![[Feedback]](/icons/banner/feedbackACT.gif){kind=icon}
![[For Subscribers]](/icons/banner/subscriberACT.gif){kind=icon}
![[Archive]](/icons/banner/archiveACT.gif){kind=icon}
![[Search]](/icons/banner/searchACT.gif){kind=icon}
![[Contents]](/icons/banner/tocACT.gif){kind=icon}
Previous Article | Next Article 
Applied and Environmental Microbiology, August 2003, p. 4814-4822, Vol. 69, No. 8
0099-2240/03/$08.00+0 DOI: 10.1128/AEM.69.8.4814-4822.2003
Copyright © 2003, American Society for Microbiology. All Rights Reserved.
Department of Cell and Molecular Biology—Interface Biophysics,1 Department of Cell and Molecular Biology—Microbiology, Göteborg University, 405 30 Göteborg, Sweden2
Received 21 January 2003/ Accepted 12 May 2003
 |
ABSTRACT |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
INTRODUCTION |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Adsorption of a conditioning layer of molecules, determined by the substances in the solution, precedes adhesion of bacteria to surfaces (37). Attached to surfaces in contact with process water, bacteria produce large amounts of extracellular polysaccharides (EPS) or slime while forming biofilms. EPS bind a biofilm together as a matrix and anchor the biofilm to the surface (9, 13, 48). Bacteria in biofilms generally express starvation phenotypes and defense mechanisms, such as EPS, multidrug efflux pumps (acrAB), and alteration of the membrane protein composition (1, 33, 47). As a consequence, bacteria become more resistant to biocides and disinfectants while they are attached to a surface (10, 18, 45).
N-Acetyl-L-cysteine (NAC), which is used in medical treatment of chronic bronchitis, cancer, and paracetamol intoxication (36, 43), is one of the smallest drug molecules in use (28), and it has antibacterial properties. The molecule is a thiol-containing antioxidant that disrupts disulfide bonds in mucus (3, 41) and competitively inhibits amino acid (cysteine) utilization (46, 50).
Despite its importance as a medicine, the effect of NAC on bacteria has been poorly studied. Perez-Giraldo et al. (32) showed that NAC reduced the formation of biofilms of Staphylococcus epidermidis on a polystyrene surface but did not distinguish between the effects on bacteria and bacterial EPS. Recently, it was shown that NAC reduced adhesion of Streptococcus pneumoniae and Haemophilus influenzae to oropharyngeal epithelial cells in vitro (36). Similarly, NAC also reduced attachment of Moraxella catarrhalis to pharyngeal epithelial cells, possibly by removing a layer of ruthenium red-positive extracellular material from the epithelial cells (49). The effect of NAC on bacteria and bacterial biofilms is still relatively unknown, and a better understanding of bacterial responses to NAC may facilitate efficient use of this compound as a biofilm inhibitor.
The objective of this study was, in general, to provide to a better understanding of how NAC influences biofilm processes. More specifically, we wanted to study the mode of action of NAC and to test its potential as a biofilm inhibitor. Since the initial adhesion is dependent on the substratum wettability, the contact angle of the stainless steel surface was determined in the presence and absence of NAC. Furthermore, we investigated the effect of NAC on (i) bacterial adhesion to, and detachment from, a stainless steel surface and (ii) bacterial growth and production of EPS. Ten different bacteria isolated from paper machines, all frequently found in this environment, were tested for the potential to form biofilms as monocultures and/or as a multispecies community in the presence and absence of NAC.
|
MATERIALS AND METHODS |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
cm-1) and cultivated at 37°C. Colonies that occurred frequently and were morphologically unique were picked and purified by the streak dilution method. Bacteria were identified by the Culture Collection, University of Göteborg, Göteborg, Sweden (CCUG) on the basis of phenotyping with classical and commercial tests (approximately 35 to 40 tests), API 20NE tests, and cellular fatty acid analysis (fatty acid methyl esters) with a gas chromatograph. Strain matching and cluster analysis for numerical analysis were performed by comparison with previously processed reference strains in the CCUG database (Table 1). Stock cultures were frozen in media amended with 40% glycerol and stored at -70°C. Experiments were performed with monocultures and a seven-member multispecies community (Acinetobacter baumannii, Acinetobacter lwoffii, Bacillus sp., Enterobacter cloacae, Klebsiella pneumoniae, Pseudomonas mendocina, and Staphylococcus warneri) comprised of bacteria frequently found in paper mill environments (Table 1) (8, 17, 26, 34).
|
Measurement of EPS. (i) Sampling procedure.
Bacteria were cultivated overnight at a neutral pH in 20% LB at 37°C in the presence or absence of NAC (0.25 or 0.5 mg ml-1) since most EPS-producing bacteria prefer to grow and produce the largest amount of EPS under these conditions (8, 21). The pH was determined in liquid media with and without NAC. Cells were then harvested by centrifugation, washed twice in physiological saline (PS) (0.9% [wt/vol] NaCl), and then starved in PS for 48 h in the presence or absence of NAC. The strains of A. lwoffii, B. cereus, Bacillus sp., P. mendocina, and S. warneri did not grow in the presence of 0.5 mg of NAC ml-1 in media. These bacteria were grown in 20% LB overnight without NAC, washed, and subsequently exposed to 0.5 mg of NAC ml-1 in PS during the starvation phase.
(ii) Carbohydrate assay.
The amount of carbohydrates produced by each strain, as well as by the multispecies community, was determined by using a modified version of the acid hydrolysis method of Dall and Herndon (14). Briefly, polysaccharides but not monomers were precipitated with ethanol and then dehydrated with concentrated acid to a furfural. When tryptophan reacted with furfural, a condensation product was formed, which developed a brownish violet color. The amounts of carbohydrate were determined spectrophotometrically by measuring the absorbance at 500 nm.
Samples were sonicated at 10% power for 30 s with an xL Ultrasonic processor (Novakemi AB, Enskede, Sweden) and then centrifuged at 950 x g for 10 min. One milliliter of the supernatant fluid, which contained the polysaccharide, was precipitated by adding it in drops to cold absolute alcohol and kept overnight at 4°C. Then the polysaccharides were pelleted by centrifugation (2,400 x g, 15 min) and resuspended in 1 ml of distilled water, after which they were digested with 7 ml of sulfuric acid (77%) into monosaccharides. Samples were cooled for 10 min in an ice bath. One milliliter of cold tryptophan (1%, wt/wt) was then added to each tube and mixed. After exposure to a boiling bath for 20 min, the tubes were cooled on ice, and the absorbance at 500 nm was read with a spectrophotometer (Novaspec II; Pharmacia Biotech, Uppsala, Sweden) by using a PS blank which was subjected to the same procedure. Dry weights of cells suspended in PS, as well as PS alone, were measured by using a dry weight oven (CEM AVC-80). Samples were corrected for salinity by subtracting the weight of the PS. The amount of EPS was expressed in milligrams per gram (dry weight) of cells. The assay was calibrated by using a dextran standard which was subjected to the same procedure. Measurements were obtained for three different cultures in duplicate for all strains.
The contributions of proteins to the EPS were measured spectrophotometrically with isolated EPS at 562 nm by using a protein assay (BCA protein assay kit), and the proteins accounted for <2% of the total amount of EPS measured. The protein assay was calibrated by using a bovine serum albumin standard.
Measurement of EPS degradation.
Degradation of EPS was tested by adding 0.5 mg of NAC ml-1 (final concentration) to 1 ml of the supernatant fluid containing EPS from P. mendocina, K. pneumoniae, E. cloacae, B. cereus, B. megaterium, and B. subtilis, as well as dextran, for triplicate samples from two different cultures. Controls without NAC were also included. The tubes were incubated for 1 h at room temperature. The amount of carbohydrates was determined by using the acid hydrolysis method described above.
Preparation and characterization of stainless steel surfaces.
Before use, the stainless steel (SIS 2343) surfaces (26 by 26 mm) were cleaned and sterilized. The surfaces were washed in water, chloroform, methanol, and acetone for 5 min each. The surfaces were dried and autoclaved for 20 min at 121°C as described by Johansen et al. (22). To determine the surface wettability, the surfaces were incubated in petri dishes with HRW (18 M cm-1) or NAC (0.5 mg ml-1) at room temperature for 24 h. The advancing contact angles at the air-dried surfaces were then measured with a goniometer (FTA 200). The contact angles of 10-µl drops of HRW (18 M cm-1) were measured eight times per surface, and the results shown below are the means for five sample surfaces.
Bacterial adhesion to stainless steel surfaces.
The effect of NAC on adhesion of a multispecies community was evaluated by using 0.5 mg of NAC ml-1 in media since growth was not inhibited at this concentration. The experiment was performed both in 20% LB and in process water (white water). The process water was collected from a Swedish paper mill, sterile filtered (pore size, 0.20 µm), and used within 3 h.
One milliliter containing approximately 4 x 106 actively growing (exponential-phase) cells of each strain was centrifuged and resuspended in 20 ml of fresh 20% LB or sterile process water to obtain a concentration of 1 x 106 cells per ml. Stainless steel surfaces were exposed to the multispecies community in the presence and absence of NAC (0.5 mg ml-1) during shaking (100 rpm) at 37°C. Adhesion of the bacterial mixture was measured after 24, 48, and 72 h of incubation.
Since K. pneumoniae and E. cloacae were the two strains that grew at this NAC concentration, we tested the adhesion of these strains when they were grown separately in 20% LB. The adhesion experiments were performed as described above.
To measure the ability of NAC to detach bacteria from the stainless steel, we incubated surfaces in 20% LB for 48 h to allow adhesion of the multispecies community. The surfaces were then rinsed in PS and immersed in fresh PS in the presence or absence of NAC (0.5 mg ml-1). The stainless steel surfaces were then incubated for a further 24 h. To count attached cells, the surfaces were rinsed in PS for 10 min with shaking (100 rpm), which allowed only irreversibly attached cells to be present; the surfaces were stained with acridine orange and eventually rinsed with PS. The surfaces were examined with an Olympus epifluorescence microscope (magnification, x1,250) by using blue excitation light. At least 30 fields of view were counted on each surface. Three surfaces were examined for each incubation time from three different cultures.
Visualization of capsular polysaccharide (CPS) and EPS of K. pneumoniae by a fluorescent lectin probe.
The lectin Triticum vulgaris wheat germ agglutination (WGA) (Sigma) was labeled with the fluorescent probe tetramethyl rhodamine isocyanate (excitation wavelength range, 360 to 557 nm; emission wavelength maximum, 576 nm). WGA was used because it specifically binds to N-acetylneuraminic acid and N-acetyl-ß-D-glucosamine (40). These saccharide residues are frequently encountered in bacterial EPS (9).
Surfaces with attached K. pneumoniae were washed three times with 1 ml of phosphate-buffered saline (PBS) (8.5 g of NaCl, 1.34 g of Na2HPO4, 0.39 g of NaH2PO4·2H2O, 1,000 ml of HRW [18 M cm-1]) and then with 0.2 M cacodylate-HCl (sodium salt), and then they were covered with 1 ml of CT buffer (0.5% casein, 0.05% Tween 20, 0.1 mM Ca2+, and 0.1 mM Mn2+ in PBS [pH 7.4]) (38) for 1 h at room temperature. The surfaces were washed again with PBS and covered with 500 µl of CT buffer containing lectin (300 µg/ml of CT buffer). The slides were placed at room temperature for 1 h to allow the lectin to bind to the saccharide residues of EPS. Unbound lectin was washed off with PBS. Samples were stored at 4°C overnight and examined within 24 h. Surfaces were covered with SlowFade (DABCO antifade reagent in glycerol and PBS) and examined with an Olympus epifluorescence microscope.
|
RESULTS |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
After 24 h of growth, the total numbers of the multispecies community bacteria were higher in cultures supplemented with 0.5 mg of NAC ml-1 (1.8 x 109 ± 3.1 x 107 CFU ml-1), as determined by the higher optical density, than in cultures without NAC (1.2 x 109 ± 1.1 x 107 CFU ml-1) (Fig. 1f). In the absence of NAC in 20% LB K. pneumoniae and E. cloacae dominated the community, while Bacillus sp. and S. warneri were present at low levels and the levels of A. baumannii, A. lwoffii, and P. mendocina were below the detection limit (1 x 104 CFU ml-1).
Effect of NAC on EPS production.
The bacteria used in this study produced various amounts of EPS (approximately 60 to 600 mg g [dry weight] of cells-1) (Fig. 2). K. pneumoniae and A. baumannii produced the largest amount of EPS, about 10 times more than S. warneri and Bacillus sp. produced. The results showed that EPS production decreased significantly in the presence of NAC (0.25 mg ml-1) for B. megaterium, B. subtilis, B. cereus, A. lwoffii, P. mendocina, S. warneri (P < 0.005, as determined by Student's t test), K. pneumoniae, Bacillus sp., and the multispecies community (P < 0.05, as determined by Student's t test), although growth was not affected for most of the bacteria tested. The amount of EPS produced by B. megaterium and B. subtilis decreased by as much as 96% ± 30%, while NAC tended only to affect the EPS production of A. baumannii. The average reduction in the amount of the EPS produced was 58% ± 20% for the bacteria tested. The presence of 0.5 mg of NAC ml-1 in media resulted in the same reduction pattern as the presence of 0.25 mg ml-1 resulted in except for A. baumannii; for the latter organism the amount of EPS produced in the presence of 0.5 mg of NAC ml-1 tended to be greater than the amount produced in the presence of 0.25 mg of NAC ml-1.
|
|
EPS degradation.
EPS isolated from K. pneumoniae, P. mendocina, E. cloacae, B. cereus, B. megaterium, B. subtilis, and dextran was tested for susceptibility to degradation into monosaccharides by NAC. None of the polysaccharides were degraded by NAC (0.5 mg ml-1) (Fig. 4), suggesting that NAC instead affects EPS production.
|
cm-1) had a contact angle of 76 ± 2°. The wettability increased (i.e., the contact angle decreased) to 38 ± 3° after 24 h of incubation in the presence of 0.5 mg of NAC ml-1, indicating that NAC adsorbs to stainless steel surfaces. The number of attached cells of a multispecies community grown in 20% LB in the absence of NAC increased with time (Fig. 5), while the number of attached cells decreased with time when the preparations were incubated in process water. It was evident that NAC (0.5 mg m-1) had an inhibitory effect on the adhesion of a multispecies community in 20% LB (P < 0.0001, as determined by Student's t test) (Fig. 5) and in process water (P < 0.005) (Fig. 6). The presence of NAC reduced the number of attached multispecies community bacteria by as much as 76% ± 46% in 20% LB and by 57% ± 21% in process water after different incubation times. Since K. pneumoniae and E. cloacae were the two strains that grew at this NAC concentration, we tested the adhesion of these strains when they were grown separately in 20% LB (Fig. 5). The results show that adhesion was reduced significantly for both monocultures (P < 0.0001, as determined by Student's t test) as well. The number of attached K. pneumoniae cells was reduced by 51% ± 18%, while the number of E. cloacae cells was reduced by as much as 99% ± 16% after different incubation times. These reductions were seen despite the fact that the growth of these bacteria was not reduced by 0.5 mg of NAC ml-1 (Fig. 1c and d).
|
|
|
|
|
DISCUSSION |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
NAC is considered to be a nonantibiotic drug but to have antibacterial (bacteriostatic) properties (31, 32). NAC did indeed reduce growth to various degrees for all strains tested, as well as the multispecies community (Fig. 1). The gram-positive strains were generally more sensitive to NAC than the gram-negative strains. B. megaterium showed a very long lag phase (13 h) in the presence of 0.5 mg ml-1 before growth started. The reason for this delay of growth is not known. The inhibitory effect of NAC on growth could not be explained in terms of gram-positive or gram-negative bacteria, since P. mendocina was much more sensitive that the other gram-negative organisms. This result fits well with previous studies in which different Pseudomonas strains were shown to be more sensitive to NAC than Staphylococcus aureus, K. pneumoniae, or E. cloacae (31). The peculiar growth curve of P. mendocina can be explained by aggregation in the stationary phase, which resulted in a reduced optical density and subsequent dissociation of aggregates after longer times, which caused a markedly increased optical density after 25 h.
The growth response to NAC in the multispecies community culture was complex. First, the maximum biomass (optical density) of the community increased after addition of NAC at concentrations up to 0.5 mg ml-1 compared with the maximum biomass of the control (Fig. 1f). The higher maximum optical density at an NAC concentration of 0.25 mg ml-1 is difficult to explain, since most strains grow at this NAC concentration. At a concentration of 0.5 mg ml-1 the response may have been the result of elimination of sensitive strains, leaving more substrate for the resistant organisms K. pneumoniae, E. cloacae, and A. baumannii, although A baumannii was less numerous than the other two species. The fact that the final optical density was higher in the multispecies community (Fig. 1f) than in the corresponding monocultures of A. baumannii, E. cloacae, and K. pneumoniae (Fig. 1b to d) may have been the result of better substrate utilization by the particular strains together, resulting in higher total biomass. Why the multispecies community was not able to grow at all in the presence of 1.0 mg of NAC ml-1 is puzzling, since several of the strains grew at this NAC concentration as monocultures. The results show that the growth inhibition effects of NAC can be different in monocultures, with no interspecies interactions, compared to the effects in multispecies communities. In the natural situation, with even more complex community structures and interactions, there may well be combinations of factors that cause quite unforeseen results when inhibitors are used.
Adhesion of bacteria to surfaces in general involves a complex interplay of physical, chemical, and biological factors (5, 20). It is generally thought that bacterial adhesion is preceded by the adsorption of a conditioning layer of molecules that influence adhesion of bacteria to stainless steel surfaces (37). The conditioning layer is obviously determined by the substances in the solution. NAC adsorbs to a steel surface, as seen by the increased wettability of the surface (see Results). Therefore, it is interesting that NAC (0.5 mg ml-1) dramatically reduced the adhesion of bacteria to steel surfaces both in 20% LB (Fig. 5) and in the paper mill process water (Fig. 6), even if there was a higher total number of bacteria in the bulk phase in the presence of 0.5 mg ml-1 NAC than in the control without NAC.
In the multispecies community, K. pneumoniae and E. cloacae dominate because of their resistance to 0.5 mg of NAC ml-1 and their high growth potential. When the organisms were tested separately, NAC reduced adhesion of K. pneumoniae, while adhesion of E. cloacae was totally inhibited (Fig. 5). It is baffling that the effect on adhesion was greater for E. cloacae than for K. pneumoniae, since NAC decreased the EPS production by K. pneumoniae (Fig. 2) (thereby changing its surface character [Olofsson and Hermansson, unpublished data]), while NAC had no effect on E. cloacae EPS (Fig. 2).
It is worth stressing that NAC not only reduced adhesion but in fact also detached adhered cells from a steel surface. This has some importance since the initial adhesion often develops into a stronger interaction with time (bond ageing) (27). Busscher et al. (7) suggested that designs of new antiadhesive coatings should be based on the ease with which the initially adhering bacteria can be detached. The conditioning film should have a low cohesive strength, resulting in a weak bond between the biofilm and the surface (7). Allison et al. (1) showed that following extended incubation, Pseudomonas fluorescens biofilm detachment coincided with a reduction in the amount of EPS (2). Both statements could be possible explanations for the detachment of cells by NAC.
EPS is one of the major components in biofilms, and it is therefore interesting that EPS production is significantly reduced for many strains in the presence of NAC, even at concentrations at which the growth is not affected (Fig. 2). We suggest that NAC reduces EPS production directly or indirectly since no degradation of EPS by NAC was detected. NAC reduced the EPS production in cells that were incubated in the absence of energy and nutrient sources as well (Fig. 3). This has some importance for the possible use of NAC as a biofilm control agent, because in most environments, both natural and man-made, nutrient limitation is the rule (6, 16).
The binding specificity of lectins for simple sugars is a specific way to quantify EPS, and WGA was chosen because it binds to saccharide residues that are frequently encountered in bacterial EPS (9). In the absence of NAC K. pneumoniae produced polysaccharides that were tightly bound to the cells in a capsule-like structure (CPS) and also in a form that was excreted from the cells (EPS) (Fig. 8a). The presence of NAC excluded EPS (Fig. 8b). It is possible that the EPS composition changed as a result of cell exposure to NAC, so that the lectin no longer recognized the EPS. However, the reduction in the amount of EPS that we observed with the lectin was also seen in the EPS assay. Therefore, the simplest explanation of the results seems to be that the total amount of EPS was reduced. Moreover, K. pneumoniae was unable to form large microcolonies; only single cells and small microcolonies were present, which changed the texture of the biofilm formed. At this time, it is not possible to offer a detailed explanation for this, but we speculate that CPS and EPS are composed of similar polysaccharides and that reduced production of the polysaccharides allows only a small amount of polysaccharide, which accumulates on the cell surface rather than being released into the surroundings. Alternatively, the polysaccharides may be different and therefore may be affected differently by NAC. Whatever the explanation, the effect is significant since cells exposed to NAC which possess only CPS are not able to form the extensive EPS-embedded large surface colonies present in the control (Fig. 8a). Danese et al. found that colanic acid is required not for surface attachment but rather for the formation of the complex three-dimensional structure of an E. coli biofilm (15).
The regulation of EPS production is known in detail in only a few cases, such as the production of alginate by Pseudomonas aeruginosa (35, 39), but most often the possible environmental cues for regulation of these important polysaccharides are not known. The reduction in the amount of EPS in the presence of NAC may have many explanations. The more direct effects of NAC include a possible reaction of its sulfydryl group with disulfide bonds in enzymes involved in EPS production or excretion, which renders these molecules less active, or competitive inhibition of cysteine utilization. Also, the possibility of interference of NAC with control or signaling systems that direct the EPS production at translation or at the enzymatic level cannot be excluded. The fact that NAC is an antioxidant may have indirect effects on cell metabolism and EPS production. Investigations aimed at elucidating the cellular responses to NAC of some of these bacterial strains by a proteomic approach are under way in our laboratory.
We propose that the reduction of bacterial adhesion to stainless steel by NAC is chemical as well as biological. Our results show that the presence of NAC increases the wettability of surfaces. Moreover, NAC detached bacteria that were adhering to steel surfaces. Growth of the various bacteria, as monocultures or in a multispecies community, was inhibited at different concentrations of NAC. We also found that there was no detectable degradation of EPS by NAC, indicating that NAC reduced the production of EPS in most bacteria tested, even at concentrations at which growth was not affected. Altogether, the presence of NAC changes the texture of the biofilm formed and makes NAC an interesting candidate for use as a general inhibitor of bacterial biofilm formation on stainless steel surfaces.
|
ACKNOWLEDGMENTS |
|---|
|
FOOTNOTES |
|---|
|
REFERENCES |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| J. Bacteriol. | Microbiol. Mol. Biol. Rev. | Eukaryot. Cell | All ASM Journals |
|---|
, 0.5 mg of NAC ml-1; 
, 1.0 mg of NAC ml-1; 
, 2.0 mg of NAC ml-1. The growth of A. lwoffii (a), A. baumannii (b), E. cloacae (c), K. pneumoniae (d), P. mendocina (e), a multispecies community consisting of A. baumannii, P. mendocina, K. pneumoniae, A. lwoffii, E. cloacae,S. warneri, and Bacillus sp. (f), Bacillus sp. (g), B. cereus (h), B. megaterium (i), B. subtilis (j), and S. warneri (k) was examined. The error bars indicate the standard deviations of the means for three experiments.
