Scientific Publications - Work Done by Microbiology Reader Bioscreen C

Journal of Applied Microbiology, February 2004, Volume 96 Issue 2 Page 244-253

The synergistic effect  of EDTA/antimicrobial combinations  on Pseudomonas aeruginosa

ABSTRACT

Aims: To demonstrate that the nonlinear concentration-dependent inhibition of Pseudomonas aeruginosa to EDTA can be used to successfully model and predict the potentiation of antimicrobials by EDTA.

Methods and Results: A model used successfully to describe the concentration-dependent inhibition of bacterial growth caused by many antimicrobials was unable to describe the inhibition of P. aeruginosa by EDTA. Examination of the inhibition profiles for EDTA against P. aeruginosa revealed a biphasic inhibitory pattern suggesting different mechanisms of action at different concentrations. A modelled, two-stage inhibitory process was shown to fit the observations. This model was then used to examine the effect of combining EDTA with other antimicrobials. The apparent synergy of mixtures of EDTA with quaternary ammonium surfactants (QAC) and specific antibiotics was successfully modelled. Minimum inhibitory concentrations (MIC) of the QAC and that of oxacillin and cefamandole were reduced by a factor of 3-10, whereas ampicillin was reduced by a factor of 70 from an MIC of 1524 to 21 mg l ¹ in the presence of 500 mg l ¹ of EDTA.

Conclusions: A nonlinear concentration-dependent inhibition of P. aeruginosa by EDTA gives rise to apparent observation of synergy with other antimicrobials.

Significance and Impact of the Study:This is a further example where the current methodology for the examination of antimicrobial synergy (the summed fractional inhibitory concentrations) leads to false conclusions.

INTRODUCTION

The metal chelator EDTA is not recognized as an important antimicrobial agent in its own right. In general, EDTA is regarded as a 'potentiator' of the activity of other antimicrobial agents (Brown and Richards 1965). As such, much literature has been written on its synergistic or potentiating action with common preservatives, antibiotics and cationic surfactants, e.g. quaternary ammonium compounds (Weiser et al. 1969; Sheikh and Parker 1972; Hart 1984; Vaara 1992).

One of the recognized modes of action of EDTA is the disruption of the lipopolysaccharide structure in the outer membrane of Gram-negative bacteria. Through this disruption the membrane becomes more permeable to other agents, hence the potentiating action. If this is taken further, the action of EDTA in conjunction with lysozyme to degrade the peptidoglycan layer can result in the production of spheroplasts, in which the cell wall has been totally stripped away (MacGregor and Elliker 1958; Haque and Russell 1974a,b).

Weiser et al. (1968) reported the reversal of antibiotic resistance in strains of Pseudomonas aeruginosa through the combined use of EDTA with tetracycline, penicillin or ampicillin. For example in the presence of ca 1000 mg l ¹ of EDTA, the minimum inhibitory concentrations (MIC) of ampicillin was reduced from 500 to <2 mg l ¹. MacGregor and Elliker (1958) had shown similar results with quaternary ammonium surfactants (QAC) as had Voss (1967).

The observation of potentiation or synergy between two antimicrobials invariably uses a chessboard-type approach and the results are analysed using the sum of the fractional inhibitory concentrations (FIC), which is based on the MIC of the two antimicrobials (Weiser et al. 1969). The findings are often portrayed on an 'isobologram'- a chart showing contours of equivalent biological activity. On such a chart, if a straight line joins the MIC of the two agents and the MIC of mixtures of the two agents lie on this line, then combinations are considered to give a simple additive effect; if the line is concave, synergy is reported and antagonism if convex (Denyer et al. 1985).

Recently, it was reported that the FIC method was only applicable to mixtures of antimicrobials, which individually have similar dose responses (Lambert and Lambert 2003). Mixtures of antimicrobials with significantly different dose responses when analysed using the FIC method could give rise to an apparent synergy or antagonism, but which reflect a wholly additive physiological behaviour. However, one particular combination was used to highlight that 'true' synergy could exist: the EDTA/QAC combinations against P. aeruginosa. The analysis of combinations of EDTA with several quaternary ammonium and related compounds could not be reconciled with the mathematical model (the fa _comb model), which had been used successfully with other antimicrobial mixtures. As there was such a marked deviation from the expected additive effect, these observations were labelled as examples of possible 'true synergy'- essentially ratifying the observations of Brown and Richards (1965). It was noted, however, that the inhibition profile of EDTA did not fit with the Lambert-Pearson model (LPM) of inhibition (Lambert and Pearson 2000). This suggested that the mismatch between the fa _comb model, which is based on the LPM, and the observations, lay in the analysis of the EDTA inhibition/concentration profile.

A study was begun with the hypothesis that the observed synergy exhibited by mixtures of EDTA and other antimicrobials was due to the odd behaviour of the EDTA inhibition profile.

MATERIALS AND METHODS
 

Organisms

Staphylococcus aureus (ATCC 6538) or P. aeruginosa (ATCC 15442) was grown overnight in a flask containing 80 ml tryptone soya broth (TSB; Oxoid CM 129, Oxoid Basingstoke, UK) shaking at 30 C. The cells were harvested, centrifuged to a pellet, washed and resuspended in peptone. The optical density (O.D.) of a 100-fold dilution of the concentrated solution was measured, the dilution adjusted to achieve an ca O.D. = 0·5 at 600 nm. Diluted inoculum (1 ml) was added to 50 ml of TSB to give our standard inocula.

Antimicrobials

All antimicrobials (surfactants and antibiotics) were purchased from Sigma-Aldrich (Dorset, UK) and used as received. The QAC were labelled by their longest alkyl chain, e.g. dodecyltrimethyl ammonium bromide was labelled as C12QAC.

Inhibition profiles of single compounds

The technique described provides data in terms of a 'concentration profile'. Every well of the micro-titre plate was used to provide information on the inhibition observed, including sub-MIC.

The method of Lambert and Pearson (2000) was used. A stock solution of the antimicrobium under study was prepared in the growth media (TSB). Nine primary dilutions of the stock solution were prepared (dilutions of 1·0, 0·9, 0·8...0·1 of the stock). To all wells of a micro-titre plate (100 wells, 10  10) except those of column 10 were added 250  l of TSB. To the wells of column 10 were added 500  l of the primary dilutions. Using a multiwell pipette, these primary dilutions were diluted half-fold across the plate to column 2, where, after mixing, 250  l of solution were discarded. The wells of column 1 were used as the positive and negative controls. To all wells except those of the negative controls (receiving a further 50  l of TSB only) were added 50  l of the inocula.

The plate was incubated for the desired length of time (between 12 and 48 h) at 30 C with continuous shaking in the case of plates containing P. aeruginosa. O.D. measurements were performed every 10 min using a Bioscreen Microbiological Growth Analyser (Biosystems, Helsinki, Finland).

Inhibition profiles of antimicrobial mixtures

Stock solutions of the two antimicrobial agents were made up in TSB (solutions A and B). From stock solution B, a set of 10 half-fold dilutions was prepared ca 5 ml each. Micro-titre plates (100 wells, 10  10) were used in these experiments. Each well except the last column (column 10) received 200  l of TSB. Using a multi-well pipette, each well of column 10 received 400  l of the stock solution A. From each of the wells (200  l) of column 10 were removed and mixed with the 200  l of TSB in column 9. The half-fold dilution was then carried on across the plate, but without placing antimicrobial into wells 1-3 (used as controls: two positive controls and one negative control which received only 400  l of growth broth). Starting with the lowest concentration from the dilution series from stock B, 200  l was pipetted into the first row, the second lowest concentration into the second row and so on, excepting the first three wells in column 1. To each well was added 50  l of culture except well 1 (negative control). A total of 97 distinct mixtures of two antimicrobials was obtained using this method. The procedure then followed was the one given for individual inhibition profiles (above).

Data analysis

Data obtained were in the form of an array of O.D. readings per well per 10-min interval. Bespoke software operating on Microsoft Excel took the data and analysed it on the basis of the area under the O.D./time curve. The positive controls provided the maximum area under the curve; test solutions had areas ranging from those of the positive control to the negative. Test areas were listed relative to the positive controls as fractional areas (fa) against the inhibitor concentrations. This data set was then subject to a nonlinear fitting procedure carried out on the JMP statistics software package (SAS Institute, Cary, NC, USA).

Single compounds

Data were analysed using the nonlinear fitting procedure of the JMP statistics package and the profile parameters, P ₁ and P ₂ along with an indication of fit (RMSE) obtained using the LPM, eqn  (1) (Lambert and Pearson 2000).

where fa is a measurement of growth relative to a standard inoculum, x is the inhibitor concentration (mg l ¹), P ₁ and P ₂ are parameters obtained from the point of greatest slope from a plot of fa against log₁₀  x.

Antimicrobial mixtures

For a twin combination of inhibitors the following expression [fa _comb eqn  (2)] satisfies several requirements for the addition of two inhibitors, which individually can be described by eqn  (1) (Lambert and Lambert 2003).

Where C _1,1 and C _2,2 are equivalent to their respective P ₁ values obtained from eqn  (1), and where C _1,2, C _2,2 and C _Q are particular functions of the combined P ₂ values.

As in the case of the individual inhibition profile, the RMSE is used as a guide to the goodness-of-fit. Where possible the approximate standard errors, and the 95% confidence limits are also given. As a nonlinear fitting procedure is used, the normal statistical parameters are less well defined. Where the commonly used R ² term is quoted, this refers to the R ² of the observed vs calculated plot.

RESULTS

EDTA inhibition profiles

Micro-titre plates were prepared using a stock solution of EDTA and inoculated with P. aeruginosa or S. aureus. The fa concentration profiles and the inhibition parameters, obtained from the LPM, are given in Fig. 1 and Table 1, respectively. The LPM was able to fit the observed data for S. aureus whereas for P. aeruginosa a more complex inhibition profile was observed. In this latter case the simple LPM was of little utility.

Modelling the EDTA inhibition of P. aeruginosa

The inhibition profile of EDTA against P. aeruginosa appears to show a biphasic inhibition. Using the LPM gives an unsatisfactory fit as it assumes a single inhibition process. By assuming that two concentration-dependent processes occur, each of which obeys the LPM, a simple model can be produced, eqn  (3).

Parameter P _c is the EDTA concentration, which separates the two inhibitory processes. This parameter is arrived at by nonlinear regression modelling, adjusting the fit of the model to the observables until a value for P _c gives the lowest RMSE of the model to the data. The parameter r is the fa value when the concentration of EDTA = P _c . The modelled parameters are given in Table 2 and the fit of the model to the observed data is shown in Fig. 1 (a linear regression analysis gave fa _observed = 0·989 fa _calculated +  0·007; r ² = 0·997).

Hypothesis for combinations of antimicrobials with EDTA

As a working hypothesis, the observed synergy of EDTA with antimicrobials against P. aeruginosa is due to the biphasic nature of the inhibition profile of EDTA against P. aeruginosa. As such, a modified form of the combination model [eqn  (2)] taking the biphasic nature into account will allow for the prediction of the supposed synergistic action. Further, as EDTA against S. aureus does not have a biphasic nature, combinations of EDTA with antimicrobials against S. aureus will agree with the form of the fa _comb given in eqn  (2).

Combination studies of EDTA and alkyl QAC

1

1

1

1

Essentially what this model is suggesting is that the fa _comb model is applicable to each phase of the inhibition profile of EDTA against P. aeruginosa.

An isobologram contour plot using linear concentration scales for mixtures of C10QAC and EDTA is given in Fig. 2 (results of a 10  10 checkerboard arrangement). The FIC method would suggest this figure describes a substantial synergistic effect. When the EDTA concentration is 500 mg l ¹, the MIC occurs at a C10QAC concentration of ca 250 mg l ¹ giving a FIC = 0·2. Figure 3a shows the results from a 20  20 checkerboard arrangement using logarithmic concentration scales; Fig. 3b shows the fit of model [eqn  (4)] to the observed data (parameters given in Table 4).

Figure 4a,b gives the observed and calculated fa contours for mixtures of C12QAC and EDTA. A regression plot of the observed fa against the fitted fa gave a linear plot with equation fa _observed = 0·98 fa _calculated + 0·017,r ² = 0·989. Similar experiments were carried out with mixtures of C8QAC, C14QAC and C12Choline with EDTA. The linear regression lines found were for C8QAC/EDTA: fa _observed = 1·005 fa _calculated   0·006,r ²=0·995;C14QAC EDTA:fa _observed = 1·006 fa _calculated   0·01, r ² = 0·987;C12 Choline/EDTA/fa _observed=0·974fa _calculated+0·016,r ²=0·974.

Mixtures of C10QAC and C14QAC with EDTA against P. aeruginosa

Three mixtures of C10QAC and C14QAC [1 : 1, 2 : 1 and 1 : 2 ratio of C10QAC (3500 mg l ¹) to C14QAC (367 mg l ¹)] were prepared and three checkerboard arrangements with EDTA produced. Although experimentally complex, a model for the triple combination was attempted [eqn  (5)]. Using the values in Table 4 as a guide, values for P ₁, P ₃, P ₅, P ₈, P ₁₀ and P ₁₂ were fixed in the model. When the nonlinear fitting procedure was performed, values for the other parameters (P ₂, P ₄, P ₆, P ₇, P ₉, P ₁₁, P ₁₃ and P ₁₄) were obtained and an overall RMSE of 0·04 found. When all the parameters were allowed to relax (the fixed parameters were allowed to vary), the model converged with an RMSE = 0·029 (Fig. 5) shows the calculated fit to the observed.

P _c was found to be 400 mg l ¹ of EDTA. At concentrations below P _c , the MIC of C10QAC and C14QAC were calculated to be 1188 and 194 mg l ¹, respectively; in concentrations of EDTA greater than P _c the MIC of C10QAC and C14QAC were 183 and 35 mg l ¹, respectively.

These results show that the combination of the two QAC adds together with EDTA in a predictable manner.

EDTA and antibiotic combinations

FIGURES

Fig. 1 Inhibition profiles of EDTA against Staphylococcus aureus [observed , modelled eqn  (1) dashed l...

Fig. 2 Observed isobologram contour plot (linear concentration scales) for mixtures of EDTA and decylt...

Fig. 3 (a) Observed and (b) calculated isobologram contour plot (logarithmic scales) for mixtures of E...

Fig. 4 (a) Observed and (b) calculated isobologram contour plots (logarithmic scales) for mixtures of ...

Fig. 5 Plot of observed fa against fitted fa (eqn  5) for mixtures of decyltrimethyl ammonium bromide (...

Fig. 6 Observed isobologram plot (linear concentration scales) for mixtures of EDTA and oxacillin agai...

[Full Size]

Fig. 7 (a) Observed and (b) fitted isobologram three-dimensional plots (logarithmic concentration scal...

Fig. 8 Plot of observed fa against fitted fa (eqn  4) for mixtures of EDTA and ampicillin against Pseud...

Table 1 EDTA inhibition parameters obtained from the Lambert-Pearson model

Table 2 Inhibition parameters for the biphasic inhibition of Pseudomonas aeruginosa by EDTA

Table 3 Inhibition parameters for combinations of EDTA and two quaternary ammonium surfactants against...

Table 4 Inhibition parameters for combinations of EDTA with several antimicrobials against Pseudomonas...

Table 5 MIC of antimicrobials against Pseudomonas aeruginosa with and without EDTA (500 mg l ¹)

DISCUSSION

Against S. aureus the antimicrobial activity of EDTA can be adequately described by the LPM. Furthermore, combinations of QAC with EDTA against S. aureus can be described by the nonlinear additive model [eqn  (2)], which is based on the LPM. In these cases there are no indications of any synergistic activity.

Against P. aeruginosa, EDTA has a more complex inhibition-concentration profile. Synergism between EDTA and other antimicrobials have been widely reported against P. aeruginosa and Escherichia coli. In this study, the LPM was unable to give a satisfactory fit to the observed inhibition data obtained from P. aeruginosa. By assuming that the profile against P. aeruginosa was made up of two separate inhibition processes, one of which occurs below a threshold concentration and one above, a model was developed (eqn 3) which was able to accurately fit the observed data.

From an analysis of the literature it is possible that the first (low concentration) process is a general inhibitory process (e.g. the removal of metal ions from the growth medium), the second (higher concentration) may be a process destabilizing the outer membrane leading to cell lysis. In the study by Richards and Cavill (1976), low concentrations of EDTA caused the formation of convoluted surfaces, but no lysis.

As the inhibition of EDTA against P. aeruginosa did not follow the LPM, the additive model [eqn  (2)] could not be used to model the data from combinations experiments against P. aeruginosa. As EDTA shows a biphasic inhibition/concentration profile, it was suggested that on combination with other antimicrobials, the additive model [eqn  (2)] could be applied separately to each phase of the EDTA profile. The model so developed [eqn  (4)] appears to justify this suggestion.

For combinations of QAC with EDTA, the model suggests two processes are occurring: below 400 mg l ¹ of EDTA, the inhibition contours follow those expected from the additive effect of a QAC against P. aeruginosa but with EDTA having an MIC of ca 650 mg l ¹. Above 400 mg l ¹, the model suggests EDTA has an MIC of ca 8000 mg l ¹, but the MIC of the QAC has been much reduced. For example, C10QAC has an MIC against P. aeruginosa of ca 1450 mg l ¹, in the presence of concentrations of EDTA below 400 mg l ¹, the calculated MIC was 1571 mg l ¹; when EDTA was present in concentrations >400 mg l ¹, the calculated MIC for C10QAC was 277 mg l ¹ (Table 5). For C12QAC the reduction was from 168 to 14 mg l ¹. Against S. aureus the MIC of C10QAC was found to be 92 mg l ¹, whereas that of C12QAC was 7·3 mg l ¹.

The results suggest that at concentrations of EDTA above P _c , damage to the outer membrane occurs allowing other antimicrobials to act more readily against the damaged P. aeruginosa. In general, a 10-100-fold increase in QAC concentration is normally required to inhibit P. aeruginosa relative to S. aureus. In the presence of EDTA above the threshold value, the amount required to inhibit P. aeruginosa is reduced by a factor between 5 and 30. Table 5 gives the MIC of some antimicrobials in the presence of EDTA (500 mg l ¹).

It should be noted that within the model, P _c is a fixed single concentration. Although the model gives a good fit to the observed data, P _c would perhaps be better described by a concentration range over which the changes occur. Applying this suggestion to the model would give a superior fit to the observed data, but would, however, increase its complexity beyond current practicalities.

Of interest for the topical treatment of Pseudomonas infections is the combination of EDTA with C12Choline. This QAC is known as a 'soft-drug' in that the compound has a weak ester link between the fatty acid chain and the choline head group, easily cleaved by esterases (Bodor et al. 1980; Bodor 1984). Combination with EDTA above the threshold concentration increases the effectiveness of the QAC by over 10-fold in vitro.

Enhancement of antibiotic activity through synergistic combinations is much sought after (Eliopoulos and Moellering 1996; Acar 2000; Bonapace et al. (2000). With oxacillin and cefamandole a modest increase in sensitivity was obtained in the presence of EDTA (MIC changes were from 1300 to 375 mg l ¹ and from 2600 to 320 mg l ¹, respectively). These values reflect the magnitude of changes found for the QAC and further suggests that these reductions are because of the loss of the barrier function of the outer membrane. With ampicillin, however, a much larger decrease in the MIC was found (1520-21 mg l ¹). This suggests that more than just the loss of the barrier is responsible for the increase in sensitivity, possibly an enhanced uptake mechanism or the removal of inactivating factors present in the membrane or in the periplasmic space by EDTA.

The model [eqn  (4)] adequately describes the synergism described in the literature and the hypothesis put forward to explain the model also reflects current thought on the action of EDTA. From this study, the synergy of EDTA with quaternary ammonium compounds or with antibiotics against P. aeruginosa has been attributed to a discrepancy found in the inhibition/concentration profile of EDTA. The discrepancy would not have been observed if many of the current MIC methods had been employed. Using the growth-no growth criterion of MIC testing would have overlooked the partial inhibition observed by the LPM.

It has been suggested that there are two types of synergy: apparent and real (Lambert and Lambert 2003). Apparent synergy can occur through the mixing of antimicrobials with very different P ₂ values (as defined by the LPM). In the case of EDTA with the antimicrobials described herein, we suggest the label of true-synergy be kept - and suggest that further cases of true-synergy may be found, not by large-scale screening of combinations, but by looking for antimicrobials which show significant departures from the LPM, and then combining these with other antimicrobials.

The method described can be used to study the activity of more complex mixtures, but as the number of antimicrobials in a mixture increases the experimental complexity increases dramatically. A library of the inhibition parameters for individual antimicrobials can be used to predict the efficacy of combinations as has been performed for biocidal mixtures (Johnston et al. 2003; Lambert et al. 2003). However, one problem remains with the prediction of the efficacy of combined inhibitors (as opposed to the fitting of data) - the concentration exponents in the model cannot, as yet, be predicted beforehand. Greater developments in analytical microbiology may be required before enough information is obtained for a rationalization of these exponents.

The work does suggest several avenues of new research: combinations of antibiotics in the presence of EDTA concentrations above the P _c value may be of real benefit for topical applications of anti-pseudomonal drugs; EDTA (or perhaps other chelating molecules) with ampicillin or with anti-pseudomonal penicillins (e.g. piperacillin) may be beneficial in the treatment of the 80% of adults with cystic fibrosis chronically infected with P. aeruginosa. Research into the development of new biocompatible chelating agents to act as potentiators, would also seem a reasonable suggestion.

REFERENCES

•    Acar, J.F. (2000) Antibiotic synergy and antagonism. Antibiotic Therapy (part 1) 84, 1391-1406.

•    Bodor, N. (1984) Soft drugs: principles and methods for the design of safe drugs. Medicinal Research Reviews 4, 449-469.

•    Bodor, N., Kaminski, J.J. and Selk, S. (1980) Soft drugs: 1. Labile quaternary ammonium salts as soft antimicrobials. Journal of Medicinal Chemistry 23, 469-474.

•    Bonapace, C.R., White, R.L., Friedrich, L.V. and Bosso, J.A. (2000) Evaluation of antibiotic synergy against Acinetobacter baumannii: a comparison with E-test, time-kill and checkerboard methods. Diagnostic Microbiology and Infectious Diseases 38, 43-50.

•    Brown, M.R.W. and Richards, R.M.E. (1965) Effect of ethylenediamine tetraacetate on the resistance of Pseudomonas aeruginosa to antibacterial agents. Nature 20, 1391-1393.

•    Denyer, S.P., Hugo, W.B. and Harding, V.D. (1985) Synergy in preservative combinations. International Journal of Pharmaceutics 25, 245-253.

•    Eliopoulos, G.M. and Moellering, R.C. (1996) Antimicrobial combinations. In Antibiotics in Laboratory Medicine, 4th edn. ed. Lorian, V. pp. 330-396. Baltimore: Williams and Wilkins.

•    Haque, H. and Russell, A.D. (1974a) Effect of ethylenediamine tetraacetic acid and related chelating agents on whole cells of Gram-negative bacteria. Antimicrobial Agents and Chemotherapy 5, 447-452.

•    Haque, H. and Russell, A.D. (1974b) Effect of chelating agents on the susceptibility of some strains of Gram-negative bacteria to some antibacterial agents. Antimicrobial Agents and Chemotherapy 6, 200-206.

•    Hart, J.R. (1984) Chelating agents as preservative potentiators. In Cosmetic and Drug Preservation: Principles and Practice ed. Kabara, J.J. pp. 323-337. New York: Marcell Dekker.

•    Johnston, M.D., Hanlon, G.W., Denyer, S.P. and Lambert, R.J.W. (2003) Membrane damage to bacteria caused by single and combined biocides. Journal of Applied Microbiology 94, 1015-1023.

•    Lambert, R.J.W. and Lambert, R. (2003) A model for the efficacy of combined inhibitors. Journal of Applied Microbiology 95, 734-743.

•    Lambert, R.J.W. and Pearson, J. (2000) Susceptibility testing: accurate and reproducible minimum inhibitory concentration (MIC) and non-inhibitory concentration (NIC) values. Journal of Applied Microbiology 88, 784-790.

•    Lambert, R.J.W., Johnston, M.D., Hanlon, G.W. and Denyer, S.P. (2003) Theory of antimicrobial combinations: biocide mixtures - synergy or addition? Journal of Applied Microbiology 94, 747-759.

•    MacGregor, D.R. and Elliker, P.R. (1958) A comparison of some properties of strains of Pseudomonas aeruginosa sensitive and resistant to quaternary ammonium compounds. Canadian Journal of Microbiology 4, 499-503.

•    Richards, R.M.E. and Cavill, R.H. (1976) Electron microscope study of effect of benzalkonium chloride and edetate sodium on cell envelope of Pseudomonas aeruginosa. Journal of Pharmaceutical Sciences 65, 76-80.

•    Sheikh, M.A. and Parker, M.S. (1972) The influence of ethylenediaminetetraacetic acid and phenylethanol upon the fungistatic action of aminacrine hydrochloride. Journal of Pharmacy and Pharmacology 24, 118.

•    Vaara, M. (1992) Agents that increase the permeability of the outer membrane. Microbiological Reviews 56, 395-411.

•    Voss, J.G. (1967) Effects of organic cations on the Gram-negative cell wall and their bactericidal activity with ethylenediamine-tetra-acetate and surface active agents. Journal of General Microbiology 48, 391-400.

•    Weiser, R., Asscher, A.W. and Wimpenny, J. (1968) In vitro reversal of antibiotic resistance by ethylenediamine tetraacetic acid. Nature 219, 1365-1366.

•    Weiser, R., Wimpenny, J. and Asscher, A.W. (1969) Synergistic effect of edetic-acid/antibiotic combinations in Pseudomonas aeruginosa. Lancet 619-620.

(order Full Text from publisher)

© 2005 Transgalactic Ltd (manufacturer of Bioscreen C software) | Privacy Statement | P.O. Box 1393, 00101 Helsinki, Finland, phone: +358 9 85172920, fax: +358 9 8749481, e-mail: microbiology@bionewsonline.com
 

Last modified: May 25, 2005