J BlOLUMlN CHEMILUMIN 1993; 8: 253-260

Effect of Prostaglandin E l on Chemiluminescence Response and Adherence of Human Polymorphonuclear Leukocytes to Human Cultured EndotheIiaI Cells of Prostaglandin E l Treated Polytraumatized patients

Eleonore Jonas,* Alexander Dwenger, Michael Jonas,' Michael Nerlich' and Harald Tscherne' Department of Clinical Biochemistry and ' Unfallchirurgische Klinik, Medical School Hannover, Konstanty- Gutschow Strasse 8, 3000 Hannover 61, Germany

We investigated the effect of prostaglandin El on human polymorphonuclear leuko- cytes, in vivo. Polymorphonuclear leukocytes of a prostaglandin El and placebo study group were harvested and their function, as production of oxygen-derived metabolites and adherence t o human cultured endothelial cells, was compared. Additionally, data obtained from polymorphonuclear leukocytes of a prostaglandin El and placebo group were compared wi th data obtained from polymorphonuclear leukocytes from 28 blood donors, who served as a control group.

Production of oxygen-derived metabolites by polymorphonuclear leukocytes during contact wi th endothelial cells was measured by chemiluminescence. Chemilumines- cence was significantly (p < 0.01) increased in the placebo group in comparison to the control group decreasing to values of control group after 6 d (post-trauma). Chemi- luminescence response was not significantly suppressed in patients treated wi th pros- taglandin El in comparison to the placebo group. Adherence of polymorphonuclear leukocytes (placebo group) to endothelial cells was significantly increased (p < 0.01) within the first 6d post-trauma. Following day 6, values were in the same range as values for the control group. Adherence was not significantly suppressed in patients treated wi th prostaglandin El in comparison to the placebo group. In conclusion, pros- taglandin El at a dose of 20ng/kg bw/min does not influence production of oxygen- derived metabolites and adherence in polytraumatized patients in comparison to a placebo group. Additionally, production of oxygen-derived metabolites by polymor- phonuclear leukocytes in response to endothelial cells is shown and it is evident that endothelial cells might influence production of oxygen derived metabolites by poly- morphonuclear leukocytes.

Keywords: Prostaglandin E l ; endotheliurn; polyrnorphonuclear leukocyte; cell interaction; adher- ence; oxygen-derived metabolites

*Author for correspondence.

0884-3996/93/050253-08$09.00 fc 1993 by John Wiley & Sons, Ltd.

Received 26 June 1992 Revised 7 December 1992

254 E ,IONAS, A. DWENGER, M. JONAS, M. FUNCK, M. NERLICH AND H. TSCHERNE

INTRODUCTlQN

The effects ot sepsis dlid acute lung injury involve, at lcast r x ! ! { I , , inter-actions between endothelial cells ( E C i c t i ~ ( l Lir-cula ting polymorphonuclear leu- kocytes (PMN L.s, neutrophils). These interactions involve adhesion of neutrophils due to activation followed by release of toxic products, e.g. lysoso- ma1 enzymes and generation of toxic oxygen- derived mciaboiiies. probably leading to EC damage ( 1 2 ) . Observaiions, mainly obtained in tis-

tLms suggest that prostaglandin El ts the function of neutrophils, e.g.,

release of oxygen--derived metabolites (3,4) and endothelial cell injury (5,4). Furthermore, the E series prostaglandins have been reported to have anti-inflammatory activity when examined in a variety of animal models (6,7). Apart from these observations there is a paucity of data on the effect of PG El administration from human studies on sepsis and acute lung itijury. Using PG El for the managerncnt of Adult Respiratory Distress Syn- drome (ARDS) associated with sepsis and trawna, a mluction of the mortality rate of patients was demonstrated (8).

Orie objective of the present research was to examine, i f a possible beneficial effect following treatment , I t ' ri,ultiply traumatized patients with PG El might be based upon inhibition of neutro- phi1 function Therefore, in a pre-study, five multi- ply traumatized patients received PG El by iv infusion ot a low dose of 20 ng/kg bw/min. Adher- ence and production of oxygen-derived metabolites during coincubation nleuirophils and EC were determined and compared with the function of neutrophils harvested from patients who did not undergo PG E < l treatment ( n = 5 , placebo group).

The second objective of !he present research was to cfetermirie any differences in neutrophil- endothelial cell interaction. Therefore, functions of neuirophi!s of' blood donors (control group), inultip!y tr:iurnatized patients not undergoing PG El trealnictit (placcbo group) and patients under- going PG E i treatment (PG 151 group) were com- pared.

METHODS

Study population

The 5tLid) pcjplation consisted of healthy con- trol ~r t c i i \ ; d -~~ l i . . II - X), nnultiply traumatized

patients, who were not undergoing PG El treat- ment (n = 5), patients, who received PG El ( n = 5) for 6d . 20ng PG E l (Alprostadil, kindly provided by Schwarz Pharma, Monheim, Ger- many) were infused per kilogram body weight per minute. Control injections consisted of 0.9% NaCl solution. Blood was drawn daily.

The experimental design was approved by the Ethics Committee of the Medical School, Hann- over. Informed constent was obtained from PG E l patients or their relatives.

According to the injury severity scale (ISS) (9) patients with an average of 34 .6 f4 .5 ISS points (PG E l group) and 25.2 f 4.9 ISS points (placebo group) were selected to guarantee a nearly homo- geneous study population. Injuries included blunt thoracic trauma with rib fractures, blunt abdomi- nal trauma and multiple fractures.

Endothelial cells

Human umbilical cord veing endothelial cells were harvested by collagenase digestion (10,ll). Endothelial cells were transferred to 75 cm2 cul- ture flasks (Primaria, Becton Dickinson, Germany) and grown to confluence in RM culture medium (mixture of RPMI medium and M 199 medium, Flow Lab.) which contained 12% human AB serum. At confluence, the typical cobblestone mor- phology was identified by phase contrast micro- scopy. In addition, the cells expressed Factor VIII antigen as detected by fluorescein conjugated rabbit antihuman Factor VIII antibody (Behring, Ger- many). For experimental use, endothelial cells were plated on sterilized coverslips (Lux Scientific Corporation, USA). The final plating density was 10,000 EC/coverslip providing a visually confluent monolayer after overnight incubation in culture medium.

Neutrophils

Neutrophils were prepared using Percoll density gradient centrifugation (12), and were resus- pended in phosphate-buffered saline (PBS): 70,000 neutrophils were added to 10,000 EC/test.

Oxygen -derived meta bol it es

Oxygen-derived metabolites were measured by lucigen-enhanced chemiluminescence, using a

EFFECT OF PROSTAGLANDIN E l ON CHEMILUMINESCENCE RESPONSE 255

2000-

1500 -

1000 -

500 -

0

six-channel Biolumat (LB 9505, Berthold, Ger- many) for simultaneous measurement. Chemilumi- nescence parameters were calculated from peak maximum counters/min values. The reaction mix- ture consisted of 515 pL Minimal Essential Med- ium (MEM) buffer solution, 1OpL lucigenin (0.23 mmol/L), 20 pL human standard plasma and 25 pL PMNLs suspension containing 70,000 neutrophils. When interaction between neutro- phils and endothelial cells was examined, 10,000 endothelial cells/coverslip providing a visually con- fluent monolayer, were added. Release of oxygen- derived metabolites was determined in duplicate or triplicate reaction mixtures.

*

PLACEBO 1 I I I I

Adherence

Adherence of neutrophils to EC was determined by 'I1 In (Amersham Corp, Braunschweig, Germany) labelling (1 3). Neutrophils were added to endothe- lial cell monolayers in microtiter wells and were incubated for 1 h at 37°C. "'In-oxine does not alter the viability, chemotactic responsiveness, bac- tericidal capacity, or ultrastructure of cells (14). Neutrophils were allowed to settle onto endothe-

lial cell monolayers by gravity. The number of neu- trophils that remained adherent to the endothelial cell monolayers was determined by counting gamma emissions (13). Results are expressed as percentage adherence.

Calculations

Values represent mean plus or minus standard error of the mean (SEM). For statistical analysis the U-test according to Mann-Whitney was used.

RESULTS

0 xyg en -derived m eta bo I i tes

The unstimulated chemiluminescence response of 70,000 neutrophils isolated from healthy control individuals was 454 f 85.8 x lo3 cpm (Fig. 1) and could be increased significantly by endothelial cells to 1386 f 243.9 x lo3 cpm (n = 28, X f SEM, p < 0.01; Fig. 2).

To exclude any differences between standard and autologous plasma, nine additional experiments

3 cpm x 10 2500-1

Oxygen de I i ved met a bol it es (PMNs alone)

256

2000 4 I

E. JONAS, A. DWENGER, M. JONAS, M. FUNCK, M. NERLICH AND H. TSCHERNE

Figure 2. Time course of chemiluminescence response of neutrophils during interaction wi th endothelial cells 20 ng PG E l / kg bw/min during the first 6 d post-trauma were infused The blank block area shows values for healthy controls Data are shown as Xi- SEM significance versus values for healthy controls

Table 1. Effect of standard and autologous plasma on chemiluminescence from PMNLs and PMNLs incubated with endothelial cells

PMNLs PMNLs + EC 632 -f 21 6 x 1 O3 1669 + 593 x 1 O3 cpm standard plasma 81 9 i- 350 x 1 O3 1 51 2 i- 557 x 1 O3 cpm autologous plasma

n = 9, x & SEM, significance PMNLs versus PMNLs + EC p < 0 01, no significance between autologous and standard plasma

were performed. Table 1 shows that the chemilumi- nescence response of neutrophils or neutrophils incubated with endothelial cells was the same in standard plasma and autologous plasma.

Fig. 1 illustrates the time course of unstimulated release of oxygen-derived metabolites from neutro- phils of multiply traumatized patients in compari- son to values of a control group. Neutrophils of multiply traumatized patients (placebo group) showed significantly ( p < 0.01) higher chemilumi- nescence than neutrophils of healthy control indivi- duals within the first 6 d (post-trauma). Following day 6, no difference between the placebo group and

control group could be demonstrated. Unstimu- lated chemiluminescence of neutrophils of the PG El group were also significantly elevated in com- parison to values of the control group within the first 6 d post-trauma decreasing to values of the control group after day 6. However, there was no significant difference in the chemiluminescence values between the PG El and placebo group (Fig. 1).

In Fig. 2 chemiluminescence response of neutro- phils during interaction with endothelial cells is illustrated. Values for the PG El and placebo group are in the same range as values for the control group. Following PG El treatment the values were not significantly different from values obtained from patients without drug treatment.

Neutrophils from control individuals can be stimulated to increased oxygen radical production by endothelial cells, as already shown in Table 1. Neither neutrophils from the placebo group nor neutrophils from the PG El group could be stimulated to increased chemiluminescence response by human endothelial cells, as shown in Fig. 3.

EFFECT OF PROSTAGLANDIN E l ON CHEMILUMINESCENCE RESPONSE 257

3 cpm x 10

looo1 I 800

600

400

200

0

-400

- 600 I

Oxygen derived metabolites (PMNs + EC) - PMNs

Y l T ' I

I 1 I I 1

1/2 3)4 5/6 71 8 9/10

days

control 0 placebo PG E l

Figure 3. Effect of endothelial cells on neutrophils. Only neutrophils of healthy controls were stimulated to increased produc- tion of oxygen-derived metabolites

Adherence

Neutrophils of healthy control individuals demon- strate a baseline level of adherence to cultured human endothelial cells (8.8 f 1.6%, x f SEM, n = 28). To exclude any differences between stand- ard and autologous plasma, nine additional experi- ments were performed. No statistical difference could be observed by comparing autologous and standard plasma. Adherence, measured using standard plasma was 9.3 f 1.15% ( n = 9, X f SEM), adherence measured using autologous plasma was 8.86 f 1.05% (n = 9, X f SEM). Adherence of neutrophils from the placebo group to endothelial cells was significantly different ( p < 0.01) from the control group within the first 6d post-trauma. Following day 6, values were in the same range as the controls (Fig. 4). Time course of adherence of neutrophils of the PG El group exhibited an elevated adherence within the first 6 d post-trauma followed by a decrease to con- trol values on day 6 (Fig. 3). No significant differ- ence between the placebo and PG El group could be demonstrated.

DISCUSSION

Oxygen-derived metabolites are implicated in endothelial cell injury in acute lung injury (15,2). Ex vivo effects of a low dose of prostaglandin El on chemiluminescence response has been investi- gated using in vitro cell culture and these showed inhibition of function of neutrophils by PG E l and inhibition of endothelial cell injury (4).

Unstimulated chemiluminescence response by neutrophils of multiply traumatized patients was in general significantly elevated in comparison to control individuals. This shows evidence of in vivo preactivation of neutrophils. Activation of neutro- phils is a pathophysiologic contributor to several diseases, e.g. Adult Respiratory Distress Syn- drome (16). Therefore the present data confirm studies made in ARDS patients that demonstrate in vivo neutrophil activation (17,18). No signifi- cant difference could be demonstrated between the PG El and placebo group in the production of oxygen-derived metabolites during adherence to endothelial cells. This is evidence that treatment of patients with PG El at a dose of 20ng/kg

258 E. JONAS, A. DWENGER, M. JONAS, M. FUNCK, M. NERLICH AND H. TSCHERNE

%

40 r

30 : *

Adherence of PMNs to. endothelial cells

W

PG El

PLACEBO 1/2 314 516 7/8 9/10

dcrys Figure 4. Time course of adherence of neutrophils to endothelial cells 20 ng PG E l /kg bw/rnin were infused during the first 6 d post-trauma Values are shown as F f SEM, significance versus values for healthy controls

bw/min does not appear to blunt chemilumines- cence response during interaction between neutro- phils and endothelial cells. Beneficial effects of PG El in patients with ARDS have been shown, and furthermore improved survival in patients with ARDS complicated by sepsis and trauma was demonstrated following infusion of 0.03 pg PG El/kg bw/min for 7 d (8). It was shown that superoxide anion production of neutrophils was inhibited following PG El treatment (3,19). Other studies have failed to confirm the observations described above. Studies in animal models show that even high doses of PG El did not ameliorate lung injury processes in adult rabbits or primates. PG El infusions (up to 0.1 pg/kg bw/min) could inhibit influx of neutrophils into the lung; treat- ment, however, did not result in any significant amelioration of the hyperoxic lung injury process or improve survival (20). The precise mechanism by which PG El influences phagocyte function is as yet unknown. Because of its lipophilic nature, PG El could diffuse into neutrophils.

The present investigation shows that neutrophils of healthy volunteers can be stimulated to increased production of oxygen-derived metabo- lites by endothelial cells, whereas neutrophils from the PG El and placebo group could not be

stimulated by endothelial cells to increased produc- tion of oxygen-derived metabolites. Culture condi- tions were always identical and the results were independent of the passage time of the endothelial cells. These preliminary results rule out the pos- sibility that endothelial cells might influence pro- duction of oxygen-derived metabolites by neutrophils, and further investigation is required on the mechanisms underlying neutrophil- endothelial cell interactions.

The present results support published data that the vascular lining can play an active role in neutrophil<ndothelial cell interactions (21). Recently, it was shown that bovine pulmonary artery endothelial cells (BPAE) or human umbili- cal cord vein endothelial cells are able to inhibit production of 02- by added human neutrophils stimulated by fMLP or heat-killed Staphylococci but not by activators as phorbol-myristate acetate (PMA) and delta-hexachlorocyclohexane (22).

Adherence

Since neutrophil adherence to endothelial cells is thought to be an early event leading to injury of endothelium, neutrophil adherence to EC was

EFFECT OF PROSTAGLANDIN E l ON CHEMILUMINESCENCE RESPONSE 259

evaluated. Baseline level of adherence of neutro- phils from the control group to EC were similar to those reported by other authors (13,21). Increased adherence of neutrophils from the PG El group in comparison to the control group, indi- cated in vivo activation of neutrophils, but showed the inability of PG El to diminish neutrophil adherence to EC. The present adherence data are in contrast to in vitro data from other studies (5) and have failed to confirm observations made by other investigators (7). These authors showed diminished adherence of neutrophils following sub- cutaneous injection of 1 mg/kg PG E 1 . One expla- nation for the differing results could be that these authors worked with nylon wool instead of endothelial cells.

Taken together, the present research demon- strates that PG El at a dose of 20ng/kg bw/min does not influence neutrophikndothelial cell interaction ex vivo. Furthermore, there seems to be evidence that endothelial cells may influence production of oxygen-derived metabolites by neu- trophils.

Acknowledgement

The authors thank the staff members and nurses of the ‘Friederikenstift’ Delivery Room for their invaluable help in collecting umbilical cords.

REFERENCES

1. Worthen GS, Haslett C, Rees AJ, Gumbay RS, Henson JE, Henson PM. Neutrophil-mediated pulmonary vascular injury: synergistic effect of trace amounts of lipopolysaccharide and neu- trophil stimuli in vascular permeability and neutrophil sequestration in the lung. Am Rev Repir Dis 1987;136:19-28.

2. Jonas E, Dwenger A, Lueken B, Boehme U. Simultaneous measurement of endothelial cell damage, elastase release and chemilumines- cence response during interaction between polymorphonuclear leukocytes and endothe- lial cells. J Biolumin Chemilumin 1991 ;6: 19- 27.

3. Fantone JC, Kinnes DA. Prostaglandin El and prostaglandin I2 modulation of superox- ide production by human neutrophils. Bio- chem Biophys Res Commun 1983;113:50&12.

4. Jonas E. Dwenger A. Funck M. Effect of

varying concentrations of prostaglandin El on chemiluminescence response and endothe- lial cell damage during interaction between polymorphonuclear leukocytes and human endothelial cells. Fres J Anal Chem 1990;337:90-1.

5. Chopra J, Webster 0. PG El inhibits neutro- phi1 adherence and neutrophil-mediated injury to cultured endothelial cells. Am Rev Respir Dis 1988; 138:915-20.

6. Fantone JC, Kinkel SL, Zwier RB. Effects of prostaglandins on in vivo immune and inflam- matory reactions. In: Goodwin JS, editor. Prostaglandins and Immunity. Boston: Marti- nus Nijhoff, 198Y123-46.

7. Ulich TR, Dakay EB, Williams JH, Ni RX. In vivo induction of neutrophilia, lymphopenia, and diminution of neutrophil adhesion by stable analogs of prostaglandins El , E2 and F2 alpha. Am J Pathol 1986;124:53-8.

8. Holcroft JW, Vassar MJ, Weber CJ. Prosta- glandin E l and survival in patients with the adult respiratory distress syndrome. Ann Surg 1986;203:371-78.

9. Baker SP, O’Neill B, Haddon W Jr. The injury severity score: a method for describing patients with multiple injury and evaluating emergency care. J Trauma 1974; 14: 187-96.

10. Jaffe EA, Hoyer LW, Nachman RL. Synthesis of anti-hemophilic factor antigen by cultured human endothelial cells. J Clin Invest

11. Gimbrone MA, Shefton EJ, Cruise SA. Isola- tion and primary culture of endothelial cells from human umbilical vessels. Tissue Culture Association Manual 1978;4813-17.

12. Dwenger A, Schweitzer G, Regel G. Bronch- oalveolar lavage fluid and plasma proteins, chemiluminescence response and protein con- tents of polymorphonuclear leukocytes from blood and lavage fluid in traumatized patients. J Clin Chem Clin Biochem

13. Charo JF, Yuen C, Goldstein J. Adherence of human polymorphonuclear leukocytes to endothelial monolayers: effects of tempera- ture, divalent cations and chemotactic fac- tors on the strength of adherence measured with a new centrifugation assay. Blood

14. Zakhireh B, Thakur ML, Malech HL, Cohen MS. Gottschalk A. Root RK. Indium-

1973;52:2757-64.

1986;24:73-88.

1985;65:473-9.

‘2 1 1 1:labeled human polymorphonuclear

260 E. JONAS, A. DWENGER, M. JONAS, M. FUNCK, M. NERLICH AND H. TSCHERNE

leukocytes: viability, random migration, che- motaxis, bactericidal capacity, and ultrastruc- ture. J Nucl Med 1979; 20:741-7.

15. Worthen GS, Haslett C , Smedly LA, Rees AJ, Gumbay RS, Henson JE, Henson PM. Lung vascular injury induced by chemotactic fac- tors: enhancement by bacterial endotoxins. Fed Proc 1986;45:7-12.

16. Tate RM, Repine JE. Neutrophils and the adult respiratory distress syndrome. Am Rev Respir Dis 1983;128:552-9.

17. Zimmerman GA, Renzetti AD, Hill HR. Func- tional and metabolic activity of granuloctyes from patients with Adult Respiratory Distress Syndrome. Am Rev Respir Dis 1983;127:290- 300.

18. Fletcher MP, Kassar MJ, Holcroft JW. Patients with adult respiratory distress syn- drome demonstrate in vivo neutrophil activa- tion associated with diminished binding of neutrophil-specific monoclonal antibody 31D8. Inflammation 1988;12:455-73.

19. Fantone JC, Marasco WA, Elgas LJ, Ward PA. Stimulus specificity of prostaglandin inhi- bition of rabbit polymorphonuclear leukocyte lysosomal enzyme release and superoxide anion production. Am J Pathol 1984; 1 15:6-9.

20. Hageman JR, Lee SE, Zemaitis J, Smith LJ, Hunt CE. Prostaglandin El influsion fails to prevent hyperoxic lung injury in adult rab- bits. Crit Care Med 1989;17:33944.

21. Tonnesen MG. Neutrophikndothelial cell interactions: mechanisms of neutrophil adher- ence to vascular endothelium. J Invest Derma-

22. Basford RE, Clark RL, Stiller RA, Kaplan SS, Kuhns DB, Rinaldo JE. Endothelial cells inhi- bit receptor-mediated superoxide-anion pro- duction by human polymorphonuclear leukocytes via a soluble inhibitor. Am J Respir Cell Mol Biol 1990;2:23543.

to1 1989;93:535-85.