Journal of Antimicrobial Chemotherapy (1991) 27, 263-271
Effect of non-pMactam antibiotics on penicillin-binding protein synthesis of Enterococcus hirae ATCC 9790 A. Grossato", R. Sartori and R. Fontana* of
Fosfomycin, bacitracin and vancomycin in combination with penicillin exhibit a synergic effect against Enterococcus hirae ATCC 9790. This strain, when incubated in presence of the MIC of non-/?-lactam antibiotics, showed an alternated pattern of PBPs. Bacitracin and vancomycin caused a decrease in the density of all PBPs while fosfomycin only reduced that of PBP 6. It is suggested that the observed synergy is a consequence of the inhibition.of PBP synthesis by antibiotics which act on the early stages of peptidoglycan synthesis prior to the formation of cross-links.
Introduction Peptidoglycan synthesis is a complex process which can be divided into three stages. In the first step the basic monomer (the UDP-acetylmuramyl-pentapeptide) is assembled in the cytoplasm, in the second the precursor unit is bound to a membrane lipid and carried from inside the cell membrane to outside and thirdly, it is cross-linked into the pre-existing peptidoglycan layers of the cell wall. (Shockman & Barrett, 1983.) The individual steps can be inhibited by different classes of antibiotics; fosfomycin (Kalian et al., 1974); cycloserine (Roze & Strominger, 1966) inhibit some enzymes involved in monomer synthesis; bacitracin (Storm, 1974) and vancomycin (Perkins, 1969) interfere, by different mechanisms, with the second stage. /?-lactams inhibit the terminal reactions leading to cross-linking of the peptidoglycan through covalent binding to several membrane proteins (the penicillin-binding proteins or PBPs) which are known to possess carboxypeptidase and transpeptidase activities (Spratt, 1975; Waxman & Strominger, 1983). Combinations of inhibitors of the first or second step in peptidoglycan synthesis with /Mactams have been shown to be synergic (Donabedian & Andriole, 1977; Perea, Torres & Borobio, 1978; Neu & Kamimura, 1982). It has been suggested that such synergy is the result of activity exerted on different reactions within the same biochemical process (Sande & Scheld, 1980; Krogstad & Moellering, 1986) but the mechanism leading to the increased antibacterial activity has never been investigated. In the present work we have studied the effect of fosfomycin, bacitracin and vancomycin on the PBP pattern of Enterococcus hirae ATCC 9790 and found that all the drugs tested reduced the synthesis of one or more PBPs. 263 0305-7453/91/030263 + 09 $02.00/0
© 1991 The British Society for Antimicrobial Chemotherapy
Downloaded from http://jac.oxfordjournals.org/ at Dalhousie University on June 19, 2015
"Institute of Microbiology, University of Padua, 35100-Padua;bInstitute Microbiology, University of Verona, 37134-Verona, Italy
264
A. Grossato et al. Materials and methods
Bacterial strains and conditions of growth All experiments were performed with E. hirae ATCC 9790, a penicillin susceptible strain used in previous studies (Fontana et al., 1985). Bacteria were grown in streptococcal broth (SB), a yeast extract/peptone medium, at 37°C and the growth monitored at 650 nm in a Perkin-Elmer spectrophotometer. Antibiotics and chemicals
Determination of susceptibility and synergy The susceptibility of E. hirae to all agents was determined by two-fold serial dilutions in SB with a final inoculum of 106 cfu/ml. The MIC was determined by visual inspection for lack of turbidity after 18 h of incubation at 37°C without shaking. Synergy studies were performed by the chequerboard titration method. Synergy was defined as a fractional inhibitory index (FIC) of < 0-5 which was determined by the following formula: (MIC of drug A in combination with drug B/MIC of drug A alone) + (MIC of drug B in combination with drug A/MIC of drug B alone). Killing kinetics Exponentially growing cultures were diluted with fresh medium in order to obtain a starting density of 106 cfu/ml. These suspensions were divided into sets of subcultures and various concentrations of fosfomycin, bacitracin and vancomycin were added to these. At one hour intervals 0-5 ml samples were taken, suitably diluted and plated on Brain Heart (BH Difco) agar. After an incubation of 18 h at 37°C the colonies on each plate were counted and averaged. Membrane preparation Exponentially growing cells were harvested by centrifugation, and washed with ice-cold distilled water and membranes were isolated as described previously (Fontana et al., 1985). In this procedure cellular lysis was carried out in phosphate buffer (pH 7) containing 100 mg/1 lysozyme, 5 mg/1 DNAse, 2 mg/1 RNAse, 1 HIM MgCl2 (for 1 h at 37°C). PBP analysis The PBPs of E. hirae ATCC 9790 were analysed by the method previously described (Cojette et al., 1980; Fontana et al., 1985). Experiments to determine competitive binding of fosfomycin, bacitracin and vancomycin to PBPs were performed by preincubation of membranes with various concentrations of non-radioactive antibiotic at 37°C for 30 min before the addition of 14C-benzyl penicillin. To study the effect of non/Mactam antibiotics upon the PBP pattern of actively growing cells, cultures were incubated with the antibiotics for 30 min, then centrifuged and lysed in order to obtain
Downloaded from http://jac.oxfordjournals.org/ at Dalhousie University on June 19, 2015
Fosfomycin-trometamol was provided by Zambon (Milano, Italy), and penicillin and vancomycin by Eli Lilly Italia (Sesto Fiorentino, Italy). l4C-benzylpenicillin was purchased from Amersham (Buckinghamshire, England). Bacitracin and chemicals for the preparation of polyacrylamide gels were obtained from Sigma (St. Louis, USA).
265
NoD-p-lactams and E. hirae PBPs
membranes. These were then incubated with radioactive penicillin for 60 min at 37°C. All samples were processed by sodium dodecyl sulphate-polyacrylamide slab gel electrophoresis and the PBPs were detected by fluorography. Results Effect of non fl-lactams alone and in combination with penicillin
E. hirae ATCC 9790 showed different susceptibility to peptidoglycan synthesis inhibitors; it was resistant to fosfomycin-trometamol (MIC = 128 mg/1) but susceptible to
Antibiotic Fosfomycin Vancomycin Bad trad n
antibiotic alone
MIC (mg/1) combination with penicillin"
FIC
128 1 2
32 0-25 05
0-37 0-37 0-37
Downloaded from http://jac.oxfordjournals.org/ at Dalhousie University on June 19, 2015
Table I. Antimicrobial activity of fosfomycin and other antibiotics, alone and in combination with penicillin against E. hirae ATCC 9790
"The concentration of penicillin was 0-03 mg/1 (1/8 MIC).
24
24
Time (h) Figure 1. Effect of fosfomycin (a), (d), vancomycin (b), (e) and bacitracin (c), (0 on OD (a, b, c) and viable count (d, e, f) of E. hirae ATCC 9790. All antibiotics were added at the following concentrations: • , MIC, A , 1/2 MIC; D , 1/4 MIC at zero time. Untreated cultures ( • ) were used as control.
A. Grossato et al.
266
PBP1»-
2-
b
c
d
e
f
g
Figure 2. PBP pattern of E. hirae ATCC 9790 membranes. The membranes were incubated for 60 min without any drug (lane a), with MIC (b) and 1/2 MIC (c) of fosfomycin, with MIC (d) and 1/2 MIC (e) of vancomycin and with MIC (!) and 1/2 MIC (g) of bacitracin. PBP detection was carried out by incubating membranes with l4C-penicillin for 60 min.
vancomycin (MIC = 1 mg/1), bacitracin (MIC = 2 mg/1) and penicillin (MIC = 0-25 mg/1) (Table I). Figure 1 shows the optical density (OD) and the viable count of cultures of ATCC 9790 incubated in the presence of 1/4, 1/2 and the MIC of each antibiotic. The increase in cell numbers was inhibited within 30 min of incubation with MIC and 1/2 the MIC of all drugs. During this time no cell lysis occurred. After 30 min a rapid decrease in viable count and absorbance occurred in samples treated with the MIC of fosfomycin and vancomycin, whereas bacitracin exhibited a smaller lytic and bactericidal effect. Fosfomycin showed a good inhibitory effect at 1/2 MIC for up to 4 h, after which time active growth resumed. The effect of the combination of fosfomycin, vancomycin and bacitracin with penicillin was also examined (Table I). A synergic effect was demonstrated for all combinations, shown by a fourfold decrease in the MIC of the individual antibiotics when combined with penicillin. Effect of non-fl-lactams on the PBP pattern In a preliminary experiment the PBP pattern of membranes of E. hirae treated with MIC and 1/2 MIC of the three antibiotics was analysed to exclude possible
Downloaded from http://jac.oxfordjournals.org/ at Dalhousie University on June 19, 2015
a
Non-0-lactains and E. hirae PBPs
. PBP1H
267
Mw 10 3
_ ~
e f g h
unspecific competition of these antibiotics with penicillin for binding to PBPs. As shown in Figure 2, PBP patterns of isolated membranes pretreated with fosfomycin, or vancomycin or bacitracin were similar to those of untreated samples. In order to evaluate the effect of these antibiotics upon the PBP pattern of growing cells, cultures in exponential phase of growth were incubated with several antibiotic concentrations for 30 min, an incubation time which did not cause lysis of the treated cultures (Figure 1). Cells were then collected and membranes isolated. Figure 3 shows the protein the PBP patterns of membranes from cells grown with or without fosfomycin. Growth for 30 min in the presence of the MIC and 1/2 MIC caused several alterations in protein profile (Figure 3, lanes e, f). Three proteins (60, 52, 43 kD) appeared to decrease, and four increased (140, 100, 50, 40 kD). In the PBP pattern (Figure 3, lane a) PBP 6 appeared to be the protein which was the most affected by fosfomycin as the density of this protein was reduced by 76% at the MIC. It is interesting to note that in Coomassie Blue stained gels the amount of a protein of 43 kD (the molecular weight of PBP 6) also decreased compared with untreated cells. The membrane proteins and PBP pattern of cells grown in the presence of MIC, 1/2, and 1/4 MIC of vancomycin and bacitracin are shown in Figures 4 and 5. Growth in the presence of vancomycin caused an increase of proteins with high molecular weight (140 and 50 kD) and a decrease in proteins of the same low molecular weight as with
Downloaded from http://jac.oxfordjournals.org/ at Dalhousie University on June 19, 2015
a b e d
Figure 3. PBP and membrane protein patterns of fosfomycin-treated cells of E. hirae ATCC 9790. PBPs were detected by binding "C-penicillin for 60 min to membranes obtained from cells grown in the presence of fosfomycin at MIC (a), 1/2 the MIC (b), 1/4 the MIC (c) and in drug-free medium (lane d). Membrane proteins were stained by Coomassie Blue: treated samples correspond to lanes e, f, g and untreated to lane h.
268
A. Grossato et al.
25»"
b
e
d
e
f
g
h
fosfomycin (Figure 4, lane e). In the presence of MIC and 1/2 MIC all PBPs had a similar decrease in density. (Figure 4, lanes a, b.) A decrease in five high molecular weight proteins (140, 95, 60, 52, 43 kD) were observed in the presence of MIC and 1/2 MIC of bacitracin (Figure 5). The PBP pattern showed that the density of all PBPs was affected. The density of PBPs 1, 2 and 6 was reduced more than that of PBP 4 and 5 at the MIC (Table II, Figure, lanes a, b). Discussion The synergy observed between drugs that act upon peptidoglycan synthesis is explained by the generally accepted mechanism of antibacterial synergy, that is sequential inhibition of a common biochemical pathway (Kronstad & Moellering, 1986). Our results, that treatment of E. hirae with the MIC of fosfomycin, bacitracin and vancomycin, showed a decrease of one or more PBPs suggests an additional mechanism: inhibition of a prior step that interferes with the synthesis of enzymes involved in subsequent reactions, thus decreasing their activity and, as a consequence, the amount of the second drug required to inhibit these enzymes. The decrease in PBP density observed in membranes of cells grown in the presence of fosfomycin, bacitracin and vancomycin could not be due to inhibition of PBPs as these
Downloaded from http://jac.oxfordjournals.org/ at Dalhousie University on June 19, 2015
a
Figure 4. PBP and membrane protein patterns of vancomycin-treated cells of E. hirae ATCC 9790. PBPs were detected by binding "C-penicillin for 60 min to membranes obtained from cells grown in the presence of vancomycin at MIC (a), 1/2 MIC (b), 1/4 MIC (c) in drug-free medium (lane d). Membrane proteins were stained by Coomassie Blue. Treated samples correspond to lanes e, f, g and untreated to lane h.
269
Non-p-lactams and E. hirae PBPs
2*~
b
d
e f
Figure 5. PBP and membrane protein patterns of bacitracin-treated cells of E. hirae ATCC 9790. PBP were detected by binding 14C-penicillin for 60 min to membranes obtained from cells grown in the presence of bacitracin at MIC (a), 1/2 MIC (b), 1/4 MIC (c) and in drug-free medium (lane d).
drugs did not compete with radioactive penicillin for binding to PBPs (Figure 2), nor to a general degradation of membrane proteins during treatment with the drugs as in Coomassie Blue stained gels the concentrations of some high molecular weight components were found to increase. Thus, the most likely explanation for the alteration in the PBP pattern is that these antibiotics inhibit the synthesis of these proteins. In addition the decrease in PBP synthesis did not appear to be an indirect consequence of growth inhibition as a different pattern of PBP alteration was observed for the different drugs. Bacitracin and vancomycin appeared to interfere in a similar way with the synthesis of all PBPs, whereas fosfomycin appeared to act specifically on the synthesis Table II. Density of the PBPs of E. hirae after treatment with the MIC of each antibiotic
Agent Fosfomycin Vancomycin Bacitracin
1
2
PBP 3/4'
5
6
100 55 49
100 58 65
100 65 85
100 52 68
24 38 36
100% density = untreated control. °PBPs 3 and 4 were not separated by the scanning densitomeler.
Downloaded from http://jac.oxfordjournals.org/ at Dalhousie University on June 19, 2015
a
270
A. Grossato et al.
Acknowledgements We are grateful to Mr Massimo Guida for his secretarial assistance. This investigation was supported by a grant from the Ministero per la Ricerca Scientifica (60%). Parts of this study were presented at the 16th International Congress of Chemotherapy, Jerusalem, 1989. References Buchanan, C. E. & Sowell, M. O. (1982). Synthesis of penicillin-binding protein 6 by stationaryphase Escherichia coli. Journal of Bacteriology 151, 491-4. Coyette, J., Ghuysen, J. M. & Fontana, R. (1980). The penicillin-binding proteins in Streptococcus faecalis ATCC 9790. European Journal of Biochemistry 110, 445-56. de la Rosa, E. J., de Pedro, M. A. & Vazques, D. (1982). Modification of penicillin-binding proteins of Escherichia coli associated with changes in the state of growth of the cells. FEMS Microbiology Letters 14, 91-4. Donabedian, H. & Andriole, V. T. (1977). Synergy of vancomycin with penicillins and cephalosporins against pseudomonas, klebsiella, and serratia. Yale Journal of Biology and Medicine, 50, 165-76. Fontana, R., Grossato, A., Rossi, L., Cheng, Y. R. & Satta, G. (1985). Transition from resistance to hypersusceptibility to 0-lactam antibiotics associated with loss of a low-affinity penicillin-binding protein in a Streptococcus faecium mutant highly resistant to penicillin. Antimicrobial Agents and Chemotherapy 28, 678-83. Kalian, F. M., Kahan, J. S., Cassidy, P. J. & Kropp, H. (1974). The mechanism of action of fosfomycin (phosphonomycin). Annals of the New York Academy of Sciences 235, 364-86. Kronstad, D. J. & Moellering, R. C. (1986). Antimicrobial combinations. In Antibiotics in Laboratory Medicine, 2nd edn (Lorian, V., Ed.), pp. 537-95. Williams & Wilkins, Baltimore, MD. Neu, H. C. & Kamimura, T. (1982). Synergy of fosmidomycin (FR-31564) and other antimicrobial agents. Antimicrobial Agents and Chemotherapy 22, 560-3. Perea, E. J., Torres, M. A. & Borobio, M. V. (1978). Synergism of fosfomycin-ampicillin and fosfomycin-chloramphenicol against Salmonella and Shigella. Antimicrobial Agents and Chemotherapy 18, 705-9. Perkins, H. R. (1969). Specificity of combination between mucopeptide precursors and vancomycin or ristocetin. Biochemical Journal 111, 195-205. Rossi, L., Tonin, E., Cheng, Y. R. & Fontana, R. (1985). Regulation of penicillin-binding protein activity: description of a methicillin-inducible penicillin-binding protein in Staphylococcus aureus. Antimicrobial Agents and Chemotherapy 27, 828-31.
Downloaded from http://jac.oxfordjournals.org/ at Dalhousie University on June 19, 2015
of PBP 6. This protein was shown to possess carboxypeptidase activity and to be a nonessential target for the mechanism of growth inhibition by penicillin (Coyette, Ghuysen & Fontana, 1980). However the role played by this enzyme in cell physiology may be equally important to justify the increased activity of fosfomycin in combination with penicillin. Utsui et al. (1986) reported that fosfomycin causes a decrease in PBPs 2, 2a and 4 of a methicillin-resistant Staphylococcus aureus (MRSA). Thus the interference by fosfomycin with regulation of PBP synthesis may be a general phenomenon that occurs in more than one species. Several studies suggest that the activities of some PBPs might be regulated at the level of synthesis, for instance the decrease of PBP 2 and 3 (de la Rosa et al., 1982) and the increase of PBP 6 synthesis in Escherichia coli stationary cells (de la Rosa et al., 1982; Buchanan & Sowell, 1982) or the induction by /Mactams of PBP 2a synthesis in MRSA (Rossi et al., 1985). The proposal that inhibitors of early stages of peptidoglycan synthesis may decrease the amount of PBP of the cells, in turn suggests that the synthesis of PBPs may be regulated by a feed-back mechanism involving the availability of certain precursors.
Non-fMactams and E. hirae PBPs
271
{Received 20 July 1990; accepted 5 October 1990)
Downloaded from http://jac.oxfordjournals.org/ at Dalhousie University on June 19, 2015
Roze, U. & Strominger, J. L. (1966). Alanine racemase from Staphylococcus aureus: conformation of its substrates and its inhibitor, D-cycloserine. Molecular Pharmacology 2, 92-4. Sande, M. A. & Scheld, W. M. (1980). Combination antibiotic therapy of bacterial endocarditis. Annals of Internal Medicine 92, 390-5. Shockman, G. D. & Barrett, J. F. (1983). Structure, function, and assembly of cell walls of Gram-positive bacteria. Annual Review of Microbiology 37, 501-27. Spratt, B. G. (1975). Distinct penicillin binding proteins involved in the division, elongation, and shape of Escherichia coli K12. Proceedings of the National Academy of Sciences of the United States of America 72, 2999-3003. Storm, D. R. (1974). Mechanism of bacitracin action: a specific lipid peptide interaction. Annals of the New York Academy of Sciences 235, 387-98. Utsui, Y., Ohya, S., Magaribuchi, T., Tajiama, M. & Yokota, T. (1986). Antibacterial activity of cefmetazole alone and in combination with fosfomycin against methicillin- and cephamresistant Staphylococcus aureus. Antimicrobial Agents and Chemotherapy 30, 917-22. Waxman, D. J. & Strominger, J. L. (1983). Penicillin-binding proteins and the mechanism of action of beta-lactam antibiotics. Annal Review of Biochemistry 52, 825-69.
Effect of non-pMactam antibiotics on penicillin-binding protein synthesis of Enterococcus hirae ATCC 9790 A. Grossato", R. Sartori and R. Fontana* of
Fosfomycin, bacitracin and vancomycin in combination with penicillin exhibit a synergic effect against Enterococcus hirae ATCC 9790. This strain, when incubated in presence of the MIC of non-/?-lactam antibiotics, showed an alternated pattern of PBPs. Bacitracin and vancomycin caused a decrease in the density of all PBPs while fosfomycin only reduced that of PBP 6. It is suggested that the observed synergy is a consequence of the inhibition.of PBP synthesis by antibiotics which act on the early stages of peptidoglycan synthesis prior to the formation of cross-links.
Introduction Peptidoglycan synthesis is a complex process which can be divided into three stages. In the first step the basic monomer (the UDP-acetylmuramyl-pentapeptide) is assembled in the cytoplasm, in the second the precursor unit is bound to a membrane lipid and carried from inside the cell membrane to outside and thirdly, it is cross-linked into the pre-existing peptidoglycan layers of the cell wall. (Shockman & Barrett, 1983.) The individual steps can be inhibited by different classes of antibiotics; fosfomycin (Kalian et al., 1974); cycloserine (Roze & Strominger, 1966) inhibit some enzymes involved in monomer synthesis; bacitracin (Storm, 1974) and vancomycin (Perkins, 1969) interfere, by different mechanisms, with the second stage. /?-lactams inhibit the terminal reactions leading to cross-linking of the peptidoglycan through covalent binding to several membrane proteins (the penicillin-binding proteins or PBPs) which are known to possess carboxypeptidase and transpeptidase activities (Spratt, 1975; Waxman & Strominger, 1983). Combinations of inhibitors of the first or second step in peptidoglycan synthesis with /Mactams have been shown to be synergic (Donabedian & Andriole, 1977; Perea, Torres & Borobio, 1978; Neu & Kamimura, 1982). It has been suggested that such synergy is the result of activity exerted on different reactions within the same biochemical process (Sande & Scheld, 1980; Krogstad & Moellering, 1986) but the mechanism leading to the increased antibacterial activity has never been investigated. In the present work we have studied the effect of fosfomycin, bacitracin and vancomycin on the PBP pattern of Enterococcus hirae ATCC 9790 and found that all the drugs tested reduced the synthesis of one or more PBPs. 263 0305-7453/91/030263 + 09 $02.00/0
© 1991 The British Society for Antimicrobial Chemotherapy
Downloaded from http://jac.oxfordjournals.org/ at Dalhousie University on June 19, 2015
"Institute of Microbiology, University of Padua, 35100-Padua;bInstitute Microbiology, University of Verona, 37134-Verona, Italy
264
A. Grossato et al. Materials and methods
Bacterial strains and conditions of growth All experiments were performed with E. hirae ATCC 9790, a penicillin susceptible strain used in previous studies (Fontana et al., 1985). Bacteria were grown in streptococcal broth (SB), a yeast extract/peptone medium, at 37°C and the growth monitored at 650 nm in a Perkin-Elmer spectrophotometer. Antibiotics and chemicals
Determination of susceptibility and synergy The susceptibility of E. hirae to all agents was determined by two-fold serial dilutions in SB with a final inoculum of 106 cfu/ml. The MIC was determined by visual inspection for lack of turbidity after 18 h of incubation at 37°C without shaking. Synergy studies were performed by the chequerboard titration method. Synergy was defined as a fractional inhibitory index (FIC) of < 0-5 which was determined by the following formula: (MIC of drug A in combination with drug B/MIC of drug A alone) + (MIC of drug B in combination with drug A/MIC of drug B alone). Killing kinetics Exponentially growing cultures were diluted with fresh medium in order to obtain a starting density of 106 cfu/ml. These suspensions were divided into sets of subcultures and various concentrations of fosfomycin, bacitracin and vancomycin were added to these. At one hour intervals 0-5 ml samples were taken, suitably diluted and plated on Brain Heart (BH Difco) agar. After an incubation of 18 h at 37°C the colonies on each plate were counted and averaged. Membrane preparation Exponentially growing cells were harvested by centrifugation, and washed with ice-cold distilled water and membranes were isolated as described previously (Fontana et al., 1985). In this procedure cellular lysis was carried out in phosphate buffer (pH 7) containing 100 mg/1 lysozyme, 5 mg/1 DNAse, 2 mg/1 RNAse, 1 HIM MgCl2 (for 1 h at 37°C). PBP analysis The PBPs of E. hirae ATCC 9790 were analysed by the method previously described (Cojette et al., 1980; Fontana et al., 1985). Experiments to determine competitive binding of fosfomycin, bacitracin and vancomycin to PBPs were performed by preincubation of membranes with various concentrations of non-radioactive antibiotic at 37°C for 30 min before the addition of 14C-benzyl penicillin. To study the effect of non/Mactam antibiotics upon the PBP pattern of actively growing cells, cultures were incubated with the antibiotics for 30 min, then centrifuged and lysed in order to obtain
Downloaded from http://jac.oxfordjournals.org/ at Dalhousie University on June 19, 2015
Fosfomycin-trometamol was provided by Zambon (Milano, Italy), and penicillin and vancomycin by Eli Lilly Italia (Sesto Fiorentino, Italy). l4C-benzylpenicillin was purchased from Amersham (Buckinghamshire, England). Bacitracin and chemicals for the preparation of polyacrylamide gels were obtained from Sigma (St. Louis, USA).
265
NoD-p-lactams and E. hirae PBPs
membranes. These were then incubated with radioactive penicillin for 60 min at 37°C. All samples were processed by sodium dodecyl sulphate-polyacrylamide slab gel electrophoresis and the PBPs were detected by fluorography. Results Effect of non fl-lactams alone and in combination with penicillin
E. hirae ATCC 9790 showed different susceptibility to peptidoglycan synthesis inhibitors; it was resistant to fosfomycin-trometamol (MIC = 128 mg/1) but susceptible to
Antibiotic Fosfomycin Vancomycin Bad trad n
antibiotic alone
MIC (mg/1) combination with penicillin"
FIC
128 1 2
32 0-25 05
0-37 0-37 0-37
Downloaded from http://jac.oxfordjournals.org/ at Dalhousie University on June 19, 2015
Table I. Antimicrobial activity of fosfomycin and other antibiotics, alone and in combination with penicillin against E. hirae ATCC 9790
"The concentration of penicillin was 0-03 mg/1 (1/8 MIC).
24
24
Time (h) Figure 1. Effect of fosfomycin (a), (d), vancomycin (b), (e) and bacitracin (c), (0 on OD (a, b, c) and viable count (d, e, f) of E. hirae ATCC 9790. All antibiotics were added at the following concentrations: • , MIC, A , 1/2 MIC; D , 1/4 MIC at zero time. Untreated cultures ( • ) were used as control.
A. Grossato et al.
266
PBP1»-
2-
b
c
d
e
f
g
Figure 2. PBP pattern of E. hirae ATCC 9790 membranes. The membranes were incubated for 60 min without any drug (lane a), with MIC (b) and 1/2 MIC (c) of fosfomycin, with MIC (d) and 1/2 MIC (e) of vancomycin and with MIC (!) and 1/2 MIC (g) of bacitracin. PBP detection was carried out by incubating membranes with l4C-penicillin for 60 min.
vancomycin (MIC = 1 mg/1), bacitracin (MIC = 2 mg/1) and penicillin (MIC = 0-25 mg/1) (Table I). Figure 1 shows the optical density (OD) and the viable count of cultures of ATCC 9790 incubated in the presence of 1/4, 1/2 and the MIC of each antibiotic. The increase in cell numbers was inhibited within 30 min of incubation with MIC and 1/2 the MIC of all drugs. During this time no cell lysis occurred. After 30 min a rapid decrease in viable count and absorbance occurred in samples treated with the MIC of fosfomycin and vancomycin, whereas bacitracin exhibited a smaller lytic and bactericidal effect. Fosfomycin showed a good inhibitory effect at 1/2 MIC for up to 4 h, after which time active growth resumed. The effect of the combination of fosfomycin, vancomycin and bacitracin with penicillin was also examined (Table I). A synergic effect was demonstrated for all combinations, shown by a fourfold decrease in the MIC of the individual antibiotics when combined with penicillin. Effect of non-fl-lactams on the PBP pattern In a preliminary experiment the PBP pattern of membranes of E. hirae treated with MIC and 1/2 MIC of the three antibiotics was analysed to exclude possible
Downloaded from http://jac.oxfordjournals.org/ at Dalhousie University on June 19, 2015
a
Non-0-lactains and E. hirae PBPs
. PBP1H
267
Mw 10 3
_ ~
e f g h
unspecific competition of these antibiotics with penicillin for binding to PBPs. As shown in Figure 2, PBP patterns of isolated membranes pretreated with fosfomycin, or vancomycin or bacitracin were similar to those of untreated samples. In order to evaluate the effect of these antibiotics upon the PBP pattern of growing cells, cultures in exponential phase of growth were incubated with several antibiotic concentrations for 30 min, an incubation time which did not cause lysis of the treated cultures (Figure 1). Cells were then collected and membranes isolated. Figure 3 shows the protein the PBP patterns of membranes from cells grown with or without fosfomycin. Growth for 30 min in the presence of the MIC and 1/2 MIC caused several alterations in protein profile (Figure 3, lanes e, f). Three proteins (60, 52, 43 kD) appeared to decrease, and four increased (140, 100, 50, 40 kD). In the PBP pattern (Figure 3, lane a) PBP 6 appeared to be the protein which was the most affected by fosfomycin as the density of this protein was reduced by 76% at the MIC. It is interesting to note that in Coomassie Blue stained gels the amount of a protein of 43 kD (the molecular weight of PBP 6) also decreased compared with untreated cells. The membrane proteins and PBP pattern of cells grown in the presence of MIC, 1/2, and 1/4 MIC of vancomycin and bacitracin are shown in Figures 4 and 5. Growth in the presence of vancomycin caused an increase of proteins with high molecular weight (140 and 50 kD) and a decrease in proteins of the same low molecular weight as with
Downloaded from http://jac.oxfordjournals.org/ at Dalhousie University on June 19, 2015
a b e d
Figure 3. PBP and membrane protein patterns of fosfomycin-treated cells of E. hirae ATCC 9790. PBPs were detected by binding "C-penicillin for 60 min to membranes obtained from cells grown in the presence of fosfomycin at MIC (a), 1/2 the MIC (b), 1/4 the MIC (c) and in drug-free medium (lane d). Membrane proteins were stained by Coomassie Blue: treated samples correspond to lanes e, f, g and untreated to lane h.
268
A. Grossato et al.
25»"
b
e
d
e
f
g
h
fosfomycin (Figure 4, lane e). In the presence of MIC and 1/2 MIC all PBPs had a similar decrease in density. (Figure 4, lanes a, b.) A decrease in five high molecular weight proteins (140, 95, 60, 52, 43 kD) were observed in the presence of MIC and 1/2 MIC of bacitracin (Figure 5). The PBP pattern showed that the density of all PBPs was affected. The density of PBPs 1, 2 and 6 was reduced more than that of PBP 4 and 5 at the MIC (Table II, Figure, lanes a, b). Discussion The synergy observed between drugs that act upon peptidoglycan synthesis is explained by the generally accepted mechanism of antibacterial synergy, that is sequential inhibition of a common biochemical pathway (Kronstad & Moellering, 1986). Our results, that treatment of E. hirae with the MIC of fosfomycin, bacitracin and vancomycin, showed a decrease of one or more PBPs suggests an additional mechanism: inhibition of a prior step that interferes with the synthesis of enzymes involved in subsequent reactions, thus decreasing their activity and, as a consequence, the amount of the second drug required to inhibit these enzymes. The decrease in PBP density observed in membranes of cells grown in the presence of fosfomycin, bacitracin and vancomycin could not be due to inhibition of PBPs as these
Downloaded from http://jac.oxfordjournals.org/ at Dalhousie University on June 19, 2015
a
Figure 4. PBP and membrane protein patterns of vancomycin-treated cells of E. hirae ATCC 9790. PBPs were detected by binding "C-penicillin for 60 min to membranes obtained from cells grown in the presence of vancomycin at MIC (a), 1/2 MIC (b), 1/4 MIC (c) in drug-free medium (lane d). Membrane proteins were stained by Coomassie Blue. Treated samples correspond to lanes e, f, g and untreated to lane h.
269
Non-p-lactams and E. hirae PBPs
2*~
b
d
e f
Figure 5. PBP and membrane protein patterns of bacitracin-treated cells of E. hirae ATCC 9790. PBP were detected by binding 14C-penicillin for 60 min to membranes obtained from cells grown in the presence of bacitracin at MIC (a), 1/2 MIC (b), 1/4 MIC (c) and in drug-free medium (lane d).
drugs did not compete with radioactive penicillin for binding to PBPs (Figure 2), nor to a general degradation of membrane proteins during treatment with the drugs as in Coomassie Blue stained gels the concentrations of some high molecular weight components were found to increase. Thus, the most likely explanation for the alteration in the PBP pattern is that these antibiotics inhibit the synthesis of these proteins. In addition the decrease in PBP synthesis did not appear to be an indirect consequence of growth inhibition as a different pattern of PBP alteration was observed for the different drugs. Bacitracin and vancomycin appeared to interfere in a similar way with the synthesis of all PBPs, whereas fosfomycin appeared to act specifically on the synthesis Table II. Density of the PBPs of E. hirae after treatment with the MIC of each antibiotic
Agent Fosfomycin Vancomycin Bacitracin
1
2
PBP 3/4'
5
6
100 55 49
100 58 65
100 65 85
100 52 68
24 38 36
100% density = untreated control. °PBPs 3 and 4 were not separated by the scanning densitomeler.
Downloaded from http://jac.oxfordjournals.org/ at Dalhousie University on June 19, 2015
a
270
A. Grossato et al.
Acknowledgements We are grateful to Mr Massimo Guida for his secretarial assistance. This investigation was supported by a grant from the Ministero per la Ricerca Scientifica (60%). Parts of this study were presented at the 16th International Congress of Chemotherapy, Jerusalem, 1989. References Buchanan, C. E. & Sowell, M. O. (1982). Synthesis of penicillin-binding protein 6 by stationaryphase Escherichia coli. Journal of Bacteriology 151, 491-4. Coyette, J., Ghuysen, J. M. & Fontana, R. (1980). The penicillin-binding proteins in Streptococcus faecalis ATCC 9790. European Journal of Biochemistry 110, 445-56. de la Rosa, E. J., de Pedro, M. A. & Vazques, D. (1982). Modification of penicillin-binding proteins of Escherichia coli associated with changes in the state of growth of the cells. FEMS Microbiology Letters 14, 91-4. Donabedian, H. & Andriole, V. T. (1977). Synergy of vancomycin with penicillins and cephalosporins against pseudomonas, klebsiella, and serratia. Yale Journal of Biology and Medicine, 50, 165-76. Fontana, R., Grossato, A., Rossi, L., Cheng, Y. R. & Satta, G. (1985). Transition from resistance to hypersusceptibility to 0-lactam antibiotics associated with loss of a low-affinity penicillin-binding protein in a Streptococcus faecium mutant highly resistant to penicillin. Antimicrobial Agents and Chemotherapy 28, 678-83. Kalian, F. M., Kahan, J. S., Cassidy, P. J. & Kropp, H. (1974). The mechanism of action of fosfomycin (phosphonomycin). Annals of the New York Academy of Sciences 235, 364-86. Kronstad, D. J. & Moellering, R. C. (1986). Antimicrobial combinations. In Antibiotics in Laboratory Medicine, 2nd edn (Lorian, V., Ed.), pp. 537-95. Williams & Wilkins, Baltimore, MD. Neu, H. C. & Kamimura, T. (1982). Synergy of fosmidomycin (FR-31564) and other antimicrobial agents. Antimicrobial Agents and Chemotherapy 22, 560-3. Perea, E. J., Torres, M. A. & Borobio, M. V. (1978). Synergism of fosfomycin-ampicillin and fosfomycin-chloramphenicol against Salmonella and Shigella. Antimicrobial Agents and Chemotherapy 18, 705-9. Perkins, H. R. (1969). Specificity of combination between mucopeptide precursors and vancomycin or ristocetin. Biochemical Journal 111, 195-205. Rossi, L., Tonin, E., Cheng, Y. R. & Fontana, R. (1985). Regulation of penicillin-binding protein activity: description of a methicillin-inducible penicillin-binding protein in Staphylococcus aureus. Antimicrobial Agents and Chemotherapy 27, 828-31.
Downloaded from http://jac.oxfordjournals.org/ at Dalhousie University on June 19, 2015
of PBP 6. This protein was shown to possess carboxypeptidase activity and to be a nonessential target for the mechanism of growth inhibition by penicillin (Coyette, Ghuysen & Fontana, 1980). However the role played by this enzyme in cell physiology may be equally important to justify the increased activity of fosfomycin in combination with penicillin. Utsui et al. (1986) reported that fosfomycin causes a decrease in PBPs 2, 2a and 4 of a methicillin-resistant Staphylococcus aureus (MRSA). Thus the interference by fosfomycin with regulation of PBP synthesis may be a general phenomenon that occurs in more than one species. Several studies suggest that the activities of some PBPs might be regulated at the level of synthesis, for instance the decrease of PBP 2 and 3 (de la Rosa et al., 1982) and the increase of PBP 6 synthesis in Escherichia coli stationary cells (de la Rosa et al., 1982; Buchanan & Sowell, 1982) or the induction by /Mactams of PBP 2a synthesis in MRSA (Rossi et al., 1985). The proposal that inhibitors of early stages of peptidoglycan synthesis may decrease the amount of PBP of the cells, in turn suggests that the synthesis of PBPs may be regulated by a feed-back mechanism involving the availability of certain precursors.
Non-fMactams and E. hirae PBPs
271
{Received 20 July 1990; accepted 5 October 1990)
Downloaded from http://jac.oxfordjournals.org/ at Dalhousie University on June 19, 2015
Roze, U. & Strominger, J. L. (1966). Alanine racemase from Staphylococcus aureus: conformation of its substrates and its inhibitor, D-cycloserine. Molecular Pharmacology 2, 92-4. Sande, M. A. & Scheld, W. M. (1980). Combination antibiotic therapy of bacterial endocarditis. Annals of Internal Medicine 92, 390-5. Shockman, G. D. & Barrett, J. F. (1983). Structure, function, and assembly of cell walls of Gram-positive bacteria. Annual Review of Microbiology 37, 501-27. Spratt, B. G. (1975). Distinct penicillin binding proteins involved in the division, elongation, and shape of Escherichia coli K12. Proceedings of the National Academy of Sciences of the United States of America 72, 2999-3003. Storm, D. R. (1974). Mechanism of bacitracin action: a specific lipid peptide interaction. Annals of the New York Academy of Sciences 235, 387-98. Utsui, Y., Ohya, S., Magaribuchi, T., Tajiama, M. & Yokota, T. (1986). Antibacterial activity of cefmetazole alone and in combination with fosfomycin against methicillin- and cephamresistant Staphylococcus aureus. Antimicrobial Agents and Chemotherapy 30, 917-22. Waxman, D. J. & Strominger, J. L. (1983). Penicillin-binding proteins and the mechanism of action of beta-lactam antibiotics. Annal Review of Biochemistry 52, 825-69.
