Chelator-Induced Dispersal and Killing of Pseudomonas aeruginosa Cells in a Biofilm†
http://aem.asm.org/content/72/3/2064.full
ABSTRACT
Biofilms consist of groups of bacteria attached to surfaces and encased in a hydrated polymeric matrix. Bacteria in biofilms are more resistant to the immune system and to antibiotics than their free-living planktonic counterparts. Thus, biofilm-related infections are persistent and often show recurrent symptoms. The metal chelator EDTA is known to have activity against biofilms of gram-positive bacteria such as Staphylococcus aureus. EDTA can also kill planktonic cells of Proteobacteria like Pseudomonas aeruginosa. In this study we demonstrate that EDTA is a potent P. aeruginosa biofilm disrupter. In Tris buffer, EDTA treatment of P. aeruginosa biofilms results in 1,000-fold greater killing than treatment with the P. aeruginosa antibiotic gentamicin. Furthermore, a combination of EDTA and gentamicin results in complete killing of biofilm cells. P. aeruginosa biofilms can form structured mushroom-like entities when grown under flow on a glass surface. Time lapse confocal scanning laser microscopy shows that EDTA causes a dispersal of P. aeruginosa cells from biofilms and killing of biofilm cells within the mushroom-like structures. An examination of the influence of several divalent cations on the antibiofilm activity of EDTA indicates that magnesium, calcium, and iron protect P. aeruginosa biofilms against EDTA treatment. Our results are consistent with a mechanism whereby EDTA causes detachment and killing of biofilm cells.
----
The metal chelator EDTA has been shown to cause lysis, loss of viability, and increased sensitivity of planktonic Proteobacteria to a variety of antibacterial agents (reference 13; reviewed in references 25, 29, and 40). This has led to the use of EDTA as a preservative in many products. Little is known about the influence of EDTA on biofilms of Proteobacteria. Raad et al. (32, 33) have shown that EDTA combined with minocycline is an effective treatment for microorganisms embedded in biofilms on catheter surfaces. Their studies focused on Staphylococcus epidermidis, Staphylococcus aureus, and Candida albicans; however, they also reported two cases of P. aeruginosa-infected catheters where the EDTA-minocycline treatment caused a large decrease in the number of viable biofilm cells (32). Recently, Kite et al. (23) reported that tetrasodium EDTA could be used to eradicate biofilms on catheters. Ayres et al. (3) have examined the effects of permeabilizing agents on antibacterial activity against a P. aeruginosa biofilm grown on a metal disk. Their results further suggest increased anti-P. aeruginosa biofilm activity for several antibiotics when combined with EDTA (3).
We have further characterized the activity of EDTA against P. aeruginosa biofilms. We show that EDTA treatment of Pseudomonas biofilms results in 1,000-fold greater killing than treatment with gentamicin, an antibiotic commonly used to treat P. aeruginosa infections. Furthermore, a combination of EDTA and gentamicin can result in eradication of P. aeruginosa in our model biofilms. We present evidence that, in addition to killing, EDTA causes a rapid dispersion of P. aeruginosa cells from biofilms. Our data suggest that magnesium, calcium, and iron are involved in P. aeruginosa biofilm maintenance.
RESULTS
EDTA enhances loss of P. aeruginosa biofilm-associated cells.We examined the reduction in biofilm cells in response to EDTA treatment. Biofilms grown in a spinning disk reactor were exposed to various concentrations of EDTA, gentamicin, or both, and cell viability was measured (Fig. 1). In PBS, 50 mM EDTA reduced the number of biofilm-associated cells by >99%, while gentamicin (50 μg per ml, a concentration well over 10 times higher than the MIC for planktonic P. aeruginosa PAO1) caused a reduction of <10% in the number of biofilm cells. Furthermore, treatment with EDTA (50 mM) and gentamicin (50 μg per ml) together was more effective than EDTA alone (Fig. 1). Because Tris and EDTA can work synergistically to permeabilize planktonic Proteobacteria (25), we examined the effect of EDTA on P. aeruginosa biofilms in Tris buffer and found that all treatments were more effective in Tris buffer than in PBS (Fig. 1). In fact, the combination of EDTA (50 mM) and gentamicin (50 μg/ml) in Tris buffer completely eradicated biofilm-associated cells.
Effect of EDTA on P. aeruginosa biofilm structure.To examine the effect EDTA has on P. aeruginosa biofilm architecture, we used a flow cell system. As expected, P. aeruginosa biofilms developed mushroom-like structures in our experiments (Fig. 2). Addition of 50 mM EDTA resulted in preferential killing of cells inside the mushroom-like structures. The structures remained intact, as indicated by a shell of green fluorescent cells around the edge. Dead cells stained with propidium iodide (red) within the shell (Fig. 2). We chose 50 mM EDTA based on previous experiments (Fig. 1), which indicate that this is close to the minimal EDTA concentration for maximal killing of P. aeruginosa.
EDTA induces dispersal of cells from biofilms. Imaging of bacteria in flow cell biofilms over time suggested that EDTA caused not only killing but also dispersal of cells from the biofilm. To further examine the EDTA effect, we collected time lapse images by CSLM (see movies S1 to S3 in the supplemental material), and we determined total and viable cell numbers in the flow cell effluent at 10-min intervals (Fig. 3). Whereas an untreated flow cell showed a constant level of viable, dispersed cells in the effluent, an increase in the number of cells in the effluent was detected 50 min after addition of EDTA to the medium reservoir, corresponding to the time at which EDTA reached the flow cell (Fig. 3, top). This increase in the number of cells in the effluent correlated with a dispersion of green fluorescence (i.e., cells) in the flow cell as observed by CSLM (Fig. 3, green channel, 50 to 90 min). After 90 min, the cell numbers in the effluent decreased. We believe this decrease is due to washout of detached cells from the flow cell and EDTA killing of cells, as can be seen by a difference of approximately 100-fold between the direct and viable counts (Fig. 3, top). The interior of the biofilm is affected at this time as dying cells lose their GFP and the propidium iodide staining is exposed. Here we used a low concentration of propidium iodide (4 μM versus 30 μM for the dead-cell staining) that stains the extracellular matrix but not P. aeruginosa cells (4, 37). With further incubation in the presence of EDTA, the decrease in cell numbers progresses and the internal regions of the mushroom-like biofilm structures become devoid of viable cells (Fig. 3, 150 min).
Role of divalent cations in detachment and lysis of biofilms.Previous work on EDTA-treated planktonic P. aeruginosa focused on cell lysis. Our results suggest that treatment of biofilms with EDTA facilitates two processes: cell detachment and killing (Fig. 2 and 3). To gain insight into which divalent cations are involved in cell detachment and lysis, we utilized the difference in the stability constant [log(Kc)] that EDTA has for various divalent cations (barium [Kc = 7.78], magnesium [Kc = 8.83], calcium [Kc = 10.61], and iron [Kc = 25.0]) (35). We reasoned that any EDTA effect blocked by the addition of a specific cation (at a concentration that will completely saturate EDTA) is due to the role that cation (or a cation for which EDTA has lower affinity) plays in stabilizing the biofilm. In flow cell experiments, the addition of barium had no effect on EDTA-mediated killing and detachment. Addition of magnesium appears to block killing (GFP is not lost in the internal region of the mushroom), but some detachment is evident (Fig. 4). When calcium or iron is added, killing and detachment are completely blocked (Fig. 4). Similar results were obtained in spinning disk reactor experiments (Fig. 5), although in this system addition of calcium led to some detachment (40% of the cells detached and were in the planktonic state [Fig. 5, bottom]). Detachment was effectively blocked by addition of iron. This suggests that both calcium and iron may be important in stabilizing biofilms.
{...}
DISCUSSION
EDTA has a detrimental effect on the outer membrane permeability of free-living planktonic Proteobacteria (15, 25, 29, 40). By chelating divalent cations from their binding sites in lipopolysaccharide (LPS), EDTA facilitates the release of a significant proportion of LPS from the cell (26). Although prolonged treatments with EDTA are lethal, short treatments increase the permeability of the outer membrane to hydrophobic molecules (25, 29). Thus, there can be synergy between EDTA and other antibacterial agents (2, 8, 24). In this study we report that EDTA not only kills P. aeruginosa planktonic cells but also affects P. aeruginosa biofilms (Fig. 1 and 2).
Exposure of P. aeruginosa biofilms to EDTA killed P. aeruginosa cells and triggered detachment of cells from biofilms (Fig. 3 to 5). CSLM revealed that the majority of the cell population affected by the EDTA treatment resides in the inner regions of the mushroom-like structures. This type of killing or detachment pattern has been observed in P. aeruginosa biofilms exposed to various conditions (6, 34, 41). We note that sloughing of cells from the outer regions of the biofilms might also contribute to the detachment process. Chen and Stewart (9) have previously tested the abilities of various chemical treatments to remove mixed P. aeruginosa-Klebsiella pneumoniae biofilms. They reported that EDTA treatment (10 mM) resulted in a 49% reduction in cell counts, and they presented some evidence that this was due to dispersal of biofilm bacteria. The authors hypothesized that calcium was important for stabilizing the biofilm matrix (9). Other studies have also suggested a role for calcium in stabilizing biofilms of bacteria (18, 22, 39).
To better understand how P. aeruginosa biofilms are affected by EDTA treatment, we examined the abilities of different divalent cations to block EDTA-induced detachment and killing. Barium addition did not block killing, but the addition of magnesium, calcium, or iron did (Fig. 4 and 5). The relative stability constants of EDTA for the divalent cations may be ranked in ascending order as follows: barium, magnesium, calcium, and iron. Thus, our data support previous conclusions that magnesium can block lysis of planktonic P. aeruginosa by EDTA (1, 7). EDTA is thought to chelate stabilizing magnesium ions from the LPS, causing release of LPS from the outer membrane (5, 26). Magnesium did not completely block EDTA-induced detachment, but the addition of either calcium or iron did (Fig. 4 and 5). Based on previous work, one might have anticipated an involvement of iron and calcium in biofilm maintenance. In P. aeruginosa, addition of calcium to growth media increased biofilm cohesiveness, resulting in decreased detachment (38). Turakhia et al. (39) demonstrated that addition of EGTA (a calcium-specific chelator) to a mixed aerobic sewage sludge biofilm resulted in immediate detachment of cells from the biofilm. We found similar EGTA effects on detachment from the biofilm, but killing was fivefold lower than that found with EDTA (data not shown). Chen and Stewart (10) have tested the viscosity of a mixed P. aeruginosa-Klebsiella pneumoniae biofilm suspension following addition of various cations. They report that addition of iron salts significantly increased biofilm viscosity. The authors concluded that electrostatic interactions contribute to biofilm cohesion and that iron cations are potent cross-linkers of the biofilm matrix (10).
The use of EDTA to treat biofilm-related infections is being evaluated by several groups, with promising results (23, 32, 33); however, little is known about how EDTA causes increased killing of biofilm cells. The results of this study suggest that the activity of EDTA against biofilm cells is mediated by chelation of several divalent cations that are required to stabilize the biofilm matrix. Future work will be required to determine their specific role in this process. Our results imply that EDTA chelation of magnesium, calcium, and iron can enhance detachment of cells from the biofilm. EDTA also facilitates the killing of biofilm cells by chelating magnesium associated with the LPS. This dispersal process and the increased cell permeability facilitated by EDTA may also explain the enhanced killing observed in combined EDTA and gentamicin treatment (Fig. 1). This combination may have therapeutic utility.