From www.bloodjournal.org by guest on July 31, 2017. For personal use only.
Release of Oxygen Metabolites From Chemoattractant-Stimulated
Neutrophils Is Inhibited by Resting Platelets: Role of Extracellular
Adenosine and Actin Polymerization
By Torbjtirn Bengtsson, Stefan Zalavary, Olle Stendahl, and Magnus GrenegArd
The effect of human platelets on chemoattractant-induced
generation of oxygen metabolites in neutrophils was investigated, using luminol-enhanced chemiluminescence (CL).
Resting platelats inhibited the extracellular, but not the intracellular, production of oxygen radicals in formyl-methionyl-leucyl-phenylalanine(fMet-Leu-Phe)-stimulated neutrophils. Maximal effect was obtained at the physiological
neutrophil/platelet ratio of 1/50. Similar resutts were acquired by adding supernatants of platelets, indicating a role
for a soluble factor. Removal of extracellular adenosine by
adenosine deaminase (ADA), or blocking of adenosine-receptors by theophylline, antagonized the inhibitory effects
of platelets (or the equivalent supernatant) on the neutrophil
respiratory burst. In contrast, accumulation of adenosine by
apyrase enhanced the inhibition. Exogenous adenosine
mimicked the effects of platelets on the Met-Leu-Phe-induced respiratory burst. To further assess the role of platelet-derived adenosine, the platelets were fixed with paraformaldehyde. We found that fixed platelets, as well as their
supernatant, inhibited the Met-Leu-Phe-induced CL-response t o the same extent as viable cells. These effects were
also reversed by ADA and theophylline, respectively. A prior
removal of adenosine in the platelet suspension by ADA,
followed by treatment with erythro-9-(2-hydroxy-3-nonyl)adenine (EHNA)t o inactivate ADA, did not reverse the inhibitory action of platelets on the Met-Leu-Phe-induced CL-response in neutrophils. However, if adenosine receptors of
the neutrophil at the same time were blocked with theophylline, the inhibition was significantly reduced. Platelets markedly increased the generation of adenosine in a neutrophil
suspension. This effect was antagonized by S-(4-Nitrobenzyl)-&thioguanosine (NBTG), but unaffected by @-methyleneadenosine5’diphosphate (AMP-CP), indicating that the
platelet-dependent accumulation of adenosine is due t o an
increased release of endogenous adenosine from neutrophils and not t o a degradation of extracellular AMP. In correlation, NBTG, but not AMP-CP, reversed the platelet-mediated inhibition of the fMet-Leu-Phe-induced CL-response in
neutrophils. Consequently, these data suggest that a platelet-derived factor increases the release of endogenously
formed adenosine from neutrophils, terminating the production of oxygen radicals. The inhibition of oxidase activity
was also associated with a platelet-induced polymerization
of actin in the margin of the neutrophils. Treatment of neutrophils with cytochalasin B reversed the effects of platelets,
both on F-actin content and CL-response. In summary, resting platelets limit the release of oxygen radicals from chemoattractant-stimulated neutrophils, thus preventing excessive damage t o host tissues in the vascular space. This effect
is suggested t o be associated with an increased generation
of neutrophil-derived adenosine enhancing an autoregulatory inhibitory pathway, and a peripheral accumulation of
actin filaments forming a barrier for extracellular release of
reactive oxygen radicals.
0 1996 by The American Society of Hematology.
N
EUTROPHIL GRANULOCYTES play a central role
in human host defense against microbial invasion. Release of reactive oxygen metabolites and lysosomal enzymes
by activated neutrophils may, however, cause considerable
tissue damage following ischemia and during inflammatory
processes. It is, therefore, likely that there are regulatory
mechanisms operating to control these destructive activities.
Several endogenous, soluble inhibitors of the neutrophil respiratory burst have been described, eg, adenosine, nitric oxide, fibronectin, platelet-derived growth factor, and soluble
P-~electin.’-~Furthermore, neutrophil activation may be
modulated through interaction with other cells at inflammatory sites, eg, erythrocytes, platelets, and endothelial
of oxygen species, as well as the secretion of granule products, are inhibited by a peripheral accumulation of actin
Consequently, a platelet-mediated elevation of
the neutrophil F-actin content might also affect the microbicidal systems of the neutrophil.
There is increasing evidence suggesting that adenine nucleotides and adenosine serve as important extracellular factors modifying the biologic responses of inflammatory
cells.22,23
For instance, occupation of adenosine triphosphate
(ATP) receptors on neutrophils triggers mobilization of intracellular free calcium, which is considered to prime the cells
for enhanced superoxide anion (0;) responses to Met-Leu-
Platelets or platelet-derived products have been shown
to modify different neutrophil functions including adhesion,
aggregation, chemotaxis, superoxide anion production, and
degranulation.
The intercellular mechanisms during
platelet-neutrophil interaction are poorly characterized, but
involve both soluble mediators, as well as contact-dependent
mechanisms.”-’6We have recently demonstrated that coincubation of neutrophils and platelets, isolated separately from
human blood, results in a cytoskeletal activation shown by
sustained elevation of filamentous actin (F-actin) content in
both cells.” We also found that platelets markedly potentiate
formyl-methionyl-leucyl-phenylalanine(met-Leu-Phe)-induced actin assembly and calcium mobilization in neutrophils.” Several studies have shown that the function of the
respiratory burst oxidase, especially the extracellular release
From the Departments of Medical Microbiology and Pharmacology, Faculty of Health Sciences, Linkoping University, Linkoping,
Sweden.
Submitted September 21, 1994; accepted January 26, 1996.
Supported by grants from The King Gustaf V 80-year Foundation,
The Swedish Society of Medicine, The Swedish Medical Research
Council, The County of Ostergotland, and The Swedish Association
Against Rheumatism.
Address reprint requests to Torbjorn Bengtsson, PhD, Department
of Medical Microbiology, Faculty of Health Sciences, Linkoping
University, S-581 85 Linkoping, Sweden.
The publication costs of this article were defrayed in part by page
charge payment. This article must therefore be hereby marked
“advertisement” in accordance with 18 U.S.C. section 1734 solely to
indicate this fact.
0 1996 by The American Society of Hematology.
0006-4971/96/8710-0010$3.00/0
Blood, Vol87, No 10 (May 15). 1996: pp 4411-4423
441 1
From www.bloodjournal.org by guest on July 31, 2017. For personal use only.
4412
BENGTSSON ET AL
Phe.24Studies by Cron~tein~~
and others indicate that adenosine suppresses the neutrophil respiratory burst induced by
chemoattractants through activation of a specific receptor of
the A2 type.26Platelets release from their dense granules
substantial amounts of ATP and adenosine diphosphate
(ADP), which are subsequently converted to adenosine, and
may thus function as potent regulators of the oxygen radical
production by activated neutrophils. In addition, adenosine
derived from stimulated neutrophils is consideredto function
as an autocrine regulator attenuating cell activation.*'
In the present study, the role of platelets in the control of
the neutrophil respiratory burst was investigated and related
to accumulation of extracellular adenosine and activation of
the actin cytoskeleton.
MATERIALS AND METHODS
The materials and their sources were as follows: dextran and Ficoll-Paque (Pharmacia, Uppsala, Sweden); sodium-metrizoate (Nycomed Pharma, Oslo, Norway); bodipy phallacidin (Molecular Probes, Eugene, OR); S-(4-Nitrobenzyl)-6-thioguanosine
(NBTG) (Research Biochemicals Int, Natick, MA); catalase (Boeringer Mannheim, Germany); adenosine, adenosine deaminase
(ADA), adenosine and 5'nucleotidase kit, apyrase, cytochalasin B
(cyt B), cytochrome c, erythro-9-(2-hydroxy-3-nonyl)-adenine
(EHNA), formyl-methionyl-leucyl-phenylalanine (Met-Leu-Phe),
horseradish peroxidase (HRP), 5-amino-2,3-dihydro-1,4-phtalazinedione (luminol), lysophosphatidylcholine, a,P-methyleneadenosine S'diphosphate (AMP-CP), superoxide dismutase (SOD),
theophylline and thrombin (Sigma Chemical Co, St Louis, MO). All
reagents used were of an analytical grade.
Cell preparation. Polymorphonuclear neutrophil leukocytes
(neutrophils) were isolated from heparinized human peripheral
blood, essentially by the method of Boyum.28In short, erythrocytes
were removed by dextran-sodium metrizoate sedimentation and brief
hypotonic lysis in distilled water. The preparation was then centrifuged on a Ficoll-paque gradient to exclude lymphocytes, monocytes
and platelets. A calcium-free Krebs-Ringer phosphate buffer, supplemented with IO mmom glucose and 1.5 mmol/L MgS04 (KRG; pH
7.3), was used for washing. After isolation, the neutrophils were
suspended in KRG supplemented with 1.1 mmom CaCI2,and stored
on ice before use. Contamination of the neutrophil preparation was
usually lower than 1 platelet to 10 neutrophils. Washed platelets were
obtained essentially as previously de~cribed.'~
In short, a mixture of
five parts of the blood and one part of an acid-citrate-dextrose (ACD)
solution was centrifuged twice for 20 minutes (220g followed by
480g) and as much as possible of the plasma was removed. Finally,
the platelets were resuspended in KRG and stored in plastic tubes at
room temperature before use. To obtain functional but nonactivated
platelets, the isolation was performed without any platelet inhibitors
(except the ACD-solution), and due to this, extra care was taken
when handling the cells. The majority of the isolated platelets exhibited disc shape, and the cells did not aggregate. In some experiments,
intact cells and debris were removed by centrifugation at 9,OOOg for
5 minutes at room temperature in a Microfuge (Beckman Instruments, Inc, Palo Alto, CA) or by ultracentrifugation at 100,OOOg for
30 minutes at 20°C in a XL-70 ultracentrifuge (Beckman Instruments, Inc). Altematively, platelets (5 X 107/mL) were fixed with
paraformaldehyde (2%) for 1 hour at room temperature. Fixation
was stopped by incubating the cells 15 minutes at room temperature
in a buffer containing tris(hydroxymethy1)aminomethane (1.25 X
IO-' mom) and glycine (2.5 X IO-' mom), followed by washing
three times in buffer.
Chemiluminescence. Neutrophils and platelets were preincuMaterials.
bated separately in multiwell-plates (Nunc Intermed, Roskilde, Denmark) at 37°C under stimng conditions. After 5 minutes, the cells
were mixed (giving a final concentration of 2 x I06/mL and 2 x
lo6 to 1 X 108/mL of neutrophils and platelets, respectively, and a
neutrophil/platelet ratio of 111 to 1/50), incubated for another 2
minutes and then stimulated with Met-Leu-Phe
mom). Unstimulated cell suspensions were run in parallel. The cell suspensions
were then rapidly converted to polycarbonate tubes with luminol
(final concentration 50 pmoVL), stimulated with Met-Leu-Phe and
monitored for chemiluminescence (CL) at 37°C in a 6-channel Biolumat LB 9505 (Berthold Co, Wildbaden, Germany). Because peroxidase is often a limiting factor in luminol-dependent CL, extra peroxidase (HRP, 4 U) was added to the tubes in some experiments.
Measurements in the HRP system reflected the total CL activity.
The intracellular part of the CL response was defined as the activity
obtained in the presence of SOD (200 U) and catalase (2,000 U),
large molecular 0; and Hz02 scavengers, respectively. Dimethyl
sulfoxide, used as a solvent vehicle for some substances, was tested
and had, without exception, no effect on CL. Furthermore, inactivation of enzymes (ADA or apyrase) by boiling, totally abrogated their
effects on Met-Leu-Phe-induced CL.
Production of superoxide anions. 0, production was assayed
spectrophotometrically by the reduction of fenicytochrome c (cyt
c), essentially as described by Cohen and C h ~ v a n i e cNeutrophils
.~~
and platelets were incubated and mixed as described in the chemiluminescence method. The assay mixture contained 1 mL cell suspension (2 X IO6 neutrophils/mL with or without 2 X lo6 to 1 x 10'
plateletdml) with 15 mg cyt c. The absorbance change accompanying cyt c reduction was monitored at 550 nm using a Beckman
DU-68 spectrophotometer (Beckman Instruments, Inc, Palo Alto,
CA) equipped with a thermostated (37°C) cuvette holder. Scavering
of 0, by platelets or platelet products was evaluated by using the
cell-free 0; -generating system of xanthine oxidase (0.02 U/mL)
and hypoxanthine (0.6 mmolL).
Generation of extracellular adenosine. The formation of extracellular adenosine in a neutrophil suspension was determined by
using an enzymatic, spectrophotometrical method. The method is
based on the following reactions: Adenosine deaminase (ADA) deaminates adenosine, producing inosine and ammonium ions
(NH:). In a coupled reaction, catalyzed by L-glutamate dehydrogenase (GLDH), NH: reacts with 2-oxo-glutarate in the presence of
reduced nicotinamide adenine dinucleotide (NADH) to form glutamate and NAD. Formation of NAD, which decreases the absorbance
at 340 nm, is directly proportional to the extracellular release of
adenosine. Neutrophils and platelets were incubated and mixed as
described in the chemiluminescence method. The cell suspensions
were then rapidly converted to cuvettes containing an assay mixture
with the following composition: ADA (400 UL), 2-oxoglutarate
(3.7 mmol/L), NADH (0.2 mmoW), GLDH (1 1,OOO UL), P-glycerophosphate (0.2 "om),
and AMP (3.2 mmol/L). The neutrophils
were unstimulated or stimulated with Met-Leu-Phe (IO-' mom),
and the change in absorbance at 340 nm accompanying NAD-formation was monitored at 37°C with a 6-channel Beckman DU-68 spectrophotometer (Beckman Instruments). The method was calibrated
by analyzing the absorbance change induced by buffers containing
known amounts of adenosine.
Determination of F-actin content. Neutrophils and platelets were
incubated and mixed essentially as described in the chemiluminescence assay. After various periods of time, aliquots of cell suspensions were removed, fixed with ice-cold paraformaldehyde (4%),
and stained for F-actin by incubation in a mixture of bodipy phallacidin (0.6 pg/mL) and lysophosphatidylcholine (100 pg/mL). The bodipy fluorescence of individual neutrophils was determined with a
cytofluorometer based on a Leitz MPV I1 microscope photometer
From www.bloodjournal.org by guest on July 31, 2017. For personal use only.
4413
PLATELETS INHIBIT NEUTROPHIL OXIDASE ACTIVITY
connected to a computer system.30Only a single cell was exposed
shortly (<0.8 s) at a time to avoid fading. For each sample, a
minimum of 50 cells was analyzed. Background fluorescence and
the fluorescence of unstained neutrophils were routinely analyzed
and automatically substracted from the fluorescence values of the
cells. In comparison, we found a good correlation between the results
obtained with the procedure described above and the original procedure using NBD phallacidin, respecti~ely.”.~~
RESULTS
(0; and H2O2),the luminol-amplified CL reaction depends
on myeloperoxidase released from primary granules.33Previous studies have indicated that the release of myeloperoxidase could be a limiting factor in the met-Leu-Phe-induced
A
a,
Platelets inhibit the extracellular release of oxygen radicals from fMet-Leu-Phe-stimulated neutrophils. The respiratory burst of neutrophil granulocytes is initiated by
activation of a membrane-bound enzyme complex (NADPHoxidase) that catalyzes the reduction of oxygen by an electron transfer from cytoplasmic NADPH. In this study, we
have used luminol-enhanced chemiluminescence (CL) as an
indicator for oxidase activity.” Exposure of human neutrophils to met-Leu-Phe
m o m ) led to a rapid biphasic
CL response (Fig 1A). The first phase (within 3 minutes) is
considered to mainly reflect extracellular release of oxygen
metabolites, whereas the second phase (around 5 minutes)
is due to intracellular generation.34 Platelets alone did not
show any detectable CL response on met-Leu-Phe stimulation, nor did they induce production of oxygen radicals by
coincubated neutrophils. However, platelets dose-dependently inhibited the met-Leu-Phe-induced activation of the
neutrophil oxidase (Fig 1A). As previously shown, regarding
F-actin and [Ca2’]i,17 this effect was most prominent at a
physiological neutrophNplatelet (NE) ratio between 1/50 to
MOO. At low N/P ratios (1/1 to 1/5), platelets inhibited
the initial peak of the bimodal response, whereas higher
concentrations of platelets affected the overall CL response
(Fig 1A). Stimulating the platelets, at different N/P ratios,
with thrombin (0.5 U/mL) did not significantly change the
extent of inhibition of the neutrophil oxidase activity. Previous studies have shown that neutrophil chemiluminescence
is dependent on cell c~ncentration.~’
Consequently, the observed effects by adding platelets to a neutrophil suspension
might be nonspecific due to an increased cell number. However, increasing the amount of neutrophils, in cell volume
equivalent with lo8 platelets, did not mimic the inhibitory
effects of platelets, but rather increased the CL-response
(data not shown).
In addition to the amount of reactive oxygen species
20
0
C
al
0
u)
0)
.-C
E
-.-
3
10
E
al
c
0
Time
(min)
B
I
al
100
I-‘,
I
\
0
S
0)
0
u)
Q)
.-S
-.-5
50
E
al
c
u
5
10
Time (mln)
C
al
20
0
C
al
*
Fig 1. Time traces of chemiluminescence emitted from Met-LeuPhe-stimulated neutrophils in the absence or presence of platelets.
(A) Neutrophils and platelets were preincubated separately at 37°C
under stirring conditions. After 5 minutes, the neutrophils (2 x loe/
mL, final concentration) were mixed with platelets (2 x l V to 1 x
108/mL, final concentration; NIP ratio 111 to 1/50] or with buffer
(control; broken line) and incubated for another 2 minutes. The cell
suspensions were then monitored for luminol-amplffied chemiluminescence triggered by Met-Leu-Phe (lo-’ mol/LI. IB) Same as (A),
but in the presence of extra peroxidase, HRP I4 U). (C) Same as (A),
but in the presence of the scavengers SOD 1200 UI and catalase (2.000
U). Ordinate, chemiluminescence in cpm x 10’. The figure shows
representative recordings of at least five different experiments.
0
u)
9)
.-C
E
-.-
3
10
E
al
S
0
5
Time
1‘0
(min)
From www.bloodjournal.org by guest on July 31, 2017. For personal use only.
BENGTSSON ET AL
4414
-
100
Total acliviw
H tntracell. activity
80
60
40
20
0
control
1/1
115
1\10
NeutrophillPlatelet
1125
1/50
ratio
Effects of platelets on extra- and intracellular release of
oxygen radicals. Neutrophils and platelets were preincubated separately at 37°C under stirring conditions. After 5 minutes, the neutrophils (2x 106/mL, final concentration)were mixed with platelets 12
x 10‘ to 1 x 108/mL, final concentration; N/P ratio 1/1 to 1/50] or
with buffer (control) and incubated for another 2 minutes. The cell
suspensions were then monitored for luminol-amplified chemiluminescence triggered by Wet-Leu-Phe (lo-’ mol/L) in the presence of
extra peroxidase, HRP (4 U) (total activity;
or the scavengers SOD
1200 U) and catalase (2.000 U) (intracellularactivity; E).The data are
based on integral values over 10 minutes and are expressed as percent of the neutrophil control stimulated in the presence of HRP
(total-control).The values represent the mean ? SEM of six separate
experiments.
CL response in neutrophils.” It is possible that the plateletmediated inhibition of the fMet-Leu-Phe-triggered CL response is due to a depressed granular release of myeloperoxidase. This is not likely, however, as the difference in CL
response between neutrophils incubated with or without
platelets remained, or was even amplified. when extra peroxidase (HRP, 4 U) was added to the system (Fig IB). High
concentrations of platelets almost totally abolished the initial
peak, thus indicating that they negatively modulate the extracellular release of oxygen metabolites.
To separate the intracellular activity of the CL-response,
we have used SOD and catalase. These proteins are large,
membrane-impermeable 0; and HzOz scavengers, respectively. fMet-Leu-Phe stimulation of neutrophils in the presence of SOD (200 U) and catalase (2.000 U) resulted in a
delayed response, without the initial peak (Fig IC). Platelets
reduced only slightly, at a 1/50 N/P ratio, the fMet-LeuPhe-triggered CL response under these conditions (Fig IC),
ie, platelets do not mainly interfere with the intracellular
generation of oxygen metabolites. This is also illustrated in
Fig 2 showing the relation between the intra and extracellular
parts of the met-Leu-Phe-induced CL responses in neutrophils incubated in the presence of increasing numbers of
platelets (analyzed on the integral values over the measuring
time of I0 minutes). Whereas the extracellular release of
oxygen metabolites was markedly reduced, the intracellular
part of the fMet-Leu-Phe-triggered CL response was only
slightly affected and proportionally increased from 18% in
the control (without platelets) to about 3 3 9 in neutrophils
coincubated with platelets at a 1/50 N/P ratio.
To study if similar patterns of inhibition by platelets were
obtained when using another method of assaying the oxidative burst, the effects of platelets on cytochrome c reduction
in stimulated neutrophils were analyzed. This method specifically monitored the extracellular release of 0;. Indeed.
we found that platelets dose-dependently reduced the m e t Leu-Phe-induced 0; production, with an apparent inhibition at a N/P ratio of 1/5 and reaching a maximal inhibition
at 1/50 (85.2% 2 14.9% and 37.2% 2 8.49,respectively.
of the control, mean 2 SEM, n = 5).
Adenosine as nn e.wncelliilnr messenger in the plntelerinrlirced inhihition. Adenosine is a powerful inhibitor of
the respiratory burst of neutrophils and might be derived
from the metabolism of platelet adenine nucleotides by ectonucleotidases in a mixed cell suspension.’’ We found, by
using an ATP-bioluminescent assay, that addition of resting
platelets caused a threefold to fourfold increase in the extracellular concentration of ATP (from about 30 n m o l n in the
control to over 100 nmol/L in the presence of platelets).
met-Leu-Phe stimulation did not further increase the ATP
level in the mixed suspension. Addition of thrombin-stimulated platelets to the neutrophil suspension raised the ATPconcentration to approximately 700 nmolk, thus showing
that the platelets are potent reservoirs of adenine nucleotides.
To elucidate the role of adenosine in our experiments, we
pretreated the cell suspensions with either ADA. which hydrolyzes adenosine to inosine, or with apyrase, which converts ATP and ADP to AMP. presumably leading to an
accumulation of adenosine. The presence of ADA in the
neutrophil control increased the met-Leu-Phe-induced CL
response with 3 1%, whereas apyrase had no significant effect
(Fig 3). A combination of ADA and apyrase caused an enhancement of 43% (Fig 3). Addition of ADA to the platelet
suspension before addition of neutrophils, or during the neutrophil-platelet incubation. markedly neutralized the suppressive effects of platelets on the Met-Leu-Phe-induced
CL response in neutrophils (Fig 3). The met-Leu-Phe-induced generation of oxygen radicals during a 10-minute period was instead higher in a N/P (1/50) mixture containing
ADA compared with the neutrophil control (1 30.6% 2 7.4%
of the control: mean 2 SEM, n = 12). However, the time
courses of the CL responses were quite different. with a
delayed but markedly amplified unimodal response in the
ADA-treated N/P-suspension (Fig 3). These counteracting
effects of ADA were due to an increased intracellular generation of oxygen species (data not shown). This was also reflected by the observation that ADA only partly restored
0; production measured with the cytochrome c reduction
assay, which primarily registers the extracellular release of
0; (data not shown). In contrast, apyrase exerted an additive
effect on the inhibited met-Leu-Phe-induced CL response
in neutrophils coincubated with platelets, further supporting
a role for adenosine (Fig 3). The observed effects of ADA
and apyrase, respectively, were not due to the solvent vehicle, as it had no effect by itself. Furthermore, the specific
action of ADA was confirmed by preincubating the cells
with the ADA inhibitor EHNA. This substance effectively
From www.bloodjournal.org by guest on July 31, 2017. For personal use only.
PLATELETS INHIBIT NEUTROPHIL OXIDASE ACTIVITY
al
0
c
al
0
20
I
4415
A
v)
-
d)
c
10
Q)
0
.E
-.-a
2ol
B
10
E
Q)
c
0
10
5
Time
c
0
1504
T
(mln)
blunted the effects of ADA on the fMet-Leu-Phe-induced
CL response in neutrophils. both in the absence and presence
of platelets (Fig 3). In addition. EHNA decreased the CL
response in the neutrophil control, indicating an endogenous
ADA activity (Fig 3).
To further elucidate the role of adenosine, we preincubated
the neutrophils with the adenosine-antagonist theophylline.
Similar to ADA. theophylline abolished the inhibitory effects
of platelets on the fMet-Leu-Phe-trigsered CL response in
neutrophils (Fig 3). Theophylline also increased the oxygen
radical production in the neutrophil control. suggesting an
accumulation of adenosine in the neutrophil suspension on
activation (Fig 3).
A platelet-mediated reduction of the neutrophil oxidase
activity due to adenosine does not appear to involve conversion of adenosine to inosine because this nucleoside had no
effect on the fMet-Leu-Phe-induced CL response in neutrophils (100.9% t 4.6% of control. mean t SEM. n = 3).
This also argues against a role for inosine in the reversing
effects of ADA on the Wet-Leu-Phe-induced CL response.
The effects of siipern[rtant jliiidv of plnrelers nti oxidcise
crctirir?.i i i neirtropliils. A role for adenosine was supported
by the finding that the supernatant of the platelet suspension
( I X IOX/mL.9.000,q for 5 minutes). devoid of intact cells.
inhibited the fMet-Leu-Phe-induced CL response to the
same extent (33.2% t 3.4% of the control. mean t SEM.
n = 6) as intact platelets, and that this effect was antagonized
by ADA and theophylline. respectively (105.9% t 12.1%
and 64.6% t 1.4%. respectively, of the control: mean t
SEM, n = 6 and 3, respectively) (Fig 4). Apyrase further
magnified the inhibition by the supernatant fluid (not shown).
An extensive inhibition (43.8% 5 5.6% of the control, mean
? SEM. n = 4) was also achieved when mixing neutrophils
with the supematant obtained after an ultracentrifugation of
the platelet suspension (lOO.OO0,q. 30 minutes), an effect
that was antagonized by ADA and theophylline, respectively
(data not shown). However, preliminary experiments show
that the inhibitory capacity of this supematant on neutrophil
oxidase activity is markedly reduced (from 5 5 8 to 20%
inhibition) after passing through an ultrafilter with a molecular weight cut-off of I O kD.
E.ro,qenoiisnderiosine mimics the effectsof plnrelets. The
data so far implicates adenosine as the responsible factor
I
Fig 3. The role of extracellular adenosine in the platelet-mediated
inhibition of the respiratory burst in neutrophils. Neutrophils and
platelets were preincubated separately at 37'C under stirring conditions. After 5 minutes, the neutrophils (2 x 106/mL,final concentration) were mixed with platelets (1 x 108/mL, final concentration, N I
P ratio 1/50; P, solid lines) or with buffer (broken lines) in the absence
or presence of adenosine deaminase (0.25 UImL; ADA), apyrase (10
pg/mL; Ap), theophylline (10 pmol/L; Theo) andlor erythro-942-hydroxy-3-nonyl)-adenine (10 pmol/L; EHNA) and then incubated for
another 2 minutes. The suspensions were thereafter monitored for
luminol-amplified chemiluminescence during stimulation with fMetLeu-Phe (lo-' mol/L). The data are presented as time traces (A and
B; ordinate, chemiluminescence in cpm x lo6) or based on integral
values over 10 minutes and expressed as percent of the untreated
neutrophil control (C). (A) and (B) show representative recordings,
and (C) the mean 2 SEM of six separate experiments.
From www.bloodjournal.org by guest on July 31, 2017. For personal use only.
441 6
BENGTSSON ET AL
A
1‘0
Time
(min)
enhanced on addition of 1 pmol/L adenosine (13.8% 2 2.1%
of control, mean 5 SEM, n = 4), an effect that exceeded
the inhibition caused by 10 pmol/L adenosine in the absence
of platelets (21.0% +. 1.9% of control, mean +. SEM, n =
4). This suggests that an additional factor(s) is involved in
the mechanisms behind platelet-induced restriction of neutrophil oxidase.
The effects offuced platelets on oxidase activity in neutrophils. To further clarify the role of platelet-derived adenosine in the reduction of oxidase activity in neutrophils, we
used paraformaldehyde-fixed platelets, ie, cells unable to
generate adenosine. Interestingly, we found that fixed platelets, dose-dependently and to the same extent as viable platelets, inhibited the met-Leu-Phe-induced CL response (Fig
6). Surprisingly, the same effect was found using a supernatant of the paraformaldehyde-fixed platelets (1 X 108/mL,
9,00Og, 5 minutes) (Fig 7). The inhibitory effects of fixed
platelets (or the equivalent supernatant fluids) were antagonized by both ADA and theophylline (Fig 7). Consequently,
these experiments suggest that the platelet-mediated inhibition of the neutrophil respiratory burst is due to an increased
generation of adenosine from neutrophils.
The role of neutrophil-derived adenosine in the regulation
of oxygen radical production. As previously shown, activated neutrophils release adenosine, which is considered to
negatively modulate cellular
In support, we
found in this study that removal of extracellular adenosine
with ADA, or blocking of adenosine receptors with theophyl-
control
P-sup
P-sup
P-supP
ADA
The0
Fig 4. Effect of platelet supernatants on fMet-Leu-Phe-induced
chemiluminescence. Neutrophils and platelet supernatants (9,OOOg.
5 minutes) werepreincubated separately at 37°C under stirring conditions. After 5 minutes, the neutrophils (2 x 106/mL, final concentration) were mixed with the platelet
supernatants (the supernatant of
1 x 108platelets/mL;P-sup, solid lines) or with buffer (control; broken
line) in the absence or presence of adenosine deaminase (0.25 U/
mL; ADA), or theophylline ( I O pmol/L; Theo), and then incubated for
another 2 minutes. In parallel, neutrophils were incubated
with intact
platelets (N/P ratio 1/50) as previously described (P, solid line). The
suspensions were thereafter monitored for luminol-amplified
chemiluminescence during stimulation with
fMet-Leu-Phe ( I O ” mol/L). The
data are presented as time traces (A; ordinate, chemiluminescence
in cpm x IO6) or based on integral values over 10 minutes and expressed as percent of the untreated neutrophil control
(B). (A) Shows
representative recordings, and (B) the mean k SEM of five separate
experiments.
in platelet-mediated inhibition of neutrophil oxygen radical
generation. In support, addition of exogenous adenosine
(lo-’ to
mol/L) mimicked the inhibitory effects of platelets on Met-Leu-Phe-induced generation of oxygen radicals
in neutrophils (Fig 5). The inhibition at 0.1 pmol/L of adenosine (33.2% +. 3.9% of control, mean +- SEM, n = 4) corresponded with the inhibition obtained at a N/P ratio of 1/50.
Interestingly, the platelet-mediated inhibition was markedly
-9
-8
-7
-6
-5
log [Adenosine](M)
Fig 5. Effects of exogenous adenosine on fMet-Leu-Phe-induced
chemiluminescence. Neutrophils were preincubated at 37°C under
stirring conditions in the absence or presence of cytochalasin B (5
pg/mL; cyt B). After 5 minutes, the neutrophils (2 x 106/mL, final
concentration) were mixed with buffer containing differentconcentrations
of
adenosine
to
mol/L) and then incubated for
another 2 minutes. The suspensions were thereafter monitored for
luminol-amplified chemiluminescence during stimulation with fMetLeu-Phe
mol/L). The data are based onintegral values over 10
minutes and expressed as percent of the neutrophil control (without
adenosine). The values represent the mean rf: SEM of five separate
experiments.
From www.bloodjournal.org by guest on July 31, 2017. For personal use only.
4417
PLATELETS INHIBIT NEUTROPHIL OXIDASE ACTIVITY
E
0
0.20 pmol/minute, mean ? SEM. n = 5)) and a Net-LeuPhe-stimulated (1.96 ? 0.24 pmol/minute mean 2 SEM. n
= 5) neutrophil suspension. Similar effects were obtained
when mixing neutrophils with platelet supernatants ( 1 X IO’/
mL, 9.000g. 5 minutes). The generation of adenosine in a
platelet suspension was approximately 0.05 pmol/minute.
To identify the mechanisms by which adenosine is generated in neutrophil suspensions, with or without platelets. the
neutrophils were treated with the S’nucleotidase inhibitor
100
r
0
E
9)
0
C
Q
0
v)
Q
-E
-
80
60
40
A
C
3
20
I
E
Q
r
-
c>
control
1/1
1/5
1/10
NeutrophillPiatelet
1/25
1/50
ratio
Fig 6. Effects of fixed platelets on met-Leu-Phe-induced chemiluminescence. Platelets were fixed with paraformaldehyde (4%) and
then thoroughly washed (see Materials and Methods). Neutrophils
and fixed platelets were preincubated separately at 37°C under stirring conditions. After 5 minutes, the neutrophils (2 x 106/mL, final
concentration) were mixed with platelets (2 x lo6 t o 1 x 10*/mL,
final concentration; N/P ratio 1/1 t o 1/50) or with buffer (control) and
incubated for another 2 minutes. The cell suspensions were then
monitored for luminol-amplified chemiluminescence triggered by
fMet-Leu-Phe (lo-’ mol/L) in the presence of extra peroxidase, HRP
(4 U). The data are based on integral values over 10 minutes and
expressed as percent of the neutrophil control. The values represent
the mean 2 SEM of four separate experiments.
5
Time
P-fix
El
100
line, potentiated oxygen radical production in neutrophils
triggered by fMet-Leu-Phe (Fig 3). Furthermore, inhibition
of endogenous ADA activity by EHNA reduced the response
(Fig 3). To further assess whether the platelet-mediated inhibition of the neutrophil respiratory burst was attributable
to adenosine soley derived from neutrophils, the following
experiments were performed. Viable platelets or their supernatant were treated with ADA to remove adenosine. ADA
was then inactivated with EHNA, on which neutrophils were
added and stimulated with met-Leu-Phe. This treatment did
not reverse the inhibitory effects of platelets on fMet-LeuPhe-induced oxygen radical production. However, if adenosine receptors of the neutrophil at the same time were
blocked with theophylline, the inhibition was significantly
reduced. This suggests that a platelet-derived factor increases
the generation of adenosine from neutrophils, terminating
the activation of the oxidase.
Platelets increase the generation <$adenosine from new
trophils. The extracellular release of adenosine was analyzed with an enzymatic, spectrophotometrical assay (Fig 8).
We found that a suspension of neutrophils (2 X IO”/mL),
incubated at 37°C under stirring conditions, generated adenosine at rate of 0.17 ? 0.03 pmol/minute (mean 2 SEM, n
= 5 ) . This generation was almost doubled on addition of
IO-’ mol/L Net-Leu-Phe (0.31 ? 0.07 pmol/minute, mean
? SEM, n = 5). Platelets (1 X lO*/mL) markedly enhanced
the release of adenosine both in an unstimulated (1.50 ?
10
(min)
B
P-firrup
TT
50
0
untreated
ADA
Theophylline
Fig 7. Fixed platelets inhibit fMet-Leu-Phe-induced generation of
oxygen radicals in neutrophils through an adenosine-dependent
mechanism. Platelets were fixed with paraformaldehyde (4961, thoroughly washed, and centrifuged in a microfuge (see Materials and
Methods). Neutrophils and fixed platelets, or the equivalent supernatant, were preincubated separately at 37°C under stirring conditions.
After 5 minutes, the neutrophils (2 x 106/mL, final concentration)
were mixed with platelets (1 x 1O8/mL, final concentration; N/P ratio
1/50; P-fix, solid lines or filled bars), the corresponding supernatant
(P-fixsup, solid lines or hatched bars) or with buffer (broken line), in
the absence or presence of adenosine deaminase (0.25 U/mL; ADA),
or theophylline (10 pmol/L; Theo), and then incubated for another
2 minutes. The cell suspensions were then monitored for luminolamplified chemiluminescence triggered by fMet-Leu-Phe (lo-’ m o l l
L). The data are presented as time traces (A; ordinate, chemiluminescence in cpm x lo6) or based on integral values over 10 minutes
and expressed as percent of the untreated neutrophil control IB). (AI
Shows representative recordings, and (B) the mean rt: SEM of five
separate experiments.
From www.bloodjournal.org by guest on July 31, 2017. For personal use only.
BENGTSSON ET AL
4418
-eE
h
m~
>
T
N+P
0
In
a
0,
2!
l,o
al
.-e
In
0
C
a,
-
U
Q:
0,o
unstim.
fMLP
Fig 8. Effect of platelets on the adenosine release in a neutrophil
suspension. Neutrophils and platelets were preincubated separately
at 37'C under stirring conditions. After 5 minutes, the neutrophils (2
x 106/mL, final concentration) were mixed with the platelets (1 x
108/mL, final concentration; N P, hatched bars) or with buffer IN,
filled bars) and then incubated for another 2 minutes. The suspensions, unstimulated or stimulated with fMet-Leu-Phe (lo-' mol/L;
fMLP1, were thereafter monitored for adenosine release using an enzymatic, spectrophotometrical method described in Materials and
Methods. The values represent the mean rt SEM of five separate
experiments.
+
These effects are optimal at a physiological N/P ratio of I /
SO." I n this study. we further characterize the effects of
platelets on the morphology and F-actin distribution in coinciibated neutrophils. Resting neutrophils showed an uniform
cytoplasmic distribution of F-actin (Fig 1 I A). Stimulation
with fMet-Leu-Phe ( I O ' mol/L) caused an extensive transient actin polymerization in the margin of the neutrophil
(Fig 1 IB). which thereafter became more polarized uith Factin mainly localized :it one end of the cell (not shown).
Platelets induced a peripheral formation of actin filaments
in coincubated neutrophils (Fig I IC). and potentiated severalfold a subsequent fMet-Leu-Phe-induced actin polymerization (Fig 1 ID). In these neutrophils. ;i broad peripheral
ring of F-actin was prominent for at least IO minutes (not
shown). As indicated in Fig I 1 D. fMet-Leu-Phe stimulation
increased the binding between neutrophils and platelets. Cytochnlasin R (cyt B). which totally blocked the fMet-LeuPhc-induced changes in both neutrophil shape and F-actin
content (Fig I IE), also reduced the effects of platelets on
actin polymerization in resting and Met-Leu-Phe-stimulated neutrophils (Fig I 1 F). Preliminary experiments indicate that platelet supernatants increase the F-actin content in
both unstimulated and t~et-Leu-Phe-stimulatedneutrophils.
L
"
"
.N
AMP-CP or the adenosine transport inhibitor NRTG. We
found that the cxtrxellular accumulation of ndcnosine. both
in the absence and presence of platelets. was unaffected by
AMP-CP. but reduced by NRTG (data not shown). Conscqiicntly. these data suggest that adenosine accumulation in
;I neutrophil suspension is not due to ;i S'nucleotidase-mediatcd tlcgradation of AMP. but to a release of endogenously
generated ndenosinc. This outward transport of endogenous
aclcnosine is obviously enhanced by platelets.
To correlate the degree of adenosine generation with oxygen radical production. the inhibitors above were used in the
chemiluminescence assay. We found that NBTG potentiated
the IMet-Leu-Phe-induced CL response in neutrophils.
while AMP-CP was ineffective. and moreover. the plateletmediated inhibition of the response was reversed by NBTG,
hut not irifliicnced by AMP-CP (Fig 9). As a comparison,
exogenous AMP ( IO j " / L ) reduced the chemotactic peptide-induced generation of oxygen species, both in the absence nntl presence of platelets (Fig 9). However. these effects were reversed by AMP-CP. Consequently, these data
indicate that neutrophils express. to a certain extent.
S'nuclcotidase activity. However. an adenosine accumulation clue t o dcgrad:ition of AMP is apparently not involved
in the platelet-mediated inhibition of the respiratory burst in
ne ut roph i Is.
PINri~lctsillcl//cY~crc'till flssel?lhl~
ill horh r'esrill<qf / l l f l j M ~ +
L i ~ r r - P h ~ ~ - . r t i r i i r r l ~~icvrtrophils.
/re~/
The major part of the
neutrophil function is dependent on a dynamic reorganization of the actin cytoskeleton. As recently reported. and
shown in Fig IO. the presence of platelets induces a rapid
and sustained elevation of the F-actin content in unstimulated
neutrophils and markedly potentiates a subsequent actin polymerization in fMct-Leu-Phe-stiniiilatcd neutrophils."
N+P
T
"
U
n
a
a
U
I
W
I
m
I
+
n
E
L
C
0
3
e;
I
a
Fig 9. The role of adenosine transport and AMP degradation in the
platelet-mediated inhibition of the respiratory burst in neutrophils.
Neutrophils and platelets were preincubated separately at 37'C under
stirring conditions. After 5 minutes, the neutrophils (2 x 106/mL, final
concentration) were mixed with platelets (1 x 108/mL,final concentration, NIP ratio 1/50; N P, hatched bars) or with buffer IN, filled
bars) in the absence or presenceof S-(4-Nitrobenzyl)-6-thioguanosine
(100 pmol/L; NBTG), n,,3"thyleneadenosine5'diphosphate
(100
pmol/L; AMP-CP) andlor adenosine monophosphate (10 pmol/L;
AMP), and then incubated for another 2 minutes. The suspensions
were thereafter monitored for luminol-amplified chemiluminescence
during stimulation with fMet-Leu-Phe (10 ' mol/L). The data are
based on integral values over 10 minutes and expressed as percent
of the untreated neutrophil control. The values represent the mean
? SEM of four separate experiments.
+
From www.bloodjournal.org by guest on July 31, 2017. For personal use only.
4419
PLATELETS INHIBIT NEUTROPHIL OXIDASE ACTIVITY
actin filament system. Comparison of the effects of platelets
and exogenous adenosine, respectively, on the Met-LeuPhe-induced CL response in cyt B-treated neutrophils suggests that the platelet suspension contributes with an adenosine concentration around
moVL. However, because
adenosine at this level reduced the CL response with only
25% in neutrophils nontreated with cyt B, this indicates that
some other factor, possibly increasing the F-actin content, is
involved in the platelet-induced inhibition of oxygen radical
production in neutrophils.
DISCUSSION
0
1
2
3
4
5
Time(min)
Fig 10. Effect of platelets on the F-actin content in resting and
fMet-Leu-Phe-stimulated neutrophils. Neutrophils and platelets
were preincubatedseparately a t 37°C under stirring conditions. After
5 minutes, the neutrophils (2 x 1OB/mL,final concentration) were
mixed with platelets (1 x lV/mL, final concentration; N/P ratio I /
50) or with buffer and incubated for another 2 minutes followed
by stimulation with Met-Leu-Phe (lo-’ molll; arrow). After various
periods of time, aliquots of the cell suspension were fixed in paraformaldehyde (4%) and then selectivaly stained for F-actinwith bodipy phallacidin.The fluorescencewas quantified in single neutrophils
that had been incubated in the absence [open squares) or presence
of platelets (closed squares). The valuea representthe mean & SEM
of six separate experiments[the error bars of the points at 2 minutes
lie within the sire of the symbols).
Platelet-mediated inhibition of oxygen radical generation
is coupled to elevated F-actin levels in neutrophils. The
role of the actin cytoskeleton in the platelet-mediated effects
on the neutrophil oxidase activity was studied by pretreating
the neutrophils with the actin-polymerizing inhibitor cyt B.
Cyt B alone did not induce any CL response, but caused an
amplified and prolonged generation of oxygen radicals in
Met-Leu-Phe-stimulated neutrophils (Fig 12). The suppressive effects of platelets on the Met-Leu-Phe-induced
CL response were totally abolished in neutrophils pretreated
with cyt B (Fig 12). Instead, platelets potentiated the M e t Leu-Phe-triggered generation of oxygen metabolites in cyt
B-treated neutrophils, with a maximum at a N/P ratio of 1/
10 (132% 5 29% of the control; mean 2 SEM, n = 4, Fig
12). Similar results were obtained when replacing cyt B with
cytochalasin D (not shown).
To elucidate the relationship between changes in the actin
cytoskeleton and the extracellular level of adenosine in the
platelet mediated inhibition of oxygen radical production,
we analyzed the effects of different concentrations of adenosine on the Met-Leu-Phe-induced CL response in neutrophils, treated or nontreated with cyt B (Fig 5). Although
adenosine was still able to inhibit Met-Leu-Phe-triggered
production of oxygen radicals in cyt B-treated neutrophils,
the sensitivity for exogenous adenosine was significantly
reduced. This indicates that adenosine regulates the neutrophil oxidase activity through a mechanism involving the
Several studies indicate an intimate interaction between
neutrophils and platelets, modulating their inflammatory and
hemostatic functions. The present work suggests that platelets act as regulators preventing inappropriate release of reactive oxygen metabolites by activated neutrophils.
We found that resting platelets inhibit Met-Leu-Phe-induced generation of oxygen metabolites in neutrophils, measured as luminol-enhanced chemiluminescence. These effects are not due to inactivation of Met-Leu-Phe or
inhibition of its receptor, as we have previously shown that
other neutrophil responses triggered by Met-Leu-Phe, ie,
actin polymerization and calcium mobilization, are potentiated.I7The inhibition in chemiluminescence showed a maximum effect around a N/P ratio of 1/50, which is close to the
situation in blood. Stimulation of neutrophils with Met-LeuPhe induces both a rapid extracellular release of oxygen
metabolites and a slow, more sustained intracellular generat i ~ n Previous
.~~
studies have indicated that the Met-LeuPhe-induced CL response in neutrophils is dependent on
the presence of myeloper~xidase.~~
It is possible that the
platelet-mediated inhibition is associated with a depressed
release of myeloperoxidase from primary granules of the
neutrophil. However, we found that the difference in CL
response between neutrophils incubated with or without
platelets remained when extra peroxidase was added to the
system. This indicates that platelets do not affect the release
of myeloperoxidase, but instead modulate the activity of the
0;-producing NADPH-oxidase in neutrophils. In support,
platelets inhibited the Met-Leu-Phe-induced extracellular
release of O;, assayed with the cytochrome c-reduction
method. Addition of SOD and catalase to the CL system
enabled us to determine the intracellular activity. Interestingly, platelets only slightly reduced the intracellular part of
the Met-Leu-Phe-induced respiratory burst. Consequently,
platelets inhibit the neutrophil ability to release oxygen metabolites to the extracellular millieu.
Adenine nucleotides have been reported to both enhance
and inhibit 0; production in activated n e ~ t r o p h i l s . ~
A~ * ~ ~
possible explanation for these contradictory results might be
differences in how the neutrophils are exposed to adenine
nucleotides and the degree of extracellular conversion to
adenosine by ATPases and nucleotidases. Several investigations have suggested that adenosine functions as an extracellular messenger controlling the activity of the respiratory
burst oxidase and that its effects are related to ATP and ADP
released from platelet^.'^.^^ In this study, we present several
data supporting a role for adenosine as the responsible factor
From www.bloodjournal.org by guest on July 31, 2017. For personal use only.
4420
BENGTSSON ET AL
Fig 11. Platelets affect the Factin distribution in resting and
Wet-Leu-Phe-stimulated neutrophils. Neutrophils (2 x 106/mL,
final concentration) and platelets
(1 x W/mL, final concentration)
were incubated at 37°C separately
or together (N/P ratio 1/50). stimulated with Met-Leu-Phe 110-’
mol/L), fixed with paraformaldehyde 14%) and selectively stained
for F-actin with bodipy phallacidin. The cell suspensions were
placed batween a microscopic
glass and a coverslip, and the
fluorescence of the cells was examined and photographed in a
confocal scanning laser microscope (Sarastro 1OOO). The micrographs represent thin (0.6 am)
sections (A) nontreated neutrophils, bar 5 pn; (BI Wet-Leu-Phestimulated (lo-’ mol/L, 30 s)neutrophils; IC) Neutrophils in the
presence of platelets; (D) M e t Leu-Phe-stimulated 110-’ mol/L,
30 s) neutrophils in the presence
of platelets, (E) Cytochalasin Btreated (5 pg/mL) neutrophils
stimulated with Wet-Leu-Phe
(10 mol/L 30 SI, (Fl Cytochalasin
B-treated (5 pg/mL) neutrophils
stimulated with Wet-Leu-Phe
(lo-’ mol/L, 30 SI in the presence
of platelets.
in the platelet-mediated inhibition of the Net-Leu-Phe-induced generation of oxygen metabolites: ( 1 ) platelets enhance the generation of adenosine in a neutrophil suspension,
(2) removal of adenosine by ADA reversed, whereas accumulation of adenosine by apyrase potentiated the plateletmediated effects, (3) blocking of adenosine receptors of the
neutrophil with the adenosine receptor antagonist theophyl-
line abolished the inhibitory effects of platelets, (4) a supernatant fluid of the platelet suspension had the same effect
as intact cells, an effect that was reversed by ADA and
theophylline, respectively, and (5) addition of exogenous
adenosine mimicked the effects of platelets. The counteracting effects of ADA on the platelet-mediated inhibition of
the neutrophil oxidase activity might be due to an accumula-
From www.bloodjournal.org by guest on July 31, 2017. For personal use only.
PLATELETSINHIBITNEUTROPHILOXIDASEACTIVITY
4421
A
i
Time
10
(min)
1
150
IS
1
control
1111/15
1/10
2/ 5 0
Neutrophil/Platelet
ratio
Fig12.Chemiluminescence
emitted from cyt B-treated neutrophils stimulated with fMet-Leu-Phe in the absence or presence of
platelets. Neutrophils (2 x 106/mL, final concentration), pretreated
with cyi B (5 pg/mL) for 5 minutes, were incubated in the absence
(control; broken line) or presence of platelets (2 x 10' to 1 x 10*/mL,
final concentration; N/P ratio 1/1 to 1/50) for 2 minutes and then
monitored for luminol-amplified chemiluminescence triggered by
fMet-Leu-Phe (10" mol/L). The data are presented as time traces (A;
ordinate, chemiluminescence in cpm x 10') or based on integral values overIO minutes and expressed aspercent of the neutrophil control (B). (A) shows representative recordings,and (B) the mean f
SEM of five separate experiments.
tion of inosine and hypoxanthine, ie, the products of deamination of adenosine. However, neither inosine (this study)
nor hypoxanthine' affect Net-Leu-Phe-induced production
of oxygen radicals in neutrophils.
To further assess the role of adenosine and other soluble
factors in the platelet-mediated inhibition, the platelets were
fixed with paraformaldehyde and then thoroughly washed.
These cells had no metabolism and were thus incompatible to
generate adenosine and its related metabolites. Interestingly,
fixed platelets, as well as their equivalent supernatants, inhibited the Met-Leu-Phe-triggered production of oxygen radi-
cals to the same extent as viable platelets and, furthermore,
these effects were antagonized by ADA and theophylline,
respectively. Pretreatment of a suspension of viable platelets
(or the equivalent supernatant) with ADA to remove extracellular adenosine, followed by EHNA to inactivate ADA,
did not reverse the inhibitory effects on the Net-Leu-Pheinduced production of reactive oxygen species in neutrophils. However, if adenosine receptors of the neutrophil at
the same time were blocked with theophylline, the inhibition
was significantly reduced. Consequently, these experiments
suggest that platelets induce a generation of adenosine or
its related metabolites from coincubated neutrophils. It has
previously been shown that neutrophils can be stimulated,
eg, by Met-Leu-Phe, to release adenosine and related nucleotide and nucleoside metabolite^.^^ In this study, we found
that incubation of neutrophils at 37°C under stirring conditions is associated with a release of adenosine, which is
markedly amplified in the presence of platelets. This release
reaction has been suggested to be an autoregulatory loop,
where adenosine bind to A,-receptors on the neutrophil surface and thereby inhibit further neutrophil activation, presumably by elevating the intracellular ~AMP-level.*'*~*
This
is also supported in our study, where removal of extracellular
adenosine (with ADA) or blocking of adenosine receptors
(with theophylline) potentiate Net-Leu-Phe-induced respiratory burst in neutrophils, whereas inhibition of endogenous
ADA activity of the neutrophil (with EHNA) instead reduces
the response. Furthermore, we found that the extracellular
accumulation of adenosine was reduced by the adenosinetransport inhibitor NBTG, and that this effect was associated
with an increased production of oxygen metabolites in m e t Leu-Phe-stimulated neutrophils, both in the absence and
presence of platelets. In contrast, the S'nucleotidase inhibitor
AMP-CP was unable to affect either adenosine generation
or oxygen radical production, thus arguing against a role for
adenosine formed by degradation of extracellular AMP. In
conclusion, this study indicates that a platelet-derived factor
increases the release of endogenously generated adenosine
from neutrophils, leading to a termination of the respiratory
burst.
The actin filament network may regulate the activity of
the respiratory burst oxidase. We and others have shown
that inhibitors of actin polymerization potentiate agonistinduced generation of oxygen radicals.'"21 Furthermore,
Woodman et a139have suggested not only a functional, but
even a physical link, between the respiratory burst oxidase
and the actin filament system. Consequently, our observation
that platelets induce a peripheral accumulation of F-actin
in resting and Met-Leu-Phe-stimulated neutrophils might
further explain the inhibitory effect on Met-Leu-Phe-induced CL in neutrophils. This idea was tested by mixing
platelets with neutrophils pretreated with cyt B, followed by
Net-Leu-Phe stimulation. We found that both the formation
of actin filaments and the inhibition of oxidase activity induced by platelets were abolished. The extensive peripheral
accumulation of actin filaments in neutrophils incubated with
platelets might thus inhibit the movement and the assembly
of the different components leading to a functional oxidase,
and thereafter its access to the plasma membrane for extra-
From www.bloodjournal.org by guest on July 31, 2017. For personal use only.
4422
cellular release of reactive oxygen species. Recent data also
show that the platelet-induced effects on the actin cytoskeleton in the neutrophil correlate with a potentiated phagocytosis of IgG-opsonized particles.40
Several data presented in this study disagree with the idea
that the inhibitory effects of platelets on oxygen radical generation in neutrophils are due to mechanisms of platelets
consuming oxygen metabolites, eg, superoxide dismutase
activity or the glutathione cycle: (1) platelets did not affect
0; production in a cell-free system in which 0; was generated with xanthine-xanthine oxidase, (2) the platelet-mediated inhibition was reversed by ADA and theophylline, respectively, and (3) no significant inhibition was observed by
platelets on neutrophils pretreated with cyt B. Furthermore,
we have recently shown that platelets markedly enhance the
oxidative burst triggered by IgG-opsonized yeast particles
in human ne~trophils.~"
The intracellular processes underlying adenosine regulation of neutrophil activities are still uncertain, but have been
suggested to involve changes in receptor activity, cytoskeletal dynamics, and generation of intracellular messenger^.'^
We have recently shown that adenosine, through A2-receptor
activation, inhibits FcR-mediated phagocytosis by elevating
the intracellular concentration of cyclic AMP:' However, a
role for adenosine-triggered elevation of CAMPin the inhibitory effects by platelets is less possible, as CAMP has been
shown to inhibit increases in both F-actin content and cytosolic free calcium in Met-Leu-Phe-triggered neutrophil~,'".~~
responses shown to be potentiated in the presence
of platelets." Cronstein et alU have suggested that adenosine
inhibits Met-Leu-Phe-induced 0; production by promoting
a rapid association of occupied Met-Leu-Phe receptors with
the cytoskeleton. This idea is attractive as we found that (1)
platelets increase the submembraneous actin cytoskeleton,
which might favor an adenosine-induced receptor-cytoskeleton association, and (2) the suppressive effects by platelets,
as well as by exogenous adenosine, on the Met-Leu-Pheinduced CL response are reduced when actin polymerization
in neutrophils is inhibited by cyt B. However, we have previously shown that platelets markedly potentiate other MetLeu-Phe-induced processes, eg, calcium mobilization,"
thus arguing against a mechanism involving an inactivation
of the Met-Leu-Phe receptor. Consequently, our study suggests that the effects of platelets are attributable to an accumulation of adenosine inhibiting the neutrophil oxidase activity through a pathway yet unidentified. The nature of the
platelet-derived factor, generating adenosine from neutrophils, as well as the mechanisms of action of adenosine,
require further studies.
In summary, we report that resting platelets inhibit the
neutrophil ability to release reactive oxygen metabolites in
response to a chemoattractant. We suggest that this effect is
attributable to a platelet-derived factor, which increases the
generation of adenosine from neutrophils and thereby enhances an autocrine inhibitory pathway. Furthermore, the
inhibition is associated with an increased peripheral formation of actin filaments, which might form a barrier for an
extracellular release of oxygen radicals. These platelet-related defense mechanisms might be relevant in controlling
BENGTSSON ET AL
the harmful consequences of an excessive neutrophil activation during hemostasis and inflammation.
REFERENCES
1. Cronstein BN, Kramer SB, Weissmann G, Hirschhorn R:
Adenosine: A physiological modulator of superoxide anion generation by human neutrophils. J Exp Med 158:1160, 1983
2. Rubanyi GM, Ho EH, Cantor EH, Lumma WC, Parker Botelho
LH: Cytoprotective function of nitric oxide: Inactivation of superoxide radicals produced by human leukocytes. Biochem Biophys Res
Commun 181:1392, 1991
3. Yang KD, Augustine NH, Shaio M-F, Bohnsack JF, Hill HR:
Effects of fibronectin on actin organization and respiratory burst
activity in neutrophils, monocytes and macrophages. J Cell Physiol
158:347, 1994
4. Wilson E, Laster SM, Gooding LR, Lambeth JD: Plateletderived growth factor stimulates phagocytosis and blocks agonistinduced activation of the neutrophil oxidative burst: A possible cellular mechanism to protect against oxygen radical damage. Proc Natl
Acad Sci USA 84:2213, 1987
5. Wong CS, Gamble JR, Skinner MP, Lucas CM, Berndt MC,
Vadas MA: Adhesion protein GMP140 inhibits superoxide anion
release by human neutrophils. Proc Natl Acad Sci USA 88:2397,
1991
6. Forslid J, Hed J, Stendahlo: Erythrocyte enhancement of C3bmediated phagocytosis by human neutrophils in vitro: A combined
effect of the erythrocyte complement receptors CR1 and erythrocyte
scavengers to reactive oxygen metabolites (ROM). Immunology
55:97, 1985
7. Moon DG, van Der Zee H, Weston LK, Gudewicz PW, Fenton
I1 JW, Kaplan JE:Platelet modulation of neutrophil superoxide anion
production. Thromb Haemost 63:91, 1990
8. Del Maschio A, Bazzoni G, Dejana E: Polymorphonuclear
leukocyte adhesion to endothelial cells is inhibited by resting platelets. Arterioscler Thromb 12:1420, 1992
9. Godin C, Caprani A, Dufaux J, Flaud P: Interactions between
neutrophils and endothelial cells. J Cell Sci 106:441, 1993
10. Del Maschio A, Dejana E, Bazzoni G: Bidirectional modulation of platelet and polymorphonuclear leukocyte activities. Ann
Hematol 67:23, 1993
11. Weksler BB: Platelets, in Gallin JI, Goldstein IM, Snyderman
R (eds): Inflammation: Basic Principles and Clinical Correlates, (ed
2). New York, NY, Raven, 1992, p727
12. de Bruijne-Admiraal LG, Modderman PW, Von dem Borne
AEGKr, Sonnenberg A: P-selectin mediates Caz'-dependent adhesion of activated platelets to many different types of leukocytes:
Detection by flow cytometry. Blood 80:134, 1992
13. Deuel TF, Senior RM, Huang JS, Griffin GL: Chemotaxis of
monocytes and neutrophils to platelet-derived growth factor. J Clin
Invest 69:1046, 1982
14. Ruf A, Schlenk RF, Maras A, Morgenstern E, Patscheke H:
Contact-induced neutrophil activation by platelets in human cell
suspensions and whole blood. Blood 80:1238, 1992
15. Spisani S, Giuliani AL,Cavalletti T, Zaccarini M, Milani L,
Gavioli R, Traniello S: Modulation of neutrophil functions by activated platelet release factors. Inflammation 16:147, 1992
16. Ward PA, Cunningham TW, McCullough KK, Phan SH,
Powell J, Johnson KJ: Platelet enhancement of 0; responses in
stimulated human neutrophils. Identification of platelet factor as adenine nucleotide. Lab Invest 58:37, 1988
17. Bengtsson T, Grenegkd M: Platelets amplify chemotactic
peptide-induced changes in F-actin and calcium in human neutrophils. Eur J Cell Biol 63:345, 1994
18. AI-Mohanna FA, Hallet MB: Actin polymerization modifies
From www.bloodjournal.org by guest on July 31, 2017. For personal use only.
PLATELETS INHIBIT NEUTROPHIL OXIDASE ACTIVITY
stimulus-oxidase coupling in rat neutrophils. Biochim Biophys Acta
927:366, 1987
19. Al-Mohanna FA, Ohishi I, Hallet MB: Botulinum Cz toxin
potentiates activation of the neutrophil oxidase. Further evidence of
a role for actin polymerization. FEBS Lett 21940, 1987
20. Norgauer J, Kownatzki E, Seifert R, Aktories K Botulinum
C2 toxin ADP-ribosylates actin and enhances 0; production and
secretion but inhibits migration of activated human neutrophils. J
Clin Invest 82:1376, 1988
21. Bengtsson T, Dahlgren C, Stendahlo, Andersson T: The role
of actin assembly in the regulation of neutrophil function: Effects
on chemotactic peptide-induced 0; production and degranulation. J
Leukocyte Biol49:236, 1991
22. McGarrity ST,Stephenson AH, Webster RO: Regulation of
human neutrophil functions by adenine nucleotides. J Immunol
142:1986, 1989
23. Dubyak GR, El-Moatassim C: Signal transduction via Pzpurinergic receptors for extracellular ATP and other nucleotides. Am
J Physiol 265:C577, 1993
24. Kuhns DB, Wright DG, Nath J, Kaplan SS,Basford RE:ATP
induces transient elevations of [Ca2'Ii in human neutrophils and
primes these cells for enhanced 0; generation. Lab Invest 58:448,
1988
25. Cronstein BN, Rosenstein, ED, Kramer SB, Weissmann G,
Hirschhom R: Adenosine, a physiologic modulator of superoxide
anion generation by human neutrophils. Adenosine acts via an AI
receptor on human neutrophils. J Immunol 135:1366, 1985
26. Schrier DJ, Imre KM: The effects of adenosine agonists on
human neutrophil function. J Immunol 137:3284, 1986
27. Iannone MA, Wolberg G, Zimmerman, Tp: Chemotactic peptide induces CAMPelevation in human neutrophils by amplification
of the adenylate cyclase response to endogenously produced adenosine. J Biol Chem 264:20177, 1989
28. B6yum A: Isolation of mononuclear cells and granulocytes
from human blood. Scand J Clin Lab Invest: 97:77, 1968
29. Cohen HJ, Chovaniec ME: Superoxide generation by digitonin-stimulated guinea-pig granulocytes. A basis for a continuous
assay for monitoring superoxide production and for the study of the
activation of the generating system. J Clin Invest 61:1081, 1978
30. Bengtsson T, Rundquist I, Stendahlo, Wymann MP, Andersson T: Increased breakdown of phosphatidylinositol 4,5-bisphosphate is not an initiating factor for actin assembly in human neutrophils. J Biol Chem 263:17385, 1988
31. Barak Ls,Yokum RR, Nothnagel EA, Webb WW: Fluorescence staining of the actin cytoskeleton with 7-nitrobenz-2-oxa-l,3diazole-phallacidin. Proc Natl Acad Sci USA 77:980, 1980
4423
32. Bengtsson T, Stendahlo, Andersson T: The role of the cytosolic free Caz' transient for fMet-Leu-Phe induced actin polymerization in human neutrophils. Eur J Cell Biol 42:338, 1986
33. Dahlgren C, Follin P, Johansson A, Lock R, Lundqvist H,
Walan A: Chemiluminescence as a means of following the function
of phagocytic cells. Trends Photochem Photobiol 2:427, 1991
34. Bender JG, Van Epps D E Analysis of the bimodal chemiluminescence pattem stimulated in human neutrophils by chemotactic
factors. Infect Immun 41:1062, 1983
35. Dahlgren C: Polymorphonuclear leukocyte chemiluminescence induced by formylmethionyl-leucyl-phenylalanineand phorbol
myristate acetate: Effects of catalase and superoxide dismutase.
Agents Actions 21:104, 1987
36, Mann JS, Renwick AG, Holgate ST: Release of adenosine
and its metabolites from activated human leucocytes. Clin Sci
70:461, 1986
37. Cronstein BN: Adenosine, an endogenous anti-inflammatory
agent. J Appl Physiol 76:5, 1994
38. Smolen JE,Korchak HM, Weissmann G: Increased levels of
cyclic adenosine-3',5'-monophosphate in human polymorphonuclear
leukocytes after surface stimulation. J Clin Invest 65:1077, 1980
39. Woodman RC, Ruedi JM, Jesaitis AJ, Okamura N, Quinn
MT, Smith RM, Curnutte JT, Babior BM: Respiratory burst oxidase
and three of four oxidase-related polypeptides are associated with
the cytoskeleton of human neutrophils. J Clin Invest 87:1345, 1991
40. Zalavary S, Stendahl 0, Bengtsson T Platelets enhance Fc
receptor-triggered phagocytosis and respiratory burst in human neutrophils: Role of purinergic modulation. Eur J Clin Invest 25:157,
1995 (abstr)
41. Zalavary S, Stendahl 0, Bengtsson T The role of cyclic
AMP,calcium and filamentous actin in adenosine modulation of
Fc receptor-mediated phagocytosis in human neutrophils. Biochim
Biophys Acta 1222249, 1994
42. Kat0 H, Ishitoya J, Takenawa T Inhibition of inositol phospholipids metabolism and calcium mobilization by cyclic AMPincreasing agents and phorbol ester in neutrophils. Biochim Biophys
Res Commun 139:1272, 1986
43. Downey GP, Elson EL, Schwab 111 B, Erzurum SC, Young
SK, Worthen GS: Biophysical properties and microfilament assembly in neutrophils: Modulation by cyclic AMP. J Cell Biol 114:1179,
1991
44. Cronstein BN, Daguma L, Nichols D, Hutchison A, Williams
M: The adenosineheutrophil paradox resolved Human neutrophils
possess both A, and A2 receptors that promote chemotaxis and inhibit 0; generation, respectively. J Clin Invest 85:1150, 1990
From www.bloodjournal.org by guest on July 31, 2017. For personal use only.
1996 87: 4411-4423
Release of oxygen metabolites from chemoattractant-stimulated
neutrophils is inhibited by resting platelets: role of extracellular
adenosine and actin polymerization
T Bengtsson, S Zalavary, O Stendahl and M Grenegard
Updated information and services can be found at:
http://www.bloodjournal.org/content/87/10/4411.full.html
Articles on similar topics can be found in the following Blood collections
Information about reproducing this article in parts or in its entirety may be found online at:
http://www.bloodjournal.org/site/misc/rights.xhtml#repub_requests
Information about ordering reprints may be found online at:
http://www.bloodjournal.org/site/misc/rights.xhtml#reprints
Information about subscriptions and ASH membership may be found online at:
http://www.bloodjournal.org/site/subscriptions/index.xhtml
Blood (print ISSN 0006-4971, online ISSN 1528-0020), is published weekly by the American
Society of Hematology, 2021 L St, NW, Suite 900, Washington DC 20036.
Copyright 2011 by The American Society of Hematology; all rights reserved.