Aims: To evaluate both the antimicrobial activity and the effectiveness of a combination of sodium hypochlorite and hydrogen peroxide (Ox-B) for killing Pseudomonas aeruginosa ATCC 19142 cells and removing P. aeruginosa biofilms on aluminum or stainless steel surfaces.
Methods and Results: Pseudomonas aeruginosa biofilms were developed in tryptic soy broth containing vertically suspended aluminium or stainless steel plates. Biofilms were exposed to a mixed sodium hypochlorite and hydrogen peroxide solution as a sanitizer for 1, 5 and 20 min. The sanitizer was then neutralized, the cells dislodged from the test surfaces, and viable cells enumerated. Cell morphologies were determined using scanning (SEM) and transmission electron microscopy (TEM). Cell viability was determined by confocal scanning laser microscopy (CSLM). Biofilm removal was monitored by Fourier transform infrared (FTIR) spectrophotometry. Cell numbers were reduced by 5-log to 6-log after 1 min exposure and by 7-log after 5 min exposure to Ox-B. No viable cells were detected after a 20 min exposure. Treatment with equivalent concentrations of sodium hypochlorite reduced viable numbers by 3-log to 4-log after 1 min exposure and by 4-log to 6-log after 5 min, respectively. A 20 min exposure achieved a 7-log reduction. Hydrogen peroxide at test concentration treatments showed no effect. FTIR analysis of treated pseudomonad biofilms on aluminium or stainless steel plates showed either a significant reduction or complete removal of biofilm material after a 5 min exposure to the mixed sodium hypochlorite and hydrogen peroxide solution. SEM and TEM images revealed damage to cell wall and cell membranes.
Conclusions: A combination of sodium hypochlorite and hydrogen peroxide effectively killed P. aeruginosa cells and removed biofilms from both stainless steel and aluminium surfaces.
Significance and Impact of the Study: The combination of sodium hypochlorite and hydrogen peroxide can be used as an alternative disinfectant and/or biofilm remover of contaminated food processing equipment.
Biofilms are the result of reversible and/or irreversible bacterial attachment and multiplication of cells on a surface (biotic or abiotic) with the associated production of extracellular polymeric substances (EPS). EPS are hydrated polyanionic polysaccharide matrices produced by polymerases, some of which are affixed to the lipopolysaccharide components of the cell walls (Chimielewski and Frank 2003; Jass et al. 2003). EPS not only provide protection to biofilm inhabitants by preventing access of biocides and sequestering toxins, acquiring nutrients and reducing desiccation but also by inducing phenotypic variation and intercellular communication (Heinzel 1998; Lechevallier et al. 1988;Cochran et al. 2000; Wirtanen et al. 2001; Chimielewski and Frank 2003; Jass et al. 2003). Biofilms are highly complex microbial communities that occur in both natural and artificial environments.
Studies by Gibson et al. (1995) on attached micro-organisms in 17 different food processing environments reported that 79% of the isolates were gram-negative rods, 8·6% gram-positive cocci, 6·5% gram-positive rods and 1·2% yeast, with Pseudomonas spp., staphylococci and Enterobacter spp. being the most commonly isolated organisms. Many pseudomonads are psychotrophic bacteria that can attach to surface materials with relative ease and are spoilage organisms in chilled foods (Lindsay and von Holy 1999; Holah and Gibson 2000). The survival of spoilage bacteria or pathogens in food products reduces shelf life and may be hazardous to human health.
The nature and degree of roughness of surface materials, temperature, pH, available nutrient, time availability and organic and inorganic materials remaining on the substratum enhance attachment of micro-organisms to surfaces and the development of complex biofilm communities (Mettler and Carpentier 1998; Holah and Gibson 2000). Stainless steel is a commonly used food preparation surface. Microbial attachment to this surface can lead to contamination and deterioration of foods producing economic loss. Other commonly found surfaces that support the growth of biofilms containing foodborne pathogens and spoilage bacteria include aluminium, glass, BunaN, Teflon seals and nylon (Kumar and Anand 1998; Holah and Gibson 2000).
In order to kill or remove biofilms, antimicrobials must penetrate the EPS to gain access to the microbial cells. Oxidizing biocides, such as chlorine and peroxides are nonspecific in their mode of action. These compounds are often preferred for general sanitation because of their availability and low cost.
An evaluation was conducted of the efficacy of a mixed sodium hypochlorite and hydrogen peroxide solution (US Patent 6,866,870) for killing and removing Pseudomonas aeruginosa ATCC 19142 cells, grown as biofilms, on aluminium and stainless steel surfaces. Pseudomonas aeruginosa is an important human pathogen generally associated with healthcare-related infections (Mattila-Sandholm and Wirtanen 1992; Lindsay et al. 1996; Chimielewski and Frank 2003; Thomas et al. 2005). However, this particular strain produces high amounts of extracellular polymeric substances thus making it a suitable organism for simulating a severe case of contaminated food processing equipment.
Materials and methods
Pseudomonas aeruginosa ATCC 19142 was obtained from the American Type Culture Collection (ATCC, Manassas, VA,USA). Cells were grown on tryptic soy broth (TSB) at 35°C (Difco, Sparks, MD, USA) and stored at 4˚C on tryptic soy agar (TSA) slants.
Stainless steel (Type 304) and aluminium (Alloy 3003) plates measuring 2·5-cm wide by 6·3-cm long by 3-mm deep were sand blasted to remove surface scratches. Before each experiment, the stainless steel plates were cleaned by dipping in acetone, air dried, rinsed in nanopure water, soaked in 2 N HCl for 2 h, rinsed once with deionized water and once with nanopure water and then air dried (Stewart et al. 2001). Aluminium plates were dipped in acetone, air dried and rinsed at least three times with nanopure water and air dried. Plates were autoclaved for 15 min at 121°C prior to use.
The sanitizer solution was prepared by combining 2500 mg l−1 hydrogen peroxide and 25 000 mg l−1 sodium hypochlorite from stocks of 30% hydrogen peroxide (Sigma-Aldrich, St. Louis, MO, USA) and 10–13% sodium hypochlorite (Sigma-Aldrich, Allentown, PA, USA) by the method listed in US Patent 6,866,870. This sanitizer solution is referred to as Ox-B throughout the text.
Pseudomonas aeruginosa was grown overnight in TSB, on a shaker-incubator at 35˚C. Each aluminium or stainless steel plate was suspended vertically in a 2-l beaker containing 1200 ml of TSB, which was inoculated with 5 ml of an overnight culture of the test organism. About 800 ml of the culture broth was replaced with the equivalent volume of fresh medium every 2 days, for a total of 6 days.
Each plate was removed from the broth on day six and the surface rinsed with 10 ml of distilled water. Separate plates were then vertically suspended in either 30 ml of Ox-B solution, 25 000 mg l−1 of sodium hypochlorite, 2500 mg l−1 of hydrogen peroxide or distilled water, for 1, 5 or 20 min contact time, followed by immersion in 30 ml of D/E neutralizing broth (Difco) for an additional 20 min. Plates were then shaken for 1 min, with 7·5 g sterile glass beads (3 mm) in 20 ml phosphate buffer, to dislodge attached cells. The suspension was serially diluted and the cells enumerated on Pseudomonas PTM agar using duplicate spread plates. Enumeration of surface-attached P. aeruginosa on both aluminium and stainless steel plates was performed prior to treatment with sanitizers or distilled water by plating. Plates were incubated for 24–48 h at 35˚C prior to counting. A minimum of two assays were conducted for each surface type and sanitizer.
Plates exposed to P. aeruginosa were sampled on day six as previously described, before and after sanitizer treatment, and prepared for scanning electron microscopy (SEM), transmission electron microscopy (TEM) and confocal scanning laser microscopy (CSLM).
SEM sample preparation
Biofilms on plates were fixed with 6% glutaraldehyde in 0·2 mol l−1 cacodylate buffer for 1 h, rinsed three times with buffer for a total of 30 min, exposed to osmium tetroxide vapour for 1 h, briefly rinsed with 0·2 mol l−1 of cacodylate buffer, dehydrated in an ethanol series and air dried. Sample plates were coated with gold/palladium in an Edwards S-150 sputter coater and the biofilm surface imaged using a Cambridge S-260 SEM. The average length of P. aeruginosa cells was estimated by measuring the length of at least 10 individual cells before and after treatment.
TEM sample preparation
Six-day-old P. aeruginosa biofilms growing on aluminium plates were treated with Ox-B for 20 min, neutralized with D/E neutralizing broth and the biofilm dislodged as previously described. Cells were centrifuged at 8000 g for 10 min and the pellet immediately fixed in 4% glutaraldehyde in 0·2 mol l−1 cacodylate buffer for at least 30 min and rinsed twice in buffer. This step was followed by postfixation with 2% osmium tetroxide for 30 min. Cells were then filtered with a 0·2 μm polycarbonate filter and dehydrated with increasing concentrations of ethanol over 2 h. Cells were infiltrated with ethanol : LR White embedding resin (1 : 1) for 1 h, followed by 100% LR White for 1 h. Cells were removed from the filter, embedded in LR White and polymerized at 60˚C overnight. Thin 60–90-nm sections were cut with a glass knife on a DuPont 5000 ultramicrotome, stained with Reynolds lead citrate and imaged with a JEOL 100CX TEM.
CSLM sample preparation
Coupons were rinsed with distilled water and flooded with 150 µl of each component of the Baclight Bacterial Viability Kit™ (Molecular Probes, Leiden, The Netherlands), diluted to the manufacturer’s instructions (Lindsay et al. 2002). Samples were kept in the dark at room temperature for 10 min, rinsed several times with sterile distilled water and viewed on a Leica DM IRE2 inverted scope.
Fourier transform infrared analysis
Stainless steel (Type 304) and aluminium (Alloy 3003) plates measuring 2·5-cm wide by 6·3-cm long by 3-mm deep were polished to a mirror finish. The plates were first coarse-polished using increasingly finer grades of emery paper (500, 1200 and 4000 mesh) mounted on a Struers LaboPol-5 lapping machine (Struers Inc, Denmark). Colloidal silica with an OP-NAP polishing cloth was used for further polishing of the plates. The finishing step involved using water and an oil-based lubricant mixed with a suspension of polycrystalline diamonds (particle size, 3 μm).
Biofilm and sample preparation
Cultures of P. aeruginosa ATCC 19142 were grown overnight as previously described. A 2-ml inoculum of this culture was inoculated into separate 1-l beakers each containing 500 ml of TSB and the vertically suspended aluminium or stainless steel plates. Approximately 300-ml culture broth were removed and replaced with the same volume of fresh medium every two days, for a total of 18 days. Duplicate plates were retrieved on days 6, 12 and 18 and each rinsed once with 10 ml of phosphate buffer. One of the duplicates was submerged in 30 ml of Ox-B for 5 min and then the sanitizer was neutralized using 30 ml D/E neutralizing broth as described previously. The second sample served as control and was treated only with phosphate buffer. Samples were air dried, and kept in a desiccator until all samples were collected. Samples were analysed using Fourier transform infrared (FTIR) spectrophotometry.
A Spectrum 4000 FTIR spectrophotometer (Nicolet, Madison, WI, USA) equipped with a narrow-band liquid nitrogen-cooled HgCdTe (MCT) detector was used for collecting the infrared spectrum. Spectra from 4000 to 700 cm−1 were collected in the reflectance mode with a resolution of 4 cm−1. Omnic software was used to express the ordinate as absorbance. Each spectrum was an average of 512 scans.
Means were calculated and compared with each other and with the corresponding controls by multifactor analysis of variance (anova; the SAS System for Windows 9·0, the SAS Institute, USA) at the 90% or 95% confidence level.
Effect of the surface composition on sanitizer efficacy
Ox-B was the most effective sanitizer tested against attached pseudomonads on either surface. After a 1 min exposure to Ox-B, cell numbers were reduced significantly (P < 0·05) by 5-log for aluminium and by 6-log for stainless steel, whereas treatment with equivalent concentrations of sodium hypochlorite produced 3-log or 4-log reductions for aluminium or stainless steel plates, respectively. A 5 min exposure to Ox-B reduced counts by 8-log or by 7-log, respectively. Sodium hypochlorite reduced counts by 4-log and 6-log, which were significantly (P < 0·05) different from Ox-B-treated samples. A reduction of 7-log was observed after 20-min treatments with sodium hypochlorite on both aluminium and stainless steel plates. No cells could be recovered after 20 min treatment with Ox-B from either surface. Hydrogen peroxide treatments did not produce changes significantly (P > 0·05) different from controls.
SEM images of 6-day-old P. aeruginosa biofilms growing on stainless steel (Fig. 1a) or aluminium (Fig. 1b) plates revealed cells embedded within a layer of EPS. This EPS layer appeared thicker on aluminium than on stainless steel surfaces although P. aeruginosa numbers on either surface averaged between 8 and 9 log CFU cm−2 (Fig. 2a,b). Cells averaged 0·74 μm in length and showed no visible malformations (Fig. 3a). In CSLM images, cells stained green (viable) and appeared spread out over the stainless steel and the aluminium surfaces (data not shown). In contrast, P. aeruginosa cells exhibited shape distortions, indentations and slight elongation after 1 min (Fig. 3b) or 5 min (data not shown) exposure to Ox-B. Cells exposed to Ox-B for 20 min showed severe indentations, distortions and roughness of surfaces (Fig. 3c). TEM images of 20-min treated cells revealed massive perforations to cell walls and cell membranes (Fig. 3d). The cell length of P. aeruginosa on aluminium surfaces did not change significantly (P > 0·05) after 1 or 5 min treatment with Ox-B. However, a significant (P < 0·05) increase in cell length, from 0·74 to 1·44 μm, was observed after 20 min. Significant (P < 0·05) differences in cell length were observed for stainless steel plates after treatment with Ox-B. The average length increased from 0·8 μm before treatment to 0·97 μm after 1 min and then to 1·02 and 1·5 μm after 5 and 20 min treatments, respectively. Lindsay and von Holy (1999) have reported that increases in cell length after oxidant treatment are an indication of cell injury. In this study, increased cell lengths and surface malformations after sanitizer treatment may have indicated cell death, which was supported by CSLM images showing red (death) cells (data not shown).
FTIR spectra of P. aeruginosa biofilms on aluminium and stainless steel surfaces revealed an array of bands between 1800 and 1000 cm−1, with different intensities, typical of biofilms. The control spectra lacked any bands that corresponded to organic and/or inorganic materials characteristic of EPS (data not shown). FTIR spectrum of P. aeruginosa biofilms, before and after treatment on days 6, 12 and 18, are presented in Figs 4 and 5. The FTIR bands between 1800 and 900 cm−1 correlate with the presence of proteins, polysaccharides and nucleic acids, macromolecules commonly found in biofilms (Suci et al. 1998). The protein region is characterized by the amide I (1652–1648 cm−1) and amide II (1550–1548 cm−1) bands (Schmitt et al. 1995; Schmitt and Flemming 1998). Prominent amide I bands (∼1668–1640 cm−1) and amide II bands (∼1551–1538 cm−1) were observed in untreated biofilms on both aluminium and stainless steel surfaces (Figs 4A, 5A). Amide I and amide II bands became more intense with time, an indication of protein accumulation over the 18 days of the experiment. Relative intensity of features near 1500 and 1400 cm−1 increased between days 6 and 18. According to Cheung et al. (2000), the band near 1500 cm−1 corresponds to the symmetric deformation of zwitterions adsorbed onto surfaces. The intensity of the band near 1400 cm−1 increased with time, an indication of an interaction of carboxylate ions with the surface (Doyle et al. 1982; Schmitt and Flemming 1998). Bands centered at 1081 and 1090 cm−1 corresponded to the C–O polysaccharide stretching mode (Bremer and Geesey 1991; Schmitt and Flemming 1998). These bands became more intense by day 18, indicative of a significant increase in polysaccharide concentration.
Removal of biofilm from stainless steel and aluminium surfaces
Spectra of biofilms on treated aluminium and stainless steel plates are shown in Figs 4B and 5B. The bands corresponding to those of proteins, nucleic acids and carbohydrates were greatly reduced compared with the untreated samples. Complete removal of proteins, nucleic acids and carbohydrates were observed for Ox-B-treated stainless steel surfaces from days 6 and 12. Only traces of carbohydrates remained attached to stainless steel surfaces on which pseudomonad biofilms had developed for 18 days.
The combination of sodium hypochlorite and hydrogen peroxide (Ox-B) had a synergistic effect and was a better sanitizer of P. aeruginosa in biofilms than matching concentrations of sodium hypochlorite or hydrogen peroxide. A 1 min treatment with Ox-B reduced P. aeruginosa counts by 5- or 6-log CFU cm−2 and no cells could be recovered after a 20 min treatment. Guidelines established by the Association of Official Analytical Chemists Germicidal and Detergent Sanitizer test state that decreases in attached cell numbers by ≥3 log units is an acceptable target value for sanitizer efficacy against biofilms (Lindsay and von Holy 1999; Lindsay et al. 2002). Hypochlorite and its active form hypochlorous acid penetrate biofilms poorly, because of the neutralization of the chlorine which reacts more rapidly with organic materials found on the surfaces than its rate of diffusion to biofilm interiors (Chen and Stewart 1996; Suci et al. 1998; Stewart et al. 2001). Stewart et al. (2001) reported reductions of 1·1- and 0·4-log after 60 min treatment of 6-day-old P. aeruginosa and Klebsiella pneumonaie biofilms with 1000 mg l−1 of chlorosulfamate or alkaline sodium hypochlorite. Similar penetration effects because of a reaction–diffusion interaction have been observed with hydrogen peroxide for P. aeruginosa biofilms (Stewart et al. 1998; Stewart et al. 2000). Cochran et al. (2000) reported that disinfection rate coefficients for P. aeruginosa averaged 0·551 mg−1 min−1 for monochloramine and 3·1 × 10−4 mg−1 min−1 for hydrogen peroxide; compared with 24 h biofilm disinfection rate coefficients of 0·291 mg−1 min−1 and 9·2 × 10−5 mg−1 min−1 for monochloramine and hydrogen peroxide, respectively. Oxidizing formulations based on hydrogen peroxide (1·5% hydrogen peroxide plus 5–15% peracetic and acetic acid; 50% hydrogen peroxide plus 0·05% silver ions) have been reported to be effective against gram-negative bacteria and biofilm constructs of Bacillus subtilis (Wirtanen et al. 2001). Eginton et al. (1998) reported that biofilms from P. aeruginosa and Staphylococcus epidermis were less susceptible than their planktonic cells to exposure for 5 min at 20˚C to sodium hypochlorite (0·02% or 0·015% w/v), DodigenTM (0·0015% w/v or 0·0006% w/v), sodium dodecylsulfate (6% w/v or 0·1% w/v) or Tween-80 (6% w/v). However, the attachment of these micro-organisms to surfaces was loosened by such treatments, with the exception of Tween-80, which strengthened the attachment of S.epidermis to stainless steel.
SEM and TEM images of Ox-B-treated attached P. aeruginosa cells revealed roughness, indentations and perforations to cell walls and membranes as well as an increase in cell length. The morphological changes observed on cell surfaces after treatment with Ox-B can be attributed to radical formation and changes in cell membrane permeability.
The increase in cell length as an indication of cell damage is also supported by CSLM images. CSLM images revealed that all cells stained red, indicating cell death. Lindsay et al. (2002) indicated that oxidizing agents such as chlrorine dioxide-containing sanitizers increase the length of Pseudomonas fluorescens M2 cells in biofilms, but CSLM images revealed that 90% of the cells stained yellow, an indication of cell injury, but not cell death.
Similar surface changes have been observed with planktonic and sessile B. subtilis and P. fluorescens cells after treatment with a mixture of hydrogen peroxide and peracetic acid (Lindsay and von Holy 1999). The hydrogen peroxide and peracetic acid combination acts by disrupting the sulfhydryl and sulphur bonds within enzymes, leading to disruption of cell membrane components. They also disrupt the chemiosmotic function of the membrane transport system by damaging the cell wall (Gougeon et al. 1996).
Pseudomonas aeruginosa cells attached on stainless steel were more sensitive to Ox-B or sodium hypochlorite than cells attached to aluminium surfaces. These results agree with another finding that sanitizer efficacy against attached cells in part depends on the type of the surface material (Frank and Chimielewski 1997). Vandevivere and Kirchman (1993) have proposed that cell attachment stimulates exopolymer synthesis independently of growth changes or nutrient limitation. Pseudomonas aeruginosa biofilms appeared thicker on aluminium than stainless steel surfaces. This difference in EPS composition was also observed through FTIR spectrometry. The changes observed in the spectra taken during biofilm development (days 6 through 18) suggest changes in chemical bonding with surface atoms involving COO– and NH3+ interactions (Cheung et al. 2000). Fletcher and Marshall (1982) reported similar findings in that surface-conditioning proteins can function as adhesins in specific attachment mechanisms. It is also possible that functional groups of a given polymer or different polymers may act cooperatively, binding to a particular surface at different times depending on environmental conditions, medium compositions and substratum chemistries (Doyle et al. 1982; Paul and Jeffrey 1985; Eginton et al. 1998)
Techniques such as physical scrubbing of surfaces and the use of high-pressure water sprays are often applied in food processing plants for the removal of bacterial biofilms (Gibson et al. 1999). A downside is the generation of aerosols which results in the dispersion of the surviving micro-organisms over an extensive area (Holah et al. 1993). Methods such as electric fields (Blenkinsopp et al. 1992) and ultrasound (Mott et al. 1998) have been used to enhance the removal of biofilms as well as the efficacy of sanitizers against biofilms. Detergents are also used in food processing environments but their formulations often focus on the removal of particular types of food soils rather than of biofilms (Holah and Gibson 2000). An exception is ethylenediaminetetraacetic acid (EDTA), which has been reported to improve the removal of biofilms from surfaces (Wirtanen et al. 1996; Kite et al. 2004; Percival et al. 2005). Ox-B was not only capable of penetrating and effectively killing P. aeruginosa cells embedded within biofilms, but also removed the biofilms from aluminium and stainless steel surfaces. FTIR spectra of 6, 12 and 18 days old P. aeruginosa biofilms treated with Ox-B for 5 min revealed drastic reductions or complete removal of proteins, nucleic acids and carbohydrates from both aluminium and stainless steel surfaces. Chemical compounds such as peracetic acid, mercuric chloride and formaldehyde have been found to have no effect on biofilms (Carpentier and Cerf 1993). Oxidizing substances such as chlorine or peroxoacetic acid are often used in the removal of biofilms (Meyer 2003). However, their activity is reduced in the presence of organic matter.
This new formulation of sodium hypochlorite and hydrogen peroxide (Ox-B) penetrates, kills and removes P. aeruginosa cells and biofilms from stainless steel and aluminium surfaces. Ox-B may provide a safe, effective and easy solution for killing potential pathogens as well as for disinfecting surfaces and removing biofilms. However, in situ work on food contact surfaces common to the food processing industry is necessary because of the complexity of biofilms.
This publication was made possible by National Institute of Health (NIH) Grant Number P20 RR16456 from the BRIN Program of the National Center for Research Resources. Its contents are solely the responsibility of the authors and do not necessarily represent the official views of NIH. We thank Cindy Henk for her technical assistance with SEM and TEM.