© 1991 S. Karger AG, Basel 0301-0147/91 /0213-0141 $2.75/0
Haemostasis 1991;21:141-146
Effect of Fibrinogen Fragments D and E on the Adhesive Properties of Human Granulocytes to Venous Endothelial C ells1 E G. Fischer. P. Vogel. A S. Ziegler. C.J. Kirkpatrick Institute of Pathology, RWTH-Klinikum, Aachen, FRG
Key Words. Cell adhesion • Fibrinogen degradation products • Neutrophils • Endothelial cells
Introduction As a prognostic parameter in polytrauma or severe burns, the measurement of fibrino gen degradation product (FDP) levels is well established in clinical medicine. While un der physiological conditions the concentra tions of FDP fragment D (FDP-D) and frag ment E (FDP-E) in peripheral blood are less than 8 and 0.1 pg/ml, respectively [1,2], lev els up to 100 pg/ml (in extreme cases 1,000 pg/ml) are found under pathological condi
1 This study is part of the dissertation of A.S.Z.
tions, e.g. in disseminated intravascular co agulation. Furthermore, a pathogenetic role of FDPs is discussed in the development of the adult respiratory distress syndrome (ARDS) [3]. In that study, endothelial alterations could not be correlated with high concentrations of FDPs. In contrast, in vitro it was found that FDP-D induces disorganization of bovine endothelial cell monolayers [4], We have in vestigated the influence of FDP-D and FDPE on endothelial cell attachment as an im portant repair mechanism. The influence of large FDP fragments on some granulocyte functions has been re ported recently [5]. We studied the cell at-
Downloaded by: Access provided by the University of Michigan Library 141.216.78.40 - 1/11/2019 6:14:48 AM
Abstract. In hyperfibrinolytic conditions, e.g. in disseminated intravascular coagulation or the adult respiratory distress syndrome, high levels of fibrinogen degradation products (FDPs) D and E are found in human plasma. This study investigates the influence of these fragments on cell attachment of human granulocytes in vitro. While leaving unaffected the adhesion of human umbilical vein endothelial cells (HUVEC) on gelatine-coated glass, both FDP fragments at 50 pg/ml inhibited granulocyte attachment to glass as well as to FIUVEC monolayers. At the same concentration, the fragments diminished the superoxide release of stimulated granulocytes. These results suggest a modulatory role of pathologically elevated FDPs on the granulocyte function cascade.
Fischer/Vogel/Ziegler/Kirkpatrick
tachment and the superoxide release as im portant steps in the granulocyte function cas cade. Materials and Methods To investigate the initial contact between endo thelial cells or granulocytes and the appropriate sur face, cells were allowed to settle and attach under the influence of fibrinogen degradation products for a definite time period. The experiments were carried out in 16-mm wells of multiwell plates (Falcon, Becton Dickinson. Heidelberg. FRG) at 37 °C. The meth ods used for this attachment assay are essentially the same as described earlier [6], In brief, human endothelial cells derived from umbilical cord veins (HUVEC) were isolated and cul tured up to the second passage in medium 199, enriched with ¿-glutamine, penicillin, streptomycin and 20% pooled human serum. To perform an assay with HUVEC, cells of 1 donor were harvested from the culture flask using 0.2% collagénase 1 (CLS I, Worthington, Interchem, Munich, FRG), followed by a washing step in assay medium. To investigate the attachment of HUVEC, cells were suspended in bovine serum albumin (BSA)enriched N-(2-hydroxyethyl)piperazine-N'-(2-ethane sulfonic) acid (HEPES)-buffered Hanks’ solution and were allowed to settle on the surface of gelatine-coated round 12-mm glass coverslips for 2 h at 37 °C. For each assay with human polymorphonuclear leukocytes (hPMN), cells of different young healthy donors were separated at room temperature from heparinized venous blood by dextran sedimentation and Ficoll-Paque (Pharmacia, Stockholm, Sweden) density centrifugation [6], followed by two washing steps in assay medium. hPMN cell purity was always above 90%, and cell viability amounted to 98 ± 1 % (trypan blue exclusion). The attachment of hPMN to sterile glass coverslips or to the surface of HUVEC monolayers was investigated. For the latter, HUVEC of the second passage were seeded in wells of multiwell plates con taining gelatine-coated round 12-mm coverslips. Cells were cultured to confluence within 3-5 days. Prior to the assay, the cell monolayers were washed twice with serum-free medium 199. The influence of the following substances was tested: the highly purified fibrinogen degradation
products FDP-D (85 kD) and FDP-E (50 kD; Diag nostica Stago, Asnières-sur-Seine, France), and the synthetic chemoattractant for hPMN, N-formyl-methionine-leucine-phenylalanine (fMLP; Sigma, Deisenhofen. FRG). Attachment assays were carried out in duplicate with cells of at least 3 different donors. To start the assay, each well received the appropriate solution of FDP, followed within 3 s by a suspension of HUVEC (3.5 X 104 cells/well) or hPMN (0.5 X or 1.0 X 106 cells/well on HUVEC or on glass, respectively). The assay medium was Hanks’ buffered salt solution (HBSS). enriched with 20 mM HEPES and 0.5% BSA. The final volume was 800 pl/well in HUVEC attachment assays and 500 pl/well in hPMN assays. Cells were allowed to settle for a definite time period and then washed twice with HBSS in order to remove nonadherent cells. After fixation with para formaldehyde, cells were stained with Giemsa solu tion. The number of adherent cells was counted under the light microscope in ten fields of 0.5 mm2 each located on a randomly chosen diameter of the coverslip. The distance between each of these areas was 1 mm. The comparison of different locations on the coverslip revealed mathematically negligible varia tions of cell distribution. Hence, a weighing proce dure was not necessary. The average number of cells per square millimeter found for the control in the HUVEC on-glass assays was 425 and 295 for hPMN on-glass, and 473 and 1,158 for hPMN on-HUVEC after 4 and 12 min, respectively. As there was a considerable interassay variance (s/x = 0.42; intra-assay variance: 0.21, total), the data of each assay had to be expressed relative to their control group by setting the control as the arbitrary unit. After calculation of the means and their stan dard deviations, the significance of the results was evaluated using multiple variance analysis (Peritz’ F test), setting the experimentwise confidence level to 99%. To investigate the influence of FDPs. another granulocytic cell function, chemiluminescence, was measured. Isolated granulocytes (5 X 106 cells/ml) were incubated with luminol ( 1 m V/, Lumac, Schaesberg, The Netherlands) and FDP-D or -E (25, 50 pg/ml) or HBSS for the control, as described by Allen and Roose [7], Without delay, opsonized zymo san was added, and light emission was measured as counts per minute in the chemiluminometer (Bio counter M2000, Lumac).
Downloaded by: Access provided by the University of Michigan Library 141.216.78.40 - 1/11/2019 6:14:48 AM
142
143
FDP Effects on Cell Adhesion
Results Attachment o f HUVEC to Glass Cell attachment of HUVEC to a gelatinecoated glass surface was not significantly in fluenced by FDP-D or FDP-E in concentra tions up to 50 pg/ml within 2 h (table 1). In addition, we found no visible influence of FDPs on HUVEC under microscopic obser vation. No obvious changes in cell integrity and shape could be seen either with con fluent HUVEC monolayers or within the at tachment assay after 2 or 4 h. HUVEC at tachment in vitro seems to be completed within 2 h, since there was no increase in the number of attached cells after 4 h. hPMN Attachment to HUVEC Monolayer or to Glass Fibrinogen degradation products, FDP-D or FDP-E, each in concentrations of 50 pg/ml, inhibited the 4-min attachment of hPMN to the surfaces of HUVEC mono
layers or glass (p < 0.01). In 12-min assays on HUVEC monolayers, the inhibition was highly significant only with 50 pg/ml FDP-E. At 10 pg/ml or less, neither FDP displayed any significant inhibitory effect on the at tachment (table 1). After a 30-min preincubation of PMN with FDP-D or -E ( 10 and 50 pg/ml) in poly ethylene tubes, followed by a washing step, we observed no significant influence of FDPs on hPMN attachment on glass (con trol = 1.00; FDP-D (50 pg/ml): 0.85 ± 0.33; FDP-E (50 pg/ml): 0.79 ± 0.17, relative to control; n = 3). Under assay conditions, fMLP increased hPMN attachment on glass 1.4-fold. On HUVEC, the stimulation was 2.6-fold within 3 min and 1.6-fold within 12 min versus their controls (table 1). In contrast, within 4 min, the simultaneous addition of fMLP to FDP-D or FDP-E just showed a slight in crease in the number of attached cells com pared to the amount observed with FDPs
Table 1. Attachment of HUVEC to glass and attachment of hPMN
after 12 min
after 4 min
Control FDP-D 10pg/ml FDP-D 50 pg/ml FDP-E 10 pg/ml FDP-E 50 pg/ml fMLP 10 nM
HUVEC to glass after 2 h
mean ± SD
n
mean ± SD
n
mean ± SD
n
mean ± SD
n
1.00 0.90 ±0.15 0.12 ± 0.041 0.73 ±0.24 0.25 ± 0 .1 11 1.38 ± 0.091
4 4 4 4 4 4
1.00 0.75 ±0.50 0.27 ±0.1 O' 0.70±0.37 0.18 ± 0.071 2.64±0.93'
10 9 10 9 10 10
1.00 0.91 ±0.17 0.41 ±0.262 0.39 ±0.18 0.14±0.08‘ 1.57 ±0.88*
3 3 3 3 3 3
1.00 1.09±0.24 0.92 ±0.07 0.93 ±0.02 0.91 ±0.08
3 3 3 3 3
-
-
Relative counts of attached cells after sedimentation under various test conditions. Means ± SD are expressed in relation to their controls. Assays with n different cell donors were done at least in duplicate. 1 Significant vs. control in the multivariance analysis at the 99% level of experimentwise confidence. 2 At the 95 % level.
Downloaded by: Access provided by the University of Michigan Library 141.216.78.40 - 1/11/2019 6:14:48 AM
hPMN to HUVEC monolayer
hPMN to glass after 4 min
Fischer/Vogel/Ziegler/Kirkpatrick
144
(! !
H'-ft i i :j
; i !|
jj A
i
jj
ß ■ •
. - f v
:
..
12 Min
/O
;1 v J l i n i l
Haemostasis 1991;21:141-146
Effect of Fibrinogen Fragments D and E on the Adhesive Properties of Human Granulocytes to Venous Endothelial C ells1 E G. Fischer. P. Vogel. A S. Ziegler. C.J. Kirkpatrick Institute of Pathology, RWTH-Klinikum, Aachen, FRG
Key Words. Cell adhesion • Fibrinogen degradation products • Neutrophils • Endothelial cells
Introduction As a prognostic parameter in polytrauma or severe burns, the measurement of fibrino gen degradation product (FDP) levels is well established in clinical medicine. While un der physiological conditions the concentra tions of FDP fragment D (FDP-D) and frag ment E (FDP-E) in peripheral blood are less than 8 and 0.1 pg/ml, respectively [1,2], lev els up to 100 pg/ml (in extreme cases 1,000 pg/ml) are found under pathological condi
1 This study is part of the dissertation of A.S.Z.
tions, e.g. in disseminated intravascular co agulation. Furthermore, a pathogenetic role of FDPs is discussed in the development of the adult respiratory distress syndrome (ARDS) [3]. In that study, endothelial alterations could not be correlated with high concentrations of FDPs. In contrast, in vitro it was found that FDP-D induces disorganization of bovine endothelial cell monolayers [4], We have in vestigated the influence of FDP-D and FDPE on endothelial cell attachment as an im portant repair mechanism. The influence of large FDP fragments on some granulocyte functions has been re ported recently [5]. We studied the cell at-
Downloaded by: Access provided by the University of Michigan Library 141.216.78.40 - 1/11/2019 6:14:48 AM
Abstract. In hyperfibrinolytic conditions, e.g. in disseminated intravascular coagulation or the adult respiratory distress syndrome, high levels of fibrinogen degradation products (FDPs) D and E are found in human plasma. This study investigates the influence of these fragments on cell attachment of human granulocytes in vitro. While leaving unaffected the adhesion of human umbilical vein endothelial cells (HUVEC) on gelatine-coated glass, both FDP fragments at 50 pg/ml inhibited granulocyte attachment to glass as well as to FIUVEC monolayers. At the same concentration, the fragments diminished the superoxide release of stimulated granulocytes. These results suggest a modulatory role of pathologically elevated FDPs on the granulocyte function cascade.
Fischer/Vogel/Ziegler/Kirkpatrick
tachment and the superoxide release as im portant steps in the granulocyte function cas cade. Materials and Methods To investigate the initial contact between endo thelial cells or granulocytes and the appropriate sur face, cells were allowed to settle and attach under the influence of fibrinogen degradation products for a definite time period. The experiments were carried out in 16-mm wells of multiwell plates (Falcon, Becton Dickinson. Heidelberg. FRG) at 37 °C. The meth ods used for this attachment assay are essentially the same as described earlier [6], In brief, human endothelial cells derived from umbilical cord veins (HUVEC) were isolated and cul tured up to the second passage in medium 199, enriched with ¿-glutamine, penicillin, streptomycin and 20% pooled human serum. To perform an assay with HUVEC, cells of 1 donor were harvested from the culture flask using 0.2% collagénase 1 (CLS I, Worthington, Interchem, Munich, FRG), followed by a washing step in assay medium. To investigate the attachment of HUVEC, cells were suspended in bovine serum albumin (BSA)enriched N-(2-hydroxyethyl)piperazine-N'-(2-ethane sulfonic) acid (HEPES)-buffered Hanks’ solution and were allowed to settle on the surface of gelatine-coated round 12-mm glass coverslips for 2 h at 37 °C. For each assay with human polymorphonuclear leukocytes (hPMN), cells of different young healthy donors were separated at room temperature from heparinized venous blood by dextran sedimentation and Ficoll-Paque (Pharmacia, Stockholm, Sweden) density centrifugation [6], followed by two washing steps in assay medium. hPMN cell purity was always above 90%, and cell viability amounted to 98 ± 1 % (trypan blue exclusion). The attachment of hPMN to sterile glass coverslips or to the surface of HUVEC monolayers was investigated. For the latter, HUVEC of the second passage were seeded in wells of multiwell plates con taining gelatine-coated round 12-mm coverslips. Cells were cultured to confluence within 3-5 days. Prior to the assay, the cell monolayers were washed twice with serum-free medium 199. The influence of the following substances was tested: the highly purified fibrinogen degradation
products FDP-D (85 kD) and FDP-E (50 kD; Diag nostica Stago, Asnières-sur-Seine, France), and the synthetic chemoattractant for hPMN, N-formyl-methionine-leucine-phenylalanine (fMLP; Sigma, Deisenhofen. FRG). Attachment assays were carried out in duplicate with cells of at least 3 different donors. To start the assay, each well received the appropriate solution of FDP, followed within 3 s by a suspension of HUVEC (3.5 X 104 cells/well) or hPMN (0.5 X or 1.0 X 106 cells/well on HUVEC or on glass, respectively). The assay medium was Hanks’ buffered salt solution (HBSS). enriched with 20 mM HEPES and 0.5% BSA. The final volume was 800 pl/well in HUVEC attachment assays and 500 pl/well in hPMN assays. Cells were allowed to settle for a definite time period and then washed twice with HBSS in order to remove nonadherent cells. After fixation with para formaldehyde, cells were stained with Giemsa solu tion. The number of adherent cells was counted under the light microscope in ten fields of 0.5 mm2 each located on a randomly chosen diameter of the coverslip. The distance between each of these areas was 1 mm. The comparison of different locations on the coverslip revealed mathematically negligible varia tions of cell distribution. Hence, a weighing proce dure was not necessary. The average number of cells per square millimeter found for the control in the HUVEC on-glass assays was 425 and 295 for hPMN on-glass, and 473 and 1,158 for hPMN on-HUVEC after 4 and 12 min, respectively. As there was a considerable interassay variance (s/x = 0.42; intra-assay variance: 0.21, total), the data of each assay had to be expressed relative to their control group by setting the control as the arbitrary unit. After calculation of the means and their stan dard deviations, the significance of the results was evaluated using multiple variance analysis (Peritz’ F test), setting the experimentwise confidence level to 99%. To investigate the influence of FDPs. another granulocytic cell function, chemiluminescence, was measured. Isolated granulocytes (5 X 106 cells/ml) were incubated with luminol ( 1 m V/, Lumac, Schaesberg, The Netherlands) and FDP-D or -E (25, 50 pg/ml) or HBSS for the control, as described by Allen and Roose [7], Without delay, opsonized zymo san was added, and light emission was measured as counts per minute in the chemiluminometer (Bio counter M2000, Lumac).
Downloaded by: Access provided by the University of Michigan Library 141.216.78.40 - 1/11/2019 6:14:48 AM
142
143
FDP Effects on Cell Adhesion
Results Attachment o f HUVEC to Glass Cell attachment of HUVEC to a gelatinecoated glass surface was not significantly in fluenced by FDP-D or FDP-E in concentra tions up to 50 pg/ml within 2 h (table 1). In addition, we found no visible influence of FDPs on HUVEC under microscopic obser vation. No obvious changes in cell integrity and shape could be seen either with con fluent HUVEC monolayers or within the at tachment assay after 2 or 4 h. HUVEC at tachment in vitro seems to be completed within 2 h, since there was no increase in the number of attached cells after 4 h. hPMN Attachment to HUVEC Monolayer or to Glass Fibrinogen degradation products, FDP-D or FDP-E, each in concentrations of 50 pg/ml, inhibited the 4-min attachment of hPMN to the surfaces of HUVEC mono
layers or glass (p < 0.01). In 12-min assays on HUVEC monolayers, the inhibition was highly significant only with 50 pg/ml FDP-E. At 10 pg/ml or less, neither FDP displayed any significant inhibitory effect on the at tachment (table 1). After a 30-min preincubation of PMN with FDP-D or -E ( 10 and 50 pg/ml) in poly ethylene tubes, followed by a washing step, we observed no significant influence of FDPs on hPMN attachment on glass (con trol = 1.00; FDP-D (50 pg/ml): 0.85 ± 0.33; FDP-E (50 pg/ml): 0.79 ± 0.17, relative to control; n = 3). Under assay conditions, fMLP increased hPMN attachment on glass 1.4-fold. On HUVEC, the stimulation was 2.6-fold within 3 min and 1.6-fold within 12 min versus their controls (table 1). In contrast, within 4 min, the simultaneous addition of fMLP to FDP-D or FDP-E just showed a slight in crease in the number of attached cells com pared to the amount observed with FDPs
Table 1. Attachment of HUVEC to glass and attachment of hPMN
after 12 min
after 4 min
Control FDP-D 10pg/ml FDP-D 50 pg/ml FDP-E 10 pg/ml FDP-E 50 pg/ml fMLP 10 nM
HUVEC to glass after 2 h
mean ± SD
n
mean ± SD
n
mean ± SD
n
mean ± SD
n
1.00 0.90 ±0.15 0.12 ± 0.041 0.73 ±0.24 0.25 ± 0 .1 11 1.38 ± 0.091
4 4 4 4 4 4
1.00 0.75 ±0.50 0.27 ±0.1 O' 0.70±0.37 0.18 ± 0.071 2.64±0.93'
10 9 10 9 10 10
1.00 0.91 ±0.17 0.41 ±0.262 0.39 ±0.18 0.14±0.08‘ 1.57 ±0.88*
3 3 3 3 3 3
1.00 1.09±0.24 0.92 ±0.07 0.93 ±0.02 0.91 ±0.08
3 3 3 3 3
-
-
Relative counts of attached cells after sedimentation under various test conditions. Means ± SD are expressed in relation to their controls. Assays with n different cell donors were done at least in duplicate. 1 Significant vs. control in the multivariance analysis at the 99% level of experimentwise confidence. 2 At the 95 % level.
Downloaded by: Access provided by the University of Michigan Library 141.216.78.40 - 1/11/2019 6:14:48 AM
hPMN to HUVEC monolayer
hPMN to glass after 4 min
Fischer/Vogel/Ziegler/Kirkpatrick
144
(! !
H'-ft i i :j
; i !|
jj A
i
jj
ß ■ •
. - f v
:
..
12 Min
/O
;1 v J l i n i l
