Zinc as an agent for the prevention of biofilm formation by pathogenic bacteria

Authors


Correspondence

Mario Jacques, Faculté de médecine vétérinaire, Université de Montréal, 3200 rue Sicotte, St-Hyacinthe, J2S 7C6 QC, Canada. E-mail: mario.jacques@umontreal.ca

Abstract

Aims

Biofilm formation is important for the persistence of bacteria in hostile environments. Bacteria in a biofilm are usually more resistant to antibiotics and disinfectants than planktonic bacteria. Our laboratory previously reported that low concentrations of zinc inhibit biofilm formation of Actinobacillus pleuropneumoniae. The aim of this study is to evaluate the effect of zinc on growth and biofilm formation of other bacterial swine pathogens.

Methods and Results

To determine the effect of zinc on biofilm formation, biofilms were grown with or without zinc in 96-well plates and stained with crystal violet. At micromolar concentrations (0–250 μ mol l−1), zinc weakly inhibited bacterial growth and it effectively blocked biofilm formation by A. pleuropneumoniae, Salmonella Typhymurium and Haemophilus parasuis in a dose-dependent manner. Additionally, biofilm formation of Escherichia coli, Staphylococcus aureus and Streptococcus suis was slightly inhibited by zinc. However, zinc did not disperse preformed biofilms. To determine whether zinc inhibits biofilm formation when poly-N-acetylglucosamine (PGA) is present, PGA was detected with the lectin wheat germ agglutinin. Only A. pleuropneumoniae and Staph. aureus biofilms were found to contain PGA.

Conclusion

Zinc used at nonbactericidal concentrations can inhibit biofilm formation by several Gram-negative and Gram-positive bacterial swine pathogens.

Significance and Impact of Study

The antibiofilm activity of zinc could provide a tool to fight biofilms, and the nonspecific inhibitory effect may well extend to other important human and animal bacterial pathogens.

Background

Biofilms are a structured community of micro-organisms enclosed in a self-produced extracellular polymer matrix adhered to biological or nonbiological surfaces (Costerton et al. 1999). In nature, bacteria predominantly exist in a sessile form (biofilm) rather than a free-swimming form (planktonic) (O'Toole et al. 2000; Stoodley et al. 2002). The biofilm matrix is responsible for adhesion to surfaces and for cohesion in the biofilm and may contain polysaccharides, proteins and extracellular DNA (eDNA). The composition of the matrix varies greatly between different micro-organisms (Flemming and Wingender 2010). Biofilm formation by bacterial pathogens is important for the transmission of infections and persistence of bacteria in hostile environments (Hall-Stoodley et al. 2004; Lewis 2010). Furthermore, bacteria grown as a biofilm are protected against a variety of environmental stresses such as antibiotics, disinfectants and host defences (Hall-Stoodley and Stoodley 2009; Hoiby et al. 2010; Bridier et al. 2011).

The negative impact of biofilm formation by pathogens of medical and veterinary importance on the efficacy of antibiotics and disinfectants is a major problem in animal and human health (Parsek and Singh 2003; Clutterbuck et al. 2007; Jacques et al. 2010). Due to the general properties of biofilms, the prevention, diagnosis and treatments of diseases associated with biofilms require novel approaches. The discovery and development of agents with the ability to limit biofilm formation or eradicate established biofilms would have the potential to enhance the efficacy of biocides that are relatively ineffective against biofilm bacteria (for recent reviews see Rendueles et al. 2013; Worthington et al. 2012). Furthermore, there is a growing interest in the discovery of nonbiocidal antibiofilm molecules because the selective pressure on bacteria to develop resistance to nonbiocidal agents should be significantly reduced (Worthington et al. 2012).

Our laboratory recently reported that low concentrations of zinc inhibit biofilm formation by Actinobacillus pleuropneumoniae (Labrie et al. 2010). A. pleuropneumoniae is the Gram-negative bacterium responsible for porcine pleuropneumonia, a respiratory disease of swine (Chiers et al. 2010). Biofilm formation by A. pleuropneumoniae on polystyrene microtitre plates is dependent on the production of PGA, a polymer of β-1, 6-N-acetyl-D-glucosamine (Kaplan et al. 2004). Biofilm formation has also been demonstrated in other swine pathogen including Bordetella bronchiseptica, Escherichia coli, Haemophilus parasuis, Salmonella Typhymurium, Staphylococcus aureus and Streptococcus suis (Jacques et al. 2010). The observation we made concerning the antibiofilm activity of zinc is of interest because zinc supplementation has been associated with the reduction in diarrhoeal and respiratory diseases in humans and in animals (Aggarwal et al. 2007; Crane et al. 2011; Molist et al. 2011) and is frequently added to piglet feed (Molist et al. 2011; Shelton et al. 2011).

Thus, the aim of the present study was to evaluate the effect of zinc on bacterial growth and biofilm formation of other important swine pathogens.

Materials and methods

Bacterial strains

All strains used in this study are listed in Table 1. A. pleuropneumoniae and H. parasuis were grown on Brain Heart Infusion (BHI; Oxoid Ltd, Basingstoke, Hampshire, UK) agar supplemented with 15 μg/mL nicotinamide adenine dinucleotide (NAD). Bord. bronchiseptica, Salmonella, Staph. aureus, Strep. suis were grown on BHI agar. E. coli strains were grown on Luria-Bertani (LB; Becton, Dickinson and Company, Sparks, MD, USA) agar. Plates were incubated overnight at 37°C with 5% CO2.

Table 1. Bacterial strains used in this study
Bacterial speciesStrainsRelevant traitsZinc concentration required to inhibit growth of planktonic cellsSource
ZnCl2 (μ mol l−1)ZnO (μ mol l−1)
  1. a

    Faculté de médecine vétérinaire, Université de Montréal, St-Hyacinthe, QC, Canada.

  2. b

    Ontario Veterinary College, University of Guelph, Guelph, ON, Canada.

  3. c

    Institute for Research on Animal Disease, Compton, UK.

Actinobacillus pleuropneumoniae S4074Serotype 1; reference strain10001000K.R. Mittala
719Serotype 1750750D. Slavicb
L20Serotype 5b; reference strain10001000K.R. Mittal
Bordetella bronchiseptica 276 >1000>1000J.M. Rutterc
Escherichia coli ECL 17608STb: AIDA: EAST1>1000>1000J. M. Fairbrothera
ECL 17659F18: AIDA>1000>1000J. M. Fairbrother
ECL 17635Eae: Paa>1000>1000J. M. Fairbrother
Haemophilus parasuis NagasakiSerotype 5; reference strain>1000>1000M. Gottschalka
Salmonella Typhimurium

ATCC 14028

STF07-8567-3

 

>1000

>1000

>1000

>1000

A. Letelliera

A. Letellier

Salmonella HeidelbergSTF08-453 >1000>1000A. Letellier
Staphylococcus aureus 154NMethicillin-resistant nasal isolate>1000>1000M. Archambaulta
294Methicillin-resistant skin isolate>1000>1000M. Archambault
327NMethicillin-resistant nasal isolate>1000>1000M. Archambault
Streptococcus suis 735Serotype 2; reference strain250250M. Gottschalk

Biofilm assay

Growth conditions to obtain mature biofilms for the assay are summarized in Table 2. Briefly, overnight cultures of A. pleuropneumoniae, Bord. bronchiseptica, E. coli, Salmonella, Staph. aureus or Strep. suis were diluted 1/100 in their corresponding broth and a volume (100 μl) was aliquoted in triplicate in a flat-bottom 96-well polystyrene plate. For H. parasuis, colonies from overnight agar cultures were resuspended in BHI with 5 μg/mL NAD, and the suspension was aliquoted (100 μl) in triplicate in a flat-bottom 96-well polystyrene plate. With the exception of E. coli, the microtitre plate used was a Costar® 3599 96-well plate (Corning, NY, USA). For E. coli, Costar® 3370 96-well plates were used. Wells containing sterile broth were used as negative control.

Table 2. Growth conditions used for each bacterial species to obtain mature biofilms
Bacterial SpeciesGrowth ConditionsCorning Plate IDIncubation temperature and atmosphereIncubation time (h)Reference
A. pleuropneumoniae O/N in 5 ml of BHI with NAD (5 μg ml−1) at 37°C with shaking (200 rpm); dilution 1/100 in BHI with NAD (5 μg ml−1)359937°C, 5% CO25Labrie et al. (2010)
Bord. bronchiseptica O/N in 5 ml of BHI at 37°C with shaking (200 rpm); dilution 1/100 in BHI359937°C, 5% CO224This study
E. coli O/N in 5 mL of M9 minimal medium at 30°C with shaking (200 rpm); dilution 1/100 in M9 minimal medium337030°C24Charbonneau et al. (2006)
H. parasuis Resuspend colonies from a BHI with NAD (15 μg ml−1) agar plate in 3 ml of BHI with NAD (15 μg ml−1)359937°C, 5% CO248This study
Salmonella O/N in 5 ml of colonization factor antigen (CFA) medium at 37°C with shaking (200 rpm); dilution 1/100 in CFA359930°C48Suzuki et al. (2002)
Staph. aureus O/N in 5 ml of BHI with glucose [0·25% (w/v)] at 37°C with shaking (200 rpm); dilution 1/100 in BHI with glucose [0·25% (w/v)]359937°C, 5% CO224This study
Strep. suis O/N in 5 ml of basal broth medium (BBM) with fibrinogen (5 mg ml−1) at 37°C with shaking (200 rpm); dilution 1/100 in BBM with fibrinogen (5 mg ml−1)359937°C, 5% CO224Bonifait et al. (2008)

Following incubation (Table 2), biofilms were treated as described by Labrie et al. (2010) with some modifications. Briefly, the liquid medium was removed using a vacuum, and unattached cells were removed by immersing the plate once in Milli-Q water. The water was removed with a vacuum and excess water was removed by inverting plates onto a paper towel. Biofilms were then stained with 0·1% (w/v) crystal violet for 2 min. Biofilms were washed once with distilled water and then dried at 37°C for 15 min. The stain was then released with 100 μl of 70% (v/v) ethanol, and the amount of released stain was quantified by measuring the absorbance at 590 nm with a microplate reader (Powerwave; BioTek Instruments, Winooski, VT, USA).

Confocal laser scanning microscopy (CLSM)

Biofilms were prepared as described above. After the desired incubation time (Table 2), biofilms were stained with FilmTracerTM FM® 1-43 fluorescent marker (Molecular Probes; Eugene, OR, USA) according to manufacturer's instructions. To determine the composition of the biofilm matrix, biofilms were stained with wheat germ agglutinin (WGA-Oregon Green 488; Molecular Probes), FilmTracerTM SYPRO® Ruby biofilm matrix stain (Molecular Probes) or BOBOTM-3 iodide (Molecular Probes) according to manufacturer's instructions. After a 30 min incubation at room temperature, the fluorescent marker solution was removed, biofilms were washed with water and the wells were then filled with 100 μl of water or PBS for WGA-stained biofilms. Stained biofilms were visualized by CLSM (Olympus FV1000 IX81, Markham, ON, Canada).

Effect of zinc on biofilm formation

Biofilms were prepared as described above with some modifications. Prior to inoculation, varying concentration of zinc was added to the biofilm medium by adding an identical volume of serial dilutions of a stock solution of ZnCl2 or ZnO in water. With the exception of Strep. suis, 0, 100, 250, 500, 750 or 1000 μ mol l−1 of zinc was added to the biofilm medium. As Strep. suis growth was more sensitive to zinc (Table 1), 0, 50, 100, 150, 200 or 250 μ mol l−1 of zinc was added. Plates were prepared in duplicate for each experiment: one plate was used to measure biofilm formation and the other plate was used to measure growth of the bacteria in the presence of Zn. Both plates were incubated as described in Table 2, and the biofilm plate was processed as described in 'Biofilm assay'. The unstained replicate plate was used to evaluate growth by measuring the absorbance at 600 nm.

Dispersion of preformed biofilms by zinc

Biofilms were prepared as described in 'Biofilm assay'. After the desired incubation time, the biofilms were washed with water and aliquots (100 μl) of growth medium containing different concentration of ZnCl2 (0, 100, 250, 500, 750 or 1000 μ mol l−1) were added to preformed biofilms. The biofilms were incubated for an additional 24 h in the presence of ZnCl2. Biofilms were then stained with crystal violet as described above. Dispersion was measured by comparing the amount of stained biofilm (OD590) in the control and treated wells. A biofilm was considered dispersed if the amount of zinc-treated biofilm was significantly reduced.

Statistical analysis

The effect of zinc concentration on the per cent biofilm formation from the untreated control was compared with one-way analysis of variance (anova) taking into account the bacterial growth and considering the runs of the ELISA as random effects [package lme4 (Bates et al. 2011) of R statistical software (R Development Core Team 2012)]. Multiple comparisons to the control concentration were realized by Dunnett's test.

Results

Biofilm formation

Optimal conditions for biofilm formation

In this study, optimal conditions for biofilm formation by different bacterial swine pathogens were determined based on information available in the literature (Table 2). The incubation period was selected to yield mature biofilms for each species. For E. coli and Salmonella, the bacteria need to be incubated at 30°C to form biofilms. The other species were able to form mature biofilms at 37°C. The typical time of incubation was 24 h, but Salmonella and H. parasuis required 48 h to form a mature biofilm, whereas 5 h was sufficient for A. pleuropneumoniae to form a mature biofilm. With the exception of E. coli, every species formed a biofilm in a polystyrene microtitre plate that was treated for tissue culture (TC) (Costar® 3599). E. coli did not form biofilms on the TC-treated polystyrene and required nontreated polystyrene to form biofilms (Costar® 3370). Biofilm formation for all bacteria was tested first in BHI. Most bacteria formed biofilms in BHI; however, E. coli, Strep. suis and Salmonella required the defined minimal media M9, basal broth medium (BBM) and colonization factor antigen medium, respectively. Finally, the Staph. aureus and Strep. suis biofilm medium required supplementation with glucose and fibrinogen, respectively, to form mature biofilms.

Typical biofilm assay results

Biofilm formation was assayed using a static microtitre plate assay and by staining the biofilm with crystal violet (Table 3). In a typical assay, A. pleuropneumoniae and Strep. suis strains were the strongest biofilm producers with an average A590 of dye that ranged from 1 to 3, followed by E. coli, Salmonella Typhimurium and Staph. aureus with A590 approx. 1·0. Bordbronchiseptica, H. parasuis and Salmonella. Heidelberg were the weakest biofilm formers with an average A590 ranging from 0·34 to 0·77.

Table 3. Biofilm formation in a microtitre plate
Bacterial strainsRange of OD590 nm after staining with crystal violetBiofilm thickness (in μm) as determined by CLSM
A. pleuropneumoniae S40741·85 ± 0·2435
Bord. bronchiseptica 2760·60 ± 0·2725
E. coli ECL 176081·04 ± 0·0723
H. parasuis Nagasaki0·77 ± 0·4720
SalmTyphimurium ATCC 140280·99 ± 0·1721
Salm. Heidelberg STF08-4530·34 ± 0·1020
Staph. aureus 154N1·09 ± 0·6340
Strep. suis 7352·63 ± 0·2635

Confocal laser scanning microscopy

To confirm the results obtained with the crystal violet assay, biofilms were visualized by CLSM. The biofilms were stained with FilmTracerTM FM® 1-43, a molecule that becomes fluorescent once it is inserted in the cell membrane. Biofilm structure characteristics varied among the different bacterial species. Representative CLSM images of the different biofilms are shown in Fig. 1. To further characterize the biofilms, 15 images of biofilm layers were recorded and stacked, and 3D-images of the biofilms were generated (Fig. 2a). Based on these reconstructions, the thickness of the biofilm produced by each bacterial species was evaluated. The thickness of A. pleuropneumoniae serotype 5b strain L20 biofilm was around 60 μm (Fig. 2b). The biofilm thickness for the other bacterial species ranged from 20 to 40 μm (Table 2).

Figure 1.

Confocal laser scanning microscopy of FilmTracerTM FM® 1–43 stained biofilms of Actinobacillus pleuropneumoniae S4074 (a), Bordetella bronchiseptica 276 (b), Escherichia coli ECL17608 (c), Haemophilus parasuis Nagasaki (d), Salmonella Typhimurium ATCC14028 (e), Salm. Heidelberg STF08-453 (f), Staphylococcus aureus 154N (g), Streptococcus suis 735 (h).

Figure 2.

Confocal laser scanning microscopy three-dimensional images of biofilm formation by Actinobacillus pleuropneumoniae strain L20 stained with FilmTracerTM FM® 1–43 (a) and stack of sections of the X–Z plane of the biofilm (b).

Effect of zinc on biofilm formation

Once growth conditions for optimal biofilm formation were determined, the effect of different zinc (ZnCl2) concentration on biofilm formation was assessed (Fig. 3). To test the relationship between the effect of zinc on bacterial growth and biofilm formation, A600 was recorded to assess the growth of the bacteria and, with a replicate plate, crystal violet staining was measured to assess the amount of biofilm formed. For the purpose of statistical analyses, the values were transformed so that they are represented as the percentage of the no-treatment (without zinc) control. Similar results were obtained with ZnO and are, therefore, not shown.

Figure 3.

Effect of ZnCl2 on the formation of biofilm and growth of Actinobacillus pleuropneumoniae S4074 (a), Bordetella bronchiseptica 276 (b), Escherichia coli ECL17608 (c), Haemophilus parasuis Nagasaki (d), Salmonella Typhimurium ATCC14028 (e), Salm. Heidelberg STF08-453 (f), Staphylococcus aureus 154N (g), Streptococcus suis 735 (h). Values are represented as percentage of the no-treatment control. Box and whisker plots represent biofilm formation and the diamonds represent bacterial growth. Black dots outside the box and whiskers are considered outliers. Statistical significance was established by analysis of variance (anova). Multiple comparisons to the control concentration were realized by the Dunnett's test. *< 0·01; ** 0·001

Actinobacillus pleuropneumoniae

Zinc inhibited biofilm formation by A. pleuropneumoniae in a dose-dependent manner as previously shown by our group (Labrie et al. 2010; Fig. 3a). Zinc treatment up to 250 μ mol l−1 did not affect bacterial growth. However, biofilm formation by A. pleuropneumoniae was significantly decreased (P < 0·001) when the same concentration of zinc (250 μ mol l−1) was added.

Bordetella bronchiseptica

Zinc did not have any effect on bacterial growth and on biofilm formation of Bord. bronchiseptica strain 276 (Fig. 3b).

Escherichia coli

When zinc was added, a significant decrease (P < 0·001) in biofilm formation was observed for E. coli (Fig. 3c). However, the amount of biofilm was approx. 50% of the no-treatment control. Bacterial growth was reduced to at least 80% of the control at 100, 250 and 500 μ mol l−1 (see Fig. 3c). Therefore, zinc was considered to have a slight effect on E. coli biofilm formation.

Haemophilus parasuis

Biofilm formation by H. parasuis was significantly reduced (P < 0·001) at a concentration as low as 100 μ mol l−1 of ZnCl2 (Fig. 3d). At higher concentrations (250–1000 μ mol l−1), the percentage of biofilm formation was stable at approx. 40% of the no-treatment control. Furthermore, zinc did not have a bactericidal effect (Figure 3d).

Salmonella Typhimurium

Salmonella Typhimurium formed significantly (P < 0·001) less biofilms in the presence of zinc. At lower concentrations (100 and 250 μ mol l−1), biofilm formation was reduced to approx. 60% (P < 0·001) and bacterial growth was reduced to approx. 80% of the control (Fig. 3e). At higher concentrations of zinc, there was almost no biofilm formed and growth remained at approx. 80% of the control (Fig. 3e).

Salmonella Heidelberg

In the case of Salm. Heidelberg, zinc decreased biofilm formation compared with the no-treatment control, but a similar decrease was observed in bacterial growth (Fig. 3f). Therefore, the effect of zinc on biofilm formation was not considered to be significant.

Staphylococcus aureus

Unlike most bacteria tested, biofilm formation by Staph. aureus seemed stimulated in the presence of 100 μ mol l−1 of zinc (Fig. 3g). Biofilm formation was reduced to 80% of the control when 500 μ mol l−1 ZnCl2 was added (P < 0·001). Bacterial growth was slightly affected by zinc. Therefore, high concentrations of zinc slightly decreased biofilm formation by Staph. aureus.

Streptococcus suis

Growth of Strep. suis was more sensitive to zinc when compared with the other bacterial swine pathogens tested. A significant (P < 0·001) reduction in biofilm formation was observed at 150 μ mol l−1 ZnCl2 (Fig. 3h). At 200 μ mol l−1 ZnCl2, Strep. suis biofilm formation was completely inhibited but growth was also markedly affected. Despite the effect of zinc on growth, we concluded that zinc had a significant effect on biofilm formation of Strep. suis at higher concentrations of zinc.

Inhibitory effect of zinc confirmed by CLSM

To confirm the inhibitory effect of zinc on biofilm formation, we used CLSM and fluorescent staining to visualize the zinc-treated and control biofilms. Biofilm formation by Salm. Typhimurium in the presence of ZnCl2 is shown as an example (Fig. 4). As observed with the microtitre plate assay and crystal violet staining, Salm. Typhimurium formed markedly less biofilm than the control when grown in the presence of 250 and 500 μ mol l−1 of ZnCl2 (Fig. 4).

Figure 4.

Confocal laser scanning microscopy images of Salmonella Typhimurium ATCC 14028 biofilms grown in the presence of different ZnCl2 concentrations (0:a, 250:b or 500 μ mol l−1:c). Biofilms were stained with FilmTracerTM FM® 1–43.

Effect of zinc on dispersion of preformed biofilms

The ability of zinc to disperse preformed biofilms was also evaluated. The addition of zinc (ZnCl2) followed by an additional incubation for 24 h did not result in a reduction in the amount of biofilm when compared with the control biofilm. Therefore, it was concluded that zinc did not disperse preformed biofilms (data not shown).

Composition of biofilm matrix

The matrix of the different biofilms was stained with fluorescent probes specific for poly-N-acetyl-D-glucosamine (PGA) (Wheat Germ Agglutinin), eDNA (BOBOTM-3 iodide) and proteins (FilmTracerTM SYPRO® Ruby). The composition of the matrix for the different bacterial pathogens is summarized in Table 4. The biofilm matrices of A. pleuropneumoniae and Staph.aureus were positive for all three components (PGA, eDNA and proteins), whereas Bord. bronchiseptica was negative for all three (Fig. 5; Table 4). E. coli was also negative for the three components and H. parasuis and Salm. Heidelberg were only positive for eDNA. Both Strep. suis and SalmTyphimurium were positive for eDNA and proteins.

Table 4. Composition of the biofilm matrix as determined by staining and CLSM
 Component
Bacterial strainsPGA (WGA)Extracellular DNA (BOBO-3)Protein (SYPRO Ruby)
A. pleuropneumoniae S4074+++
Bord. bronchiseptica 276
E. coli ECL 17608
H. parasuis Nagasaki+
Salm Typhimurium ATCC 14028++
Salm. Heidelberg STF08-453+
Staph. aureus 154N+++
Strep. suis 735++
Figure 5.

Confocal laser scanning microscopy images of the biofilm matrix of Actinobacillus pleuropneumoniae S4074 (a) and Bordetella bronchiseptica 276 (b) stained with FilmTracerTM FM® 1–43, wheat germ agglutinin (WGA) conjugated to Oregon Green 488, BOBO-3TM and Film TracerTM SYPRO Ruby.

Discussion

Given that biofilm-associated infections are often chronic and difficult to eradicate, the identification of antibiofilm molecules is of high importance (Hall-Stoodley and Stoodley 2009; Jacques et al. 2010). The use of metal ions to eradicate biofilms has received some attention (Harrison et al. 2005; Workentine et al. 2008). The potential of zinc as antibiofilm molecule has not been fully explored, but recent study have demonstrated that biofilms of enteroaggregative E. coli (EAEC; Pereira et al. 2010), uropathogenic E. coli (UPEC; Hancock et al. 2010) and dental plaque bacteria (Gu et al. 2012) were sensitive to zinc. Furthermore, our laboratory previously demonstrated that sub-bactericidal concentration of zinc could inhibit biofilm formation of the swine pathogen, A. pleuropneumoniae (Labrie et al. 2010). The objective of our study was to evaluate the effect of zinc on biofilm formation of other important bacterial swine pathogens including Bord. bronchiseptica, E. coli, H. parasuis, Salmonella, Staph. aureus and Strep. suis. Under optimal conditions for biofilm formation, the addition of sub-bactericidal concentration of zinc (ZnCl2 or ZnO) effectively blocked biofilm formation of A. pleuropneumoniae, Salm. Typhymurium and H. parasuis in a dose-dependent manner. Additionally, biofilm formation of E. coli, Staph. aureus and Strep. suis was slightly inhibited by the presence of zinc.

In our study, zinc was able to inhibit the biofilm formation of both intestinal and respiratory pathogens. Furthermore, the use of zinc to reduce diarrhoeal and respiratory diseases in humans and animals has already been demonstrated (Aggarwal et al. 2007; Crane et al. 2011; Molist et al. 2011). Thus, the reduction in intestinal and respiratory diseases by zinc can probably be attributed to both the antimicrobial and antibiofilm activity of zinc. In addition to preventing infectious diseases, zinc supplementation has been used in the diet of pigs to improve feed intake. For example, ZnO supplementation altered the development of the small intestine mucosa of weaned pigs (Slade et al. 2011) and improved feed intake and growth of piglets (Molist et al. 2011). Additionally, the combination of an antibacterial agent and ZnO supplementation leads to an improvement of performance markers (Hill et al. 2001). In combination with our data, these indicate the clinical value of zinc as an additive in diet of pigs. Furthermore, it highlights the possibility that zinc may act synergistically with biocides.

Synergic effects of metal ions and biocides on biofilms have also been investigated. For example, when copper was combined with quaternary ammonium cations, synergistic bactericidal and antibiofilm activity were observed against Pseudomonas aeruginosa (Harrison et al. 2008). This suggested that zinc could increase the effectiveness of disinfectants. If such phenomenon is observed, zinc could be used with other disinfectant to control environmental biofilms in the farm and food processing plants. Environmental biofilms are important in the persistence of bacterial pathogens, such as E. coli and Vibrio cholerae (Shikuma and Hadfield 2010).

Despite our positive results, the use of zinc may face some limitation given that bacterial evolution is a fairly rapid process. For example, the use of zinc could apply selective pressure for strains that are able to form biofilm in the presence of antibiofilm concentration of zinc. Furthermore, a selection pressure could also be applied on pathogens to increase the subpopulations that do not form biofilm. In our study, strains from a species respond similarly to presence of zinc; however, a larger set of isolates representing different genotypes should be included in future studies to ensure that antibiofilm effect of zinc is not genotype specific within a species.

The mechanism behind the antibiofilm activity of zinc has yet to be characterized, but zinc could interact with components of the matrix. It has been recently reported that PgaB activity, involved in de-N-acetylation of E. coli PGA, is decreased by zinc (Little et al. 2012). However, in our study, only A. pleuropneumoniae and Staph. aureus biofilms were found to contain PGA. Under our experimental conditions, species that did not produce PGA were also inhibited, indicating that the inhibitory effect of zinc does not appear to be solely dependent on the presence of PGA in the biofilm matrix. eDNA was one component that was present in most of the biofilm matrices. eDNA can act as a zinc chelator and this interaction has an impact on biofilm stability (Mulcahy et al. 2008). Zinc may also interfere with other cellular mechanisms such as signalling and gene regulation. Zinc can bind to the ferric uptake regulator and may affect iron homoeostasis (Klemm et al. 2010). For example, Zn interference has been observed in E. coli and Klebsiella pneumoniae (Hancock et al. 2010). Zinc can also interfere with a toxin–antitoxin (TA) system, MqsR/MqsA, which is associated with biofilm formation and the development of persister cells (Papadopoulos et al. 2012). Homologues to this TA system are found in many pathogenic bacteria (Gerdes et al. 2005). Finally, zinc can inhibit the EAL domains of cyclic diguanylate phosphodiesterases and this inhibition blocks the degradation of c-di-GMP (Tamayo et al. 2005; Jenal and Malone 2006). c-di-GMP is an important player in the regulation of biofilm formation, and interference in the c-di-GMP pathway will likely have consequences on the biofilm formation process (Jonas et al. 2009).

In conclusion, micromolar concentrations of zinc can inhibit biofilm formation by several Gram-negative and Gram-positive bacteria of porcine origin. The mechanism behind the antibiofilm activity of zinc has yet to be characterized. It does not, however, appear to be solely dependent on the presence of PGA in the biofilm matrix as initially thought. Given that zinc is a simple and inexpensive molecule, it would be worth to test whether synergic effects are observed with antibiotic and disinfectant treatments, and thus reduce the virulence, persistence and transmission of pathogenic bacteria. In addition, the nonspecific inhibitory effect of zinc on biofilm formation may well extend to other important human and animal bacterial pathogens.

Acknowledgements

This work has been supported by a Discovery Grant (003428 to MJ) from the Natural Sciences and Engineering Research Council of Canada and a New Initiative grant from the ‘Centre de recherche en infectiologie porcine’, a Strategic Network supported by the ‘Fonds de recherche du Québec – Nature et technologies’. CW received a scholarship from the China Scholarship Council.

Ancillary