In vitro and in vivo uptake of azithromycin (CP-62,993) by phagocytic cells: possible mechanism of delivery and release at sites of infection.

Azithromycin, a novel azalide antibiotic, concentrated in human and mouse polymorphonuclear leukocytes (PMNs), murine peritoneal macrophages, and mouse and rat alveolar macrophages, attaining intracellular concentrations up to 226 times the external concentration in vitro. In murine peritoneal macrophages, azithromycin achieved concentration gradients (internal to external) up to 26 times higher than erythromycin. The cellular uptake of azithromycin was dependent on temperature, viability, and pH and was decreased by 2,4-dinitrophenol. Azithromycin did not decrease phagocyte-mediated bactericidal activity or affect PMN or macrophage oxidative burst activity (H2O2 release or Nitro Blue Tetrazolium reduction, respectively). Azithromycin remained in cells for several hours, even after extracellular drug was removed. However, its release was significantly enhanced by phagocytosis of Staphylococcus aureus (82 versus 23% by 1.5 h). In vivo, 0.05 micrograms of azithromycin was found in peritoneal fluids of mice 20 h after oral treatment with a dose of 50 mg/kg. Following caseinate-induced PMN infiltration, there was a sixfold increase in peritoneal cavity azithromycin to 0.32 micrograms, most of which was intracellular. Therefore, the uptake, transport, and later release of azithromycin by these cells demonstrate that phagocytes may deliver active drug to sites of infection.     

Stored red blood cells selectively activate human neutrophils to release IL-8 and secretory PLA2

Packed red blood cell (PRBC) transfusion has been invoked previously with immunosuppression and increased infections, but it has now been demonstrated that stored PRBCs (>14 days) can prime PMNs and provoke multiple organ failure. Recently, the role of PMNs in the genesis of MOF has been extended to their release of inflammatory cytokines, notably IL-1, IL-8, TNFalpha, and secretory phospholipase A2 (sPLA2). We hypothesize that stored PRBCs can act as a second event via stimulating the release of inflammatory cytokines from PMNs. Isolated human PMNs were incubated for 24 h in RPMI with either 20% fresh plasma or plasma from 42 day old PRBC (day of outdate) and release of IL-8, IL-1beta, TNFalpha, and sPLA2 were measured. Plasma from stored PRBCs contained small amounts of IL-8, sPLA2, and TNFalpha (102.1 +/-5.6 pg/ml, 87.6+/-6.0 pg/ml and 9.7+/-.7 pg/ml). Levels of IL-1beta were below detection (<1 pg/ml). Day 42 PRBC plasma stimulated significant PMN release of both IL-8 and sPLA2 as compared to both control and day 0 plasma (*P < .05), but PRBC plasma did not stimulate PMN release of either IL-1beta or TNFalpha. Transfused blood is emerging as an inflammatory agent that is capable of producing PMN priming. In this study we have demonstrated that PRBC plasma selectively activates PMNs to release both IL-8 and sPLA2. Thus, transfusion of PRBCs may represent a preventable inflammatory insult via modification of both blood banking and transfusion practices.     

Pharmacokinetic and in vivo studies with azithromycin (CP-62,993), a new macrolide with an extended half-life and excellent tissue distribution.

Azithromycin (CP-62,993), a new acid-stable 15-membered-ring macrolide, was well absorbed following oral administration in mice, rats, dogs, and cynomolgus monkeys. This compound exhibited a uniformly long elimination half-life and was distributed exceptionally well into all tissues. This extravascular penetration of azithromycin was demonstrated by tissue/plasma area-under-the-curve ratios ranging from 13.6 to 137 compared with ratios for erythromycin of 3.1 to 11.6. The significance of these pharmacokinetic advantages of azithromycin over erythromycin was shown through efficacy in a series of animal infection models. Azithromycin was orally effective in treating middle ear infections induced in gerbils by transbulla challenges with amoxicillin-resistant Haemophilus influenzae or susceptible Streptococcus pneumoniae; erythromycin failed and cefaclor was only marginally active against the H. influenzae challenge. Azithromycin was equivalent to cefaclor and erythromycin against Streptococcus pneumoniae. In mouse models, the new macrolide was 10-fold more potent than erythromycin and four other antibiotics against an anaerobic infection produced by Fusobacterium necrophorum. Similarly, azithromycin was effective against established tissue infections induced by Salmonella enteritidis (liver and spleen) and Staphylococcus aureus (thigh muscle); erythromycin failed against both infections. The oral and subcutaneous activities of azithromycin, erythromycin, and cefaclor were similar against acute systemic infections produced by Streptococcus pneumoniae, Streptococcus pyogenes, Streptococcus viridans, or S. aureus, whereas azithromycin was more potent than erythromycin and cefaclor against the intracellular pathogen Listeria monocytogenes. The pharmacokinetic advantage of azithromycin over erythromycin in half-life was clearly demonstrated in prophylactic treatment of an acute mouse model of S. aureus infection. These properties of azithromycin strongly support the further evaluation of this new macrolide for use in community-acquired infections of skin or soft tissue and respiratory diseases.     

A functional granulocyte colony-stimulating factor receptor is required for normal chemoattractant-induced neutrophil activation

Granulocyte colony-stimulating factor (G-CSF) is a hematopoietic growth factor that is widely used to treat neutropenia. In addition to stimulating polymorphonuclear neutrophil (PMN) production, G-CSF may have significant effects on PMN function. Because G-CSF receptor (G-CSFR)-deficient mice do not have the expected neutrophilia after administration of human interleukin-8 (IL-8), we examined the effect of the loss of G-CSFR on IL-8-stimulated PMN function. Compared with wild-type PMNs, PMNs isolated from G-CSFR-deficient mice demonstrated markedly decreased chemotaxis to IL-8. PMN emigration into the skin of G-CSFR-deficient mice in response to IL-8 was also impaired. Significant chemotaxis defects were also seen in response to N-formyl-methionyl-leucyl-phenylalanine, zymosan-activated serum, or macrophage inflammatory protein-2. The defective chemotactic response to IL-8 does not appear to be due to impaired chemoattractant receptor function, as the number of IL-8 receptors and chemoattractant-induced calcium influx, actin polymerization, and release of gelatinase B were comparable to those of wild-type PMNs. Chemoattractant-induced adhesion of G-CSFR-deficient PMNs was significantly impaired, suggesting a defect in beta2-integrin activation. Collectively, these data demonstrate that selective defects in PMN activation are present in G-CSFR-deficient mice and indicate that G-CSF plays an important role in regulating PMN chemokine responsiveness.     

Storage of neutrophil granulocytes (PMNs) in additive solution or in autologous plasma for 72 h

Neutropenia may necessitate polymorphonuclear (PMN) transfusion, but among other reasons, PMN short shelf-life complicates realization of innovative transfusion strategies. In 18 donors, PMNs were mobilized using rHuG-CSF + dexamethasone. (8.3 +/- 1.6) x 10(10) PMNs were harvested in 203 +/- 8.7 mL. PMNs were stored undiluted (1, n = 18) and diluted 1-in-2, 1-in-4, 1-in-8 using T-Sol (2, n = 6), T-Sol + 1% HSA (3, n = 6), or autologous plasma (4, n = 6) for 72 h. Haemograms, pH values, phagocytosis, oxidative burst, and interleukin (IL)-1beta, IL-8 and tumour necrosis factor (TNF)-alpha levels were assessed every 24 h. PMN count decreased from (4.3 +/- 0.8) x 10(10) to (2.2 +/- 1.0) x 10(10), and pH value dropped from 6.4 +/- 0.3 to 5.4 +/- 0.2 within 72 h (1), whereas 1-in-4 and 1-in-8 dilutions exhibited consistent haemograms and pH values above 6.0. 1-in-8 dilution (4) stabilized pH at 7.1 +/- 0.4 after 72 h. Function deteriorated to about 50% within 24 h (1), but 1-in-8 (3), 1-in-4 and 1-in-8 diluted PMNs (4) kept it >90% for 72 h. In all collectives, cytokine levels increased during storage. After all, IL-1beta ranged between 31.0 +/- 16.3 (1-in-4, 4) and 100.0 +/- 21.4 (1-in-4, 2), IL-8 from 513 +/- 454 (1) to 3180 +/- 760 (1-in-8, 2), and TNF-alpha between 3.8 +/- 1.7 (1-in-2, 2) and 23.2 +/- 11.8 (1-in-8, 4) (pg mL(-1)). PMN function may be preserved for 72 h in vitro by dilution of PMN apheresates with, preferably, autologous plasma.     

Glycogen metabolism in stored granulocytes

Optimal function of transfused granulocytes (PMNs) requires adequate glycogen metabolism. Previous studies in our laboratory suggested that stored PMNs had decreased glycogen. We report here the glycogen content and chemotaxis of stored PMNs, and the ability of fresh and stored PMNs to use glycogen as the fuel source for chemotaxis. PMNs were prepared from 8 fresh units of blood drawn into citrate-phosphate-dextrose-adenine, suspended at 2 or 8 X 10(7) PMN per ml in autologous plasma with or without 15 mM sodium bicarbonate, and stored at 22 to 24 degrees C in transfer packs for 48 hours. Glycogen was measured on resting PMNs, and after challenge with opsonized zymosan and F-Met-Leu-Phe (FMLP). The chemotaxis of fresh and stored PMNs was measured in the presence or absence of extracellular glucose. Fresh PMNs contained 10.3 +/- 0.5 (mean +/- SEM) micrograms of glycogen per 10(6) PMN. Glycogen decreased by 4.2 +/- 0.9 micrograms per 10(6) PMN after challenge with opsonized zymosan and by 1.1 +/- 0.6 micrograms per 10(6) PMN after FMLP. After 48 hours of storage, neutrophil glycogen increased by 18 percent, except in units stored at a concentration of PMN of 8 X 10(7) per ml without sodium bicarbonate. In PMNs from these units stored without bicarbonate, glycogen decreased by 9 percent (p less than .05), and there was a 19 and 55 percent decrease in the ability of PMN from these units to metabolize glycogen after exposure to opsonized zymosan and FMLP, respectively (p less than 0.05). In addition, in PMNs from units stored at a concentration of PMN of 8 X 10(7) per ml without bicarbonate, there was a 47 and 70 percent decrease in chemotaxis at 24 and 48 hours, respectively (p less than 0.05).(ABSTRACT TRUNCATED AT 250 WORDS)     

Differences in neutrophil death among beta-lactam antibiotics after in vitro killing of bacteria

Antibiotic therapy is an essential treatment for gram-negative bacterial infections. Antibiotic-induced endotoxin release and subsequent production of inflammatory cytokines reportedly depend on the type of antibiotic action. This study examined the effects of various beta-lactam antibiotics on cell death of human polymorphonuclear neutrophils (PMNs) cocultured with Escherichia coli (E. coli) in vitro. E. coli morphology after antibiotic treatment was determined. PMNs and E. coli were cocultured with antibiotics for 0, 4, or 12 h. Levels of endotoxin and cytokines (TNF-alpha, IL-1beta, and IL-6) in the supernatants were measured. The filtrates of antibiotic-treated E. coli supernatants were cocultured with PMNs for 0, 4, or 12 h. In all experiments, ampicillin (ABPC), cefazolin sodium (CEZ), cefoperazone sodium (CPZ), latamoxef sodium (LMOX), imipenem (IPM), and polymyxin B sulfate (PLB) were used at 30 microg/mL. PMNs were isolated from healthy volunteers. PMN cell death was assessed by flow cytometry and light microscopy. ABPC, CEZ, CPZ, and LMOX, which induce bacterial filament formation with lysis, caused PMN necrosis when cocultured with E. coli. In contrast, IPM, which induces bacterial spheroplast formation with lysis, caused PMN apoptosis. Levels of endotoxin, TNF-alpha and IL-6 in the supernatants with IPM and PLB were significantly lower than in those with other beta-lactam antibiotics. The filtrates of IPM- and PLB-treated E. coli supernatants induced PMN apoptosis, whereas those treated with other beta-lactam antibiotics increased PMN necrosis. Beta-lactam antibiotics have different impacts on the types of PMN cell death after E. coli killing. Underlying mechanisms and the clinical relevance of IPM-induced PMN apoptosis in severe gram-negative infection warrant further investigation.     

TNFalpha-induced suppression of PMN apoptosis is mediated through interleukin-8 production

Dysregulated neutrophil (polymorphonuclear PMN) apoptosis is thought to contribute to the onset of adult respiratory distress syndrome (ARDS) in critically ill patients. Tumor necrosis factor-alpha (TNFalpha), which is present in elevated levels in the bronchoalveolar lavage fluid in patients with ARDS, is thought to play a central role in regulating PMN function in the lungs. Studies have shown that short-term culture with TNFalpha increases apoptosis yet extended culture with TNFalpha suppresses apoptosis. However, it is unclear whether this latter effect of TNFalpha is directly or indirectly mediated through production of anti-apoptotic cytokines such as interleukin (IL)-8. To investigate the role of IL-8 in TNFalpha-induced apoptosis PMN were exposed to TNFalpha (100 ng/mL) in the presence or absence of antibodies to IL-8, and the extent of apoptosis was assessed. An enzyme-linked immunoassay was used to measure levels of the anti-apoptotic cytokine IL-8, induced by TNFalpha-stimulation. Because TNFalpha may mediate its effect through various cell-signaling pathways, we next assessed the effect of kinase inhibition on the ability of TNFalpha to effect apoptosis and IL-8 production. Treatment with TNFalpha had a biphasic effect: at 4-8 h, apoptosis was increased but was markedly suppressed at 24 h (P < 0.05). PMN cultured for 24 h with TNFalpha also showed markedly increased levels of IL-8. Neutralization of IL-8 inhibited the ability of TNFalpha to suppress apoptosis (P < 0.05). Incubation of TNFalpha + p38-mitogen-activated protein kinase (MAPK) inhibitor SB202190 increased apoptosis (P < 0.01) and decreased IL-8 production to PMN control. To a lesser extent, incubation of TNFalpha with inhibitors to NF-kappaB (SN50) and PI3K (LY294002) also increased apoptosis and decreased IL-8 production (P < 0.05). These data illustrate a novel mechanism by which TNFalpha can indirectly elicit an anti-apoptotic effect via p38-MAPK induced release of the anti-apoptotic chemokine IL-8. The exploitation of such a pathway represents a potential target for regulation of PMN-mediated acute lung injury.     

Ratio of bacteria to polymorphonuclear neutrophils (PMNs) determines PMN fate

This study was designed to investigate how live Escherichia coli influence the fate of polymorphonuclear neutrophils (PMNs) in vitro. PMNs from 10 healthy volunteers were cocultured with or without live E. coli at different ratios. Heat-killed E. coli (Hk) were also added to PMNs at a ratio of 1:10. The PMNs were then analyzed by flow cytometry for cell death, reactive oxygen intermediates (ROI) production, and CD16 expression. Morphologic features were also assessed. PMN apoptosis was confirmed by DNA gel electrophoresis. Low doses of E. coli (PMN:E. coli ratios of 1:0.01 and 1:0.1) inhibited PMN apoptosis. In contrast, a high dose of E. coli (PMN:E. coli ratio of 1:10) increased PMN necrosis. ROI production was significantly greater at PMN:E. coli ratios of 1:10 and 1:10 (Hk) than at ratios of 1:0.01 and 1:0.1, or in PMNs cultured alone after a 15 or 30 minute coculture. CD16 expressions were significantly lower in PMNs cocultured with E. coli than in those cultured alone after a 4 or 12-h coculture. Tumor necrosis factor-alpha, interleukin (IL)-1beta and IL-6 levels in cell-free supernatants were also measured. The mean percentages of apoptosis at PMN:E. coli ratios of 1:0.01 and 1:10 (Hk), and in PMNs cultured alone after a 12-h coculture showed significant inverse correlations with these cytokine levels in cell-free supernatants at 12 h. Our results demonstrate that low doses of live E. coli inhibits predominantly PMN apoptosis, whereas a high dose of E. coli increases necrosis. Augmented PMN bactericidal function, via inhibition of PMN cell death, may be beneficial for host defense against bacterial infection and/or sepsis.     

Effects of antiretroviral dideoxynucleosides on polymorphonuclear leukocyte function.

Dideoxynucleosides (zidovudine[AZT], dideoxycytidine[ddC], and dideoxyinosine[ddI]) are promising new agents for the management of human immunodeficiency virus type 1 (HIV-1) infections. In light of recent data demonstrating defects in the polymorphonuclear leukocyte (PMN) bactericidal activity of HIV-1-infected patients and since many chemotherapeutic agents affect PMN function, we examined their effects on the function of PMNs from both healthy and HIV-1-infected individuals in vitro. AZT (0.1 to 25 microM), ddC (0.01 to 1 microM), and ddI (0.2 to 50 microM) had no effect on viability, chemotaxis to N-fromylmethionyl leucyl phenylalanine, phagocytosis of Candida albicans or Staphylococcus aureus, or superoxide production following stimulation by N-formylmethionyl leucyl phenylalanine. Killing of C. albicans was not affected by AZT but was enhanced by 0.1 and 1 microM ddc (a 1 microM, killing was 26.0 +/- 2.02% compared with 17.0 +/- 0.73% for controls: P = 0.006) and 0.2 to 50 microM ddI (at 10 microM, killing was 25.0 +/- 0.68% compared with 17.8 +/- 0.91% for controls; P = 0.002). Killing of S. aureus was unchanged by AZT and ddC but was significantly enhanced by ddI at 0.2 to 20 microM (at 2 microM, killing was 71.2 +/- 5.57% compared with 51.4 +/- 6.29% for controls; P = 0.0045). In addition, the preexisting defective bactericidal capacity of PMNs from HIV-1-infected patients was enhanced by ddI (P less than 0.025). Potential enhancement by these dideoxynucleosides of certain PMN functions of HIV-1-infected patients deserves further study.     

Chemiluminescence of Human and Canine Polymorphonuclear Leukocytes in the Absence of Phagocytosis

Polymorphonuclear leukocytes (PMNs) have increased oxidative metabolism during phagocytosis and emit light (chemiluminescence, CL) as a result of metabolic activation. The present study examined PMN CL in the absence of phagocytosis using sodium fluoride (NaF), a nonparticulate agent and known stimulator of cellular oxidative metabolism. Normal human and canine PMNs were assayed in a CL spectrometer which permitted continuous sample mixing and constant temperature regulation during CL measurement. PMNs treated with 20 mM NaF demonstrated maximum CL responses of 10,000-20,000 cpm above background, 13-17 min after addition of NaF at 37 degrees C. Temperature regulation of reaction mixtures was found to be a critical factor in assaying PMN CL responses to NaF, because a small decrease in temperature (i.e. 1.5 degrees C) substantially depressed and delayed the CL response. Superoxide anion production correlated closely with CL responses in NaF-treated human PMNs. CL responses were completely suppressed in the presence of the oxidative metabolic inhibitors, iodoacetamide, and N-ethylmalemide; and were partially suppressed in the presence of either superoxide dismutase or sodium azide.CL responses of NaF-treated PMNs were significantly lower than responses generated by PMNs phagocytizing opsonized yeast. When NaF was evaluated for its effect on light generation from a singlet oxygen dependent CL reaction, it was found that NaF did not quench singlet oxygen light. This study demonstrates that PMN CL can occur in the absence of phagocytosis, and it proposes that a nonphagocytic PMN CL assay may be useful in evaluating leukocyte metabolic defects.     

Neutrophil dysfunctions in sickle cell disease

The abnormal susceptibility towards certain infections in SCD patients has a partial explanation in the well described functional defects of the spleen and of the alternative complement pathway; such defects probably account for the etiology of fulminant, often fatal, childhood infections with encapsulated organisms (Streptococcus pneumoniae, Haemophilus influenzae). On the other hand, the frequent systemic infections with enteric organisms in SCD patients, particularly of the salmonella species, and also with Staphylococcus aureus, are more difficult to explain. We therefore reviewed the potential contribution of neutrophil (PMN) dysfunctions to the increased infective tendency of SCD patients and included some previously unpublished data from our laboratory. While notable discrepancies still exist--and need further clarification--a tentative working hypothesis can be extracted from the available data: dysfunctions of neutrophils affect their locomotion (as reflected by decreased chemotaxis and in vivo migration), their phagocytic processes and their bactericidal performance. The latter concerns the ineffective killing of Staphylococcus aureus, Candida albicans, and Streptococcus pneumoniae. Dysfunctional bactericidal activity, in turn, apparently relates to a poor or at times non-existent PMN oxidative activity, which prevents the prompt disposal of microorganisms. Under certain circumstances salmonella species seem to further paralyze the oxidative machinery of PMNs in SCD. Serum from some patients contains a poorly defined inhibitor, or lacks an enhancing factor, and such serum abnormalities aggravate the existing defects just described. Interestingly recent findings suggest that dysfunctional PMNs may originate from the mandatory demargination of leukocytes secondary to the functional asplenia of SCD; a predominance of non-rosetting (EA-) PMNs among such leukocytes could produce the operational explanation for an exaggerated representation of dysfunctional PMNs in SCD patients with leukocytis.     

Neutrophil heterogeneity in patients with blunt trauma

To determine whether neutrophil (PMN) dysfunction observed in patients with blunt trauma could be explained by alterations in PMN functional subpopulations and to further study the origins of PMN heterogeneity, we studied PMN subpopulations in 18 patients with severe blunt trauma by using a micropore filter chemotactic assay and a mouse monoclonal antibody (31D8 Mab). A major PMN subpopulation binds 31D8 Mab avidly (31D8 "bright") and depolarizes and responds chemotactically to formyl peptide (fMLP) and C5a; a minor PMN subpopulation binds 31D8 Mab weakly (31D8 "dull") and fails to depolarize and responds poorly to fMLP and C5a. Fourteen patients with trauma had marked alteration of PMN 31D8 expression compared with healthy controls 52% +/- 20% versus 92% +/- 4% bright PMNs, respectively (p less than 0.01). These patients also had significantly decreased PMN chemotaxis and increased band counts compared with controls 30 +/- 10 micron versus 53 +/- 19 micron (p less than 0.01) and 34% +/- 14% versus 6% +/- 1% (p less than 0.01), respectively. Four patients with less-severe injuries had unaltered 31D8 PMN expression and normal PMN chemotaxis. In patients whose band counts exceeded 20%, there was a strong correlation between the number of bands and the percentage of 31D8 dull PMNs. PMNs that weakly express the 31D8 antigen appear to be less mature than PMNs that strongly express the antigen regardless of cell morphology (i.e., bands, multilobed cells). The data suggest that the decreased PMN chemotaxis and increased infection rate in patients with blunt trauma is caused partly by an increase in the number of poorly functioning 31D8 dull PMNs.     

Electrochemical detection of nitric oxide production in human polymorphonuclear neutrophil leukocytes.

The detection of nitric oxide (NO) release by human polymorphonuclear neutrophil leukocytes (PMNs) presents several difficulties, mainly due to concomitant production of O2- and H2O2, which could interfere with the measurements. A Nafion and nickel porphyrin-coated microelectrode was used to measure NO production in PMNs in vitro. It allowed detection of 6.3 +/- 1.9 nM NO in a PMN-containing system and was unaffected by added chemicals. Addition of the chemotactic oligopeptide f-met-leu-phe (fMLP; 100 nM) induced a NO release which reached a value of 71 +/- 30 pmol NO/10(6) PMN x ml(-1) 5 min after stimulation in the presence of SOD (150 U/ml). If SOD was omitted, the corresponding value was 36 +/- 20 pmol NO/10(6) PMN x ml(-1). Presence or absence of catalase did not alter the amount of NO measured. Addition of the NO-synthase inhibitor N(G)-monomethyl-L-arginine (LNMMA; 1 mM) reduced the current by 82 +/- 20%. These results agree with the rate of NO production in human PMNs when measured spectrophotometrically using the NO-dependent oxidation of oxyhaemoglobin to methaemoglobin. The NO production in human PMN was dependent on fMLP concentrations, but independent of cell-concentrations of 0.5-3.5 x 10(6)/ml. This paper shows that a electrochemical method, e.g. Nafion and porphyrin-coated microelectrode, is suitable for studies of NO release from stimulated human PMNs.     

Lipid complexing decreases amphotericin B inflammatory activation of human neutrophils compared with that of a desoxycholate-suspended preparation of amphotericin B (Fungizone).

Amphotericin B (AmB) has toxic effects and alters neutrophil (polymorphonuclear leukocyte [PMN]) function. A lipid-complexed formulation of AmB (AmB-LC) has been reported (A. S. Janoff, L. T. Boni, M. C. Popescu, S. R. Minchey, P. R. Cullis, T. D. Madden, T. Taraschi, S. M. Gruner, E. Shyamsunder, M. W. Tate, R. Mendelsohn, and D. Bonner, Proc. Natl. Acad. Sci. USA 85:6122-6126, 1988) to be less toxic than a desoxycholate-suspended preparation of AmB (AmB-des; Fungizone). In this study we compared the effects of AmB-des and AmB-LC on in vitro PMN function. Neither form of AmB stimulated PMN chemiluminescence, but AmB-des (2 micrograms/ml) nearly tripled PMN chemiluminescence in response to f-Met-Leu-Phe (fMLP), a phenomenon known as priming. Because AmB stimulates monocytes to release cytokines which can affect PMN function, we studied the effects of AmB on PMNs in mixed leukocyte cultures. AmB-des (1 to 2 micrograms/ml) increased the chemiluminescence of PMNs plus mixed mononuclear leukocytes (MNLs) to fMLP. The activity was about three times that of PMNs plus MNLs and seven times the activity of PMNs stimulated with fMLP in the absence of MNLs. Cell-free AmB-des (2 micrograms/ml)-stimulated, MNL-conditioned medium primed pure PMNs to a level equal to that of whole MNLs treated with AmB-des. AmB-LC was much less potent. AmB-LC (20 micrograms/ml) increased fMLP-stimulated chemiluminescence to two times that of PMNs plus MNLs without AmB-LC. AmB-des (2 micrograms/ml) (but not AmB-LC [2 micrograms/ml]) increased nitroblue tetrazolium reduction by PMNs in whole blood from 31 to 52% of positive cells. Neither form of AmB increased Mac-1 (the CD11b/CD18 integrin) expression of pure PMNs. AmB-des (0.5 to 2 micrograms/ml) (but not AmB-LC [< or = 40 micrograms/ml]) nearly doubled PMN Mac-1 expression in the presence of MNLs, and cell-free AmB-des (2 micrograms/ml)-stimulated, MNL-conditioned medium stimulated PMN Mac-1 to 125% of the control level. AmB-des (0.2 to 2 micrograms/ml) (but not AmB-LC [< or = 40 micrograms/ml]) decreased chemotaxis of pure PMNs to fMLP by as much as 35% and that of PMNs in the presence of MNLs by as much as 50%. Desoxycholate by itself had no effect on PMN function. These differences in activity between AmB-des and AmB-LC may explain the lessened toxicity observed with AmB-LC.     

Effects of azithromycin on shiga toxin production by Escherichia coli and subsequent host inflammatory response

Shiga toxin (Stx)-producing Escherichia coli (STEC) colonizes the human intestinal mucosa, produces Stx from phage, and causes the development of hemolytic-uremic syndrome via Stx-induced inflammatory cytokine production. Azithromycin exhibited strong in vitro activity against STEC without inducing Stx-converting phage, in marked contrast to norfloxacin. Azithromycin decreased the tumor necrosis factor alpha (TNF-alpha), interleukin-1beta (IL-1beta), and IL-6 production from Stx-treated human peripheral mononuclear cells or monocytes to a greater extent than did clarithromycin. In Stx-injected mice, azithromycin significantly suppressed Stx-induced TNF-alpha, IL-1beta, and IL-6 levels in serum and improved the outcome as assessed by survival rate. In the STEC oral infection experiment using immature mice immediately after weaning (weaned immature-mouse model), all mice died within 7 days postinfection. Azithromycin administration gave the mice 100% protection from killing, while ciprofloxacin administration gave them 67% protection. The data suggest that azithromycin (at least at higher concentrations) has a strong effect on Stx production by STEC and on the Stx-induced inflammatory host response and prevents death in mice. Azithromycin may have a beneficial effect on STEC-associated disease.     

Azithromycin reduces Chlamydia pneumoniae-induced attenuation of eNOS and cGMP production by endothelial cells

BACKGROUND: Intracellular infections with cytomegalovirus (CMV) or Chlamydia pneumoniae (Cp) may play a role in the aetiology of atherosclerosis. Nitric oxide (NO) is a key regulator of endothelial function. Under pathological conditions uncoupling of endothelial nitric oxide synthase (eNOS) leads to vessel damage as a result of production of oxygen radicals instead of NO. We hypothesized that infection-induced atherosclerosis is initiated by changes in NO metabolism and may be reversed by azithromycin treatment. METHODS: Confluent human umbilical vein endothelial cells (HUVECs) were infected with Cp or CMV. After 48 h of infection, production of eNOS, cyclic guanosine monophosphate (cGMP) and reactive oxygen species (ROS) was measured. Detection of cGMP was used as a reporter assay for the bioavailability of NO. Subsequently, Cp- and CMV-infected HUVECs were coincubated with 0.016 mg L(-1) and 1 mg L(-1) azithromycin. RESULTS: Infection with Cp (MOI 1 and MOI 0.1) and CMV (MOI 1) caused a dose- and time-dependent reduction of eNOS production in the HUVECs: Cp MOI 1: 1141 +/- 74 pg mL(-1) (P < 0.01); Cp MOI 0.1: 3189 +/- 30 pg mL(-1) (P < 0.01); CMV: 3213 +/- 11 pg mL(-1) (P < 0.01) vs. 3868 +/- 83 pg mL(-1) for uninfected HUVECs. Chlamydia pneumoniae- but not CMV-infection also reduced cGMP-production (Cp: 0.195 +/- 0.030 pmol mL(-1) (P < 0.01); CMV: 0.371 +/- 27 pmol mL(-1) (P > 0.05) vs. 0.378 +/- 0.019 pmol mL(-1) for uninfected HUVECs). CMV-infection did not affect ROS production either, but Cp-infection reduced ROS-production by 21% (P > 0.05; Cp MOI 0.1) to 68% (P < 0.01; Cp MOI 1). Azithromycin treatment restored Cp-induced eNOS, cGMP and ROS production in a dose-dependent manner. CONCLUSIONS: Infection with Cp in endothelial cells in vitro attenuates eNOS, cGMP and ROS production in HUVECs and azithromycin reverses Cp-induced effects on eNOS, cGMP and ROS-production. The results from our in vitro research support the role of antibiotic therapy for infection-induced atherosclerosis by indicating that azithromycin does actually improve endothelial function.     

Intracellular and extracellular defenses of human phagocytes against Blastomyces dermatitidis conidia and yeasts

The lesions of blastomycosis are characterized by both suppuration and granuloma formation, but the relative roles of human neutrophils, monocytes, and macrophages against Blastomyces dermatitidis are poorly defined. Our studies reveal that B. dermatitidis yeasts are generally too large to be ingested by polymorphonuclear neutrophils (PMNs), and are killed predominantly by external PMN attachment and degranulation, whereas conidia are first ingested, then killed. PMN function is maximal in the presence of serum, divalent cations, and complement, and killing is more efficient for conidia (approximately 50%) than for yeasts (approximately 20%). PMNs that have degranulated, but remain attached to yeasts, block access by contiguous PMNs. When degranulated PMNs are removed, allowing access by fresh PMNs, there is a further increment in yeast killing. Both conidia and yeasts are killed by predominantly oxidative PMN mechanisms, with conidia being greater activators of the respiratory burst, and proportionately more influenced by oxidative inhibitors. Peripheral blood monocytes can kill conidia (approximately 35%), but are feebly active against yeasts (approximately 5%). Monocyte-derived macrophages kill approximately 90% of conidia and 40% of yeasts. The dramatic susceptibility of conidia, the infective particles of B. dermatitidis, to nonspecific phagocytic host defenses may help to explain the relative rarity of blastomycosis as a clinical problem. The presence of PMNs in lesions of blastomycosis may indicate an active, although limited, role of these cells in host defense against B. dermatitidis yeasts.