Evidence of a charge-density threshold for optimum efficiency of biocidal cationic surfaces

R. Kügler{dagger}, O. Bouloussa and F. Rondelez

Laboratoire de Physico Chimie Curie, Institut Curie, Section de Recherche, 11 rue Pierre et Marie Curie, 75005 Paris, France

Correspondence
F. Rondelez
rondelez{at}curie.fr


   ABSTRACT
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS AND DISCUSSION
REFERENCES
 
The deposition of organic monolayers containing quaternary ammonium groups has been shown by many authors to confer biocidal properties on a large variety of solid surfaces. In a search for the controlling factors, the authors have grafted quaternized poly(vinylpyridine) chains on glass surfaces by two different methods and varied the charge density within the organic layer between 1012 and 1016 positive charges per cm2. The measurements show that this parameter has a large influence on the killing efficiency. Bacterial death occurs in less than 10 min in the quiescent state above a threshold value. The value is smaller for bacteria in the growth state. It also depends on the bacterial type. An electrostatic mechanism based on the exchange of counterions between the functionalized cationic surface and the bacterial membrane is proposed and appears consistent with the results.


Abbreviations: DMF, dimethyl formamide; OLCD, outer-layer charge density; PVP, poly(vinylpyridine); QPVP, quaternized poly(vinylpyridine); TSB, tryptophan soya broth

{dagger}Present address: Sony Deutschland GmbH, Materials Science Laboratory, Hedelfinger Str. 61, D-70327 Stuttgart, Germany.


   INTRODUCTION
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS AND DISCUSSION
REFERENCES
 
Most of the common antibiotics are designed to affect the metabolism of bacteria. However, a growing number of bacterial species show resistance to these antibiotics and cause serious health problems. A promising possibility to overcome these difficulties is the development of a concept to kill bacteria based on physical interactions.

It is quite well known that charged molecules in solution are able to kill bacteria (Endo et al., 1987; Fidai et al., 1997; Friedrich et al., 2000; Isquith et al., 1972). However, it has been realized more recently that charges attached to surfaces can kill bacteria upon contact. All bear cationic, positively charged groups, such as quaternary ammonium (Thome et al., 2003) or phosphonium (Kanazawa et al., 1993; Popa et al., 2003). Various architectures have been tested: self-assembled monolayers (Atkins, 1990; Gottenbos et al., 2002; Rondelez & Bezou, 1999), polyelectrolyte layers (Lee et al., 2004; Lin et al., 2002, 2003; Popa et al., 2003; Sauvet et al., 2000; Thome et al., 2003; Tiller et al., 2001) and hyperbranched dendrimers (Cen et al., 2003; Chen & Cooper, 2000, 2002). An important advantage of this approach is that the biocidal molecules are attached covalently to the substrates, which allows their reusability after cleaning processes and prevents uncontrolled material release to the environment. However, the key parameters of the effects involved in the biocidal process have not yet been identified.

We report here on the existence of a charge-density threshold above which bacterial death occurs quickly upon adsorption on substrates bearing cationic quaternary ammonium groups. For a given bacterium, the threshold value is different in the quiescent and the dividing states. It also depends on the bacteria type, and we have observed a difference by a factor of 10 between Escherichia coli and Staphylococcus epidermidis under high-division conditions. Our data support a lethal mechanism based on ion exchange between the bacterial membrane and the functionalized surface.


   METHODS
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS AND DISCUSSION
REFERENCES
 
Bacterial strains and media.
Staphylococcus epidermidis (ATCC 12228) and Escherichia coli (ATCC 11775) were chosen as examples of Gram-positive and Gram-negative bacteria, respectively. They were transferred from the frozen state onto agar plates (diameter 90 mm, Columbia agar+5 % sheep blood, Bio-Rad) and incubated overnight at 37 °C. One colony was then transferred to a test tube containing a liquid nutrient solution (tryptophan casein soya broth, TSB, 10 ml, Bio-Rad) at 37 °C and allowed to divide. Growth was followed for 7–8 h in the case of S. epidermidis and 5–6 h in the case of E. coli. After growth, the bacterial culture was centrifuged at 10 000 r.p.m. for 3 min and the pellet was resuspended in PBS containing 0·85 % NaCl. This washing procedure was repeated three times. The final concentration of bacteria was typically 108 per ml, as counted from the number of colony-forming units (c.f.u.) after plating on an agar medium.

Preparation of bioactive coatings and chemicals.
Polymer brushes of poly(4-vinyl-N-alkylpyridinium) bromide (QPVP) layers were formed on fused silica beads (mean diameter 1·5 mm, Patinal, Merck), planar glass slides covered with an aminosilane layer (Sigma-Aldrich) and oxidized silicon wafers (diameter 25·4 mm, one side polished, Siltronix). Two different methods of deposition were used. In the ‘grafting from’ approach (Biesalski & Rühe, 1999), the polymer chains are grown gradually from the surface by continuous addition of vinylpyridine monomers, using radical polymerization. In the ‘grafting to’ approach (Tiller et al., 2001), poly(vinylpyridine) (PVP) chains of large molecular masses are directly attached to the surface. In both cases, the quaternization step is performed after completing the grafting.

Fig. 1 shows the details of the ‘grafting to’ approach. PVP of molecular mass 1·6x105 Da was obtained commercially (Sigma-Aldrich). It was grafted on amino-terminated silanes at random positions along the chain, using a homo bifunctional cross-linking agent ({alpha},{alpha}'-dibromo-p-xylene). The attached chains were later transformed into poly(4-vinyl-N-butylpyridinium bromide) by quaternization (QPVP), using a bromo-terminated short-chain alkane (butyl).



View larger version (20K):
[in this window]
[in a new window]
 
Fig. 1. Chemical functionalization of SiO2 beads and amino-terminated glass slides by poly(4-vinyl-N-butylpyridinium) bromide polymer chains, using the ‘grafting to’ approach.

 
In the ‘grafting from’ approach, PVP chains were grown directly from the surface. The chain densities on the surface are higher than in the ‘grafting to’ process because the steric repulsions are minimized. Consequently, the charge densities can be as high as a few 1016 N+ cm–2. A similar approach has been described very recently by Lee et al. (2004).

Surface analysis.
The thickness of the organic cationic layers grafted on the silicon wafers was measured by ellipsometry (rotating analyser ellipsometer, Plasmos model SD 2300). Data were taken at nine different spots on each sample with a beam spot of 1·2 mm. The determination of the layer thickness was obtained with an accuracy of ±0·1 nm. The thickness of the oxide layer was taken to be 2·5 nm and was subtracted from the data.

The surface density of quaternary ammonium groups was measured by a colorimetric method based on fluorescent complexation and UV-VIS spectroscopy, as described by Tiller et al. (2001). Samples were first immersed in a solution of fluorescein disodium salt (1 % in distilled water) for 10 min. Due to their negative charges, the fluorescent markers bind strongly to the cationic sites and the unreacted molecules can then be removed by exhaustive washing with distilled water. The bound fluorescein molecules were then exchanged by immersing the modified samples in a small volume (9 ml) of a solution of monovalent salt (hexadecyltrimethylammonium chloride, C16H36N+Cl, 0·5 % in distilled water), and sonicated for 15 min. After adding 1 ml PBS (100 mM, pH 8), the absorbance of the resultant solution was measured at 501 nm and the concentration of fluorescein was calculated, taking a value of 77 000 M–1 cm–1 for the extinction coefficient. The density of cationic groups was then derived from this concentration: the charges measured are those corresponding to quaternary ammonium groups capable of forming an ionic complex with fluorescein.

Determination of bactericidal efficiency.
A technique based on epifluorescence microscopy was used in order to study the antimicrobial activity of the samples in situ. A 50 µl sessile droplet of a bacterial suspension in PBS (corresponding to about 5x106 cells) was deposited on the functionalized surface. The droplet diameter was 5–10 mm. The bacteria were allowed to sediment on the substrate for several minutes until a bacterial density convenient for the subsequent experiments was reached. After that, the bacteria remaining in the solution were washed away by flushing the substrate with PBS. A 1–3 µl aliquot of a mixture of two fluorescent markers (LiveDead, Molecular Probes) was then added to PBS. The adsorbed bacteria appear as green dots if still viable and as orange/red dots if their membrane has been damaged following contact with the functionalized substrate. A commercial epifluorescence microscope (DMR Leica) equipped with a 64x water-immersion objective was used for the optical observations. The images were recorded with a colour CCD camera (CoolSnap, RS Photometrics) and analysed with a computer imaging system. The presence of planktonic bacteria in the supernatant was checked by spreading an aliquot of the PBS on an agar plate and counting the number of c.f.u.

When it was necessary to study the biocidal effect of the treated wafers in high-division conditions, the bacteria were deposited on the wafers using a sessile droplet of PBS. The wafers were then quickly transferred to TSB. At the end of the incubation period, the nutrient solution was replaced by the PBS medium and the wafer was immediately observed by fluorescence microscopy. This procedure was chosen to solve the problem of having to observe bacteria in a solution containing a high concentration of proteins, and therefore turbid.


   RESULTS AND DISCUSSION
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS AND DISCUSSION
REFERENCES
 
Bactericidal effect of charged surfaces
E. coli and S. epidermidis in contact with planar surfaces.
Fig. 2 shows epifluorescence images of E. coli and S. epidermidis that have been left to sediment from a PBS solution onto oxidized silicon wafers functionalized with QPVP brushes by the ‘grafting from’ approach. The charge densities are {sigma}=1·4x1015 N+ cm–2 for E. coli (a) and {sigma}=1·7x1016 N+ cm–2 for S. epidermidis (b). Bacterial death is evident on both images taken after 5 min of contact. Similar images (not shown) were also obtained at charge densities of 3·4x1015 N+ cm–2 for E. coli. The surface treatment is thus effective to kill non-dividing bacteria after a period of contact, and the bactericidal effect exists for Gram-negative as well as Gram-positive bacteria.



View larger version (86K):
[in this window]
[in a new window]
 
Fig. 2. Fluorescence microscopy images of sessile E. coli (a) and S. epidermidis (b) cells deposited on silicon wafers and stained with a marker of viability. Bacteria were deposited by sedimentation in a PBS droplet at 20 °C and were therefore in low-division conditions. Both types of bacteria become non-viable within 10 min of contact with the charged surface, staining orange/red and displaying perforated membranes. A few viable bacteria can still be detected (green). The density of cationic quaternary ammonium groups was 1·5x1015 N+ cm–2 for E. coli and 1·7x1016 N+ cm–2 for S. epidermidis. Bar, 25 µm.

 
S. epidermidis in contact with beads.
In these experiments, inocula of S. epidermidis were put in contact with several hundred 1·5 mm diameter silica beads functionalized by the ‘grafting to’ method, and of charge density {sigma}=5x1015 N+ cm–2. Control experiments with uncharged beads ({sigma}=0 N+ cm–2) were also systematically performed. The beads were first placed in a beaker, which was filled with the inoculum until they were just covered with fluid. The minimum volume required was 2 ml and the total surface of contact was a few tens of cm2. The bead : bacteria ratio was varied by adding a 100 % excess volume of inoculum. Bacteria adsorbed on the beads after a period of incubation were observed by epifluorescence microscopy. The number of viable bacteria in the solution was separately measured by transferring aliquots of known volumes to agar plates and counting the c.f.u.

Fig. 3 shows images of S. epidermidis bacteria adsorbed on the surface of non-functionalized and functionalized beads after a contact time of 5·5 h in a nutrient TSB solution at 37 °C. It can be seen that only few bacteria are present when the surface is uncharged, and they are of green colour (a). On the charged surface, numerous bacteria are present, and they are of orange/red colour (b). Morphological differences are also obvious: the orange/red bacteria have lost their distinctive spherical shape and their margins are blurred. This indicates that their membrane is damaged, resulting in the leakage of some of the cellular constituents. All these differences demonstrate that the charged beads are biocidal for bacteria in dividing conditions. The cationic surface treatment is therefore effective against bacteria in different metabolic states, in either growth or quiescent conditions. The large difference in surface coverage between the two types of bead shows that the charged surfaces exert a much stronger attraction towards the bacterial cells. We have indeed observed that the bacteria adhere firmly on such surfaces and can withstand repeated transfers of the beads from PBS to TSB media (and vice versa), in addition to extensive rinsing. This increased bonding strength is due to the electrostatic attraction between the negative charges of the bacterial membrane and the positive charges carried by the substrate.



View larger version (65K):
[in this window]
[in a new window]
 
Fig. 3. Fluorescence microscopy images of sessile S. epidermidis bacteria on silica beads with and without charges. Bacteria were left in contact with the beads in a nutrient solution (TSB) for 5·5 h at 37 °C and were therefore in dividing conditions. The cells were stained with a marker of viability. Most of the bacteria adsorbed on the beads without charges (a) are viable and stain green. Most of bacteria adsorbed on beads with a high density of quaternary ammonium cationic groups ({sigma}=5x1015 N+ cm–2) (b) are dead and stain orange/red. Bar, 25 µm.

 
The number of viable bacteria present in the solution after incubation with the beads is given in Table 1. To allow comparison between the different experimental conditions, it is expressed as the ratio of the c.f.u. counted at initial and final times. Column 2 shows the data for uncharged beads as control. Columns 3 and 4 correspond to charged beads without and with excess volume, respectively. Line 3 is for low-division conditions (saline PBS solution at 22 °C). Line 4 is for high-division conditions (nutrient TSB medium at 37 °C).


View this table:
[in this window]
[in a new window]
 
Table 1. Viability comparison for an inoculum of S. epidermidis in contact with QPVP-functionalized and non-functionalized (control) SiO2 beads

Several hundred beads of 1·5 mm mean diameter were used in each experiment. Functionalization was obtained via the ‘grafting to’ pathway of Fig. 1. Charge density was 5x1015 N+ cm–2. The experiments were performed in PBS and in nutrient solution (TSB). In either case, adding an excess of solution varied the beads to solution ratio. The initial bacterial concentrations were 6·5x106 ml–1 and 2·9x106 ml–1, respectively. The final concentration of surviving bacteria was measured by taking 10 µl aliquots of the suspensions and counting the colonies on agar plates. The multiplication factor c.f.u.final/c.f.u.initial was then calculated.

 
In non-dividing conditions, the values of the multiplication factors are extremely small, meaning that the number of viable planktonic bacteria has been drastically reduced if the beads are charged. This is clearly due to a charge effect, since the control experiments show no variation in the number of planktonic bacteria when the beads are uncharged. The bacteria are adsorbed on the beads due to the electrostatic forces, and we have already concluded from Fig. 2 that this adsorption leads to their death. The effect does not depend much on the volume of the inoculum relative to the beads volume. The reduction factor changes from 4000 to 500 when excess volume is added.

In dividing conditions, the results depend crucially on the experimental conditions. The multiplication factor changes from a low value of 2x10–2 to a large value of 20 when excess volume is added. This latter value is comparable to the 22x increase obtained in the control experiment. Obviously, the charged beads have little influence in this case: the bacteria still multiply in solution with a doubling generation time of 74 min, in agreement with the literature (Pelmont, 1993). The rationale is that growth occurs because the rate of cell division is higher than the rate of adsorption on the charged beads. Taking a cell size of 1 µm for S. epidermidis (Davis et al., 1968), one estimates from the Einstein law of diffusion (Atkins, 1990) that a bacterium submitted to Brownian motion has a low probability to encounter a bead during one cell cycle if it is located more than 100 µm away. As the solution volume gets larger, this effect becomes more prominent. This result provides clear evidence that the bacteria become non-viable only after they have adsorbed on the charged surface. The possibility that bacterial death could result from molecules leaching out from the surface was eliminated by separate experiments in which bacteria were added to nutrient solutions that had previously contained functionalized beads. Regular growth was always observed.

Influence of charge density on cell mortality for S. epidermidis
To investigate the interplay between cell division, substrate charge density and cell mortality, we performed a series of experiments on silicon wafers with charge densities {sigma} in the range of 1012–1016 N+ cm–2. This was achieved by varying the QPVP layer thickness through two preparation factors: the graft density and the length of the polymer chains. Depending on the method of preparation, the layer thickness was between 2 and 200 nm. In some instances, self-assembled monolayers and multilayers of aminated trimethoxy- and trichlorosilanes were also grafted to the surface to get charge densities of 1013 N+ cm–2. Interestingly, no simple relationship was observed between the thickness and the charge density, which proves that not all the charges present in the organic layer are detected by the fluorescein complexation method. The backbone of the polymer chains is hydrophobic and that hinders the penetration of the ionic fluorescein probes. This effect is obviously more important for the thickest organic layers.

The results obtained for S. epidermidis after 10 min contact in low-division conditions (PBS at 20 °C and 37 °C, or TSB at 20 °C), and after an incubation time of 5·5 h in high-division conditions (TSB at 37 °C) are summarized in Table 2. In the growth phase, cell death is observed if the charge density is 1014 N+ cm–2 or greater. In the stationary phase, rapid cell death is only observed if the charge density is 1016 N+ cm–2, but not if it is 1015 N+ cm–2 or lower. This data shows i) that the biocidal activity of the polyelectrolyte layers requires a minimum charge density on the substrate for instant killing and ii) that this threshold depends on the bacterial metabolism. All the experiments were repeated two to three times on a given substrate to check repeatability and also to prove that the biocidal effect is still present after repeated exposure to bacterial inocula.


View this table:
[in this window]
[in a new window]
 
Table 2. Viability of S. epidermidis adsorbed on silicon wafers with different charge densities under low- and high-division conditions

For each range of charge density {sigma}, experiments were performed on at least two different samples and/or repeated several times on the same wafer. The percentage of dead cells was determined by epifluorescence microscopy after 10 min of contact for the low-division conditions and after overnight incubation for the high-division conditions. The number of bacterial cells in the field of view of 150 µm by 100 µm was close to 950. The first row of results represents the control experiment, using an uncharged silicon wafer. OLCD, outer-layer charge density.

 
If the charge density is lower than the threshold, cell death can still occur, but only after a period of incubation. As shown in Table 1, S. epidermidis cells in low-division conditions are killed by silica beads with a charge density of 5x1015 N+ cm–2 after 2 h contact. We have not attempted a systematic study of the death rate as a function of the charge density. It seems that it varies sharply with surface density and we have been unable to control this parameter well enough with our current samples. Cen et al. (2003), working with cellulose fibres, have also reported a dramatic increase in the killing rate of E. coli when the density of grafted pyridinium groups is changed from 2·4 to 15 nmol cm–2. This would indicate a charge-density threshold between 1·4x1015 and 9x1015 N+ cm–2 under their experimental conditions.

Ion-exchange mechanism for the biocidal effect
Many groups have now observed the antibacterial effect of cationic surfaces on Gram-positive as well as Gram-negative bacteria. This suggests that the mechanism is not system-specific, contrary to that which is generally the case with antibiotics. We surmise, as already hinted by others (Tiller et al., 2001), that the death process involves electrostatics and is related to the high density of cationic charges present on the surface. The bacterial membranes possess a large number of negative charges and it is therefore natural that they adsorb on cationic substrates. However, this cannot be the complete explanation, since it does not explain the existence of a charge-density threshold above which the death process is extremely rapid.

We are helped in the search for a new lethal mechanism by recent advances in the understanding of the electrostatic interactions between polyelectrolyte chains and oppositely charged surfaces (Park et al., 1999; Record et al., 1978). It has been gradually realized that adsorption in such cases is driven by the release in solution of the counterions initially confined within the respective electrical double layers. For instance, the many low-valency counterions that surround the polyelectrolyte in solution are no longer necessary to ensure electroneutrality if they are replaced by the charges present on the substrate. Their release leads to a large entropy increase that more than offsets the loss of entropy associated with polymer immobilization. The same process applies to bacteria, which can be crudely considered as large two-dimensional polyelectrolytes. Upon adsorption on a cationic solid substrate, the electrostatic compensation of the negative charges of the bacterial envelope is provided by the cationic charges of the substrate, and the bacteria lose their natural counterions.

We propose that this counterion release initiates bacterial death. In the case of Gram-negative bacteria such as E. coli, stabilization of the outer membrane is provided by divalent cations that neutralize and bridge the phosphate groups of the lipopolysaccharide molecules, which otherwise would strongly repel each other (Nikaido & Nakae, 1993). If such Mg2+ and Ca2+ ions are expelled during the adsorption of the bacteria on the charged substrate, the outer membrane is destabilized, leading to non-viable cells. Similar effects have been described for bacteria in solution upon addition of complexing agents such as EDTA (Brass, 1986). Such a model is also compatible with our observation that the charge density required for efficient killing is smaller for dividing cells, since their envelope is more fragile and sensitive to external disturbance.

Our experiments show that charged surfaces also have a biocidal effect on Gram-positive bacteria such as S. epidermidis. Although the observations are qualitatively similar to the ones for Gram-negative bacteria, it is unclear if the mechanism of counterion exchange leading to membrane disruption can be invoked here as well. The structure of the Gram-positive bacterial envelope is different, and it is probably significant that a much larger charge density is required to induce bacterial death in dividing conditions. In a recent paper (Lin et al., 2003), it has been suggested that grafted polyelectrolyte chains can penetrate within the bacterial cell if their molecular mass is high enough. This would provide another mechanism lethal to bacteria.

Electrostatic interactions between the charged substrate and the bacteria
So far, we have discussed our results in terms of the charge density measured by the fluorescein complexation method. However, it is not the most relevant parameter, since bacterial cells most probably interact only with the outer part of the polymer layer. First, they are too bulky to penetrate deeply into the dense grafted organic layer. Second, if one assumes that the interactions between the bacteria and the charged substrate are electrostatic in nature, the characteristic distance of interaction is very small. Ionic effects are no longer effective beyond the so-called Debye–Hückel screening length, {kappa}–1. Its value is controlled by the ionic strength of the solution and is of the order of 1 nm in both PBS and TSB media (NaCl concentrations are 137 and 178 mM, respectively). That means that electrostatic interactions are only felt at very short distances, and a more correct picture would be of a bacterium interacting with the charges contained in the first nanometer of the polyelectrolyte layer.

One can get a rough estimate of this ‘outer-layer charge density’ (OLCD) by considering that the charges are homogeneously distributed throughout the polymer layer. In that case, the OLCD over a 1 nm thickness is simply the charge density measured by the fluorescein staining method normalized by the ellipsometric thickness (in nanometers). Taking the different polymer layers discussed in Table 2, we obtain an OLCD of 1013 N+ cm–2 for the layers active against S. epidermidis in the dividing state and 1014 N+ cm–2 in the quiescent state. Experiments with E. coli are presented in Table 3. An OLCD of 1012 N+ cm–2 is sufficient to kill E. coli in the dividing state, whereas an OLCD of 1014 N+ cm–2 is required in the quiescent state.


View this table:
[in this window]
[in a new window]
 
Table 3. Viability of E. coli adsorbed on silicon wafers with different charge densities under low- and high-division conditions

For each range of charge density {sigma}, experiments were performed on at least two different samples and/or repeated several times on the same wafer. The percentage of dead cells was determined by epifluorescence microscopy after 10 min of contact for the low-division conditions and after overnight incubation for the high-division conditions. The first row of results represents the control experiment, using an uncharged silicon wafer. OLCD, outer-layer charge density.

 
It is interesting to compare this OLCD to the density of negative charges on the bacterial envelope. Bacteria possess in the order of 105 charges according to the literature (Poortinga et al., 2002). Taking an outer surface of 1 µm2 as typical, we obtain a density of 1013 N+ cm–2. This is in the same range as the OLCD values necessary for efficient killing of S. epidermidis. The solid substrate becomes biocidal when its number of cationic sites is large enough to replace all the small counterions surrounding the bacteria in solution.

To go beyond these rough estimates, it would be necessary to measure precisely the density of charges present at the extreme surface of the polymer layer. Our OLCD is an underestimate, since it is based on the fluorescein data for the total charge density and assumes that it is homogeneously distributed throughout the layer. X-Ray photoelectron spectroscopy can probe the first few atomic layers of a substrate and should allow direct measurements of the OLCD values, at least on a relative basis between different polymer layers. In preliminary experiments on self-assembled monolayers of long-chain silanes with a terminal amino group, we have been able to detect cationic charge densities as low as a few 1013 N+ cm–2 in a 2 nm thick organic layer.

CONCLUSION
In conclusion, our experiments provide evidence for the importance of the charge density of cationic substrates to the induction of bacterial death. The effect occurs for Gram-positive as well as Gram-negative bacteria after they adsorb on the functionalized substrate, and depends on the metabolic state. In low-division conditions, it is necessary to have an OLCD of 1014 N+ cm–2 for E. coli and S. epidermidis. In high-division conditions, the density required is lower, of the order of 1013 N+ cm–2 for S. epidermidis and 1012 N+ cm–2 for E. coli. We propose that the removal of divalent counterions from the bacteria during adsorption on charged surfaces induces disruption of the bacterial envelope and non-viability. Although other mechanisms may also contribute, especially in the case of Gram-positive bacteria, we believe that electrostatics is important in the search for controlling bacterial proliferation by non-chemical methods. Highly charged cationic surfaces can provide a substitute to antibiotics for bacteria that have developed resistant strains.


   ACKNOWLEDGEMENTS
 
The authors wish to thank J. Rühe, O. Prucker and H. Zhang (Institute for Microsystem Technology, Freiburg, Germany) for the preparation of the quaternized PVP brushes, and L. Hirschbein for many helpful and interesting discussions.


   REFERENCES
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS AND DISCUSSION
REFERENCES
 
Atkins, P. W. (1990). Physical Chemistry. New York: W. H. Freeman & Company.

Biesalski, M. & Rühe, J. (1999). Preparation and characterization of a polyelectrolyte monolayer covalently attached to a planar solid surface. Macromolecules 32, 2309–2316.[CrossRef]

Brass, J. M. (1986). The cell envelope of Gram-negative bacteria: new aspects of its function in transport and chemotaxis. Curr Top Microbiol Immunol 129, 1–92.[Medline]

Cen, L., Neoh, K. G. & Kang, E. T. (2003). Surface functionalization technique for conferring antibacterial properties to polymeric and cellulosic surfaces. Langmuir 19, 10295–10303.[CrossRef]

Chen, C. Z. & Cooper, S. L. (2000). Recent advances in antimicrobial dendrimers. Adv Materials 12, 843–846.[CrossRef]

Chen, C. Z. & Cooper, S. L. (2002). Interactions between dendrimer biocides and bacterial membranes. Biomaterials 23, 3359–3368.[CrossRef][Medline]

Davis, B. D., Dulbecco, R., Eisen, H. N., Ginsberg, H. S. & Wood, W. B. (1968). Staphylococci. In Microbiology, chapter 24, p. 728. New York, Evanston & London: Harper and Row.

Endo, Y., Tani, T. & Kodama, M. (1987). Antimicrobial activity of tertiary amine covalently bonded to a polystyrene fiber. Appl Environ Microbiol 53, 2050–2055.[Medline]

Fidai, S., Farer, S. W. & Hancock, R. E. (1997). Interaction of cationic peptides with bacterial membranes. Methods Mol Biol 78, 187–204.[Medline]

Friedrich, C. L., Moyles, D., Beverige, T. J. & Hancock, R. E. W. (2000). Antibacterial action of structurally diverse cationic peptides on gram-positive bacteria. Antimicrob Agents Chemother 44, 2086–2092.[Abstract/Free Full Text]

Gottenbos, B., van der Mei, H. C., Klatter, F., Nieuwenhuis, P. & Busscher, H. J. (2002). In vitro and in vivo antimicrobial activity of covalently coupled quaternary ammonium silane coatings on silicone rubber. Biomaterials 23, 1417–1423.[CrossRef][Medline]

Isquith, A. J., Abbott, E. A. & Walters, P. A. (1972). Surface-bonded antimicrobial activity of an organosilicon quaternary ammonium chloride. Appl Microbiol 24, 859–863.[Medline]

Kanazawa, A., Ikeda, T. & Endo, T. (1993). Polymeric phosphonium salts as a novel class of cationic biocides. III. Immobilization of phosphonium salts by surface photografting and antibacterial activity of the surface-treated polymer films. J Polym Sci Part A Polym Chem 31, 1467–1472.[CrossRef]

Lee, S. B., Koepsel, R. R., Morley, S. W., Matyajaszewski, K., Sun, Y. & Russell, A. J. (2004). Permanent, non-leaching antibacterial surfaces. 1. Synthesis by atom transfer radical polymerization. Biomacromolecules 5, 877–882.[CrossRef][Medline]

Lin, J., Qiu, S., Lewis, K. & Klibanov, A. M. (2002). Bactericidal properties of flat surfaces and nanoparticules derivatized with alkylated polyethyleneimines. Biotechnol Prog 18, 1082–1096.[CrossRef][Medline]

Lin, J., Qiu, S., Lewis, K. & Klibanov, A. M. (2003). Mechanism of bactericidal and fungicidal activities of textile covalently modified with alkylated polyethyleneimine. Biotechnol Bioeng 83, 168–172.[CrossRef][Medline]

Nikaido, H. & Nakae, T. (1993). The outer membrane of Gram-negative bacteria. In Bacillus subtilis and Other Gram-positive Bacteria, pp. 163–250. Edited by A. Sonenshein, J. Hoch & R. Losick. Washington, DC: American Society for Microbiology.

Park, S. Y., Bruinsma, R. F. & Gelbart, W. M. (1999). Spontaneous overcharging of macro-ion complexes. Europhys Lett 46, 454–460.[CrossRef]

Pelmont, J. (1993). La culture des bactéries. In Bactéries et Environnement, chapter 11, p. 258. Grenoble: Presses Universitaires de Grenoble.

Poortinga, A. T., Bos, R., Norde, W. & Busscher, H. J. (2002). Electric double layer interactions in bacterial adhesion to surfaces. Surface Sci Rep 47, 1–32.[CrossRef]

Popa, A., Davidescu, C. M., Trif, R., Ilia, Gh., Iliescu, S. & Dehelean, Gh. (2003). Study of quaternary "onium" salts grafted on polymers: antibacterial activity of quaternary phosphonium salts grafted on "gel-type" styrene–divinylbenzene copolymers. React Funct Polym 55, 151–158.[CrossRef]

Record, M. T., Anderson, C. F. & Lohman, T. M. (1978). Thermodynamic analysis of ion effects on the binding and conformational equilibria of proteins and nucleic acids: the roles of ion association or release, screening, and ion effects on water activity. Q Rev Biophys 11, 103–178.[Medline]

Rondelez, F. & Bezou, P. (1999). Des surfaces antibactériennes et autostériles. Actual Chim 10, 4–8.

Sauvet, G., Dupond, S., Kazmierski, K. & Chojnowski, J. (2000). Biocidal polymers active by contact. V. Synthesis of polysiloxanes with biocidal activity. J Appl Polym Sci 75, 1005–1012.[CrossRef]

Thome, J., Holländer, A., Jaeger, W., Trick, I. & Oehr, C. (2003). Ultrathin antibacterial polyammonium coatings on polymer surfaces. Surface Coating Technol 174–175, 584–587.

Tiller, J. C., Liao, C., Lewis, K. & Klibanov, A. M. (2001). Designing surfaces that kill bacteria on contact. Proc Natl Acad Sci U S A 98, 5981–5985.[Abstract/Free Full Text]

Received 28 July 2004; revised 28 December 2004; accepted 4 January 2005.



This Article
Abstract
Full Text (PDF)
Alert me when this article is cited
Alert me if a correction is posted
Services
Email this article to a friend
Similar articles in this journal
Similar articles in PubMed
Alert me to new issues of the journal
Download to citation manager
Google Scholar
Articles by Kügler, R.
Articles by Rondelez, F.
Articles citing this Article
PubMed
PubMed Citation
Articles by Kügler, R.
Articles by Rondelez, F.
Agricola
Articles by Kügler, R.
Articles by Rondelez, F.


HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
INT J SYST EVOL MICROBIOL MICROBIOLOGY J GEN VIROL
J MED MICROBIOL ALL SGM JOURNALS
Copyright © 2005 Society for General Microbiology.