Polymorphonuclear Leukocytes

Transcription

1 ANTIMICROBIAL AGENTS AND CHEMOTHERAPY, OCt. 1987, p /87/ $02.00/0 Copyright C 1987, American Society for Microbiology Vol. 31, No. 10 Entry of Roxithromycin (RU 965), Imipenem, Cefotaxime, Trimethoprim, and Metronidazole into Human Polymorphonuclear Leukocytes W. LEE HAND,* NEVA KING-THOMPSON, AND JOHN W. HOLMAN Veterans Administration Medical Center (Atlanta), Decatur, Georgia 30033,* and Department of Medicine, Emory University School of Medicine, Atlanta, Georgia Received 26 March 1987/Accepted 6 July 1987 Entry of antibiotics into phagocytes is necessary for activity against intracellular organisms. Therefore, we examined the uptake of five of the newer antibiotics-roxithromycin (RU 965), imipenem, cefotaxime, trimethoprim, and metronidazole-by human polymorphonuclear leukocytes (PMN). Antibiotic uptake by PMN was determined by a velocity gradient centrifugation technique and expressed as the ratio of the cellular concentration of antibiotic to the extracellular concentration (C/E). Cefotaxime, like other I-lactam antibiotics, was taken up poorly by phagocytes (C/E ' 0.3). The metronidazole concentration within PMN was similar to the extracellular level. Imipenem bound rapidly to phagocytes (C/E = 3), but cell-associated drug progressively declined during the period. Trimethoprim was well concentrated by PMN (C/E = 9 to 13), and uptake was unexpectedly greater at 25 C than at 37 C. The most striking finding was that roxithromycin was more avidly concentrated by PMN (C/E = 34) than any other antibiotic we studied. Entry of roxithromycin into phagocytes was an active process and displayed saturation kinetics characteristic of a carrier-mediated membrane transport system. Ingestion of microbial particles by PMN slightly decreased the ability of these cells to accumulate roxithromycin (C/E = 24 to 31). These studies identified two antibiotics, trimethoprim and especially roxithromycin, which are markedly concentrated within human PMN and may prove useful in treatment of infections caused by susceptible intracellular organisms. The interactions of antibiotics with phagocytes, and the influence of these interactions on the fate of intraphagocytic bacteria, may be of therapeutic importance. Obviously, entry of antibiotics into phagocytic cells is essential for activity against intracellular organisms. The use of therapeutic agents which are able to penetrate phagocytic cells and demonstrate intracellular activity is especially desirable in therapy of infections caused by facultative intracellular organisms (9). Unfortunately, as we and others have demonstrated, most antimicrobial agents have limited cellular penetration, and only a few antibiotics are taken up well by phagocytes. For example, various,-lactam antibiotics and aminoglycosides enter phagocytes poorly, whereas lipidsoluble antibiotics such as rifampin and chloramphenicol attain intracellular levels which exceed their extracellular concentrations (6, 10, 11). In our studies to date, the antibiotics which have achieved the highest intraphagocytic concentrations are clindamycin and erythromycin. These antimicrobial agents are markedly concentrated by means of active transport systems in both macrophages and polymorphonuclear leukocytes (PMN) (6, 7, 10, 11, 13). In the present study we examined the ability of several of the newer antibiotics to enter human PMN, the crucial phagocytic cells, in protection against most bacterial pathogens. The antimicrobial agents evaluated included roxithromycin (RU 965; an investigational macrolide), imipenem, cefotaxime, trimethoprim, and metronidazole. Of great interest was the observation that roxithromycin was massively * Corresponding author accumulated by PMN, achieving the highest intraphagocytic levels of any antibiotic studied thus far. MATERIALS AND METHODS Preparation of human PMN. Peripheral venous blood was collected by venipuncture from normial human volunteers. Granulocytes (largely PMN) were isolated by dextran sedimentation and Hypaque-Ficoll density gradient centrifugation (3, 11, 13). The cells were washed and suspended in tissue culture medium 199 (TC199) or Hanks balanced salt solution (HBSS) (GIBCO Laboratories, Grand Island, N.Y.). Determination of antibiotic entry into human PMN. Uptake of radiolabeled antibiotics by PMN was evaluated by means of a velocity gradient centrifugation technique which we have described in detail elsewhere (6, 7, 10, 11, 13). Human PMN in TC199 or HBSS were incubated with "therapeutic" concentrations of radiolabeled antibiotics (-2.5 x 10-5 M). The antibiotics used in this study were [14C]roxithromycin (54.5 mci/mmol) and [thiazolyl-2-14c]cefotaxime (9.5 mci/ mmol) (Hoechst-Roussel Pharmaceuticals Inc., Somerville, N.J.), ['4C]metronidazole (12 mci/mmol; Searle Research and Development), [ring-2-'4c]imipenem (15.6 mnci/mmol; Merck, Sharp and Dohme, West Point, Pa.), and [3H]trimethoprim (60.3 mci/mmol; Hoffman-LaRoche Inc., Nutley, N.J.). At intervals, phagocytes with their associated radioactive antibiotic were separated from the extracellular antibiotic by velocity gradient centrifugation in a microcentrifuge tube (6, 7, 10, 11, 13). This was accomplished by centrifugation of PMN through a water-impermeable barrier of silicone oil

2 1554 HAND ET AL. ANTIMICROB. AGENTS CHEMOTHER. TABLE 1. Entry of antibiotics into human PMN Antibiotic uptake (C/E)a Time of (min) Roxithromycin Trimethoprim Imipenem Metronidazole Cefotaxime ± 0.55 (13) 4.30 ± 0.32 (15) 3.29 ± 0.57 (13) 0.85 ± 0.11 (3) 0.13 ± 0.04 (3) 5 b _ 2.77 ± 0.73 (8) ± 2.16 (34) 9.02 ± 0.77 (16) 1.03 ± 0.21 (9) 0.97 ± 0.14 (3) 0.31 ± 0.09 (3) ± 2.08 (37) 8.50 ± 0.68 (15) 0.84 ± 0.36 (8) 1.00 ± 0.20 (3) 0.27 ± 0.10 (3) ± 2.94 (22) 8.23 ± 0.86 (13) 0.09 ± 0.11 (5) 1.03 ± 0.22 (3) 0.04 ± 0.05 (3) a Data are means ± standard error of the mean. The number of experiments is shown in parentheses. b -, No studies performed. into formic acid, which dissolves the cells. The tubes were then frozen, and the layers were separated by slicing with a razor. The radioactive content of the lower layer, containing the radiolabeled antibiotic which entered the cells, and the upper layer, containing the antibiotic still in solution, was determined in a liquid scintillation counter. Antibiotic uptake was expressed as the ratio of the cellular concentration of antibiotic to the extracellular concentration (C/E). Characterization of antibiotic uptake. Studies were performed to define the mechanisms and characteristics of antibiotic uptake by PMN. First, we evaluated the environmental and metabolic requirements of the uptake process for those antibiotics which enter phagocytes readily (6, 7, 10, 11). Thus, the influences of cell viability, environmental temperature, ph, and metabolic inhibitors on antibiotic accumulation by PMN were examined. The metabolic inhibitors employed in this study were sodium cyanide (Sigma Chemical Co., St. Louis, Mo.) and potassium fluoride (Mallinckrodt Inc., St. Louis, Mo.). Cells in HBSS with and without an inhibitor were incubated for 30 min at 37 C. Radiolabeled antibiotic was added, and uptake was measured as described above. The influence of phagocytosis and other cell membrane stimulation.on antibiotic entry into PMN was also evaluated (8, 13). PMN (107 cells per ml) were incubated with ingestible particles (opsonized zymosan and Staphylococcus aureus) or with a soluble membrane-perturbating agent (concanavalin A) for 30 min at 37 C, washed, and suspended in TC199, after which antibiotic uptake was determined. Kinetic analysis of antibiotic uptake during the initial exposure of PMN to the drug was performed. Cells were exposed to a wide range of radiolabeled antibiotic concentrations for 1 min. Velocity of transport was determined for each concentration, and a double-reciprocal (Lineweaver- Burke) plot of uptake velocity versus drug concentration data was constructed. This allowed calculation of the apparent binding affinity (Kin) and maximum velocity or rate of transport (Vmax) (7, 11). Since phagocytes have specific carrier-mediated membrane transport systems for hexoses, amino acids, and nucleosides (2, 14), we examined the possibility that one or more of these systems might transport the tested antibiotics into PMN. We have previously demonstrated that clindamycin is transported into phagocytes by the cell membrane nucleoside system (7, 13). Substances evaluated for possible inhibition of antibiotic uptake in PMN included L-amino acids (glycine, leucine, lysine, glutamic acid, and aspartic acid; Pierce Chemical Co., Rockford, Ill.), D-glucose (Sigma), and nucleosides. These nucleosides were adenosine, 2-chloroadenosine, nitrobenzylthioinosine (all from Sigma), and N6-phenylisopropyladenosine (Boehringer Mannheim Biochemicals, Indianapolis, Ind.). These substances were preincubated with PMN for 20 min prior to determination of radiolabeled antibiotic uptake. Statistical analysis of data. Comparisons between experimental groups were carried out by means of two-sample or paired t tests with a Tektronic 4051 computer and an appropriate statistical program. RESULTS Characteristics of the cell population. At least 97% of the granulocytic cells were neutrophils, and >90% of these were PMN. Thus, the cells are referred to as PMN. More than 95% of these phagocytes were viable, as judged by trypan blue exclusion. Antibiotic uptake by human PMN. Cefotaxime, like the,b-lactam antibiotics which we evaluated previously, penetrated phagocytic cells poorly. The cellular concentration of this drug was much lower than the extracellular level (C/ E 0.3; Table 1). Imipenem, a novel,-lactam antibiotic, demonstrated an unusual interaction with PMN. The drug bound to phagocytes rapidly, but cell-associated antibiotic declined steadily during a 1-h period. Metronidazole achieved a concentration in PMN which was approximately equal to the extracellular concentration (C/E 1). Cellular entry of this drug was rapid (probably due to lipid solubility), and there was little change in the intracellular concentration over time (Table 1). In contrast to the above antibiotics, trimethoprim and roxithromycin were markedly concentrated by PMN. Trimethoprim was concentrated eight- to ninefold by PMN, and the characteristics of this entry process were somewhat unusual (see below). Roxithromycin was rapidly and massively concentrated by human PMN (Table 1). Uptake of this antibiotic by human PMN (C/E = 34) was greater than that of any other antimicrobial agent we tested. Characterization of roxithromycin uptake in human PMN. The avid uptake of roxithromycin by human PMN stimulated us to examine this process in considerable detail. Entry of roxithromycin into PMN was greater when studies were performed in HBSS (C/E = 91) than in TC199 (C/E = 34) TABLE 2. Uptake of roxithromycin by human PMN Time of Uptake (C/E)0 (min) TC199 TC199 + serum HBSS ± 0.54 (12) 3.75 ± 0.55 (15) 5.44 ± 0.83 (14) ± 1.71 (12) ± 2.16 (34) ± 6.02 (27) ± 2.47 (12) ± 2.08 (37) ± 5.71 (30) ± 4.07 (6) ± 2.94 (22) ± 5.43 (23) a Data are means ± standard error of the mean. The number of experiments is shown in parentheses.

3 VOL. 31, 1987 ENTRY OF ANTIBIOTICS INTO HUMAN PMN 1555 TABLE 3. Influence of cell viability and environmental temperature on entry of roxithromycin and trimethoprim into human PMN Antibiotic uptake (C/E)l Time (min) of with antibiotic Viable cells Dead cells, 37 C 37 C 250C 4 C Roxithromycin ± 0.54 (12) 2.20 ± 0.45 (3) 0.32 ± 0.26 (3) ± 1.71 (12) 7.52 ± 0.56 (3) 0.62 ± 0.43 (3) ± 2.47 (12) ± 2.01 (3) 0.68 ± 0.37 (3) ± 4.07 (6) ± 4.14 (3) 0.85 ± 0.44 (3) 6.37 Trimethoprim ± 0.32 (15) 2.48 ± 0.49 (6) 0.39 ± 0.28 (5) 3.11 ± 0.71 (3) (16) ± 1.12 (7)b 1.82 ± 0.50 (5) 2.71 ± 0.57 (3) ± 0.68 (15) ± 1.31 (7)' 2.49 ± 0.67 (5) 2.29 ± 0.58 (3) (13) ± 1.66 (7)d 6.01 ± 1.69 (5) 1.93 ± 0.36 (3) a Data are means + standard error of the mean. The number of experiments is shown in parentheses. P values reflect differences between trimethoprim uptake by viable PMN at 25 C and at 37 C. b p Pp = d p = (Table 2). This observation suggested that one or more of the components in the complex tissue culture medium inhibited entry of the antibiotic. The presence of serum had little effect on roxithromycin uptake by PMN. The environmental and metabolic requirements of the roxithromycin uptake process were characterized. Cellular uptake of the antibiotic was dependent on cell viability and a physiologic environmental temperature (Table 3). Inhibitors of cellular metabolism were examined for their influence on antibiotic accumulation. Both sodium cyanide, an inhibitor of mitochondrial oxidative respiration, and potassium fluoride, which inhibits glycolysis, modestly decreased roxithromycin entry into human PMN (Table 4). The ph profile for roxithromycin entry into phagocytes is shown in Fig. 1. The optimal ph for this process was approximately 8, similar to what we found with other weakly basic antibiotics (clindamycin, erythromycin) which are actively transported into phagocytes (10). A kinetic analysis of roxithromycin uptake in human PMN (Lineweaver-Burk reciprocal plot) is shown in Fig. 2. Roxithromycin uptake displayed saturation kinetics characteristic of a carrier-mediated transport system. This transport system had a relatively high binding constant (Kin) and rate of uptake (Vmax). Phagocytes have specific carrier-mediated membrane transport systems for hexoses, amino acids, and nucleosides, so we evaluated the possibility that roxithromycin might enter PMN by one of these systems. Roxithromycin transport was not influenced by the presence of hexose or amino acids (data not shown) but was inhibited by certain nucleosides (adenosine and chloroadenosine) (Table 5). The means by which nucleosides inhibited roxithromycin uptake is not clear, since formal kinetic (velocity concentration) studies showed no competitive inhibition of initial (1-min) roxithromycin transport by nucleosides. It should be noted that nucleosides were not toxic for human PMN, as assessed by trypan blue exclusion and lactate dehydrogenase release. Finally, we evaluated the influence of phagocytosis and other cell membrane stimulation on uptake of radiolabeled roxithromycin by PMN, since we had demonstrated previously that entry of antibiotics (clindamycin, erythromycin) actively transported by cell membrane systems, as well as specific nucleoside transport, was affected by phagocytosis (8, 13). In the present study we found that ingestion of phagocytic particles (specifically S. aureus) inhibited the entry of roxithromycin into human PMN (Table 6). In contrast, concanavalin A had no effect on the roxithromycin transport process. Environmental and metabolic requirements of trimethoprim and imipenem uptake by human PMN. Entry of trimethoprim into human PMN was dependent on cell viability (Table 3). A small amount of drug bound rapidly to dead cells, but this slowly decreased over time. The effect of environmental temperature on trimethoprim uptake was TABLE 4. Effects of metabolic inhibitors on entry of roxithromycin and trimethoprim into human PMN Antibiotic uptake (C/E)' Time of with antibiotic (min) Control (no addition) NaCN (1 mm) KF (1 mm) Roxithromycin ± 2.94 (8) ± 3.17 (7), P = ± 3.36 (8), P = ± 3.26 (8) ± 2.88 (7), P = ± 3.31 (8), P = ± 2.72 (6) ± 2.52 (5), P = ± 3.05 (6), P = Trimethoprim ± 0.37 (9) 5.03 ± 0.83 (8), P = ± 0.97 (9), P = ± 0.78 (10) 7.45 ± 1.13 (10), P = ± 1.09 (9), P = ± 1.10 (9) 8.66 ± 1.03 (7), P = ± 1.01 (9), P = ± 1.05 (5) 8.59 ± 1.48 (6), P = ± 1.19 (6), P = a Data are means ± standard error of the mean. The number of experiments is shown in parentheses. P values reflect differences between control and experimental (metabolic inhibitor) groups.

4 1556 HAND ET AL. ANTIMICROB. AGENTS CHEMOTHER. C/ E F lop I I, I FIG. 1. Influence of ph on uptake (C/E) of roxithromycin by human PMN. Means of uptake determinations at each ph value are shown. interesting. As expected, at 4 C there was a very slow entry of antibiotic into PMN. However, the significantly greater entry of trimethoprim into PMN at 25 C than at 37 C was quite surprising. This type of "inverse" temperature-uptake relationship was not observed with any of the antibiotics we had studied previously. Entry of trimethoprim into PMN was slightly, but not significantly, decreased in the presence of potassium fluoride. Sodium cyanide had no effect on drug uptake (Table 4). Next, we attempted to elucidate the mechanism of the unusual decrease in imipenem associated with PMN over time. Antibiotic uptake studies were performed in TC199 and HBSS with and without serum. In some experiments, the various media were preincubated with radiolabeled imipenem (to detect any drug "breakdown") prior to determination of antibiotic uptake. None of these maneuvers had any consistent effect on imipenem uptake by human PMN. Trypsinization of PMN failed to inhibit uptake of imipenem. DISCUSSION These studies provided substantial information about the interactions of several important antimicrobial agents and human phagocytes. It is obvious that entry of antibiotics into phagocytic cells is a prerequisite for activity against intracellular organisms. Unfortunately, as we demonstrated previously, certain types of antibiotics fail to penetrate phagocytes readily (6, 10, 11). Thus, cefotaxime, like other,-lactam antibiotics we have studied, was taken up poorly by phagocytes. Metronidazole also had a somewhat limited ability to enter human PMN, achieving a cellular concentration which was similar to the extracellular level. It would be interesting to examine the uptake of this drug under anaerobic conditions, since metronidazole exhibits antimicrobial activity only against anaerobic bacteria (5). ph 8.0 I 4n I o 9.0 E Vmax= 2.86nmol/min/2xlO6cells Km= 238pM I II I ojo I/ M FIG. 2. Kinetic analysis of 1t4CJroxithromycin transport in human PMN (Lineweaver-Burk reciprocal plot of uptake velocity versus antibiotic concentration data). Cells were incubated for 1 min with various concentrations of roxithromycin, and the transport of drug at each concentration was determined. Means of observations at each roxithromycin concentration are shown. The interaction of imipenem with human PMN was unusual. This antibiotic bound rapidly to phagocytes, but the amount of cell-associated drug progressively declined during the period. There are several possible explanations for this phenomenon, including rapid binding to cell membrane followed by dissociation, extracellular hydrolysis or other alteration of the antibiotic, and cellular metabolism of the drug. The ability of trimethoprim to penetrate phagocytes (C/E = 9 to 13) may contribute to its success (in the trimethoprimsulfamethoxazole combination) against intracellular pathogens such as Pneumocystis carinii. An unexpected finding, for which we do not have a ready explanation, was that trimethoprim uptake by human PMN was greater at 25 C (C/E = 13) than at 37 C (C/E = 9). This inverse temperaturedrug uptake relationship has not been seen with other antibiotics. Obviously, the most striking finding was that roxithromycin was taken up and concentrated more efficiently by human PMN than any of the other 18 antibiotics we tested. Thus, PMN uptake of roxithromycin (C/E = 34) was substantially greater than the previously "best-transported" antibiotics, clindamycin (C/E = 11) and two macrolides, erythromycin and erythromycin propionate (C/E = 10 to 15). TABLE 5. Effects of competitive inhibitors on entry of roxithromycin into human PMN Time of Uptake (CIE)' (min) Control (HBSS) Adenosine (1 mm) 2-Chloroadenosine (1 mm) NBTI (0.01 mm) PIA (0.1,uM) (14) 5.76 ± 0.64 (6), P = ± 3.43 (4), P = ± 2.64 (3) ± 1.02 (2) ± 6.02 (27) ± 6.99 (5), P = ± 8.72 (6), P = ± 5.45 (5), P = ± 0.14 (2) ± 5.71 (30) ± 9.92 (7), P = ± 7.07 (7), P = ± 5.82 (5), P = (2) ± 5.43 (23) ± 8.07 (6), P = ± (5), P = (5), P = a Data are means ± standard error of the mean. The number of experiments is shown in parentheses. P values reflect differences between control (HBSS only) and experimental (nucleoside) groups. NBTI, nitrobenzylthioinosine; PIA, N6-phenylisopropyladenosine.

5 VOL. 31, 1987 ENTRY OF ANTIBIOTICS INTO HUMAN PMN 1557 TABLE 6. Effects of membrane stimulation on uptake of roxithromycin by human PMN Time of Uptake (C/E)I (min) Control (TC199) Zymosan S. aureus Concanavalin A ± 0.55 (15) 1.88 ± 0.50 (5), P = ± 0.20 (5), P = ± 0.84 (7), P = (34) ± 2.85 (9), P = ± 1.89 (21), P = ± 4.09 (9), P = ± 2.08 (37) ± 3.17 (10), P = ± 2.16 (24), P = ± 4.27 (9), P = ± 2.94 (22) ± 3.71 (6), P = ± 3.38 (12), P = ± 5.76 (4), P = a Data are means - standard error of the mean. The number of experiments is shown in parentheses. P values reflect differences between control (TC199 only) and experimental groups. Entry of roxithromycin into human PMN proved to be an active process, dependent on cell viability, a physiological environmental temperature, ph, and metabolic energy. Kinetic analysis of the uptake process revealed that it displayed saturation kinetics typical of a carrier-mediated membrane transport system. These characteristics are similar to what was observed with the two erythromycin preparations. The uptake of roxithromycin by PMN was inhibited in the presence of nucleosides. However, this inhibitory effect did not appear to be competitive. Therefore, the membrane transport system responsible for roxithromycin entry into PMN has not been clearly identified. The influence of phagocytosis and other cell membrane stimulation on uptake of radiolabeled roxithromycin by PMN was of interest, since the ability of antimicrobial agents to enter phagocytes may be a major determinant of therapeutic efficacy during infections caused by bacteria which are ingested but not killed efficiently by phagocytes. Ingestion of microbial particles (zymosan, S. aureus) by human PMN decreased the ability of these phagocytes to accumulate roxithromycin, probably as a consequence of cell membrane internalization. This inhibitory effect of microbial particle ingestion on roxithromycin uptake is similar to what we demonstrated previously with erythromycin. However, postingestion uptake of roxithromycin by PMN was still very impressive (C/E = 31 after zymosan ingestion, C/E = 24 after S. aureus ingestion). The results of these studies with roxithromycin are quite interesting. This macrolide antibiotic is generally similar to erythromycin in its antimicrobial spectrum (1). Because of certain pharmacokinetic advantages and its remarkable ability to enter phagocytes, roxithromycin might prove superior to erythromycin in treatment of infections caused by susceptible intracellular organisms (e.g., Legionella, Listeria, Chlamydia, and Toxoplasma spp.) (4, 12). Obviously, additional in vitro and in vivo studies are essential to evaluate this area of interest. These studies have identified two antibiotics, trimethoprim and especially roxithromycin, which are markedly concentrated within human PMN. Entry into phagocytes is a necessary, but certainly not the only, determinant of the intracellular antibacterial activity of an antibiotic. Therefore, it is essential to evaluate the consequences of roxithromycin and trimethoprim uptake by human PMN on intraphagocytic bactericidal activity and other phagocyte functions. ACKNOWLEDGMENTS This work was supported by the Medical Research Service of the Veterans Administration and by a grant from Hoechst-Roussel Pharmaceuticals Inc. Searle Research and Development, Merck, Sharp and Dohme Research Laboratories, Hoffman-LaRoche Inc., and Hoechst- Roussel Pharmaceuticals Inc. kindly provided the radiolabeled antibiotics used in these studies. We thank Lynne Wilson for preparation of the manuscript. LITERATURE CITED 1. Barlam, T., and H. C. Neu In vitro comparison of the activity of RU 28965, a new macrolide, with that of erythromycin against aerobic and anaerobic bacteria. Antimicrob. Agents Chemother. 25: Berlin, R. D Membrane transport in macrophages, p In R. Van Furth (ed.), Mononuclear phagocytes. Blackwell Scientific Publications, Inc., Oxford. 3. Boyum, A Isolation of lymphocytes, granulocytes and macrophages. Scand. J. Immunol. 5(Suppl. 5): Chan, J., and B. J. Luft Activity of roxithromycin (RU 28695), a macrolide, against Toxoplasma gondii infection in mice. Antimicrob. Agents Chemother. 30: Feingold, S. M Metronidazole, p In G. L. Mandell, R. G. Douglas, Jr., and John E. Bennett (ed.), Principles and practice of infectious diseases, 2nd ed. John Wiley & Sons, Inc., New York. 6. Hand, W. L., R. W. Corwin, T. H. Steinberg, and G. D. Grossman Uptake of antibiotics by human alveolar macrophages. Am. Rev. Respir. Dis. 129: Hand, W. L., and N. L. King-Thompson Membrane transport of clindamycin in alveolar macrophages. Antimicrob. Agents Chemother. 21: Hand, W. L., and N. L. King-Thompson Contrasts between phagocyte antibiotic uptake and subsequent intracellular bactericidal activity. Antimicrob. Agents Chemother. 29: Horwitz, M. A Phagocytosis of microorganisms. Rev. Infect. Dis. 4: Johnson, J. D., W. L. Hand, J. B. Francis, N. King-Thompson, and R. W. Corwin Antibiotic uptake by alveolar macrophages. J. Lab. Clin. Med. 95: Prokesch, R. C., and W. L. Hand Antibiotic entry into human polymorphonuclear leukocytes. Antimicrob. Agents Chemother. 21: Stamm, W. E., and R. Suchland Antimicrobial activity of U-70138F (paldimycin), roxithromycin (RU 965), and ofloxacin (ORF 18489) against Chlamydia trachomatis in cell culture. Antimicrob. Agents Chemother. 30: Steinberg, T. H., and W. L. Hand Effects of phagocytosis on antibiotic and nucleoside uptake by human polymorphonuclear leukocytes. J. Infect. Dis. 149: Zuckerman, S. H., and S. D. Douglas Dynamics of the macrophage plasma membrane. Annu. Rev. Microbiol. 33: