/mL) Ab/Nis Z

FICI

Ab/Ped

FICI

Ab/col

FICI

A/P. fluorescens LRC-R73 Penicillin G Ampicillin Vancomycin Lysozyme Kanamycin Tetracycline Streptomycin Lincomycin Chloramphenicol Rifampicin

0.024/0.156 64.06/2.5 0.39/0.625 0.78/1.25 0.09/16 0.045/8 0.031/0.5 0.048/0.625 0.011/0.78 0.076/4

0.01 1.3 0.01 0.02 0.25 0.2 0.01 0.01 0.01 0.07

0.048/0.625 16.01/5 0.39/5 0.78/1.3 0.073/1.25 0.225/5 0.019/0.312 0.78/1.3 0.015/0.125 0.076/5

0.01 0.4 0.01 0.02 0.2 0.75 0.01 0.02 0.03 0.07

0.048/0.006 64/0.005 6.25/0.023 0.78/0.0028 0.0017/0.0008 0.09/0.046 0.0312/0.0014 0.78/0.0028 0.09/0.023 0.076/0.0115

0.08 1.4 0.8 0.7 0.01 0.91 0.03 0.06 0.45 0.25

B/Penicillin-resistant variant Penicillin G Ampicillin Vancomycin Lysozyme Kanamycin Tetracycline Streptomycin Lincomycin Chloramphenicol Rifampicin

15.62/1.56 0.49/3.12 0.06/3.12 0.03/12.5 0.03/6.25 0.06/12.5 0.27/25 0.48/6.25 0.06/6.25 0.15/12.5

31.24/12.5 0.49/3.12 0.25/6.25 0.06/4.68 0.03/3.12 0.03/6.25 1.1/6.25 0.24/12.5 0.12/1.56 0.03/6.25

0.062 0.13 0.26 0.24 0.5 0.06 0.23 0.02 0.13 0.01

15.62/0.01 0.24/0.01 0.13/0.005 0.01/0.005 0.015/0.01 0.03/0.01 0.13/0.02 0.12/0.005 0.03/0.005 0.15/0.002

0.26 0.29 0.24 0.18 0.39 0.29 0.23 0.13 0.14 0.06

0.031 0.13 0.06 0.24 0.5 0.12 0.06 0.03 0.06 0.01

C/Streptomycin-resistant variant Penicillin G 0.015/0.66 Ampicillin 0.03/4.67 Vancomycin 0.01/1.32 Lysozyme 0.06/1.32 Kanamycin 0.045/12.5 Tetracycline 0.122/6.25 Streptomycin 25/6.25 Lincomycin 0.97/9.37 Chloramphenicol 0.056/0.66 Rifampicin 0.32/0.16

0.12 0.01 0.03 0.25 0.37 0.5 0.03 0.12 0.05 0.03

0.008/6.25 0.97/1.56 0.01/12.5 0.12/0.78 0.015/3.12 0.09/12.5 100/25 0.72/6.25 0.083/6.25 0.48/12.5

0.03 0.24 0.03 0.5 0.13 0.37 0.25 0.09 0.09 0.05

0.03/0.043 0.12/0.011 0.006/0.011 0.03/0.086 0.015/0.043 0.03/0.043 100/0.086 0.13/0.011 0.11/0.032 0.24/0.011

0.36 0.05 0.04 0.37 0.25 0.24 0.5 0.05 0.2 0.06

D/Lincomycin-resistant variant Penicillin G 0.03/25 Ampicillin 1.46/3.12 Vancomycin 0.12/6.25 Lysozyme 0.12/12.5 Kanamycin 0.061/6.25 Tetracycline 0.06/25 Streptomycin 0.27/1.56 Lincomycin 125/6.25 Chloramphenicol 0.09/12.5 Rifampicin 2.57/25

0.24 0.25 0.49 0.49 0.25 0.12 0.06 0.25 0.12 0.25

0.06/6.25 2.19/12.5 0.031/1.56 0.09/1.2 0.03/3.12 0.22/12.5 0.81/6.25 62.5/25 0.36/3.12 0.64/6.25

0.5 0.37 0.12 0.36 0.13 0.5 0.18 0.12 0.5 0.06

0.015/0.021 0.183/0.021 0.031/0.032 0.061/0.016 0.03/0.021 0.011/0.032 0.06/0.043 125/0.086 0.04/0.021 0.25/0.043

0.13 0.09 0.22 0.29 0.07 0.11 0.14 0.06 0.11 0.15

E/Rifampicin-resistant variant Penicillin G 0.06/25 Ampicillin 1.46/6.25 Vancomycin 0.06/6.25 Lysozyme 0.12/12.5 Kanamycin 0.03/3.12 Tetracycline 0.015/25 Streptomycin 0.25/1.56 Lincomycin 62.5/6.25 Chloramphenicol 0.36/50 Rifampicin 0.956/12.5

0.5 0.24 0.24 0.5 0.12 0.03 0.05 0.12 0.5 0.18

0.06/18.75 2.2/12.5 0.03/3.12 0.06/25 0.06/1.56 0.02/12.5 1.01/3.12 23.5/3.12 0.14/12.5 0.482/25

0.5 0.38 0.13 0.24 0.49 0.04 0.22 0.04 0.19 0.09

0.015/0.07 0.73/0.032 0.016/0.024 0.03/0.064 0.015/0.008 0.015/0.012 0.06/0.016 7.8/0.024 0.09/0.032 0.956/0.06

0.39 0.25 0.16 0.3 0.16 0.08 0.08 0.06 0.25 0.41

Nis Z e nisin Z; Ped e pediocin PA-1/AcH; col e colistin Minimal inhibitory concentration (MIC) is expressed in mg/ml, median of two independent repetitions. FICI (fractional inhibitory concentration index): FICIA þ FICIB ¼ (MIC of drug A in combination/MIC of drug A alone) þ (MIC of drug B in combination/MIC of drug B alone). The FICI index was interpreted as follows: synergy (FICI 0.5), additive (0.5 < FICI  1), indifference (1 < FICI  2) and antagonism (FICI index > 2).

K. Naghmouchi et al. / Research in Microbiology 163 (2012) 101e108

develop resistant variants of P. fluorescens. The respective variants thereof obtained were named RvP (Penicillin), RvS (Streptomycin) and RvL (Lincomycin). A variant resistant to rifampicin, RvR, was also obtained.

105

compared to 26.9%, 2.6%, 6.5% and 28.6% respectively for RvP, RvS, RvL and RvR. 3.4. Kþ efflux from bacterial cells in the presence of antimicrobial agents

3.2. Antibacterial sensitivity of the resistant variants Resistant variants RvP, RvS, RvL and RvR developed from P. fluorescens retained the resistant phenotype after at least 50 generations (eight transfers at 1% volume) of culture in TSB without their respective antibiotics. The MIC of the various antibiotics and antimicrobial peptides are shown in Table 1. Nisin Z and pediocin PA-1/AcH showed no inhibitory activity against any of the variants or against the parental P. fluorescens strain. Ongoing research has revealed very strong synergistic effects (FICI 0.5) of combinations of antimicrobial peptides and antibiotics on the resistant variants (Table 2). 3.3. Fatty acid composition The fatty acid composition of P. fluorescens and its antibiotic-resistant variants is shown in Fig. 1. Individual fatty acid contents are expressed as percentages of total fatty acids. The predominant fatty acid found in all organisms was palmitic acid (C16:0), a saturated fatty acid. The proportion of C16:0 was significantly higher in antibiotic-resistant variants than in the parent strain, accounting for 32.3%, 46.49%, 43.3%, 40.1% and 44.1% of the total fatty acids respectively in P. fluorescens, RvP, RvS, RvL and RvR. RvP and RvR showed conspicuous decreases in heptadecanoic acid and tetradecanoic acid, but an increase in octadecenoic acid, and in the case of RvR, in pentadecanoic acid as well. Octadecenoic acid accounted for 8.96% of the total fatty acid in P. fluorescens,

After 30 min in the presence of 0.076 mg/ml of colistin, the efflux of extracellular potassium ion from the P. fluorescens parental strain reached about 0.45 mM, compared to 4.43  0.32 mM in the absence of the peptide. The effect of nisin Z or pediocin PA-1/AcH (>100 mg/mL) was quite similar, as the quantity extracellular potassium concentration was of about 4.46 and 4.39 mM respectively. No significant deviations from these values were observed for the resistant variants in the presence of their corresponding antibiotic alone or combined with antimicrobial peptide (Table 3). 4. Discussion The increasing emergence of antibiotic-resistant bacteria is a leading public health concern. Although increased awareness has spurred research, the development of new antibiotics is becoming increasingly expensive and difficult. The use of combinations of antibiotics with different mechanisms (i.e. targets sites) of action forces bacteria to undergo multiple mutations in order to become resistant, thus slowing the emergence of resistant strains considerably (Olofsson and Cars, 2007). In this study, P. fluorescens, a Gram-negative bacterium, was selected for the study of sensitivity to antimicrobial peptides, antibiotics and combinations thereof. Using selective media each containing single antibiotic, resistant variants of strain P. fluorescens were developed, thus possibly representing a diverse range of mechanisms of resistance. These

Fig. 1. Fatty acid composition of P. fluorescens LRC-R73 parent strain ( ); penicillin-resistant ( ); streptomycin-resistant ( ); lincomycin-resistant ( ); rifampicin-resistant ( ). Values are means of duplicate analyses.

106

K. Naghmouchi et al. / Research in Microbiology 163 (2012) 101e108

Table 3 Potassium efflux from washed cells of Pseudomonas fluorescens LRC-R73 antibiotic-resistant variants suspended in PBS containing antimicrobial agents. Penicillin-resistant

[Kþ] (mM)*        

e

No agent Penicillin (>500) Pediocin (100) Nisin Z (100) Colistin (0.043) Penicillin (31.24)/Pediocin (12.5) Penicillin (15.62)/Nisin (1.56) Penicillin (15.62)/Colistin (0.01)

4.10 3.98 4.44 4.38 3.84 4.29 4.36 4.12

Lincomycin-resistant

[Kþ] (mM)*

No agent Lincomycin (>500) Pediocin (100) Nisin Z (100) Colistin (0.344) Lincomycin (62.5)/Pediocin (25) Lincomycin (125)/Nisin Z (6.25) Lincomycin (125)/Colistin (0.086)

3.81 4.12 4.08 3.94 3.98 4.81 4.77 4.04

       

0.01 0.01f 0.01a 0.01b 0.01g 0.02d 0.01c 0.01e h

0.02 0.01c 0.02d 0.01g 0.01f 0.01a 0.01b 0.04e

Streptomycin-resistant

[Kþ] (mM)*

No agent Streptomycin (400) Pediocin (100) Nisin Z (100) Colistin (0.344) Streptomycin (100)/Pediocin (25) Streptomycin (25)/Nisin (6.25) Streptomycin (100)/Colistin (0.086)

3.95 4.02 4.05 3.99 4.01 4.05 4.08 4.03

Rifampicin-resistant

[Kþ] (mM)*

No agent Rifampicin (3.86) Pediocin (100) Nisin Z (100) Colistin (0.26) Rifampicin (0.482)/Pediocin (25) Rifampicin (0.956)/Nisin Z (12.5) Rifampicin (0.956)/Colistin (0.06)

4.39 4.23 4.05 4.61 4.27 4.31 4.36 4.12

               

0.01g 0.01de 0.01b 0.00f 0.01e 0.01bc 0.01a 0.01cd 0.01b 0.02f 0.01h 0.02a 0.02e 0.02d 0.01c 0.01g

Means with the same letters are not significantly different (P < 0.05). Concentration of antimicrobial agents in mg/ml. * Concentration measured after 30 min; means of triplicate tests.

mechanisms may be presumed to address the mechanism of action of the antibiotic, penicillin G inhibiting cell wall synthesis, streptomycin inhibiting protein mRNA translation by binding to the 30S ribosomal subunit, lincomycin doing likewise on the 50S subunit and rifampicin binding to the betasubunit of DNA-dependent RNA polymerase to inhibit transcription. Cell fatty acid composition and Kþ efflux were examined to provide insights on the resistance mechanisms developed. An increase in saturated fatty acids, especially C:16 and C:18, should increase the rigidity of the bacterial cell membranes, thereby impeding penetration of antibiotics and/ or antimicrobial peptides (Naghmouchi et al., 2007). Our results show significant high proportions of C16:0 in antibiotic-resistant variants than in the parent strain, accounting for 46.49%, 43.3%, 40.1% and 44.1% of the total fatty acids respectively in RvP, RvS, RvL and RvR, versus 32.32% in P. fluorescens. RvP, RvL and RvR showed slightly increased resistance to tetracycline (MIC increased 0.183 mg/ mL), while RvP and RvS showed similar increases in resistance to chloramphenicol (MIC increased 0.238 mg/mL). The significance of these small increases is uncertain, but they may be associated with the increased saturated fatty acid content of the cells. The four resistant strains were more sensitive than the parent strain to all the other antibiotics, suggesting that development of resistance to one antibiotic can decrease resistance to another type. For a long time, studies focused on class IIa bacteriocins revealed the absence of activity against Gram-negative bacteria (Drider et al., 2006). It was expected that pediocin AcH would have no inhibitory activity against P. fluorescens and its variants. Notwithstanding that recent studies have pointed out activities of class IIa bacteriocins against Gramnegative bacterium Campylobacter jejuni (Stern et al., 2005, 2006; Messaoudi et al., 2011). Furthermore, Naghmouchi

et al. (2010) reported that although nisin Z (a class I bacteriocin) has a broad spectrum of activity against Gram-positive bacteria, it is ineffective against Gram-negative bacteria when used alone. The results obtained in the present study are in agreement with the previous literature, in the sense that neither pediocin AcH nor nisin Z was effective against P. fluorescens (even at concentrations > 100 mg/mL) or its variants when used alone. However, we note with interest the synergistic effects of combining pediocin AcH with antibiotics, tetracycline excepted (additive effect, FICI ¼ 0.75) and nisin Z with antibiotics, ampicillin excepted (indifferent effect, FICI ¼ 1.3). Helander and Mattila-Sandholm (2000) showed that nisin could in fact stabilize the outer membrane of Gramnegative bacteria, resulting in less damage to the permeability barrier. The indifferent effect observed for the nisin Z/ampicillin combination may be due to this phenomenon. Falagas et al. (2008) reported that one promising solution to the increase in antibiotic resistance might be the use of old antibiotic compounds such as colistin, a cationic detergent-like cyclic lipopeptide. Produced by a Gram-positive organism and active against many Gram-negative bacteria, colistin is currently observed as a last resort for treating infections by resistant Pseudomonas. Our results showed that P. fluorescens and variants RvP, RvS, RvL and RvR are highly sensitive to colistin, with MIC values of 0.076, 0.043, 0.344, 0.344 and 0.26 mg/ml respectively. Combinations of colistin with antibiotics showed synergistic effects, except in the case of vancomycin, lysozyme and streptomycin (additive effect, FICI ¼ 0.8, 0.7 and 0.91 respectively) as well as ampicillin (indifferent effect, FICI ¼ 1.4). In this study, no significant effect on Kþ efflux from the resistant variants in the presence of antibiotics or peptides alone or combined is observed. Colistin is known to bind to the anionic bacterial outer membrane with high affinity for the lipid moiety of the lipopolysaccharide and can preferentially

K. Naghmouchi et al. / Research in Microbiology 163 (2012) 101e108

displace Mg2þ and Ca2þ from cationic binding sites (Groisman et al., 1997). Friedrich et al. (2000) reported that many antimicrobial peptides kill microorganisms by causing membrane permeabilization, although not necessarily as their sole mode of action. In P. aeruginosa cells treated with polymyxyin, the secondary leakage of UV-absorbing materials and ribose containing molecules is a consequence of an enzymic degradation of RNA (Newton, 1956). Pseudomonas sp. resistance to the polymyxins is postulated to be induced by a gene ( pmrA) that, in the presence of low magnesium concentrations, modifies the lipopolysaccharide, resulting in reduced binding affinity of colistin and perhaps related antimicrobial peptides to the outer membrane (Evans et al., 1999; Groisman et al., 1997). Lambert (2002) reported that alginate, an associated layer of the anionic polysaccharide, can bind cationic antibiotics such as the aminoglycosides (gentamicin, amikacin) and restrict their diffusion (Nichols et al., 1988). P. fluorescens was found highly sensitive to kanamycin, tetracycline and chloramphenicol (MIC ¼ 0.366, 0.305 and 0.732 mg/ml respectively). A synergistic effect against P. fluorescens was produced with 90% (27/30) of the combinations of the class I or class IIa bacteriocins with antibiotics and 60% (6/10) of the combinations of colistin with antibiotic. Strong synergistic effects (FICI < 0.5) were noted against all four variants with at least one antimicrobial peptide and antibiotic combination. These results suggest that lower concentrations of antibiotics could be used to control (bacteriostatic effect) or eliminate (bactericidal effect) Pseudomonas if combined with antimicrobial peptides. The use of lower concentrations of antibiotic should decrease the emergence of bacterial populations resistant to these agents as well as reduce the cost associated with their use. Further studies should be conducted to investigate the possibility of obtaining synergistic effects by combining nisin Z (class I) or pediocin PA-1/AcH (class IIa) bacteriocins with antibiotics in the treatment of nosocomial infections due to Gram-negative and Gram-positive bacteria. Acknowledgments This research was supported by The National Science and Engineering Research Council of Canada, Agriculture and Agri-Food Canada and BEST, Biotechnology, Edmonton, Alberta Canada. References Belguesmia, Y., Naghmouchi, K., Chihib, N.E., Drider, D., September 2011. Class IIa bacteriocins: current knowledge and perspectives. In: Drider, D., Rebuffat, S. (Eds.), Prokaryotic Antimicrobial Peptides: From Genes to Applications. Editions Springer, NY-USA. doi:10 1007/978-1-4419-76925 10. Brockhurst, M.A., Morgan, A.D., Fenton, A., Buckling, A., 2007. Experimental coevolution with bacteria and phage: the Pseudomonas fluorescens model system. Infect. Genet. Evol. 7, 547e552. Carpentier, B., Chassaing, D., 2004. Interactions in biofilms between Listeria monocytogenes and resident microorganisms from food industry premises. Int. J. Food Microbiol. 97, 111e122.

107

Daba, H., Lacroix, C., Huang, J., Simard, R.E., Lemieux, L., 1994. Simple method of purification and sequencing of a bacteriocin produced by Pediococcus acidilactici UL5. J. Appl. Bacteriol. 77, 682e688. Daoudi, L., Turcotte, C., Lacroix, C., Fliss, I., 2001. Production and characterization of anti-nisin Z monoclonal antibodies: suitability for distinguishing active from inactive forms through a competitive enzyme immunoassay. Appl. Microbiol. Biotechnol. 56, 114e119. de Man, J.C., Rogosa, M., Sharpe, M.E., 1960. A medium for the cultivation of lactobacilli. 473. J. Appl. Bacteriol. 23, 130e135. Dogan, B., Boor, K.J., 2003. Genetic diversity and spoilage potentials among Pseudomonas spp. isolated from fluid milk products and dairy processing plants. Appl. Environ. Microbiol. 69, 130e138. Donnarumma, G., Buommino, E., Fusco, A., Paoletti, I., Auricchio, L., Tufano, M.A., 2010. Effect of temperature on the shift of Pseudomonas fluorescens from an environmental microorganism to a potential human pathogen. Int. J. Immunopathol. Pharmacol. 23, 227e234. Drider, D., Fimland, G., He´chard, Y., McMullen, L.M., Pre´vost, H., 2006. The continuing story of class IIa bacteriocins. Microbiol. Mol. Biol. Rev. 70, 564e582. Ennahar, S., Sashihara, T., Sonomoto, K., Ishizaki, A., 2000. Class Ha bacteriocins: biosynthesis, structure and activity. FEMS Microbiol. 24, 85e106. Evans, M.E., Feola, D.J., Rapp, R.P., 1999. Polymyxin B sulfate and colistin: old antibiotics for emerging multiresistant gram-negative bacteria. Ann. Pharmacother. 33, 960e967. Falagas, M.E., Kanellopoulou, M.D., Karageorgopoulos, D.E., Dimopoulos, G., Rafailidis, P.I., Skarmoutsou, N.D., Papafrangas, E.A., 2008. Antimicrobial susceptibility of multidrug-resistant Gram negative bacteria to fosfomycin. Eur. J. Clin. Microbiol. Infect. Dis. 27, 439e443. Falagas, M.E., Kasiakou, S.K., 2005. Colistin: the revival of polymyxins for the management of multidrug-resistant gram-negative bacterial infections. J. Clin. Microbiol. Infect. Dis. 40, 1333e1341. Friedrich, C.L., Moyles, D., Beveridge, T.J., Hancock, R.E., 2000. Antibacterial action of structurally diverse cationic peptides on gram-positive bacteria. Antimicrobial. Agents Chemother. 44, 2086e2092. Gaussier, H., Morency, H., Lavoie, M., Subirade, M., 2002. Remplacement of trifluoroacetic acid with HCl in the hydrophobic purification steps of pediocin PA-1: a structural effect. Appl. Environ. Microbiol. 68, 4803e4808. Groisman, E.A., Kayser, J., Soncini, F.C., 1997. Regulation of polymyxin resistance and adaptation to low-Mg2þ environments. J. Bacteriol. 179, 7040e7045. Helander, I.M., Mattila-Sandholm, T., 2000. Fluorometric assessment of Gram-negative bacterial permeabilization. J. Appl. Microbiol. 88, 213e219. Lambert, P.A., 2002. Mechanisms of antibiotic resistance in Pseudomonas aeruginosa. J. R. Soc. Med. 95, 22e26. Lim, L.M., Ly, N., Anderson, D., Yang, J.C., Macander, L., Jarkowski, A., Forrest, A., Bulitta, J.B., Tsuji, B.T., 2010. Resurgence of colistin: a review of resistance, toxicity, pharmacodynamics, and dosing. Pharmacotherapy 30, 1279e1291. Lowry, O.H., Rosebrough, N.J., Farr, A.L., Randall, R.J., 1951. J. Biol. Chem. 193, 265e275. Meghrous, J., Lacroix, C., Bouksaim, M., LaPointe, G., Simard, R.E., 1997. Genetic and biochemical characterization of nisin Z produced by Lactococcus lactis ssp. lactis biovar. diacetylactis UL719. J. Appl. Microbiol. 83, 133e138. Messaoudi, S., Kergourlay, G., Rossero, A., Ferchichi, M., Pre´vost, H., Drider, D., Manai, M., Dousset, X., 2011. Identification of lactobacilli residing in chicken caeca with antagonism against Campylobacter. Int. Microbiol. 14, 96e103. Naghmouchi, K., Drider, D., Baah, J., Teather, R., 2010. Nisin A and polymyxin B as synergistic inhibitors of Gram-positive and Gram-negative bacteria. Prob. Ant. Prot. 2, 98e103. Naghmouchi, K., Drider, D., Kheadr, E., Lacroix, C., Pre´vost, H., Fliss, I., 2006. Multiple characterizations of Listeria monocytogenes sensitive and insensitive variants to divergicin M35, a new pediocin-like bacteriocin. J. Appl. Microbiol. 100, 29e39.

108

K. Naghmouchi et al. / Research in Microbiology 163 (2012) 101e108

Naghmouchi, K., Kheadr, E., Lacroix, C., Fliss, I., 2007. Class I and class II a bacteriocins cross re´sistance phenomenon of Listeria monocytogenes LSD530. Food Microbiol. 4, 718e727. Newton, B.A., 1956. The proprieties and mode action of the polymyxin. Bacteriol. Rev. 20, 14e27. Nichols, W.W., Dorrington, S.M., Slack, M.P., Walmsley, H.L., 1988. Inhibition of tobramycin diffusion by binding to alginate. Antimicrob. Agents Chemother. 32, 518e523. Olofsson, K., Cars, Otto, 2007. Optimizing drug exposure to minimize selection of antibiotic resistance. Clin. Infect. Dis 45, 129e136. Rossignol, G., Sperandio, D., Guerillon, J., Duclairoir Poc, C., SoumSoutera, E., Orange, N., Feuilloley, M.G.J., Merieau, A., 2009. Phenotypic variation in the Pseudomonas fluorescens clinical strain MFN1032. Res. Microbiol. 160, 337e344. Sasser, M., 1990. Identification of Bacteria by Gas Chromatography of Cellular Fatty Acids. MIDI Inc, Newark, DE. MIDI Technical Note 101.

Scott, J., Boulton, F.E., Govan, J.R., Miles, R.S., McClelland, D.B., Prowse, C.V., 1988. A fatal transfusion reaction associated with blood contaminated with Pseudomonas fluorescens. Vox. Sang. 54, 201e204. Sillankorva, S., Neubauer, P., Azeredo, J., 2008. Pseudomonas fluorescens biofilms subjected to phage phiIBB-PF7A. BMC Biotechnol. 8, 79. Stern, N.J., Svetoch, E.A., Eruslanov, B.V., Kovalev, Y.N., Volodina, L.I., Perelygin, V.V., Mitsevich, E.V., Mitsevich, I.P., Levchuk, V.P., 2005. Paenibacillus polymyxa purified bacteriocin to control Campylobacter jejuni in chickens. J. Food Prot. 68, 1450e1453. Stern, N.J., Svetoch, E.A., Eruslanov, B.V., Perelygin, V.V., Mitsevich, E.V., Mitsevich, I.P., Pokhilenko, V.D., Levchuk, V.P., Svetoch, O.E., Seal, B.S., 2006. Isolation of a Lactobacillus salivarius strain and purification of its bacteriocin which is inhibitory to Campylobacter jejuni in the chicken gastrointestinal system. Antimicrobial. Agents. Chemother. 50, 3111e3116. Wolf, C.E., Gibbons, W.R., 1996. Improved method for quantification of the bacteriocin nisin. J. Appl. Bacteriol. 80, 453e457.