Influence of Low Molecular Weight Heparin (Certoparin) - Springer Link

Inflammation, Vol. 25, No. 5, October 2001 ( 2001)

Influence of Low Molecular Weight Heparin (Certoparin) and Unfractionated Heparin on the Release of Cytokines from Human Leukocytes M. Ko¨ ller,1,3 F. Kutscha-Lissberg,1 J. Brom,2 G. Weidinger,2 and G. Muhr1

Abstract—We analyzed the influence of heparins (unfractionated heparin, UFH and low molecular weight heparin certoparin) on the generation of IL-1ra, IL-6, IL-10, and IL-12p40 and from leukocyte fractions in vitro. Polymorphonuclear neutrophil leukocytes (PMN) and peripheral blood mononuclear cells (PBMC) from 16 different healthy donors were isolated and adjusted to 1 × 106 cells/ ml supplemented RPMI 1640. Leukocyte fractions were differentially stimulated (PMN with 1 mg and 5 mg LPS, PBMC with 10 ng TSST-1 or 2 mg ConA) in the presence or absence of heparins (1 U/ ml, 2 U/ ml, and 4 U/ ml) for 24 h at 378 C. Cytokine release was analyzed by ELISA. Certoparin but not UFH led to a dose-dependent increase in IL-6 from non-stimulated PBMC. In contrast, the release of IL-1ra, IL-10, and IL-12p40 was not modulated by heparins in a dose-dependent fashion. Increases in these cytokines occurred only as single incidents at intermediate heparin levels. An influence of the heparins on the apoptosis of PMN (measured as DNA-fragmentation in non-stimulated or LPS-stimulated cell-fractions) was not observed. KEY WORDS: Heparin; cytokine; leukocytes; neutrophils.

INTRODUCTION

hesion to endothelial cells, inhibition of L- and P-selectin expression or inhibition of reactive oxygen species generation (3, 4, 5). These anti-inflammatory activities of heparins may be related to the modulation of cytokine synthesis. The influence of heparins on cytokine generation was shown using isolated leukocyte fractions or whole blood assays (6, 7, 8, 9, 10). In these studies inconsistant results were reported, especially for the generation of proinflammatory TNF-a. Thus, it was the purpose of this study to analyze the influence of heparins (UFH and certoparin) on the generation of cytokines with known anti-inflammatory activities (IL-1ra, IL-6, IL-10), and of IL-12p40 from human leukocyte fractions.

Anticoagulant therapy by heparin is extensively used in clinical settings for reduction of thromboembolic complications in surgery. Main target of heparin action is its ability to potentate the activity of the endogenous coagulation cofactor antithrombin-III (1). Beneath unfractionated heparin (UFH) low molecular weight heparins (LMWH, e.g. certoparin) became availabe for clinical use (2). Substitution of LMWH for UFH decreases the incidence of heparin-induced thrombocytopenia and may be related to a lower hemorrhagic risk (1, 2). In addition to its anticoagulant activities there is accumulating evidence that heparins significantly affect inflammatory processes. It was reported that heparins exert anti-inflammatory activities such as inhibition of leukocyte ad-

MATERIALS AND METHODS Materials

1 Department

of Surgery, BG Kliniken Bergmannsheil, Bochum, Germany. 2 Novartis Pharma, Nu ¨ rnberg, Germany. 3 To whom correspondence should be addressed.

Histopaque 1119, Histopaque 1077, RPMI1640 medium supplemented with L-glutamine, toxic shock syn-

331 0360-3997/ 01/ 1000-0331/ 0  2001 Plenum Publishing Corporation

332 drome toxin-1 (TSST-1), lipopolysaccharide (LPS) from Escherichia coli (O55:B5), concanavalin-A (ConA), N(2-hydroxyethyl)-piperazine-N′ -(2-ethanesulfonic acid), and Tween 20 were obtained from Sigma (Deisenhofen, Germany). Protease-, peroxidase-, alkaline phosphatasefree bovine serum albumine (BSA, fraction V) was from Serva (Heidelberg, Germany). Propidium iodide was obtained from Molecular Probes Inc., Eugene, Oregon. Fetal calf serum (FCS) was from Life Technologies (Eggenstein, Germany). Phosphate-buffered saline (PBS, 10 mM Na2 PO4 / 138 mM NaCl/ 2.7 mM KCl, pH 7.2) was from Sigma. Triton-X-100 was provided from Fluka BioChemika, Deisenhofen, Germany. Certoparin sodium (LMWH) was obtained from Novartis Pharma (Nu¨ rnberg, Germany) and heparin sodium (unfractionated heparin, UFH) was from Ratiopharm (Ulm, Germany). Isolation and Stimulation of Leukocyte Fractions Polymorphonuclear neutrophil leukocytes (PMN) and peripheral blood mononuclear cells (PBMC) were isolated by a one step procedure based on a discontinuous double Ficoll-gradient described by English and Andersen (11). Briefly, EDTA-anticoagulated peripheral blood (9 ml MonovetteTM, Sarstedt, Nu¨ rnbrecht, Germany) obtained from healthy volunteers were diluted with an equal volume of 0.9% NaCl and carefully overlain on a double-gradient formed by layering 10 ml of polysucrose/ sodium diatrizoate adjusted to a density of 1.077 g/ ml (Histopaque 1077) on 10 ml Histopaque 1119 in 50 ml Falcon tubes (Becton Dickinson, Heidelberg, Germany). The tubes were subsequently centrifugated at 700 × g for 30 min at room temperature. After centrifugation two distinct leukocyte cell layers (PBMC and PMN including eosinophil granulocytes) were obtained above the bottom sediment of erythrocytes. The cell layers were carefully aspirated and both PMN and PBMC were transfered to separate 50 ml-tubes which were subsequently filled with PBS. Centrifugation followed at 200 × g for 15 min at 48 C. After this first washing contaminating erythrocytes within the PMN-fraction were removed by hypotonic lysis using 0.3% NaCl for 2 min at room temperature. After reconstitution of physiological osmotic strength, cells were washed again with PBS. This method led to more than 95% pure and viable neutrophils including eosinophils. Cell counting was performed using Tuerk staining solution (Sigma-Aldrich Chemie, Deisenhofen, Germany). Viability was measured by trypan blue (SigmaAldrich) exclusion test. Differential cell counts were

Ko¨ ller, Kutscha-Lissberg, Brom, Weidinger, and Muhr performed by a modified Pappenheim staining (MayGru¨ nwald solution and Giemsa solution, Sigma-Aldrich) of respective cell smears. Isolated cells were adjusted to 1 × 106 cells/ ml RPMI 1640 supplemented with L-glutamine (0.3 g/ L), sodium bicarbonate (2.0 g/ L), 10% fetal calf serum (FCS), and 20 mM N-(2-hydroxyethyl)-piperazine-N′ (2-ethanesulfonic acid). Leukocyte fractions were differentially stimulated in the presence or absence of different concentrations of heparins (1 U/ ml, 2 U/ ml, and 4 U/ ml). Final dilutions of heparins were made with RPMI1640. The concentrations of heparins were chosen according to a theoretical model of dilution of a therapeutic heparin dose in a distribution volume of 70 ml/ kg. Incubations of cells without heparins were performed by the addition of an adjusted volume RPMI1640. PBMC were stimulated with TSST-1 (10 ng/ ml) and ConA (2 mg/ ml). PMN were stimulated with LPS (1 mg/ ml and 5 mg/ ml) for 24 h at 378 C in a humidified atmosphere (5% carbon dioxide) in 24-well tissue culture plates (Becton Dickinson). Due to the different numbers of isolated cells the experimental replications ranged from n c 10 (PMN) to n c 16 (PBMC). Supernatants were harvested and subsequently centrifugated at 2000 × g for 2 min at room temperature. Resulting supernatants were stored at − 708 C. ELISA-System Mouse monoclonal antibodies (capture antibodies) and biotinylated goat polyclonal antibodies (detection antibodies) as well as recombinant human cytokines (standards) were supplied by R & D Systems (Wiesbaden, Germany). Volumina of 100 ml were used throughout ELISA procedures except 200 ml for blocking solution (PBS/ containing 1% BSA) and washing buffer (PBS/ Tween 0.05%). Cytokines were quantitated following the manufacturer’s ELISA-protocol. The capture antibodies were diluted in PBS and added to 96-well ELISA plates (MaxiSorp, Nunc, Roskilde, Danmark). The plates were incubated overnight at room temperature and then washed three times with wash buffer (AM60 Plate Washer, Dynex Technologies, Denkendorf, Germany). After addition of samples and standards, the plates were left at 2 h at room temperature on a shaker (200 shakes/ min), washed three times, and the biotinylated detection antibodies (diluted with PBS-Tween 0.05%) were added. After a subsequent three times washing step streptavidin-peroxidase (S5512, Sigma) was used as a 400-fold dilution in PBS of a stock

333

Heparin and Cytokine Release solution (0.1 mg/ ml PBS) for 20 min at room temperature. After three times washing the ELISA-microtiter plates were developed with 3,3′ ,5,5′ -tetramethylbenzidine (TMB) liquid substrate system (Sigma). The reaction was stopped by the addition of 1 M H3 PO4 . ELISA-microtiter plates were read by an ELISA reader (MRX Revelation, Dynex Technologies, Denkendorf, Germany) set to 450 nm. Quantitation of cytokine concentrations was performed by the use of recombinant human cytokines as standards. The sensitivities of the immunoassays were 25 pg/ ml (IL-1ra), 0.6 pg/ ml (IL6), 6 pg/ ml (IL-10) and 7 pg/ ml (IL-12p40).

in phosphate-buffered saline containing 0.1% Triton X-100). Fluorescence signals were measured using a flow cytometer (FACSCalibur, Becton Dickinson, Heidelberg, Germany) and for each sample data of 10.000 acquired cells were analyzed (CELLQuestTM 1.2.2 Becton Dickinson). The flow rate and all instrument settings were maintained unchanged for all measurements. Fluorescence signals of the freshly isolated PMN from each donor were used to determine the signal threshold of unfragmentated DNA. Fluorescence signals below unfragmentated DNA were gated and expressed as percentage of cells undergoing DNA-fragmentation.

Analysis of DNA-Fragmentation

Statistical Analysis

DNA-fragmentation of PMN was analyzed using flow cytometry after DNA-staining with propidium iodide as we decribed previously (12). Briefly, after the incubation period of 24 h with and without heparins the neutrophils were centrifuged at 200 g for 10 min at 48 C. The pellet was gently resuspended in 500 ml fluorochrome solution (50 mg/ ml propidium iodide

Differences between cytokine release and DNAfragmentation from leukocyte fractions in the presence of various heparin concentrations were determined by Wilcoxon matched pairs test after assessment of distribution characteristics of values using KomolgorovSmirnov test (STATISTICA 5.0, StatSoft Inc., Tulsa, USA). The minimum level of significance was p < 0.05.

Table 1. Influence of heparins on DNA-fragmentation and cytokine release from human polymorphonuclear granulocytes (PMN) isolated from different healthy donors (n). Data represent mean values ± standard deviation (SD). 1 × 106 cells/ ml supplemented RPMI 1640 were incubated in the presence or absence (None) of lipopolysaccharide (LPS) and in the presence or absence of different concentrations of unfractionated heparin (UFH) or certoparin (1 U/ ml, 2 U/ ml, or 4 U/ ml). Significant differences in cytokine release in the absence or presence of respective heparins , *( p < 0.05). % of PMN, percentage of cells undergoing DNA-fragmentation. Significant differences in DNA-fragmentation between non-stimulated and LPS-stimulated PMN, ( p < 0.05).

334 RESULTS Influence of Heparins on the Generation of IL-1ra from Isolated Polymorphonuclear Granulocytes (PMN) PMN have long been considered as the major leukocyte subpopulation involved in acute inflammatory processes. However, the capacity of PMN to produce large quantities of IL-1ra beneath IL-8 (13) indicates that PMN play an active role also in anti-inflammatory regulation. As is shown in Table 1 there are minor not significant increases in spontaneous as well as LPS-stimulated release of IL-1ra from PMN in the presence of certoparin and unfractionated heparin (UFH), which only reached statistical significance in the presence of certoparin (2 U). However, LPS-stimulated PMN synthesized significantly higher amounts of IL-1ra compared to non-stimulated cells (1 mg LPS p c 0.021; 5 mg LPS p c 0.002). Influence of DNA-Fragmentation on Cytokine Synthesis of PMN During the Culture Period As we and others described previously (12, 14) PMN undergo apoptosis during prolonged cultivation in medium which is not supplemented by e.g. granulocytemacrophage colony stimulating factor (GM-CSF). Thus, under the used experimental conditions ongoing apoptosis of PMN may influence the release capacity of PMN for IL-1ra in the presence and absence of heparins. Thus apoptosis was analyzed flow-cytometrically as cellular DNA-fragmentation. As is shown in Table 1 the number of PMN with positive DNA-fragmentation raised to 8.5 ± 3.8% (range 2.4 to 12.1%, n c 7) after the cultivation period in supplemented RPMI1640 compared to freshly isolated fractions. As was also described previously (14) this DNA-fragmentation significantly decreased for 49.3 ± 21.8% ( p < 0.02) when the cells were concomitantly stimulated with LPS (Table 1). However, an influence of the heparins on DNA-fragmentation in non-stimulated or LPS-stimulated PMN-fractions was not observed (Table 1). Influence of Heparins on the Generation of IL-1ra, IL-6, IL-10 and IL-12p40 from Isolated Peripheral Blood Mononuclear Cells (PBMC) In contrast to PMN, the PBMC produce a larger panel of cytokines. Table 2 (certoparin) and Table 3 (unfractionated heparin) summarize the obtained results on the influence of heparins for spontaneous or stimu-

Ko¨ ller, Kutscha-Lissberg, Brom, Weidinger, and Muhr lated release of cytokines from PBMC. Stimulation of PBMC was differentially performed either by the superantigen toxic shock syndrome toxin-1 (TSST-1) or by the lectin concanavalin A (ConA), which were typical experimentally used stimuli for lymphocytes (Con A) or lymphocytes/ monocytes (TSST-1). As expected the stimulation of the cells led to highly significant increases in cytokine synthesis compared to non-stimulated cells (for all cytokines p < 0.001 except IL-12p40 p < 0.003). However, modulation of cytokine formation in the presence of heparins was not influenced when the PBMC were concomitantly stimulated (Tables 2 and 3). Statistical calculation revealed increases only in TSST-1 induced IL-6 (UFH 1U) and IL-10 (certoparin, 4U). The spontaneous release of IL-1ra and IL-12p40 was not and the release of IL-10 was partly increased in the presence of certoparin and unfractionated heparin (UFH). In contrast, certoparin significantly led to an increase in IL6 from non-stimulated cells (Fig. 1). This was the only dose-dependent significant effect of heparins observed under the experimental conditions.

DISCUSSION In this study we have demonstrated that unfractionated heparin and certoparin only marginally modulate the generation of cytokines from isolated leukocyte fractions except an increase in IL-6 from PBMC induced by certoparin, which was dose-dependent in a monotonic fashion. Significant calculated increases in other cytokines may be spurious incidents since the statistically significant points occurred at intermediate heparin levels rather than the highest heparin level with partly fractionally differences to background levels. However, inflammation and coagulation are clearly linked by different molecular signals. Inflammation promotes coagulation by leading to intravascular tissue factor expression, eliciting the expression of leukocyte adhesion molecules on the intravascular cell surfaces, and down regulating the fibrinolytic pathways. Thrombin, in turn, can promote inflammatory responses (15, 16). We focussed on the synthesis of cytokines with known or presumed anti-inflammatory activities (IL-1ra, IL-6, IL-10, IL-12p40). IL-1ra (IL-1 receptor antagonist) is a member of the IL-1 gene family and occupancy of the IL-1 receptor I results in effective prevention of IL-1 signal transduction (17). IL-6 is a pleiotropic cytokine involved in various immune responses and hematopoiesis. Although initially thought to be a proin-

335

Heparin and Cytokine Release

Table 2. Influence of certoparin on cytokine release from human peripheral blood mononuclear cells (PBMC) isolated from different healthy donors (n). Data represent mean values ± standard deviation (SD). 1 × 106 cells/ ml supplemented RPMI 1640 were incubated in the presence or absence (None) of different stimuli and in the presence or absence of different certoparin concentrations (1 U/ ml, 2 U/ ml, or 4 U/ ml). Significant differences in cytokine release in the absence or presence of certoparin are indicated by asterics (*, p < 0.05). TSST-1, Toxic shock syndrome toxin-1; ConA, Concanavalin A.

flammatory cytokine, recent findings (for summary see 18) suggest that IL-6 has additionally anti-inflammatory and immunosuppressive effects (e.g. stimulation of most acute phase proteins or induction of IL-1ra). IL-10 was

discovered as a cytokine synthesis inhibitory factor and leads to the suppression of chemokines, proinflammatory cytokines (e.g. IL-1 and TNF-a) or granulocyte survival, thus limiting the duration of inflammatory pro-

Table 3. Influence of unfractionated heparin (UFH) on cytokine release from human peripheral blood mononuclear cells (PBMC) isolated from different healthy donors (n). Data represent mean values standard deviation (SD). 1 × 106 cells/ ml supplemented RPMI 1640 were incubated in the presence or absence (None) of different stimuli and in the presence or absence of different UFH concentrations (1 U/ ml, 2 U/ ml, or 4 U/ ml). Significant differences in cytokine release in the absence or presence of UFH are indicated by asterics (*, p < 0.05). TSST-1, Toxic shock syndrome toxin-1; ConA, Concanavalin A.

336

Ko¨ ller, Kutscha-Lissberg, Brom, Weidinger, and Muhr

Fig. 1. Influence of certoparin on the release of IL-6 from human peripheral blood mononuclear cells (PBMC) isolated from different healthy donors (n). Data represent mean values ± standard deviation (SD). 1 × 106 cells/ ml supplemented RPMI 1640 were incubated in the presence or absence of different certoparin concentrations (1 U/ ml, 2 U/ ml, or 4 U/ ml). Significant differences in cytokine release in the absence or presence of certoparin are indicated by asterics (*, p < 0.05).

cesses (19). Whereas the biological active IL-12p70 has potent immunoregulatory and proinflammatory activities the role of human IL-12p40 is still not clear. However, in murine models the homodimer IL-12p40 serves as a functional antagonist to the IL-12p70 heterodimer (20). Heparins are characterized by specific binding sites for proteins such as antithrombin III and by nonspecific electrostatic binding to a variety of proteins depending upon their polyanionic nature. Obviously, the latter type of binding may be involved in anti-inflammatory effects of heparins (4, 5, 21, 22). However, it have been also reported that the release of proinflammatory cytokines such as TNF-a from LPS-stimulated leukocytes or whole blood was increased in vitro in the presence of heparin (6, 7, 8). On the other hand in a porcine model of endotoxinemia it was shown that heparin pretreatment of animals exert protective effects and attenuated LPS-mediated TNF-activity obviously by increasing the concentration of circulating soluble TNF-receptors, which are naturally occurring inhibitors of TNF-activity (21). Thus, the relative concentrations of pro- and anti-inflammatory signals will finally determine the modulation of inflammatory reactions in the microenvironment. One may suggest that LMWH have fewer effects on immunological processes due to fewer specific and nonspecific binding sites (2). However, we observed a pro-

nounced response of certoparin on IL-6 generation when same units of UFH and certoparin were compared. Any influence of LPS contaminations could be excluded since the heparin diluent was the same as for UFH and certoparin was certified LPS negative. One possible explanation for this increase in IL-6 may be the binding of IL6 to the heparin molecules. It has been reported recently, that IL-6 is heparin-binding protein (23). The binding of released cytokines may protect them from proteolytic degradation. However, stabilisation of cytokine halflife is obviously not the sole effect of heparins on cytokines. The binding of heparin alter the secondary and tertiary structure of cytokines and inhibit binding to respective cytokine receptors (24), thus, the heparins additionally influence the biological activity of distinct cytokines. There is accumulating evidence that heparins modulate inflammatory reactions by different mode of actions. From the experimental design of our study it was not possible to relate anti-inflammatory activities of therapeutical administered heparins to increased generation of anti-inflammatory cytokines. Acknowledgments—This work was supported by a Grant from the Wissenschaftskommission of the BG Kliniken Bergmannsheil, Bochum, Germany. We are indepted to Mrs. Peter for excellent technical assistance.

Heparin and Cytokine Release REFERENCES 1. Fareed, J., D. A. Hoppensteadt, and R. L. Bick. 2000. An update on heparins at the beginning of the new millennium. Semin. Thromb. Hemost. 26 (Suppl 1):5–21 2. Weitz, J. I. 1997. Low-molecular-weight heparins. New Engl. J. Med. 337:688–698. 3. Lever, R., J. R. Hoult, and C. P. Page. 2000. The effects of heparin and related molecules upon the adhesion of human polymorphonuclear leucocytes to vascular endothelium in vitro. Br. J. Pharmacol. 129: 533–540. 4. Donata, P., T. Qutob, W. Hamouda, F. Bakri, A. Aljada, and Y. Kumbkarni. 1999. Heparin inhibits reactive oxygen species generation by polymorphonuclear and mononuclear leucocytes. Thromb. Res. 96:437–443. 5. Nelson, R. M., O. Cecconi, W. G. Roberts, A. Aruffo, R. J. Linhardt, and M. P. Bevilacqua. 1993. Heparin oligosaccharides bind L- and P-selectin and inhibit acute inflammation. Blood 82:3253–3258. 6. McBride, W. T., M. A. Armstrong, and T. J. McMurray. 1996. An investigation of the effects of heparin, low molecular weight heparin, protamine, and fentanyl on the balance of pro- and antiinflammatory cytokines in in-vitro monocyte cultures. Anaesthesia 51:634–640. 7. Heinzelmann, M., M. Miller, A. Platz, L. E. Gordon, D. O. Herzig, H. C. Polk Jr. 1999. Heparin and enoxaparin enhance endotoxininduced tumor necrosis factor-alpha production in human monocytes. Ann. Surg. 229:542–550. 8. Call, D. R., and D. G. Remick. 1998. Low molecular weight heparin is associated with greater cytokine production in a stimulated whole blood model. Shock 10:192–197. 9. Baram, D., M. Rashkovsky, R. Hershkoviz, I. Drucker, T. Reshef, S. Ben-Shitrit, and Y. A. Mekori. 1997. Inhibitory effects of low molecular weight heparin on mediator release by mast cells: preferential inhibition of cytokine production and mast cell-dependent cutaneous inflammation. Clin. Exp. Immunol. 110:485–491. 10. Engstad, C. S., T. J. Gutteberg, and B. Osterud. 1997. Modulation of blood cell activation by four commonly used anticoagulants. Thromb. Haemost. 77:690–696. 11. English, D., and B. R. Andersen. 1974. Single step separation of red blood cells, granulocytes and mononuclear leukocytes on discontinuous density gradient of ficoll-hypaque. J. Immunol. Methods 5:249–252.

337 12. Ko¨ ller, M., P. Wachtler, A. David, G. Muhr, and W. Ko¨ nig. 1997. Arachidonic acid induces DNA-fragmentation in human polymorphonuclear neutrophil granulocytes. Inflammation 21:463– 474. 13. Altstaedt, J., H. Kirchner, and L. Rink. 1996. Cytokine production of neutrophils is limited to interleukin-8. Immunology 89:563– 568. 14. Sweeney, J. F., P. K. Nguyen, G. M. Omann, and D. B. Hinshaw. 1997. Ultraviolet irradiation accelerates apoptosis in human polymorphonuclear leukocytes: protection by LPS and GM-CSF. J. Leuko. Biol. 62:517–523. 15. Esmon, C. T., K. Fukudome, T. Mather, W. Bode, L. M. Regan, D. J. Stearns-Kurosawa, and S. Kurosawa. 1999. Inflammation, sepsis, and coagulation. Haematologica 84:254–259. 16. Hoffman, M., and S. T. Cooper. 1995. Thrombin enhances monocyte secretion of tumor necrosis factor and interleukin-1 beta by two distinct mechanisms. Blood Cells Mol. Dis. 21:156–167. 17. Dinarello, C. A. 1998. Interleukin-1. In: The Cytokine Handbook. Ed. A. Thomson. Academic Press, San Diego, pp. 35–72. 18. Tilg, H., C. A. Dinarello, and J. W. Mier. 1997. IL-6 and APPs: anti-inflammatory and immunosuppressive mediators. Immunol. Today 9:428–432. 19. De Waal Malefyt, R., H. Yssel, M. G. Roncarolo, H. Spits, and J. E. De Vries. 1992. Interleukin-10. Curr. Opinion Immunol. 4:314–320. 20. Storkus, W. J., H. Tahara, and M. T. Lotze. 1998. Interleukin-12. In: The Cytokine Handbook. Ed. A. Thomson. Academic Press, San Diego, pp. 391–426. 21. Darien, B. J., J. Fareed, K. S. Centgraf, A. P. Hart, P. S. MacWilliams, M. K. Clayton, H. Wolf, and K. T. Kruse-Elliott. 1998. Low molecular weight heparin prevents the pulmonary hemodynamic and pathomorphologic effects of endotoxin in a porcine acute lung injury model. Shock 9:274–281. 22. Schwartz, D., D. Engelhard, R. Gallily, I. Matoth, and T. Brenner. 1998. Glial cells production of inflammatory mediators induced by Streptococcuspneumoniae: inhibition by pentoxifylline, lowmolecular-weight heparin and dexamethasone. J. Neurol. Sci. 18:13–22. 23. Mummery, R. S., and C. C. Rider. 2000. Characterization of the heparin-binding properties of IL-6. J. Immunol. 165:5671–5679. 24. Balasubramanian, V., and M. Ramanathan. 2000. Glycosaminoglycans alter the conformation of interferon-gamma. Cytokine 12:466–471.