Antibiotic and Pentoxifylline Inhibition of Antimicrobial Pathways in Human Polymorphonuclear Leukocytes:

Considerations for the Treatment
of Severe Infection

 

W. Lee Hand, MD^*†

Neva L. Thompson, MD^‡

Debra L. Hand, MD^†

 

^*Assistant Dean for Research,
^†Department of Research Development and
^‡Department of Internal Medicine,

Texas Tech University Health Sciences Center,

4800 Alberta Avenue, El Paso, TX 79905

and VA Medical Center, Decatur, GA 30033

 

This study was supported in part by a grant from Hoechst-Roussel Pharmaceuticals and the State of Texas Tobacco Settlement Endowment earnings.

 

KEY WORDS: antibiotics; pentoxifylline; inhibition of PMN antimicrobial function

ABSTRACT

Interactions between antibiotic agents and leukocytes may influence the outcome of therapy for bacterial infections. Certain antibiotics (clindamycin, macrolides) that are highly concentrated by human polymorphonuclear neutrophilic leukocytes (PMN), as well as the biologic response modifier pentoxifylline, inhibit the respiratory burst response in these cells. As human trials of pentoxifylline use in severe sepsis continue, both this drug and modulatory antibiotics will be administered to the same patients. This prompted us to study the effects of these agents on antimicrobial activities in PMN. Pentoxifylline slightly reduced microbial particle-induced extracellular release of granular enzymes, whereas clindamycin had no effect on PMN degranulation. Clindamycin inhibited the production of superoxide by stimulated cytoplasts, which contain very few granules, in a manner identical to that observed with intact cells. Each of the evaluated respiratory burst modulators, pentoxifylline, clindamycin, and roxithromycin, demonstrated a similar inhibitory effect on intraphagocytic killing of antibiotic-resistant Streptococcus aureus by PMN, apparently by altering the production of oxidative antimicrobial factors. Pentoxifylline and antimicrobial agents influence the respiratory burst activation pathway at different sites; therefore, inhibition of inflammatory and antibacterial activities of activated PMN may be additive. When both types of drugs are given to the same patients for treatment of severe infections, the result may be a “double-edged sword.” Inhibition of PMN antimicrobial activity might have adverse consequences, but the possible control of excessive inflammation due to oxidative radical release has obvious therapeutic potential. These interactions between modulatory antibiotics, pentoxifylline, and human PMN require additional evaluation to define their potential role in improving the management of severe bacterial infections.

INTRODUCTION

The therapeutic effectiveness of an antimicrobial agent may depend in part on the interactions between the drug and phagocytic cells. An optimal antibiotic would kill extracellular organisms, would enter phagocytic cells, have no adverse impact on phagocytic cell function, and eradicate surviving intracellular organisms.

Unfortunately, most antibiotics (especially ß-lactam drugs) fail to enter leukocytes efficiently and, therefore, are unable to kill intraphagocytic organisms. Conversely, even antibiotics that achieve high cellular levels may not kill antibiotic-sensitive, intraphagocytic bacteria.1,2 One potential cause for the failure of some highly concentrated drugs to kill intraphagocytic organisms is antibiotic-induced inhibition of phagocytic cell antimicrobial activity. Indeed, we found that several antibiotics that are highly concentrated within human polymorphonuclear neutrophilic leukocytes (PMN; clindamycin, macrolides, and trimethoprim) inhibit the stimulated respiratory burst response in these cells.3–6

We also examined the ability of pentoxifylline, a drug with major effects on cell membrane functions in mammalian cells, to modulate cellular activation and other plasma membrane—associated functions in human PMN. These studies demonstrated that pentoxifylline reduced both the agonist-induced respiratory burst response (superoxide production) and membrane transport of certain substances (e.g., adenosine) in human PMN.6,7 Two pentoxifylline derivatives, HWA-448 (torbafylline) and HWA-138 (ambuphylline), also inhibited the respiratory burst response in PMN.6 Furthermore, pentoxifylline and these 2 derivatives increased the uptake by PMN of several modulatory antibiotics (clindamycin, roxithromycin, dirithromycin), especially during phagocytosis. The combination of pentoxifylline and a modulatory antibiotic inhibited superoxide generation to a greater extent than either agent alone.6 It is noteworthy that pentoxifylline is protective in animal models of infection and injury.8–16 These protective effects are due in part to drug-induced suppression of excessive and injurious phagocytic cell (PMN and macrophage) responses.6,7,17–27 Pentoxifylline is currently under investigation in human sepsis and septic shock as a potential protective biological response modulator.28

Although we investigated pathways by which certain antibiotics and pentoxifylline might influence PMN oxidative function, other biological consequences of cellular interaction with these drugs were not defined in our previous studies.3,5,7 It is important to determine the impact of these modulatory drugs on antimicrobial function in human PMN, because both types of drugs may be administered to patients with severe infection. Therefore, we examined the effects of inhibitory antibiotics and pentoxifylline on nonoxidative (cytoplasmic granule) antimicrobial functions and on intracellular bactericidal activity in human PMN.

METHODS

Bacteria

A clinical isolate of Staphylococcus aureus, resistant to clindamycin and macrolides (but not methicillin), was stored at -70˚C. Radiolabeled organisms were prepared by growth in Trypticase soy broth (TSB; BBL Microbiology Systems, Cockeysville, MD), containing 10 mCi of [methyl-3H] thymidine per mL. Bacteria were grown overnight, washed, and used in studies with human PMN and antibiotic or pentoxifylline as described.

Preparation of human PMN

Peripheral venous blood from normal volunteers was collected by venipuncture. Granulocytes were isolated by dextran sedimentation and Hypaque-Ficoll density gradient centrifugation, washed, and resuspended in tissue culture medium 199 (TC 199) or Hanks balanced salt solution (HBSS).29–31

Antibiotics, Pentoxifylline, Adenosine

Antibiotics (standard susceptibility powders) used in these studies were clindamycin (Upjohn Co. [now Pharmacia and Upjohn], Kalamazoo, MI), roxithromycin (Hoechst-Roussel Pharmaceuticals, [now a component of Aventis], Somerville, NJ), and penicillin G (Pfizer, New York, NY). Pentoxifylline (Hoechst-Roussel) was stored at -70˚C, reconstituted in HBSS, and used in experiments at concentrations of 10–3 M to 10–5 M. Adenosine (Sigma Chemical Co., St. Louis, MO) was used at final concentrations of 10-4 M to 10-6 M in studies of PMN cytoplast superoxide production. Zymosan A (Sigma) was boiled, washed, opsonized with fresh human serum, stored at -70˚C until needed, and used at a final concentration of 0.1 mg/mL in experiments. A clinical isolate of S. aureus was grown overnight in TSB, and then was heat-killed, opsonized, and stored at -70˚C until use. Concanavalin A (Con A;Sigma) (30 mg/mL) was used as a soluble membrane-stimulating agent.

Assays of Phagocyte Granular and Cytoplasmic Enzymes

Degranulation induced by microbial particle ingestion was evaluated by assaying the extracellular release of neutrophil granular enzymes.3 b-glucuronidase serves as a marker for azurophil granules, and lysozyme is found in both azurophil and specific granules. Release of lactic dehydrogenase (LDH), a cytoplasmic enzyme, was monitored as an indicator of cellular damage. PMN (107 mL in HBSS) were preincubated in the presence or absence of antibiotic or pentoxifylline for 15 minutes, followed by addition of cytochalasin B (5 mg/mL) (Sigma) to enhance exocytosis during ingestion, and then the particulate stimulating agent (S. aureus or zymosan). After additional incubation (with rotation) for 20 minutes at 37˚C, cells were recovered by centrifugation, resuspended, and lysed in 0.2% Triton X-100. Enzyme activity was determined in supernatants, pellets, and whole-cell suspensions. b-glucuronidase was measured as the release of phenolphthalein from phenolphthalein b-glucuronide. Lysozyme activity was determined by measuring the rate of lysis of Micrococcus lysodeikticus. LDH was assayed by determining the increase in A340 (Beckman 640 B spectrophotometer) resulting from the reduction of NAD to form NADH.

Effects of clindamycin and adenosine on PMN cytoplast superoxide production. Clindamycin and adenosine (nucleosides) inhibit respiratory burst activity in human PMN.3,5 In this study, we examined the effect of these agents on respiratory burst activity in PMN cytoplasts, which contain no nuclei and very few granules. Cytoplasts accumulate nearly as much clindamycin as intact cells.32 Enucleated human PMN (cytoplasts) were prepared by ultracentrifugation of isolated peripheral blood granulocytes in 12.5% Ficoll with 20 mM cytochalasin B on a discontinuous Ficoll gradient (6 mL of 16% Ficoll over 6 mL of 25% Ficoll) containing cytochalasin B.32,33

After centrifugation, the cytoplasts present at the interface of the 12.5% and 16% Ficoll solution were collected with a pipette and washed 5 times to remove the cytochalasin B. Confirmation that these cytoplasts contain virtually no granules was obtained by monitoring the content of 2 granular enzymes, b-glucuronidase and lysozyme, in PMN and cytoplasts. The cytoplasmic LDH content of intact and enucleated cells was also quantified. Superoxide generation by intact PMN and cytoplasts was determined as superoxide dismutase-inhibitable reduction of ferricytochrome C.3,7 Intact or enucleated cells (2 x 106 cells per 3 mL) in HBSS with 100 mM ferricytochrome C were preincubated in the presence or absence of clindamycin, penicillin G, or adenosine for 15 minutes. Cytochalasin B (5 m/mL) and the stimulating agent (zymosan or Con A) or control media were then added. Duplicate samples contained superoxide dismutase (50 mg/mL). After incubation, the cells (intact or enucleated) were removed by centrifugation at 4˚C. The OD of the supernatants was determined at 550 nM in a Beckman spectrophotometer.

Influence of Antibiotics and Pentoxifylline on Bacterial Ingestion and Killing by Human PMN

The influence of drugs that modulate the respiratory burst response on ingestion and killing of bacteria by PMN was evaluated by means of a phagocytic-bactericidal assay, as previously described.1,34 PMN (5 x 106/mL) in TC199 5% serum were preincubated at 37˚C for 30 minutes with clindamycin, roxithromycin, pentoxifylline, or control media, followed by the addition of antibiotic-resistant, radiolabeled S. aureus (5 x 107 to 1 x 108 bacteria/mL). After a 30-minute phagocytosis period, the leukocytes from some tubes were recovered by centrifugation, washed, lysed, appropriately diluted, and used for radioactive counting and bacterial enumeration. The remaining tubes containing PMN and bacteria, with or without test agents, were incubated for an additional period of time before quantification of viable intracellular bacteria. Calculations of phagocytosis and of intraphagocytic bactericidal activity were performed as previously described.1,6,35 The effects of PMN exposure to antibiotics or pentoxifylline on ingestion of S. aureus and survival of intraphagocytic bacteria were determined and compared to controls.

Statistical Analysis of Data

Experimental group results were expressed as the mean plus/minus standard error of the mean. Differences between experimental groups were evaluated by means of paired or unpaired t-tests, using a computer statistical program.

RESULTS

Characteristics of the Cell Population

The granulocytic cell population was 97% or more neutrophils; more than 90% were mature polymorphonuclear neutrophils. Therefore, the cells used in these studies will be referred to as PMN. More than 95% of these cells were viable, as judged by trypan blue exclusion.

Assays of PMN Granular and Cytoplasmic Enzymes

Exposure of PMN to microbial particles causes degranulation, with release of certain granular enzymes to the extracellular environment. Pentoxifylline, at concentrations of 10-3 M to 10-5 M, inhibited the release of lysozyme by PMN ingesting S. aureus or zymosan. The phagocytosis-induced extracellular release of b-glucuronidase from azurophil granules was also inhibited by pentoxifylline (Table 1). Pentoxifylline itself had no effect on the measurement of lysozyme or b-glucuronidase. At nontoxic concentrations, clindamycin had no inhibitory effect on microbial particle—stimulated release of granular enzymes. Cell viability (trypan blue exclusion) and extracellular release of cytoplasmic LDH were monitored as indicators of cellular damage. The ability of cells to exclude trypan blue was not altered by antibiotic or pentoxifylline. Release of cytoplasmic LDH to the external environment by PMN was not influenced by tested concentrations of clindamycin or pentoxifylline.

Superoxide Production by Human
PMN Cytoplasts

Enucleated human PMN contain very few granules.32,33 Therefore, stimulated superoxide production in PMN cytoplasts will be independent of granule function. Both clindamycin and adenosine modulate the respiratory burst response in intact PMN, but the inhibitory effects of these agents differ in regard to stimulus specificity and site of action.3,5 In this study, clindamycin was a potent inhibitor of superoxide generation in cytoplasts activated by zymosan and Con A (Table 2). These findings were similar to those observed with intact neutrophils.3 The modulatory effects of adenosine were also consistent with those seen in experiments with intact PMN.3 Penicillin G, which enters PMN poorly and has no effect on respiratory burst activity in intact cells, was used as a “negative” control and failed to alter superoxide production in cytoplasts.

Influence of Antibiotics and Pentoxifylline on Bacterial Ingestion
and Killing by Human PMN

The effects of highly concentrated modulatory antibiotics or pentoxifylline on PMN antimicrobial function were evaluated in a phagocytic-bactericidal assay system. An isolate of S. aureus highly resistant to the antibiotics used in these experiments was used as the test organism. The minimal inhibitory concentrations of clindamycin and roxithromycin against this organism were more than 1000 mg/mL. Thus, in the absence of PMN, the antibiotics themselves (10-4 M to 10-3 M [clindamycin 42.5 to 425 mg/mL; roxithromycin 83.7 to 837 mg/mL]) did not affect bactericidal survival during a 2-hour incubation period. In the phagocytic-bactericidal assays, a mean 8.4 x 107 bacteria were added to each milliliter of PMN suspension (5 x 106 cells). Of these radiolabeled bacteria, 56% to 59.5% (4.7 to 5.0 x 107 organisms) were ingested by neutrophils (Table 3). The intraphagocytic location of organisms was confirmed by microscopic examination.

Using this same system, we previously demonstrated that the numbers of PMN-associated staphylococci were similar in the presence or absence of lysostaphin.1 Therefore, the calculation of “intraphagocytic” radiolabeled S. aureus in this study is an accurate reflection of bacterial ingestion by PMN (as opposed to binding without ingestion). As might be expected, clindamycin and roxithromycin failed to alter the ingestion of these highly antibiotic-resistant organisms by PMN (Table 3). Pentoxifylline was also without effect on phagocytosis in this assay. In contrast, each of the study drugs did inhibit the bactericidal activity of PMN. Thus, clindamycin (5 x 10-4 M), roxithromycin (10-4 M), and pentoxifylline (10-3 M) all decreased the ability of neutrophils to kill S. aureus after ingestion (Table 3).

DISCUSSION

In previous studies we demonstrated that several weakly basic antibiotics (particularly clindamycin and macrolides) that are highly concentrated within phagocytic cells inhibit the respiratory burst response in human PMN.3,5 Pentoxifylline, a methylxanthine derivative with major effects on mammalian cell membrane function, modulates superoxide production in activated neutrophils.6,7 Two pentoxifylline derivatives, HWA-448 (torbafylline) and HWA-138 (ambuphylline) also inhibit the respiratory burst response in PMN.6

This study addressed specific questions concerning the effects of these drugs on other antimicrobial functions in PMN. Initially, we evaluated the influence of the antibiotics and pentoxifylline on degranulation induced by phagocytosis. Pentoxifylline had a modest effect on zymosan- and S. aureus—induced degranulation, whereas clindamycin failed to alter the release of granular enzymes during phagocytosis. Previously we had found that clindamycin and macrolides had no effect on FMLP-stimulated release of granular enzymes by human PMN.3 Clindamycin has been reported to inhibit degranulation of PMN lysosomes.36 However, this occurs only at high concentrations (10-3 M or greater), which we found to be toxic.3

Stimulated PMN cytoplasts (which contain virtually no cytoplasmic granules) produce superoxide, generation of which is inhibited by clindamycin in a manner identical to that observed with intact PMN. In previous studies, we documented that cytoplasts accumulate nearly as much clindamycin as intact cells.32 Thus, the presence of cytoplasmic granules is not required for either uptake of clindamycin or modulation of the respiratory burst by this drug.

These agents that modulate the respiratory burst response in PMN also exhibited an inhibitory effect on intraphagocytic bactericidal activity. The concentrations of clindamycin, roxithromycin, and pentoxifylline required to modulate superoxide production are similar to those that inhibited killing of S. aureus by PMN. Clindamycin activity is not dependent on nonoxidative (cytoplasmic granule) antimicrobial functions, and this agent apparently inhibits PMN bactericidal activity by suppressing production of oxidative antimicrobial factors. Pentoxifylline caused a very modest inhibition of degranulation in stimulated PMN. However, the similarity of the pentoxifylline and antibiotic effects on antibacterial activity suggests that modulation of the oxidative respiratory burst function is the important factor in pentoxifylline-induced inhibition of PMN bactericidal activity against S. aureus. This is in keeping with the known essential role of granulocyte oxidative bactericidal function in protecting against infection due to S. aureus.37

Clindamycin- and roxithromycin-induced modulation of the respiratory burst response is mediated through an intracellular site or sites of action. We found that these agents inhibit the phospholipase D/phosphatidic acid phosphohydrolase (PAH) pathway in activated neutrophils. This antibiotic-mediated inhibition of PAH activity leads to decreased diradylglycerol generation and, as a consequence, its activation of the NADPH (respiratory burst)-oxidase.5 Pentoxifylline does not alter phospholipase D/PAH activity, but does inhibit production of phosphatidic acids, which are important intracellular messengers for cellular activation.38,39 This pentoxifylline-induced change could cause decreased production of diracylglycerol with diminished activation of the respiratory burst oxidase. Inhibition of phosphodiesterase activity by pentoxifylline might also alter the respiratory burst activation sequence40 and other cellular activation processes, perhaps by increasing intracellular cAMP.41

In summary, our studies have shown that clindamycin, roxithromycin, and pentoxifylline, agents that modulate the respiratory burst response in PMN, also inhibit intraphagocytic bactericidal activity. These appear to be casually linked phenomena. In a previous study, we found that the combination of both pentoxifylline and a modulatory antibiotic inhibited PMN superoxide generation to a greater extent than either agent alone. This additive effect might be expected, because pentoxifylline and the modulatory antibiotics impact the respiratory burst activation pathway at different sites.

Pentoxifylline is protective in animal model studies of infection and injury, and is under evaluation in human sepsis and septic shock as a possible protective biological response modifier. In animal studies, the beneficial effects attributed to pentoxifylline have included improved microcirculatory blood flow, preservation of visceral organ function, inhibition of endothelial activation, stimulation of fibrinolysis, and improved survival in some shock models.8–16 In human sepsis and septic shock studies, pentoxifylline administration was associated with improved cardiopulmonary function and decreased serum tumor necrosis factor a (TNF-a levels.28,42–45 In one small study of premature infants with documented bacteremia and sepsis, pentoxifylline administration lowered IL-6 and TNF-a blood levels and appeared to reduce mortality.45 Pentoxifylline also decreased mortality and the length of hospital stay in a small study of patients with peritonitis due to a perforated viscus.46

These protective effects of pentoxifylline are due, at least in part, to drug-mediated suppression of excessive and injurious phagocytic cell (PMN and macrophage) responses. For example, we have shown that pentoxifylline modulates the respiratory burst response in human PMN, and the combination of pentoxifylline and a modulatory antibiotic inhibited PMN superoxide production more than either agent alone. Furthermore, this additive effect may be amplified during an infection because pentoxifylline increases PMN uptake of modulatory antibiotics, especially during ingestion of microbial particles.6

As human trials of pentoxifylline use in severe sepsis continue, it is obvious that both this drug and modulatory antibiotics will be administered to the same patients. We must carefully monitor such cases, because this scenario can be viewed as a “double-edged sword.” The possibility that inhibition of PMN antimicrobial activity might have adverse clinical consequences is of obvious concern. Nevertheless, the potential for these agents to control excessive inflammation and tissue damage due to release of toxic oxidative radicals and granular enzymes has obvious therapeutic appeal.

The exciting possibilities regarding the interactions among modulatory antibiotics, pentoxifylline (and its derivatives), and phagocytic cells require additional evaluation. We have reason to hope that this combination of pentoxifylline and current or future modulatory antibiotics will improve therapy for certain severe bacterial infections.

REFERENCES

1. Hand WL, King-Thompson NL: Contrasts between phagocyte antibiotic uptake and subsequent intracellular bactericidal activity. Antimicrob Agents Chemother 29:135–140, 1986.

2. Labro M-T: Interference of antibacterial agents with phagocyte functions: Immunomodulation or “immuno-fairy tales”? Clin Microbiol Rev 13:615–650, 2000.

3. Hand WL, Hand DL, King-Thompson NL: Antibiotic inhibition of the respiratory burst response in human polymorphonuclear leukocytes. Antimicrob Agents Chemother 34:863–870, 1990.

4. Hand WL, Hand DL: Interactions of dirithromycin with human polymorphonuclear leukocytes. Antimicrob Agents Chemother 37:2257–2562, 1993.

5. Perry DK, Hand WL, Edmondson DE, Lambeth JD: Role of phospholipase D-derived diradylglycerol in the activation of the human neutrophil respiratory burst oxidase: Inhibition by phosphatidic acid phosphohydrolase inhibitors. J Immunol 149:2749–2758, 1992.

6. Hand WL, Hand DL: Influence of pentoxifylline and its derivatives on antibiotic uptake and superoxide generation by human phagocytic cells. Antimicrob Agents Chemother 39:1574–1579, 1995.

7. Hand WL, Butera ML, King-Thompson NL, Hand DL: Pentoxifylline modulation of plasma membrane functions in human polymorphonuclear leukocytes. Infect Immunol 57: 3520–3526, 1989.

8. Bjornson HS, Cave CM, Bjornson AB: Effects of pentoxifylline on survival and chemotaxis of polymorphonuclear leukocytes in a rat model of bacterial peritonitis. In Mandell GL, Novick WJ Jr (eds): Proceedings of a Symposium, Pentoxifylline and Leukocyte Function. Somerville, NJ: Hoechst-Roussel Pharmaceuticals, pp138–153, 1988.

9. Till GO, Warren JS, Gannon DE, et al: Effects of pentoxifylline on phagocyte responses in vitro and acute and chronic inflammatory reactions in vivo. In Mandell GL, Novick WJ Jr (eds). Proceedings of a Symposium, Pentoxifylline and Leukocyte Function. Somerville, NJ: Hoechst-Roussel Pharmaceuticals, pp124–137, 1988.

10. Welsh CH, Lien D, Worthen GS, Weil JV: Pentoxifylline decreases endotoxin-induced pulmonary neutrophil sequestration and extravascular protein accumulation in the dog. Am Rev Resp Dis 138:1106–1114, 1988.

11. Lundblad R, Ekstrom P, Giercksky K-E: Pentoxifylline improves survival and reduces tumor necrosis factor, interleukin-6, and endothelin-1 in fulminant intra-abdominal sepsis in rats. Shock 3:210–215, 1995.

12. Waxman K: Pentoxifylline preserves intestinal function (and more) following shock. Crit Care Med 26:9–10, 1998.

13. Hotchkiss RS, Karl IE: Pentoxifylline and modulation of the inflammatory response. Crit Care Med 26:427–428, 1998.

14. Heller S, Weber K, Heller A, et al: Pentoxifylline improves bacterial clearance during hemorrhage and endotoxemia. Crit Care Med 27:756–763, 1999.

15. Reynolds H: Pentoxifylline: More evidence that it improves host defenses during sepsis. Crit Care Med 27:681–683, 1999.

16. Howe LM: Novel agents in the therapy of endotoxic shock. Exp Opin Invest Drugs 9:1363–1372, 2000.

17. Hammerschmidt DE, Kotasek T, McCarthy T, et al: Pentoxifylline inhibits granulocyte and platelet function, including granulocyte priming by platelet activating factor. J Lab Clin Med 112:254–263, 1988.

18. Sullivan GW, Carper HT, Novick WJ Jr, Mandell GL: Inhibition of the inflammatory action of interleukin-1 and tumor necrosis factor (alpha) on neutrophil function by pentoxifylline. Infect Immunol 56:1722–1729, 1988.

19. Josaki K, Contrino J, Kristie J, et al: Pentoxifylline-induced modulation of human leukocyte function in vitro. Am J Pathol 136:623–630, 1990.

20. Endres S, Fülle H-S, Sinha B, et al: Cyclic nucleotides differentially regulate the synthesis of tumor necrosis factor-a and interleukin-1b by human mononuclear cells. Immunology 72:56-60, 1991.

21. Neuner P, Klosner G, Pourmojib M, et al: Pentoxifylline in vitro and in vivo down-regulates the expression of the intercellular adhesion molecule-1 in monocytes. Immunology 90:435-439, 1997.

22. Oka Y, Hasegawa N, Nakayama M, et al: Selective down regulation of neutrophils by a phosphatidic acid generation inhibitor in a porcine sepsis model. J Surg Res 81:147-155, 1999.

23. Porter MH, Hrupka BJ, Altreuther G, et al: Inhibition of TNF-a production contributes to the attenuation of LPS-induced hypophagia by pentoxifylline. Am J Physiol Regulatory Integrative Comp Physiol 279:R2113-R2120, 2000.

24. Alkharfy KM, Kellum JA, Matzke GR: Unintended immunomodulation: Part II. Effects of pharmacological agents on cytokine activity. Shock 13:346-360, 2000.

25. Krakauer T, Stiles BG: Pentoxifylline inhibits superantigen- induced toxic shock and cytokine release. Clin Diag Lab Immunol 6: 594-598, 1999.

26. Van Crevel R, Vonk AG, Netea MG, et al: Modulation of LPS-,PHA-, and M.tuberculosis-mediated cytokine production by pentoxifylline and thalidomide. Eur Cytokine Net 11: 574-579, 2000.

27. Bahra PS, Rainger GE, Wautier J-L, Nash GB: Effects of pentoxifylline on the different steps during adhesion and transendothelial migration of flowing neutrophiles. Cell Biochem Funct 19: 249-257, 2001.

28. Zimmerman JJ: Appraising the potential of pentoxifylline in septic premies. Crit Care Med 27:695-697, 1999.

29. Boyum A: Isolation of lymphocytes, granulocytes, and macrophages. Scand J Immunol 5 (Suppl 5):9-15, 1976.

30. Prokesch RC, Hand WL: Antibiotic entry into human polymorphonuclear leukocytes. Antimicrob Agents Chemother 21:373-380, 1982.

31. Steinberg TH, Hand WL: Effects of phagocytosis on antibiotic and nucleoside uptake by human polymorphonuclear leukocytes. J Infect Dis 149:397-403, 1984.

32. Hand WL, King-Thompson NL: Uptake of antibiotics by human polymorphonuclear leukocyte cytoplasts. Antimicrob Agents Chemother 34:1189-1193, 1990.

33. Roos D, Voetman AA, Meerhof LJ: Functional activity of enucleated human polymorphonuclear leukocytes. J Cell Biol 97:368-377, 1993.

34. Steinberg TH, Hand WL: Effect of phagocyte membrane stimulation on antibiotic uptake and intracellular bactericidal activity. Antimicrob Agents Chemother 31:271-276, 1987.

35. Hand WL, King-Thompson NL, Johnson JD: Influence of bacterial-antibiotic interactions on subsequent antimicrobial activity of alveolar macrophages. J Infect Dis 149:271-276, 1984.

36. Klempner MS, Styrt B: Alkalinization of the intralysosomal ph by clindamycin and its effects on neutrophil function. J Antimicrob Chemother 12 (Suppl C):39-50, 1983.

37. Waldvogel, FA: Staphylococcus aureus (including staphylococcal toxic shock). In Mandell, GL, Bennett, JE, Dolin, R (eds): Principles and Practice of Infectious Diseases, 5th Edn.            Philadelphia: Churchill Livingstone, pp2069-2092, 2000.

38. Yasui K, Komiyama A, Molski TFP, Sha’afi RI: Pentoxifylline and CD14 antibody additively inhibit priming of polymorphonuclear leukocytes for enhanced release of superoxide by lipopolysaccharide: possible mechanisms of these actions. Infect Immunol 62:922-927, 1994.

39. Rice GC, Brown PA, Nelson RJ, et al: Protection from endotoxin shock in mice by pharmacologic inhibition of phosphatidic acid. Proc Natl Acad Sci U S A 91:3857-3861, 1994.

40. Thiel MH, Bardenheuer H, Poch G, et al: Pentoxifylline does not act via adenosine receptors in the inhibition of the superoxide anion production of human polymorphonuclear leukocytes. Biochem Biophys Res Commun 180:53-58, 1991.

41. Reimund J-M, Raboisson P, Pinna G, et al: Anti-TNF-( properties of new 9-benzyladenine derivatives with selective hosphodiesterase-4-inhibiting properties. Biochem Biophys Res Commun 288: 427-434, 2001.

42. Zeni F, Pain P, Vindimian M, et al: Effects of pentoxifylline on circulating cytokine concentrations and hemodynamics in patients with septic shock: results from a double-blind, randomized, placebo-controlled study. Crit Care Med 24:207-214, 1996.

43. Staudinger T, Presterl E, Graninger W, et al. Influence of pentoxifylline on cytokine levels and inflammatory parameters in septic shock. Intensive Care Med 22:888-893, 1996.

44. Staubach K-H, Schröder J, Stüber F, et al: Effect of pentoxifylline in severe sepsis: Results of a randomized, double-blind, placebo-controlled study. Arch Surg 133:94-100, 1998.

45. Lauterbach R, Pawlik D, Kowalczyk D, et al: Effect of the immunomodulating agent, pentoxifylline, in the treatment of sepsis in prematurely delivered infants: a placebo-controlled, double-blind trial. Crit Care Med 27:807-814, 1999.

46. Shukla VK, Ojha AK, Pandey M, Pandey BL: Pentoxifylline in perforated peritonitis: results of a randomized, placebo controlled trial. Eur J Surg 167: 622-624, 2001.

 

 

 

 

 

 

 

 

Table 1. Influence of Antibiotics and Pentoxifylline on Release of Granular and Cytoplasmic Enzymes by Zymosan- and Staphylococcus aureus—Stimulated Human PMN

 

 

 

   Zymosan

   Control (PMN only)                13.4±1.1(8)*                                          6.8±1.4(9)                        10.7±1.2(9)

 

   Zymosan stimulated

 

   Clindamycin

   5x10-4M                         28.6±3.8(4), P=0.37‡             16.0±2.1(4), P=0.70          16.3±2.8(4), P=0.29

 

   Pentoxifylline

   10-3M                             20.0±1.3(4), P=0.002              6.6±1.0(4), P=0.02           12.9±1.8(4), P=0.97

   10-4M                             22.7±6.1(4), P=0.008             10.2±4.0(4), P=0.26          13.8±3.6(4), P=0.78

 

   S. aureus            

   Control (PMN only)                15.0±1.2(13)                           6.8±1.2(11)                      12.5±1.3(13)

 

   S. aureus stimulated

 

   Clindamycin

   5x10-4M                         28.5±2.9(8), P=0.93‡              17.6±3.6(6), P=0.31          21.7±5.1(7), P=0.18

 

   Pentoxifylline

   10-3M                             17.0±2.6(4), P=0.009              7.2±1.5(4), P=0.09           14.2±1.1(4), P=0.83

   10-4M                             20.2±1.9(9), P=0.008              8.2±2.2(6), P=0.11           10.8±1.5(8), P=0.17

   10-5M                             20.0±0.8(7), P=0.006              9.1±3.0(5), P=0.23           10.8±1.7(6), P=0.23

 

*Data are means ± standard errors of the mean. The number of experiments is in parentheses.

†P values reflect differences between control (PMN only) and experimental (zymosan-or S. aureus-stimulated) groups.

‡Remaining P values reflect differences between zymosan- or S. aureus-stimulated (no addition) and zymosan- or S. aureus-stimulated (plus antibiotic or pentoxifylline) groups.

 

 

 

Table 2. Effect of Clindamycin and Adenosine on Stimulated Superoxide Production by Human PMN and Cytoplasts

 

 

 

   Clindamycin

   Con A

      PMN                       56.7±6.9(6)†     51.6±15.3(5)†      65.7±19.2(4)         95.0(6.7(5)               —

 

   Zymosan

      PMN                       51.0±6.7(8)†       82.4±6.2(4)            125.0(1)           91.4±15.2(4)             —

 

   Adenosine

   Con A

      PMN                       30.4±5.8(6)†               —                        —               33.9±13.9(6)†    49.0±18.4(4)

 

   Zymosan

      PMN                       42.8±8.5(7)b               —                        —                49.2±14.5(3)     50.5±24.5(2)

      Cytoplasts             73.3±9.5(6)b               —                        —               40.1±13.1(3)†    67.8±14.5(2)

 

*Data are means ± standard errors of the mean. The number of experiments is in parentheses.

†Significant difference between control (PMN + stimulus) and experimental (PMN + stimulus + clindamycin or adenosine) group; P < 0.05.

 

 

Table 3. Effects of Clindamycin, Roxithromycin, and Pentoxifylline on Phagocytosis and Killing of Antibiotic-Resistant S. aureus by Human PMN*

 

                     Phagocytosis            Bactericidal Activity

 

 

 

   PMN + Drug

     Clindamycin                          4.7 (± 0.7)x107[8]                  1.3 (± 0.3)x107[8]†       8.1 (± 0.9)x106[8]

     Roxithromycin                      5.0 (± 0.7)x107[8]                  1.5 (± 0.3)x107[8]†      1.4 (± 0.3)x107[8]†

     Pentoxifylline                        4.9 (± 0.6)x107[7]                  1.2 (± 0.2)x107[7]†      1.3 (± 0.1)x107[7]†

 

*Data are means (± standard errors of the mean) of bacterial numbers. The number of experiments is in brackets.

†Significant difference between control (PMN only) and experimental (PMN + antibiotic or pentoxifylline) group; P < 0.05.