Inhibition of Neutrophil Superoxide Production by Adenosine Released from Vascular Endothelial Cells Gary R. Gunther, PhD, Malcolm B. Herring, MD, Indianapolis, Indiana
To investigate the inhibitory effect of adenosine released by endothelium on neutrophil superoxide (02-) production, we treated confluent monolayers of cultured human umbilical vein endothelial cells with the enzyme adenosine deaminase, and then added human neutrophils. Superoxide (02-) production by human neutrophils stimulated with 10 -6 M formyl-methionyl-leucyl-phenylalanine was inhibited by 49% in the presence of a confluent monolayer of human umbilical vein endothelial cells (5.1 -+ 0.1 versus 2.6 _+ 0.3 nmols O2-/106 neutrophils). Addition of 0.25 U/ml adenosine deaminase to neutrophils plus endothelial cells restored formyl-methionyl-leucyl-phenylalanine-stimulated neutrophil superoxide production to the level seen with neutrophils alone. Deoxycoformycin (10 -4 M), an inhibitor of adenosine deaminase activity, prevented the increase in superoxide production associated with adenosine deaminase addition. The adenosine analogue 5'-(N-ethylcarboxamido)adenosine (3 x 10 -4 M) caused increased inhibition of formyl-methionyl-leucylphenylalanine-stimulated superoxide release by neutrophils in the presence of endothelial cells and prevented neutrophil-mediated endothelial cell damage, as measured by release of 3H-2-deoxy-D-glucose. Pairing 2-chloroadenosine (10 -s M) or 5'-(N-ethylcarboxamido)-adenosine (3 x 10 -4 M) with a cyclic adenosine monophosphate phosphodiesterase inhibitor, 3-isobutyl-l-methyl-xanthine (10 -4 M), produced greater inhibition of neutrophil superoxide production than occurred with either compound alone. The results support the hypothesis that vascular endothelial cells protect themselves from neutrophil attack by releasing adenosine to inhibit superoxide production. (Ann Vasc Surg 1991;5:325-330). KEY WORDS:
Endothelial seeding; superoxide production; adenosine; neutrophils.
Studies of endothelial cell seeding [1] of prosthetic vascular grafts reveal extensive cell loss within 24 hours of restoring blood flow [2]. Emerick and co-workers [3] were able to prevent the loss of endothelial cells from seeded grafts in dogs by making the animals leukopenic. Inspection of the From the Department of Surgery, Indiana University School of Medicine, Indianapolis, Indiana. Reprint requests: Gary R. Gunther, PhD, Department of Physiology and Biophysics MS374, Indiana University School of Medicine, 1100 West Michigan Street, Indianapolis, Indiana 46202.
flOW surfaces of culture-lined canine grafts by scanning electron microscopy six hours after implantation demonstrated neutrophil attachment and endothelial cell loss [4]. These findings support the concept [4] that neutrophils mediate endothelial cell detachment from the graft surface. One of the ways that neutrophils can damage endothelial cells is by the release of superoxide (02-) radicals [5,6]. Adenosine, which is released by human neutrophils [7] and endothelial cells [8], inhibits neutrophil Oe production [7]. Adenosine analogues can inhibit neutrophil adhesion to endothelial cells and neutrophil-mediated endothelial
325
I N H I B I T I O N OF NEUTROPH1L SUPEROXIDE P R O D U C T I O N
326
cell damage in vitro [8]. Adenosine deaminase converts adenosine to inosine, which does not inhibit neutrophil function [7]. As one might expect, adenosine deaminase increases neutrophil-mediated injury to endothelial cells in vitro [8]. We report that adenosine deaminase greatly increases neutrophil O2- production in the presence of endothelial cells. This supports the hypothesis [8] that adenosine release is a mechanism by which endothelial cells protect themselves from damage by neutrophil release of O2-.
ANNALS OF VASCULAR SURGERY
search, Waltham, Massachusetts. Gelatin was obtained from Difco Laboratories, Inc., Detroit, Michigan. D-glucose and sodium chloride were obtained from Mallinckrodt, Inc., Paris, Kentucky. 2-deoxy-D[1-3H] glucose (3H-DOG) (17 Ci/mmol) was purchased from Amersham Corporation, Arlington Heights, Illinois. Ready-Protein liquid scintillation cocktail was obtained from Beckman Instruments, Palo Alto, California. Endothelial cell culture
MATERIAL AND METHODS Experimental design
To analyze the effect of endothelial cells on neutrophil 0 2 - production, human neutrophils were stimulated with the peptide N-formyl-L-methionylL-leucyl-L-phenylalanine (FMLP) in the presence or absence of a monolayer of human umbilical vein endothelial cells in vitro. To determine the role of endogenous adenosine, adenosine deaminase was added and compared to culture wells containing the adenosine deaminase inhibitor deoxycoformycin along with the enzyme. To show that inhibition of neutrophil 0 2 - production was related to a decrease in injury to endothelial cells, an adenosine analogue was added to cultures containing endothelial cells plus FMLP-stimulated neutrophils. 0 2 - production and endothelial cell injury were measured. Materials
Adenosine, 2-chloroadenosine (2-ChlAdn), 5'-(Nethylcarboxamido)-adenosine (NECA), 3-isobutyl-lmethyl-xanthine (IBMX), adenosine deaminase (Type VI, from calf intestinal mucosa), deoxycoformycin, FMLP, cytochrome c (Type VI, from horse heart), superoxide dismutase (from bovine erythrocytes), sodium azide, bovine serum albumin (BSA) (Cat. No. A-2153), M-ethylmaleimide, dextran (MW 225,000), and Ficoll-Hypaque leukocyte separation medium (Histopaque 1077) were purchased from Sigma Chemical Company, St. Louis, Missouri. Collagenase (CLS Type I) was obtained from Worthington Biochemical Corporation, Freehold, New Jersey. Medium M199 with Earle's balanced salts and 25 mM HEPES, fetal bovine serum, 200 mM L-glutamine, 100 mM sodium pyruvate, nonessential amino acids (100X), 2.5% trypsin, phosphate-buffered saline (PBS), Dulbecco's phosphate-buffered saline (DPBS), and deionized water were purchased from Whittaker Bioproducts, Inc., Walkersville, Maryland. Penicillin-streptomycin solution (100X) was obtained from GIBCO, Grand Island, New York. Insufin-transferrin-selenous acid mixture (ITS concentrate) was a product of Collaborative Re-
Human umbilical vein endothelial cells were harvested from umbilical cords obtained from a local hospital. Endothelial cells were dissociated from the vessel wall using 0.13% collagenase in PBS according to an established procedure [9,10]. Endothelial cells were cultured at 37~ in a humidified 5% CO2 atmosphere in M199 medium containing 10% heat-inactivated fetal bovine serum and supplemented with 2 mM L-glutamine, 1 mM pyruvate, 1% nonessential amino acid supplement, 100 U/ml penicillin plus 100 /zg/ml streptomycin, 5 /zg/ml insulin plus 5 /xg/ml transferrin plus 5 ng/ml selenous acid (added as ITS concentrate), and 0.1% retinal-derived endothelial cell growth factor prepared from bovine retina [11]. Each experiment used endothelial cells from a single umbilical cord. The endothelial cells were initially cultured in a 60 mm dish and then transferred to a 75 cm 2 flask after four to seven days when the cells had become confluent. All surfaces on which endothelial cells were cultured were first gelatin-coated by coverage with 0.2% gelatin for one hour at 37~ Endothelial cells were passaged by removing them from the dish with 0.25% trypsin. For experiments in which endothelial cells were exposed to neutrophils, the endothelial cells were used in the second passage by transferring them from a 75 c m 2 flask to 72 individual 16 mm tissue culture wells and culturing them for an additional two to three days to again reach confluence. Preparation of neutrophils
Human neutrophils were prepared from the leukocyte-rich "buffy coat" from one unit of blood by removal of erythrocytes by gravity sedimentation in 1.63% (W/v) Dextran (final concentration), followed by density centrifugation of the neutrophils through a Ficoll-Hypaque solution to remove mononuclear cells, and hypotonic lysis of remaining erythrocytes [12]. Source leukocytes ("buffy coat"), obtained from a local blood bank, were removed from the collection bag with a 60 ml syringe and combined with a rinse of the bag with 10 ml PBS (154 mM NaC1, 5.59 mM NazHPO4, 1.06 mM KHzPO4, pH 7.4) to which had been added I mg/ml D-glucose
VOLUME 5 N o 4 - 1991
I N H I B I T I O N OF N E U T R O P H I L SUPEROX1DE P R O D U C T I O N
(PBS-glucose). A mixture of two parts PBS-glucose plus one part 4% (W/v) Dextran (D-7265, Sigma Chemical Company, St. Louis, Missouri) in 0.85% NaC1 was prepared, and a volume of this mixture equal to one half the volume of source leukocytes plus rinse was drawn up in the syringe and mixed by inversion. The cell suspension was divided equally using a second 60 cc syringe, and a volume of 4% Dextran in 0.85% NaC1 equal to one half the volume in each syringe was drawn up. After mixing, the syringes were placed vertically, outlet uppermost, just until the red cells settled to form a sharp line (10 to 14 minutes). The upper layer was transferred to a 50 ml centrifuge tube through a piece of tubing by pushing the syringe plunger down against the work surface. The cells were centrifuged for seven minutes at 400 g, the supernatant was discarded, and the cell pellet was resuspended in 40 ml PBSglucose. The cell suspension was layered equally onto eight 3-ml cushions of Histopaque 1077 density centrifugation medium in 15 ml centrifuge tubes, and the gradients centrifuged for 23 minutes at 450 g. The upper layers of mononuclear cells were removed, and the pellets were combined and centrifuged at 320 g for four minutes. The resulting pellet was resuspended in 6 ml deionized water for 20 seconds to lyse remaining red cells, followed by addition of 2 ml of 3.4% NaC1 to restore tonicity. The neutrophils were centrifuged again and resuspended for counting in 20 ml Dulbecco's phosphatebuffered saline (DPBS) (137 mM NaC1, 8.06 mM Na2HPO 4, 1.47 mM KHzPO4, 2.68 mM KCI, 0.88 mM CaCI2, 0.49 mM MgC12) to which had been added 1 mg/ml glucose (DPBS-glucose). To prevent clumping, the cells were resuspended every three minutes until used. All steps were carried out at ambient temperature. To minimize exposure of the neutrophils to endotoxin, PBS, DPBS, and the deionized water used to make all other solutions were obtained commercially from a supplier that routinely tests for endotoxin.
Measurements of neutrophil 02- production
0 2- production was measured by the in situ reduction of ferricytochrome c [13]. Incubation was for 15 to 30 minutes at 37~ in 16 mm tissue culture wells (24-well plates) in a final volume of 0.3 ml DPBS-glucose, containing 3 x 106 neutrophils, 0.119 mM cytochrome c, 2 mM sodium azide, and test reagents, as indicated. Each condition was tested in triplicate; and a second group of three wells for each condition, containing all reagents and cells plus 0.2 mg/ml superoxide dismutase, was included to determine the absorbence of the mixture without cytochrome c reduction. Reagents were usually added as 10-fold concentrated solu-
327
tions in 30 txl volumes. 02 generation was initiated by the addition of the peptide FMLP at 10-6 M or 1 0 - 7 M, as indicated. The tissue culture wells either contained a monolayer of cultured endothelial cells or were precoated with 0.2 ml of an 0.2% gelatin solution for one hour at 37~ In both cases, the wells were rinsed three times with DPBS-glucose before use. Adenosine deaminase, when present, was incubated with the endothelial cells for five minutes before the addition of neutrophils, to remove adenosine. Substances to be tested as inhibitors of neutrophil O2- production were incubated with the complete mixture for five minutes at 37~ before the addition of FMLP. Following the incubation with FMLP, the plates were placed on ice and 0.3 ml of ice-cold 0.125 mg/ml N-ethylmaleimide in 0.85% NaC1 was added to each well to prevent any further change in the oxidation state of the cytochrome c. The contents of the wells were transferred to 12 x 75 mm tubes and centrifuged at 300 g for three minutes. The absorbence of the supernatants at 550 nm was determined spectrophotometrically. The nmoles of O2- that were produced were taken to be equal to the nmoles of cytochrome c reduced. The values were calculated taking the average absorbence of the wells without superoxide dismutase minus the average absorbence of the wells with superoxide dismutase for each condition. We used an extinction coefficient (reduced minus oxidized) of 19,100 M 1 [13]. Since O2- production by human neutrophils in response to FMLP is complete within five to 10 minutes [1,2], the results are expressed as total nmoles 02- produced per 106 neutrophils.
Measurement of endothelial cell damage
Endothelial cell damage by FMLP-stimulated neutrophils was assessed by the increase in release of 3H-DOG, by a modification of a published procedure [14]. Endothelial cell monolayers in 16 mm tissue culture wells were preincubated for four hours in 0.3 ml M199 medium containing 10% heat-inactivated fetal calf serum and 1 txCi/ml 3HDOG. The cells were then rinsed three times with 0.5 ml DPBS (without glucose), containing 0.2% BSA (DPBS-BSA), with the third rinse left on the monolayer for one hour at 37~ Incubating the cells in glucose-free medium results in increased sensitivity to oxidant injury [14,15], possibly due to depletion of cellular reserves of reduced glutathione [15]. At the end of the one-hour incubation, the medium was removed and the appropriate volume of fresh DPBS-BSA was added, followed by the addition of 3 x 1 0 6 neutrophils in 30 /xl, and the agent to be tested in an additional 30/xl DPBS-BSA. The mixture was incubated for five minutes at 37~ before adding FMLP in 30/xl and incubating for 90
7
"ca .-=
~e
4
o
=
%
6
~
[] []
WITH EC WITHOUT EC
*
~/'r
p
To investigate the inhibitory effect of adenosine released by endothelium on neutrophil superoxide (02-) production, we treated confluent monolayers of cultured human umbilical vein endothelial cells with the enzyme adenosine deaminase, and then added human neutrophils. Superoxide (02-) production by human neutrophils stimulated with 10 -6 M formyl-methionyl-leucyl-phenylalanine was inhibited by 49% in the presence of a confluent monolayer of human umbilical vein endothelial cells (5.1 -+ 0.1 versus 2.6 _+ 0.3 nmols O2-/106 neutrophils). Addition of 0.25 U/ml adenosine deaminase to neutrophils plus endothelial cells restored formyl-methionyl-leucyl-phenylalanine-stimulated neutrophil superoxide production to the level seen with neutrophils alone. Deoxycoformycin (10 -4 M), an inhibitor of adenosine deaminase activity, prevented the increase in superoxide production associated with adenosine deaminase addition. The adenosine analogue 5'-(N-ethylcarboxamido)adenosine (3 x 10 -4 M) caused increased inhibition of formyl-methionyl-leucylphenylalanine-stimulated superoxide release by neutrophils in the presence of endothelial cells and prevented neutrophil-mediated endothelial cell damage, as measured by release of 3H-2-deoxy-D-glucose. Pairing 2-chloroadenosine (10 -s M) or 5'-(N-ethylcarboxamido)-adenosine (3 x 10 -4 M) with a cyclic adenosine monophosphate phosphodiesterase inhibitor, 3-isobutyl-l-methyl-xanthine (10 -4 M), produced greater inhibition of neutrophil superoxide production than occurred with either compound alone. The results support the hypothesis that vascular endothelial cells protect themselves from neutrophil attack by releasing adenosine to inhibit superoxide production. (Ann Vasc Surg 1991;5:325-330). KEY WORDS:
Endothelial seeding; superoxide production; adenosine; neutrophils.
Studies of endothelial cell seeding [1] of prosthetic vascular grafts reveal extensive cell loss within 24 hours of restoring blood flow [2]. Emerick and co-workers [3] were able to prevent the loss of endothelial cells from seeded grafts in dogs by making the animals leukopenic. Inspection of the From the Department of Surgery, Indiana University School of Medicine, Indianapolis, Indiana. Reprint requests: Gary R. Gunther, PhD, Department of Physiology and Biophysics MS374, Indiana University School of Medicine, 1100 West Michigan Street, Indianapolis, Indiana 46202.
flOW surfaces of culture-lined canine grafts by scanning electron microscopy six hours after implantation demonstrated neutrophil attachment and endothelial cell loss [4]. These findings support the concept [4] that neutrophils mediate endothelial cell detachment from the graft surface. One of the ways that neutrophils can damage endothelial cells is by the release of superoxide (02-) radicals [5,6]. Adenosine, which is released by human neutrophils [7] and endothelial cells [8], inhibits neutrophil Oe production [7]. Adenosine analogues can inhibit neutrophil adhesion to endothelial cells and neutrophil-mediated endothelial
325
I N H I B I T I O N OF NEUTROPH1L SUPEROXIDE P R O D U C T I O N
326
cell damage in vitro [8]. Adenosine deaminase converts adenosine to inosine, which does not inhibit neutrophil function [7]. As one might expect, adenosine deaminase increases neutrophil-mediated injury to endothelial cells in vitro [8]. We report that adenosine deaminase greatly increases neutrophil O2- production in the presence of endothelial cells. This supports the hypothesis [8] that adenosine release is a mechanism by which endothelial cells protect themselves from damage by neutrophil release of O2-.
ANNALS OF VASCULAR SURGERY
search, Waltham, Massachusetts. Gelatin was obtained from Difco Laboratories, Inc., Detroit, Michigan. D-glucose and sodium chloride were obtained from Mallinckrodt, Inc., Paris, Kentucky. 2-deoxy-D[1-3H] glucose (3H-DOG) (17 Ci/mmol) was purchased from Amersham Corporation, Arlington Heights, Illinois. Ready-Protein liquid scintillation cocktail was obtained from Beckman Instruments, Palo Alto, California. Endothelial cell culture
MATERIAL AND METHODS Experimental design
To analyze the effect of endothelial cells on neutrophil 0 2 - production, human neutrophils were stimulated with the peptide N-formyl-L-methionylL-leucyl-L-phenylalanine (FMLP) in the presence or absence of a monolayer of human umbilical vein endothelial cells in vitro. To determine the role of endogenous adenosine, adenosine deaminase was added and compared to culture wells containing the adenosine deaminase inhibitor deoxycoformycin along with the enzyme. To show that inhibition of neutrophil 0 2 - production was related to a decrease in injury to endothelial cells, an adenosine analogue was added to cultures containing endothelial cells plus FMLP-stimulated neutrophils. 0 2 - production and endothelial cell injury were measured. Materials
Adenosine, 2-chloroadenosine (2-ChlAdn), 5'-(Nethylcarboxamido)-adenosine (NECA), 3-isobutyl-lmethyl-xanthine (IBMX), adenosine deaminase (Type VI, from calf intestinal mucosa), deoxycoformycin, FMLP, cytochrome c (Type VI, from horse heart), superoxide dismutase (from bovine erythrocytes), sodium azide, bovine serum albumin (BSA) (Cat. No. A-2153), M-ethylmaleimide, dextran (MW 225,000), and Ficoll-Hypaque leukocyte separation medium (Histopaque 1077) were purchased from Sigma Chemical Company, St. Louis, Missouri. Collagenase (CLS Type I) was obtained from Worthington Biochemical Corporation, Freehold, New Jersey. Medium M199 with Earle's balanced salts and 25 mM HEPES, fetal bovine serum, 200 mM L-glutamine, 100 mM sodium pyruvate, nonessential amino acids (100X), 2.5% trypsin, phosphate-buffered saline (PBS), Dulbecco's phosphate-buffered saline (DPBS), and deionized water were purchased from Whittaker Bioproducts, Inc., Walkersville, Maryland. Penicillin-streptomycin solution (100X) was obtained from GIBCO, Grand Island, New York. Insufin-transferrin-selenous acid mixture (ITS concentrate) was a product of Collaborative Re-
Human umbilical vein endothelial cells were harvested from umbilical cords obtained from a local hospital. Endothelial cells were dissociated from the vessel wall using 0.13% collagenase in PBS according to an established procedure [9,10]. Endothelial cells were cultured at 37~ in a humidified 5% CO2 atmosphere in M199 medium containing 10% heat-inactivated fetal bovine serum and supplemented with 2 mM L-glutamine, 1 mM pyruvate, 1% nonessential amino acid supplement, 100 U/ml penicillin plus 100 /zg/ml streptomycin, 5 /zg/ml insulin plus 5 /xg/ml transferrin plus 5 ng/ml selenous acid (added as ITS concentrate), and 0.1% retinal-derived endothelial cell growth factor prepared from bovine retina [11]. Each experiment used endothelial cells from a single umbilical cord. The endothelial cells were initially cultured in a 60 mm dish and then transferred to a 75 cm 2 flask after four to seven days when the cells had become confluent. All surfaces on which endothelial cells were cultured were first gelatin-coated by coverage with 0.2% gelatin for one hour at 37~ Endothelial cells were passaged by removing them from the dish with 0.25% trypsin. For experiments in which endothelial cells were exposed to neutrophils, the endothelial cells were used in the second passage by transferring them from a 75 c m 2 flask to 72 individual 16 mm tissue culture wells and culturing them for an additional two to three days to again reach confluence. Preparation of neutrophils
Human neutrophils were prepared from the leukocyte-rich "buffy coat" from one unit of blood by removal of erythrocytes by gravity sedimentation in 1.63% (W/v) Dextran (final concentration), followed by density centrifugation of the neutrophils through a Ficoll-Hypaque solution to remove mononuclear cells, and hypotonic lysis of remaining erythrocytes [12]. Source leukocytes ("buffy coat"), obtained from a local blood bank, were removed from the collection bag with a 60 ml syringe and combined with a rinse of the bag with 10 ml PBS (154 mM NaC1, 5.59 mM NazHPO4, 1.06 mM KHzPO4, pH 7.4) to which had been added I mg/ml D-glucose
VOLUME 5 N o 4 - 1991
I N H I B I T I O N OF N E U T R O P H I L SUPEROX1DE P R O D U C T I O N
(PBS-glucose). A mixture of two parts PBS-glucose plus one part 4% (W/v) Dextran (D-7265, Sigma Chemical Company, St. Louis, Missouri) in 0.85% NaC1 was prepared, and a volume of this mixture equal to one half the volume of source leukocytes plus rinse was drawn up in the syringe and mixed by inversion. The cell suspension was divided equally using a second 60 cc syringe, and a volume of 4% Dextran in 0.85% NaC1 equal to one half the volume in each syringe was drawn up. After mixing, the syringes were placed vertically, outlet uppermost, just until the red cells settled to form a sharp line (10 to 14 minutes). The upper layer was transferred to a 50 ml centrifuge tube through a piece of tubing by pushing the syringe plunger down against the work surface. The cells were centrifuged for seven minutes at 400 g, the supernatant was discarded, and the cell pellet was resuspended in 40 ml PBSglucose. The cell suspension was layered equally onto eight 3-ml cushions of Histopaque 1077 density centrifugation medium in 15 ml centrifuge tubes, and the gradients centrifuged for 23 minutes at 450 g. The upper layers of mononuclear cells were removed, and the pellets were combined and centrifuged at 320 g for four minutes. The resulting pellet was resuspended in 6 ml deionized water for 20 seconds to lyse remaining red cells, followed by addition of 2 ml of 3.4% NaC1 to restore tonicity. The neutrophils were centrifuged again and resuspended for counting in 20 ml Dulbecco's phosphatebuffered saline (DPBS) (137 mM NaC1, 8.06 mM Na2HPO 4, 1.47 mM KHzPO4, 2.68 mM KCI, 0.88 mM CaCI2, 0.49 mM MgC12) to which had been added 1 mg/ml glucose (DPBS-glucose). To prevent clumping, the cells were resuspended every three minutes until used. All steps were carried out at ambient temperature. To minimize exposure of the neutrophils to endotoxin, PBS, DPBS, and the deionized water used to make all other solutions were obtained commercially from a supplier that routinely tests for endotoxin.
Measurements of neutrophil 02- production
0 2- production was measured by the in situ reduction of ferricytochrome c [13]. Incubation was for 15 to 30 minutes at 37~ in 16 mm tissue culture wells (24-well plates) in a final volume of 0.3 ml DPBS-glucose, containing 3 x 106 neutrophils, 0.119 mM cytochrome c, 2 mM sodium azide, and test reagents, as indicated. Each condition was tested in triplicate; and a second group of three wells for each condition, containing all reagents and cells plus 0.2 mg/ml superoxide dismutase, was included to determine the absorbence of the mixture without cytochrome c reduction. Reagents were usually added as 10-fold concentrated solu-
327
tions in 30 txl volumes. 02 generation was initiated by the addition of the peptide FMLP at 10-6 M or 1 0 - 7 M, as indicated. The tissue culture wells either contained a monolayer of cultured endothelial cells or were precoated with 0.2 ml of an 0.2% gelatin solution for one hour at 37~ In both cases, the wells were rinsed three times with DPBS-glucose before use. Adenosine deaminase, when present, was incubated with the endothelial cells for five minutes before the addition of neutrophils, to remove adenosine. Substances to be tested as inhibitors of neutrophil O2- production were incubated with the complete mixture for five minutes at 37~ before the addition of FMLP. Following the incubation with FMLP, the plates were placed on ice and 0.3 ml of ice-cold 0.125 mg/ml N-ethylmaleimide in 0.85% NaC1 was added to each well to prevent any further change in the oxidation state of the cytochrome c. The contents of the wells were transferred to 12 x 75 mm tubes and centrifuged at 300 g for three minutes. The absorbence of the supernatants at 550 nm was determined spectrophotometrically. The nmoles of O2- that were produced were taken to be equal to the nmoles of cytochrome c reduced. The values were calculated taking the average absorbence of the wells without superoxide dismutase minus the average absorbence of the wells with superoxide dismutase for each condition. We used an extinction coefficient (reduced minus oxidized) of 19,100 M 1 [13]. Since O2- production by human neutrophils in response to FMLP is complete within five to 10 minutes [1,2], the results are expressed as total nmoles 02- produced per 106 neutrophils.
Measurement of endothelial cell damage
Endothelial cell damage by FMLP-stimulated neutrophils was assessed by the increase in release of 3H-DOG, by a modification of a published procedure [14]. Endothelial cell monolayers in 16 mm tissue culture wells were preincubated for four hours in 0.3 ml M199 medium containing 10% heat-inactivated fetal calf serum and 1 txCi/ml 3HDOG. The cells were then rinsed three times with 0.5 ml DPBS (without glucose), containing 0.2% BSA (DPBS-BSA), with the third rinse left on the monolayer for one hour at 37~ Incubating the cells in glucose-free medium results in increased sensitivity to oxidant injury [14,15], possibly due to depletion of cellular reserves of reduced glutathione [15]. At the end of the one-hour incubation, the medium was removed and the appropriate volume of fresh DPBS-BSA was added, followed by the addition of 3 x 1 0 6 neutrophils in 30 /xl, and the agent to be tested in an additional 30/xl DPBS-BSA. The mixture was incubated for five minutes at 37~ before adding FMLP in 30/xl and incubating for 90
7
"ca .-=
~e
4
o
=
%
6
~
[] []
WITH EC WITHOUT EC
*
~/'r
p
