Notice: Wiley Online Library will be unavailable on Saturday 30th July 2016 from 08:00-11:00 BST / 03:00-06:00 EST / 15:00-18:00 SGT for essential maintenance. Apologies for the inconvenience.
Kalmokoff Bureau of Microbial Hazards, Food Directorate, Health Products and Food Branch, Banting Research Centre, Tunney’s Pasture, P.L. 2204A2, Ottawa, Canada, K1A 0L2 (e-mail: Martin_Kalmokof@hc-sc.gc.ca).
Aims: To determine whether isolates of Listeria monocytogenes differ in their ability to adsorb and form biofilms on a food-grade stainless steel surface.
Methods and Results: Strains were assessed for their ability to adsorb to a test surface over a short time period. Although some differences in numbers of bound cells were found among the strains, there were no correlations between the degree of adsorption and either the serotype or source of the strain. The ability of each strain to form a biofilm when grown with the test surface was also assessed. With the exception of a single strain, all strains adhered as single cells and did not form biofilms. Significant differences in adherence levels were found among strains. Strains demonstrating enhanced attachment produced extracellular fibrils, whereas those which adhered poorly did not. A single strain formed a biofilm consisting of adhered single cells and aggregates of cells.
Conclusions: Significant differences were found in the ability of various L. monocytogenes strains to attach to a test surface. In monoculture, the majority of strains did not form biofilms.
Significance and Impact of the Study: Differences in attachment and biofilm formation among strains provide a basis to study these characteristics in L. monocytogenes.
If you can't find a tool you're looking for, please click the link at the top of the page to "Go to old article view". Alternatively, view our Knowledge Base articles for additional help. Your feedback is important to us, so please let us know if you have comments or ideas for improvement.
While it has been clearly demonstrated that L. monocytogenes adsorbs to processing surfaces, there has been much less information concerning the ability of this species to actually form biofilms. A biofilm is a multicellular layer of adherent bacteria surrounded by a matrix of extracellular polysaccharide (Costerton et al. 1987). In nature, biofilms may be composed of a single species or represent a consortium of numerous species. Biofilm formation occurs in two distinct phases. The first stage involves adsorption of a cell to a surface. This step is reversible and the bacteria can be removed from the surface by gentle washing. Following the initial adsorption, the cells can become attached to the surface through the production of extracellular polysaccharides. The biofilm develops as a result of both the adsorption and adherence of new planktonic cells, combined with the continued growth of those which are already adhered. Biofilms represent a serious problem in food processing as they are a source of contamination. Furthermore, both attached cells and those contained within a biofilm are much more resistant to cleaning (Ren and Frank 1993; Lindsay and Vonholy 1999).
Although it appears to be widely accepted that L. monocytogenes forms biofilms on food-processing surfaces (see Gravani 1999), there has been very little direct microscopical evidence to support this. Scanning electron micrographs demonstrating the presence of individual cells on various surfaces have been reported (Herald and Zottola 1988; Mafu et al. 1990; Blackman and Frank 1996). There has also been evidence documenting the formation of microcolonies in some strains of L. monocytogenes following long-term incubation with multiple transfers of the test surfaces into fresh medium (Blackman and Frank 1996). In contrast, Hood and Zottola (1997a) reported that individual cells of L. monocytogenes V7 were able to attach to a stainless steel surface, but did not form a biofilm under conditions which would allow growth on this surface. Others have found that L. monocytogenes will only grow as a biofilm as part of a consortium of bacterial species (Sasahara and Zottola 1993; Jeong and Frank 1994a).
In this study, we have examined short-term adsorption and longer-term biofilm formation using different strains of L. monocytogenes. Our findings indicate that there appears to be very little difference among isolates in terms of the ability of cells to adsorb to a model surface. However, significant differences were found among these strains in terms of adherence and the ability to form a biofilm on a model stainless steel surface under conditions in which growth was allowed to occur during contact with the test surface.
MATERIALS AND METHODS
Isolates of L. monocytogenes and L. innocua used in this study are listed in Table 1 and Fig. 1, respectively. Listeria strains, Escherichia coli (O157:H7) and Salmonella enteritidis were obtained from the Bureau of Microbial Hazards culture collection. Isolates of other eubacteria used in this study were as follows: E. coli C-600, Pseudomonas aeruginosa ATCC 27853, Ps. aeruginosa ATCC 35032 and Enterococcus faecium ATCC 35667. Cultures were maintained frozen at −70°C. Cultures were grown at 21°C using brain heart infusion (BHI) broth and/or agar plates.
Table 1. Test strains of Listeria monocytogenes used in this study (also appear in the same order in Figs 2 and 6)
Stainless steel coupons (type 304, no. 4 finish, supplier unknown), used for both the adsorption assay and biofilm formation, were boiled for 5 min in a solution of 1% (w/v) sodium-dodecyl-sulphate and then rinsed in 100% isopropanol to remove any residual detergent. The coupons were stored in 100% ethanol and sterilized by flaming prior to placement within the test system.
Cells used for the adsorption assay were grown for 12 h at 21°C in 10 ml BHI broth. The inocula consisted of 10 μl from a 24-h 2·0-ml BHI culture. Cells were recovered by centrifugation (10 000 g, 10 min), washed once using an equal volume of Tris-buffered saline (TBS; 20 mmol l–1 Tris-HCl, 150 mmol l–1 NaCl, pH 7·5) and resuspended to their original optical density in TBS. A volume (8 ml) of the washed cell suspension was placed into the well of a sterile 12 × 10 ml polystyrene tissue culture plate with a lid (Corning, New York, NY, USA). A sterile steel coupon was then aseptically transferred to the well. The cells were incubated in the presence of the coupons for 2 h at room temperature with gentle rocking (60 rev min–1). For testing involving non-Listeria spp., densities of washed cell suspensions were adjusted to approximately 106 cells ml–1, as was the control strain.
After incubation, the coupons were removed and then rinsed by repeated immersion in 100 ml TBS for 10 s. The coupons were then stained by immersion in a 0·01% (w/v) solution of filter-sterilized acridine orange for 5 min. The coupons were gently rinsed by repeated immersion in distilled water to remove the excess stain and allowed to air dry. The adsorbed bacterial cells were counted using direct epifluorescence light microscopy (see below). The cells present in 10 randomly selected fields of view (at 1000 × magnification) were counted.
Isolates were screened in multiple batches. In all cases, an internal control consisting of L. monocytogenes Scott A was included with each batch of isolates. The mean value counts from each batch of isolates screened were normalized against the counts derived from this control.
For the assessment of biofilm formation, a sterile clean stainless steel coupon (see above) was immersed in 8 ml BHI contained in the well of a 12-well sterile polystyrene tissue culture plate. Each well was inoculated with 10 μl from a 24-h liquid culture and incubated with gentle agitation (60 rev min–1) for 72 h at room temperature. After incubation, coupons were removed and washed (see above), fixed and processed for examination by scanning electron microscopy (SEM; as described below).
Attached cells on the stainless steel chips processed for SEM were enumerated. Counting was carried out by photographing five randomly selected fields using 72-mm film at 2500 × magnification. Cells were visually counted from each negative and converted to cells mm–2.
Epifluorescence light microscopy
Adherent cells were visualized using a light microscope (Axiophot; Carl Zeiss Canada, Toronto, ON, Canada). The microscope was equipped for epifluorescent imaging (546 nm emission wavelength, 450–490 nm emission wavelength). Digital images of the adsorbed cells were recorded using a Coolsnap™TM camera (Photometrics; Roper Scientific Trenton, NJ, USA) connected to a PC running Metaview software (Universal Imaging Corp., Downington, NJ, USA). Micrographs were printed and adsorbed bacteria visually counted.
Scanning electron microscopy
The stainless steel coupons were removed from the BHI cultures, washed and fixed for 12 h in 0·2 mol l–1 cacodylate buffer, pH 7·4, containing 2·5% (w/v) glutaraldehyde. Sample preparation for SEM was as described previously (Austin and Bergeron 1995). Specimens were sputter-coated with 30 nm platinum and viewed under a scanning electron microscope (LE2100; Vickers Nanolab, Ottawa, ON, Canada) operated at an accelerating voltage of 15 kV.
Adsorption to a model stainless steel surface
The serovar and source of L. monocytogenes strains used in our experiments are listed in Table 1. Many of these strains have been involved in foodborne outbreaks. The ability of each strain to adsorb to the surface over a short contact period (2 h) was assessed using food-grade stainless steel as the model surface. Images from 10 randomly selected fields were recorded for each washed and stained coupon and adsorbed cells were enumerated. A control consisting of L. monocytogenes Scott A was included for each set of coupons; all of the counts (adsorbed cells mm–2) were normalized based on this control. Results for all 36 strains are shown in Fig. 2.
Most strains (21 of 36) showed levels of adsorption similar to the control strain L. monocytogenes Scott A (Fig. 2). Three strains demonstrated significantly lower adsorption and 12 strains adsorbed significantly better than L. monocytogenes Scott A. There were no obvious correlations between the numbers of cells bound to the surface and serotype or source of the strain. The response of strains involved in foodborne outbreaks was comparable to that of other test strains and showed adsorption levels of approximately 104 cells mm–2. Listeria monocytogenes 1/2a (strain 399) demonstrated the highest level of adsorbed cells (105 mm–2); however, the binding efficiency of this strain was not significantly better than that of other strains within the same serogroup.
A second series of experiments involving a 2-h contact period compared the ability of L. monocytogenes Scott A, L. innocua and other Gram-positive and -negative eubacteria to adsorb to the test surface (Fig. 1). Among the six non-listeria species, four strains (E. coli C600, Ps. aeruginosa ATCC 27853, Ps. aeruginosa ATCC 35032 and Ent. faecium ATCC 35667) demonstrated higher numbers of cells adsorbed than with the control L. monocytogenes Scott A. Two isolates (E. coli O157:H7 and Salm. enteritidis) were found to be no different, in terms of cells adsorbed, from the control. Similarly, a comparison of adsorbed cells between L. monocytogenes Scott A and isolates of L. innocua (Fig. 1) demonstrated no difference in adsorption characteristics among six of seven of these strains. A single isolate (L. innocua strain 227) demonstrated a higher number of adsorbed cells than found with L. monocytogenes Scott A. However, L. innocua strain 227 was equally adsorptive when compared with other L. monocytogenes strains which demonstrated higher levels of adsorption (Fig. 2). Overall, strains of Listeria spp. appeared to adsorb sparingly to the test surface compared with other species of eubacteria (Fig. 1).
Assessment of the ability to adhere and form a biofilm on a model surface
The ability of each strain of L. monocytogenes to form a biofilm on food-grade stainless steel over a 72-h growth and incubation period was also assessed. Initially, the biofilm test procedure was validated by examining biofilm formation using a variety of Gram-positive and -negative species (see above) which are known to form biofilms under similar growth conditions. An example contrasting the lack of biofilm formation in L. monocytogenes Scott A (Fig. 3a) with the biofilm formed by E. coli C600 (Fig. 3b) is shown. All of the non-listeria species attached to the steel surface and formed either extensive microcolonies (Salm. enteritidis and Ent. faecium) or thick biofilms (E. coli and Ps. aeruginosa).
The majority of L. monocytogenes strains (35 of 36) did not form a biofilm under these test conditions. Each of these strains formed a uniform distribution of cells across the entire steel surface (Fig. 4a and b). While there were differences in the density of adherent cells between strains, no microcolonies and few clusters of cells were observed (Fig. 4a and b). A single strain (L. monocytogenes CLIP 23485) was significantly different from the other stains of L. monocytogenes and formed a rudimentary biofilm. This isolate demonstrated a uniform distribution of cells on the test surface, as well as the formation of microcolonies and larger cell aggregates (Fig. 4c and d ).
Significant differences in the numbers of attached cells on the chips processed for SEM were apparent among all 36 strains examined. Total numbers of adherent cells ranged from 102 to > 105 cells mm–2; on this basis, the strains could be divided into two major groups, those which demonstrated low adherence (< 104 cells mm–2) and those which demonstrated a higher level of adherence (> 104 cells mm–2; Fig. 5). Five strains (L. monocytogenes 563, 568, 642, 913 and 6861) adhered at less than 103 cells mm–2 and nine strains (L. monocytogenes 412, 1340, Scott A, F7032, 82-131, L3357, L3353d, CLIP 22573 and ATCC 19116) adhered at more than 103 but less than 104 cells mm–2. In general, serotype 1/2a (five of 10) demonstrated the poorest overall adherence (< 104 cells mm–2), whereas serotype 1/2b was the best overall (six of seven > 104 cells mm–2). Four isolates demonstrated very high adherence (L. monocytogenes 399, 492, 503 and CLIP 23485). Half of the strains (18 of 36) demonstrated adherence levels between 104 and 105 cells mm–2 (L. monocytogenes 410, 413, 976, 2070, 394, 395, 658, 1367, ATCC 19112, 673, 81-861, F7032, L3503a, L3257, LI531, 89-85, SLU 2157 and 42).
In general, strains showing higher densities of adhered cells (compared with L. monocytogenes Scott A) also displayed the presence of extracellular fibrils (Fig. 6d–f). The fibrils extended both between cells as well as between cells and the stainless steel surface. Isolates which demonstrated average or poor binding characteristics either lacked these fibrils or possessed them at a much reduced level (Fig. 6a–c). The presence of similar structures has been noted previously in other isolates of L. monocytogenes bound to solid surfaces (Herald and Zottola 1988).
Short-term adsorption and longer-term biofilm formation in a variety of L. monocytogenes strains has been investigated. The ability of L. monocytogenes to adhere to surfaces in the food-processing environment is well known and numerous studies have examined the effect of environmental factors and growth conditions on binding to these surfaces. However, much of the previous research has utilized a limited number of test strains and, in many instances, only L. monocytogenes Scott A. More recent work involving larger numbers of isolates has suggested that there are significant differences in the ability of individual strains to attach to stainless steel, with members of the 1/2c serogroup exhibiting the highest levels of attachment (Norwood and Gilmour 1999; Lunden et al. 2000). An important, and as yet unanswered, question is whether enhanced binding or attachment to these surfaces bears any relationship to pathogenicity in foodborne isolates.
Our initial experiments assessed the ability of non-growing cells to adsorb to stainless steel over a short 2-h contact period. As L. monocytogenes Scott A has been widely utilized in similar studies, it is of interest to compare the results from this strain with other strains. Overall, the numbers of adsorbed cells in 21 of 36 strains of L. monocytogenes were not significantly different from L. monocytogenes Scott A, although there were strains which were more (12 of 36) or less (three of 36) adsorbed than the reference strain. However, all of the strains adsorbed to the surface to an approximate level of 104–105 mm–2. There were no direct correlations between the serotype or source and the levels of adsorbed cells, nor were the foodborne outbreak strains different from other test isolates. Furthermore, both L. monocytogenes and L. innocua strains adsorbed to stainless steel at much lower levels than some of the other Gram-positive and -negative isolates, as has been previously reported (Jeong and Frank 1994a, 1994b; Hood and Zottola 1997a, 1997b). Taken together, these findings indicate that the adsorption characteristics, determined using standardized testing conditions, are generally similar among this diverse selection of L. monocytogenes isolates.
The second series of experiments assessed the biofilm formation ability among the strains. Marked differences were noted among strains, including the density of attached cells, the presence/absence of extracellular fibrils and the ability of strains to actually form a biofilm. In contrast to L. monocytogenes CLIP 23485 and the non-listeria spp., most of the L. monocytogenes strains (35 of 36) did not form a biofilm under the test conditions, but adhered to the surface as isolated cells, similar to that reported by other workers (Herald and Zottola 1988). There were, however, large differences among the strains in terms of the numbers of cells adhering to the model surface. In general, adherence of the 1/2b strains was less variable than that found with the 1/2a and 4b serogroups. However, this finding may only reflect the limited numbers of strains tested within each serogoup. There were also no obvious trends in terms of an enhanced level of adherence among either the food or clinical isolates, including those from foodborne outbreaks.
In contrast to the short-term adsorption experiments where the strains behaved in a similar manner, much more marked differences in the number of adhered cells were found on the chips processed for SEM. For example, the number of adhered cells of L. monocytogenes Scott A on the surface after processing for SEM was approximately 10 × lower than that found in the short-term adsorption assay. Other strains (L. monocytogenes 563) exhibited 100-fold lower cell numbers than found in the initial adsorption assay. Overall, the differences in cell numbers found between the short-term adsorption assay and the biofilm assay probably reflect the difference between cells which are ‘adsorbed’ onto the surface and those which have ‘adhered’ to the surface. In the short-term adsorption experiment, the test surface was subjected to a relatively gentle wash and it is probable that this wash was insufficient to dislodge cells which were only adsorbed to the surface. Other studies, in which significant differences in adsorption were reported in short-term assays (Lunden et al. 2000), employed more stringent washing conditions for the test surface. In contrast to the short-term adsorption assay, the processing steps to prepare the chips for SEM (fixation, dehydration and sputter coating) are quite harsh and probably only preserved the cells which were firmly attached to the surface.
The second obvious difference found among these strains related to the presence of surface fibrils. The presence of similar structures has previously been reported in other isolates of L. monocytogenes (Herald and Zottola 1988). The fibrils were generally associated with highly adherent strains and were absent on strains demonstrating low levels of adherence. The precise nature of these fibrils is unknown, although they may represent extracellular polysaccharides which have condensed to form a fibril as a result of specimen preparation (Herald and Zottola 1988). In this study, critical point drying was replaced with chemical drying using hexamethyldisilazane, which has been found to aid in the preservation of bacterial pili and exopolysaccharides (Dekker et al. 1991; Austin and Bergeron 1995; Austin et al. 1998; Warburton et al. 1998). On this basis, these structures may not be artefactual. Both the precise nature and their possible role in surface adherence will require further investigation.
Although it is widely stated throughout the literature that L. monocytogenes forms biofilms on food-processing surfaces, our results indicate that this may not be correct. Under our test conditions, only a single isolate (L. monocytogenes CLIP 23485) formed a rudimentary biofilm. The strain was atypical and formed both microcolonies and larger cell aggregates on the test surface, morphologically similar to the biofilms formed by some of the other bacterial species tested. This strain was originally isolated from a large foodborne outbreak (Jacquet et al. 1995) and does indicate that there is a potential within this species to form a biofilm. Other reports examining longer-term biofilm formation have noted that L. monocytogenes is a poor former of biofilms and this has led to suggestions that these strains may use a primary colonizing bacteria of a different species to form a biofilm consortium on a surface (Sasahara and Zottola 1993).
We have investigated both short-term adsorption and longer-term biofilm formation in a wide range of isolates using stainless steel as a model surface. Our findings indicate that, while there appears to be little difference in terms of the ability of a given isolate to adsorb to our test surface, there were obvious differences among these strains in terms of their ability to adhere and form a biofilm under conditions in which growth was coupled with attachment. Three distinct phenotypes were observed in our biofilm assay: poor/good adherence, presence/absence of surface fibrils and the ability to form a biofilm, all of which will probably be reflected in terms of differences in the cell surface among the strains. Little is known concerning the molecular basis of either attachment or longer-term biofilm formation in L. monocytogenes. However, the involvement of flagellar filaments in surface binding, the initial step of biofilm formation, was recently demonstrated (Vatanyoopaisarn et al. 2000). The important question remaining is whether these differences in adherence, or the ability to form a biofilm, may influence the survival of pathogenic strains within the food-processing environment.