(g) 1990 S. Karger AG, Basel 0009-3157/90/0366-0422 $ 2.75/0

Chemotherapy 1990;36:422-427

Effect of Azithromycin, Roxithromycin and Erythromycin on Human Polymorphonuclear Leukocyte Function against Staphylococcus aureus A. Pascual, G. López-López, J. Aragón, E.J. Perea Department of Microbiology, School of Medicine, Seville, Spain

Key Words. Azithromycin • Roxithromycin • Erythromycin • Staphylococcus aureus

Introduction The phagocytosis of bacteria by human polymorphonuclear leukocytes (PMNs) constitutes one of the most important host defense mechanisms against Staphylococ­ cus aureus. This process can be modified by different antimicrobial agents [1]. Even at subinhibitory concentrations some anti­ microbial agents induce morphological and functional changes on some bacteria rendering them more susceptible to phago­ cytosis [1,2]. Other antimicrobials are able

to penetrate into phagocytes and remain active, this phenomenon being very im­ portant in the fight against intracellular pathogens [3]. In the last years there has been an in­ creasing interest in the macrolides, and new compounds have been developed. Among them, roxithromycin and azith­ romycin have been shown to offer some advantages in terms of half-life and tissue distribution over erythromycin, and, moreover, these antimicrobials penetrate into phagocytes reaching intracellular

Downloaded by: Univ. of California Santa Barbara 128.111.121.42 - 3/8/2018 12:16:53 PM

Abstract. The effect of three macrolides (azithromycin, roxithromycin and erythro­ mycin) on the interaction in vitro of human polymorphonuclear leukocytes (PMNs) with Staphylococcus aureus was examined. The exposure of S. aureus to 0.25 MIC of roxithromycin and erythromycin but not of azithromycin significantly increased the up­ take of opsonized bacteria by human PMNs. The preincubation of PMNs with 1,10 and 25 mg/1 of the three antimicrobial agents did not affect either the uptake of S. aureus or the superoxide radical production by human PMNs. At these same concentrations the three agents showed slight but not significant intracellular activity in PMNs against S. aureus. It is concluded that treatment of S. aureus with subinhibitory concentrations of roxithromycin and erythromycin enhanced phagocytosis by PMNs, but the three macrolides tested did not directly affect the functions of human PMNs against S. aureus.

concentrations considerably higher than those of erythromycin [4], There are, how­ ever, some controversies on the intracellu­ lar activity of these compounds [3, 5]. The purpose of this study was to evalu­ ate the effect of azithromycin and roxith­ romycin in comparison with that of eryth­ romycin on the interaction of human PM Ns with Staphylococcus aureus. Firstly, the phagocytic susceptibility of S. aureus exposed to subinhibitory concentrations of these agents was investigated. Secondly, the effect of the macrolides on the uptake of S. aureus and superoxide radical pro­ duction by human PM Ns was evaluated. Finally, the intracellular activity of these agents against S. aureus was assessed. Material and Methods Bacteria and Sensitivity Testing S. aureus ATCC 25923 was used for the study. The minimal inhibitory concentrations (MICs) of azithromycin (Pfizer), roxithromycin (Roussell) and erythromycin (Abbott) were determined by the broth dilution method. MICs were as follows: azithromy­ cin and roxithromycin, 0.5 mg/1; erythromycin, 0.25 mg/1. To evaluate the effect of subinhibitory concentrations of these antimicrobials, the men­ tioned strain was grown overnight at 37°C in Mueller-Hinton broth (Difco Laboratories, Detroit, Mich., USA) in the presence or absence of 0.25 MIC of the different antimicrobials. Isolation o f PM Ns Human PMNs were recovered from heparinized venous blood of healthy donors and purified using a previously described method [6]. PMN preparations were 97% pure. Final cell suspensions were adjusted to 5 x 106 PMNs/ml in Hanks’ balanced salt solution containing 0.1% gelatin (GHBSS). Cells were 95% vi­ able by trypan blue exclusion criteria. Phagocytosis Assay The uptake of radioactively labelled bacteria by PMNs was measured by the method of Verhoef et al.

423

[6]. The bacteria were labelled by adding 10 pCi JHadenine (Amersham International, UK) to 5 ml Mueller-Hinton broth. After overnight incubation, the bacteria were washed three times in phosphatebuffered saline (PBS) and adjusted photometrically to yield 2.5 x 10s CFU /m l. For opsonization, radio­ actively labelled bacteria were incubated in 5% pooled human serum for 15 min. At indicated times, opsonization was stopped by the addition of 2.5 ml ice-cold PBS. The opsonized bacteria were centri­ fuged for 15 min at 1,600 g and resuspended in GHBSS. A 0.1-ml amount of preopsonized staphylo­ cocci ( 5 x l0 7/m l) was incubated with 0.1 ml of PMNs (5 x 106/m l) in a 37°C shaking water bath. Af­ ter 2, 6 and 12 min, phagocytosis was stopped by the addition of 2.5 ml ice-cold PBS. The percentage of staphylococci taken up by the PMNs at each time in­ terval was calculated from the uptake of radioactivity determined in a separate vial. Non-phagocyte-associated radioactive material was separated from PMN-associated radioactive material by three cycles of differential centrifugation (160 g). Radioactivity was determined by liquid scintillation counting. To determine the total bacteria-associated radioactivity (representing leukocyte-associated bacteria plus extracellular bacteria) duplicate 0.1-ml samples were taken at the end of the assay period, placed in 2.5 ml ice-cold PBS and centrifuged at 1,600 g for 15 min. The pellets were resuspended in 2.5 ml scintillation liquid and counted. This assay was performed using bacteria exposed to 0.25 MIC of the different agents. In another series of experiments, PMNs were incubated (30 min, 37 °C) with different concentrations (1, 10 and 25 mg/1) of azithromycin, roxithromycin and eryth­ romycin before assessing the uptake of radiolabelled bacteria. Superoxide Production After incubation with different concentrations of the antimicrobial agents at 37 °C, the PMN were stim­ ulated with phorbol myristate acetate (200 n M), and superoxide production was measured using a modifi­ cation of the ferricytochrome c reduction microassay described by Pick and Mizel [7], Briefly, aliquots con­ taining 2.5 x 10s PMNs in 130 pi of GHBSS with or without antimicrobial agents were pipetted into wells of a 96-well flat-bottomed microtiter plate (Flow Lab­ oratories, USA). Half of the cell suspensions con­ tained 60 units of superoxide dismutase (Sigma,

Downloaded by: Univ. of California Santa Barbara 128.111.121.42 - 3/8/2018 12:16:53 PM

Effect of New Macrolides on Neutrophil Functions

424

Pascual/López-López/Aragón/Perea

Fig. 1. Effect of exposure of S. aureus to 0.25 MIC of azythromycin ( • ) , roxithromycin (■) and erythromycin (A) on the uptake of opsonized bacteria by human PMNs. O = Control, a 5% HPS. b 10% HPS. * p < 0.05. Exposure, min

USA). Plates were then incubated for 30 min at 37°C (5% C 0 2). Finally, 70 [xl of a suspension containing PMNs or GHBSS (controls) and 60 units of cyto­ chrome c (Sigma, USA) were added to the corre­ sponding wells, and the plates were again incubated for 60 min at 37 °C. Absorbance was read at 550 nm on a Multiskan MC photometer (Flow Laboratories, USA). The amount of superoxide was expressed as nanomoles superoxide per milligram of protein. PMN protein was determined by the method of Lowry et al. [8]. Every experiment was done using PMNs from at least four donors, and four replicates were tested in each experiment.

agar. Colonies were counted after 24 h of incubation at 37 °C. The data were expressed as the percentage of surviving staphylococci compared to the controls (without antimicrobial agent) at time zero. In addi­ tion to determining bacterial survival, morphologic studies were routinely performed at time zero and af­ ter 3 h of incubation to evaluate any changes in bac­ teria (cell-associated or extracellular). Fifty-microli­ ter samples were removed from biovials and depo­ sited on glass slides. After staining with Wright’s stain, samples were examined by light microscopy. All assays were performed in duplicate using PMNs from six different donors.

Intracellular Activity o f Antimicrobial Agents To evaluate the intracellular activity of antimi­ crobials, a previously described method was used [9]. Briefly, 0.1 ml of opsonized bacterial suspension (5 x 107 CFU /m l) and 0.1 ml of PMNs ( 5 x l0 6 PMNs/ml) were combined in a series o f polypropyl­ ene biovials (Beckman Instruments, USA) and incu­ bated in a shaker incubator (250 rpm) for 60 min at 37°C. Subsequent to incubation, extracellular bacte­ rial were removed by differential centrifugation. Cells were then resuspended in 0.2 ml of RPMI me­ dium (Gibco, UK). At this time (designated time zero) the different antimicrobial agents were added and the vials reincubated in a shaker (250 rpm) at 37 °C. Vials were revomed at time zero (control) and after incubation for 3 h (control and samples with an­ timicrobial agents). Cells were lysed in distilled wa­ ter, and samples were diluted and pour-plated in

Statistical Analysis o f Data Data were expressed as means ± SD. Statistical analysis was performed by means of Student’s t test, p ^ 0.05 being considered significant.

Results Phagocytosis of S. aureus The exposure of S. aureus to 0.25 MIC of roxithromycin and erythromycin signi­ ficantly increased the uptake of opsonized S. aureus (fig. 1). This phenomenon was not observed with azithromycin-exposed bacteria. Preincubation of PMN, instead of bacteria, with different concentrations

Downloaded by: Univ. of California Santa Barbara 128.111.121.42 - 3/8/2018 12:16:53 PM

Exposure, min

Effect of New Macrolides on Neutrophil Functions

425

Table I. Effect of azithromycin, roxithromycin and erythromycin on superoxide radical production by human PMNs (n = 4)

Table 2. Effect of azithromycin, roxithromycin and erythromycin on the survival of S. aureus within human PMNs(n = 5)

Antimicrobial

Antimicrobial

Roxithromycin

Erythromycin

1 10 25 1 10 25 1 10 25

Superoxide anions, nmol/mg protein 194 ± 25 181 ± 19 170123 148131 180149 175137 165149 190121 187123 190121

of the three macrolides did not affect the uptake of S. aureus by these phagocytes (data not shown). Superoxide Anion Production The preincubation of PMNs with 1, 10 and 25 mg/1 of the three antimicrobials for 30 min at 37°C did not significantly affect the production of superoxide radi­ cals by these phagocytes as was measured by a ferricytochrome c reduction assay (table 1). Intracellular Activity The intracellular activity of different concentrations of azithromycin, roxith­ romycin and erythromycin against S. au­ reus was evaluated in a 3-hour assay (table 2). The three antimicrobials tested induced a slight but not significant reduction in the survival of intracellular S. aureus at both 90 and 180 min.

No antimicrobial Azithromycin

Roxithromycin

Erythromycin

Concen­ tration mg/l

90 min

180 min

1 10 25 1 10 25 1 10 25

38.71 12 34.3111 31.3110 30.91 10 35.5110 32.0111 31.1112 32.1112 28.1113 28.01 13

25.517 18.717 16.116 15.016 18.317 15.116 15.017 17.317 14.515 15.117

S. aureus, %

Discussion The effect of three macrolides, azith­ romycin, roxithromycin and erythromy­ cin, on human PMN activity against S. aureus has been evaluated. The expo­ sure of these microorganisms to 0.25 MIC of roxithromycin and erythromycin in­ creased the uptake by human PMNs. This effect seems to be the result of a direct in­ fluence on opsonization, since these anti­ microbials did not affect the uptake of un­ opsonized bacteria as has been observed with other antimicrobials [10]. Azithromy­ cin, which differs structurally from eryth­ romycin by a methyl-substituted nitrogen in the macrolide ring, did not show this ef­ fect. This discrepancy may be related to the reduced in vitro activity of azithromy­ cin, as compared to roxithromycin and erythromycin at low pH values. Studies are being performed in order to study the

Downloaded by: Univ. of California Santa Barbara 128.111.121.42 - 3/8/2018 12:16:53 PM

No antimicrobial Azithromycin

Concentration mg/l

effect of the subinhibitory concentrations of these three antimicrobial agents on the cell wall of S. aureus directed to find out the possible explanation of these differ­ ences observed in the opsonization of anti­ microbial-exposed bacteria. The oxygen-dependent bactericidal mechanism of human PMNs has been shown to be essential for the killing of S. aureus [11]. The exposure of PMNs to different concentrations of the three macrolides did not affect the production of su­ peroxide radicals by these cells. Similar re­ sults were obtained by other authors eval­ uating the hydrogen peroxide production and nitroblue tétrazolium reduction as­ says by human phagocytes in the presence of azithromycin [12]. Labro et al. [13] ob­ served, however, that high concentrations of roxithromycin (50 and 100 mg/1), higher than the therapeutic ones, strongly impaired the oxidative burst of PMNs as­ sessed by chemiluminescence, superoxide anion generation and myeloperoxidasemediated iodination of proteins. Erythromycin, roxithromycin and azithromycin have been shown to reach high intracellular concentrations. The cellular/extracellular concentration ratios described in human PMNs have been around 10 for eryhtromycin, 20-30 for roxithromycin and even higher than 200 for azithromycin [3,4, 12]. The intracellular activity of macrolides, however, is very controversial, and the re­ sults obtained were conflicting [3, 5, 14]. We tried to evaluate whether or not this in­ creased penetration of the new macrolides induced higher intracellular activity. Our results showed that even at a high concen­ tration, azithromycin and roxithromycin showed a poor intracellular activity

Pascual/López-López/Aragón/Perea

against S. aureus, similar to that observed with erythromycin. It must be empha­ sized, however, that the relatively short time of exposure (3 h) of intracellular bac­ teria to the antimicrobial agent is not re­ presentative of the therapeutic situation, especially with the new macrolides, but on the other hand it is very difficult to keep S. aureus intracellularly in PMNs at longer incubation periods. The limitations of the colony-counting method and the susceptibility of the strain used in this study to the bactericidal mechanisms of PMNs can be other factors to be consid­ ered, but we have observed the intracellu­ lar activity of other antimicrobials such as ofloxacin, ciprofloxacin and rifapentine using the same model [9, 14, 15]. A possi­ ble explanation of this discrepancy be­ tween the high intracellular penetration and poor activity of the macrolides within the phagocytes could be related to a loss of activity of these basic compounds induced by the low intraphagolysosomal pH, as has been demonstrated for clindamycin [14] and some macrolides [5]. In summary erythromycin and roxith­ romycin but not azythromycin could indi­ rectly affect the phagocytic process by altering the opsonization of S. aureus, in­ creasing its susceptibility to phagocytosis by human PMNs. At therapeutic concen­ trations these agents did not directly affect the bactericidal mechanisms of PMNs and did not show significant intracellular ac­ tivity against S. aureus. Acknowledgements The authors thank M.C. Guzman for her techni­ cal assistance and C. Acosta for her assistance with the preparation of the manuscript.

Downloaded by: Univ. of California Santa Barbara 128.111.121.42 - 3/8/2018 12:16:53 PM

426

Effect of New Macrolides on Neutrophil Functions

1 Milatovic D: Antibiotics and phagocytosis. Eur J Clin Microbiol 1983:2:414-425. 2 Elliot GR, Peterson PK, Verbrugh HA, et al: In­ fluence of subinhibitory concentrations of peni­ cillin, cephalotin and clindamycin on Staphylo­ coccus aureus growth in human phagocytic cells. Antimicrob Agents Chemother 1982:22:781-784. 3 Hand WL, King-Thompson NL: Contrast be­ tween phagocyte antibiotic uptake and subse­ quent intracellular bactericidal activity. Antimi­ crob Agents Chemother 1986:29:135-140. 4 Hand WL, King-Thompson N, Holman JW: En­ try of roxithromycin (RU 965), imipenem, cefo­ taxime, trimethoprim and metronidazole into hu­ man polymorphonuclear leukocytes. Antimicrob Agents Chemother 1987;31:1553-1557. 5 Milatovic D: Intraphagocytic activity of eryth­ romycin, roxithromycin and azithromycin. Eur J Clin Microbiol Infect Dis 1990;9:33-35. 6 Verhoef J, Peterson PK, Quie PG: Kinetics of staphylococcal opsonization, attachment, inges­ tion and killing by human polymorphonuclear leukocytes: A quantitative assay using (3H)-thymidine-labelled bacteria. J Immunol Methods 1977:14: 303-311. 7 Pick E, Mizel D: Rapid microassay for the mea­ surement of superoxide and hydrogen peroxide production by macrophages in culture using an automatic enzyme immunoassay reader. J Immu­ nol Methods 1981:46:211-226. 8 Lowry OH, Rosebrough NJ, Farr AL, et al: Pro­ tein measurement with the Folin phenol reagent. J Biol Chem 1951;193:265-275. 9 Pascual A, Tsukayama D, Kovarik J, et al: Up­ take and activity of rifapentine in human perito­ neal macrophages and polymorphonuclear leu­ kocytes. Eur J Clin Microbiol 1987;6:152-157.

10 Pascual A, Martinez-Martinez L, Aragon J, et al: Effect of amoxycillin and clavulanic acid, alone and in combination, on human polymorphonu­ clear leukocyte function against Staphylococcus aureus. Eur J Clin Microbiol Infect Dis 1989:8: 277-281. 11 Quie PG, White JG, Holmes B, et al: In vitro bac­ tericidal capacity of human polymorphonuclear leukocytes: Diminished activity in chronic granu­ lomatous disease of childhood. J Clin Invest 1967;46:668-679. 12 Gladue RP, Bright GM, Newborg MF: In vitro and in vivo uptake of azithromycin (CP-62,993) by phagocytic cells: Possible mechanism of deliv­ ery and release at sites of infection (No 714). Program Abstr 28th Intersci Conf Antimicrob Agents Chemother, Los Angeles 1988. 13 Labro MT, El Benna J, Babin-Chevaye C: Com­ parison of the in vitro effect of several macrolides on the oxidative burst of human neutrophils. J Antimicrob Chemother 1989;24:561-572. 14 Anderson R, Joone G, van Rensburg CEJ: An in vitro investigation of the intracellular bioactivity of amoxicillin, clindamycin and erythromycin for Staphylococcus aureus. J Infect Dis 1986:153: 593-600. 15 Pascual A, Martinez-Martinez L, Perea EJ: Effect of ciprofloxacin and ofloxacin on human poly­ morphonuclear leukocyte activity against staph­ ylococci. Chemotherapy 1989:35:17-22.

A. Pascual Department of Microbiology School of Medicine Apdo. 914 E -4 1080 Seville (Spain)

Downloaded by: Univ. of California Santa Barbara 128.111.121.42 - 3/8/2018 12:16:53 PM

References

427

Effect of azithromycin, roxithromycin and erythromycin on human polymorphonuclear leukocyte function against Staphylococcus aureus.

The effect of three macrolides (azithromycin, roxithromycin and erythromycin) on the interaction in vitro of human polymorphonuclear leukocytes (PMNs)...
633KB Sizes 0 Downloads 0 Views