Erythromycin and Azithromycin Transport into Haemophilus influenzae ...

5 downloads 59 Views 870KB Size Report
CAPOBIANCO* AND ROBERT C. GOLDMAN. Anti-Infective Research Division, Abbott Laboratories, Abbott Park, Illinois 60064-3500. Received 26 January ...
ANTIMICROBIAL AGENTS AND CHEMOTHERAPY, Sept. 1990, p. 1787-1791

Vol. 34, No. 9

0066-4804/90/091787-05$02.00/0 Copyright © 1990, American Society for Microbiology

Erythromycin and Azithromycin Transport into Haemophilus influenzae ATCC 19418 under Conditions of Depressed Proton Motive Force (A'H) JOHN 0. CAPOBIANCO* AND ROBERT C. GOLDMAN Anti-Infective Research Division, Abbott Laboratories, Abbott Park, Illinois 60064-3500 Received 26 January 1990/Accepted 29 May 1990

The effect of collapsing the electrochemical proton gradient (AjiH) on [3H]erythromycin and ['4C]azithromycin transport in Haemophilus influenzae ATCC 19418 was studied. The proton gradient and membrane potential were determined from the distribution of [2-14C]dimethadione and rubidium-86, respectively. A.LH was reduced from 124 to 3 mV in EDTA-valinomycin-treated cells at 22°C with 150 mM KCI and 0.1 mM carbonyl cyanide m-chlorophenylhydrazone. During the collapse of A(LH, macrolide uptake increased. Erythromycin efflux studies strongly suggested that this increase was not due to an energy-dependent efflux pump but was likely due to increased outer membrane permeability. These data indicated that macrolide entry was not a AiiH-driven active transport process but rather a passive diffusion process.

Intersci. Conf. Antimicrob. Agents Chemother., abstr. no. 648, 1988) and azithromycin (18), with potential for general use against H. influenzae required a closer look at the macrolide transport process. In particular, clarithromycin contains one nitrogen and azithromycin contains two nitrogens which are protonated at physiological pH, raising the question of the role of membrane potential (A*4) (inside negative) in uptake. We have investigated the effects of ALiH on macrolide accumulation and the possibility of a macrolide efflux pump in H. influenzae ATCC 19418. Our findings are consistent with the hypothesis that these antibiotics passively enter bacterial cells and accumulate intracellularly with the assistance of the internal ribosomal binding site. H. influenzae ATCC 19418 was grown in brain heart infusion broth plus 5% Fildes enrichment or Mueller-Hinton broth plus 0.5% yeast extract, 0.2% glucose, 15 jig of P-NAD per ml, and 15 ,ug of hematin per ml at 37°C to a cell density of 2 x 109 to 6 x 109 CFU/ml. Cells were harvested by centrifugation (5,000 x g for 10 min) and washed once with Mueller-Hinton medium in the drug efflux experiments or with 120 mM Tris hydrochloride (pH 8.0) in the A1iH studies. For the concurrent study of macrolide transport and protein synthesis inhibition, mid-log cells (2 x 109 CFU/ml) in Mueller-Hinton medium were incubated at 37°C for 3 h with 4 ,ug of [N-methyl-3H]erythromycin (39 mCi/mmol) per ml or 1.5 ,ug of [N-methyl-'4C]azithromycin (7 mCi/mmol) per ml. These concentrations represent three times the MICs for this organism. MICs were determined by broth dilutions with inocula of 105 and 106 CFU/ml at 37°C for 24 h. In the transport studies, cells (1 ml) were removed from incubation at 10- to 20-min time intervals, and the cells were separated from medium on glass fiber filters (GFF; Whatman, Inc.). Trapped cells were washed three times with 10 ml of 0.9% saline, placed in scintillation vials, and solubilized with 0.5 ml of NCS tissue solubilizer (Amersham Corp.) at 50°C for 30 min. Instagel (10 ml; Packard Instrument Co., Inc.) was added, and samples were counted in a Tri-carb 300 scintillation counter (Packard). At various time points, cell-associated radioactivity was adjusted for counts at time zero (cells plus label at 4°C for 4.1 0.36 >4.7

150

EDTA-valinomycin-

3

0.63 >3.7 0.55 >4.4

Cell treatment

ErythromycinC at:

22°C 37°C 22°C 37°C

1

1.1 0.07 3.4 0.06 3.7 0.07

CCCPd a 5 mM Tris HEPES (pH 7.5)-16 mM succinate. b Azithromycin was tested at 0.5 pLg/ml. c Erythromycin was tested at 1.0 ,ug/ml. d

0.1 mM CCCP.

0.4 1.5 1.1

1789

fluidity changes which allow for greater penetration of lipophilic agents (21). The rate increases observed between the 22 and 37°C incubations of EDTA-treated cells may be due to Tris EDTA-mediated outer membrane structural damage (16) occurring at 37°C. Uptake rates increased fiveto eightfold for both macrolides at 22°C when KCl concentrations were increased to 150 mM without (0.44 pmol/,ll per min for azithromycin; 0.36 pmol/,l per min for erythromycin) or with (0.63 pmol/,ul per min for azithromycin; 0.55 pmol/,ll per min for erythromycin) CCCP pretreatment. At 37°C, uptake was so rapid that only minimum rates could be established. These data demonstrated that at 22°C, under conditions of AjiH collapse, macrolide uptake did not decrease; therefore, drug influx was not an active transport process. The increase in macrolide uptake as A,uH was reduced may indicate an energy-dependent efflux system or an increase in outer membrane permeability due to a decreased Donnan potential across this barrier (16). We investigated the possibility of a AjiH-driven macrolide efflux pump in H. influenzae. Erythromycin was used to

8

3

0 -J .1

w C)

.5 0

E

0

20

40

60

80

100

MIN AT 370C FIG. 1. Erythromycin efflux from CCCP-treated and untreated H. influenzae ATCC 19418 cells. Cells (5 x 1010 CFU/ml) in MuellerHinton medium were incubated at 37°C for 30 min with 1.5 ,g of [3H]erythromycin per ml. Labeled cells were harvested (arrow) and suspended in fresh medium (37°C) containing unlabeled antibiotic. Symbols: l, control; *, 0.1 mM CCCP. Results are expressed as the mean + standard deviation (n = 3). (Inset) Macrolide efflux plot of ln (CL/CLO) versus t. The efflux data were plotted to determine the drug efflux t1/2. CL, Cellular-label concentration at time t; CLo, cellular label concentration at zero time (40-min sample). t1/2, 0.693/slope.

1790

NOTES

ANTIMICROB. AGENTS CHEMOTHER.

12-

o

y

(

0

=-0036 - 0.1462X R =1.00

-9 -2

.1 Lu

C) -3

y = 0.133- 0.1356x R =0.98 ~~~~~~~~~~~~~~~~(El) 5

0

O E

10

15

20

MIN AT 370C

0.

3

0

20

40

60

80

100

MIN AT 370C FIG. 2. Erythromycin efflux from 150 mM KCl- and 1 mM KCl-exposed EDTA-valinomycin-treated cells. Cells (5 x 1010 CFU/ml) in 5 mM Tris hydrochloride (pH 7.5) containing 1 mM KCl, 16 mM succinate, and 150 mM choline chloride were preloaded with 1.5 ,ug of [3H]erythromycin per ml at 370C for 30 min. Labeled cells were harvested (arrow) and suspended in fresh buffer (37°C) containing 1 mM KCl (El) or 150 mM KCl (*). Results are expressed as the mean ± standard deviation (n = 3). (Inset) Macrolide efflux plot of ln (CL/CLO) versus t. The efflux data were plotted to determine the drug efflux t1y2, as described in the legend to Fig. 1.

measure drug efflux in this organism, because erythromycin has a faster ribosomal binding off rate than azithromycin (R. Goldman, unpublished data), allowing for the observation of any decrease in efflux on deenergizing. Drug efflux was measured in Mueller-Hinton medium with or without the energy uncoupler CCCP in intact cells preloaded with 1.5 ,ug of [3H]erythromycin per ml at 37°C for 30 min (Fig. 1). The zero time for kinetic analysis was 40 min into the incubation, since this represented the first measurable time period following centrifugation and suspension (30- to 40-min time points). This processing period allowed for the decoupling effect of CCCP. The differences in initial intracellular macrolide concentrations between CCCP-treated and untreated cells at the 40-min time point may be due to cell loss during the processing period. This discrepancy was not germane to the analysis of the dissociation rate constant. The efflux of drug followed first-order kinetics for the untreated control cells with a t1/2 of 9.8 min, determined from a plot of ln (CL/CLO) versus t (Fig. 1, inset). Similar observations were made with CCCP-treated cells for the initial 30 min (t112, 10.8 min), after which time the rate of efflux appears to decrease.

The last two time points on the efflux curve (Fig. 1) gave measurements which were close to background and therefore not reliable for kinetic analysis. Drug efflux was also measured in EDTA-valinomycin treated cells preloaded with [3H]erythromycin, as described above (Fig. 2). Because of the permeabilized outer membrane caused by EDTA treatment, the efflux rates were higher than in intact cells; however, the efflux of drug still followed first-order kinetics, with t1,2s of 5.1 and 4.6 min for cells in 1 and 150 mM KCl, respectively (Fig. 2, inset). The t112 data for the EDTAvalinomycin-treated cells approach values observed in ribosome-binding studies (R. Goldman, unpublished data). Although our observations did not detect a macrolide efflux pump in this organism, we cannot rule out the existence of an inefficient efflux system masked by the ribosome binding off rate. There have been efflux pumps detected in E. coli for tetracycline (13) and norfloxacin (4), but as yet there are no published data for macrolides. In summary, the driving force for the accumulation of erythromycin and azithromycin into H. influenzae was not dependent upon A/iH but, rather, was dependent on intra-

NOTES

VOL. 34, 1990

cellular binding to ribosomes. Therefore, macrolide transport into this organism was by a passive process. Since an energy-dependent macrolide efflux pump was not detected, the increase observed in antibiotic uptake under conditions of depressed A,uH was likely the result of increased outer membrane permeability. The inhibition of protein synthesis paralleled macrolide uptake, confirming drug entry into these cells. We thank J. Martin and P. Johnson for the MIC determinations and the velocity-gradient centrifugation methods which were adapted from tissue culture procedures and modified for bacterial separations. We also thank G. Rotert for the radiolabeled macrolides. 1.

2.

3. 4.

5.

6.

7.

8.

9. 10. 11.

LITERATURE CITED Azzone, G. F., D. Pietrobon, and M. Zoratti. 1984. Determination of the proton electrochemical gradient across biological membranes. Curr. Top. Bioenerg. 13:1-76. Bakker, E. P. 1982. Membrane potential in a potassium transport-negative mutant of Escherichia coli K-12. The distribution of rubidium in the presence of valinomycin indicates a higher potential than that of the tetraphenylphosphonium cation. Biochim. Biophys. Acta 681:474-483. Barre, J., M.-P. Fournet, R. Zini, L. Deforges, J. Duval, and J.-P. Tillement. 1986. In vitro [3H]erythromycin binding to Staphylococcus aureus. Biochem. Pharmacol. 35:1001-1004. Cohen, S. P., D. C. Hooper, J. S. Wolfson, K. S. Souza, L. M. McMurry, and S. B. Levy. 1988. Endogenous active efflux of norfloxacin in susceptible Escherichia coli. Antimicrob. Agents Chemother. 32:1187-1191. Dette, G. A., H. Knothe, and S. Kaula. 1987. Modifying effects of pH and temperature on (14C)erythromycin uptake into Staphylococcus aureus-relation to antimicrobial activity. Zentralbl. Bakteriol. Hyg. A 265:393-403. Fierro, J. F., C. Hardisson, and J. A. Salas. 1988. Involvement of cell impermeability in resistance to macrolides in some producer streptomycetes. J. Antibiot. 41:142-144. Godfrey, A. J., and L. E. Bryan. 1984. Intrinsic resistance and whole cell factors contributing to antibiotic resistance, p. 113145. In L. E. Bryan (ed.), Antimicrobial drug resistance. Academic Press, Inc., New York. Harold, F. M., E. Paviasova, and J. R. Baarda. 1970. A transmembrane pH gradient in Streptococcus faecalis: origin, and dissipation by proton conductors and N,N'-dicyclohexylcarbodiimide. Biochim. Biophys. Acta 196:235-244. Inzana, T. J. 1983. Electrophoretic heterogeneity and interstrain variation of the lipopolysaccharide of Haemophilus influenzae. J. Infect Dis. 148:492-499. Johnson, J. D., W. L. Hand, J. B. Francis, N. King-Thompson, and R. W. Corwin. 1980. Antibiotic uptake by alveolar macrophages. J. Lab. Clin. Med. 95:429-439. Lampson, B. C., W. von David, and J. T. Parisi. 1986. Novel mechanism for plasmid-mediated erythromycin resistance by

12. 13. 14.

15. 16. 17. 18.

19. 20.

21.

22.

23.

24.

25. 26.

1791

pNE24 from Staphylococcus epidermidis. Antimicrob. Agents Chemother. 30:653-658. Mao, J. C.-H., and M. Putterman. 1968. Accumulation in Gram-positive and Gram-negative bacteria as a mechanism of resistance to erythromycin. J. Bacteriol. 95:1111-1117. McMurry, L. M., D. A. Aronson, and S. B. Levy. 1983. Susceptible Escherichia coli cells can actively excrete tetracyclines. Antimicrob. Agents Chemother. 24:544-551. Menninger, J. R., and D. P. Otto. 1982. Erythromycin, carbomycin, and spiramycin inhibit protein synthesis by stimulating the dissociation of peptidyl-tRNA from ribosomes. Antimicrob. Agents Chemother. 21:811-818. Muto, Y., K. Bandoh, K. Watanabe, N. Katoh, and K. Ueno. 1989. Macrolide accumulation by Bacteroides fragilis ATCC 25285. Antimicrob. Agents Chemother. 33:242-244. Nikaido, H., and M. Vaara. 1985. Molecular basis of bacterial outer membrane permeability. Microbiol. Rev. 49:1-32. Padan, E., D. Zilberstein, and H. Rottenberg. 1976. The proton electrochemical gradient in Escherichia coli cells. Eur. J. Biochem. 63:533-541. Retsema, J., A. Girard, W. Schelkly, M. Manousos, M. Anderson, G. Bright, R. Borovoy, L. Brennan, and R. Mason. 1987. Spectrum and mode of action of azithromycin (CP-62,993), a new 15-membered-ring macrolide with improved potency against gram-negative organisms. Antimicrob. Agents Chemother. 31:1939-1947. Rottenberg, H. 1975. The measurement of transmembrane electrochemical proton gradients. J. Bioenerg. 7:61-74. Sabath, L. D., V. Lorian, D. Gerstein, P. B. Loder, and M. Finland. 1968. Enhancing effect on alkalinization of the medium on the activity of erythromycin against gram-negative bacteria. Appl. Microbiol. 16:1288-1292. Stein, W. D. 1981. Permeability for lipophilic molecules, p. 1-28. In S. L. Bonting and J. J. dePont (ed.), Membrane transport. Elsevier/North-Holland Biomedical Press, Amsterdam. Taubman, S. B., N. R. Jones, F. E. Young, and J. W. Corcoran. 1966. Sensitivity and resistance to erythromycin in Bacillus subtilis 168: the ribosomal binding of erythromycin and chloramphenicol. Biochim. Biophys. Acta 123:438-440. Tucker, A. N., and D. C. White. 1970. Release of membrane components from viable Haemophilus parainfluenzae by ethylenediaminetetraacetic acid-Tris(hydroxymethyl)-aminomethane. J. Bacteriol. 102:498-507. Vachon, V., D. J. Lyew, and J. W. Coulton. 1985. Transmembrane permeability channels across the outer membrane of Haemophilus influenzae type b. J. Bacteriol. 162:918-924. West, I., and P. Mitchell. 1972. Proton-coupled ,-galactoside translocation in non-metabolizing Escherichia coli. J. Bioenerg. 3:445-462. Williams, L. T., and R. J. Lefkowitz. 1978. Theory of ligandreceptor interactions, p. 27-41. In L. T. Williams and R. J. Lefkowitz (ed.), Receptor binding studies in adrenergic pharmacology. Raven Press, New York.