Pergamon Thrombosis Research, Vol. 77, No. 4, pp. 321-327, 1995

Copyright 8 1995 Elsevier Science Ltd Printed in the USA. All rights reserved

0049-3848&j $9.50 + .OO

0049-3848(9S)EO0003-8

ACTIVATED NEUTROPHIL BY ENDOTHELIN- 1 CAUSED TISSUE DAMAGE IN HUMAN UMBILICAL CORD

Abdul Halim, Naohiro Kanayama, Emad El Maradny, Kayoko Maehara, Toshihiko Terao. Department of Obstetrics and Gynaecology, Hamamatsu University School of Medicine,

3600 Handa-cho, Hamamatsu 43 l-3 1, Japan.

(Received 18 July 7994 by Editor M. Fur/an; revisedaccepted 28 November 1994)

Abstract Immunostaining of human neutrophils incubated with endothelin- 1 (ET- 1) showed intense and spreading pattern of anti human granulocyte elastase within the cytosol. That reflected neutrophil activation followed by the release of granule contents by ET- 1. In contrast, PBS (phosphate buffered saline) treated neutrophils showed localized and faintly stained granules. Intracellular calcium in fura- loaded neutrophils was measured at 340/380 nm. A dose and time dependent increase in intracellular calcium by ET- 1 occurred in human single neutrophil. Elastase activity assay was done with chromogenic substrate S2484. ET-1 induces dose and time dependent increase in elastase activity in neutrophil suspensions like ionophore A23187. A similar time dependent increase in elastase activity was retained even after repeated wash and ET- I treatment. That confirmed the viability of most of the neutrophils after each treatment. In umbilical cord preparations, ET-l treated neutrophils could migrate from the venous lumen into the tissue matrix of the umbilical cords. Hematoxylin and eosin staining revealed a massive tissue destruction in ET-l activated neutrophil treated cords when compared to sham control and untreated neutrophil injected cords. Immunostaining with monoclonal anti human elastase revealed an intense staining in former sections when compared to the others. We suggest that ET-l activated neutrophil might play a major role in endothelial injury and tissue damage in conditions with high blood level of endothelin.

Endothelin (ET) is a potent vasoconstrictor peptide derived from the endothelial cells (1). High levels of endothelin are usually associated with conditions like myocardial infarction (2), preeclampsia-eclampsia, HELLP (hemolysis, elevated liver enzymes and low platelet) syndrome (3, 4) and disseminated intravascular coagulation (5). Activation of neutrophils was observed in severe preeclampsia, HELLP syndrome (6). Activated neutrophil has been shown to release polymorphonuclear (PMN) elastase, free radicals, cytokines etc. Increased neutrophil elastase was also found in unstable (compared to stable one) angina pectoris and acute myocardial infarction (7). Intra arterial administration of ET-l increased myeloperoxidase activity in ischemic heart tissues (8). The released products from activated neutrophils are known to cause tissue and endothelial damage (9, IO). A recent study revealed that ET- I increased adhesion of neutrophils on endothelial cells (8). Endothelial injury, organ damage and activation of coagulation in these diseases with multiple organ faliure could be a result of neutrophil activation and its interactions with the endothelium (I 1). Endothelin evoked vasoconstriction and ischemia were suggested in these diseases (2 - 4) and in preeclampsia to cause renal dysfunction and reduced placental blood flow ( 12). ET- 1 evoked in vitro enhanced coagulation and in vivo intravascular thrombosis (13).

Key words: Endothelin-1, Neutrophil activation, Intracellular calcium, Elastase, Tissue damage. Correspondence to: Md. Abdul Halim, Department of Obstetrics and Gynecology, Hamamatsu

University School of Medicine, 3600 Handa-cho, Hamamatsu 43 1-3 I, Japan.

321

322 NEUTROPHIL ACTIVATION Vol. 77, No. 4

In a recent study, we obtained a HELLP syndrome like phenomenon by ET-l injections in rabbits where hepatic vasospasm and ischemic hepatic necrosis were observed (14). Therefore, we assumed that high level of endothelin might have a role on neutrophils through their activation in circulation and tissues besides its vasoconstrictive function. We studied the effect of ET- 1 on human neutrophil and ET- 1 treated neutrophil in human umbilical cords.

MATERIALS AND METHODS

Neutrophilpreparation: Blood was collected from known healthy donors and neutrophils were prepared as described previously (15). Neutrophils were washed and re suspended with Hank’s buffered salt (Sigma Chem. Co., USA) solution (pH 7.3) at a concentration of l- 1.2~10” cell/ml.

Immunostoining of the neutrophil smear incubated with ET-I: 50 pmol of ET-l or same volume of PBS (phosphate buffered saline) was added in test tubes (n=7) containing 200 ul of neutrophil suspension. They were incubated at 37” C for 2 hours with 5% COz. The neutrophils were smeared on slides and fixed with 3% paraformaldehyde. The avidin peroxidase method of immunohistochemistry was performed as described previously (16). Monoclonal anti human elastase antibody (DAKO, USA) was used as primary antibody.

Measurement of intracellular Calcium concentration ( [Ca++]i ) in single Neutrophil Suspended neutrophils in BSS (134 mM NaCl, 3 mM MgCL2,S mM KCl, 5 mM D-Glucose, 15 mM Tris-HCl buffer, pH 7.4) were loaded with Fura- (Fura- AM, Kumamoto, Japan) and digital imaging microscopy was carried out as described previously ( 17, 18). The Fura- loaded neutrophils layered on a thin cover glass were placed on the microscope stage previously warmed to 37°C. The microscopic system is similar to that was used in our previous study (18). Fura- loaded cells were excited by ultraviolet light at 340/380 nm excitation. Sequential images were collected and digitized by an image analyzer ARGUS 100 (Hamamatsu Photonics K. K., Japan). After the acquisition of two ratio images, 25, 50 and 100 pmol/ml of ET-l (Peptide Institute, Osaka, Japan) and 2, 5 and 10 pg/ml of 4-Bromo-Calcium Ionophore A23187 (Sigma Chem. Co., USA) were administered into neutrophil suspension (in presence of 1 mM CaC12) respectively. image acquisitions were done at an interval of 30 seconds (set) for 180 sec. [Ca++]i at single cell level were calculated from 340/380 ratio values using a calibration curve as described earlier (19). We calculated the Mean + SD of [Ca++]i in 20 single neutrophils taken randomly from similar experiments (n=6) for each dose and time. Statistical analysis was done with Students’ ‘t’ test.

Eiastase activity assay with substrute S2484 in Neutrophil culture : Materials: L-Pyroglutamyl-L-prolyl-1-valine--p-nitroanilide (S-2484, KABI Diagnostic, Sweden), a specific chromogenic substrate for granulocyte elastase (20). Buffer: pH 8.3, Tris 100 mM/l, NaCl960 mM/l in distilled water. Incubation was done at 37” C with 5% CO2 in all cases. Group 1: 100 ul of buffer and 100 1.11 of neutrophil suspension were taken in each wells of a EIA plate (preheated to 37” C) and was incubated at 37” C for 5 min. 0,25,50, 100 pmol/ml of ET-l and 0, 1, 2, and 5 ug/ml of ionophore A23 187 were added to separate wells and were incubated for 0, 10, 30,60 min. Then 100 ul of substrate solution (S2484) was added to each well and the absorbencies in them were measured (at 405 nm with a photometer BIO-RAD, USA) at 5, 10, 15, 20, 25 and 30 min. The tests for each dose of agonists and time of incubations were repeated six times. Elastase activity was calculated as described previously (20). Group 2: 50 pmol/ml of ET- 1 was added to neutrophil suspension (in wells) and was incubated for 0, 30, 60, 90 min. respectively. Elastase activities in them were measured as described in group 1. After substrate reaction, each well was washed twice with buffer and incubated with Hank’s solution (at 37°C with 5% CO2) for 30 min. Then, they were treated again with 50 pmol/ml of ET-l as in first trial. The elastase activities were measured in them similarly. After second trial substrate reactions, each well was washed again (twice) with buffer and incubated with Hank’s solution for 30 min. They were treated again with ET-I similar to previous trials. Elastase activities in them (third trial) were calculated and analyzed using the students’ ‘t’ test.

Vol. 77, No. 4 NEUTROPHIL ACTIVATION 323

Immunohistochemical study in human umbilical cord.7: Umbilical cords (10 cm) were collected from normal pregnant mothers undergone cesarean section for cephalopelvic disproportion (n=2), previous cesarean section (n=3) and normal spontaneous delivery (n=2). They had no perinatal complication. The umbilical veins were washed three times with PBS. Each umbilical cord was divided equally into three parts. 3 ml of PBS (sham control) and 3 ml of washed neutrophil suspension treated with or without ET-l (100 pmol/ml) were injected and kept within venous lumen of either part of the cords by clumping both ends. After 4 hours of incubation at 37°C they were fixed in 3% paraformaldehyde for hematoxylin and eosin staining and immunostaining with anti human elastase antibody as described before in this text.

RESULTS

Immunostaining of the neutrophil smeur incubated with ET-I: In fig. la and b show the staining pattern of neutrophils stained with monoclonal anti human elastase antibody. The staining was localized as few granules within the cytosol and less intensely stained in ET-l untreated neutrophils (fig. la). Fig 1 b shows intense staining with anti human elastase antibody in neutrophils incubated with 50 pmol of ET- 1. ET- 1 increased the intensity and

FIG. I Immunostaining with monoclonal anti human elastase antibody in neutrophil smear incubated with or without ET- 1. Fig. lb shows intense staining with elastase antibody in neutrophils incubated with 50 pmol of ET-I compared to untreated neutrophils (fig. la).

600 ] Doses of ionophore:

0 50 p’mol/ml .25 pmol/ml

0 J t a 0 30 60 90 120 150 180 b ’ o 30 60 go 120 I;O lie

Time (seconds) Time (seconds)

FIG. 2 Measurement of [Ca++]i by digital image microscopy in Fura- loaded single neutrophils and expressed as mean + SD of 20 cells. ET- I (Fig. 2a) like ionophore A23 I87 (fig. 2b) increased [Ca++]i dose and time dependently (p < 0.05 in all compared to 0 min. values).

324 NEUTROPHIL ACTIVATION Vol. 77, No. 4

25 * t = m- l

2

* 15 l

l :: ti :a l

4: 10 t

*

l significant difference

1 compared to NO ET-1 value 0

Nil 25 50 100 a Doses of ET-l (pmol/ml)

Incubation time:

n 60 min. 0 30 min. l 10 min. 0 0 min.

7o _ l significant compared to

Incubation

time:

0 60min. . 30min. (I 10min.

n 0 min.

Dose of lonophore

FIG. 3 Elastase activity assay with S2484 in neutrophil culture incubated with and without ET- 1. A time and dose dependent increase in elastase activity (mean f SD) was observed by ET- I (fig. 3a) as with A23 187 (fig. 3b). * = p < 0.05 compared to initial values.

caused spreading pattern of the staining character with in the cytosol reflecting neutrophil activation and some events of the release of granule contents.

Measurement of [Ca++]i changes in single Neutrophil: Fig. 2 shows the changes in [Ca++]i in human neutrophils (n=20) at single cell level. Fig. 2a demonstrated a dose and time dependent increase in [Ca++]i by the effect of 25, 50 and 100 pmol/ml of ET-l. [Ca++]i in neutrophils was also increased significantly with 2, 5, 10 ug/ml of ionophore A231 87 (Fig. 2b). Change in [Ca++]i by 100 pmoYml of ET- I was almost comparable to that with 5ug/ml of ionophore.

Elustase activity assay with substrate S2484 in Neutrophil culture : Group 1: ET- 1 induces dose and time dependent increased elastase activity in neutrophil culture (fig. 3a). Similar increased elastase activity in neutrophil culture was observed when treated with 1,2 and 5 ug of ionophore (fig 3b).

Group 2: Fig. 4 showed the elastase activity in neutrophils treated by 50 pmol/ml of ET-l in three successive trials with two washes and incubation with Hank’s solution in between. A similar time dependent increase in elastase activity was observed in the same cells, That confirmed the viability of the neutrophils after each ET- I treatment and excluded any cytotoxic effect of ET- 1.

‘“‘q Duration of ET-l incubatton (mmute)

* p< 0.001 NS: not significant

FIG. 4

Elastase activity in neutrophils suspension incubated with 50 pmol/ml of ET- 1 in three successive trials as described in methods. A time dependent increase in elastase activity was observed by ET-l in each trial in same cells. * p < 0.001, compared to initial values.

Vol. 77, No. 4 NEUTROPHIL ACTIVATION 325

FIG. 5 Fig. 5a is the I-I/E stained umbilical cord section treated intravenously with ET-l activated neutrophils. It demonstrates the migration of neutrophils into the tissue matrix from the venous lumen. Fig. 5b and c show the immunostaining in umbilical cords with anti human elastase antibody. Staining was intense and diffuse in ET- 1 activated neutrophil treated cords (fig 5c) when compared to ET- 1 untreated ones (fig. 5b).

tmmunohistochemical study in human umbilical cords: ET- 1 treated neutrophil could migrate from the venous lumen into the tissue matrix of the umbilical cords (fig. 5a). Hematoxylin and eosin (l-I/E) staining revealed a massive tissue destruction in ET- 1 activated neutrophil treated cords when compared to sham control and neutrophil (not activated by ET-l) treated cords (figures not shown). The anti human elastase is stained strongly and diffusely around the vein of umbilical cord (fig. 5c) treated with ET-l activated neutrophils. However, those, injected with untreated neutrophils (fig. 5b) or sham control (figure not shown) showed no or minimum staining with anti human monoclonal elastase.

DISCUSSION

Several studies have shown that the circulating immmunoreactive ET-l levels are elevated in conditions where neutrophil elastase level and the interaction between neutrophils and endothelium are relevant (21, 22). However this study could be the first report about the migration of ET-l activated neutrophils in to the tissue followed by tissue damage.

In this study, immunostaining of the neutrophils incubated with ET- 1 revealed that ET- 1 increased the intensity and spreading pattern of the stained elastase within the cytosol. That reflected neutrophil activation and some events of release reaction. In contrast to ET- 1 treated neutrophils, PBS treated neutrophils showed localized and faintly stained granules within the cytoplasm of the neutrophils. Increased intracellular free cytosolic calcium, a dynamic second messenger, is known to reflect the activation and release reaction/secretion in many cells. We observed a dose and time dependent increment of [Ca++]i by ET- 1 in human neutrophils that agrees with the previous report ( 15). ET- 1 induced calcium change is comparable to that with ionophore A231 87. The activation of neutrophils was also presumed from increased [Ca++]i by ET- 1,

Activated neutrophil has been shown to release elastase, a lysosomal enzyme (9). In this connection we performed the dose and time related elastase activity assay with chromogenic substrate S2484 in human neutrophils treated with ET- 1 and A23 187. ET- 1 induced a dose and time dependent increase in elastase activity in neutrophils suspension. A similar dose and time dependent effect was also observed in neutrophils treated with ionophore A23 187. Therefore, we

NEUTROPHIL ACTIVATION Vol. 77, No. 4

suggested that the increased elastase activity might occur due to high level of endothelin in vivo. In this study, ET-l caused mobilization of [Ca++]i in neutrophils leading to release of elastase. We did not measure the other contents of neutrophil granules along with the elastase release. However, we also excluded any possible cytotoxic effect of ET-I on neutrophils in group-2 experiments. An almost similar elastase activity of neutrophils was observed by repeated ET- 1 treatment. Two washes with buffer and incubation with Hank’s solution was done in between trials. Almost similar time dependent increase in elastase activity was observed in three successive trials. That confirmed the viability of most of the neutrophils after each trial. From these observations, it was also presumed that ET- 1 acts on the neutrophils not only through the release but also through increased synthesis of elastase in the neutrophils. ET-l induced intracellular calcium mobilization reflected the same phenomenon. The critical link between Ca++ fluxes and Ca++ -sensitive protein is the cytosolic free Ca ++. An increase in cytosolic [Ca++]i might act as a second messenger and eventually activation and release reaction in neutrophils resulted. In an unpublished experiment, we confirmed the release of elastase in neutrophils suspension media by ET- 1. We could not exclude the other effects of ET- 1 in this study.

In a previous study (1 S), we found that ET- 1 caused expression of vWF by endothelial cells reflecting endothelial injury or activation. In umbilical cord preparations, ET-l treated neutrophils could migrate from the venous lumen into the tissue matrix of the umbilical cords. Hematoxylin and eosin staining revealed a massive tissue destruction in ET- 1 activated neutrophil treated cords when compared to sham control and untreated neutrophils injected cords. The lysosomal constituents from the activated and adherent neutrophils might cause endothelial damage. The injured endothelium became vulnerable for neutrophil migration. The neutrophil derived enzymes (elastase, collagenase, cathepsins, and so forth) might attack collagen, basement. membrane, fibrin and elastin resulting tissue destruction (10). Free radicals from the activated neutrophils caused further damage (10). Because, these tissues were outside the living body, the proper protective mechanism against neutrophil attack might be compromised. Potentially this might lead to unopposed protease activity with increased tissue destruction. The results from this in vitro study does not necessarily reflect the events taking place in vivo. Moreover, in in vivo situation, many other factors should infuence the ET-l effects. Although umbilical vessels contract after birth which may influence the results of this study, the human umbilical cord could be a useful tissue model (close to in vivo experiment) to study the migration of blood cells from intravascular space to the tissue and their effects. We compared our findings with appropriate sham controls that suggested the validity of this results. Endothelin induced vasospasm and ischemia were suggested to be important in the pathophysiologic events in such diseases (2-5). However, our present observations strongly suspected that ET- 1 activated neutrophil might play an additional important role in tissue damage and endothelial injuries in diseases with high level of endothelin (2-5).

We could not exclude other effects of ET-I and activated neutrophil. However further experiments are necessary to reveal the mechanism of and pathophysiological consequences by ET-l. This study could be a clue to know the exact relation between circulating neutrophils and endothelin.

REFERENCES

1. YANAGISAWA, M., KURIHARA, H., KIMURA, S., TOMOBE, Y., KOBAYASHI, M., MITSUI, Y., YAZAKI, Y., GOTO, K., and MASAKI, T. A novel potent vasoconstrictor peptide produced by vascular endothelial cells, Nature 332,4 1 I-41 5, 1988. 2. DOUSSET, B. and JACOB, C. Endothelin, a peptidic vasoconstrictor. Presse-Med 21(14), 665-669, 1988. 3. TAYLOR, RN., VARMA, M., NELSON, N.H. TENG and ROBERTS, JM. Women with preeclampsia have higher endothelin levels than women with normal pregnancies. J Clin Endocrinol Metab 71(6), 1675-1677, 1990. 4. IHARA, Y., SAGAWA, N., HASEGAWA, M., OKAGAKI, A., LI, X.M., INAMORI, K., H., MORI, T., SAITO, Y., SHIRAKAMI, G., NAKAO, K. and IMURA, H. Concentration of endothelin in maternal and umbilical cord blood at various stages of pregnancy. J Cardiovasc Pharmacol 17 Suppl 7, ~443-445, 1991.

Vol. 77, No. 4 NEUTROPHIL ACTIVATION 327

5. ISHIBASHI, M., ITO, N., FUJITA, M., FURUE, H. and YAMAJI, T. Endothelin-1 as an aggravating factor of disseminated intravascular coagulation associated with malignant neoplasm. Cancer 73(l), 191-195, 1994. 6. HAEGER, M., UNANDER, M., NORDER HANSSON, B., TYLMAN, M. and BENGTSSON, A. Complement, Neutrophil and Macrophage activation in women with severe preeclampsia and the syndrome of hemolysis, elevated liver enzymes and low platelet count. Obstet Gynecol79 (I), 19 - 26. 7. DINERMAR, JL., MEHTA, JL., SALDEEN, TG., EMERSON, S., WALLIN, R., DAVDA, R. and DAVIDSON, A. Increased neutrophil elastase in unstable angina pectoris and acute myocardial,infarction. J Am Colt Cardiol 15, 1559- 1563, 1990. 8. FARRE, L.A., RIESCO, A., ESPINOSA, G., DIGIUNI, E., CERNADAS, M R., ALVAREZ, V., MONTON, M., RIVAS, F., GALLEGO, M.J., EGIDO, J., CASADO, S. and CARAMELO, C. Effect of Endothelin-1 on Neutrophii Adhesion to Endothelial cells and perfused heart. Circulation 88, 1166 -I 17 1, 1993. 9. FRITZ, H., JOCHUM, M., GEIGER, R., DUSWALD, KH., DITTMER, H., KORTMANN, H., NEUMANN, S. and LANG, H. Granulocyte proteinases as a mediators of unspecific proteolysis in inflamation: A reveiw. Folia Histochem Cytobiol 24,99- I IS, 1986. 10. JANOFF, A. and CARP, H. Protease, antiproteases and oxidants: Pathway of tissue injury during inflamation. in MAGNO, G. and COTRAN, R.S. (eds.): Current Topics in inflamation and infection. Baltimore, Williums & Wilkins Co., 1982, p. 62-70. 11. PACHER, R., REDL, H., FRASS, M., PETZL, DH., SCHUSTER, E. and WOLOSZCZUK, W. Relationship between neopterine and granulocyte elastase plasma levels and the severity of multiple organ failure. Crit Care Med 17,757-760, 1989. 12. CLARK, BA., HALVORSON, L., SACHS, B. and EPSTEIN, FH. Plasma endothelin levels in preeclampsia: Elevation and correlation with uric acid levels and renal impairment. Am J Obstet Gynecol 166, 962-968, 1992. 13. HALIM, A., KANAYAMA, N, MARADNY E El, MAEHARA, K., TERAO, T. Coagulation in vivo microcirculation and in vitro caused by endothelin- I. Thromb Res 72,203-209, 1993. 14. HALIM, A., KANAYAMA, N., MAEHARA, K., TAKAHASHI, M., TERAO, T. HELLP Syndrome-like biochemical parameters obtained with Endotheiin-1 injections in rabbits. Gynecol Obstet Invest 35, 193-198, 1993. 15. FARRE, L.A,, RIESCO, A., MOLIZ, M., EGIDO, J., CASADO, S., HERNANDO, L. and CARAMEL0 C. Inhibition by L-arginine of endothelin-mediated increase in cytosolic calcium in human neutrophils. Biochem Biophys Res-commun 178, 884-89 1, 199 I. 16. SHI, ZR., ITZKOWITZ, SH., KIM, YS. A comparison of three immunoperoxidase technique for antigen detection in colorectal carcinoma tissues, J Histochem Cytochem 36, 3 17-322, 1988. 17. NISHIO, H., IKEGAMI, Y., and SEGAWA, T. Fluorescence digital image analysis of Serotonin induced calcium oscillation in single blood platelets. Cell Calcium 12, 177- 184, 199 1. 18. HALIM, A., KANAYAMA, N, MARADNY E El, MAEHARA K., HIRANO, M. and TERAO, T. Endothelin- I increased immunoreactive von Willebrand factor in endothelial cells and induced micro thrombosis in rats. Thromb Res 76,7 I-78, 1994. 19. MIYATA, I-I., HAYASHI, H., KOBAYASHI, A., and YAMAZAKI, N. Effect of Strophanthidin on intracellular Ca2+ Cardiovasc Res. 23, 378-384, 1989.

concentration and cellular morphology of Guini pig myocyte.

20. KRAMPS, JA, VANTWISK, CH, VANDER L.AC. L-Pyroglutamyl-L-prolyl-I-valine-p-nitro anilide, a highly specific substrate for granulocyte elastase. Stand J Lab invest 43,427-432, 1983. 21. ENGLER, RL., DAHLGREN, M-D., PETERSON, MA., DOBBS, A. and SCHIMD -SCONBEIN, GW. Accumulation of polymorphoneuclear leukocytes during 3 h experimental myocardial ischemia. Am J Physiol 251, H93-H-100, 1986. 22. MEHTA, JL., NICHOLS, WW., MEHTA, P. Neutrophils as potential participants in acute myocardial ischemia: relevance to repurfusion. J Am Coll Cardiol I 1, I 309- 13 16, 1988.