- Email: [email protected]
Characteristics and mechanisms of azithromycin accumulation and efflux in human polymorphonuclear leukocytes
International Journal of Antimicrobial Agents 18 (2001) 419– 425 www.ischemo.org
Original article
Characteristics and mechanisms of azithromycin accumulation and efflux in human polymorphonuclear leukocytes W. Lee Hand *, Debra L. Hand Departments of Research De6elopment and Internal Medicine, Texas Tech. Uni6ersity Health Sciences Center, 4800 Alberta A6enue, El Paso, TX 79905, USA Received 25 April 2001; accepted 8 June 2001
Abstract Azithromycin achieves prolonged, high tissue concentrations in spite of low serum levels and obviously must be active at tissue sites of infection to be effective. These unique features prompted us to evaluate the interactions of azithromycin and human polymorphonuclear leukocytes (PMN). Uptake of radiolabeled antibiotic by PMN was determined by a velocity-gradient centrifugation technique and expressed as the ratio of cellular to extracellular drug concentration (C/E). Azithromycin was massively accumulated by human PMN (C/E =387.2 at 2 h). Uptake was not influenced by inhibitors of cellular metabolism, but phagocytosis slightly inhibited the entry of azithromycin into PMN. After removal of extracellular drug, the release (efflux) of azithromycin from PMN was extremely slow. Agents which neutralize lysosomal pH, preventing protonation and trapping of azithromycin, markedly increased antibiotic efflux. Active concentration and prolonged retention of azithromycin by phagocytic cells should allow delivery and subsequent release of accumulated drug at sites of infection. © 2001 Elsevier Science B.V. and International Society of Chemotherapy. All rights reserved. Keywords: Azithromycin; Antibiotic interactions with human PMN; Intracelluar uptake of antibiotic
1. Introduction During the decade in which azithromycin has been available, potential new uses for this azolide antibiotic have increased substantially. Azithromycin was initially used for treatment of respiratory tract infections caused by susceptible pathogens. Subsequently, the drug has been used for such diverse infections as typhoid fever and shigellosis, treatment and prophylaxis of Mycobacterium a6ium-intracellulare infection in AIDS patients and prophylaxis against malaria [1 – 5]. Azithromycin has also been used in trials to treatment in patients with symptomatic coronary atherosclerosis presumptively caused by Clamydia pneumoniae [6 – 10]. The pharmacokinetics of azithromycin are characterized by very low serum concentrations and wide-spread tissue distribution [11]. This drug has an exceptionally long tissue half life estimated to be greater than 2 days. * Corresponding author. Tel.: + 1-915-545-6630; fax: +1-915-5458960. E-mail address: [email protected] (W.L. Hand).
Such pharmacokinetics suggests azithromycin must be biologically active at the tissue level in order to be clinically effective. Indeed, high tissue concentrations of azithromycin correlate with efficacy in the treatment of infections in animals and humans [12,13]. The newly established and potential uses for azithromycin have been pursued because of these high, prolonged tissue and cellular (including phagocytic cells) concentrations [14]. Thus, this antimicrobial agent may be clinically active against microorganisms which are not exquisitely susceptible to the drug. High intracellular drug levels may help to explain the relatively low number of reported clinical failures in azithromycin treatment of respiratory tract infections due to ‘resistant’ Streptococcus pneumoniae [15,16]. The overall therapeutic effectiveness of an antibiotic will depend in part upon the agent’s entry into and influence on host phagocytic cells and their ingested organisms, as well as the extracellular drug-bacterial contact. Azithromycin is highly concentrated by murine and human phagocytic cells [3,14,17 –23] but certain
0924-8579/01/$ - $20 © 2001 Elsevier Science B.V. and International Society of Chemotherapy. All rights reserved. PII: S 0 9 2 4 - 8 5 7 9 ( 0 1 ) 0 0 4 3 0 - 7
420
W.L. Hand, D.L. Hand / International Journal of Antimicrobial Agents 18 (2001) 419–425
features of azithromycin uptake, intraphagocytic accumulation, intracellular distribution and release have not been elucidated. The azithromycin– phagocytic cell association is particularly interesting as this might constitute an ‘antibiotic delivery system’ if large quantities of this agent are targeted and delivered to sites of tissue infection. We have, therefore, evaluated specific characteristics and mechanisms of azithromycin interactions with human polymorphonuclear leukocytes (PMN). These studies included, (1) quantification of azithromycin uptake (accumulation) by PMN over time; (2) characterization of specific influences that alter the uptake and intracellular distribution of the agent. (3) Evaluation of azithromycin efflux from the cell.
2. Methods
2.1. Preparation of human PMN Peripheral venous blood was collected from normal volunteers by venipuncture. Granulocytes were isolated by dextran sedimentation and Hypaque– Ficoll density gradient centrifugation [24– 26].
2.2. Determination of azithromycin entry into human PMN Radiolabeled azithromycin ([14C] azithromycin, 45 mCi/mmol; Central Research Division, Pfizer Inc, Groton, Conn.) and other radiolabeled antibiotics were incubated with human PMN in tissue culture medium 199 (TC 199; Gibco, Grand Island, NY) containing 5% normal human serum at 37 °C. The radiolabeled antibiotics were utilized at similar molar concentrations ( 2.5×10 − 4 M) for comparability in antibiotic uptake experiments. Specific concentrations (micrograms per millilitre) for azithromycin and other antibiotics used in this and our previous studies were azithromycin-20, roxithromycin-21, clindamycin-10, erythromycin-18, rifampicin-20, penicillin G-20. After designated periods of time the cellular uptake of these drugs was determined by a velocity-gradient centrifugation technique as previously described [25–33]. PMN with their associated radiolabeled antibiotic were separated from the extracellular drug by velocity-gradient centrifugation in a microcentrifuge (Beckman Microfuge 11, Beckman Instruments Inc, Fullerton, CA). This was accomplished by centrifugation of PMN through a water-impermeable layer of silicone oil into formic acid, which dissolved the cells. The microcentrifuge tubes were then frozen, and the layers were separated by slicing with a razor. The radioactive contents of the lower layer, containing radiolabeled antibiotic which entered the cells, and the upper layer, which
contains the drug that was still in solution, were quantitated in a liquid scintillation counter (Beckman LS 6500 Scintillation System). Antibiotic (azithromycin or other drug) uptake was then expressed as the ratio of the cellular concentration of antibiotic to the extracellular concentration (C/E).
2.3. Characterization of azithromycin uptake by human PMN The effects on azithromycin accumulation by PMN of cell viability, environmental temperature, pH, cellular metabolic inhibitors (sodium cyanide, sodium azide, 2,4-dinitrophenol, potassium fluoride), scavengers of reactive oxygen species (superoxide dismutase [SOD], catalase), and human myeloperoxidase (MPO) (all from Sigma Chemical Co, St. Louis, MO) were determined. The influence of pH (over a range of pH 5–9) on azithromycin uptake was evaluated. Phagocytic cells have specific carrier-mediated membrane transport systems for hexoses, amino acids and nucleosides. We have demonstrated that clindamycin is transported into PMN and alveolar macrophages by the cell membrane nucleoside system [26,28,34]. Therefore, we examined the possibility that one of these systems might transport azithromycin. Competitive inhibitors of these transport systems (potential inhibitors of azithromycin transport), including adenosine (Sigma), D-glucose (Sigma), glycine, leucine, lysine (all from Pierce Chemical Co Rockford, IL), were preincubated with PMN before determination of azithromycin uptake. We also evaluated the in vitro uptake of azithromycin by PMN under conditions that mimic in vivo infection. In these experiments we incubated PMN with microbial particles (opsonized zymosan) or with soluble membrane-perturbating agents (concanavalin A, formyl-methionyl-leucyl-phenylalanine [FMLP]) (both from Sigma) for 30 min. The cells were then washed and suspended in fresh TC 199–5% serum before determination of antibiotic uptake.
2.4. Efflux (release) of azithromycin from human PMN Release of azithromycin from PMN, after removal of extracellular drug, was quantitated over time. The requirement for metabolic energy and the impact of phagocytosis on the efflux process were determined as described above for azithromycin uptake. The role of lysosomal trapping in retention and release of azithromycin was evaluated by adding ammonium chloride (Mallinckrodt Chemical Co, St. Louis, MO) or chloroquine (Sigma), which neutralize lysosomal pH [35]. Efflux of azithromycin was determined after allowing radiolabeled antibiotic uptake by PMN during a 60 min incubation. Cells were then collected by centrifuga-
W.L. Hand, D.L. Hand / International Journal of Antimicrobial Agents 18 (2001) 419–425
tion and resuspended in fresh medium without antibiotic. Intracellular and extracellular azithromycin were quantitated initially and at subsequent designated times by the velocity-gradient centrifugation method.
421
dent upon cell viability and environmental temperature (Table 2). Thus, there was diminished uptake by dead (phenol-killed) cells (C/E = 7–8) and by viable cells incubated at either 4 °C (C/E = 5 8) or 25 °C (C/E = 17 at 2 h). The pH profile for entry of azithromycin into phagocytic cells was determined. The optimum pH for accumulation of this antibiotic was approximately 8.7. This is similar to what we observed with other weakly basic antibiotics (clindamycin, macrolides) which are selectively concentrated by phagocytic cells [31,32,36]. Inhibitors of cellular metabolism were tested for their effect on azithromycin accumulation. Potassium fluoride, an inhibitor of glycolysis (the predominant energy source in PMN), did not alter antibiotic uptake. Sodium azide (which inhibits oxidative respiration and inactivates MPO), 2,4-dinitrophenol (an inhibitor of oxidative respiration), SOD, catalase and MPO had no effect on azithromycin entry into human PMN. Sodium cyanide (which inhibits oxidative respiration and certain iron-containing enzymes) caused a substantial increase in azithromycin accumulation by PMN (Table 3). However, sodium cyanide-containing medium is strongly alkaline (pH 10.1 initially and 8.3 at conclusion of 2 h incubation) and azithromycin uptake returned to control levels when the medium was buffered to pH 7.4.
3. Results
3.1. Uptake of azithromycin by human PMN Azithromycin was massively and progressively concentrated by PMN during incubation with the radiolabeled antibiotic. The ratio of cellular to extracellular drug concentration (C/E) was 42.3 at 15 min, 61.6 at 30 min, 146.0 at 60 min, and 387.2 at 120 min (Table 1). Data regarding uptake of other antibiotics by human neutrophils is presented for comparative purposes in Table 1. Uptake of azithromycin by PMN was far greater than that of other macrolides and clindamycin, antimicrobial agents which we previously found to be selectively concentrated by these phagocytic cells [25,29,31,33,36,37].
3.2. Characterization of azithromycin uptake This accumulation of azithromycin by PMN led us to evaluate the entry process in detail. Uptake was depenTable 1 Uptake of azithromycin and other selected antibiotics by human PMN Time of incubation (min)
15 30 60 120 a b
Antibiotic uptake (C/E)a
Azithromycin
Roxithromycin
Clindamycin
Erythromycin
Rifampicin
Penicillin G Cephalosporins
42.3918.2 (4)b 61.69 3.9 (12) 146.096.1 (14) 387.2921.5 (12)
26.6 35.6 41.5 –
11.4 12.0 11.8 11.8
2.5 6.3 11.7 13.3
2.2 2.3 2.4 2.3
50.4 50.3 50.4 50.3
C/E is the ratio of the cellular concentration of antibiotic to the extracellular concentration. Data are means 9 standard errors of the mean. The number of experiments is in parentheses.
Table 2 Influence of cell viability and environmental temperature on entry of azithromycin into human PMN Time of incubation (min)
Antibiotic uptake (C/E)a Viable Cells
15 30 60 120 a b
Dead Cells
37 °C
25 °C
4 °C
37 °C
42.39 18.2 (4)b 61.69 3.9 (12) 146.09 6.1 (14) 387.29 21.5 (12)
2.8(1) 7.5 92.7 (3) 11.2 92.8 (4) 16.9 9 4.1 (3)
2.9(2) 5.5 9 3.5 (3) 8.4 9 3.0 (4) 5.8 9 1.7 (4)
7.1(1) 6.5 (2) 8.5 94.4 (3) 7.3 92.0 (3)
C/E is the ratio of the cellular concentration of antibiotic to the extracellular concentration. Data are means 9 standard errors of the mean. The number of experiments is in parentheses.
W.L. Hand, D.L. Hand / International Journal of Antimicrobial Agents 18 (2001) 419–425
422
Table 3 Effects of metabolic and competitive inhibitors on accumulation of azithromycin by human PMN Time of Incubation (min)
15 30 60 120
Antibiotic uptake (C/E)a Control (No addition)
NaCNb (1 mM)
42.39 18.2 (4)c 61.69 3.9 (12) 146.09 6.1 (14) 387.29 21.5 (12)
– – 53.6 9 0.8 (4) 100.69 6.5 (7)d 247.89 23.3 (8)b, d 131.5 918.2 (4) 607.89 31.5 (8)b, d 376.2 951.4 (4)
Na Azide (2 mM)
KF (1 mM)
Adenosine (1 mM)
D-Glucose (1 mM)
27.1(1) 64.8 94.7 (4) 166.2 913.5 (5) 419.6 943.7 (5)
– 55.1 9 2.6 (3) 148.3 9 17.3 (4) 337.8 9 27.6 (4)
– 63.2 96.1 (3) 132.8 910.1 (4) 358.1 929.1 (4)
a
C/E is the ratio of the cellular concentration of antibiotic to the extracellular concentration. Alkaline incubation medium (pH]8.3). c Data are means 9 standard errors of the mean. The number of experiments is in parentheses. d Significant difference between control (antibiotic only) and experimental (antibiotic+inhibitor) group, PB0.05. b
Table 4 Effects of cell membrane stimulation on uptake of azithromycin by human PMN Time of incubation (min)
15 30 60
Antibiotic uptake (C/E)a Control (No addition)
Zymosan
FMLP
Con A
61.69 3.9 (12)b 146.09 6.1 (14) 387.29 21.5 (12)
39.6 96.5 (3)c 85.4 96.2 (4)c 195.1 915.4 (4)c
48.9 9 8.8 (3) 113.6 920.0 (4) 307.8 949.8 (4)
50.2 9 13.6 (3) 115.8 925.9 (4) 301.3 968.5 (4)
a
C/E is the ratio of the cellular concentration of antibiotic to the extracellular concentration. Data are means 9 standard errors of the mean. The number of experiments is in parentheses. c Significant difference between control (antibiotic only) and experimental (antibiotic+stimulating agent) group, PB0.05. b
Table 5 Efflux of intracellular azithromycin from human PMN after removal of extracellular drugs Time after drug removal (min)
Remaining intracellular azithromycin (% of initial uptake) 37° Control (No addition)
30 60 120 180 a
94.99 1.4 91.6 9 3.8 83.0 9 4.5 80.7 9 6.1
(5)a (6) (5) (5)
25 °C
4 °C
NaCN (1 mM)
KF (1 mM)
No addition
No addition
96.8 92.0 (5) 87.1 97.0 (5) 91.7 (2) 71.1 94.2 (4)
90.5 9 5.7 (5) 95.1 91.9 (5) 78.8 910.5 (4) 77.2 98.5 (3)
92.2 9 6.6 (5) 96.3 93.6 (5) 90.6 95.5 (3) 80.0 910.8 (4)
95.3 9 2.5 (5) 92.8 94.3 (5) 83.1 9 11.7 (3) 83.6 9 9.6 (4)
Data are means 9standard errors of the mean. The number of experiments is in parentheses.
We looked at the possibility that azithromycin might enter human neutrophils by a specific carrier-mediated membrane transport system for nucleosides, hexoses, or amino acids. Accumulation of azithromycin by PMN was not influenced by potential competitive inhibitors of these transport systems (adenosine, D-glucose, glycine, leucine, lysine) (Table 3). The effects of phagocytosis and other cell membrane stimulation on entry of radiolabeled azithromycin into PMN were studied. Ingestion of microbial particles (opsonized zymosan) inhibited the entry of azithromycin into phagocytic cells ( 50% decrease at
60 min.). FMLP and Con A had a lesser inhibitory effect on azithromycin uptake (Table 4).
3.3. Efflux of azithromycin from human PMN The kinetics and characteristics of azithromycin release from human PMN were determined. Efflux of this antibiotic after removal of extracellular drug was extremely slow (Table 5). Greater than 90% of the intracellular drug was retained for 60 min and 80% of the cell-associated azithromycin remained after 3 h. Changes in environmental temperature and inhibitors
W.L. Hand, D.L. Hand / International Journal of Antimicrobial Agents 18 (2001) 419–425
of cellular metabolism (NaCN and KF) failed to alter drug release. However, phagocytosis of zymosan, as well as exposure to FMLP, increased the efflux of azithromycin from phagocytic cells. Of particular interest was that the addition of ammonium chloride (50 mM) and chloroquine (1 mM), agents which neutralize lysosomal pH [35], led to a striking increase in the efflux of azithromycin from PMN (Table 6). Lower concentrations of these lysosomatropic weak bases produced a similar, but less dramatic, increase in antibiotic efflux.
4. Discussion These studies have provided important information concerning the interactions of azithromycin and human PMN. An antibiotic must enter phagocytes to have any activity against viable intracellular organisms. We have shown that most antimicrobial agents fail to penetrate phagocytic cells readily [25,27,30– 33]. Nevertheless, certain antibiotics, including macrolides and clindamycin, are accumulated by phagocytic cells [25– 33,36]. We have demonstrated that azithromycin is massively concentrated by human PMN (C/E = 387.2 at 2 h), far more than any other antibiotic we have tested. Azithromycin accumulation by PMN was dependent upon cell viability and a physiological environmental temperature. The optimal pH for accumulation was : 8.7. We observed similar increased uptake at alkaline pH with other antibiotics which are weak bases and are concentrated within phagocytic cells (erythromycin, roxithromycin, dirithromycin, clindamycin) [31,32,36]. Uptake of clindamycin and several macrolide antibiotics by phagocytic cells requires metabolic energy and is diminished by inhibitors of the predominant energy source in the specific cell type [25– 28,31,32]. In contrast, accumulation of azithromycin by human PMN may not be a metabolically active process as uptake of this antibiotic was not influenced by exposure to fluoride, which inhibits glycolysis, the major source of energy in PMN. Likewise, sodium azide and 2,4-dinitrophenol, inhibitors of oxidative respiration, failed to inhibit azithromycin uptake. However, azithromycin accumula-
423
tion was strikingly increased in PMN exposed to sodium cyanide. This finding is unrelated to any effect of cyanide on mitochondrial oxidative respiration (a minor pathway in PMN, which have few mitochondria), since other inhibitors of oxidative respiration did not alter azithromycin uptake. The cyanide effect theoretically could reflect an inhibitory interaction with a heme (or other iron)-containing enzyme system, such as myeloperoxidase (MPO), a major component of PMN azurophil granules (lysosomes). However, neither sodium azide (which inhibits MPO activity), MPO itself, or inhibitors of reactive oxygen species which interact with MPO, had any influence on azithromycin uptake. Sodium cyanide is strongly alkaline in solution. This elevation in pH apparently accounts for the cyanide-associated increase in azithromycin entry into PMN. Antibiotic uptake in cyanide-containing media was the same as that at the same pH in media without cyanide. Azithromycin uptake returned to control levels with adjustment of the cyanide-containing medium to pH 7.4. We attempted to identify the mechanism(s) by which azithromycin is transported across the cell membrane to enter phagocytic cells. Previously we found that clindamycin enters rabbit AM and human PMN via the cell membrane nucleoside transport system [26,28,34]. In the current study we demonstrated no effect of inhibitors for known phagocytic cell carrier-mediated membrane transport systems on azithromycin uptake. Next, we examined the effect of microbial particle ingestion by PMN on azithromycin uptake. As was true for other macrolide antibiotics [26,29,31,36,37], the accumulation of azithromycin by PMN was moderately inhibited by ingestion of zymosan. This result might be due to internalization of cell membrane, possibly including receptors for macrolides/azolides, or discharge of lysosomal (azurophilic granule) contents, thereby decreasing the accumulation sites for azithromycin. Efflux of azithromycin from human PMN was extremely slow; 80% of the initial intracellular azithromycin was still retained within cells 3 h after removal of extracellular antibiotic. This is in distinct contrast to what we observed with some other selec-
Table 6 Efflux of intracellular azithromycin from human PMN after removal of extracellular drug Time after drug removal (Min)
Remaining intracellular azithromycin (% of initial uptake) Control (No addition)
30 60 120 180 a b
94.99 1.4 91.69 3.8 83.09 4.5 80.7 96.1
(5)a (6) (5) (5)
Zymosan
FMLP
NH4C1 (50 mM)
Chloroquine (1 mM)
80.5 96.1 (5) 76.6 9 3.8(6)b 71.2 97.2 (4) 59.7 9 7.2 (5)
85.6 (2) 71.2 9 7.4(3)b 67.7 (2) 74.3 9 5.5 (3)
12.7 9 3.0 (5)b 12.3 9 4.8 (6)b 8.2 9 3.5 (4)b 4.9 90.6 (5)b
25.1 95.7 (4)b 9.8 (2) 8.9 95.3 (3)b 3.1 9 7 (3)b
Data are means 9standard errors of the mean. The number of experiments is in parentheses. Significant difference between control and experimental group, PB0.05.
424
W.L. Hand, D.L. Hand / International Journal of Antimicrobial Agents 18 (2001) 419–425
tively concentrated antibiotics (erythromycin, clindamycin), which are promptly released from alveolar macrophages and PMN upon removal of extracellular drug [25,32]. Other investigators also reported a rapid efflux of several macrolide antibiotics from human PMN, lymphocytes and tissue culture cells [38–40]. Obviously, these macrolides are not tightly bound or trapped within intracellular compartments. The extremely slow release of azithromycin from PMN indicates that this drug differs from the macrolides noted above in regard to intracellular localization, binding, and/or trapping. Efflux of azithromycin was not altered by metabolic inhibitors or by variations in environmental temperature, observations suggesting that cellular metabolic energy is not involved in this process. Azithromycin is a weakly basic (dibasic) antibiotic which would be protonated and trapped within acidic lysosomes, accounting for the agent’s prolonged cellular retention. Ammonium chloride and chloroquine (agents which neutralize lysosomal pH, preventing protonation and trapping of weak bases) strikingly increased the efflux of azithromycin [35]. These azithromycin efflux characteristics are similar to what we observed in studies of PMN interactions with dirithromycin, another relatively new macrolide antibiotic [36]. Phagocytosis inhibited the accumulation of azithromycin by PMN, but this observation theoretically might be due to enhanced efflux (release) of drug rather than decreased uptake. Indeed, ingestion of zymosan by PMN modestly enhanced the release of azithromycin (after 60 min). This released drug presumably comes from the lysosomal compartment as a result of phagocytosis-induced degranulation. Thus, zymosan ingestion alters both the uptake and release of azithromycin by human PMN. However, the effect on efflux was much less than the inhibitory impact on intracellular accumulation of the antibiotic. In summary, azithromycin was massively accumulated by human PMN. This uptake of azithromycin by neutrophils was far greater than that of any other antibiotic we have studied. Antimicrobial effects of this accumulated intracellular drug might include, (1) a direct action against organisms within phagocytic cells [3,20,21,23,41,42]; and (2) extracellular antibacterial activity after release of intraphagocytic antibiotic at sites of infection. This latter possibility is particularly interesting. The remarkable accumulation and extremely slow release of intracellular azithromycin by human PMN are features which have exciting therapeutic implications. Avid accumulation and prolonged retention of azithromycin by phagocytic cells should allow delivery and subsequent release of drug over time at sites of infection. Ingestion of microorganisms accelerates the release of azithromycin by PMN, and would lead to high extracellular drug concentrations. Thus, the azithromycin-PMN association might be considered an
‘antibiotic delivery system,’ and the subsequent release of azithromycin to the extracellular milieu would represent a true example of targeted antimicrobial therapy. Acknowledgements This study was supported in part by a grant from Pfizer Labs Division, Pfizer Inc, New York, NY. References [1] Frenck RW Jr, Nakhla I, Sultan Y, et al. Azithromycin versus ceftriaxone for the treatment of uncomplicated typhoid fever in children. Clin Infect Dis 2000;31:1134 – 8. [2] Girgis NI, Butler T, Frenck RW, et al. Azithromycin versus ciprofloxacin for treatment of uncomplicated typhoid fever in a randomized trial in Egypt that included patients with multidrug resistance. Antimicrob Agents Chemother 1999;43:1441 –4. [3] Rakita RM, Jacques-Palaz K, Murray BE. Intracellular activity of azithromycin against bacterial enteric pathogens. Antimicrob Agents Chemother 1994;38:1915 – 21. [4] California Collaborative Treatment Group, Havlir DV, Dube´ MP, Sattler FR, et al. Prophylaxis against disseminated Mycobacterium a6ium complex with weekly azithromycin, daily rifabutin, or both. New Engl J Med 1996;335:392 – 8. [5] Horsburgh CR Jr. Advances in the prevention and treatment of Mycobacterium a6ium disease. New Engl J Med 1996;335:428 – 30. [6] Ngeh J, Gupta S. C. pneumoniae and atherosclerosis: causal or coincidental link? ASM News 2000;66:732 – 7. [7] Anderson JL, Muhlestein JB, Carlquist J, et al. Randomized secondary prevention trial of azithromycin in patients with coronary artery disease and serological evidence for Chlamydia pneumoniae infection. The azithromycin in coronary artery disease: elimination of myocardial infarction with Chlamydia (ACADEMIC) study. Circulation 1999;99:1540 – 7. [8] Gupta S, Leatham EW, Carrington D, Mendall MA, Kaski JC, Camm AJ. Elevated Chlamydia pneumoniae antibodies, cardiovascular events, and azithromycin in male survivors of myocardial infarction. Circulation 1997;96:404 – 7. [9] Dunne MW. Rationale and design of a secondary prevention trial of antibiotic use in patients after myocardial infarction: the WIZARD (weekly intervention with Zithromax [azithromycin] for atherosclerosis and its related disorders) trial. J Infect Dis 2000;181(Suppl 3):S572 – 8. [10] Jackson LA. Description and status of the azithromycin and coronary events study (ACES). J Infect Dis 2000;181(Suppl 3):S579 – 81. [11] Foulds G, Shepard RM, Johnson RB. The pharmokinetics of azithromycin in human serum and tissues. J Antimicrob Chemother 1990;25(Suppl A):73 –82. [12] Girard AE, Girard D, Retsema JA. Correlation of the extravascular pharmacokinetics with in-vivo efficacy in models of localized infection. J Antimicrob Chemother 1990;25(Suppl A):61 – 71. [13] Retsema JA, Girard AE, Girard D, Milisen WB. Relationship of high tissue concentrations of azithromycin to bactericidal activity and efficacy in vivo. J Antimicrob Chemother 1990;25(Suppl A):83 – 9. [14] Gladue RP, Bright GM, Isaacson RE, Newborg MF. In vitro and in vivo uptake of azithromycin (CP-62, 993) by phagocytic cells: possible mechanisms of delivery and release at sites of infection. Antimicrob Agents Chemother 1989;33:277 – 82.
W.L. Hand, D.L. Hand / International Journal of Antimicrobial Agents 18 (2001) 419–425 [15] Siegel RE. The significance of serum vs tissue levels of antibiotics in the treatment of penicillin-resistant Streptococcus pneumoniae and community-acquired pneumonia. Are we looking in the wrong place? Chest 1999;116:535 – 8. [16] Amsden GW. Pharmacological considerations in the emergence of resistance. Int J Antimicrob Agents 1999;11(Suppl 1):S7 – S14. [17] Baldwin DR, Wise R, Andrews JM, Ashby JP, Honeybourne D. Azithromycin concentration at the sites of pulmonary infection. Eur Respir J 1990;3:886 –90. [18] Bonnet M, Van der Auwera P. In vitro and in vivo intraleukocytic accumulation of azithromycin (CP-62,993) and its influence on ex vivo leukocyte chemiluminescence. Antimicrob Agents Chemother 1992;36:1302 – 9. [19] Wildfeuer A, Laufen H, Muller-Wening D, Haferkamp O. Interaction of azithromycin and human phagocytic cells: uptake of the antibiotic and the effect on the survival of ingested bacteria. Arzneim-Forsch/Drug Res 1989;39:755 –8. [20] Meyer AP, Bril-Bazuin C, Mattie H, van den Broek PJ. Uptake of azithromycin by human monocytes and enhanced intracellular antibacterial activity against Staphylococcus aureus. Antimicrob Agents Chemother 1993;37:2318 –22. [21] Stamler DA, Edelstein MAC, Edelstein PH. Azithromycin pharmacokinetics and intracellular concentrations in Legionella pneumophila-infected and uninfected guinea pigs and their alveolar macrophages. Antimicrob Agents Chemother 1994;38:217 – 22. [22] Fietta A, Merlini C, Grassi GG. Requirements for intracellular accumulation and release of clarithromycin and azithromycin by human phagocytes. J Chemother 1997;9:23 –31. [23] Wildfeuer A, Laufen H, Zimmerman T. Uptake of azithromycin by various cells and its intracellular activity under in vivo conditions. Antimicrob Agents Chemother 1996;40:75 – 9. [24] Boyum A. Isolation of lymphocytes, granulocytes, and macrophages. Scand J Immunol 1976;5(Suppl 5):9 –15. [25] Prokesch RC, Hand WL. Antibiotic entry into human polymorphonuclear leukocytes. Antimicrob Agents Chemother 1982;21:373 – 80. [26] Steinberg TH, Hand WL. Effects of phagocytosis on antibiotic and nucleoside uptake by human polymorphonuclear leukocytes. J Infect Dis 1984;149:397 – 403. [27] Hand WL, Corwin RW, Steinberg TH, Grossman GD. Uptake of antibiotics by human alveolar macrophages. Am Rev Resp Dis 1984;129:933 – 7. [28] Hand WL, King-Thompson NL. Membrane transport of clindamycin in alveolar macrophages. Antimicrob Agents Chemother 1982;21:241 – 7.
425
[29] Hand WL, King-Thompson NL. Contrasts between phagocyte antibiotic uptake and subsequent intracellular bactericidal activity. Antimicrob Agents Chemother 1986;29:135 – 40. [30] Hand WL, King-Thompson NL. Entry of antibiotics into human monocytes. J Antimicrob Chemother 1989;23:681 – 9. [31] Hand WL, King-Thompson NL, Holman JW. Entry of roxithromycin (RU965), imipenem, cefotaxime, trimethoprim, and metronidazole into human polymorphonuclear leukocytes. Antimicrob Agents Chemother 1987;31:1553 – 7. [32] Johnson JD, Hand WL, Francis JB, King-Thompson N, Corwin RW. Antibiotic uptake by alveolar macrophages. J Lab Clin Med 1980;95:429 – 39. [33] Steinberg TH, Hand WL. Effect of phagocyte membrane stimulation on antibiotic uptake and intracellular bactericidal activity. Antimicrob Agents Chemother 1987;31:660 – 2. [34] Hand WL, King-Thompson NL. Uptake of antibiotics by human polymorphonuclear leukocyte cytoplasts. Antimicrob Agents Chemother 1990;34:1189 –93. [35] Klempner MS, Styrt B. Alkalinizing the intralysosomal pH inhibits degranulation of human neutrophils. J Clin Invest 1983;72:1793 – 800. [36] Hand WL, Hand DL. Interactions of dirithromycin with human polymorphonuclear leukocytes. Antimicrob Agents Chemother 1993;37:2557 – 62. [37] Hand WL, Hand DL. Influence of pentoxifylline and its derivatives on antibiotic uptake and superoxide generation by human phagocytic cells. Antimicrob Agents Chemother 1995;39:1574 –9. [38] Martin JR, Johnson P, Miller MF. Uptake, accumulation, and egress of erythromycin by tissue culture cells of human origin. Antimicrob Agents Chemother 1985;27:314 – 9. [39] Dette GA, Knothe H. Kinetics of erythromycin uptake and release by human lymphocytes and polymorphonuclear leukocytes. J Antimicrob Chemother 1986;18:73 – 82. [40] Ishiguro M, Koga H, Kohno S, Hayashi T, Yamaguchi H, Hirota M. Penetration of macrolides into human polymorphonuclear leukocytes. J Antimicrob Chemother 1989;24:719 – 29. [41] Donowitz GR, Earnhardt KI. Azithromycin inhibition of intracellular Legionella micdadei. Antimicrob Agents Chemother 1993;37:2261 – 4. [42] Jonas D, Engels I, Daschner FD, Frank Y. The effect of azithromycin on intracellular Legionella pneumophila in the Mono Mac 6 cell line at serum concentrations attainable in vivo. J Antimicrob Chemother 2000;46:385 – 90.
Original article
Characteristics and mechanisms of azithromycin accumulation and efflux in human polymorphonuclear leukocytes W. Lee Hand *, Debra L. Hand Departments of Research De6elopment and Internal Medicine, Texas Tech. Uni6ersity Health Sciences Center, 4800 Alberta A6enue, El Paso, TX 79905, USA Received 25 April 2001; accepted 8 June 2001
Abstract Azithromycin achieves prolonged, high tissue concentrations in spite of low serum levels and obviously must be active at tissue sites of infection to be effective. These unique features prompted us to evaluate the interactions of azithromycin and human polymorphonuclear leukocytes (PMN). Uptake of radiolabeled antibiotic by PMN was determined by a velocity-gradient centrifugation technique and expressed as the ratio of cellular to extracellular drug concentration (C/E). Azithromycin was massively accumulated by human PMN (C/E =387.2 at 2 h). Uptake was not influenced by inhibitors of cellular metabolism, but phagocytosis slightly inhibited the entry of azithromycin into PMN. After removal of extracellular drug, the release (efflux) of azithromycin from PMN was extremely slow. Agents which neutralize lysosomal pH, preventing protonation and trapping of azithromycin, markedly increased antibiotic efflux. Active concentration and prolonged retention of azithromycin by phagocytic cells should allow delivery and subsequent release of accumulated drug at sites of infection. © 2001 Elsevier Science B.V. and International Society of Chemotherapy. All rights reserved. Keywords: Azithromycin; Antibiotic interactions with human PMN; Intracelluar uptake of antibiotic
1. Introduction During the decade in which azithromycin has been available, potential new uses for this azolide antibiotic have increased substantially. Azithromycin was initially used for treatment of respiratory tract infections caused by susceptible pathogens. Subsequently, the drug has been used for such diverse infections as typhoid fever and shigellosis, treatment and prophylaxis of Mycobacterium a6ium-intracellulare infection in AIDS patients and prophylaxis against malaria [1 – 5]. Azithromycin has also been used in trials to treatment in patients with symptomatic coronary atherosclerosis presumptively caused by Clamydia pneumoniae [6 – 10]. The pharmacokinetics of azithromycin are characterized by very low serum concentrations and wide-spread tissue distribution [11]. This drug has an exceptionally long tissue half life estimated to be greater than 2 days. * Corresponding author. Tel.: + 1-915-545-6630; fax: +1-915-5458960. E-mail address: [email protected] (W.L. Hand).
Such pharmacokinetics suggests azithromycin must be biologically active at the tissue level in order to be clinically effective. Indeed, high tissue concentrations of azithromycin correlate with efficacy in the treatment of infections in animals and humans [12,13]. The newly established and potential uses for azithromycin have been pursued because of these high, prolonged tissue and cellular (including phagocytic cells) concentrations [14]. Thus, this antimicrobial agent may be clinically active against microorganisms which are not exquisitely susceptible to the drug. High intracellular drug levels may help to explain the relatively low number of reported clinical failures in azithromycin treatment of respiratory tract infections due to ‘resistant’ Streptococcus pneumoniae [15,16]. The overall therapeutic effectiveness of an antibiotic will depend in part upon the agent’s entry into and influence on host phagocytic cells and their ingested organisms, as well as the extracellular drug-bacterial contact. Azithromycin is highly concentrated by murine and human phagocytic cells [3,14,17 –23] but certain
0924-8579/01/$ - $20 © 2001 Elsevier Science B.V. and International Society of Chemotherapy. All rights reserved. PII: S 0 9 2 4 - 8 5 7 9 ( 0 1 ) 0 0 4 3 0 - 7
420
W.L. Hand, D.L. Hand / International Journal of Antimicrobial Agents 18 (2001) 419–425
features of azithromycin uptake, intraphagocytic accumulation, intracellular distribution and release have not been elucidated. The azithromycin– phagocytic cell association is particularly interesting as this might constitute an ‘antibiotic delivery system’ if large quantities of this agent are targeted and delivered to sites of tissue infection. We have, therefore, evaluated specific characteristics and mechanisms of azithromycin interactions with human polymorphonuclear leukocytes (PMN). These studies included, (1) quantification of azithromycin uptake (accumulation) by PMN over time; (2) characterization of specific influences that alter the uptake and intracellular distribution of the agent. (3) Evaluation of azithromycin efflux from the cell.
2. Methods
2.1. Preparation of human PMN Peripheral venous blood was collected from normal volunteers by venipuncture. Granulocytes were isolated by dextran sedimentation and Hypaque– Ficoll density gradient centrifugation [24– 26].
2.2. Determination of azithromycin entry into human PMN Radiolabeled azithromycin ([14C] azithromycin, 45 mCi/mmol; Central Research Division, Pfizer Inc, Groton, Conn.) and other radiolabeled antibiotics were incubated with human PMN in tissue culture medium 199 (TC 199; Gibco, Grand Island, NY) containing 5% normal human serum at 37 °C. The radiolabeled antibiotics were utilized at similar molar concentrations ( 2.5×10 − 4 M) for comparability in antibiotic uptake experiments. Specific concentrations (micrograms per millilitre) for azithromycin and other antibiotics used in this and our previous studies were azithromycin-20, roxithromycin-21, clindamycin-10, erythromycin-18, rifampicin-20, penicillin G-20. After designated periods of time the cellular uptake of these drugs was determined by a velocity-gradient centrifugation technique as previously described [25–33]. PMN with their associated radiolabeled antibiotic were separated from the extracellular drug by velocity-gradient centrifugation in a microcentrifuge (Beckman Microfuge 11, Beckman Instruments Inc, Fullerton, CA). This was accomplished by centrifugation of PMN through a water-impermeable layer of silicone oil into formic acid, which dissolved the cells. The microcentrifuge tubes were then frozen, and the layers were separated by slicing with a razor. The radioactive contents of the lower layer, containing radiolabeled antibiotic which entered the cells, and the upper layer, which
contains the drug that was still in solution, were quantitated in a liquid scintillation counter (Beckman LS 6500 Scintillation System). Antibiotic (azithromycin or other drug) uptake was then expressed as the ratio of the cellular concentration of antibiotic to the extracellular concentration (C/E).
2.3. Characterization of azithromycin uptake by human PMN The effects on azithromycin accumulation by PMN of cell viability, environmental temperature, pH, cellular metabolic inhibitors (sodium cyanide, sodium azide, 2,4-dinitrophenol, potassium fluoride), scavengers of reactive oxygen species (superoxide dismutase [SOD], catalase), and human myeloperoxidase (MPO) (all from Sigma Chemical Co, St. Louis, MO) were determined. The influence of pH (over a range of pH 5–9) on azithromycin uptake was evaluated. Phagocytic cells have specific carrier-mediated membrane transport systems for hexoses, amino acids and nucleosides. We have demonstrated that clindamycin is transported into PMN and alveolar macrophages by the cell membrane nucleoside system [26,28,34]. Therefore, we examined the possibility that one of these systems might transport azithromycin. Competitive inhibitors of these transport systems (potential inhibitors of azithromycin transport), including adenosine (Sigma), D-glucose (Sigma), glycine, leucine, lysine (all from Pierce Chemical Co Rockford, IL), were preincubated with PMN before determination of azithromycin uptake. We also evaluated the in vitro uptake of azithromycin by PMN under conditions that mimic in vivo infection. In these experiments we incubated PMN with microbial particles (opsonized zymosan) or with soluble membrane-perturbating agents (concanavalin A, formyl-methionyl-leucyl-phenylalanine [FMLP]) (both from Sigma) for 30 min. The cells were then washed and suspended in fresh TC 199–5% serum before determination of antibiotic uptake.
2.4. Efflux (release) of azithromycin from human PMN Release of azithromycin from PMN, after removal of extracellular drug, was quantitated over time. The requirement for metabolic energy and the impact of phagocytosis on the efflux process were determined as described above for azithromycin uptake. The role of lysosomal trapping in retention and release of azithromycin was evaluated by adding ammonium chloride (Mallinckrodt Chemical Co, St. Louis, MO) or chloroquine (Sigma), which neutralize lysosomal pH [35]. Efflux of azithromycin was determined after allowing radiolabeled antibiotic uptake by PMN during a 60 min incubation. Cells were then collected by centrifuga-
W.L. Hand, D.L. Hand / International Journal of Antimicrobial Agents 18 (2001) 419–425
tion and resuspended in fresh medium without antibiotic. Intracellular and extracellular azithromycin were quantitated initially and at subsequent designated times by the velocity-gradient centrifugation method.
421
dent upon cell viability and environmental temperature (Table 2). Thus, there was diminished uptake by dead (phenol-killed) cells (C/E = 7–8) and by viable cells incubated at either 4 °C (C/E = 5 8) or 25 °C (C/E = 17 at 2 h). The pH profile for entry of azithromycin into phagocytic cells was determined. The optimum pH for accumulation of this antibiotic was approximately 8.7. This is similar to what we observed with other weakly basic antibiotics (clindamycin, macrolides) which are selectively concentrated by phagocytic cells [31,32,36]. Inhibitors of cellular metabolism were tested for their effect on azithromycin accumulation. Potassium fluoride, an inhibitor of glycolysis (the predominant energy source in PMN), did not alter antibiotic uptake. Sodium azide (which inhibits oxidative respiration and inactivates MPO), 2,4-dinitrophenol (an inhibitor of oxidative respiration), SOD, catalase and MPO had no effect on azithromycin entry into human PMN. Sodium cyanide (which inhibits oxidative respiration and certain iron-containing enzymes) caused a substantial increase in azithromycin accumulation by PMN (Table 3). However, sodium cyanide-containing medium is strongly alkaline (pH 10.1 initially and 8.3 at conclusion of 2 h incubation) and azithromycin uptake returned to control levels when the medium was buffered to pH 7.4.
3. Results
3.1. Uptake of azithromycin by human PMN Azithromycin was massively and progressively concentrated by PMN during incubation with the radiolabeled antibiotic. The ratio of cellular to extracellular drug concentration (C/E) was 42.3 at 15 min, 61.6 at 30 min, 146.0 at 60 min, and 387.2 at 120 min (Table 1). Data regarding uptake of other antibiotics by human neutrophils is presented for comparative purposes in Table 1. Uptake of azithromycin by PMN was far greater than that of other macrolides and clindamycin, antimicrobial agents which we previously found to be selectively concentrated by these phagocytic cells [25,29,31,33,36,37].
3.2. Characterization of azithromycin uptake This accumulation of azithromycin by PMN led us to evaluate the entry process in detail. Uptake was depenTable 1 Uptake of azithromycin and other selected antibiotics by human PMN Time of incubation (min)
15 30 60 120 a b
Antibiotic uptake (C/E)a
Azithromycin
Roxithromycin
Clindamycin
Erythromycin
Rifampicin
Penicillin G Cephalosporins
42.3918.2 (4)b 61.69 3.9 (12) 146.096.1 (14) 387.2921.5 (12)
26.6 35.6 41.5 –
11.4 12.0 11.8 11.8
2.5 6.3 11.7 13.3
2.2 2.3 2.4 2.3
50.4 50.3 50.4 50.3
C/E is the ratio of the cellular concentration of antibiotic to the extracellular concentration. Data are means 9 standard errors of the mean. The number of experiments is in parentheses.
Table 2 Influence of cell viability and environmental temperature on entry of azithromycin into human PMN Time of incubation (min)
Antibiotic uptake (C/E)a Viable Cells
15 30 60 120 a b
Dead Cells
37 °C
25 °C
4 °C
37 °C
42.39 18.2 (4)b 61.69 3.9 (12) 146.09 6.1 (14) 387.29 21.5 (12)
2.8(1) 7.5 92.7 (3) 11.2 92.8 (4) 16.9 9 4.1 (3)
2.9(2) 5.5 9 3.5 (3) 8.4 9 3.0 (4) 5.8 9 1.7 (4)
7.1(1) 6.5 (2) 8.5 94.4 (3) 7.3 92.0 (3)
C/E is the ratio of the cellular concentration of antibiotic to the extracellular concentration. Data are means 9 standard errors of the mean. The number of experiments is in parentheses.
W.L. Hand, D.L. Hand / International Journal of Antimicrobial Agents 18 (2001) 419–425
422
Table 3 Effects of metabolic and competitive inhibitors on accumulation of azithromycin by human PMN Time of Incubation (min)
15 30 60 120
Antibiotic uptake (C/E)a Control (No addition)
NaCNb (1 mM)
42.39 18.2 (4)c 61.69 3.9 (12) 146.09 6.1 (14) 387.29 21.5 (12)
– – 53.6 9 0.8 (4) 100.69 6.5 (7)d 247.89 23.3 (8)b, d 131.5 918.2 (4) 607.89 31.5 (8)b, d 376.2 951.4 (4)
Na Azide (2 mM)
KF (1 mM)
Adenosine (1 mM)
D-Glucose (1 mM)
27.1(1) 64.8 94.7 (4) 166.2 913.5 (5) 419.6 943.7 (5)
– 55.1 9 2.6 (3) 148.3 9 17.3 (4) 337.8 9 27.6 (4)
– 63.2 96.1 (3) 132.8 910.1 (4) 358.1 929.1 (4)
a
C/E is the ratio of the cellular concentration of antibiotic to the extracellular concentration. Alkaline incubation medium (pH]8.3). c Data are means 9 standard errors of the mean. The number of experiments is in parentheses. d Significant difference between control (antibiotic only) and experimental (antibiotic+inhibitor) group, PB0.05. b
Table 4 Effects of cell membrane stimulation on uptake of azithromycin by human PMN Time of incubation (min)
15 30 60
Antibiotic uptake (C/E)a Control (No addition)
Zymosan
FMLP
Con A
61.69 3.9 (12)b 146.09 6.1 (14) 387.29 21.5 (12)
39.6 96.5 (3)c 85.4 96.2 (4)c 195.1 915.4 (4)c
48.9 9 8.8 (3) 113.6 920.0 (4) 307.8 949.8 (4)
50.2 9 13.6 (3) 115.8 925.9 (4) 301.3 968.5 (4)
a
C/E is the ratio of the cellular concentration of antibiotic to the extracellular concentration. Data are means 9 standard errors of the mean. The number of experiments is in parentheses. c Significant difference between control (antibiotic only) and experimental (antibiotic+stimulating agent) group, PB0.05. b
Table 5 Efflux of intracellular azithromycin from human PMN after removal of extracellular drugs Time after drug removal (min)
Remaining intracellular azithromycin (% of initial uptake) 37° Control (No addition)
30 60 120 180 a
94.99 1.4 91.6 9 3.8 83.0 9 4.5 80.7 9 6.1
(5)a (6) (5) (5)
25 °C
4 °C
NaCN (1 mM)
KF (1 mM)
No addition
No addition
96.8 92.0 (5) 87.1 97.0 (5) 91.7 (2) 71.1 94.2 (4)
90.5 9 5.7 (5) 95.1 91.9 (5) 78.8 910.5 (4) 77.2 98.5 (3)
92.2 9 6.6 (5) 96.3 93.6 (5) 90.6 95.5 (3) 80.0 910.8 (4)
95.3 9 2.5 (5) 92.8 94.3 (5) 83.1 9 11.7 (3) 83.6 9 9.6 (4)
Data are means 9standard errors of the mean. The number of experiments is in parentheses.
We looked at the possibility that azithromycin might enter human neutrophils by a specific carrier-mediated membrane transport system for nucleosides, hexoses, or amino acids. Accumulation of azithromycin by PMN was not influenced by potential competitive inhibitors of these transport systems (adenosine, D-glucose, glycine, leucine, lysine) (Table 3). The effects of phagocytosis and other cell membrane stimulation on entry of radiolabeled azithromycin into PMN were studied. Ingestion of microbial particles (opsonized zymosan) inhibited the entry of azithromycin into phagocytic cells ( 50% decrease at
60 min.). FMLP and Con A had a lesser inhibitory effect on azithromycin uptake (Table 4).
3.3. Efflux of azithromycin from human PMN The kinetics and characteristics of azithromycin release from human PMN were determined. Efflux of this antibiotic after removal of extracellular drug was extremely slow (Table 5). Greater than 90% of the intracellular drug was retained for 60 min and 80% of the cell-associated azithromycin remained after 3 h. Changes in environmental temperature and inhibitors
W.L. Hand, D.L. Hand / International Journal of Antimicrobial Agents 18 (2001) 419–425
of cellular metabolism (NaCN and KF) failed to alter drug release. However, phagocytosis of zymosan, as well as exposure to FMLP, increased the efflux of azithromycin from phagocytic cells. Of particular interest was that the addition of ammonium chloride (50 mM) and chloroquine (1 mM), agents which neutralize lysosomal pH [35], led to a striking increase in the efflux of azithromycin from PMN (Table 6). Lower concentrations of these lysosomatropic weak bases produced a similar, but less dramatic, increase in antibiotic efflux.
4. Discussion These studies have provided important information concerning the interactions of azithromycin and human PMN. An antibiotic must enter phagocytes to have any activity against viable intracellular organisms. We have shown that most antimicrobial agents fail to penetrate phagocytic cells readily [25,27,30– 33]. Nevertheless, certain antibiotics, including macrolides and clindamycin, are accumulated by phagocytic cells [25– 33,36]. We have demonstrated that azithromycin is massively concentrated by human PMN (C/E = 387.2 at 2 h), far more than any other antibiotic we have tested. Azithromycin accumulation by PMN was dependent upon cell viability and a physiological environmental temperature. The optimal pH for accumulation was : 8.7. We observed similar increased uptake at alkaline pH with other antibiotics which are weak bases and are concentrated within phagocytic cells (erythromycin, roxithromycin, dirithromycin, clindamycin) [31,32,36]. Uptake of clindamycin and several macrolide antibiotics by phagocytic cells requires metabolic energy and is diminished by inhibitors of the predominant energy source in the specific cell type [25– 28,31,32]. In contrast, accumulation of azithromycin by human PMN may not be a metabolically active process as uptake of this antibiotic was not influenced by exposure to fluoride, which inhibits glycolysis, the major source of energy in PMN. Likewise, sodium azide and 2,4-dinitrophenol, inhibitors of oxidative respiration, failed to inhibit azithromycin uptake. However, azithromycin accumula-
423
tion was strikingly increased in PMN exposed to sodium cyanide. This finding is unrelated to any effect of cyanide on mitochondrial oxidative respiration (a minor pathway in PMN, which have few mitochondria), since other inhibitors of oxidative respiration did not alter azithromycin uptake. The cyanide effect theoretically could reflect an inhibitory interaction with a heme (or other iron)-containing enzyme system, such as myeloperoxidase (MPO), a major component of PMN azurophil granules (lysosomes). However, neither sodium azide (which inhibits MPO activity), MPO itself, or inhibitors of reactive oxygen species which interact with MPO, had any influence on azithromycin uptake. Sodium cyanide is strongly alkaline in solution. This elevation in pH apparently accounts for the cyanide-associated increase in azithromycin entry into PMN. Antibiotic uptake in cyanide-containing media was the same as that at the same pH in media without cyanide. Azithromycin uptake returned to control levels with adjustment of the cyanide-containing medium to pH 7.4. We attempted to identify the mechanism(s) by which azithromycin is transported across the cell membrane to enter phagocytic cells. Previously we found that clindamycin enters rabbit AM and human PMN via the cell membrane nucleoside transport system [26,28,34]. In the current study we demonstrated no effect of inhibitors for known phagocytic cell carrier-mediated membrane transport systems on azithromycin uptake. Next, we examined the effect of microbial particle ingestion by PMN on azithromycin uptake. As was true for other macrolide antibiotics [26,29,31,36,37], the accumulation of azithromycin by PMN was moderately inhibited by ingestion of zymosan. This result might be due to internalization of cell membrane, possibly including receptors for macrolides/azolides, or discharge of lysosomal (azurophilic granule) contents, thereby decreasing the accumulation sites for azithromycin. Efflux of azithromycin from human PMN was extremely slow; 80% of the initial intracellular azithromycin was still retained within cells 3 h after removal of extracellular antibiotic. This is in distinct contrast to what we observed with some other selec-
Table 6 Efflux of intracellular azithromycin from human PMN after removal of extracellular drug Time after drug removal (Min)
Remaining intracellular azithromycin (% of initial uptake) Control (No addition)
30 60 120 180 a b
94.99 1.4 91.69 3.8 83.09 4.5 80.7 96.1
(5)a (6) (5) (5)
Zymosan
FMLP
NH4C1 (50 mM)
Chloroquine (1 mM)
80.5 96.1 (5) 76.6 9 3.8(6)b 71.2 97.2 (4) 59.7 9 7.2 (5)
85.6 (2) 71.2 9 7.4(3)b 67.7 (2) 74.3 9 5.5 (3)
12.7 9 3.0 (5)b 12.3 9 4.8 (6)b 8.2 9 3.5 (4)b 4.9 90.6 (5)b
25.1 95.7 (4)b 9.8 (2) 8.9 95.3 (3)b 3.1 9 7 (3)b
Data are means 9standard errors of the mean. The number of experiments is in parentheses. Significant difference between control and experimental group, PB0.05.
424
W.L. Hand, D.L. Hand / International Journal of Antimicrobial Agents 18 (2001) 419–425
tively concentrated antibiotics (erythromycin, clindamycin), which are promptly released from alveolar macrophages and PMN upon removal of extracellular drug [25,32]. Other investigators also reported a rapid efflux of several macrolide antibiotics from human PMN, lymphocytes and tissue culture cells [38–40]. Obviously, these macrolides are not tightly bound or trapped within intracellular compartments. The extremely slow release of azithromycin from PMN indicates that this drug differs from the macrolides noted above in regard to intracellular localization, binding, and/or trapping. Efflux of azithromycin was not altered by metabolic inhibitors or by variations in environmental temperature, observations suggesting that cellular metabolic energy is not involved in this process. Azithromycin is a weakly basic (dibasic) antibiotic which would be protonated and trapped within acidic lysosomes, accounting for the agent’s prolonged cellular retention. Ammonium chloride and chloroquine (agents which neutralize lysosomal pH, preventing protonation and trapping of weak bases) strikingly increased the efflux of azithromycin [35]. These azithromycin efflux characteristics are similar to what we observed in studies of PMN interactions with dirithromycin, another relatively new macrolide antibiotic [36]. Phagocytosis inhibited the accumulation of azithromycin by PMN, but this observation theoretically might be due to enhanced efflux (release) of drug rather than decreased uptake. Indeed, ingestion of zymosan by PMN modestly enhanced the release of azithromycin (after 60 min). This released drug presumably comes from the lysosomal compartment as a result of phagocytosis-induced degranulation. Thus, zymosan ingestion alters both the uptake and release of azithromycin by human PMN. However, the effect on efflux was much less than the inhibitory impact on intracellular accumulation of the antibiotic. In summary, azithromycin was massively accumulated by human PMN. This uptake of azithromycin by neutrophils was far greater than that of any other antibiotic we have studied. Antimicrobial effects of this accumulated intracellular drug might include, (1) a direct action against organisms within phagocytic cells [3,20,21,23,41,42]; and (2) extracellular antibacterial activity after release of intraphagocytic antibiotic at sites of infection. This latter possibility is particularly interesting. The remarkable accumulation and extremely slow release of intracellular azithromycin by human PMN are features which have exciting therapeutic implications. Avid accumulation and prolonged retention of azithromycin by phagocytic cells should allow delivery and subsequent release of drug over time at sites of infection. Ingestion of microorganisms accelerates the release of azithromycin by PMN, and would lead to high extracellular drug concentrations. Thus, the azithromycin-PMN association might be considered an
‘antibiotic delivery system,’ and the subsequent release of azithromycin to the extracellular milieu would represent a true example of targeted antimicrobial therapy. Acknowledgements This study was supported in part by a grant from Pfizer Labs Division, Pfizer Inc, New York, NY. References [1] Frenck RW Jr, Nakhla I, Sultan Y, et al. Azithromycin versus ceftriaxone for the treatment of uncomplicated typhoid fever in children. Clin Infect Dis 2000;31:1134 – 8. [2] Girgis NI, Butler T, Frenck RW, et al. Azithromycin versus ciprofloxacin for treatment of uncomplicated typhoid fever in a randomized trial in Egypt that included patients with multidrug resistance. Antimicrob Agents Chemother 1999;43:1441 –4. [3] Rakita RM, Jacques-Palaz K, Murray BE. Intracellular activity of azithromycin against bacterial enteric pathogens. Antimicrob Agents Chemother 1994;38:1915 – 21. [4] California Collaborative Treatment Group, Havlir DV, Dube´ MP, Sattler FR, et al. Prophylaxis against disseminated Mycobacterium a6ium complex with weekly azithromycin, daily rifabutin, or both. New Engl J Med 1996;335:392 – 8. [5] Horsburgh CR Jr. Advances in the prevention and treatment of Mycobacterium a6ium disease. New Engl J Med 1996;335:428 – 30. [6] Ngeh J, Gupta S. C. pneumoniae and atherosclerosis: causal or coincidental link? ASM News 2000;66:732 – 7. [7] Anderson JL, Muhlestein JB, Carlquist J, et al. Randomized secondary prevention trial of azithromycin in patients with coronary artery disease and serological evidence for Chlamydia pneumoniae infection. The azithromycin in coronary artery disease: elimination of myocardial infarction with Chlamydia (ACADEMIC) study. Circulation 1999;99:1540 – 7. [8] Gupta S, Leatham EW, Carrington D, Mendall MA, Kaski JC, Camm AJ. Elevated Chlamydia pneumoniae antibodies, cardiovascular events, and azithromycin in male survivors of myocardial infarction. Circulation 1997;96:404 – 7. [9] Dunne MW. Rationale and design of a secondary prevention trial of antibiotic use in patients after myocardial infarction: the WIZARD (weekly intervention with Zithromax [azithromycin] for atherosclerosis and its related disorders) trial. J Infect Dis 2000;181(Suppl 3):S572 – 8. [10] Jackson LA. Description and status of the azithromycin and coronary events study (ACES). J Infect Dis 2000;181(Suppl 3):S579 – 81. [11] Foulds G, Shepard RM, Johnson RB. The pharmokinetics of azithromycin in human serum and tissues. J Antimicrob Chemother 1990;25(Suppl A):73 –82. [12] Girard AE, Girard D, Retsema JA. Correlation of the extravascular pharmacokinetics with in-vivo efficacy in models of localized infection. J Antimicrob Chemother 1990;25(Suppl A):61 – 71. [13] Retsema JA, Girard AE, Girard D, Milisen WB. Relationship of high tissue concentrations of azithromycin to bactericidal activity and efficacy in vivo. J Antimicrob Chemother 1990;25(Suppl A):83 – 9. [14] Gladue RP, Bright GM, Isaacson RE, Newborg MF. In vitro and in vivo uptake of azithromycin (CP-62, 993) by phagocytic cells: possible mechanisms of delivery and release at sites of infection. Antimicrob Agents Chemother 1989;33:277 – 82.
W.L. Hand, D.L. Hand / International Journal of Antimicrobial Agents 18 (2001) 419–425 [15] Siegel RE. The significance of serum vs tissue levels of antibiotics in the treatment of penicillin-resistant Streptococcus pneumoniae and community-acquired pneumonia. Are we looking in the wrong place? Chest 1999;116:535 – 8. [16] Amsden GW. Pharmacological considerations in the emergence of resistance. Int J Antimicrob Agents 1999;11(Suppl 1):S7 – S14. [17] Baldwin DR, Wise R, Andrews JM, Ashby JP, Honeybourne D. Azithromycin concentration at the sites of pulmonary infection. Eur Respir J 1990;3:886 –90. [18] Bonnet M, Van der Auwera P. In vitro and in vivo intraleukocytic accumulation of azithromycin (CP-62,993) and its influence on ex vivo leukocyte chemiluminescence. Antimicrob Agents Chemother 1992;36:1302 – 9. [19] Wildfeuer A, Laufen H, Muller-Wening D, Haferkamp O. Interaction of azithromycin and human phagocytic cells: uptake of the antibiotic and the effect on the survival of ingested bacteria. Arzneim-Forsch/Drug Res 1989;39:755 –8. [20] Meyer AP, Bril-Bazuin C, Mattie H, van den Broek PJ. Uptake of azithromycin by human monocytes and enhanced intracellular antibacterial activity against Staphylococcus aureus. Antimicrob Agents Chemother 1993;37:2318 –22. [21] Stamler DA, Edelstein MAC, Edelstein PH. Azithromycin pharmacokinetics and intracellular concentrations in Legionella pneumophila-infected and uninfected guinea pigs and their alveolar macrophages. Antimicrob Agents Chemother 1994;38:217 – 22. [22] Fietta A, Merlini C, Grassi GG. Requirements for intracellular accumulation and release of clarithromycin and azithromycin by human phagocytes. J Chemother 1997;9:23 –31. [23] Wildfeuer A, Laufen H, Zimmerman T. Uptake of azithromycin by various cells and its intracellular activity under in vivo conditions. Antimicrob Agents Chemother 1996;40:75 – 9. [24] Boyum A. Isolation of lymphocytes, granulocytes, and macrophages. Scand J Immunol 1976;5(Suppl 5):9 –15. [25] Prokesch RC, Hand WL. Antibiotic entry into human polymorphonuclear leukocytes. Antimicrob Agents Chemother 1982;21:373 – 80. [26] Steinberg TH, Hand WL. Effects of phagocytosis on antibiotic and nucleoside uptake by human polymorphonuclear leukocytes. J Infect Dis 1984;149:397 – 403. [27] Hand WL, Corwin RW, Steinberg TH, Grossman GD. Uptake of antibiotics by human alveolar macrophages. Am Rev Resp Dis 1984;129:933 – 7. [28] Hand WL, King-Thompson NL. Membrane transport of clindamycin in alveolar macrophages. Antimicrob Agents Chemother 1982;21:241 – 7.
425
[29] Hand WL, King-Thompson NL. Contrasts between phagocyte antibiotic uptake and subsequent intracellular bactericidal activity. Antimicrob Agents Chemother 1986;29:135 – 40. [30] Hand WL, King-Thompson NL. Entry of antibiotics into human monocytes. J Antimicrob Chemother 1989;23:681 – 9. [31] Hand WL, King-Thompson NL, Holman JW. Entry of roxithromycin (RU965), imipenem, cefotaxime, trimethoprim, and metronidazole into human polymorphonuclear leukocytes. Antimicrob Agents Chemother 1987;31:1553 – 7. [32] Johnson JD, Hand WL, Francis JB, King-Thompson N, Corwin RW. Antibiotic uptake by alveolar macrophages. J Lab Clin Med 1980;95:429 – 39. [33] Steinberg TH, Hand WL. Effect of phagocyte membrane stimulation on antibiotic uptake and intracellular bactericidal activity. Antimicrob Agents Chemother 1987;31:660 – 2. [34] Hand WL, King-Thompson NL. Uptake of antibiotics by human polymorphonuclear leukocyte cytoplasts. Antimicrob Agents Chemother 1990;34:1189 –93. [35] Klempner MS, Styrt B. Alkalinizing the intralysosomal pH inhibits degranulation of human neutrophils. J Clin Invest 1983;72:1793 – 800. [36] Hand WL, Hand DL. Interactions of dirithromycin with human polymorphonuclear leukocytes. Antimicrob Agents Chemother 1993;37:2557 – 62. [37] Hand WL, Hand DL. Influence of pentoxifylline and its derivatives on antibiotic uptake and superoxide generation by human phagocytic cells. Antimicrob Agents Chemother 1995;39:1574 –9. [38] Martin JR, Johnson P, Miller MF. Uptake, accumulation, and egress of erythromycin by tissue culture cells of human origin. Antimicrob Agents Chemother 1985;27:314 – 9. [39] Dette GA, Knothe H. Kinetics of erythromycin uptake and release by human lymphocytes and polymorphonuclear leukocytes. J Antimicrob Chemother 1986;18:73 – 82. [40] Ishiguro M, Koga H, Kohno S, Hayashi T, Yamaguchi H, Hirota M. Penetration of macrolides into human polymorphonuclear leukocytes. J Antimicrob Chemother 1989;24:719 – 29. [41] Donowitz GR, Earnhardt KI. Azithromycin inhibition of intracellular Legionella micdadei. Antimicrob Agents Chemother 1993;37:2261 – 4. [42] Jonas D, Engels I, Daschner FD, Frank Y. The effect of azithromycin on intracellular Legionella pneumophila in the Mono Mac 6 cell line at serum concentrations attainable in vivo. J Antimicrob Chemother 2000;46:385 – 90.