Immunobiol., vol. 182, pp. 117-126 (1991)
Original Papers Depts. of I Rheumatology, The Netherlands
2
Pulmonology, and
3
Nephrology, University Hospital Leiden,
Inhibition of Polymorphonuclear Leukocyte-Mediated Endothelial Cell Detachment by Antileukoprotease: A Comparison with Other Proteinase Inhibitors PIETERS. HIEMSTRA1,JOHANNESA. KRAMPS2, THERESAM. DEVREEDE 1, FERDINAND C. BREEDVELD J, and MOHAMED R. DAHA 3 Received March 30,1990· Accepted in Revised Form November 13, 1990
Abstract The role of elastase and proteinase inhibitors in polymorphonuclear leukocyte(PMN)mediated injury to human umbilical cord venous endothelial cells (HUVEC) was investigated. Both purified human neutrophil elastase and PMN that were stimulated with serum-treated zymosan (ST2) induced detachment, but not lysis of HUVEC. PMN-, but not purified elastase-mediated detachment was enhanced by the presence of methionine, which indicates a role for reactive oxygen metabolites in PMN-mediated HUVEC detachment. Detachment of HUVEC could be inhibited by secretory leukocyte proteinase inhibitor or antileukoprotease (ALP), ai-proteinase inhibitor (aI-PI) and N-methoxy-succinyl-ala-ala-pro-val-chloromethyl ketone (CMK). At concentrations at which elastase-mediated detachment was maximally inhibited, ALP and CMK, but not at-PI, were also able to inhibit maximally PMN-mediated detachment. An explanation for this difference could be that the larger size of aI-PI reduces the access of aI-PI to the interface between the PMN and the HUVEC.
Introduction Polymorphonuclear leukocytes (PMN) are able to secrete a wide variety of reactive oxygen metabolites and enzymes after stimulation (1, 2). In pathological conditions these reactions may result in tissue damage, but they may also be involved in the passage of PMN through the endothelial lining layer of small blood vessels. PMN stimulated by serum-treated zymosan (STZ) or phorbol myristate acetate (PMA) have been shown to induce injury to cultured endothelial cells (3, 4). Both oxidative and nonoxidative processes have been shown to be involved in PMN-mediated Abbreviations: ALP = antileukoprotease; aI-PI = aI-proteinase inhibitor; CMK = Nmethoxy-succinyl-ala-ala-pro-val-chloromethyl ketone; HBSS = Hanks' balanced salt solution; HUVEC = human umbilical cord venous endothelial cells; NBCS = newborn calf serum; NE = neutrophil elastase; PMA = phorbol myristate acetate; PMN = polymorphonuclear leukocyte; STZ = serum-treated zymosan.
118 . P. S. HIEMSTRA,
J. A.
KRAMPS, T. M.
DE
VREEDE, F. C. BREEDVELD, and M. R. DAHA
endothelial cell injury (3, 4). Neutrophil-derived elastase is thought to play an important role in the observed detachment of endothelial cells from the monolayer (3, 5), and in neutrophil-mediated solubilization of the subendothelial matrix (6). In vivo the activity of neutrophil elastase is regulated by proteinase inhibitors that are present in the circulation or, more locally, in tissues. A well-studied proteinase inhibitor is a1-proteinase inhibitor (a1-PI), the main inhibitor in the circulation (7). The effectiveness of a1-PI in vivo may be limited since in vitro a1-PI is only partly able to inhibit the proteolytic activity of PMN that are in close contact to an insoluble substrate (6, 8-11). This limited effectiveness of a1-PI in the pericellular microenvironment may in part be due to the molecular mass of a1-PI (52 kD) (8, 9, 11), and to inactivation by reactive oxygen metabolites (12, 13). The secretory leukocyte proteinase inhibitor or antileukoprotease (ALP), a major regulator of elastase activity in mucous secretions (14, 15), is another potent inhibitor of elastase and, to a lesser extent, of cathepsin G (16). It is like a1-PI susceptible to oxidative inactivation (17), but has a lower molecular mass (11.7 kD). Investigations into the role of ALP in the lung reported that ALP is, on a quantitative basis, the most important proteinase inhibitor in the central airways, whereas a1-PI is most important in the peripheral airways (15, 18-21). The aim of the present study was to investigate the role of elastase and proteinase inhibitors in PMN-mediated injury to endothelial cells. Therefore the injury to endothelial cells by purified neutrophil elastase and by serum-treated zymosan (STZ) stimulated PMN was studied. In addition, the inhibition of elastase- and PMN-mediated endothelial cell injury by ALP (MW 11.7 kD) was compared to that of the larger a1-PI (52 kD) and the smaller synthetic proteinase inhibitor N-methoxy-succinyl-ala-ala-proval-chloromethyl ketone (CMK) (503 D). Materials and Methods Materials
Human aI-PI, N-methoxy-succinyl-ala-ala-pro-val-chloromethyl ketone (CMK) and zymosan were purchased from Sigma (St. Louis, MO, USA). Human ALP was isolated from mucoid sputum of bronchitic patients by affinity chromatography using polyclonal rabbit anti-ALP (22). The concentration of both elastase-inhibitors present in the preparations used was determined by active site titration using neutrophil elastase and a synthetic substrate (23). Human neutrophil elastase (NE) was isolated from purulent sputum as described (24). Human serum was obtained from healthy volunteers and stored in aliquots at -70°C. Human PMN were isolated from heparinized blood obtained from healthy volunteers as described (25). After hypotonic shock in NH 4 CI to remove contaminating erythrocytes, the PMN were washed and resuspended at a concentration of 2 x 10 7/ml in Hanks' balanced salt solution (HBSS) containing 0.5 % bovine serum albumin (BSA). Both viability, as judged by trypan blue exclusion, and purity of the PMN preparation exceeded 95 %. Serum-treated zymosan (STZ) was prepared by incubating 10 mg boiled and washed zymosan in 1 ml normal human serum for 30 min at 37°C. Next, STZ was washed twice in PBS
PMN-mediated endothelial cell injury . 119 to remove serum, and resuspended at the desired concentration in HBSS containing 0.5 % BSA.
Culture of HUVEC Human umbilical cord venous endothelial cells (HUVEC) were isolated and cultured as decribed (26). Briefly, human umbilical cord veins were flushed with phosphate-buffered saline (PBS) and incubated with 1 mg/ml collagenase (Sigma) for 20 min at 37°C. HUVEC were harvested and cultured on 1 % gelatin in M199 with Earle's salts (Seromed, Biochrom KG, Berlin, Germany) containing endothelial cell growth factor (isolated from bovine hypothalamus; 27), penicillin (100 IV/ml) and streptomycin (100 flg/ml), and supplemented with 20 % heat-inactivated normal human serum. HUVEC were passaged by trypsinization.
Assay for detachment and 5JCr release from HUVEC The assay to determine PMN-mediated injury was performed essentially as described (25) with minor modifications. Briefly, HUVEC (third to seventh passage) were passaged by trypsinization, plated in gelatin-coated 48-well tissue culture plates (Costar, Cambridge, MA, USA), and cultured in MEM containing 20 % heat-inactivated newborn calf serum (NBCS) (Gibco, Grand Island, NY, USA) until the cells reached near confluency (approximately 3-5 x 104 HUVEC/well). The wells were washed and incubated with 200 fll medium containing 10 flCi/ml 51Cr-sodium chromate (New England Nuclear, Boston, MA, USA) for 18 hat 37°C and 5 % CO 2 , and washed three times with PBS before use. Next 1.5 x 106 PMN were added in a final volume of 225 fll HBSS-O.s % BSA in the presence or absence of proteinase inhibitors and/or methionine, and the PMN were allowed to settle for 10 min at room temperature. The concentration of methionine used (5 mM) is optimal to prevent oxidative inactivation of aI-PI (6) and ALP (17). Next 75 fll STZ (2 mg/ml) in HBSS-0.5 % BSA were added, and the wells were incubated for 3 h at 37"C in a 5 % CO 2 humidified atmosphere. Experiments with neutrophil elastase instead of PMN were performed by adding a mixture of neutrophil elastase, proteinase inhibitors and/or methionine in HBSS-0.5 % BSA, to labelled HUVEC; the wells were subsequently incubated for 3 h at 37°C in a 5 % CO 2 humidified atmosphere. Next, half of the supernatant was carefully removed, and the wells were washed five times with prewarmed HBSS containing 0.1 % BSA to remove detached cells. The radioactivity in the cell-free supernatant and the detached cells was determined in a gamma counter, and used to calculate «cpm release» and «cpm detachment». Residual HUVEC were harvested with 200 fll 1 M NaOH, and counted in a gamma counter (= «cpm 1 M NaOH»). The percentage specific detachment was calculated according to the following formula: «cpm detachment test sample» - «cpm detachment control» total cpm of the well (= «cpm 1 M NaOH» + «cpm detachment» + «cpm release»)
--~----~~--~~----~~~~~~~--~-.------------~--~x
1000/0
Lysis was calculated from the radioactivity in the cell-free supernatant ( 0.3). Mean± SD of fo ur independent experiments.
elastase-mediated HUVEC detachment was compared, CMK was used at 15 !lM, and a1-PI and ALP at 1.5 !lM. In these experiments, it was found that a1-PI, relative to ALP and CMK, was significantly (p < 0.05; Student's t-test) less effective in inhibiting PMN-mediated detachment of HUVEC than in inhibiting elastase-mediated detachment (Fig. 4). At inhibitor concentrations at which elastase-mediated detachment was maximally inhibited, ALP and CMK, but not at-PI, were also able to maximally inhibit PMN-mediated detachment. Discussion In the present study the ability of STZ-stimulated PMN to cause endothelial cell detachment in the presence of different proteinase inhibitors was investigated. STZ-stimulated PMN caused detachment on HUVEC from the monolayers, but no death of HUVEC (as judged by 51 Cr-release), in agreement with other reports (3, 25). It has been shown that neutrophil elastase plays a major role in the PMN-mediated injury to HUVEC (3, 5). The observation, in the present study, that purified human neutrophil elastase causes detachment of HUVEC (Fig. 1), and that PMN-mediated detachment can be inhibited to 86 % by the elastase-specific inhibitor CMK (Fig. 4), supports this conclusion.
PMN-mediated endothelial cell injury· 123
The presence of methionine enhanced PMN-mediated HUVEC detachment but did not affect elastase-mediated detachment (Fig. 2). This can be explained by an enhanced release of active elastase from the PMN, or by an alteration in the susceptibility of HUVEC to stimulated PMN. In other experiments, we observed an enhanced release of functional elastase activity in the presence of methionine by STZ-stimulated PMN (data not shown). It is likely that methionine enhances PMN -mediated HUVEC detachment by preventing oxidative inactivation of released elastase because 1. proteins that are released from stimulated PMN contain high levels of oxidized methionine (28), and 2. the respiratory burst accompanying phagocytosis of STZ by human PMN causes oxidative inactivation of a variety of released enzymes (29). In vivo the activity of released elastase is regulated by proteinase inhibitors. This report is the first to demonstrate that ALP inhibits both elastase- and PMN-mediated detachment of HUVEC (Fig. 3 and 4). aI-PI and CMK also inhibited elastase- and PMN -mediated detachment (Fig. 3 and 4). At suboptimal concentrations, ALP inhibited elastase-mediated HUVEC detachment to a greater extent than a 1-PI (Fig. 3). This can be explained by the higher capacity of ALP to inhibit the activity of elastase that is bound to an insoluble substrate (30, 31). The maximal inhibition of elastase-mediated detachment by aI-PI, ALP or CMK in molar excess did not differ between these proteinase inhibitors (Fig. 3 and 4). In contrast, ALP and CMK were more effective than aI-PI in inhibiting PMNmediated HUVEC detachment (Fig. 4). Elastase may escape the inactivation by proteinase inhibitors due to the capacity of PMN to cause oxidative inactivation of proteinase inhibitors such as ai-PI (12, 13) and ALP (17). The capacity of neutrophils, obtained from patients with chronic granulomatous disease, to degrade subendothelial matrix in the presence of ai-PI (10) demonstrates that oxidative inactivation of proteinase inhibitors is not the only mechanism by which elastase may escape inactivation by proteinase inhibitors. In order to prevent oxidative inactivation of the proteinase inhibitors in the present study, the experiments were performed in the presence of methionine, that effectively prevents oxidative inactivation of susceptible proteinase inhibitors by stimulated PMN (6, 12, 13, 17). Since a 1-PI only partly inhibited PMN-mediated HUVEC detachment, whereas elastase-mediated detachment was inhibited to a much higher extent (Fig. 4), it is more likely that elastase escapes inactivation by aI-PI because aI-PI is limited to get access to the interface between PMN and HUVEC. As mentioned in the introduction, the size of the proteinase inhibitor influences its capacity to inhibit degradation of substrates: large molecular weight proteinase inhibitors, such as aI-PI, are less effective in inhibiting substrate degradation than synthetic low molecular weight inhibitors (8, 9, 11). Recently ALP, which is a relatively small molecule, was shown to be more potent in inhibiting PMN-mediated fibrinolysis than ai-PI (32). ALP has also been
124 . P. S. HIEMSTRA,
J. A.
KRAMPS,
T. M. DE VREEDE, F. C. BREEDVELD, and M. R. DAHA
shown to be more effective than plasma, that contains various proteinase inhibitors including al-PI, in inhibiting proteolysis of fibronectin or elastin (36). Our results demonstrate that ALP (and CMK) has a higher capacity to inhibit PMN-mediated HUVEC detachment than al-PI. It is possible that in addition to the smaller size, other factors such as the cationic charge of ALP (pI> 9.5; 33) as compared to that of al-PI (pI 4-6; 34), playa role in the higher capacity to inhibit PMN -mediated HUVEC detachment. To gain further insight into the influence of the size of proteinase inhibitors on their capacity to inhibit PMN -mediated detachment, it would be of interest to study the effect of modulating the access of large molecules such as al-PI to the pericellular microenvironment between the PMN and the endothelial cells. Possibilities to achieve this goal are to decrease the adherence of the PMN to the endothelial cells using monoclonal antibodies directed to the common B-chain of the members of the LF A family of adhesion molecules on the PMN (35, 36), or to increase the size of low molecular weight inhibitors by aggregation. Future experiments using ALP aggregated by non-blocking monoclonal antibodies directed against ALP could provide more insight in this matter. In conclusion, ALP is a proteinase inhibitor that effectively inhibits detachment of HUVEC by elastase or stimulated PMN. A comparison of three proteinase inhibitors demonstrated that the size of proteinase inhibitors partly determines their effectiveness to prevent HUVEC damage by stimulated PMN. Therefore, it can be speculated that despite the presence of aI-PI and a2-macroglobulin in the circulation, stimulation of PMN may result in local effects on the vessel wall, which may result in tissue damage or allow extravasation of PMN. In contrast, in the lung ALP may prevent tissue injury by stimulated PMN more effectively than ai-PI, due to the smaller size of ALP. Acknowledgements This study was supported in part by the Dutch league against Rheumatism. The authors thank Mrs. G. J. M. LAtEBER for expert technical assistance.
References 1. HENSON, P. M., J. E. HENSON, C. FITTSCHEN, G. KIMANI, D. L. BRATTON, and D. W. H. RICHES. 1988. Phagocytic cells: degranulation and secretion. In: Inflammation. Basic principles and clinical correlates. Ed. GALLIN, J. E., 1. M. GOLDSTEIN, and R. SNYDERMAN. Raven Press, New York. 363 pp. 2. KLEBANOPF, S. J. 1988. Phagocytic cells: products of oxygen metabolism. In: Inflammation. Basic principles and clinical correlates. Ed. GALLIN, J. E., 1. M. GOLDSTEIN, and R. SNYDERMAN. Raven Press, New York. 391 pp. 3. HARLAN, J. M., P. D. KILLEN, L. A. HARKER, G. E. STRIKER, and D. WRIGHT. 1981. Neutrophil-mediated endothelial injury in vitro. Mechanisms of cell detachment. J. Clin. Invest. 68: 1394.
PMN-mediated endothelial cell injury· 125 4. WEISS, S. J., J. YOUNG, A. F. LoBUGLIO, A. SLIVKA, and N. F. NIMEH. 1981. Role of hydrogen peroxide in neutrophil-mediated destruction of cultured endothelial cells. J. Clin. Invest. 68: 714. 5. SMEDLY, L. A., M. G. TONNESSEN, R. A. SANDHAUS, C. HASLETT, L. A. GURTHRIE, R. B. JOHNSTON JR., P. M. HENSON, and G. S. WORTHEN. 1986. Neutrophil-mediated injury to endothelial cells. Enhancement by endotoxin and essential role of neutrophil elastase. J. Clin. Invest. 77: 1233. 6. WEISS, S. J., and S. REGIANJ. 1984. Neutrophils degrade subendothelial matrices in the presence of alpha-I-proteinase inhibitor. Cooperative use of lysosomal proteinases and oxygen metabolites. J. Clin. Invest. 73: 1297. 7. TRAVIS, J., and G. S. SALVESEN. 1983. Human plasma proteinase inhibitors. Ann. Rev. Biochem. 52: 655. 8. WEITZ, J. I., A. J. HUANG, S. L. LANDMAN, S. C. NICHOLSON, and S. C. SILVERSTEIN. 1987. Elastase-mediated fibrinogenolysis by chemoattractant-stimulated neutrophils occurs in the presence of physiologic concentrations of antiproteinases. J. Exp. Med. 166: 1836. 9. CAMPBELL, E. J., R. M. SENIOR, J. A. McDONALD, and D. L. Cox. 1982. Proteolysis by neutrophils. Relative importance of cell-substrate contact and oxidative inactivation of proteinase inhibitors in vitro. J. Clin. Invest. 70: 845. 10. WEISS, S. J., J. T. CURNUTTE, and S. REGIANI. 1986. Neutrophil-mediated solubilization of the subendothelial matrix: oxidative and nonoxidative mechanisms of proteolysis used by normal and chronic granulomatous disease phagocytes. J. Immuno!. 136: 636. 11. CAMPBELL, E. J., and M. A. CAMPBELL. 1988. Pericellular proteolysis by neutrophils in the presence of proteinase inhibitors: effects of substrate opsonization. J. Cell Bio!. 106: 667. 12. CARP, H., and A. JANOff. 1980. Potential mediator of inflammation. Phagocyte-derived oxidants suppress the elastase inhibitory capacity of alpha I-proteinase inhibitor in vitro. J. Clin. Invest. 66: 987. 13. CLARK, R. A., P. J. STONE, A. E. HAG, J. D. CALORE, and C. FRANZBLAU. 1981. Myeloperoxidase-catalyzed inactivation of alpha I-proteinase inhibitor by human neutrophils. J. Bio!. Chern. 256: 3348. 14. SCHIESSLER, H., K. HOCHSTRASSER, and K. OHLSSON. 1978. In: Neutral proteases of human polymorphonuclear leukocytes. Ed. HAVEMANN, K., and A. JANOH. Urban and Schwarzenberg Inc. Baltimore, Munich. 195 pp. 15. KRAMPS, J. A., C. FRANKEN, and J. H. DIJKMAN. 1984. ELISA for quantitative measurement of low-molecular-weight bronchial protease inhibitor in human sputum. Am. Rev. Respir. Dis. 129: 959. 16. TEGNER, H., K. OHLSSON, and I. OLSSEN. 1977. The interaction between a low molecular weight protease inhibitor of bronchial mucus and chymotrypsin-like cationic proteins of granulocytes. Hoppe Seyler's Z. Physio!. Chern. 358: 431. 17. KRAMPS, J. A., CH. VAN TWISK, E. C. KLASEN, and J. H. DIJKMAN. 1988. Interactions among stimulated human polymorphonuclear leucocytes, released elastase and bronchial antileucoprotease. Clin. Science 75: 53. 18. STOCKLEY, R. A. 1987. Alpha-I-antitrypsin and the pathogenesis of emphysema. Lung 165: 61. 19. TEGNEK, H. 1978. Quantitation of human granulocyte protease inhihitors in nonpurulent bronchial lavage fluids. Acta Otolaryngo!. 85: 282. 20. KRAMPS, J. A., C. FRANKEN, and J. H. DIJKMAN. 1988. Quantity of anti-leucoprotease relative to al-proteinase inhibitor in peripheral airspaces of the human lung. Clin. Science 75: 351. 21. BoulJlEK, c., A. PELLETIER, A. GAST, J. M. TOURNIER, G. PAULI, and J. G. BIETH. 1987. The elastase inhibitory capacity and the alphal-proteinase inhibitor and bronchial inhibitor content of bronchoalveolar lavage fluids from healthy subjects. Bio!. Chern. Hoppe Seyler 368: 981. 22. KLASEN, E. c., and J. A. KRAMPS. 1985. The N-terminal sequence of antileukoprotease isolated from bronchial secretion. Biochem. Biophys. Res. Comm. 128: 285.
126 . P. S. HIEMSTRA,
J. A. KRAMPS, T. M. DE VREEDE, F. C. BREEDVELD, and M. R. DAHA
23. KRAMPS, J. A., CH. VAN TWISK, and A. C. VAN DER LINDEN. 1983. L-Pyroglutamyl-Lprolyl-L-valine-p-nitroanilide, a highly specific substrate for granulocyte elastase. Scand. J. Clin. Lab. Invest. 43: 427. 24. FEINSTEIN, G., and A. JANOFF. 1975. A rapid method of purification of human granulocyte cationic neutral proteases: purification and further characterization of human granulocyte elastase. Biochim. Biophys. Acta 403: 493. 25. BREEDVELD, F. c., A. H. M. HEURKENS, G. J. M. LAFEBER, V. W. M. VAN HINSBERGH, and A. CATS. 1988. Immune complexes in sera from patients with rheumatoid vasculitis induce polymorphonuclear cell-mediated injury to endothelial cells. Clin. Immunol. Immunopathol. 48: 202. 26. MILTENBURG, A. M. M., M. E. MEIJER-PAAPE, M. R. DAHA, and L. C. PAUL. 1987. Endothelial cell lysis induced by lymphokine-activated human peripheral blood mononuclear cells. Eur. J. Immunol. 17: 1383. 27. MACIAG, T., J. CERUNDADO, S. ILSLEY, P. R. KELLEY, and R. FORAND. 1979. An endothelial cell growth factor from bovine hypothalamus: identification and partial characterization. Proc. Natl. Acad. Sci. USA 76: 5674. 28. BECK-SPEIER, I., L. LEUSCHEL, G. LUIPPOLD, and K. L. MAIER. 1988. Proteins released from stimulated neutrophils contain very high levels of oxidized methionine. FEBS Letters 227: 1. 29. KOBAYASHI, M., T. TANAKA, and T. USUI. 1982. Inactivation of lysosomal enzymes by the respiratory burst of polymorphonuclear leukocytes. J. Lab. Clin. Med. 6: 896. 30. BRUCH, M., and J. G. BIETH. 1986. Influence of elastin on the inhibition of leucocyte elastase by ai-proteinase inhibitor and bronchial inhibitor. Potent inhibition of elastinbound elastase by bronchial inhibitor. Biochem. J. 238: 269. 31. MORRISON, H. M., H. G. WELGUS, R. A. STOCKLEY, D. BURNETT, and E. J. CAMPBELL. 1990. Inhibition of human leukocyte elastase bound to elastin. Relative effectiveness and two mechanisms of inhibitory activity. Am. J. Respir. Cell Mol. Bioi 2: 263. 32. STOLK, J., W. A. HANLON, P. DAVIES, R. MUMFORD, M. E. DAHLGREN, W. B. KNIGHT, J. A. KRAMPS, and R. J. BONNEY. 1989. The effect of antileukoprotease (ALP) and a1proteinase inhibitor (ai-PI) on the PMN-mediated degradation of fibrogen. Am. Rev. Respir. Dis. 139: A201 (abstract). 33. FRITZ, H. 1988. Human mucus proteinase inhibitor (human MPI). BioI. Chern. HoppeSeyler 369 (Supp!.): 79. 34. BIETH, J. G. 1986. Elastase: catalytic and biological properties. In: Biology of extracellular matrix: A series. Regulation of matrix accumulation. Ed. MECHAM, R. P. Academic Press, New York. 217 pp. 35. DIENER, A. M., P. G. BEATTY, H. D. OCHS, and J. M. HARLAN. 1985. The role of neutrophil membrane glycoprotein 150 (gp-150) in neutrophil-mediated endothelial cell injury in vitro. J. Immunol. 135: 537. 36. RICE, W. G., and S. J. WEISS. 1990. Regulation of proteolysis at neutrophil-substrate interface by secretory leukoprotease inhibitor. Science 249: 178. Dr. P. S. HIEMSTRA, Department of Rheumatology, Building 1, C2-P, University Hospital, P.O. Box 9600, 2300 Leiden RC, The Netherlands
Original Papers Depts. of I Rheumatology, The Netherlands
2
Pulmonology, and
3
Nephrology, University Hospital Leiden,
Inhibition of Polymorphonuclear Leukocyte-Mediated Endothelial Cell Detachment by Antileukoprotease: A Comparison with Other Proteinase Inhibitors PIETERS. HIEMSTRA1,JOHANNESA. KRAMPS2, THERESAM. DEVREEDE 1, FERDINAND C. BREEDVELD J, and MOHAMED R. DAHA 3 Received March 30,1990· Accepted in Revised Form November 13, 1990
Abstract The role of elastase and proteinase inhibitors in polymorphonuclear leukocyte(PMN)mediated injury to human umbilical cord venous endothelial cells (HUVEC) was investigated. Both purified human neutrophil elastase and PMN that were stimulated with serum-treated zymosan (ST2) induced detachment, but not lysis of HUVEC. PMN-, but not purified elastase-mediated detachment was enhanced by the presence of methionine, which indicates a role for reactive oxygen metabolites in PMN-mediated HUVEC detachment. Detachment of HUVEC could be inhibited by secretory leukocyte proteinase inhibitor or antileukoprotease (ALP), ai-proteinase inhibitor (aI-PI) and N-methoxy-succinyl-ala-ala-pro-val-chloromethyl ketone (CMK). At concentrations at which elastase-mediated detachment was maximally inhibited, ALP and CMK, but not at-PI, were also able to inhibit maximally PMN-mediated detachment. An explanation for this difference could be that the larger size of aI-PI reduces the access of aI-PI to the interface between the PMN and the HUVEC.
Introduction Polymorphonuclear leukocytes (PMN) are able to secrete a wide variety of reactive oxygen metabolites and enzymes after stimulation (1, 2). In pathological conditions these reactions may result in tissue damage, but they may also be involved in the passage of PMN through the endothelial lining layer of small blood vessels. PMN stimulated by serum-treated zymosan (STZ) or phorbol myristate acetate (PMA) have been shown to induce injury to cultured endothelial cells (3, 4). Both oxidative and nonoxidative processes have been shown to be involved in PMN-mediated Abbreviations: ALP = antileukoprotease; aI-PI = aI-proteinase inhibitor; CMK = Nmethoxy-succinyl-ala-ala-pro-val-chloromethyl ketone; HBSS = Hanks' balanced salt solution; HUVEC = human umbilical cord venous endothelial cells; NBCS = newborn calf serum; NE = neutrophil elastase; PMA = phorbol myristate acetate; PMN = polymorphonuclear leukocyte; STZ = serum-treated zymosan.
118 . P. S. HIEMSTRA,
J. A.
KRAMPS, T. M.
DE
VREEDE, F. C. BREEDVELD, and M. R. DAHA
endothelial cell injury (3, 4). Neutrophil-derived elastase is thought to play an important role in the observed detachment of endothelial cells from the monolayer (3, 5), and in neutrophil-mediated solubilization of the subendothelial matrix (6). In vivo the activity of neutrophil elastase is regulated by proteinase inhibitors that are present in the circulation or, more locally, in tissues. A well-studied proteinase inhibitor is a1-proteinase inhibitor (a1-PI), the main inhibitor in the circulation (7). The effectiveness of a1-PI in vivo may be limited since in vitro a1-PI is only partly able to inhibit the proteolytic activity of PMN that are in close contact to an insoluble substrate (6, 8-11). This limited effectiveness of a1-PI in the pericellular microenvironment may in part be due to the molecular mass of a1-PI (52 kD) (8, 9, 11), and to inactivation by reactive oxygen metabolites (12, 13). The secretory leukocyte proteinase inhibitor or antileukoprotease (ALP), a major regulator of elastase activity in mucous secretions (14, 15), is another potent inhibitor of elastase and, to a lesser extent, of cathepsin G (16). It is like a1-PI susceptible to oxidative inactivation (17), but has a lower molecular mass (11.7 kD). Investigations into the role of ALP in the lung reported that ALP is, on a quantitative basis, the most important proteinase inhibitor in the central airways, whereas a1-PI is most important in the peripheral airways (15, 18-21). The aim of the present study was to investigate the role of elastase and proteinase inhibitors in PMN-mediated injury to endothelial cells. Therefore the injury to endothelial cells by purified neutrophil elastase and by serum-treated zymosan (STZ) stimulated PMN was studied. In addition, the inhibition of elastase- and PMN-mediated endothelial cell injury by ALP (MW 11.7 kD) was compared to that of the larger a1-PI (52 kD) and the smaller synthetic proteinase inhibitor N-methoxy-succinyl-ala-ala-proval-chloromethyl ketone (CMK) (503 D). Materials and Methods Materials
Human aI-PI, N-methoxy-succinyl-ala-ala-pro-val-chloromethyl ketone (CMK) and zymosan were purchased from Sigma (St. Louis, MO, USA). Human ALP was isolated from mucoid sputum of bronchitic patients by affinity chromatography using polyclonal rabbit anti-ALP (22). The concentration of both elastase-inhibitors present in the preparations used was determined by active site titration using neutrophil elastase and a synthetic substrate (23). Human neutrophil elastase (NE) was isolated from purulent sputum as described (24). Human serum was obtained from healthy volunteers and stored in aliquots at -70°C. Human PMN were isolated from heparinized blood obtained from healthy volunteers as described (25). After hypotonic shock in NH 4 CI to remove contaminating erythrocytes, the PMN were washed and resuspended at a concentration of 2 x 10 7/ml in Hanks' balanced salt solution (HBSS) containing 0.5 % bovine serum albumin (BSA). Both viability, as judged by trypan blue exclusion, and purity of the PMN preparation exceeded 95 %. Serum-treated zymosan (STZ) was prepared by incubating 10 mg boiled and washed zymosan in 1 ml normal human serum for 30 min at 37°C. Next, STZ was washed twice in PBS
PMN-mediated endothelial cell injury . 119 to remove serum, and resuspended at the desired concentration in HBSS containing 0.5 % BSA.
Culture of HUVEC Human umbilical cord venous endothelial cells (HUVEC) were isolated and cultured as decribed (26). Briefly, human umbilical cord veins were flushed with phosphate-buffered saline (PBS) and incubated with 1 mg/ml collagenase (Sigma) for 20 min at 37°C. HUVEC were harvested and cultured on 1 % gelatin in M199 with Earle's salts (Seromed, Biochrom KG, Berlin, Germany) containing endothelial cell growth factor (isolated from bovine hypothalamus; 27), penicillin (100 IV/ml) and streptomycin (100 flg/ml), and supplemented with 20 % heat-inactivated normal human serum. HUVEC were passaged by trypsinization.
Assay for detachment and 5JCr release from HUVEC The assay to determine PMN-mediated injury was performed essentially as described (25) with minor modifications. Briefly, HUVEC (third to seventh passage) were passaged by trypsinization, plated in gelatin-coated 48-well tissue culture plates (Costar, Cambridge, MA, USA), and cultured in MEM containing 20 % heat-inactivated newborn calf serum (NBCS) (Gibco, Grand Island, NY, USA) until the cells reached near confluency (approximately 3-5 x 104 HUVEC/well). The wells were washed and incubated with 200 fll medium containing 10 flCi/ml 51Cr-sodium chromate (New England Nuclear, Boston, MA, USA) for 18 hat 37°C and 5 % CO 2 , and washed three times with PBS before use. Next 1.5 x 106 PMN were added in a final volume of 225 fll HBSS-O.s % BSA in the presence or absence of proteinase inhibitors and/or methionine, and the PMN were allowed to settle for 10 min at room temperature. The concentration of methionine used (5 mM) is optimal to prevent oxidative inactivation of aI-PI (6) and ALP (17). Next 75 fll STZ (2 mg/ml) in HBSS-0.5 % BSA were added, and the wells were incubated for 3 h at 37"C in a 5 % CO 2 humidified atmosphere. Experiments with neutrophil elastase instead of PMN were performed by adding a mixture of neutrophil elastase, proteinase inhibitors and/or methionine in HBSS-0.5 % BSA, to labelled HUVEC; the wells were subsequently incubated for 3 h at 37°C in a 5 % CO 2 humidified atmosphere. Next, half of the supernatant was carefully removed, and the wells were washed five times with prewarmed HBSS containing 0.1 % BSA to remove detached cells. The radioactivity in the cell-free supernatant and the detached cells was determined in a gamma counter, and used to calculate «cpm release» and «cpm detachment». Residual HUVEC were harvested with 200 fll 1 M NaOH, and counted in a gamma counter (= «cpm 1 M NaOH»). The percentage specific detachment was calculated according to the following formula: «cpm detachment test sample» - «cpm detachment control» total cpm of the well (= «cpm 1 M NaOH» + «cpm detachment» + «cpm release»)
--~----~~--~~----~~~~~~~--~-.------------~--~x
1000/0
Lysis was calculated from the radioactivity in the cell-free supernatant ( 0.3). Mean± SD of fo ur independent experiments.
elastase-mediated HUVEC detachment was compared, CMK was used at 15 !lM, and a1-PI and ALP at 1.5 !lM. In these experiments, it was found that a1-PI, relative to ALP and CMK, was significantly (p < 0.05; Student's t-test) less effective in inhibiting PMN-mediated detachment of HUVEC than in inhibiting elastase-mediated detachment (Fig. 4). At inhibitor concentrations at which elastase-mediated detachment was maximally inhibited, ALP and CMK, but not at-PI, were also able to maximally inhibit PMN-mediated detachment. Discussion In the present study the ability of STZ-stimulated PMN to cause endothelial cell detachment in the presence of different proteinase inhibitors was investigated. STZ-stimulated PMN caused detachment on HUVEC from the monolayers, but no death of HUVEC (as judged by 51 Cr-release), in agreement with other reports (3, 25). It has been shown that neutrophil elastase plays a major role in the PMN-mediated injury to HUVEC (3, 5). The observation, in the present study, that purified human neutrophil elastase causes detachment of HUVEC (Fig. 1), and that PMN-mediated detachment can be inhibited to 86 % by the elastase-specific inhibitor CMK (Fig. 4), supports this conclusion.
PMN-mediated endothelial cell injury· 123
The presence of methionine enhanced PMN-mediated HUVEC detachment but did not affect elastase-mediated detachment (Fig. 2). This can be explained by an enhanced release of active elastase from the PMN, or by an alteration in the susceptibility of HUVEC to stimulated PMN. In other experiments, we observed an enhanced release of functional elastase activity in the presence of methionine by STZ-stimulated PMN (data not shown). It is likely that methionine enhances PMN -mediated HUVEC detachment by preventing oxidative inactivation of released elastase because 1. proteins that are released from stimulated PMN contain high levels of oxidized methionine (28), and 2. the respiratory burst accompanying phagocytosis of STZ by human PMN causes oxidative inactivation of a variety of released enzymes (29). In vivo the activity of released elastase is regulated by proteinase inhibitors. This report is the first to demonstrate that ALP inhibits both elastase- and PMN-mediated detachment of HUVEC (Fig. 3 and 4). aI-PI and CMK also inhibited elastase- and PMN -mediated detachment (Fig. 3 and 4). At suboptimal concentrations, ALP inhibited elastase-mediated HUVEC detachment to a greater extent than a 1-PI (Fig. 3). This can be explained by the higher capacity of ALP to inhibit the activity of elastase that is bound to an insoluble substrate (30, 31). The maximal inhibition of elastase-mediated detachment by aI-PI, ALP or CMK in molar excess did not differ between these proteinase inhibitors (Fig. 3 and 4). In contrast, ALP and CMK were more effective than aI-PI in inhibiting PMNmediated HUVEC detachment (Fig. 4). Elastase may escape the inactivation by proteinase inhibitors due to the capacity of PMN to cause oxidative inactivation of proteinase inhibitors such as ai-PI (12, 13) and ALP (17). The capacity of neutrophils, obtained from patients with chronic granulomatous disease, to degrade subendothelial matrix in the presence of ai-PI (10) demonstrates that oxidative inactivation of proteinase inhibitors is not the only mechanism by which elastase may escape inactivation by proteinase inhibitors. In order to prevent oxidative inactivation of the proteinase inhibitors in the present study, the experiments were performed in the presence of methionine, that effectively prevents oxidative inactivation of susceptible proteinase inhibitors by stimulated PMN (6, 12, 13, 17). Since a 1-PI only partly inhibited PMN-mediated HUVEC detachment, whereas elastase-mediated detachment was inhibited to a much higher extent (Fig. 4), it is more likely that elastase escapes inactivation by aI-PI because aI-PI is limited to get access to the interface between PMN and HUVEC. As mentioned in the introduction, the size of the proteinase inhibitor influences its capacity to inhibit degradation of substrates: large molecular weight proteinase inhibitors, such as aI-PI, are less effective in inhibiting substrate degradation than synthetic low molecular weight inhibitors (8, 9, 11). Recently ALP, which is a relatively small molecule, was shown to be more potent in inhibiting PMN-mediated fibrinolysis than ai-PI (32). ALP has also been
124 . P. S. HIEMSTRA,
J. A.
KRAMPS,
T. M. DE VREEDE, F. C. BREEDVELD, and M. R. DAHA
shown to be more effective than plasma, that contains various proteinase inhibitors including al-PI, in inhibiting proteolysis of fibronectin or elastin (36). Our results demonstrate that ALP (and CMK) has a higher capacity to inhibit PMN-mediated HUVEC detachment than al-PI. It is possible that in addition to the smaller size, other factors such as the cationic charge of ALP (pI> 9.5; 33) as compared to that of al-PI (pI 4-6; 34), playa role in the higher capacity to inhibit PMN -mediated HUVEC detachment. To gain further insight into the influence of the size of proteinase inhibitors on their capacity to inhibit PMN -mediated detachment, it would be of interest to study the effect of modulating the access of large molecules such as al-PI to the pericellular microenvironment between the PMN and the endothelial cells. Possibilities to achieve this goal are to decrease the adherence of the PMN to the endothelial cells using monoclonal antibodies directed to the common B-chain of the members of the LF A family of adhesion molecules on the PMN (35, 36), or to increase the size of low molecular weight inhibitors by aggregation. Future experiments using ALP aggregated by non-blocking monoclonal antibodies directed against ALP could provide more insight in this matter. In conclusion, ALP is a proteinase inhibitor that effectively inhibits detachment of HUVEC by elastase or stimulated PMN. A comparison of three proteinase inhibitors demonstrated that the size of proteinase inhibitors partly determines their effectiveness to prevent HUVEC damage by stimulated PMN. Therefore, it can be speculated that despite the presence of aI-PI and a2-macroglobulin in the circulation, stimulation of PMN may result in local effects on the vessel wall, which may result in tissue damage or allow extravasation of PMN. In contrast, in the lung ALP may prevent tissue injury by stimulated PMN more effectively than ai-PI, due to the smaller size of ALP. Acknowledgements This study was supported in part by the Dutch league against Rheumatism. The authors thank Mrs. G. J. M. LAtEBER for expert technical assistance.
References 1. HENSON, P. M., J. E. HENSON, C. FITTSCHEN, G. KIMANI, D. L. BRATTON, and D. W. H. RICHES. 1988. Phagocytic cells: degranulation and secretion. In: Inflammation. Basic principles and clinical correlates. Ed. GALLIN, J. E., 1. M. GOLDSTEIN, and R. SNYDERMAN. Raven Press, New York. 363 pp. 2. KLEBANOPF, S. J. 1988. Phagocytic cells: products of oxygen metabolism. In: Inflammation. Basic principles and clinical correlates. Ed. GALLIN, J. E., 1. M. GOLDSTEIN, and R. SNYDERMAN. Raven Press, New York. 391 pp. 3. HARLAN, J. M., P. D. KILLEN, L. A. HARKER, G. E. STRIKER, and D. WRIGHT. 1981. Neutrophil-mediated endothelial injury in vitro. Mechanisms of cell detachment. J. Clin. Invest. 68: 1394.
PMN-mediated endothelial cell injury· 125 4. WEISS, S. J., J. YOUNG, A. F. LoBUGLIO, A. SLIVKA, and N. F. NIMEH. 1981. Role of hydrogen peroxide in neutrophil-mediated destruction of cultured endothelial cells. J. Clin. Invest. 68: 714. 5. SMEDLY, L. A., M. G. TONNESSEN, R. A. SANDHAUS, C. HASLETT, L. A. GURTHRIE, R. B. JOHNSTON JR., P. M. HENSON, and G. S. WORTHEN. 1986. Neutrophil-mediated injury to endothelial cells. Enhancement by endotoxin and essential role of neutrophil elastase. J. Clin. Invest. 77: 1233. 6. WEISS, S. J., and S. REGIANJ. 1984. Neutrophils degrade subendothelial matrices in the presence of alpha-I-proteinase inhibitor. Cooperative use of lysosomal proteinases and oxygen metabolites. J. Clin. Invest. 73: 1297. 7. TRAVIS, J., and G. S. SALVESEN. 1983. Human plasma proteinase inhibitors. Ann. Rev. Biochem. 52: 655. 8. WEITZ, J. I., A. J. HUANG, S. L. LANDMAN, S. C. NICHOLSON, and S. C. SILVERSTEIN. 1987. Elastase-mediated fibrinogenolysis by chemoattractant-stimulated neutrophils occurs in the presence of physiologic concentrations of antiproteinases. J. Exp. Med. 166: 1836. 9. CAMPBELL, E. J., R. M. SENIOR, J. A. McDONALD, and D. L. Cox. 1982. Proteolysis by neutrophils. Relative importance of cell-substrate contact and oxidative inactivation of proteinase inhibitors in vitro. J. Clin. Invest. 70: 845. 10. WEISS, S. J., J. T. CURNUTTE, and S. REGIANI. 1986. Neutrophil-mediated solubilization of the subendothelial matrix: oxidative and nonoxidative mechanisms of proteolysis used by normal and chronic granulomatous disease phagocytes. J. Immuno!. 136: 636. 11. CAMPBELL, E. J., and M. A. CAMPBELL. 1988. Pericellular proteolysis by neutrophils in the presence of proteinase inhibitors: effects of substrate opsonization. J. Cell Bio!. 106: 667. 12. CARP, H., and A. JANOff. 1980. Potential mediator of inflammation. Phagocyte-derived oxidants suppress the elastase inhibitory capacity of alpha I-proteinase inhibitor in vitro. J. Clin. Invest. 66: 987. 13. CLARK, R. A., P. J. STONE, A. E. HAG, J. D. CALORE, and C. FRANZBLAU. 1981. Myeloperoxidase-catalyzed inactivation of alpha I-proteinase inhibitor by human neutrophils. J. Bio!. Chern. 256: 3348. 14. SCHIESSLER, H., K. HOCHSTRASSER, and K. OHLSSON. 1978. In: Neutral proteases of human polymorphonuclear leukocytes. Ed. HAVEMANN, K., and A. JANOH. Urban and Schwarzenberg Inc. Baltimore, Munich. 195 pp. 15. KRAMPS, J. A., C. FRANKEN, and J. H. DIJKMAN. 1984. ELISA for quantitative measurement of low-molecular-weight bronchial protease inhibitor in human sputum. Am. Rev. Respir. Dis. 129: 959. 16. TEGNER, H., K. OHLSSON, and I. OLSSEN. 1977. The interaction between a low molecular weight protease inhibitor of bronchial mucus and chymotrypsin-like cationic proteins of granulocytes. Hoppe Seyler's Z. Physio!. Chern. 358: 431. 17. KRAMPS, J. A., CH. VAN TWISK, E. C. KLASEN, and J. H. DIJKMAN. 1988. Interactions among stimulated human polymorphonuclear leucocytes, released elastase and bronchial antileucoprotease. Clin. Science 75: 53. 18. STOCKLEY, R. A. 1987. Alpha-I-antitrypsin and the pathogenesis of emphysema. Lung 165: 61. 19. TEGNEK, H. 1978. Quantitation of human granulocyte protease inhihitors in nonpurulent bronchial lavage fluids. Acta Otolaryngo!. 85: 282. 20. KRAMPS, J. A., C. FRANKEN, and J. H. DIJKMAN. 1988. Quantity of anti-leucoprotease relative to al-proteinase inhibitor in peripheral airspaces of the human lung. Clin. Science 75: 351. 21. BoulJlEK, c., A. PELLETIER, A. GAST, J. M. TOURNIER, G. PAULI, and J. G. BIETH. 1987. The elastase inhibitory capacity and the alphal-proteinase inhibitor and bronchial inhibitor content of bronchoalveolar lavage fluids from healthy subjects. Bio!. Chern. Hoppe Seyler 368: 981. 22. KLASEN, E. c., and J. A. KRAMPS. 1985. The N-terminal sequence of antileukoprotease isolated from bronchial secretion. Biochem. Biophys. Res. Comm. 128: 285.
126 . P. S. HIEMSTRA,
J. A. KRAMPS, T. M. DE VREEDE, F. C. BREEDVELD, and M. R. DAHA
23. KRAMPS, J. A., CH. VAN TWISK, and A. C. VAN DER LINDEN. 1983. L-Pyroglutamyl-Lprolyl-L-valine-p-nitroanilide, a highly specific substrate for granulocyte elastase. Scand. J. Clin. Lab. Invest. 43: 427. 24. FEINSTEIN, G., and A. JANOFF. 1975. A rapid method of purification of human granulocyte cationic neutral proteases: purification and further characterization of human granulocyte elastase. Biochim. Biophys. Acta 403: 493. 25. BREEDVELD, F. c., A. H. M. HEURKENS, G. J. M. LAFEBER, V. W. M. VAN HINSBERGH, and A. CATS. 1988. Immune complexes in sera from patients with rheumatoid vasculitis induce polymorphonuclear cell-mediated injury to endothelial cells. Clin. Immunol. Immunopathol. 48: 202. 26. MILTENBURG, A. M. M., M. E. MEIJER-PAAPE, M. R. DAHA, and L. C. PAUL. 1987. Endothelial cell lysis induced by lymphokine-activated human peripheral blood mononuclear cells. Eur. J. Immunol. 17: 1383. 27. MACIAG, T., J. CERUNDADO, S. ILSLEY, P. R. KELLEY, and R. FORAND. 1979. An endothelial cell growth factor from bovine hypothalamus: identification and partial characterization. Proc. Natl. Acad. Sci. USA 76: 5674. 28. BECK-SPEIER, I., L. LEUSCHEL, G. LUIPPOLD, and K. L. MAIER. 1988. Proteins released from stimulated neutrophils contain very high levels of oxidized methionine. FEBS Letters 227: 1. 29. KOBAYASHI, M., T. TANAKA, and T. USUI. 1982. Inactivation of lysosomal enzymes by the respiratory burst of polymorphonuclear leukocytes. J. Lab. Clin. Med. 6: 896. 30. BRUCH, M., and J. G. BIETH. 1986. Influence of elastin on the inhibition of leucocyte elastase by ai-proteinase inhibitor and bronchial inhibitor. Potent inhibition of elastinbound elastase by bronchial inhibitor. Biochem. J. 238: 269. 31. MORRISON, H. M., H. G. WELGUS, R. A. STOCKLEY, D. BURNETT, and E. J. CAMPBELL. 1990. Inhibition of human leukocyte elastase bound to elastin. Relative effectiveness and two mechanisms of inhibitory activity. Am. J. Respir. Cell Mol. Bioi 2: 263. 32. STOLK, J., W. A. HANLON, P. DAVIES, R. MUMFORD, M. E. DAHLGREN, W. B. KNIGHT, J. A. KRAMPS, and R. J. BONNEY. 1989. The effect of antileukoprotease (ALP) and a1proteinase inhibitor (ai-PI) on the PMN-mediated degradation of fibrogen. Am. Rev. Respir. Dis. 139: A201 (abstract). 33. FRITZ, H. 1988. Human mucus proteinase inhibitor (human MPI). BioI. Chern. HoppeSeyler 369 (Supp!.): 79. 34. BIETH, J. G. 1986. Elastase: catalytic and biological properties. In: Biology of extracellular matrix: A series. Regulation of matrix accumulation. Ed. MECHAM, R. P. Academic Press, New York. 217 pp. 35. DIENER, A. M., P. G. BEATTY, H. D. OCHS, and J. M. HARLAN. 1985. The role of neutrophil membrane glycoprotein 150 (gp-150) in neutrophil-mediated endothelial cell injury in vitro. J. Immunol. 135: 537. 36. RICE, W. G., and S. J. WEISS. 1990. Regulation of proteolysis at neutrophil-substrate interface by secretory leukoprotease inhibitor. Science 249: 178. Dr. P. S. HIEMSTRA, Department of Rheumatology, Building 1, C2-P, University Hospital, P.O. Box 9600, 2300 Leiden RC, The Netherlands
