Penicillin-aminoglycoside synergy and post-antibiotic effect for enterococci

Journal of Antimicrobial Chemotherapy (1989) 23, 189-199

Penidllin-aminoglycoside synergy and post-antibiotic effect forenterococci

T. G. Winstanley" and J. G. M. Hastings*

Departments of "Bacteriology and bMedical Microbiology, Royal Hallamshire Hospital,Glossop Road, Sheffield. S10 2JF, UK

The post-antibiotic effect (PAE) on seven strains of enterococci was studied using anovel ATP method and conventional viable counting. There was a good correlationbetween the results of the two techniques. In strains exhibiting low-level resistanceto penicillin and streptomycin, PAEs of 1 to 2-4 h (mean 1 -8 h) resulted fromexposure to penicillin; shorter PAEs were induced by streptomycin (0-2-0-3 h; mean0-2 h). Addition of streptomycin to penicillin increased the duration of penicillin-induced PAEs by two-fold to three-fold. Enterococci exhibiting high-levelstreptomycin resistance but low-level penicillin resistance did not exhibit a PAEwith streptomycin but exhibited a short PAE with gentamicin (0-3-0-6 h). For thesestrains, the addition of gentamicin, but not streptomycin, to penicillin increased thepenicillin-induced PAEs. Penicillin alone or in combination with streptomycinor gentamicin did not induce PAEs for a single strain of Enterococcus faeciumwhich exhibited high-level resistance to both penicillin and streptomycin. Duringpenicillin-induced PAEs, extracellular ATP was detectable and only during this timeperiod were enterococci susceptible to the action of gentamicin. The addition ofaminoglycosides to penicillin not only extended the PAE for enterococci but alsothe periods when organisms leaked ATP and were susceptible to growth inhibitionby gentamicin.

Introduction

Enterococci account for between 10% and 15% of cases of streptococcal endocarditis(Wilson, Zak & Sande, 1985) and successful treatment usually requires the use ofbactericidal antibiotic combinations. Conventional dosage regimens aim to maintainserum antibiotic levels above the minimum inhibitory concentration (MIC) orminimum bactericidal concentration (MBQ for the greatest proportion of the dosinginterval (Kunin, 1981). These regimens do not however take into account the post-antibiotic effect (PAE) which is the persistent suppression of growth following limitedexposure of bacteria to antimicrobial agents. We have studied the in-vitro PAEinduced by antibiotics, singly and in combination, on enterococci of varying antibioticsusceptibility profiles. In this work we have used the technique of adenosinetriphosphate (ATP) bioluminescence, an established measure of bacterial biomass(D'Eustachio & Johnson, 1968) which has been used previously to study variousinteractions between antibiotics and bacteria including the PAE on Gram-negativebacteria (Mattie, 1981; Isaksson et al., 1988).

1890305-7453/89/020189+11 $02.00/0 © 1989 The Britiih Society for Antimicrobial Chemotherapy

at University of C

alifornia, San Francisco on Decem

ber 2, 2014http://jac.oxfordjournals.org/

Dow

nloaded from

http://jac.oxfordjournals.org/

190 T. G. Winstanky and J. G. M. Hastings

Materials and methods

Bacteria

Seven strains of enterococci were studied. All were isolated from the blood of patientsat The Royal Hallamshire Hospital, Sheffield with a commercial identification system(API 20Strep, API Laboratory Products Ltd, Basingstoke, England), six strains wereidentified as Enterococcus faecalis and one (strain 284) as Ent. faecium.

Antibiotics

Benzylpenicillin (Glaxo), streptomycin (Evans Medical Ltd) and gentamicin (RousselLaboratories) were supplied as standard antibiotic powders and solutions were freshlyprepared before each experiment. MICs were determined by a macrodilution brothmethod. Serial two-fold dilutions of antibiotic in Iso-Sensitest broth (Oxoid) wereinoculated with organisms in logarithmic phase growth to achieve a starting inoculumof approximately 103 cfu/ml. Visible growth was recorded after incubation at 37°C for24 h.

Determination of PAE

Organisms (10* cfu/ml in logarithmic phase) were exposed to penicillin (10 mg/1),streptomycin (20 mg/1) or gentamicin (10 mg/1) or to various combinations of theseantibiotics in Iso-Sensitest broth. Antibiotic concentrations were chosen to representlevels achievable in patients' serum. After incubation at 37°C for 1 h, the antibioticswere removed by centrifugation, and the organisms were washed three times and thenresuspended in pre-warmed antibiotic-free broth. Control suspensions, not exposed toantibiotics, were washed and resuspended similarly. Bacterial regrowth wasquantitated by measurement of total bacterial ATP at 30 min intervals, or by viablecounts performed as described by Miles, Misra & Irwin (1938).

Assay of bacterial ATP

Total bacterial ATP was determined by mixing a 200 //I sample of the culture with anequal volume of extraction reagent, comprising trichloroacetic acid (2-5% w/v) and4 mM ethylene diamine tetra-acetic acid (EDTA). After 2 min, 20 /il of this extract wereadded to 200 /il of buffer (Tris acetate, EDTA) and 20 (A of ATP monitoring reagent(luciferin-luciferase; LKB Wallac) in the well of a microtitration tray. Light emissionwas measured in a luminometer (Amerlite Analyser, Amersham International, LittleChalfont, UK). A 10/il volume of ATP standard (001 mM, LKB Wallac) was thenadded to each well and light output again measured. The ATP concentration in thesamples was then calculated using the formula: P = IOO(R — B)/(S—R) where P wasthe ATP concentration in picomoles, R the test reading, S the reading after theaddition of ATP standard and B the background reading of the luminometer.Extracellular ATP was determined using 10/J of culture without extraction.

at University of C



Dow

nloaded from


Synergy and PAE on enterococd 191

Quantitation of PAE

The PAE was calculated from the equation PAE = T- C where T was the timerequired for the test culture to increase five picomoles of ATP or one log10 cfu abovethe value after antibiotic removal and C was the time required fof the control cultureto increase by a similar value. Preliminary data showed this value to represent thestage at which bacteria are in logarithmic growth phase.

Gentamicin challenge during PAE phase

PAE was induced by limited (1 h) exposure of bacteria to either penicillin (10 mg/1) orto a combination of penicillin and gentamicin (each 10 mg/1). Immediately afterremoval of antibiotics (PAE time 0), a 10 ml aliquot of culture was exposed togentamicin (10 mg/1) and this was repeated with separate aliquots at 30 min intervalsthroughout the PAE period. Organisms in logarithmic phase (not in PAE) were alsoexposed to gentamicin as a control. Bacterial growth was followed by determination oftotal ATP.

Results

MICs

MICs of antibiotics are shown in Table 1. Ent. faecixan 284 was resistant to penicillin,and highly resistant to streptomycin. Ent. faecalis strains were all sensitive to penicillin,but strains 250 and 286 were highly resistant to streptomycin. No organism was highlyresistant to gentamicin.

PAE for enterococci

Penicillin-induced PAEs determined by ATP bioluminescence were essentially the samein duration as those determined by viable counting (Figure 1). ATP results obtainedwith the bioluminescence method also correlated with those obtained by viablecounting when gentamicin-induced PAEs were studied (results not shown). Strain-to-strain differences were small and the results were consistent on repeated testing.

Table L MICs of antibiotics for the test strains (mg/1)

Enterococcalstrain"

284286250288549

8923281

Penicillin

64121111

AntibioticStreptomycin

£400020002000

64646464

Gentamicin

8888888

•AD Ent.faecalU, except ftrain 284.

at University of C



Dow

nloaded from


192 T. G. WinsUnley and J. G. M. Hastings

3 -

2 -

- I

.1 //

^ — ^

1

-

7/ -/ /m.

i

I 2 3

Time (h) after antibiotic removal

20

- 15

10

- 5

- 0

- 5

Figure 1. PAE induced in strain 281 by 1 h exposure to penicillin (10 mg/1). I, control, cfu/ml; II, control,total ATP; III, penicillin-treated, cfu/ml; IV, penicillin-treated, total ATP.

With the six isolates of Ent. faecalis, PAEs ranged from 1 h to 2-4 h (mean 1 -8 h)after exposure to penicillin (Table II). Streptomycin-induced PAEs were short(0-2-0-3 h) for the four strains exhibiting low-level streptomycin resistance (MIC64 mg/1). The addition of streptomycin increased the duration of the PAE produced bypenicillin two-fold to three-fold for these strains (Figure 2(a)).

Predictably, strains 286 and 250, which exhibited high-level resistance tostreptomycin, did not exhibit a PAE with streptomycin and the addition ofstreptomycin to penicillin only slightly increased the duration of the penicillin-induced

Table n. PAEs induced in entcrococci by various antibiotics singly or in combination

Strain

288549

8923281286250284

penicillin(10 mg/1)

1-8 (±016)'11-92-4 (±018)*2-21-30

PAE (hours)

streptomycin(20 mg/1)

0-20-20-30-2000

following 1

gentamicin(10 mg/1)

—

—0-60-30-5

b exposure topenicillin

(10 mg/1) andstreptomycin

(20 mg/1)

7-35-24-94-82-31-80

penicillin(10 mg/1) and

gentamicin(10 mg/1)

———4-640-2

"Mean±sj>. of five experiments.*Mean±s.D. of three experiments.

at University of C



Dow

nloaded from



15

10

5

S 0

1

_ (o)

• /

p

r1

r/

i i

/

/

i i i

115

0

5

- (b)

p

. i

inI

/ /1

i i

/

0 1 2 3 4 5 6 7Time (h) after antibiotic removal

Figure 2. (a) PAE induced in itrain 281 by 1 h exposure to penicillin (10 mg/1) and streptomycin (20 mg/1)singly and in combination. I, Control; II, streptomycin; III, penicillin; IV, penicillin plus streptomycin.Results represent total ATP values, (b) PAE induced in strain 286 by 1 b exposure to penicillin (10 mg/1),streptomycin (20 mg/1) or gentamicin (10 mg/1) singly and in combination. I, Control; II, streptomycin orgentamicin singly. III, penicillin; IV, penicillin phis streptomycin; V, penicillin plus gentamicin. Resultsrepresent total ATP values.

PAEs. Both strains exhibited only a short PAE with gentamicin but the addition ofgentamicin to penicillin increased penicillin-induced PAEs significantly (Figure 2(b)).

The single isolate of Ent. faecium, (strain 284), exhibited both high-level resistanceto streptomycin and penicillin resistance and these antibiotics did not induce a PAEeither singly or in combination. There was a short PAE with gentamicin alone but thiswas not prolonged by the addition of penicillin.

A TP release during PAE phase

PAEs induced by penicillin could also be followed by monitoring the amount of ATPreleased into the medium (Figure 3). Extracellular ATP was detectable followingremoval of penicillin or penicillin plus gentamicin for a period of time which

at University of C



Dow

nloaded from


194 T. G. WlnsUnky and J. G. M. Hastings

2 3 4 5 6Time (h) after antibiotic removal

Figure 3. Extracellular levels of ATP following 1 h exposure of strain 288 to penicillin (lOmg/1) or topenicillin plus gentamicin (each 10 mg/1). I, Control; II, penicillin; HI, penicillin plus gentamicin.

corresponded to the PAE as measured by growth inhibition (c. 2 h for penicillin and7 h for penicillin plus gentamicin with strain 288). Extracellular ATP release was not afeature of gentamicin-induced PAEs (results not shown).

Gentamicin challenge during PAE phase

The results of exposure of two enterococcal strains to gentamicin at different timeswhilst in PAE induced by penicillin are shown in Figure 4. Logarithmic phase controlcultures not in PAE were unaffected by the addition of gentamicin. At PAE time 0,growth of both organisms was markedly inhibited by gentamicin; this inhibition wasless prominent when gentamicin was added 30min later. Growth of strain 281 (PAE21 h) was affected marginally when gentamicin was added at 1-5 h but not at 2 h.Gentamicin had no effect on growth of strain 288 (PAE 1-4 h) after 1 h. When thesame strains were tested during PAEs induced by penicillin plus gentamicin, there was,again, a time-dependent inhibition in response to the addition of gentamicin(Figure 5). Organisms recovering from penicillin plus gentamicin exposure remainedsusceptible to the action of gentamicin even after 4 h, considerably longer than thesame organisms recovering from the action of penicillin alone.

Discussion

Despite differences in detailed experimental methods, the PAEs we found withpenicillin (1-3—2-4 h) and gentamicin (0-3-0-6 h) on susceptible enterococci are similarto recent data from other workers. Hessen, Pitsakis & Levison (1987) found PAEs onenterococci of 1-8 h and 0-7 h for penicillin and gentamicin, respectively. As might beexpected, we found strains showing high-level resistance to either penicillin orstreptomycin not to exhibit a PAE with these antibiotics.

The fact that PAEs determined by ATP assays were similar to those determined byviable counts raises a number of issues. It not only gives credence to the ATP method,essentially a novel technique for this type of work, but also provides confirmation of

at University of C



Dow

nloaded from



20

15

10

5

0

5

- (a)

/

1 1 1 1

1 2 3 4 5 6Time (h) after antibiotic removal

Figure 4. Growth of organisms exposed to gentamkan (10 mg/1) whilst in PAE phase induced by 1 hexposure to penicillin (10 mg/1) and in logarithmic growth phase. I, Control; II, gentamkin added to controlat to, 0; III, penicillin-induced PAE. Gentamicin was added to PAE phase organisms at time; IV, 0 rain; V,30 rain; VI, 60 min; VII, 90 min and VIII, 120 min after removal of antibiotic. Total ATP values are shownfor (a) strain 281, (b) strain 288.

observations made using conventional viable counting methods. Assessment ofbacterial biomass by viable counting is open to a number of criticisms. Antibiotic-treated organisms may adhere to each other more avidly than untreated organisms, sothat viable counts appear reduced. Organisms may be viable but unable to producecolonies on solid media. It may be, therefore, that PAEs determined by traditionalviable counts are artefactual. The fact that PAEs followed by ATP bioluminescencewere almost identical to those measured using viable counts argues against this. It alsoimplies that the ATP content of individual bacterial cells is the same during the PAEphase as in controls, indicating that metabolism is relatively unaffected.

A number of theories have been put forward to explain the PAE following transientexposure to penicillins. The PAE may represent the time required for the organism tosynthesize new penicillin-binding proteins (PBPs). However, Talbot et al. (1986)

at University of C



Dow

nloaded from


196 T. G. WinsUnky and J. G. M. Hastings

I U

7-5

5

2-5

0

_"2 '5

Si -5

(a)

. 1

/

-

i i

/

/ f'SSB.

—*"-*-~^T' 1 L *—•2I

i i i i i i i

3 4 3 6 7Time (h) after antibiotic removal

Figure 5. Growth of organisms exposed to gentamicin (10mg/l) whilst in PAE phase induced by 1 hexposure to penicillin plus gentamicin (each 10 mg/1). I, Control; II, PAE. Gentamicin added to PAE phaseorganisms at time III, 0 min, IV, 30 min; V, 60 min; VI, 180 min; VII, 210 min and VIII, 240 afterremoval of antibiotics. Total ATP values are shown for (a) strain 281, (b) strain 288.

showed that reactivation of peptidoglycan transpeptidase activity after inhibition by/Mactam agents was not due to de-novo synthesis of the PBPs. Alternatively, the PAEmay represent the time taken for the covalently bound enzyme-antibiotic complex tobreak down to re-usable enzyme molecules (Tomasz, 1979). Autolytic enzymesnormally open-up the cell wall peptidoglycan at growth points to allow newlysynthesized sections to be inserted. They normally act in a finely balanced manner,with the enzymes responsible for peptidoglycan synthesis. When the latter are inhibitedby antibiotics, autolytic enzymes are able to continue functioning. In Staphylococcusaureus, penicillin also causes lipoteichoic acid, a natural inhibitor of autolytic enzymes,to be lost from the cell (Sabath et al., 1977). Thus, penicillin-induced PAEs mayrepresent the time taken for the balance between autolytic enzymes and enzymesresponsible for peptidoglycan synthesis to be restored.

Penicillin-aminoglycoside synergism was first reported by Hunter (1947) and

at University of C



Dow

nloaded from


Synergy and PAE on cntcrococd 197

penicillin-streptomycin combinations have been used extensively in the USA to treatenterococcal endocarditis. In this study, PAEs induced by penicillin on enterococciexhibiting low-level streptomycin resistance were more than doubled by the addition ofstreptomycin. Enterococci showing high-level streptomycin resistance (MIC> 2000 mg/1) are resistant to penicillin-streptomycin synergism both in vitro and in vivo(Standiford, deMaine & Kirby, 1970; Moellering et al., 1971). Fuursted (1987) foundbactericidal synergism and prolongation of the PAE with all streptomycin-'susceptible'enterococci but with only one of five streptomycin-highly resistant strains whenchallenged with an ampicillin-streptomycin combination. Neither of the enterococcalisolates with high-level streptomycin resistance tested by us showed any significantincrease in penicillin-induced PAE when streptomycin was added. Prolongation ofPAEs was found with a penicillin-gentamicin combination with the two enterococcalstrains showing high-level streptomycin resistance. Such strains have been shown to besusceptible to a combination of penicillin and gentamicin in vitro (Watanakunakorn,1971) and in clinical practice (Weinstein & Moellering, 1973). Recent reports havedescribed production of synergistic PAEs on enterococci by combinations of penicillinand gentamicin (Hessen et al., 1986; 1987), vancomycin and gentamicin (Bush et al.,1987) and vancomycin and streptomycin (Fuursted, 1988a).

Moellering & Weinberg (1971) showed conclusively that penicillin increases theuptake of radiolabelled aminoglycosides in enterococci. Plotz & Davis (1962) suggestedthat the mechanism was an effect on the bacterial membrane and Yee, Farber & Mates(1986) suggested direct stimulation of a transport process. In the present work, themeasurement of extracellular ATP implied that penicillin-induced PAEs wereassociated with cell 'leakiness', possibly caused by autolysin activity. Addition ofgentamicin during PAE periods induced by either penicillin or penicillin plusgentamicin resulted in suppression of growth. The time period during whichgentamicin was able to cause growth suppression correlated with the length of the PAEand the period of ATP leakage from the cell, implying that the effect of gentamicin wasrelated to increased entry into the antibiotic-damaged bacterial cell. Interestingly,increased aminoglycoside lulling during PAE is not a feature of PAEs induced by non-/Mactam antibiotics. Vogelman et al. (1983) found diminished bactericidal activity ofgentamicin and tobramycin against Staphylococcus aureus, Escherichia coli andKlebsiella pneumoniae that had been exposed previously to rifampicin or erythromycin.

The possibility that prolongation of PAEs by penicillin-aminoglycosidecombinations against enterococci is purely a response to increased levels of intra-cellular aminoglycosides is supported by recent work showing that the PAE onenterococci increased in parallel with the concentration of streptomycin in anampicillin-streptomycin combination (Fuursted, 19886). Our observation that theperiod of extracellular ATP release followed the increase in PAE seen after exposure topenicillin-aminoglycoside combinations suggests that gentamicin potentiates penicillinas well as vice versa because cell leakiness was not a feature of gentamicin-inducedPAEs. This could be related to a general effect on protein synthesis.

Knowledge of the PAE permits optimal use of antibiotics and such data could haveimportant implications for the timing of doses during combination antimicrobialchemotherapy. Conventional methods for the determination of the PAE are laboriousand time-consuming. The use of ATP bioluminescence allows the PAE to bedetermined rapidly and, in addition, provides information that cannot be obtained byviable counting.

at University of C



Dow

nloaded from


198 T. G. Winstanky and J. G. M. Hastings

Acknowledgement

Thanks are due to Dr D. M. Harris and Dr R. C. Spencer for critical reading of thismanuscript.

References

Bush, L., Pitsakis, P., Boscia, J. & Kaye, D. (1987). In vitro post-antibiotic effect of LY 146032for enterococci. In Program and Abstracts of the Twenty-Seventh Interscience Conference onAntimicrobial Agents and Chemotherapy, New York, 1987. Abstract 142, p. 121. AmericanSociety for Microbiology, Washington, DC.

D'Eustachio, A. J. & Johnson, D. R. (1968). Adenosine triphosphate content of bacteria.Federation Proceedings 27, 761.

Fuursted, K. (1987) Comparison of the post-antibiotic effect of Streptococcus faecalis andStreptococcus faecium with ampicillin alone or combined with streptomycin: studies on anovel type of antimicrobial interaction Acta Pathologica Microbiologica el ImmunologicaScandmavica Section B, 95, 351-4.

Fuursted, K. (1988a). Comparative killing activity and postantibiotic effect of streptomycincombined with ampicillin, dprofloxacin, imipenem, piperacillin or vancomycin againststrains of Streptococcus faecalis and Streptococcus faecium. Chemotherapy 34, 229-34.

Fuursted, K. (19886). Synergistic effect of ampicillin or vancomycin in combination withdecreasing concentrations of streptomycin against enterococci. APMJS 96, 395-9.

Hessen, M. T., Pitsakis, P. G. & Levison, M. E. (1986). Potcntiation of post-antibiotic effectwith antibiotic combinations. In Program and Abstracts of the Twenty-Sixth InterscienceConference on Antimicrobial Agents and Chemotherapy, New Orleans, LA, 1986. Abstract584, p. 204. American Society for Microbiology; Washington, DC.

Hessen, M. T., Pitsakis, P. G. & Levison, M. E. (1987). Diminished in vitro bactericidal activityof penicillin (P) and P plus gentamicin (G) during the post antibiotic effect (PAE) inenterococci (E). In Program and Abstracts of the Twenty-Seventh Interscience Conference onAntimicrobial Agents and Chemotherapy, New York, 1987. Abstract 442, p. 173. AmericanSociety for Microbiology, Washington, DC.

Hunter, T. H. (1947). Use of streptomycin in treatment of bacterial endocarditis. AmericanJournal of Medicine 2, 436^2.

Isaksson, B., Nilsson, L., Mailer, R & Soren, L. (1988). Postantibiotic effect of aminoglycosideson Gram-negative bacteria evaluated by a new method. Journal of AntimicrobialChemotherapy 22, 23-33.

Kunin, C. M. (1981). Dosage schedules of antimicrobial agents: a historical review. Reviews ofInfectious Diseases 3, 4-11.

Mattie, H. (1981). Kinetics of antimicrobial action. Reviews of Infectious Diseases 3, 19-27.Miles, A. A., Misra, S. S. & Irwin, J. O. (1938). The estimation of the bactericidal power of the

blood. Journal of Hygiene 38, 732-49.Moellering, R. C. & Weinberg, A. N. (1971). Studies on antibiotic synergism against

enterococci. II. Effect of various antibiotics on the uptake of '^-labeled streptomycin byenterococci. Journal of Clinical Investigation 50, 2580-4.

Moellering, R. C , Wennersten, C, Medrek, T. & Weinberg, A. N. (1971). Prevalence of high-level resistance to aminoglycosides in clinical isolates of enterococci. Antimicrobial Agentsand Chemotherapy-1970, 335-40.

Plotz, P. H. & Davis, B. D. (1962). Synergism between streptomycin and penicillin: a proposedmechanism. Science 135, 1067-8.

Sabath, L. D., Laverdiere, M., Wheeler, N., Blazevic, D. & Wilkinson, B. J. (1977). A new typeof penicillin resistance of Staphylococcus aureus. Lancet i, 443-7.

Standiford, H. D., deMaine, J. B. & Kirby, W. M. M. (1970). Antibiotic synergism ofenterococci. Archives of Internal Medicine 126, 255-9.

Talbot, M. K., Schaefcr, F., Brocks, V. & Christenson, J. G. (1986). Reactivation ofpeptidoglycan synthetases after inhibition by /Mactam antibiotics. In Program andAbstracts of the Twenty-Sixth Interscience Conference on Antimicrobial Agents and

at University of C



Dow

nloaded from



Chemotherapy, New Orleans, LA, 1986. Abstract 123, p. 118. American Society forMicrobiology, Washington, DC.

Tomasz, A. (1979). From penicillin-binding proteins to the lysis and death of bacteria: a 1979view. Reviews of Infectious Diseases 1, 434-67.

Vogehnan, B., Gudmundsson, S. & Craig, W. (1983). Reduced susceptibility of bacteria duringthe postantibiotic effect (PAE) to cidal antimicrobials. In Program and Abstracts of theTwenty-Third Interscience Conference on Antimicrobial Agents and Chemotherapy, LasVegas. NV, 1983. Abstract 895, p. 249. American Society for Microbiology, Washington,DC.

Watanakunakorn, C. (1971). Penicillin combined with gentamicin or streptomycin: synergismagainst enterococd. Journal of Infectious Diseases 124, 581-6.

Weinstein, A. J. & Moelkring, R. C. (1973). Penicillin and gentamicin therapy for enterococcalinfections. Journal of the American Medical Association 223, 1030-2.

Wilson, W. R., Zak, O. & Sande, M. A. (1985). Penicillin therapy for treatment of experimentalendocarditis caused by viridans streptococci in animals. Journal of Infectious Diseases 151,1028-33.

Yee, Y., Farber, B. & Mates, S. (1986). Mechanism of penicillin-streptomycin synergy forclinical isolates of viridans streptococci. Journal of Infectious Diseases 154, 531-4.

(Received 23 September 1988; accepted 30 October 1988)

at University of C



Dow

nloaded from


Documents

Penicillin-aminoglycoside synergy and post-antibiotic effect for enterococci