Journal of Antimicrobial Chemotherapy (1992) 30, 509-523
Investigation of the in-vitro uptake, intraphagocytic biological activity and effects on nentrophil superoxide generation of dirithromycin compared with erythromycin G. K. Joone, C. E. J. Van Rensburg and R. Anderson
The cellular uptake by human neutrophils and the intraphagocytic biological activity of the new macrolide antimicrobial agent dirithromycin (0-01-2 mg/L) compared with erythromycin was investigated in vitro. Staphylococcus aureus, LJsteria monocytogenu and Legionella pneumophila were used as the test intraccllular microbial pathogens. After coincubation (45 min at 37°Q of neutrophils with a fixed concentration of 2 mg/L of each antibiotic the respective intraccllular/ extracellular ratios for erythromycin and dirithromycin were 6-1 ±2-5 and 10-6±2 respectively (P < 0-005). Using a combination of techniques (colony counting, radiometry andfluorescencemicroscopy) both erythromycin and dirithromycin at concentrations of 0-01 and 0-5 mg/L and higher, respectively, were found to possess dose-related intraphagocytic bacteristatic activity for each of the test microbial pathogens. The effects of dirithromycin and erythromycin (1-20 mg/L) on neutrophil chemotaxis and generation of reactive oxidants by these cells were also investigated in vitro. Both antimicrobial agents caused a dose-related stimulation of neutrophil migration which was associated with inhibition of leucoattractant-activated generation of superoxide and activity of the myeloperoxidase/HjOj/halide system. However, superoxide generation by neutrophils activited with opsonized zymosan or phorbol myristate acetate was unaffected by the macrolides. These findings demonstrate that dirithromycin accumulates in human neutrophils, is biologically active intracellularly and modulates leucoattractant-activated superoxide generation and chemotaxis. Introduction Dirithromycin (LY 237216), an oxazime derivative of eTythromycin, is a new macrolide antimicrobial agent with improved pharmacokinetics (Kirst & Sides, 1989). Dirithromycin is metabolized in humans to erythromycylamine which has an extremely long half-life ( > 24 h), allowing once daily administration (Kirst & Sides, 1989). The activity of dirithromycin is similar to that of erythromycin against streptococci, pneumococci and Bordetella pertussis, but it is less active against Haemophilus influenzae and Campylobacter pylori (Garcia-Rodriguez et al., 1989; Hoppe & Eichhorn, 1989; Yu & Neu, 1990). Macrolides, which are actively concentrated by human phagocytes (Prokesch & Hand, 1982; Anderson et al., 1987; Hand, King-Thompson & Holman, 1987; Anderson, Joone & Van Rensburg, 1988; Ishiguro et al., 1989), appear to interact cooperatively with these cells to eliminate intraphagocytic microbial pathogens (Miller et 509 0305-7453/92/100509+ 15 $08.00/0
© 1992 The Britiih Society for Antimicrobial Chemotherapy
Downloaded from http://jac.oxfordjournals.org/ at Stanford Medical Center on May 11, 2015
Medical Research Council Unit For The Study of Phagocyte Function, Department of Immunology, Institute For Pathology, University of Pretoria, Pretoria 0001, Republic of South Africa
510
G. K. Jooofe et aL
al., 1984; Anderson, Joone & Van Rcnsburg, 1986; Anderson et al., 1988; Ishiguro et al., 1989). These co-operative interactions between macrolides and phagocytes have been described for erythromycin (Anderson et al., 1986), josamycin (Labro & Babin-Chevaye, 1989), roxithromycin (Labro et al., 1986; Anderson et al., 1987), clarithromycin (Anderson et al., 1988; Ishiguro et al, 1989) and azithromycin (Gladue et al., 1989; Wildfeuer et al., 1989), but not for dirithromycin. In the present study the uptake of dirithromycin and erythromycylamine compared with erythromycin by human neutrophils, and their intraphagocytic biological activities and effects on neutrophil locomotion and generation of reactive oxidants were investigated.
Antimicrobial agents Erythromycin base, dirithromycin, erythromycylamine and radiolabelled dirithromycin (3H-dirithromycin, 16-7 Ci/mmol) were kindly provided by the Lilly Research Laboratories, Indianapolis, Indiana, USA. Radiolabelled erythromycin (N-methylI4 C- erythromycin, specific activity 55 mCi/mmol) was purchased from Du Pont NEN Research Products, Boston, USA. Each of the test compounds (4 mg) was dissolved in 1 mL of 30% aqueous ethanol and diluted in distilled water to give stock solutions of 200 mg/L. The antibiotics were used in assays of neutrophil functions at final concentrations of 0-01-20 mg/L. Solvent controls were included. Chemicals and reagents Unless indicated all other chemicals and reagents were obtained from the Sigma Chemical Co., St Louis, USA and radiochemicals from Du Pont NEN. Bacteria Staphylococcus aureus (ATCC 25923), Usteria monocytogenes (laboratory isolate) and Legionella micdadei (ATCC 33218) were used as the test microbial pathogens. Antibiotic sensitivity testing in vitro The MICs and MBCs of the test antibiotics for the three microbial pathogens were determined by standard bacteriological assays (Reiner, 1982). Neutrophils These cells were prepared as previously described (Anderson et al., 1986) from venous blood taken from adult human volunteers and a child with the autosomal recessive form of chronic granulomatous disease (CGD). The neutrophils were resuspended to a concentration of 1 x 107/mL in indicator-free Hanks' balanced salt solution (HBSS) buffered with 4-2 mM HEPES, pH 7-4. Measurements of the uptake of radiolabelled antibiotics by neutrophils This was measured according to the method previously described by Prokesch & Hand (1982). Briefly 100 //L of HBSS containing 2x10* neutrophils and 2 mg/L of radio-
Downloaded from http://jac.oxfordjournals.org/ at Stanford Medical Center on May 11, 2015
Materials and methods
Intraphagocytk acthrtty of dMthromydn
511
labelled dirithromycin or erythromycin were incubated over 150 /iL of silicone oil in microcentrifuge tubes at 37°C for 45 min (a predetermined optimum incubation time). Unbound and neutrophil-associated antibiotics were separated by centrifuging the neutrophils through the oil at 12,000 g for 3 min. After freezing the tubes at - 70°C the neutrophil pellets were sliced from the bottom of each tube and solubilized in Protosol (Du Pont NEN). Solubilized cell pellets and unbound antibiotic were quantified by liquid scintillation counting. The intracellular volumes were calculated using 3H2O (Prokesch & Hand, 1982) and the uptake of the radiolabelled antimicrobial agents expressed as the ratio of the intracellular to extracellular concentration (I/E ratio).
Three different assays were used. (1) Radioassays. The standardization and application of this assay has been described previously (Anderson et al., 1986). This assay was used only for studies with S. aureus. L. monocytogenes and L. micdadei grow too slowly for this assay. Neutrophils (5x 10*) were coincubated with S. aureus at a ratio of 1 neutrophil: 20 bacterial colony forming units (cfu) for 30 min at 37CC in a final volume of 1 mL HBSS. The neutrophil: bacteria mixtures were rotated on a turntable throughout the 30 min incubation period and the neutrophils and intracellular S. aureus separated from extracellular bacteria by centrifugation at 200 g for 5 min. The neutrophils were washed once and resuspended in antibiotic-free HBSS (control systems) or in medium containing various concentrations (0-01, 0-05, 0-5 and 2 mg/L) of erythromycin, dirithromycin or erythromycylamine. The various systems were then incubated for 4 h at 37°C after which the neutrophils were washed three times to remove the antimicrobial agents. Intraphagocytic bacteria werereleasedby osmotic lysis of neutrophils (1 mL distilled H2O) and 100 /zL of the lysate inoculated into 10 mL of nutrient broth containing 5 /iCi of a radiolabelled amino acid mixture (L-amino acid mixture, L-I4C(U), specific activity 55 mCi/matom carbon). The inoculated bottles were incubated at 37°C for 4 h after which the bacteria were washed three times and the extent of incorporation of radiolabelled amino acids into bacterial cells measured in a liquid scintillation spectrophotometer. The results are expressed as uptake of radiolabelled amino acids in radioactive counts per minute (cpm)- Neutrophils were omitted from identically processed control systems containing bacteria and antibiotics only. In a single experiment designed to investigate the interactions which may occur between the oxygen-dependent antimicrobial systems of the neutrophil and erythromycin, dirithromycin or erythromycylamine in the eradication of intraphagocytic S. aureus, normal neutrophils were substituted with CGD neutrophils and exposed to a fixed 2 mg/L concentration of the antimicrobial agents. (2) Colony-counting method. Opsonized bacteria were co-incubated with neutrophils as described above (cell: bacteria ratio 1:20) and processed exactly as for the radioassay except that serial ten-fold dilutions were made of distilled water lysates of neutrophils and 100 /JL of each dilution were transferred to blood agar plates (S. aureus and L. monocytogenes) or charcoal-yeast extract plates (L. micdadei). The plates were incubated for 18 h (S. aureus), 24 h (L. monocytogenes) or four days (L. micdadei) at 37°C and the bacterial colonies enumerated (Quie et al., 1987). The antimicrobial agents were used at concentrations of 0-5 and 2 mg/L. (3) Fluorescence microassay. This method was used to measure the intracellular
Downloaded from http://jac.oxfordjournals.org/ at Stanford Medical Center on May 11, 2015
Assays of intraphagocytk bioactivity of the antimicrobial agents
512
G. K. Joonfc rt aL
Neutrophil viability This was determined after the 4 h incubation period of neutrophils and intracellular bacteria with and without antibiotics using a standard spectrophotometric lactate dehydrogenase (LDH) release assay. Measurement of superoxide generation by neutrophils This was measured using lucigcnin (bis-A^-methylacridinium-nitrate) enhanced chemiluminescence (LECL), a particularly sensitive assay for superoxide generation (Minkenberg & Ferber, 1984). Neutrophils (10*) were preincubated for 45 min at room temperature with the antimicrobial agents (1-20 mg/L) in 900 fiL HBSS containing 0-2 mM lucigenin, followed by incubation for a further period of 10 min at 37°C. This procedure was used to prevent the time-dependent decrease in the numbers of receptors for the synthetic leucotactic tripeptide N-formyl-L-methionyl-L-leucyl-L-phenylalanine (FMLP) and deterioration of FMLP-activated superoxide generating capacity of neutrophils preincubated for periods in excess of 30 min at 37°C. Spontaneous and stimulus-activated LECL was then recorded in an LKB Wallac 1251 luminometer (Turku, Finland) after the addition of 100 ^L of the stimulus of membrane-associated oxidative metabolism. Four different stimuli were used at predetermined optimal concentrations: FMLP (1 /m), phorbol 12-myristate 13-acetate (PMA, 10/zg/L), opsonized zymosan (0-5 mg/L) and the calcium ionophore (A23187, 1 /XM). LECL readings were integrated for 5 sec intervals and recorded as millivolts x seconds (mV.sec). To control for possible superoxide scavenging activity, the effects of dirithromycin, erythromycin and erythromycylamine (20 mg/L) on superoxide generation (LECL) by a neutrophil-free hypoxanthine (0-25 mM)-xanthine oxidase (100 milliunits) system were investigated. Myeloperoxidase (MPO)-mediated iodination MPO-mediated auto-iodination of FMLP-activated neutrophils was investigated according to the method of Root & Stossel (1974). Briefly, neutrophils were preincubated with the test antimicrobial agents (1-20 mg/L) as described above in 900 fiL HBSS containing 0-6 /iCi Na125I and 40 nmoles of carrier Nal. The cells were then
Downloaded from http://jac.oxfordjournals.org/ at Stanford Medical Center on May 11, 2015
biological activity of the test antibiotics for L. monocytogenes and L. micdadei. The procedure was the same as for the radioassay up to the end of the 4 h incubation period of neutrophils and intracellular bacteria with the test antibiotics. Thereafter the viability of intracellular bacteria was determined according to a modification (Anderson et al., 1988) of a previously-described method based on the differential fluorescence of dead and viable intracellular bacteria detected by acridine orange (Pruzanski, Saito & Nitzan, 1983; Smith & Rommel, 1987). After incubation of neutrophils and intracellular bacteria with the test antimicrobial agents the cells were washed three times, resuspended in HBSS and the fluorescence of residual extracellular and membrane-adherent bacteria quenched by exposure of neutrophils to crystal violet for 1 min. Neutrophils were examined, using fluorescence microscopy, for dead and viable intracellular bacteria.
Intraptugocytk actirtty of dMthromydn
513
activated with FLMP (1 /a*). After 15 min of incubation at 37°C the reactions were terminated by the addition of 10% trichloroacetic (TCA), and after three washes with TCA the amount of cell-associated I23I was determined in a solid state gamma counter. These results are expressed as nmoles of I23I per 106 neutrophils. To control for possible interactions of the test antibiotics with the assay system, 20 mg/L of each agent were included in a cell-free oxidant-generating system of glucose (5 DIM), glucose oxidase (200 munits), horseradish peroxidase (200 munits), I25I (0-6 fid), carrier iodide (40 nmoles) and bovine serum albumin (1 mg) in a final volume of 1 mL HBSS. After 15 min incubation at 37°C the extent of peroxidase-mediated iodination was measured as described above.
Neutrophils (5 x 10*/mL) were coincubated with 20 mg/L dirithromycin, erythromycin or erythromycylamine in 900 pL HBSS before the addition of 100 nL FMLP and cytochalasin B at final concentrations of 1 /ZM and 1 g/L respectively, then incubated at 37°C for 15 min. Neutrophils were removed by centrifugation and the supematants assayed for MPO using a previously described colorimetric method (Paul, Selvaraj & Sbarra, 1987). The results are expressed as the percentage of total enzyme released from Triton-XlOO (0-1% final concentration)-treated neutrophils. Neutrophil migration The effects of the macrolides (1, 2-5, 5 and 10 mg/L) were measured using modified Boyden chambers with 5-/nn pore size nitrocellulose filters (Sartorius Membranfilter, Gottingen, Germany) using the leucoattractants FMLP (20 nM) and 10% endotoxinactivated serum (EAS) as previously described (Anderson, 1989). Leucoattractants were omitted from random migration systems. The antimicrobial agents were present in both the upper and lower compartments of the chamber. After incubation for 40-60 min at 37CC the filters were removed, processed and counted. The results are expressed as the numbers of neutrophils which had traversed the filter per microscope high-powered field (cells/HPF).
Expression and statistical analysis of results The results of each series of experiments are expressed as the mean ± standard error of the mean (S.E.M.). Statistical analyses were performed using Student's /-test (paired /-statistic) by comparison of the results for control and corresponding antibiotictreated neutrophils.
Results Antibiotic sensitivity Erythromycin was the most active agent for the three strains, followed by dirithromycin and erythromycylamine (Table I).
Downloaded from http://jac.oxfordjournals.org/ at Stanford Medical Center on May 11, 2015
Neutrophil degranulation
514
G. K. Joont et al
Table I. In-vitro susceptibility of 5. aureus, L. monocytogenes and L. micdadei to erythromycin, dirithromycin and erythromycylainine Erythromycin MIC MBC (mg/L) (mg/L)
Erythromycylamine
Dirithromycin MIC MBC (mg/L) (mg/L)
MIC (mg/L)
MBC (mg/L)
S. aureus
0-5
0-5
1-0
20
1-0
1-0
L. monocytogenes L. micdadei
0-25 0-2
1-0 —
2-0 0-2
20 —
10 0-5
20 —
Both agents were actively concentrated by neutrophils and the I/E ratios were 6-1 ±2-5 and 106 ±2-0 for erythromycin and dirithromycin respectively. This difference is statistically significant (P < 0-005). Measurement of the intracellular bioactivity of the antimicrobial agents All three antimicrobial agents caused dose-related inhibition of the proliferation of intraphagocytic S. aureus (Figure 1). The lowest concentrations of erythromycin, dirithromycin and erythromycylamine at which these achieved statistical significance (P < 0-05) were 0-01,0-5 and 0-5 mg/L respectively. However, none of the antimicrobial agents completely eradicated the bacteria. The data for CGD neutrophils clearly demonstrate the requirement for intact neutrophil oxygen-dependent antimicrobial systems to achieve optimum co-operation between these cells and the test antibiotics in
6 T
h
^-i
9
M
E
trial grow
Ji 4
1
o
1
NX,
2
1
0
001
0-05
0-5
2
Antibiotic concentration (mg/L)
Figure I. Measurement or the intracellular bioactivity or erythromycylainine ( • ) . dirithromycin (#), and erythromycin (V).
Downloaded from http://jac.oxfordjournals.org/ at Stanford Medical Center on May 11, 2015
Uptake of radiolabelled erythromycin and dirithromycin by neutrophils
Intraphagocytic actirtty of dMtbxomydn
515
Table IL Intraphagocytic bioactivitics of erythromycin, dirithromycin and erythromycylaminc (2 mg/L) for 5. aureus using normal and CGD neutrophils
Antimicrobial agent None (control) Erythromycin Dirithromycin Erythromycylamine
Uptake of radiolabcllcd amino acids by S. aureus ingested by: normal neutrophils CGD neutrophils 6574± 9001 2651 ±266-19 3314 ± 45-40 3212± 17104
8721 ± 6820± 8807 ± 7787±
91-62 170-08 87-89 8713
the eradication of intracellular S. aureus (Table II). In neutrophil-free control systems a 4 h exposure of the bacteria to 2 mg/L of the antibiotics in HBSS followed by washing was associated with minimum loss of viability. The values for the antibioticfree controls and for bacteria treated with 2 mg/L of erythromycin, dirithromycin, or erythromycylamine were 14144±168, 12644±140, 9472±100 and 13720± 1692 cpm respectively. Results obtained with the colony counting method for S. aureus, L. monocytogenes and L. micdadei are shown in Table III. At the concentrations tested (05 and 2 mg/L) none of the test agents affected the intraphagocytic killing of S. aureus. Dirithromycin at 2 mg/L significantly (P < 0-005) enhanced the intraphagocytic killing of L. monocytogenes. Erythromycin at both concentrations tested (0-5 and 2 mg/L) also increased the intraphagocytic killing of L. monocytogenes {P < 0025 for 2 mg/L). The radioassay was unsuitable for the measurement of growth of surviving L. monocytogenes or L. micdadei therefore the fluorescence microassay was used for these investigations. With this assay both erythromycin and dirithromycin (2 mg/L) increased the intraphagocytic killing of L. monocytogenes (P < 0-05 for both agents; Table IIL Intracellular biological activity of the test macrolides for S. aureus, L. monocytogenes and L. micdadei using the colony counting method Antimicrobial agent and concentration Control (no antibiotic) Erythromycin 0-5 mg/L 2 mg/L Dirithromycin 0-5 mg/L 2 mg/L Erythromycylamine 0-5 mg/L 2 mg/L
Viable intracellular bacteria (cfuxlO6) 5. aureus
L. monocytogenes
L. micdadei
33±6* 35±6 32±10 42±11 27±3 32±10 37±7
24±4 18±4 19±4' 23±7 17±4* 23±4 21±5
31 ± 15
Results are expressed as the mean numbers ±S£.u. different experiments. ND. not done.
•P < O05.
±S_E.M.
of three different
Table TV). None of the test agents increased the intraphagocytic killing of L. micdadei, although erythromycin and dirithromycin (2 mg/L) interfered with the intracellular proliferation of this microorganism, indicating intraphagocytic bacteristatic activity (Table V). Neutrophil viability By comparison with the corresponding drug-free control systems, none of the test agents at concentrations of up to 2 mg/L affected the viability of neutrophils. After four hours of incubation with intraphagocytic S. aureus and antibiotics the mean percentage release of LDH from control neutrophils and those incubated with erythromycin, dirithromycin or erythromycylamine were 11-5 ±1-2, 7-6 ±2-2, 7-4 ± 0 4 and 8-7 ±0-8, respectively. Superoxide generation The effects of the three test antimicrobial agents on superoxide generation by FMLPand calcium ionophore-activated neutrophils are shown in Figures 2 and 3 respectively. Table V. Intracellular biological activities of erythromycin, dirithromycin and erythromycylamine against L. micdadei using the fluorescence microassay Antimicrobial agent and concentration Control (no antibiotic) Erythromycin 2 mg/L Dirithromycin 2 mg/L Erythromycylamine 2 mg/L
Average number of bacteria per neutrophil 5-8 ±0-2* 4-8 ±0-5* 51 ±0-4* 5-3 ± 04
'Results are expressed as the mean value ±S.E.M. of four different experiments. V>
Investigation of the in-vitro uptake, intraphagocytic biological activity and effects on nentrophil superoxide generation of dirithromycin compared with erythromycin G. K. Joone, C. E. J. Van Rensburg and R. Anderson
The cellular uptake by human neutrophils and the intraphagocytic biological activity of the new macrolide antimicrobial agent dirithromycin (0-01-2 mg/L) compared with erythromycin was investigated in vitro. Staphylococcus aureus, LJsteria monocytogenu and Legionella pneumophila were used as the test intraccllular microbial pathogens. After coincubation (45 min at 37°Q of neutrophils with a fixed concentration of 2 mg/L of each antibiotic the respective intraccllular/ extracellular ratios for erythromycin and dirithromycin were 6-1 ±2-5 and 10-6±2 respectively (P < 0-005). Using a combination of techniques (colony counting, radiometry andfluorescencemicroscopy) both erythromycin and dirithromycin at concentrations of 0-01 and 0-5 mg/L and higher, respectively, were found to possess dose-related intraphagocytic bacteristatic activity for each of the test microbial pathogens. The effects of dirithromycin and erythromycin (1-20 mg/L) on neutrophil chemotaxis and generation of reactive oxidants by these cells were also investigated in vitro. Both antimicrobial agents caused a dose-related stimulation of neutrophil migration which was associated with inhibition of leucoattractant-activated generation of superoxide and activity of the myeloperoxidase/HjOj/halide system. However, superoxide generation by neutrophils activited with opsonized zymosan or phorbol myristate acetate was unaffected by the macrolides. These findings demonstrate that dirithromycin accumulates in human neutrophils, is biologically active intracellularly and modulates leucoattractant-activated superoxide generation and chemotaxis. Introduction Dirithromycin (LY 237216), an oxazime derivative of eTythromycin, is a new macrolide antimicrobial agent with improved pharmacokinetics (Kirst & Sides, 1989). Dirithromycin is metabolized in humans to erythromycylamine which has an extremely long half-life ( > 24 h), allowing once daily administration (Kirst & Sides, 1989). The activity of dirithromycin is similar to that of erythromycin against streptococci, pneumococci and Bordetella pertussis, but it is less active against Haemophilus influenzae and Campylobacter pylori (Garcia-Rodriguez et al., 1989; Hoppe & Eichhorn, 1989; Yu & Neu, 1990). Macrolides, which are actively concentrated by human phagocytes (Prokesch & Hand, 1982; Anderson et al., 1987; Hand, King-Thompson & Holman, 1987; Anderson, Joone & Van Rensburg, 1988; Ishiguro et al., 1989), appear to interact cooperatively with these cells to eliminate intraphagocytic microbial pathogens (Miller et 509 0305-7453/92/100509+ 15 $08.00/0
© 1992 The Britiih Society for Antimicrobial Chemotherapy
Downloaded from http://jac.oxfordjournals.org/ at Stanford Medical Center on May 11, 2015
Medical Research Council Unit For The Study of Phagocyte Function, Department of Immunology, Institute For Pathology, University of Pretoria, Pretoria 0001, Republic of South Africa
510
G. K. Jooofe et aL
al., 1984; Anderson, Joone & Van Rcnsburg, 1986; Anderson et al., 1988; Ishiguro et al., 1989). These co-operative interactions between macrolides and phagocytes have been described for erythromycin (Anderson et al., 1986), josamycin (Labro & Babin-Chevaye, 1989), roxithromycin (Labro et al., 1986; Anderson et al., 1987), clarithromycin (Anderson et al., 1988; Ishiguro et al, 1989) and azithromycin (Gladue et al., 1989; Wildfeuer et al., 1989), but not for dirithromycin. In the present study the uptake of dirithromycin and erythromycylamine compared with erythromycin by human neutrophils, and their intraphagocytic biological activities and effects on neutrophil locomotion and generation of reactive oxidants were investigated.
Antimicrobial agents Erythromycin base, dirithromycin, erythromycylamine and radiolabelled dirithromycin (3H-dirithromycin, 16-7 Ci/mmol) were kindly provided by the Lilly Research Laboratories, Indianapolis, Indiana, USA. Radiolabelled erythromycin (N-methylI4 C- erythromycin, specific activity 55 mCi/mmol) was purchased from Du Pont NEN Research Products, Boston, USA. Each of the test compounds (4 mg) was dissolved in 1 mL of 30% aqueous ethanol and diluted in distilled water to give stock solutions of 200 mg/L. The antibiotics were used in assays of neutrophil functions at final concentrations of 0-01-20 mg/L. Solvent controls were included. Chemicals and reagents Unless indicated all other chemicals and reagents were obtained from the Sigma Chemical Co., St Louis, USA and radiochemicals from Du Pont NEN. Bacteria Staphylococcus aureus (ATCC 25923), Usteria monocytogenes (laboratory isolate) and Legionella micdadei (ATCC 33218) were used as the test microbial pathogens. Antibiotic sensitivity testing in vitro The MICs and MBCs of the test antibiotics for the three microbial pathogens were determined by standard bacteriological assays (Reiner, 1982). Neutrophils These cells were prepared as previously described (Anderson et al., 1986) from venous blood taken from adult human volunteers and a child with the autosomal recessive form of chronic granulomatous disease (CGD). The neutrophils were resuspended to a concentration of 1 x 107/mL in indicator-free Hanks' balanced salt solution (HBSS) buffered with 4-2 mM HEPES, pH 7-4. Measurements of the uptake of radiolabelled antibiotics by neutrophils This was measured according to the method previously described by Prokesch & Hand (1982). Briefly 100 //L of HBSS containing 2x10* neutrophils and 2 mg/L of radio-
Downloaded from http://jac.oxfordjournals.org/ at Stanford Medical Center on May 11, 2015
Materials and methods
Intraphagocytk acthrtty of dMthromydn
511
labelled dirithromycin or erythromycin were incubated over 150 /iL of silicone oil in microcentrifuge tubes at 37°C for 45 min (a predetermined optimum incubation time). Unbound and neutrophil-associated antibiotics were separated by centrifuging the neutrophils through the oil at 12,000 g for 3 min. After freezing the tubes at - 70°C the neutrophil pellets were sliced from the bottom of each tube and solubilized in Protosol (Du Pont NEN). Solubilized cell pellets and unbound antibiotic were quantified by liquid scintillation counting. The intracellular volumes were calculated using 3H2O (Prokesch & Hand, 1982) and the uptake of the radiolabelled antimicrobial agents expressed as the ratio of the intracellular to extracellular concentration (I/E ratio).
Three different assays were used. (1) Radioassays. The standardization and application of this assay has been described previously (Anderson et al., 1986). This assay was used only for studies with S. aureus. L. monocytogenes and L. micdadei grow too slowly for this assay. Neutrophils (5x 10*) were coincubated with S. aureus at a ratio of 1 neutrophil: 20 bacterial colony forming units (cfu) for 30 min at 37CC in a final volume of 1 mL HBSS. The neutrophil: bacteria mixtures were rotated on a turntable throughout the 30 min incubation period and the neutrophils and intracellular S. aureus separated from extracellular bacteria by centrifugation at 200 g for 5 min. The neutrophils were washed once and resuspended in antibiotic-free HBSS (control systems) or in medium containing various concentrations (0-01, 0-05, 0-5 and 2 mg/L) of erythromycin, dirithromycin or erythromycylamine. The various systems were then incubated for 4 h at 37°C after which the neutrophils were washed three times to remove the antimicrobial agents. Intraphagocytic bacteria werereleasedby osmotic lysis of neutrophils (1 mL distilled H2O) and 100 /zL of the lysate inoculated into 10 mL of nutrient broth containing 5 /iCi of a radiolabelled amino acid mixture (L-amino acid mixture, L-I4C(U), specific activity 55 mCi/matom carbon). The inoculated bottles were incubated at 37°C for 4 h after which the bacteria were washed three times and the extent of incorporation of radiolabelled amino acids into bacterial cells measured in a liquid scintillation spectrophotometer. The results are expressed as uptake of radiolabelled amino acids in radioactive counts per minute (cpm)- Neutrophils were omitted from identically processed control systems containing bacteria and antibiotics only. In a single experiment designed to investigate the interactions which may occur between the oxygen-dependent antimicrobial systems of the neutrophil and erythromycin, dirithromycin or erythromycylamine in the eradication of intraphagocytic S. aureus, normal neutrophils were substituted with CGD neutrophils and exposed to a fixed 2 mg/L concentration of the antimicrobial agents. (2) Colony-counting method. Opsonized bacteria were co-incubated with neutrophils as described above (cell: bacteria ratio 1:20) and processed exactly as for the radioassay except that serial ten-fold dilutions were made of distilled water lysates of neutrophils and 100 /JL of each dilution were transferred to blood agar plates (S. aureus and L. monocytogenes) or charcoal-yeast extract plates (L. micdadei). The plates were incubated for 18 h (S. aureus), 24 h (L. monocytogenes) or four days (L. micdadei) at 37°C and the bacterial colonies enumerated (Quie et al., 1987). The antimicrobial agents were used at concentrations of 0-5 and 2 mg/L. (3) Fluorescence microassay. This method was used to measure the intracellular
Downloaded from http://jac.oxfordjournals.org/ at Stanford Medical Center on May 11, 2015
Assays of intraphagocytk bioactivity of the antimicrobial agents
512
G. K. Joonfc rt aL
Neutrophil viability This was determined after the 4 h incubation period of neutrophils and intracellular bacteria with and without antibiotics using a standard spectrophotometric lactate dehydrogenase (LDH) release assay. Measurement of superoxide generation by neutrophils This was measured using lucigcnin (bis-A^-methylacridinium-nitrate) enhanced chemiluminescence (LECL), a particularly sensitive assay for superoxide generation (Minkenberg & Ferber, 1984). Neutrophils (10*) were preincubated for 45 min at room temperature with the antimicrobial agents (1-20 mg/L) in 900 fiL HBSS containing 0-2 mM lucigenin, followed by incubation for a further period of 10 min at 37°C. This procedure was used to prevent the time-dependent decrease in the numbers of receptors for the synthetic leucotactic tripeptide N-formyl-L-methionyl-L-leucyl-L-phenylalanine (FMLP) and deterioration of FMLP-activated superoxide generating capacity of neutrophils preincubated for periods in excess of 30 min at 37°C. Spontaneous and stimulus-activated LECL was then recorded in an LKB Wallac 1251 luminometer (Turku, Finland) after the addition of 100 ^L of the stimulus of membrane-associated oxidative metabolism. Four different stimuli were used at predetermined optimal concentrations: FMLP (1 /m), phorbol 12-myristate 13-acetate (PMA, 10/zg/L), opsonized zymosan (0-5 mg/L) and the calcium ionophore (A23187, 1 /XM). LECL readings were integrated for 5 sec intervals and recorded as millivolts x seconds (mV.sec). To control for possible superoxide scavenging activity, the effects of dirithromycin, erythromycin and erythromycylamine (20 mg/L) on superoxide generation (LECL) by a neutrophil-free hypoxanthine (0-25 mM)-xanthine oxidase (100 milliunits) system were investigated. Myeloperoxidase (MPO)-mediated iodination MPO-mediated auto-iodination of FMLP-activated neutrophils was investigated according to the method of Root & Stossel (1974). Briefly, neutrophils were preincubated with the test antimicrobial agents (1-20 mg/L) as described above in 900 fiL HBSS containing 0-6 /iCi Na125I and 40 nmoles of carrier Nal. The cells were then
Downloaded from http://jac.oxfordjournals.org/ at Stanford Medical Center on May 11, 2015
biological activity of the test antibiotics for L. monocytogenes and L. micdadei. The procedure was the same as for the radioassay up to the end of the 4 h incubation period of neutrophils and intracellular bacteria with the test antibiotics. Thereafter the viability of intracellular bacteria was determined according to a modification (Anderson et al., 1988) of a previously-described method based on the differential fluorescence of dead and viable intracellular bacteria detected by acridine orange (Pruzanski, Saito & Nitzan, 1983; Smith & Rommel, 1987). After incubation of neutrophils and intracellular bacteria with the test antimicrobial agents the cells were washed three times, resuspended in HBSS and the fluorescence of residual extracellular and membrane-adherent bacteria quenched by exposure of neutrophils to crystal violet for 1 min. Neutrophils were examined, using fluorescence microscopy, for dead and viable intracellular bacteria.
Intraptugocytk actirtty of dMthromydn
513
activated with FLMP (1 /a*). After 15 min of incubation at 37°C the reactions were terminated by the addition of 10% trichloroacetic (TCA), and after three washes with TCA the amount of cell-associated I23I was determined in a solid state gamma counter. These results are expressed as nmoles of I23I per 106 neutrophils. To control for possible interactions of the test antibiotics with the assay system, 20 mg/L of each agent were included in a cell-free oxidant-generating system of glucose (5 DIM), glucose oxidase (200 munits), horseradish peroxidase (200 munits), I25I (0-6 fid), carrier iodide (40 nmoles) and bovine serum albumin (1 mg) in a final volume of 1 mL HBSS. After 15 min incubation at 37°C the extent of peroxidase-mediated iodination was measured as described above.
Neutrophils (5 x 10*/mL) were coincubated with 20 mg/L dirithromycin, erythromycin or erythromycylamine in 900 pL HBSS before the addition of 100 nL FMLP and cytochalasin B at final concentrations of 1 /ZM and 1 g/L respectively, then incubated at 37°C for 15 min. Neutrophils were removed by centrifugation and the supematants assayed for MPO using a previously described colorimetric method (Paul, Selvaraj & Sbarra, 1987). The results are expressed as the percentage of total enzyme released from Triton-XlOO (0-1% final concentration)-treated neutrophils. Neutrophil migration The effects of the macrolides (1, 2-5, 5 and 10 mg/L) were measured using modified Boyden chambers with 5-/nn pore size nitrocellulose filters (Sartorius Membranfilter, Gottingen, Germany) using the leucoattractants FMLP (20 nM) and 10% endotoxinactivated serum (EAS) as previously described (Anderson, 1989). Leucoattractants were omitted from random migration systems. The antimicrobial agents were present in both the upper and lower compartments of the chamber. After incubation for 40-60 min at 37CC the filters were removed, processed and counted. The results are expressed as the numbers of neutrophils which had traversed the filter per microscope high-powered field (cells/HPF).
Expression and statistical analysis of results The results of each series of experiments are expressed as the mean ± standard error of the mean (S.E.M.). Statistical analyses were performed using Student's /-test (paired /-statistic) by comparison of the results for control and corresponding antibiotictreated neutrophils.
Results Antibiotic sensitivity Erythromycin was the most active agent for the three strains, followed by dirithromycin and erythromycylamine (Table I).
Downloaded from http://jac.oxfordjournals.org/ at Stanford Medical Center on May 11, 2015
Neutrophil degranulation
514
G. K. Joont et al
Table I. In-vitro susceptibility of 5. aureus, L. monocytogenes and L. micdadei to erythromycin, dirithromycin and erythromycylainine Erythromycin MIC MBC (mg/L) (mg/L)
Erythromycylamine
Dirithromycin MIC MBC (mg/L) (mg/L)
MIC (mg/L)
MBC (mg/L)
S. aureus
0-5
0-5
1-0
20
1-0
1-0
L. monocytogenes L. micdadei
0-25 0-2
1-0 —
2-0 0-2
20 —
10 0-5
20 —
Both agents were actively concentrated by neutrophils and the I/E ratios were 6-1 ±2-5 and 106 ±2-0 for erythromycin and dirithromycin respectively. This difference is statistically significant (P < 0-005). Measurement of the intracellular bioactivity of the antimicrobial agents All three antimicrobial agents caused dose-related inhibition of the proliferation of intraphagocytic S. aureus (Figure 1). The lowest concentrations of erythromycin, dirithromycin and erythromycylamine at which these achieved statistical significance (P < 0-05) were 0-01,0-5 and 0-5 mg/L respectively. However, none of the antimicrobial agents completely eradicated the bacteria. The data for CGD neutrophils clearly demonstrate the requirement for intact neutrophil oxygen-dependent antimicrobial systems to achieve optimum co-operation between these cells and the test antibiotics in
6 T
h
^-i
9
M
E
trial grow
Ji 4
1
o
1
NX,
2
1
0
001
0-05
0-5
2
Antibiotic concentration (mg/L)
Figure I. Measurement or the intracellular bioactivity or erythromycylainine ( • ) . dirithromycin (#), and erythromycin (V).
Downloaded from http://jac.oxfordjournals.org/ at Stanford Medical Center on May 11, 2015
Uptake of radiolabelled erythromycin and dirithromycin by neutrophils
Intraphagocytic actirtty of dMtbxomydn
515
Table IL Intraphagocytic bioactivitics of erythromycin, dirithromycin and erythromycylaminc (2 mg/L) for 5. aureus using normal and CGD neutrophils
Antimicrobial agent None (control) Erythromycin Dirithromycin Erythromycylamine
Uptake of radiolabcllcd amino acids by S. aureus ingested by: normal neutrophils CGD neutrophils 6574± 9001 2651 ±266-19 3314 ± 45-40 3212± 17104
8721 ± 6820± 8807 ± 7787±
91-62 170-08 87-89 8713
the eradication of intracellular S. aureus (Table II). In neutrophil-free control systems a 4 h exposure of the bacteria to 2 mg/L of the antibiotics in HBSS followed by washing was associated with minimum loss of viability. The values for the antibioticfree controls and for bacteria treated with 2 mg/L of erythromycin, dirithromycin, or erythromycylamine were 14144±168, 12644±140, 9472±100 and 13720± 1692 cpm respectively. Results obtained with the colony counting method for S. aureus, L. monocytogenes and L. micdadei are shown in Table III. At the concentrations tested (05 and 2 mg/L) none of the test agents affected the intraphagocytic killing of S. aureus. Dirithromycin at 2 mg/L significantly (P < 0-005) enhanced the intraphagocytic killing of L. monocytogenes. Erythromycin at both concentrations tested (0-5 and 2 mg/L) also increased the intraphagocytic killing of L. monocytogenes {P < 0025 for 2 mg/L). The radioassay was unsuitable for the measurement of growth of surviving L. monocytogenes or L. micdadei therefore the fluorescence microassay was used for these investigations. With this assay both erythromycin and dirithromycin (2 mg/L) increased the intraphagocytic killing of L. monocytogenes (P < 0-05 for both agents; Table IIL Intracellular biological activity of the test macrolides for S. aureus, L. monocytogenes and L. micdadei using the colony counting method Antimicrobial agent and concentration Control (no antibiotic) Erythromycin 0-5 mg/L 2 mg/L Dirithromycin 0-5 mg/L 2 mg/L Erythromycylamine 0-5 mg/L 2 mg/L
Viable intracellular bacteria (cfuxlO6) 5. aureus
L. monocytogenes
L. micdadei
33±6* 35±6 32±10 42±11 27±3 32±10 37±7
24±4 18±4 19±4' 23±7 17±4* 23±4 21±5
31 ± 15
Results are expressed as the mean numbers ±S£.u. different experiments. ND. not done.
•P < O05.
±S_E.M.
of three different
Table TV). None of the test agents increased the intraphagocytic killing of L. micdadei, although erythromycin and dirithromycin (2 mg/L) interfered with the intracellular proliferation of this microorganism, indicating intraphagocytic bacteristatic activity (Table V). Neutrophil viability By comparison with the corresponding drug-free control systems, none of the test agents at concentrations of up to 2 mg/L affected the viability of neutrophils. After four hours of incubation with intraphagocytic S. aureus and antibiotics the mean percentage release of LDH from control neutrophils and those incubated with erythromycin, dirithromycin or erythromycylamine were 11-5 ±1-2, 7-6 ±2-2, 7-4 ± 0 4 and 8-7 ±0-8, respectively. Superoxide generation The effects of the three test antimicrobial agents on superoxide generation by FMLPand calcium ionophore-activated neutrophils are shown in Figures 2 and 3 respectively. Table V. Intracellular biological activities of erythromycin, dirithromycin and erythromycylamine against L. micdadei using the fluorescence microassay Antimicrobial agent and concentration Control (no antibiotic) Erythromycin 2 mg/L Dirithromycin 2 mg/L Erythromycylamine 2 mg/L
Average number of bacteria per neutrophil 5-8 ±0-2* 4-8 ±0-5* 51 ±0-4* 5-3 ± 04
'Results are expressed as the mean value ±S.E.M. of four different experiments. V>
