Activated protein C inhibits tumor necrosis factor and macrophage migration inhibitory factor production in monocytes

ARTICLE

INTRODUCTION

The molecular sequelae of events leading to the adverse and sometimes lethal outcome of sepsis have not been fully elucidated [1]. Cytokines are undoubtedly involved in these processes [2-4] with the pro-inflammatory cytokines tumor necrosis factor-alpha (TNF), interleukin-1, macrophage migration inhibitory factor (MIF), and others playing a pivotal role [4-8]. Less is known, however, about the role of the anti-inflammatory cytokines and, also, the interplay between the various cytokines and the regulatory loops involved is still poorly defined. Cytokines have been considered promising clinical targets in the treatment of septic shock, but to date, anti-cytokine-based therapeutic strategies such as the use of anti-TNF antibodies, soluble TNF receptors, or interleukin-1 receptor antagonists have failed to show a clear clinical benefit [9].

Activated protein C (APC), mostly known because of its powerful anti-coagulant properties [10, 11], has recently been shown to exhibit anti-inflammatory and probable anti-sepsis activities, and APC or its precursor protein C have been implicated as therapeutical agents for the treatment of meningococcaemia [12-14].

Protein C is a plasma serine protease proenzyme that is produced in the liver and by endothelial cells and that is secreted into the circulation following various stimuli [10]. Cleavage by thrombin of a dodecapeptide from the N-terminal end of the protein C heavy chain results in the generation of APC. APC acts to promote conversion of factors Va and VIIIa to factors Vi and VIIIi, respectively, by proteolysis, thereby inhibiting the blood clotting process [10, 15, 16].

Several lines of evidence have implicated protein C/APC to be an endogeneous inhibitor of the inflammatory septic cascade. Acquired severe protein C deficiency is a poor prognostic marker in severely infected septic patients [17, 18]. In meningococcal septic shock in children, clinical severity has been directly correlated with the reduction in circulating levels of protein C [17, 19], and protein C replacement in this condition not only led to normalization of plasma protein C levels and other parameters of haemostasis, but has also resulted in significant improvement of organ dysfunction, such as adult respiratory distress syndrome (ARDS), and improvement of skin lesions [13, 20, 21]. Recently, protein C replacement therapy has been reported to markedly improve survival rates in patients with meningococcus-induced purpura fulminans [14]. In models of sepsis, preloading baboons with APC prevented the coagulopathic response and lethal effects of E. coli injections [22]. These and other studies have demonstrated that the anticoagulant and fibrinolytic properties of protein C/APC most likely contribute to its anti-inflammatory effects.

Evidence that APC has a direct anti-inflammatory action is less clear-cut. Infusions of APC have been shown to suppress the peak blood levels of TNF in rats challenged with high doses of lipopolysaccharide (LPS) [23]. In vitro studies suggest that APC directly inhibits TNF production from LPS-, interferon-gamma, and phorbol ester-stimulated monocytes [24]; and although these observations have remained controversial [25], a non-coagulation-related, direct anti-inflammatory action of APC has also been argued for by investigations into the relationship between coagulation and sepsis. For example, human APC is only a weak anticoagulant in rats [26, 27], yet it exhibits anti-sepsis effects in these animals. Inhibitors of coagulation do not show anti-sepsis effects comparable to APC itself [28].

MIF was previously known as a classical T cell cytokine that acts to inhibit the migration of monocytes/macrophages [29, 30], but it has recently been shown to be an abundant cytokine with a broad spectrum of proinflammatory properties, and to be a critical mediator of Gram-negative and Gram-positive bacterial shock [5, 7, 8, 31, 32]. MIF is unique among cytokines in that it constitutes a physiological, counter-regulatory system of the immunosuppressive and anti-inflammatory actions of glucocorticoids [31, 33] and it exhibits both cytokine and catalytic activities [34, 35]. Of note, MIF is found in the alveolar spaces of patients with ARDS, it augments pro-inflammatory cytokine secretion in ARDS, and it overrides the anti-inflammatory effects of glucocorticoids in ARDS lungs [33, 36]. ARDS is known as one of the detrimental downstream events following initiation of the septic cascade and has been closely attributed to the previous activation of the extrinsic pathway of coagulation leading to the formation of microthrombi and fibrin deposition.

In this study, we sought to establish the direct interference of APC with cytokine-mediated toxicity. Because TNF is the pivotal mediator of septic shock and to clarify the existing but controversial results available, we first investigated directly the effect of APC on TNF production by monocytes in vitro. To begin to examine whether APC action interferes with other cytokine mediators that promote the septic cascade, we analyzed potential direct effects of APC on MIF.

MATERIALS AND METHODS

Materials

APC (lot PCA 162) was kindly provided by Baxter Hyland Immuno (Vienna, Austria) and was reconstituted as recommended by the supplier. The APC provided was generated from protein C as follows: protein C was isolated from human plasma by immunoaffinity chromatography and APC was subsequently prepared by activation of protein C with human thrombin, followed by purification by ion exchange chromatography. Protein C and APC were certified as sterile and pyrogen-free by the manufacturer. Structural integrity and native folding of APC was confirmed by far-UV CD spectropolarimetry. Activity of APC preparations was tested by the Immunochrom^® PC assay essentially as described by the manufacturer. Lipopolysaccharide (LPS) was of the type O111:B4 and was from Sigma-Aldrich GmbH (Deisenhofen, Germany). Monoclonal anti-human TNF antibody was from R&D Systems (Wiesbaden, Germany) and neutralizing polyclonal anti-MIF antibody was prepared from rabbit serum as described [5]. Cell culture reagents were from Life Technologies (Eggenstein, Germany) and miscellaneous chemicals from Sigma-Aldrich GmbH or Life Technologies and were of the highest analytical grade available.

Cell culture

THP-1 monocytic cells were obtained from the German collection for microbiology and cell lines (DSMZ, Braunschweig, Germany) and were cultured at a density of 0.1-0.5 x 10⁶ cells/ml in RPMI 1640 containing 10% fetal calf serum (FCS), 2 mM L-glutamine, and penicillin-streptomycin in 5% CO₂ at 37° C. For experiments, cells were washed twice in the same medium without FCS and were plated in 6- or 24-well cell culture plates (Greiner Labortechnik, Frickenhausen, Germany) at a density of 1 x
10⁶ cells/ml in serum-free medium. Cells were rested for 3 hours. LPS stimulations were performed for 4-23 hours as indicated. APC was always added 30 min before the stimulant.

ELISA

At the end of the incubations, reactions were transferred to 1.5 ml centrifuge tubes, centrifuged gently at 400 g for 5 min, and culture supernatants transferred to fresh tubes. Supernatants were then either analyzed directly or stored at ­ 80° C for up to 3 months. Antibodies for the human TNF and human MIF enzyme-linked immunosorbent assays (ELISA) were from R&D Systems and ELISA were performed following the manufacturer's instructions. To confirm that APC proteolytic activity did not affect test protein stability and detectability, TNF ELISA standards were preincubated with 50 mug/ml APC and standard curves determined in the presence versus absence of APC. Identical TNF standard values were obtained for both curves. ELISA data are expressed as the mean ± SD of the indicated number of experiments.

RESULTS

Dependence of APC activity on serum concentration

Various preparations of APC have been used in previous studies and their quality and status of biological activity have not always been clear. Thus, we initially confirmed that the APC preparation we used was natively folded and biologically active. APC exhibited full enzymatic activity as measured by the amidolytic activity assay (data not shown), but failed to have any inhibitory effect on TNF production in vitro, when tested on peripheral blood monocytes or various monocytic cell lines using standard conditions (data not shown). Detailed analysis of assay parameters then showed that APC inhibition of TNF production was abolished in the presence of FCS at concentrations as low as 1%, while its enzymatic activity was unaffected under such conditions (data not shown). Serum-mediated inhibition of the TNF-inhibitory activity of APC, i.e. as observed in the presence of 10% FCS, could only be measurably overcome when the concentration of APC was increased to beyond physiological values of >= 200 mug/ml, which are unsuitable to be handled in routine in vitro settings.

Inhibition by APC of LPS-induced TNF secretion

When incubations were performed under serum-free conditions, APC potently inhibited TNF production in THP-1 monocytes in a dose- and time-dependent fashion (Figure 1). For example, APC at a concentration of 50 mug/ml inhibited TNF release by more than 60%. Varying stimulatory LPS concentrations from 0.1-10 mug/ml had no significant influence on the APC effect (data not shown). Together, these data confirmed previous results by Hancock and coworkers [24] suggesting that APC can act to directly inhibit TNF production in monocytes and implying that conflicting results regarding this activity may have come from different concentrations or properties of the serum supplement used during the incubations.

MIF secretion by THP-1 monocytes

Although TNF is a pivotal cytokine mediator of septic shock, the demonstration that APC has a direct anti-inflammatory action should be sustained by the observation of the down-regulation of other pro-inflammatory mediators of sepsis. Hence, we investigated whether APC could also block LPS-stimulated production of MIF, another critical mediator of septic shock that is distinct from TNF and other cytokines in that it may not act by signalling through classical receptor-mediated pathways [38-40], and in that it does not appear to follow classical secretion pathways [41]. THP-1 monocytes were stimulated with LPS as established for TNF, and production of MIF was measured by ELISA. Secretion of MIF increased over time, but unlike secretion of TNF, it plateaued at approximately 20 hours (Figure 2). Of note, resting monocytes also secreted MIF in a time-dependent manner. This latter observation was not due to serum starvation, stress, or non-specific release following cell death, as confirmed by lactate dehydrogenase (LDH) assay performed in the culture supernatants of the analyzed cells (data not shown). As with TNF, the secretion of MIF was found to gradually increase as the concentration of LPS increased and was specific up to at least 10 mug/ml of LPS. MIF concentrations measured after a 10 hours stimulation period ranged from 7.7 ± 1.8 ng/ml (resting conditions) to 44.9 ± 7.8 ng/ml (10 mug/ml LPS).

Inhibition by APC of LPS-induced MIF secretion

As for TNF, APC at a concentration of 50 mug/ml, markedly inhibited secretion of MIF, when THP-1 cells were stimulated with a standard concentration of LPS (10 mug/ml) for 23 hours (Figure 3). Inhibition by APC was also evident, when cells were stimulated for 4 hours only (data not shown) or with 100 ng/ml LPS (Figure 3). Under the latter conditions, inhibition by APC was complete, i.e. MIF secretion was reduced to levels measured for resting control cells. Direct comparison of the APC effect on TNF versus MIF release showed that secretion of MIF was inhibited to a somewhat higher degree (68 ± 5% versus 43 ± 7%).

We then examined whether APC inhibition of MIF secretion was dependent on TNF. LPS-induced release of MIF was not affected by neutralizing anti-TNF antibodies that led to a complete reduction in TNF levels and, under the conditions used, anti-MIF antibodies did not affect secretion of TNF either (data not shown).

DISCUSSION

Evidence for a direct anti-inflammatory effect of APC and the physiological significance of such observations had been controversial [23-28]. Our results clearly confirm previous findings by Grey et al. that had demonstrated a selective inhibitory effect of APC on the responses of activated monocytes [24]. How-ever, our studies also imply that serum-based inhibitors of APC anti-inflammatory activity may exist that may have interfered with APC activity in earlier experiments. Using established experimental APC concentrations, no inhibitory effect of APC on TNF secretion in monocytes was observed when 1-10% FCS was included in the culture media. Serum inhibition of APC activity could only be overcompensated when extremely high APC concentrations were applied. Although our data do not allow for any quantitations of serum inhibitor/APC ratios or affinities due to the in vitro conditions used, i.e. fetal calf rather than human serum was used, they should encourage further investigations into the molecular nature of the proposed serum inhibitor of APC. The observed serum inhibitory effect corresponds well with the fact that low PC levels are a predictor of meningococcal septic shock.

The observation that APC also markedly inhibited the secretion of MIF, another key cytokine mediator of septic shock [5, 7, 8], further confirmed the significance of the anti-inflammatory activity of APC. In fact, anti-MIF antibodies have recently been shown to potently block bacterial septic shock in an E. coli pe-ritonitis and cecal ligation and puncture (CLP) model and anti-MIF antibodies are currently considered unique among potential anti-cytokine antibody-based therapeutic strategies in that they confer protection against septic shock even when administered several hours post-challenge [8].

MIF has been suggested to promote ARDS [33], while protein C replacement therapy has been found to result in improvement of ARDS lung conditions [13, 20, 21]. Thus, inhibition of MIF secretion by APC offers an intruiging molecular pathway that could be part of the yet to be defined regulatory mechanism underlying anti-inflammatory APC effects in disease.

We found that LPS-induced secretion of MIF was fully independent on the presence of TNF and vice versa. This indicated that inhibition by APC of MIF and TNF release were either independent events or that inhibition was due to interference with a mutually utilized upstream signalling event (Scheme 1). The latter argument would be consistent with the notion that TNF and MIF, although showing several similar and overlapping pro-inflammatory activities, belong to different protein families with different mechanisms of production, secretion, and target cell action (summarized in: [39]).

Preliminary data from ongoing investigations on potential effects of APC on cellular signalling pathways that are initiated following cellular activation by LPS would suggest that inhibition of pro-inflammatory cytokine release by APC is due to interference with the NFkappaB/IkappaB transcription factor pathway. Of note, NFkappaB-mediated activation has been demonstrated to be involved in the production of TNF and a NFkappaB element has also been identified in the 5'-upstream regulatory region of the mouse MIF gene [42].

CONCLUSION

In summary, we have shown that APC can act to directly down-regulate the secretion of two critical and mechanistically distinct cytokines mediators of septic shock, indicating that inhibition by APC could represent a more general anti-inflammatory principle.

Acknowledgements. We thank H. P. Schwarz for supplying protein C, APC, and Immunochrom kits. We thank E. Wagner and D. Finkelmeier for assistance with the cell culture and ELISA and H. Brunner, T. Calandra, and O. Flieger for helpful discussions.

REFERENCES

2. Gutierrez-Ramos J C, Bluethmann H. 1997. Molecules and mechanisms operating in septic shock: lessons from knockout mice. (Review) Immunol. Today 18: 329.

3. Henderson B, Poole S, Wilson M. 1998. Bacteria-Cytokine interaction in Health and Disease. Cambridge University Press, Cambridge, England.

4. Tracey K J, Abraham E. 1998. From mouse to man: or what have we learned about cytokine-based anti-inflammatory therapies? (Review) Shock 11: 224.

5. Bernhagen J, Calandra T, Mitchell R A, Martin S B, Tracey K J, Voelter W, Manogue K R, Cerami A, Bucala R. 1993. MIF is a pituitary-derived cytokine that potentiates lethal endotoxaemia. Nature 365: 756.

6. Beutler B, Bazzoni F. 1998. TNF, apoptosis and autoimmunity: a common thread? (Review) Blood Cells Mol. Dis. 24: 216.

7. Bozza M, Satoskar A R, Lin G, Lu B, Humbles A A, Gerard C, David J R. 1999. Targeted disruption of migration inhibitory factor gene reveals its critical role in sepsis. J. Exp. Med. 189: 341.

8. Calandra T, Echtenacher B, Le Roy D, Pugin J, Metz C N, Hültner L, Heumann D, Männel D, Bucala R, Glauser M. 1998. Protection from septic shock by neutralization of macrophage migration inhibitory factor (MIF). Nat. Med. 6: 164.

9. Vincent J L. 1997. New therapies in sepsis. (Review) Chest 112: 330S.

10. Cooper D N, Krawcyak M. 1997. In: Protein C and protein C deficiency. Venous thrombosis: from genes to clinical medicine. Cooper D N, Krawcyak M, eds. BIOS Scientific Publishers Ltd, Oxford, 61.

11. Esmon C T, Schwarz H P. 1995. An update on clinical and basic aspects of the protein C anticoagulant pathway. TCM 5: 141.

12. Charlton R. 1998. Protein-C concentrate for meningococcal purpura fulminans. (letter) Lancet 351: 987.

13. Rivard G E, David M, Farrell C, Schwarz H P. 1995. Treatment of purpura fulminans in meningococcemia with protein C concentrate. J Pediatr 126: 646.

14. Smith O P, White B, Vaughan D, Rafferty M, Claffey L, Lyons B, Casey W. 1997. Use of protein-C concentrate, heparin, and haemodiafiltration in meningococcus-induced purpura fulminans. Lancet 350: 1590.

15. Esmon C T. 1993. Molecular events that control the protein C anticoagulant pathway. (Review) Thromb. Haemost. 70: 29.

16. Esmon C T, Fukudome K. 1995. Cellular regulation of the protein C pathway. (Review) Semin. Cell Biol. 6: 259.

17. Fourrier F, Chopin C, Goudemand J, Hendrycx S, Caron C, Rime A, Marey A, Lestavel P. 1992. Septic shock, multiple organ failure, and disseminated intravascular coagulation. Compared patterns of antithrombin III, protein C, and protein S deficiencies. Chest 101: 816.

18. Hesselvik J F, Malm J, Dahlback B, Blomback M. 1991. Protein C, protein S and C4b-binding protein in severe infection and septic shock. Thromb. Haemost. 65: 126.

19. Fijnvandraat K, Derkx B, Peters M, Bijlmer R, Sturk A, Prins M H, van Deventer S J, ten Cate J W. 1995. Coagulation activation and tissue necrosis in meningococcal septic shock: severely reduced protein C levels predict a high mortality. Thromb. Haemost. 73: 15.

20. Rintala E, Seppala O P, Kotilainen P, Rasi V. 1996. Protein C in the treatment of coagulopathy in meningococcal disease. Lancet 347: 1767.

21. Rintala E, Seppala O P, Kotilainen P, Pettila V, Rasi V. 1998. Protein C in the treatment of coagulopathy in meningococcal disease. Crit. Care Med. 26: 965.

22. Taylor F B Jr, Chang A, Esmon C T, d'Angelo A, Vigano-d'Angelo S, Blick K E. 1987. Protein C prevents the coagulopathic and lethal effects of Escherichia coli infusion in the baboon. J. Clin. Invest. 79: 918.

23. Murakami K, Okajima K, Uchiba M, Johno M, Nakagaki T, Okabe H, Takatsuki K. 1997. Activated protein C prevents LPS-induced pulmonary vascular injury by inhibiting cytokine production. Am. J. Physiol. 272: 197.

24. Grey S T, Tsuchida A, Hau H, Orthner C L, Salem H H, Hancock W W. 1994. Selective inhibitory effects of the anticoagulant activated protein C on the responses of human mononuclear phagocytes to LPS, IFN-gamma, or phorbol ester. J. Immunol. 153: 3664.

25. Esmon C T. 1996. In: Anticoagulant properites of vascular cells: thrombomodulin and protein C activation pathways vascular control of hemostasis. Van Hinsberg V W, ed. Harwood academic publishers, Amsterdam, 9.

26. Ono M, Fujiwara H, Okafuji T, Enjoh T, Nawa K. 1994. Recombinant rat protein C: comparative studies of structure, function and synthesis with plasma protein C. Thromb. Haemost. 71: 54.

27. Pichler L, Schramm W, Ulrich W, Varadi K, Schwarz H P. 1995. Antinociceptive properties of protein C in a model of inflammatory hyperalgesia in rats. Thromb. Haemost. 73: 252.

28. Taylor F B Jr, Chang A C, Peer G T, Mather T, Blick K, Catlett R, Lockhart M S, Esmon C T. 1991. DEGR-factor Xa blocks disseminated intravascular coagulation initiated by Escherichia coli without preventing shock or organ damage. Blood 78: 364.

29. Bloom B R, Bennett B. 1966. Mechanism of a reaction in vitro associated with delayed-type hypersensitivity. Science 153: 80.

30. David J R. 1966. Delayed hypersensitivity in vitro: its mediation by cell-free substances formed by lymphoid cell-antigen interaction. Proc. Natl. Acad. Sci. USA 56: 72.

31. Calandra T, Bucala R. 1995. Macrophage migration inhibitory factor: a counter-regulator of glucocorticoid action and critical mediator of septic shock. J. Inflamm. 47: 39.

32. Calandra T, Spiegel L A, Metz C N, Bucala R. 1998. Macrophage migration inhibitory factor is a critical mediator of the activation of immune cells by exotoxins of gram-positive bacteria. Proc. Natl. Acad. Sci USA 95: 11383.

33. Donnelly S C, Haslett C, Reid P T, Grant I S, Wallace W A, Metz C N, Bruce L J, Bucala R. 1997. Regulatory role for macrophage migration inhibitory factor in acute respiratory distress syndrome. Nat. Med. 3: 320.

34. Kleemann R, Kapurniotu A, Frank R W, Gessner A, Mischke R, Flieger O, Juttner S, Brunner H, Bernhagen J. 1998. Disulfide analysis reveals a role for macrophage migration inhibitory factor (MIF) as thiol-protein oxidoreductase. J. Mol. Biol. 280: 85.

35. Rosengren E, Bucala R, Aman P, Jacobsson L, Odh G, Metz C N, Rorsman H. 1996. The immunoregulatory mediator macrophage migration inhibitory factor (MIF) catalyzes a tautomerization reaction. Mol. Med. 2: 143.

36. Makita H, Nishimura M, Miyamoto K, Nakano T, Tanino Y, Hirokawa J, Nishihira J, Kawakami Y. 1998. Effect of anti-macrophage migration inhibitory factor antibody on lipopolysaccharide-induced pulmonary neutrophil accumulation. Am. J. Repir. Crit. Care Med. 158: 573.

37. Kleemann R, Kapurniotu A, Mischke R, Held J, Bernhagen J. 1999. Characterization of catalytic centre mutants of macrophage migration inhibitory factor (MIF) and comparison to Cys81Ser MIF. Eur. J. Biochem. 261: 753.

38. Kleemann R, Mischke R, Kapurniotu A, Brunner H, Bernhagen J. 1998. Specific reduction of insulin disulfides by macrophage migration inhibitory factor (MIF) with glutathione and dihydrolipoamide: potential role in cellular redox processes. FEBS Lett. 430: 191.

39. Bernhagen J, Calandra T, Bucala R. 1998. Regulation of the immune response by macrophage migration inhibitory factor: biological and structural features. (Review) J. Mol. Med. 76: 151.

40. Swope M D, Sun H W, Klockow B, Blake P, Lolis E. 1998. Macrophage migration inhibitory factor: interactions with glutathione and S-hexylglutathione J. Biol. Chem. 273: 14877.

41. Bernhagen J, Calandra T, Bucala R. 1994. The emerging role of MIF in septic shock and infection. (Review) Biotherapy 8: 123.

42. Mitchell R, Bacher M, Bernhagen J, Pushkarskaya T, Seldin M F, Bucala R. 1995. Cloning and characterization of the gene for mouse macrophage migration inhibitory factor (MIF). J. Immunol. 154: 3863.