FEMS MicrobiologyLetters69 (1990) 211-214 Publishedby Elsevier
211
FEMSLE04009
Antifungal drugs affect adherence of Candida albicans to acrylic surfaces by changing the zeta-potential of fungal cells Yoichiro Miyake 1 Tatsuhiko T s u n o d a L2, Shogo Minagi 2,. Yasumasa Akagawa 2, Hiromichi Tsuru 2 and Hidekazu Suginaka l : Department of Microbiology and : Department of Removable Prosthodontics. Hiroshima Universit~ School of Dentistry, Minami.ku. Hiroshirna. Japan
Received2 December1989 Revisionreceived30 January1990 Accepted 2 February1990 Key words: Candida albicans; Adherence; Antifungal drug; Acrylic resin; Zeta-potential
1. SUMMARY The effect of sub-inhibitory concentrations of antifungal drugs on the adherence of Candida albicans to acrylic surfaces was investigated. Among five antifungals tested, azalomycin F and aculeacin A significantly enhanced the adherence. The zeta-potential of fungal cells was affected by antifungal drugs, whereas no significant change in cell surface hydrophobicity was observed. The relationship obtained between the change in the adherence and that in zeta-potential suggests that the enhanced adherence was caused by decreased electric repulsive forces.
infections in healthy individuals. Adherence of microorganisms including C. albicans to surfaces of the host is an important step in the infection processes [1]. Since the microorganism is likely to be involved in the etiology of denture-induced stomatitis [2], its adherence to acrylic surfaces has been extensively studied. It has been reported that hydrophobic interactions [3] and the fibrous structure of the cell surface of C. albicans [4] are both involved in its adherence to acrylic surfaces. In the present study, the effect of antifungals on the adherence of C. albicans to acrylic surfaces was investigated.
3. MATERIALS AND METHODS 2. INTRODUCTION Candida aibicans often causes systemic infec-
tion in immunocompromised patients and topical Correspondence to: YoichiroMiyake, Departmentof Microbiology, Hiroshima University School of Dentistry, 1-2-3, Kasumi,Minami-ku,Hiroshima734, Japan. * Present address: Departmentof RemovableProsthodontics. Okayama UniversityDentalSchool,Okayama700, Japan.
3.1. Fungal cell preparation Candida albicans IFO 1385 was obtained from
Institute for Fermentation Osaka, Osaka, Japan. This strain is catalogued as ATCC 18804. The fungus was subcultured in Sabouraud glucose broth (2.0% glucose, 1.0% peptone, 0.5% yeast extract) and inoculated into 10 ml of fresh medium in a test tube (18 x 150 ram) with or without sub-inhibitory concentrations of antibiotics, and
0378-1097/90/$03.50© 1990Federationof EuropeanMicrobiologicalSocieties
212 then incubated at 37°C for 16 h. Fungal cells were harvested by centrifugation, washed in 0.01 M phosphate buffered saline (PBS, pH 7.4), and resuspended in PBS at a concentration of 107 cells/ml. All fungal cells grew in the yeast form with the above culture conditions.
3.2. Antibiotics Antibiotics used in this study were as follows: Aculeacin A (Toyo Jozo, Co., Ltd., Shizuoka, Japan), ampbotericin B (Squibb Japan Inc., Tokyo, Japan), miconazole (Mochida Pharmaceutical Co., Ltd., Tokyo, Japan), ketoconazole (Kyowa Hakko, Tokyo, Japan), azalomycin F (Sankyo Co., Ltd., Japan). Miconazole, ketoconazole an~ azaiomycin F were dissolved in a small amount of N,N-dimethylformamide, and added to the medium. Amphotericin B and aculeacin A were dissolved in dimethyl sulfoxide and ethanol, respectively. Minimal inhibitory concentrations (MICs) of antifungals for C. albicans IFO 1385, used in this study, were as follows: aculeacin A, 0.003/~g/ml; amphotericin B, 0.2 /~g/ml; mlconazole, 3.13 /~g/ml; ketoconazole, 25 /tg/ml; azalomycin F, 12.5/tg/ml. 3.3. Preparation of acrylic resin plates Heat-curing resin was prepared to make plates (10 by 10 mm) with a smooth surface. The plates were washed with running water for three days to remove the remaining monomer and five times with purified water, and then dried. 3.4. Adherence to acrylic resin plates Five acrylic resin plates were placed in a plastic Petri dish (60 mm in diameter and 15 mm in depth), and 7.0 ml of the fungal suspension was poured into the dish, After incubation at 37 ° C for 1 h, the acrylic resin plates were washed three times in PBS, fixed in formaldehyde and methanol, and then stained with 1 ~ crystal violet. Adherent cells in 10 high power fields (total, 4 mm 2) were counted under a microscope and totaled, The mean number of adherent cells + the standard deviation was calculated for five plates. 3.5. Hydrophobicity Fungal cell surface hydrophobicity was determined by the method of Rosenberg et al. [5].
Briefly, 500/LI of n-hexadecane was added to 2.0 ml of fungal suspension. The tube was mixed vigorously, and the optical density at 660 nm was measured before and after mixing.
3.6. Zeta-potential Zeta-potentials of fnngal cells were measured by particle microelectrophorusis in a glass cell with an inter-electrode length of 36 mm [6]. Fungal cells were resuspended in 1 mM phosphate buffer, pH 7.4, and the voltage used was 15 V. Electrophoretic mobility was measured at the upper stationary level of the glass cell where the velocity of the dispersing medium is zero. The zeta-potential (~) was calculated from electrophoretic mobility using the Smoluchowski equation: = 4¢r~u/¢E where u is the electrophoretic mobiiity, ~ is the viscosity of the liquid, ¢ is the permittivity, and E is field strength.
3.Z Statistical analysis Comparisons of significance were by unpaired Student's t-test.
4. RESULTS A N D DISCUSSION Although many studies have been conducted to investigate the effects of sub-inhibitory concentrations of antibiotics on bacterial adherence, there have been a limited number of studies of their effect on fungal adhesion. Odds and Webster [7] and Sobel and Obedeanu [8] reported the effect of azole antifungais on candidal adherence to vaginal epithelial cells, McCourtie et al. [9] reported a reduction of candidal adherence by pretreatment of a denture acrylic surface with amphotericin B. Few studies, however, have bccn conducted to demonstrate the effect of growing of fungal cells in the presence of sub-inhibitory concentrations of antifungals on the adherence to acrylic surfaces. The effect of five antifungal drugs on the adherence of C. albicans is shown in Fig. 1. When fungal cells were cultured in the presence of subinhibitory concentrations of azalomycin F or aculeacin A, fungal adherence was significantly
213
/" /"
/
•
• o/" I
i
i
i
?
e,,.6m~ • /t
i
o Antiblottc
concentrohon
(xMIC)
Fig. l. The effect of sub-inhibitory concentrations of antifungal drugs on candidal adherence to acrylic surfaces. Hatched columns represent the adherence of non-treated funsal cells. Bars represent a sta;~dard deviation. *. 0.0l < P < 0.05: **. P < 0.01.
increased. Ketoconazole slightly enhanced the adherence, but the difference was not statistically significant. Miconazole and amphotericin B hardly affected candidal adherence to acrylic surfaces. Minagi ¢t al. [3] reported the participation of hydrophobic interaction in the adherence of the same strain C albicans IFO 1385 to acrylic surfaces. The strain used in the present study was relatively hydrophilic regardless of whether the funsal cells were treated with an antifungal agent or not (Table 1), That is, no significant change in funsal cell surface hydrophobicity was observed by culturing the funsal cells in the presence of antifungals. N o change in fungal cell surface hydrophobicity was detected either by contact angle
Table I Cell surface hydrophobicity of C. albicuns with and without antifungal treatment
Antifungal drug Miconazole Ketoconazolc Azalomycin F Amphotericin B Aculeacln A
Anlibiolic concentration ( x MIC) 0 1/8 1/4 1/2 Percent adherence to hexadecane 0 0 0 0 0 3.0 0 0 0 0 0 0 0 i.9 2.0 3.0 0 0 0 1.0
-so
-~0 -30 -20 -10 Zeta-potential {my)
0
Fig. 2. Relationship between adherence and zeta-potential of candidal cells grown in the presence of sub-inhibitory concentrations of antifungal drugs. Symbols: o . control: ®.
miconazole: A. ketoconazole: A, azalomycin F: r-I, amphoteritin B: al, Aculeacin A; r = 0.769, P < 0.00l.
measurement or by the salting out method (data not shown). It is, hence, unlikely that the change in cell surface hydrophobicity resulted in the change in fungal adherence. in microbial adherence to solid surfaces, a repulsive force is involved as well as an attractive force such as a hydrophobic interaction [10,11]. Microbial cells are usually negatively charged, and electrostatic interaction acts as a repulsive force in microbial a d h e r e n c e to negatively c h a r g e d surfaces, such as mammalian cells and acrylic resin. A linear relationship between fungai adherence and zeta-potential of the cells is shown in Fig. 2. That is, the greater the adherence was, the smaller was the negative charge of funsal calls. Taking into account that the acrylic surface is characterized by a negative charge, it is suggested that a decrease in the electric repulsive force resulted in the enhanced adherence of antifungal treated cells. Since the zeta-potential is determined by an electrophoresis of a charged particle in a liquid medium, it represents the net charge of the electric double layer around the particle. Zetapotential, therefore, does not demonstrate directly the surface characteristic of the particle. Moreover, the surface of microbial cells possesses a complex structure, wifich has a role in microbial
214 adherence. That is, zeta-potential is unlikely to be involved in the direct interaction of adhesins on microbial cells a n d foreign surfaces. However, zeta-potential of the particle probably has some role in microbial adherence to a charged surface, especially by a long-range effect [11]. Aculeacin A has been reported to inhibit glucan formation of fungal cell walls [12]. This suggests that aculeacin A enhanced the adherence by altering the fungal cell wall characteristics. Azalomycin F, which enhanced candidal adherence, acts on the cytoplasmic m e m b r a n e of fungal cells [13]. Ketoconazole and miconazole, imidazoles [14], a n d amphotericin B, a polyene [15], also act on the cytoplasmic m e m b r a n e of fungal cells. These drugs, however, had limited effect on the adherence. These results suggest that azalomycin F altered fungal cell surface characteristics by acting o n the cytoplasmic membrane. Imidazoles a n d amphotericin B, on the contrary, apparently act on the cytoplasmic m e m b r a n e without influencing cell surface characteristics that may affect fungal adherence.
ACKNOWLEDGEMENT W e would like to t h a n k N o b u t o Y a n o for his helpful suggestions.
REFERENCES [1] Ofek, i. and Beachey, E.H. (1980) in Bactdal Adherence, Receptors and Recognition. Series B, vol. 6 (Beachey. E.H., ed.), pp. i-31, Chapman and Hall. London. New York. [2] Budtz-Jergensen, E., Theilade, E. and Theilade, J. (1983) Scand. J. Dent. Res. 91,134-142. [31 Minagl, S., Miyake, Y., lnagaki, K., Tsuru, H. and Suginaka. H. (1985) Infect. lmmun. 47. 11-14. [4] McCourtie, J. and Douglas, LJ. (1981) Infect. lmmun. 32, 1234-1241. [5] Rosenberg, M., Gmnick, D. and Rosenberg, E. (1980) FEMS Microbiol. Lett. 9, 29-33. [6] Plegue, T.H., Frank. S.G., Fruman, D.H. and Zakin, J.L (1986) J. Colloid lnterf. Sci. 114, 88-105. [7] Odds, F.C. and Webster, C.E. (1988) J. Antimicrob. Chemother. 22, 473-481. [8] Sob¢l, J.D. and Obedeanu, N. (1983) Eur. J. Clin. MicrobioL 2, 445-452. [9] McCourtie, J., MacFarlane, T.W. and Sarnaranayake, L.P. (1986) J. Antimicrob. Chemother. 17, 575-583. [10] Heckels"J.E., Blackett, B., Everson, J.S. and Ward. M.E. (1976) J. Gen. MicrobioL 96, 359-364. [11] Reynolds, E.C. and Wong, A. (1983) Infect. Immun. 39, 1285-1290. [12] Mizoguchi, J., Saito, T., Mizuno, K. and Hayano, K. (1977) J. AntibioL 30, 308-313. [13] Takesako, K., Nakamura, T., Obayashi, A., iwasaki. S., Namikoshi, M., Okuda, S. and Beppu, T. (1986) J. Antibiot. 39, 713-716. [14] Georgopapadakou, N.H., Dix, B.A., Smith, S.A., Freudenberger, J. and Funke, P.T. (1987) Amimicrob. Agents Chemother. 31, 46-51. [15] Lampen, J.O., Arnow, P.M. and Safferman, R.S. (1960) I. Bacteriol. 80, 200-206.
211
FEMSLE04009
Antifungal drugs affect adherence of Candida albicans to acrylic surfaces by changing the zeta-potential of fungal cells Yoichiro Miyake 1 Tatsuhiko T s u n o d a L2, Shogo Minagi 2,. Yasumasa Akagawa 2, Hiromichi Tsuru 2 and Hidekazu Suginaka l : Department of Microbiology and : Department of Removable Prosthodontics. Hiroshima Universit~ School of Dentistry, Minami.ku. Hiroshirna. Japan
Received2 December1989 Revisionreceived30 January1990 Accepted 2 February1990 Key words: Candida albicans; Adherence; Antifungal drug; Acrylic resin; Zeta-potential
1. SUMMARY The effect of sub-inhibitory concentrations of antifungal drugs on the adherence of Candida albicans to acrylic surfaces was investigated. Among five antifungals tested, azalomycin F and aculeacin A significantly enhanced the adherence. The zeta-potential of fungal cells was affected by antifungal drugs, whereas no significant change in cell surface hydrophobicity was observed. The relationship obtained between the change in the adherence and that in zeta-potential suggests that the enhanced adherence was caused by decreased electric repulsive forces.
infections in healthy individuals. Adherence of microorganisms including C. albicans to surfaces of the host is an important step in the infection processes [1]. Since the microorganism is likely to be involved in the etiology of denture-induced stomatitis [2], its adherence to acrylic surfaces has been extensively studied. It has been reported that hydrophobic interactions [3] and the fibrous structure of the cell surface of C. albicans [4] are both involved in its adherence to acrylic surfaces. In the present study, the effect of antifungals on the adherence of C. albicans to acrylic surfaces was investigated.
3. MATERIALS AND METHODS 2. INTRODUCTION Candida aibicans often causes systemic infec-
tion in immunocompromised patients and topical Correspondence to: YoichiroMiyake, Departmentof Microbiology, Hiroshima University School of Dentistry, 1-2-3, Kasumi,Minami-ku,Hiroshima734, Japan. * Present address: Departmentof RemovableProsthodontics. Okayama UniversityDentalSchool,Okayama700, Japan.
3.1. Fungal cell preparation Candida albicans IFO 1385 was obtained from
Institute for Fermentation Osaka, Osaka, Japan. This strain is catalogued as ATCC 18804. The fungus was subcultured in Sabouraud glucose broth (2.0% glucose, 1.0% peptone, 0.5% yeast extract) and inoculated into 10 ml of fresh medium in a test tube (18 x 150 ram) with or without sub-inhibitory concentrations of antibiotics, and
0378-1097/90/$03.50© 1990Federationof EuropeanMicrobiologicalSocieties
212 then incubated at 37°C for 16 h. Fungal cells were harvested by centrifugation, washed in 0.01 M phosphate buffered saline (PBS, pH 7.4), and resuspended in PBS at a concentration of 107 cells/ml. All fungal cells grew in the yeast form with the above culture conditions.
3.2. Antibiotics Antibiotics used in this study were as follows: Aculeacin A (Toyo Jozo, Co., Ltd., Shizuoka, Japan), ampbotericin B (Squibb Japan Inc., Tokyo, Japan), miconazole (Mochida Pharmaceutical Co., Ltd., Tokyo, Japan), ketoconazole (Kyowa Hakko, Tokyo, Japan), azalomycin F (Sankyo Co., Ltd., Japan). Miconazole, ketoconazole an~ azaiomycin F were dissolved in a small amount of N,N-dimethylformamide, and added to the medium. Amphotericin B and aculeacin A were dissolved in dimethyl sulfoxide and ethanol, respectively. Minimal inhibitory concentrations (MICs) of antifungals for C. albicans IFO 1385, used in this study, were as follows: aculeacin A, 0.003/~g/ml; amphotericin B, 0.2 /~g/ml; mlconazole, 3.13 /~g/ml; ketoconazole, 25 /tg/ml; azalomycin F, 12.5/tg/ml. 3.3. Preparation of acrylic resin plates Heat-curing resin was prepared to make plates (10 by 10 mm) with a smooth surface. The plates were washed with running water for three days to remove the remaining monomer and five times with purified water, and then dried. 3.4. Adherence to acrylic resin plates Five acrylic resin plates were placed in a plastic Petri dish (60 mm in diameter and 15 mm in depth), and 7.0 ml of the fungal suspension was poured into the dish, After incubation at 37 ° C for 1 h, the acrylic resin plates were washed three times in PBS, fixed in formaldehyde and methanol, and then stained with 1 ~ crystal violet. Adherent cells in 10 high power fields (total, 4 mm 2) were counted under a microscope and totaled, The mean number of adherent cells + the standard deviation was calculated for five plates. 3.5. Hydrophobicity Fungal cell surface hydrophobicity was determined by the method of Rosenberg et al. [5].
Briefly, 500/LI of n-hexadecane was added to 2.0 ml of fungal suspension. The tube was mixed vigorously, and the optical density at 660 nm was measured before and after mixing.
3.6. Zeta-potential Zeta-potentials of fnngal cells were measured by particle microelectrophorusis in a glass cell with an inter-electrode length of 36 mm [6]. Fungal cells were resuspended in 1 mM phosphate buffer, pH 7.4, and the voltage used was 15 V. Electrophoretic mobility was measured at the upper stationary level of the glass cell where the velocity of the dispersing medium is zero. The zeta-potential (~) was calculated from electrophoretic mobility using the Smoluchowski equation: = 4¢r~u/¢E where u is the electrophoretic mobiiity, ~ is the viscosity of the liquid, ¢ is the permittivity, and E is field strength.
3.Z Statistical analysis Comparisons of significance were by unpaired Student's t-test.
4. RESULTS A N D DISCUSSION Although many studies have been conducted to investigate the effects of sub-inhibitory concentrations of antibiotics on bacterial adherence, there have been a limited number of studies of their effect on fungal adhesion. Odds and Webster [7] and Sobel and Obedeanu [8] reported the effect of azole antifungais on candidal adherence to vaginal epithelial cells, McCourtie et al. [9] reported a reduction of candidal adherence by pretreatment of a denture acrylic surface with amphotericin B. Few studies, however, have bccn conducted to demonstrate the effect of growing of fungal cells in the presence of sub-inhibitory concentrations of antifungals on the adherence to acrylic surfaces. The effect of five antifungal drugs on the adherence of C. albicans is shown in Fig. 1. When fungal cells were cultured in the presence of subinhibitory concentrations of azalomycin F or aculeacin A, fungal adherence was significantly
213
/" /"
/
•
• o/" I
i
i
i
?
e,,.6m~ • /t
i
o Antiblottc
concentrohon
(xMIC)
Fig. l. The effect of sub-inhibitory concentrations of antifungal drugs on candidal adherence to acrylic surfaces. Hatched columns represent the adherence of non-treated funsal cells. Bars represent a sta;~dard deviation. *. 0.0l < P < 0.05: **. P < 0.01.
increased. Ketoconazole slightly enhanced the adherence, but the difference was not statistically significant. Miconazole and amphotericin B hardly affected candidal adherence to acrylic surfaces. Minagi ¢t al. [3] reported the participation of hydrophobic interaction in the adherence of the same strain C albicans IFO 1385 to acrylic surfaces. The strain used in the present study was relatively hydrophilic regardless of whether the funsal cells were treated with an antifungal agent or not (Table 1), That is, no significant change in funsal cell surface hydrophobicity was observed by culturing the funsal cells in the presence of antifungals. N o change in fungal cell surface hydrophobicity was detected either by contact angle
Table I Cell surface hydrophobicity of C. albicuns with and without antifungal treatment
Antifungal drug Miconazole Ketoconazolc Azalomycin F Amphotericin B Aculeacln A
Anlibiolic concentration ( x MIC) 0 1/8 1/4 1/2 Percent adherence to hexadecane 0 0 0 0 0 3.0 0 0 0 0 0 0 0 i.9 2.0 3.0 0 0 0 1.0
-so
-~0 -30 -20 -10 Zeta-potential {my)
0
Fig. 2. Relationship between adherence and zeta-potential of candidal cells grown in the presence of sub-inhibitory concentrations of antifungal drugs. Symbols: o . control: ®.
miconazole: A. ketoconazole: A, azalomycin F: r-I, amphoteritin B: al, Aculeacin A; r = 0.769, P < 0.00l.
measurement or by the salting out method (data not shown). It is, hence, unlikely that the change in cell surface hydrophobicity resulted in the change in fungal adherence. in microbial adherence to solid surfaces, a repulsive force is involved as well as an attractive force such as a hydrophobic interaction [10,11]. Microbial cells are usually negatively charged, and electrostatic interaction acts as a repulsive force in microbial a d h e r e n c e to negatively c h a r g e d surfaces, such as mammalian cells and acrylic resin. A linear relationship between fungai adherence and zeta-potential of the cells is shown in Fig. 2. That is, the greater the adherence was, the smaller was the negative charge of funsal calls. Taking into account that the acrylic surface is characterized by a negative charge, it is suggested that a decrease in the electric repulsive force resulted in the enhanced adherence of antifungal treated cells. Since the zeta-potential is determined by an electrophoresis of a charged particle in a liquid medium, it represents the net charge of the electric double layer around the particle. Zetapotential, therefore, does not demonstrate directly the surface characteristic of the particle. Moreover, the surface of microbial cells possesses a complex structure, wifich has a role in microbial
214 adherence. That is, zeta-potential is unlikely to be involved in the direct interaction of adhesins on microbial cells a n d foreign surfaces. However, zeta-potential of the particle probably has some role in microbial adherence to a charged surface, especially by a long-range effect [11]. Aculeacin A has been reported to inhibit glucan formation of fungal cell walls [12]. This suggests that aculeacin A enhanced the adherence by altering the fungal cell wall characteristics. Azalomycin F, which enhanced candidal adherence, acts on the cytoplasmic m e m b r a n e of fungal cells [13]. Ketoconazole and miconazole, imidazoles [14], a n d amphotericin B, a polyene [15], also act on the cytoplasmic m e m b r a n e of fungal cells. These drugs, however, had limited effect on the adherence. These results suggest that azalomycin F altered fungal cell surface characteristics by acting o n the cytoplasmic membrane. Imidazoles a n d amphotericin B, on the contrary, apparently act on the cytoplasmic m e m b r a n e without influencing cell surface characteristics that may affect fungal adherence.
ACKNOWLEDGEMENT W e would like to t h a n k N o b u t o Y a n o for his helpful suggestions.
REFERENCES [1] Ofek, i. and Beachey, E.H. (1980) in Bactdal Adherence, Receptors and Recognition. Series B, vol. 6 (Beachey. E.H., ed.), pp. i-31, Chapman and Hall. London. New York. [2] Budtz-Jergensen, E., Theilade, E. and Theilade, J. (1983) Scand. J. Dent. Res. 91,134-142. [31 Minagl, S., Miyake, Y., lnagaki, K., Tsuru, H. and Suginaka. H. (1985) Infect. lmmun. 47. 11-14. [4] McCourtie, J. and Douglas, LJ. (1981) Infect. lmmun. 32, 1234-1241. [5] Rosenberg, M., Gmnick, D. and Rosenberg, E. (1980) FEMS Microbiol. Lett. 9, 29-33. [6] Plegue, T.H., Frank. S.G., Fruman, D.H. and Zakin, J.L (1986) J. Colloid lnterf. Sci. 114, 88-105. [7] Odds, F.C. and Webster, C.E. (1988) J. Antimicrob. Chemother. 22, 473-481. [8] Sob¢l, J.D. and Obedeanu, N. (1983) Eur. J. Clin. MicrobioL 2, 445-452. [9] McCourtie, J., MacFarlane, T.W. and Sarnaranayake, L.P. (1986) J. Antimicrob. Chemother. 17, 575-583. [10] Heckels"J.E., Blackett, B., Everson, J.S. and Ward. M.E. (1976) J. Gen. MicrobioL 96, 359-364. [11] Reynolds, E.C. and Wong, A. (1983) Infect. Immun. 39, 1285-1290. [12] Mizoguchi, J., Saito, T., Mizuno, K. and Hayano, K. (1977) J. AntibioL 30, 308-313. [13] Takesako, K., Nakamura, T., Obayashi, A., iwasaki. S., Namikoshi, M., Okuda, S. and Beppu, T. (1986) J. Antibiot. 39, 713-716. [14] Georgopapadakou, N.H., Dix, B.A., Smith, S.A., Freudenberger, J. and Funke, P.T. (1987) Amimicrob. Agents Chemother. 31, 46-51. [15] Lampen, J.O., Arnow, P.M. and Safferman, R.S. (1960) I. Bacteriol. 80, 200-206.
