Journal of Antimicrobial Chemotherapy (1990) 26, 29-38
Activity of sob-minimal inhibitory concentrations of aspoxicillin in prolonging the postantibiotic effect against Staphylococcus aureus Tadahiro Oshida, Tokio Onta, NoriyukJ Nakanfehi, Tadahiro Matsushita and Tontaro Yamaguchl
Aspoxicillin, a newly developed acylureido-penicillin with a long half-life in mouse serum of 55 min, induced postantibiotic effects (PAEs) against Staphylococcus aureus Smith of 1-7 h in vitro and 5-2 h in vivo in a thigh infection model in neutropenic mice. The long serum half-life meant that in order to evaluate the in-vivo PAE, it was necessary to examine the contribution of the drug at a sub-minimal inhibitory concentration (sub-MIQ. Growth suppression by sub-MICs of aspoxicillin was examined in vitro using either previously unexposed bacterial cells or cells which had been pre-exposed to twice the MIC of aspoxicillin for 2 h. At each sub-MIC tested, the duration of growth suppression for pre-exposed cells was longer than that for unexposed cells. In an attempt to eliminate the sub-MIC effect in vivo, penicillinase was injected into mice at the time after administration when the aspoxicillin serum concentration approached the MIC. The in-vivo PAE decreased to 2-7 h when penicillinase was injected. It was concluded that aspoxicillin induced a PAE in vivo which was additional to the effect of sub-inhibitory residual drug, but that sub-MIC levels of the drug were simultaneously involved in suppressing bacterial regrowth after the drug concentration decreased below the MIC. Similar postantibiotic subMIC effects may also occur with other long half-life antibiotics. Introduction
A postantibiotic effect (PAE) is the suppression of bacterial growth that persists after limited exposure to an antibiotic. The presence of a PAE may be an important factor to consider in the design of antimicrobial dosing regimens (Vogelman & Craig, 1985) since a prolonged PAE should permit widely-spaced intermittent dosing with intermediate periods where the antibiotic concentration falls below the MIC (Gerber et al., 1982; Gengo et al., 1984). A PAE is defined as an effect resulting from previous exposure to an antibiotic rather than an effect due to the continuing presence of a drug at sub-minimal inhibitory concentrations (sub-MICs) (Volgelman & Craig, 1985). When a PAE is determined in vitro, the effect of the drug at a sub-MIC can be excluded by dilution or washing of a bacterial culture or addition of an antibiotic-degrading enzyme such as penicillinase. In contrast, it is difficult to exclude completely a sub-MIC effect in vivo, especially with a drug that has a long serum half-life. Indeed it is often unclear whether suppression of bacterial growth in vivo results from a PAE or the effect of a drug at a sub-MIC. In addition, the effect of a drug at a sub-MIC on growth of an organism during a PAE phase has not yet been investigated. 29 0305-7453/90/070029+ 10 S02.00/0
© 1990 The British Society for Antimicrobial Chemotherapy
Downloaded from http://jac.oxfordjournals.org/ at UQ Library on June 18, 2015
Biological Research Laboratory, Tanabe Seiyaku Co., Ltd., 2-2-50 Kawagishi, Toda, Saitama 335, Japan
30
T. Oshidi et aL
Figure 1. Chemical structure of aspoxicillin.
Materials and methods Organism Staph. aureus Smith was used. Antibiotics and susceptibility tests Penicillins used in this study were aspoxicillin (Tanabe Seiyaku Co., Osaka, Japan) and piperacillin (Toyama Chem. Ind., Tokyo, Japan). MICs were determined by a broth macrodilution method with a final inoculum of 1 x 10*cfu/ml in Sensitivity test broth (STB, Nissui, Tokyo, Japan). In-vitro PAE The method of Buntdzen et al. (1981) was slightly modified for use in this study. The inocula for all procedures were prepared by incubating the organism overnight at 37°C in STB. The culture was then diluted 1:50 in fresh STB and incubated until the logarithmic phase of growth was reached. Cultures with a cell density of approximately 107cfu/ml were then exposed to various drug concentrations for 1-3 h. In order to remove residual antibiotics completely, cultures were diluted lO'-lOMbld in fresh STB and penicillinase (EC 3.5.2.6; Sigma, St. Louis, USA) added to each culture at a final concentration of 0-1 U/ml. Viable counts were determined on Sensitivity Test Agar (STA, Nissui) plates for the initial inoculum, following the removal of the antibiotics and every 1 h thereafter.
Downloaded from http://jac.oxfordjournals.org/ at UQ Library on June 18, 2015
/J-Lactam antibiotics arc known to produce PAEs against Gram-positive cocci in vitro and have also been shown to produce a PAE in vivo against Staphylococcus aureus, but not against Streptococcus pneumoniae. Tauber et al. (1984) showed that growth suppression of Str. pneumoniae after ampicillin treatment in a rabbit meningitis model was caused by the small residual amount of drug in CSF. Recently, Vogelman et al. (1988) reported, in a mouse thigh infection model, that the PAE observed with Staph. aureus and /Mactam antibiotics was independent of any residual drug, but that growth suppression of Str. pneumoniae was caused by a residual drug level. Aspoxicillin (TA-058; Figure 1) is an injectable acylureido-penicillin with a high and persistent serum level compared to other known penicillins (Wagatsuma et al., 1983; Yamaguchi et al., 1984). We therefore determined the PAEs of aspoxicillin against Staph. aureus, both in vitro and in vivo, and compared them with those of piperacillin. The relationship between the PAE and the effect due to residual drug at sub-MICs was also examined.
PostantiMotic effect of aspoxfcflUn against Staph. attrttu
31
In-vitro effect of sub-MICs on Staph. aureus in a PAE phase The log-phase culture was pre-exposed to aspoxicillin for 2 h at twice the MIC and then diluted by adding 2 ml to 30 ml of fresh STB. This culture, containing 0-125 x MIC of aspoxicillin, was divided among test tubes in 5-ml portions. Aspoxicillin was added to each tube to givefinalconcentrations of 0-25, 0-5 and 1 x MIC. Penicillinase was added for the drug-free control. For the pre-exposure control, the logarithmic phase culture was diluted and exposed to 0-125, 0-25, 0-5 and 1 xMIC of aspoxicillin, and viable counts were determined. In-vivo PAE
Calculation of PAE The duration in-vitro PAE was calculated as described by Gerber & Craig (1981), using the formula PAE = T— C, where T represents the time required for the viable count of the test culture to increase ten-fold following drug removal and C is the time required for the viable count of the drug-free control culture to increase ten-fold following similar treatment. The duration of in-vivo PAE was calculated as described by Vogelman et al. (1988), using the formula PAE = T— C—M, where M represents the time for which drug serum levels exceed the MIC, Tis the time required for the viable count in thigh homogenates of treated mice to increase ten-fold above the count at time M, and C is the time required for the viable count in thigh homogenates of untreated controls to increase ten-fold above the counts immediately following bacterial inoculation. Penicillinase injection
Penicillinase (EC 3.5.2.6) was dissolved (500 U/ml) in saline and injected into mice either 1 or 2 h after drug administration in order to eliminate any residual penicillins. Each mouse received 0-2 ml iv of the enzyme solution and 0-1 ml into the thigh muscle in which the bacteria had been inoculated. The efficacy of this treatment for eliminating aspoxicillin was tested using Escherichia coli KC-14 (MIC 0-78 mg/1; a half of Staph. aureus Smith). No growth suppression was observed with E. coli following penicillinase
Downloaded from http://jac.oxfordjournals.org/ at UQ Library on June 18, 2015
The method of Gudmundsson, Vogelman & Craig (1986) was slightly modified to determine the PAE in vivo. Male 20-25 g ddy mice were made neutropenic by a single ip administration of cyclophosphamide (250mg/kg) four days prior to investigation. This treatment reduced the number of neutrophils from 807± 161 (cells/mm3±S.D., n = 5) to 29 ± 15 on day 4 and 74 ± 49 on day 5. Log-phase bacteria were harvested just before inoculation into mice, washed once with ice-cold saline, re-suspended in saline to give an OD^o value of 002 and stored in an ice bath until use. Mice were then inoculated with 0-1 ml of this suspension (1 x 10*cfu) injected into a right thigh. Each drug was administered subcutaneously 2 h after this inoculation at a dose of 50 mg/kg. Three treated mice were killed every 2 h after bacterial inoculation. Thigh muscles, together with femoral bone, were removed and homogenized in 10 ml ice-cold saline with a Politron tissue homogenizer (Kinematica, Lucerne, Switzerland) and a 20 mm brade. Portions (0-1 ml) of diluted homogenate were then mixed with 10 ml of STA in a petri dish and viable counts subsequently determined.
32
T. OshMa el aL
injection, indicating that the drug concentration had sufficiently decreased to allow normal growth (data not shown). Pharmacokinetics Serum concentrations after a single subcutaneous administration were determined with the microbiological assay described by Tani et al. (1984a). Groups of eight neutropenic mice were killed at 15, 30, 60, 120 and 240 min after drug administration. Blood samples were taken by exsanguination from the femoral artery and vein, and sera were separated by centrifugation (1500g, 15 min, 4°C). Drug levels in sera were analysed using a two-compartment model. Antimicrobial levels in muscle interstitial fluid are known to fluctuate closely with serum concentration (Volgelman et al., 1988).
To determine whether Staph. aureus was in a PAE phase in vivo, the organism was recovered from thigh muscle 2h after aspoxicillin administration and regrowth was examined in vitro. Removed thigh muscle was first homogenized with 10 ml of saline. The homogenate of the aspoxicillin-administered mouse thigh was diluted 1:20 in fresh STB and penicillinase added to a final concentration of 0-1 U/ml. Homogenate of the drug-free mouse thigh was diluted 1:600 in fresh STB supplemented with 5% of homogenate prepared from non-infected thigh. These cultures were incubated at 37°C and viable counts determined at intervals. In order to control the effect of in-vivo to in-vitro transfer, log-phase cells which had been cultured in STB were transferred to fresh STB supplemented with 5% of the non-infected thigh homogenate. Results In-vitro PAE The MICs of aspoxicillin and piperacillin for Staph. aureus were 1-56 and 0-78 mg/1 respectively. Table I shows that both aspoxicillin and piperacillin produced an in-vitro PAE against Staph. aureus Smith. The PAEs with both penicillins were prolonged by an increase in drug concentration and exposure time. The maximum duration of the PAE was almost the same with each drug: approximately 2 h. Table L In-vitro PAE obtained with various concentrations and exposure times of aspoxicillin and piperacillin against Staph. aureus Smith Drug concentration (xMIQ 1 2 4 8 16 64
PAE(h) aspoxicillin piperacillin exposure time (h) exposure time (h) 1 2 3 1 2 3 01 04 06
01 1-3
01 1-3
03 08 11
1-7 1-3 1-7
07 1-5 1-8
1-4 1-4
08 1-3
Downloaded from http://jac.oxfordjournals.org/ at UQ Library on June 18, 2015
Regrowth o/Staph. aureus recovered from mouse thighs
33
Postantibiotic effect of aspoxfcfllln against Staph. aureus
Effect ofsub-MICs on the in-vitro PAE Figure 2 shows the effect of sub-MICs of aspoxicillin, together with the influence of pre-exposure, on the growth of Staph. aureus Smith. The duration of growth inhibition with pre-exposed cells was clearly longer than that observed with previously unexposed cells. Thus, pre-exposure to twice the MIC of aspoxicillin enhanced the susceptibility of the organism to sub-MIC levels, to the extent that a 50% decrease in the viable count of the pre-exposed cells was observed 1 h after exposure to 0-5 x MIC. The time required for the viable count to increase ten-fold following the dilution step was calculated from Figure 2 for each drug concentration; these are listed as 'growth times' in Table II. In this experiment, the PAE was 08 h. A 'sub-MIC effect', defined as the prolongation of the growth time produced by the presence of aspoxicillin at a subMIC, was calculated for the previously unexposed cells. Sub-MIC effects of up to 0-7 h were obtained with increased drug concentrations. The sub-MIC effect against pre-exposed cells is listed in Table II as the 'postantibiotic sub-MIC effect'. At each sub-MIC it was found that this effect was longer than that observed against previously unexposed cells. In particular, pre-exposure followed by exposure to 0-5 x MIC produced a very long postantibiotic sub-MIC effect of 3-6 h Table n. Postantibiotic sub-MIC effect of aspoxicillin against Staph. aureus Smith Drug concentration (xMIQ 0 0-125 0-25 05
Unexposed growth sub-MIC time effect (h) (h) 1-3 1-5 1-7 2-0
0-2 0-4 0-7
growth time (h)
Pre-exposed postantibiotic sub-MIC effect (h)
pre-exposure effect (h)
21 2-5 3-0 5-7
0-4 09 3-6
02 05 2-9
Downloaded from http://jac.oxfordjournals.org/ at UQ Library on June 18, 2015
Fipat 2. Growth curves of (a) the normal log-phase culture and (b) the pre-exposed culture of Staph. aweus in the presence of aspoxicillin at 0-125 x MIC ( # ) , 0-25 x MIC (A). 0-5 x MIC ( • ) , 1 x MIC ( T ) and control (O)< Pre-exposure was to 2 x MIC of aspoxicillin for 2 h.
34
T. Oshtda et aL 100
0-3
01
and a decrease in the viable count. The time difference between the sub-MIC effect and the postantibiotic sub-MIC effect is listed in Table II as the 'pre-exposure effect'. Preexposure effects of up to 2-9 h were obtained with increased sub-MIC drug concentrations. These resulted from the combined action of the PAE and the presence of aspoxidllin at a sub-MIC. Pharmacokinetics Serum levels of aspoxidllin and piperadllin in neutropenic mice following a single dose of 50 mg/kg are shown in Figure 3. The aspoxidllin levels were always higher than those of piperadllin and the half-life of aspoxidllin in the /?-phase was 2-6 x that of piperadllin. Although the MIC of aspoxidllin for Staph. aureus (1-56 mg/1) was higher than that of piperadllin (078 mg/1), the time for which the MIC was exceeded was longer for aspoxidllin (2-4 h) than for piperadllin (1-6 h). In addition, the time for which aspoxidllin remained at a sub-MIC level was longer than that for piperadllin. In-vivo PAE Figure 4(a) shows the growth of Staph. aureus Smith in mouse thighs. Administration of dther aspoxidllin or piperadllin decreased the viable count for the first 2 h, with further suppression of regrowth even after the drug level fell below the MIC. It therefore appeared that in-vivo PAEs were being observed. Injection of penidllinase 1 h after piperadllin administration resulted in a rapid increase in the viable count. In contrast, injection of penidllinase 2 h after aspoxicillin administration did not influence regrowth until a further 2 h had elapsed. This demonstrated that aspoxidllin produced an in-vivo PAE which was independent from the sub-MIC effect of the drug. Figure 4(b) shows a similar experiment, except that the piperacillin group recdved penidllinase 2h after drug administration. This resulted in a longer suppression of regrowth.
Downloaded from http://jac.oxfordjournals.org/ at UQ Library on June 18, 2015
Figure 3. Scrum levels of aspoxidllin ( # ) (Tu •• 55 min) and piperadllin (A) (Tu - 21 min). Mean±sx>. = 8).
Postantibiotk effect of gspoxkfltin against Stapk. aureus
c .
35
( o )
i
Drug
jb A
8
c
> * -
7
6
•&
£
C
3
|
i
1r i
[
-r I
1
U
(b) 9
Drug
I
8
7 / 6-
1
A
/ Time above MIC
5
~ Aipadc l l i n E ^ i
-2
I
2
I
i
4 Time (h)
Figure 4. Effect of penicillinase injection (|) on the growth of Staph. aureus Smith in mouse thighs following single dose drug administration: control ( • — # ) , aspoxicillin ( • — • ) and piperacillin (A — A)- (a) Penicillinase injected 2 h after aspoxicillin administration ( • • ) and 1 h after pipercQlin administration (A A)- (b) Penicillinase injected 2 h after administration of both aspoxicillin ( • •) and piperacillin (A A)- Mean±s.D. (n - 3).
Table III shows the in-vivo PAE calculated from Figure 4. The results were extrapolated to determine a PAE for the groups with prolonged growth suppression, with the exception of the penicillinase-free aspoxicillin group in Figure 4(b) for which the PAE was not determined since regrowth was not observed within 6h. In the absence of pencillinase, aspoxicillin produced a longer PAE (5-2 h) than piperacillin (3-3 h). When penicillinase was injected 2 h after drug administration, the PAE of aspoxicilhn was reduced to 2-7 h. This decrease was caused by elimination of residual drug at a concentration close to the MIC and the subsequent sub-MICs. A smaller decrease in PAE was observed with piperacillin, reflecting reduced duration for which the drug level stayed within the range of sub-MICs. When penicillinase was injected 1 h and 2 h after drug administration, the PAE of piperacillin was prolonged from 0-8 to 20h, perhaps reflecting the altered time for which the drug concentration exceeded the MIC. The longer PAE (2-7 h) observed with the penicillinase-injected aspoxicillin group was probably due to the same cause. The same phenomenon was observed with PAEs in vitro: longer exposure times produced longer PAEs.
Downloaded from http://jac.oxfordjournals.org/ at UQ Library on June 18, 2015
!
36
T. Oshid* et oL
Figure 5. In-vitro growth curves of Staph. aureus Smith recovered from mouse thighs. Each of two thigh homogenates, from atpoxicinin-treated ( • and • ) or non-treated (A and A) mice, were diluted with fresh STB and incubated. The log-phase control culture ( • • ) , previously grown in STB, was incubated in fresh STB supplemented with 5% of non-infected thigh homogenate for this experiment
Growth of the recovered organism Figure 5 shows the in-vitro growth of Staph. aureus Smith recovered from mice 2 h after aspoxicillin administration. In comparison with the control log-phase culture, growth suppression of cells recovered from non-treated mice was minimal. In contrast, growth suppression was clearly observed with cells recovered from the aspoxicillin-treated mice, thereby demonstrating that the cells were in a PAE phase in vivo. Discussion
This study demonstrated that Staph. aureus Smith was more susceptible to sub-MICs of aspoxicillin in the PAE phase than in the normal log phase. The duration of growth suppression in vitro was prolonged by a combination of the PAE and the effect of a Table HI. Effect of penicillin inactivation on the in-vivo PAE against Staph. aureus Smith
Drug Aspoxicillin Aspoxicillin Pipcracillin Piperacillin Pipcracillin
Penicillinase injection time (h)
Time above MIC(h)
2 — 1 2
2-4 20 1-6 10 1-6
"Mean values of two experiments.
PAE (h) 5-2 2-7* 3-3*
08 20
Downloaded from http://jac.oxfordjournals.org/ at UQ Library on June 18, 2015
Tim* ( h )
Postantibkrtfc effect of aspoxkfflin against Staph. aureus
37
Acknowledgements We are grateful to Dr I. Maezawa and Dr Y. Ito for their helpful discussion and critical review of the manuscript
Downloaded from http://jac.oxfordjournals.org/ at UQ Library on June 18, 2015
sub-MIC of aspoxicillin. The decrease in viable count which occurred in the presence of aspoxicillin at 0-5 x MIC following pre-exposure was not observed with previously unexposed cells. Similar viable count decreases in the presence of sub-MIC drug levels are often observed during determinations of PAEs in vivo and may result from the same effect. Gerber & Craig (1981) demonstrated that cells in the PAE phase were less susceptible to the bactericidal activity of penicillins; however their study examined the bactericidal effect of drug concentrations above the MIC, as opposed to suppression of regrowth at sub-MICs. There was little difference between the in-vitro PAEs of the two drugs tested; however, the in-vivo PAE of aspoxicillin in the absence of penicillinase was longer than that of piperacillin, probably as a result of the different pharmacokinetic properties of the two penicillins. Similarly, the longer time for which the concentration of aspoxicillin remained above the MIC appeared to produce a longer in-vivo PAE than piperacillin, even after penicillinase injection. In addition, aspoxicillin remained at a sub-MIC longer than piperaciUin. Penicillinase injection shortened the in-vivo PAEs by eliminating residual drug. The prolonged suppression of regrowth by aspoxicillin can therefore be attributed to the combined effect of the PAE and the residual drug activity at concentrations close to the MIC and the following sub-MICs. This conclusion was supported by the finding that growth suppression of Staph. aureus in the PAE phase was prolonged in vitro by the presence of the drug a.t sub-MICs. This prolonged suppression is different from the postantibiotic leucocyte enhancement described by McDonald, Wetherall & Pruul (1981) since the numbers of leucocytes were extremely low in the neutropenic mice used in this study. Vogelman el al. (1988) also examined a sub-MIC effect produced after inoculation of a log-phase culture into mouse thighs at a time when drug concentrations were below the MIC. However, since the pre-exposed culture was more susceptible than the log-phase culture, we believe that their method may result in an underestimation of any sub-MIC effect, particularly with regard to a long half-life drug. Sub-MICs of penicillins are known to produce a variety of effects against Staph. aureus, including changes in cell morphology cell proliferation or sensitivity to antibiotics (Lorian, 1975; Satta et al., 1986). The antibacterial effects of sub-MICs of drugs in experimental infection have also been studied (Zak & Kradolfer, 1979; Ogawa, 1986). When an antibiotic is administered in a single dose, there is a period during which the drug concentration falls below the MIC. It is therefore advantageous if the drug produces an additional effect at sub-MICs, such as an increased sensitivity or a PAE. Drug levels in humans tend to persist longer than in rodents and we therefore conclude that the presence of a postantibiotic sub-MIC effect should also be considered when the in-vivo PAE is determined for a long half-life drug. It has been reported that the in-vivo activity of aspoxicillin is greater than that predicted from its in-vitro antibacterial activity (Tani et al., 19846; Ueno et al., 1985). This feature has been thought to be a consequence of its good pharmacokinetic properties, which include high and prolonged serum and tissue levels (Matsushita, Kouyama & Yamaguchi, 1985; Okuno et al., 1988). We believe that the long PAE and sub-MIC effect of aspoxicillin may also contribute favourably to this characteristic.
38
T. Oshida et aL References
(Received 14 July 1989; revised version accepted 2 February 1990)
Downloaded from http://jac.oxfordjournals.org/ at UQ Library on June 18, 2015
Buntdzen, R. W., Gerber, A. U., Cohn, D. L. & Craig, W. A. (1981). Postantibiotic suppression of bacterial growth. Reviews of Infectious Diseases 3, 28-37. Gengo, F. M., Mannion, T. W., Nightingale, C. H. & Schentag, J. J. (1984). Integration of pharmacokinetics and pharmacodynamics of methicillin in curative treatment of experimental endocarditis. Journal of Antimicrobial Chemotherapy 14, 619-31. Gerber, A. U. & Craig, W. A. (1981). Growth kinetics of respiratory pathogens after short exposures to ampitillin and erythromycin in vitro. Journal of Antimicrobial Chemotherapy 8, Suppl. C, 81-91. Gerber, A. U., Wiprachtiger, P., Stettler-Spichiger, U. & Lebek, G. (1982). Constant infusions vs. intermittent doses of gentamicin against Pseudomonas aeruginosa in vitro. Journal of Infectious Diseases 145, 554-#). Gudmundsson, S., Vogelman, B. & Craig, W. A. (1986). The in-vivo postantibiotic effect of imipenem and other new antimicrobials. Journal of Antimicrobial Chemotherapy 18, Suppl. E, 67-73. Lorian, V. (1975). Some effects of subinhibitory concentrations of penicillin on the structure and division of staphylococci. Antimicrobial Agents and Chemotherapy 7, 864-70. McDonald, P. J., Wetherall, B. L. & Pruul, H. (1981). Postantibiotic leukocyte enhancement: increased susceptibility of bacteria pretreated with antibiotics to activity of leukocytes. Reviews of Infectious Diseases 3, 38-44. Matsushita, T., Kouyama, K. & Yamaguchi, T. (1985). Bactericidal activity of aspoxicillin in an in vitro model simulating human serum levels. Japanese Journal of Antibiotics 38, 1819-26. Ogawa, M. (1986). In vivo effect on the sub-MIC (below the minimum inhibitory concentration) of antibiotics. Chemotherapy (Tokyo) 34, 1-7. Okuno, S., Maezawa, I., Sakuma, Y., Matsushita, T. & Yamaguchi, T. (1988). Effect of the hydroxyl group of the /vhydroxyphenyl moiety of aspoxicillin, a semisynthetic penicillin, on its pharmacokinetic property. Journal of Antibiotics 41, 239-46. Satta, G., Rossi, L., Bertoloni, G., Cornaglia, G., Canepari, P. & Fontana, R. (1986). Comparison of effects of penicillin minimal inhibitory and sub-inhibitory concentration on Staphylococcus aureus and Streptococcus faecium does not support the view that antibiotic sub-inhibitory concentrations can specifically interfere with bacterial virulence. Microbiologica 9, 305-19. Tani, K., Hayasaka, H., Ishii, N., Kasuga, O., Yamaguchi, T. & Oshima, S. (1984*). In vivo antibacterial activity of TA-058. Chemotherapy (Tokyo) 32, Suppl. 2, 90-8. Tani, K., Maezawa, I., Sakuma, Y., Ishii, N., Yoshida, H., Yamaguchi, T. et al. (1984a). Assay methods of TA-058 concentrations in body fluids. I. Microbiological assay method. Chemotherapy (Tokyo) 32, Suppl. 2, 99-^111. Tfiuber, M. G., Zak, O., Scheld, W. M., Hengstler, B. & Sande, M. A. (1984). The postantibiotic effect in the treatment of experimental meningitis caused by Streptococcus pneumoniae in rabbits. Journal of Infectious Diseases 149, 575-83. Ueno, K., Wantanabe, K., Bunai, M., Kobayashi, T., Aoki, M., Maezawa, I. et al. (1985). Effect of aspoxicillin on anaerobic bacteria. Japanese Journal of Antibiotics 38, 1516-28. Vogelman, B. S. & Craig, W. A. (1985). Postantibiotic effects. Journal of Antimicrobial Chemotherapy 15, Suppl. A, 37-46. Vogelman, B., Gudmundsson, S., Turnidge, J., Leggett, J. & Craig, W. A. (1988). In vivo postantibiotic effect in a thigh infection in neutropenic mice. Journal of Infectious Diseases 157, 287-98. Wagatsuma, M., Seto, M., Miyagjshima, T., Kawazu, M., Yamaguchi, T. & Ohshima, S. (1983). Synthesis and antibacterial activity of asparagine derivatives of aminobenzylpenicillin. Journal of Antibiotics 36, 147-54. Yamaguchi, T., Maezawa, I., Yoshida, H., Tani, K., Sakuma, Y., Murata. K. et al. (1984). Pharmacokinetics of TA-058 in experimental animals. Chemotherapy (Tokyo) 32, Suppl. 2, 119-32. Zak, O. & Kradolfer, F. (1979). Effects of subminimal inhibitory concentrations of antibiotics in experimental infections. Reviews of Infectious Diseases 1, 862-79.
Activity of sob-minimal inhibitory concentrations of aspoxicillin in prolonging the postantibiotic effect against Staphylococcus aureus Tadahiro Oshida, Tokio Onta, NoriyukJ Nakanfehi, Tadahiro Matsushita and Tontaro Yamaguchl
Aspoxicillin, a newly developed acylureido-penicillin with a long half-life in mouse serum of 55 min, induced postantibiotic effects (PAEs) against Staphylococcus aureus Smith of 1-7 h in vitro and 5-2 h in vivo in a thigh infection model in neutropenic mice. The long serum half-life meant that in order to evaluate the in-vivo PAE, it was necessary to examine the contribution of the drug at a sub-minimal inhibitory concentration (sub-MIQ. Growth suppression by sub-MICs of aspoxicillin was examined in vitro using either previously unexposed bacterial cells or cells which had been pre-exposed to twice the MIC of aspoxicillin for 2 h. At each sub-MIC tested, the duration of growth suppression for pre-exposed cells was longer than that for unexposed cells. In an attempt to eliminate the sub-MIC effect in vivo, penicillinase was injected into mice at the time after administration when the aspoxicillin serum concentration approached the MIC. The in-vivo PAE decreased to 2-7 h when penicillinase was injected. It was concluded that aspoxicillin induced a PAE in vivo which was additional to the effect of sub-inhibitory residual drug, but that sub-MIC levels of the drug were simultaneously involved in suppressing bacterial regrowth after the drug concentration decreased below the MIC. Similar postantibiotic subMIC effects may also occur with other long half-life antibiotics. Introduction
A postantibiotic effect (PAE) is the suppression of bacterial growth that persists after limited exposure to an antibiotic. The presence of a PAE may be an important factor to consider in the design of antimicrobial dosing regimens (Vogelman & Craig, 1985) since a prolonged PAE should permit widely-spaced intermittent dosing with intermediate periods where the antibiotic concentration falls below the MIC (Gerber et al., 1982; Gengo et al., 1984). A PAE is defined as an effect resulting from previous exposure to an antibiotic rather than an effect due to the continuing presence of a drug at sub-minimal inhibitory concentrations (sub-MICs) (Volgelman & Craig, 1985). When a PAE is determined in vitro, the effect of the drug at a sub-MIC can be excluded by dilution or washing of a bacterial culture or addition of an antibiotic-degrading enzyme such as penicillinase. In contrast, it is difficult to exclude completely a sub-MIC effect in vivo, especially with a drug that has a long serum half-life. Indeed it is often unclear whether suppression of bacterial growth in vivo results from a PAE or the effect of a drug at a sub-MIC. In addition, the effect of a drug at a sub-MIC on growth of an organism during a PAE phase has not yet been investigated. 29 0305-7453/90/070029+ 10 S02.00/0
© 1990 The British Society for Antimicrobial Chemotherapy
Downloaded from http://jac.oxfordjournals.org/ at UQ Library on June 18, 2015
Biological Research Laboratory, Tanabe Seiyaku Co., Ltd., 2-2-50 Kawagishi, Toda, Saitama 335, Japan
30
T. Oshidi et aL
Figure 1. Chemical structure of aspoxicillin.
Materials and methods Organism Staph. aureus Smith was used. Antibiotics and susceptibility tests Penicillins used in this study were aspoxicillin (Tanabe Seiyaku Co., Osaka, Japan) and piperacillin (Toyama Chem. Ind., Tokyo, Japan). MICs were determined by a broth macrodilution method with a final inoculum of 1 x 10*cfu/ml in Sensitivity test broth (STB, Nissui, Tokyo, Japan). In-vitro PAE The method of Buntdzen et al. (1981) was slightly modified for use in this study. The inocula for all procedures were prepared by incubating the organism overnight at 37°C in STB. The culture was then diluted 1:50 in fresh STB and incubated until the logarithmic phase of growth was reached. Cultures with a cell density of approximately 107cfu/ml were then exposed to various drug concentrations for 1-3 h. In order to remove residual antibiotics completely, cultures were diluted lO'-lOMbld in fresh STB and penicillinase (EC 3.5.2.6; Sigma, St. Louis, USA) added to each culture at a final concentration of 0-1 U/ml. Viable counts were determined on Sensitivity Test Agar (STA, Nissui) plates for the initial inoculum, following the removal of the antibiotics and every 1 h thereafter.
Downloaded from http://jac.oxfordjournals.org/ at UQ Library on June 18, 2015
/J-Lactam antibiotics arc known to produce PAEs against Gram-positive cocci in vitro and have also been shown to produce a PAE in vivo against Staphylococcus aureus, but not against Streptococcus pneumoniae. Tauber et al. (1984) showed that growth suppression of Str. pneumoniae after ampicillin treatment in a rabbit meningitis model was caused by the small residual amount of drug in CSF. Recently, Vogelman et al. (1988) reported, in a mouse thigh infection model, that the PAE observed with Staph. aureus and /Mactam antibiotics was independent of any residual drug, but that growth suppression of Str. pneumoniae was caused by a residual drug level. Aspoxicillin (TA-058; Figure 1) is an injectable acylureido-penicillin with a high and persistent serum level compared to other known penicillins (Wagatsuma et al., 1983; Yamaguchi et al., 1984). We therefore determined the PAEs of aspoxicillin against Staph. aureus, both in vitro and in vivo, and compared them with those of piperacillin. The relationship between the PAE and the effect due to residual drug at sub-MICs was also examined.
PostantiMotic effect of aspoxfcflUn against Staph. attrttu
31
In-vitro effect of sub-MICs on Staph. aureus in a PAE phase The log-phase culture was pre-exposed to aspoxicillin for 2 h at twice the MIC and then diluted by adding 2 ml to 30 ml of fresh STB. This culture, containing 0-125 x MIC of aspoxicillin, was divided among test tubes in 5-ml portions. Aspoxicillin was added to each tube to givefinalconcentrations of 0-25, 0-5 and 1 x MIC. Penicillinase was added for the drug-free control. For the pre-exposure control, the logarithmic phase culture was diluted and exposed to 0-125, 0-25, 0-5 and 1 xMIC of aspoxicillin, and viable counts were determined. In-vivo PAE
Calculation of PAE The duration in-vitro PAE was calculated as described by Gerber & Craig (1981), using the formula PAE = T— C, where T represents the time required for the viable count of the test culture to increase ten-fold following drug removal and C is the time required for the viable count of the drug-free control culture to increase ten-fold following similar treatment. The duration of in-vivo PAE was calculated as described by Vogelman et al. (1988), using the formula PAE = T— C—M, where M represents the time for which drug serum levels exceed the MIC, Tis the time required for the viable count in thigh homogenates of treated mice to increase ten-fold above the count at time M, and C is the time required for the viable count in thigh homogenates of untreated controls to increase ten-fold above the counts immediately following bacterial inoculation. Penicillinase injection
Penicillinase (EC 3.5.2.6) was dissolved (500 U/ml) in saline and injected into mice either 1 or 2 h after drug administration in order to eliminate any residual penicillins. Each mouse received 0-2 ml iv of the enzyme solution and 0-1 ml into the thigh muscle in which the bacteria had been inoculated. The efficacy of this treatment for eliminating aspoxicillin was tested using Escherichia coli KC-14 (MIC 0-78 mg/1; a half of Staph. aureus Smith). No growth suppression was observed with E. coli following penicillinase
Downloaded from http://jac.oxfordjournals.org/ at UQ Library on June 18, 2015
The method of Gudmundsson, Vogelman & Craig (1986) was slightly modified to determine the PAE in vivo. Male 20-25 g ddy mice were made neutropenic by a single ip administration of cyclophosphamide (250mg/kg) four days prior to investigation. This treatment reduced the number of neutrophils from 807± 161 (cells/mm3±S.D., n = 5) to 29 ± 15 on day 4 and 74 ± 49 on day 5. Log-phase bacteria were harvested just before inoculation into mice, washed once with ice-cold saline, re-suspended in saline to give an OD^o value of 002 and stored in an ice bath until use. Mice were then inoculated with 0-1 ml of this suspension (1 x 10*cfu) injected into a right thigh. Each drug was administered subcutaneously 2 h after this inoculation at a dose of 50 mg/kg. Three treated mice were killed every 2 h after bacterial inoculation. Thigh muscles, together with femoral bone, were removed and homogenized in 10 ml ice-cold saline with a Politron tissue homogenizer (Kinematica, Lucerne, Switzerland) and a 20 mm brade. Portions (0-1 ml) of diluted homogenate were then mixed with 10 ml of STA in a petri dish and viable counts subsequently determined.
32
T. OshMa el aL
injection, indicating that the drug concentration had sufficiently decreased to allow normal growth (data not shown). Pharmacokinetics Serum concentrations after a single subcutaneous administration were determined with the microbiological assay described by Tani et al. (1984a). Groups of eight neutropenic mice were killed at 15, 30, 60, 120 and 240 min after drug administration. Blood samples were taken by exsanguination from the femoral artery and vein, and sera were separated by centrifugation (1500g, 15 min, 4°C). Drug levels in sera were analysed using a two-compartment model. Antimicrobial levels in muscle interstitial fluid are known to fluctuate closely with serum concentration (Volgelman et al., 1988).
To determine whether Staph. aureus was in a PAE phase in vivo, the organism was recovered from thigh muscle 2h after aspoxicillin administration and regrowth was examined in vitro. Removed thigh muscle was first homogenized with 10 ml of saline. The homogenate of the aspoxicillin-administered mouse thigh was diluted 1:20 in fresh STB and penicillinase added to a final concentration of 0-1 U/ml. Homogenate of the drug-free mouse thigh was diluted 1:600 in fresh STB supplemented with 5% of homogenate prepared from non-infected thigh. These cultures were incubated at 37°C and viable counts determined at intervals. In order to control the effect of in-vivo to in-vitro transfer, log-phase cells which had been cultured in STB were transferred to fresh STB supplemented with 5% of the non-infected thigh homogenate. Results In-vitro PAE The MICs of aspoxicillin and piperacillin for Staph. aureus were 1-56 and 0-78 mg/1 respectively. Table I shows that both aspoxicillin and piperacillin produced an in-vitro PAE against Staph. aureus Smith. The PAEs with both penicillins were prolonged by an increase in drug concentration and exposure time. The maximum duration of the PAE was almost the same with each drug: approximately 2 h. Table L In-vitro PAE obtained with various concentrations and exposure times of aspoxicillin and piperacillin against Staph. aureus Smith Drug concentration (xMIQ 1 2 4 8 16 64
PAE(h) aspoxicillin piperacillin exposure time (h) exposure time (h) 1 2 3 1 2 3 01 04 06
01 1-3
01 1-3
03 08 11
1-7 1-3 1-7
07 1-5 1-8
1-4 1-4
08 1-3
Downloaded from http://jac.oxfordjournals.org/ at UQ Library on June 18, 2015
Regrowth o/Staph. aureus recovered from mouse thighs
33
Postantibiotic effect of aspoxfcfllln against Staph. aureus
Effect ofsub-MICs on the in-vitro PAE Figure 2 shows the effect of sub-MICs of aspoxicillin, together with the influence of pre-exposure, on the growth of Staph. aureus Smith. The duration of growth inhibition with pre-exposed cells was clearly longer than that observed with previously unexposed cells. Thus, pre-exposure to twice the MIC of aspoxicillin enhanced the susceptibility of the organism to sub-MIC levels, to the extent that a 50% decrease in the viable count of the pre-exposed cells was observed 1 h after exposure to 0-5 x MIC. The time required for the viable count to increase ten-fold following the dilution step was calculated from Figure 2 for each drug concentration; these are listed as 'growth times' in Table II. In this experiment, the PAE was 08 h. A 'sub-MIC effect', defined as the prolongation of the growth time produced by the presence of aspoxicillin at a subMIC, was calculated for the previously unexposed cells. Sub-MIC effects of up to 0-7 h were obtained with increased drug concentrations. The sub-MIC effect against pre-exposed cells is listed in Table II as the 'postantibiotic sub-MIC effect'. At each sub-MIC it was found that this effect was longer than that observed against previously unexposed cells. In particular, pre-exposure followed by exposure to 0-5 x MIC produced a very long postantibiotic sub-MIC effect of 3-6 h Table n. Postantibiotic sub-MIC effect of aspoxicillin against Staph. aureus Smith Drug concentration (xMIQ 0 0-125 0-25 05
Unexposed growth sub-MIC time effect (h) (h) 1-3 1-5 1-7 2-0
0-2 0-4 0-7
growth time (h)
Pre-exposed postantibiotic sub-MIC effect (h)
pre-exposure effect (h)
21 2-5 3-0 5-7
0-4 09 3-6
02 05 2-9
Downloaded from http://jac.oxfordjournals.org/ at UQ Library on June 18, 2015
Fipat 2. Growth curves of (a) the normal log-phase culture and (b) the pre-exposed culture of Staph. aweus in the presence of aspoxicillin at 0-125 x MIC ( # ) , 0-25 x MIC (A). 0-5 x MIC ( • ) , 1 x MIC ( T ) and control (O)< Pre-exposure was to 2 x MIC of aspoxicillin for 2 h.
34
T. Oshtda et aL 100
0-3
01
and a decrease in the viable count. The time difference between the sub-MIC effect and the postantibiotic sub-MIC effect is listed in Table II as the 'pre-exposure effect'. Preexposure effects of up to 2-9 h were obtained with increased sub-MIC drug concentrations. These resulted from the combined action of the PAE and the presence of aspoxidllin at a sub-MIC. Pharmacokinetics Serum levels of aspoxidllin and piperadllin in neutropenic mice following a single dose of 50 mg/kg are shown in Figure 3. The aspoxidllin levels were always higher than those of piperadllin and the half-life of aspoxidllin in the /?-phase was 2-6 x that of piperadllin. Although the MIC of aspoxidllin for Staph. aureus (1-56 mg/1) was higher than that of piperadllin (078 mg/1), the time for which the MIC was exceeded was longer for aspoxidllin (2-4 h) than for piperadllin (1-6 h). In addition, the time for which aspoxidllin remained at a sub-MIC level was longer than that for piperadllin. In-vivo PAE Figure 4(a) shows the growth of Staph. aureus Smith in mouse thighs. Administration of dther aspoxidllin or piperadllin decreased the viable count for the first 2 h, with further suppression of regrowth even after the drug level fell below the MIC. It therefore appeared that in-vivo PAEs were being observed. Injection of penidllinase 1 h after piperadllin administration resulted in a rapid increase in the viable count. In contrast, injection of penidllinase 2 h after aspoxicillin administration did not influence regrowth until a further 2 h had elapsed. This demonstrated that aspoxidllin produced an in-vivo PAE which was independent from the sub-MIC effect of the drug. Figure 4(b) shows a similar experiment, except that the piperacillin group recdved penidllinase 2h after drug administration. This resulted in a longer suppression of regrowth.
Downloaded from http://jac.oxfordjournals.org/ at UQ Library on June 18, 2015
Figure 3. Scrum levels of aspoxidllin ( # ) (Tu •• 55 min) and piperadllin (A) (Tu - 21 min). Mean±sx>. = 8).
Postantibiotk effect of gspoxkfltin against Stapk. aureus
c .
35
( o )
i
Drug
jb A
8
c
> * -
7
6
•&
£
C
3
|
i
1r i
[
-r I
1
U
(b) 9
Drug
I
8
7 / 6-
1
A
/ Time above MIC
5
~ Aipadc l l i n E ^ i
-2
I
2
I
i
4 Time (h)
Figure 4. Effect of penicillinase injection (|) on the growth of Staph. aureus Smith in mouse thighs following single dose drug administration: control ( • — # ) , aspoxicillin ( • — • ) and piperacillin (A — A)- (a) Penicillinase injected 2 h after aspoxicillin administration ( • • ) and 1 h after pipercQlin administration (A A)- (b) Penicillinase injected 2 h after administration of both aspoxicillin ( • •) and piperacillin (A A)- Mean±s.D. (n - 3).
Table III shows the in-vivo PAE calculated from Figure 4. The results were extrapolated to determine a PAE for the groups with prolonged growth suppression, with the exception of the penicillinase-free aspoxicillin group in Figure 4(b) for which the PAE was not determined since regrowth was not observed within 6h. In the absence of pencillinase, aspoxicillin produced a longer PAE (5-2 h) than piperacillin (3-3 h). When penicillinase was injected 2 h after drug administration, the PAE of aspoxicilhn was reduced to 2-7 h. This decrease was caused by elimination of residual drug at a concentration close to the MIC and the subsequent sub-MICs. A smaller decrease in PAE was observed with piperacillin, reflecting reduced duration for which the drug level stayed within the range of sub-MICs. When penicillinase was injected 1 h and 2 h after drug administration, the PAE of piperacillin was prolonged from 0-8 to 20h, perhaps reflecting the altered time for which the drug concentration exceeded the MIC. The longer PAE (2-7 h) observed with the penicillinase-injected aspoxicillin group was probably due to the same cause. The same phenomenon was observed with PAEs in vitro: longer exposure times produced longer PAEs.
Downloaded from http://jac.oxfordjournals.org/ at UQ Library on June 18, 2015
!
36
T. Oshid* et oL
Figure 5. In-vitro growth curves of Staph. aureus Smith recovered from mouse thighs. Each of two thigh homogenates, from atpoxicinin-treated ( • and • ) or non-treated (A and A) mice, were diluted with fresh STB and incubated. The log-phase control culture ( • • ) , previously grown in STB, was incubated in fresh STB supplemented with 5% of non-infected thigh homogenate for this experiment
Growth of the recovered organism Figure 5 shows the in-vitro growth of Staph. aureus Smith recovered from mice 2 h after aspoxicillin administration. In comparison with the control log-phase culture, growth suppression of cells recovered from non-treated mice was minimal. In contrast, growth suppression was clearly observed with cells recovered from the aspoxicillin-treated mice, thereby demonstrating that the cells were in a PAE phase in vivo. Discussion
This study demonstrated that Staph. aureus Smith was more susceptible to sub-MICs of aspoxicillin in the PAE phase than in the normal log phase. The duration of growth suppression in vitro was prolonged by a combination of the PAE and the effect of a Table HI. Effect of penicillin inactivation on the in-vivo PAE against Staph. aureus Smith
Drug Aspoxicillin Aspoxicillin Pipcracillin Piperacillin Pipcracillin
Penicillinase injection time (h)
Time above MIC(h)
2 — 1 2
2-4 20 1-6 10 1-6
"Mean values of two experiments.
PAE (h) 5-2 2-7* 3-3*
08 20
Downloaded from http://jac.oxfordjournals.org/ at UQ Library on June 18, 2015
Tim* ( h )
Postantibkrtfc effect of aspoxkfflin against Staph. aureus
37
Acknowledgements We are grateful to Dr I. Maezawa and Dr Y. Ito for their helpful discussion and critical review of the manuscript
Downloaded from http://jac.oxfordjournals.org/ at UQ Library on June 18, 2015
sub-MIC of aspoxicillin. The decrease in viable count which occurred in the presence of aspoxicillin at 0-5 x MIC following pre-exposure was not observed with previously unexposed cells. Similar viable count decreases in the presence of sub-MIC drug levels are often observed during determinations of PAEs in vivo and may result from the same effect. Gerber & Craig (1981) demonstrated that cells in the PAE phase were less susceptible to the bactericidal activity of penicillins; however their study examined the bactericidal effect of drug concentrations above the MIC, as opposed to suppression of regrowth at sub-MICs. There was little difference between the in-vitro PAEs of the two drugs tested; however, the in-vivo PAE of aspoxicillin in the absence of penicillinase was longer than that of piperacillin, probably as a result of the different pharmacokinetic properties of the two penicillins. Similarly, the longer time for which the concentration of aspoxicillin remained above the MIC appeared to produce a longer in-vivo PAE than piperacillin, even after penicillinase injection. In addition, aspoxicillin remained at a sub-MIC longer than piperaciUin. Penicillinase injection shortened the in-vivo PAEs by eliminating residual drug. The prolonged suppression of regrowth by aspoxicillin can therefore be attributed to the combined effect of the PAE and the residual drug activity at concentrations close to the MIC and the following sub-MICs. This conclusion was supported by the finding that growth suppression of Staph. aureus in the PAE phase was prolonged in vitro by the presence of the drug a.t sub-MICs. This prolonged suppression is different from the postantibiotic leucocyte enhancement described by McDonald, Wetherall & Pruul (1981) since the numbers of leucocytes were extremely low in the neutropenic mice used in this study. Vogelman el al. (1988) also examined a sub-MIC effect produced after inoculation of a log-phase culture into mouse thighs at a time when drug concentrations were below the MIC. However, since the pre-exposed culture was more susceptible than the log-phase culture, we believe that their method may result in an underestimation of any sub-MIC effect, particularly with regard to a long half-life drug. Sub-MICs of penicillins are known to produce a variety of effects against Staph. aureus, including changes in cell morphology cell proliferation or sensitivity to antibiotics (Lorian, 1975; Satta et al., 1986). The antibacterial effects of sub-MICs of drugs in experimental infection have also been studied (Zak & Kradolfer, 1979; Ogawa, 1986). When an antibiotic is administered in a single dose, there is a period during which the drug concentration falls below the MIC. It is therefore advantageous if the drug produces an additional effect at sub-MICs, such as an increased sensitivity or a PAE. Drug levels in humans tend to persist longer than in rodents and we therefore conclude that the presence of a postantibiotic sub-MIC effect should also be considered when the in-vivo PAE is determined for a long half-life drug. It has been reported that the in-vivo activity of aspoxicillin is greater than that predicted from its in-vitro antibacterial activity (Tani et al., 19846; Ueno et al., 1985). This feature has been thought to be a consequence of its good pharmacokinetic properties, which include high and prolonged serum and tissue levels (Matsushita, Kouyama & Yamaguchi, 1985; Okuno et al., 1988). We believe that the long PAE and sub-MIC effect of aspoxicillin may also contribute favourably to this characteristic.
38
T. Oshida et aL References
(Received 14 July 1989; revised version accepted 2 February 1990)
Downloaded from http://jac.oxfordjournals.org/ at UQ Library on June 18, 2015
Buntdzen, R. W., Gerber, A. U., Cohn, D. L. & Craig, W. A. (1981). Postantibiotic suppression of bacterial growth. Reviews of Infectious Diseases 3, 28-37. Gengo, F. M., Mannion, T. W., Nightingale, C. H. & Schentag, J. J. (1984). Integration of pharmacokinetics and pharmacodynamics of methicillin in curative treatment of experimental endocarditis. Journal of Antimicrobial Chemotherapy 14, 619-31. Gerber, A. U. & Craig, W. A. (1981). Growth kinetics of respiratory pathogens after short exposures to ampitillin and erythromycin in vitro. Journal of Antimicrobial Chemotherapy 8, Suppl. C, 81-91. Gerber, A. U., Wiprachtiger, P., Stettler-Spichiger, U. & Lebek, G. (1982). Constant infusions vs. intermittent doses of gentamicin against Pseudomonas aeruginosa in vitro. Journal of Infectious Diseases 145, 554-#). Gudmundsson, S., Vogelman, B. & Craig, W. A. (1986). The in-vivo postantibiotic effect of imipenem and other new antimicrobials. Journal of Antimicrobial Chemotherapy 18, Suppl. E, 67-73. Lorian, V. (1975). Some effects of subinhibitory concentrations of penicillin on the structure and division of staphylococci. Antimicrobial Agents and Chemotherapy 7, 864-70. McDonald, P. J., Wetherall, B. L. & Pruul, H. (1981). Postantibiotic leukocyte enhancement: increased susceptibility of bacteria pretreated with antibiotics to activity of leukocytes. Reviews of Infectious Diseases 3, 38-44. Matsushita, T., Kouyama, K. & Yamaguchi, T. (1985). Bactericidal activity of aspoxicillin in an in vitro model simulating human serum levels. Japanese Journal of Antibiotics 38, 1819-26. Ogawa, M. (1986). In vivo effect on the sub-MIC (below the minimum inhibitory concentration) of antibiotics. Chemotherapy (Tokyo) 34, 1-7. Okuno, S., Maezawa, I., Sakuma, Y., Matsushita, T. & Yamaguchi, T. (1988). Effect of the hydroxyl group of the /vhydroxyphenyl moiety of aspoxicillin, a semisynthetic penicillin, on its pharmacokinetic property. Journal of Antibiotics 41, 239-46. Satta, G., Rossi, L., Bertoloni, G., Cornaglia, G., Canepari, P. & Fontana, R. (1986). Comparison of effects of penicillin minimal inhibitory and sub-inhibitory concentration on Staphylococcus aureus and Streptococcus faecium does not support the view that antibiotic sub-inhibitory concentrations can specifically interfere with bacterial virulence. Microbiologica 9, 305-19. Tani, K., Hayasaka, H., Ishii, N., Kasuga, O., Yamaguchi, T. & Oshima, S. (1984*). In vivo antibacterial activity of TA-058. Chemotherapy (Tokyo) 32, Suppl. 2, 90-8. Tani, K., Maezawa, I., Sakuma, Y., Ishii, N., Yoshida, H., Yamaguchi, T. et al. (1984a). Assay methods of TA-058 concentrations in body fluids. I. Microbiological assay method. Chemotherapy (Tokyo) 32, Suppl. 2, 99-^111. Tfiuber, M. G., Zak, O., Scheld, W. M., Hengstler, B. & Sande, M. A. (1984). The postantibiotic effect in the treatment of experimental meningitis caused by Streptococcus pneumoniae in rabbits. Journal of Infectious Diseases 149, 575-83. Ueno, K., Wantanabe, K., Bunai, M., Kobayashi, T., Aoki, M., Maezawa, I. et al. (1985). Effect of aspoxicillin on anaerobic bacteria. Japanese Journal of Antibiotics 38, 1516-28. Vogelman, B. S. & Craig, W. A. (1985). Postantibiotic effects. Journal of Antimicrobial Chemotherapy 15, Suppl. A, 37-46. Vogelman, B., Gudmundsson, S., Turnidge, J., Leggett, J. & Craig, W. A. (1988). In vivo postantibiotic effect in a thigh infection in neutropenic mice. Journal of Infectious Diseases 157, 287-98. Wagatsuma, M., Seto, M., Miyagjshima, T., Kawazu, M., Yamaguchi, T. & Ohshima, S. (1983). Synthesis and antibacterial activity of asparagine derivatives of aminobenzylpenicillin. Journal of Antibiotics 36, 147-54. Yamaguchi, T., Maezawa, I., Yoshida, H., Tani, K., Sakuma, Y., Murata. K. et al. (1984). Pharmacokinetics of TA-058 in experimental animals. Chemotherapy (Tokyo) 32, Suppl. 2, 119-32. Zak, O. & Kradolfer, F. (1979). Effects of subminimal inhibitory concentrations of antibiotics in experimental infections. Reviews of Infectious Diseases 1, 862-79.
