| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
RIIIb Induces the Production of Granulocyte Colony-Stimulating Factor and Granulocyte-Macrophage Colony-Stimulating Factor by Polymorphonuclear Neutrophils
Laboratory of Immunology, Institut de Synergie des Sciences et de la Santé, Brest University Medical School, Brest, France
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
RIIIb (CD16) protect
these cells from spontaneous apoptosis. In this study, we used
anti-CD16 F(ab')2 to delineate the mechanism(s) whereby
the PMN life span is extended. As documented using four methods, CD16
cross-linking impeded spontaneous apoptosis, whereas anti-CD18
F(ab')2 exerted no effect. Incubation of PMNs with
anti-CD16 prevented the up-regulation of 
2
integrins, particularly CD11b, which is the 
-chain of complement
receptor type 3, but also CD18, which is its -chain, as well as
CD11a and CD11c. Anti-CD16-conditioned supernatant of PMNs diminished
the percentage of annexin V-binding fresh PMNs after another 18 h
in culture, whereas the negative control anti-CD18 had no effect.
The expression of mRNA for G-CSF and GM-CSF was induced by
anti-CD16, followed by the release of G-CSF and GM-CSF in a
dose-dependent manner. Anti-G-CSF and anti-GM-CSF mAbs abrogated
the antiapoptotic effect of the related growth factors. The delay in
apoptosis was accompanied by a down-regulated expression of
Bax, and a partial reduction of caspase-3 activity.
These data suggest an autocrine involvement of anti-CD16-induced
survival factors in the rescue of PMNs from spontaneous apoptosis.
Thus, apoptosis of aged PMNs can be modulated by signaling through
FcRIIIb, which may occur in patients with PMN-binding
anti-FcRIIIb autoantibodies. |
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
(4) and IL-6 (5), but delayed by IL-1,
G-CSF, GM-CSF, IFN- (6), and IL-8 (7).
Regarding the intracellular pathways involved in spontaneous apoptotic
mechanism, the inhibitory effect of Bcl-2 remains uncertain
in PMNs (8), although the balance between Bax
(9) and Bcl-xL has been shown
to regulate the machinery governing the activity of caspase-3 in these
cells (10).
The low affinity receptors for the Fc region of IgG, FcRIIIb (CD16),
and FcRIIa (CD32) are naturally expressed in PMNs, unlike the high
affinity receptor FcRI (CD64), in which the synthesis may be induced
by IFN- (reviewed in Ref. 11). Autoantibodies directed
against CD16 have been described in autoimmune mice (12)
and patients (13). In addition, on the basis of subsequent
results obtained by indirect immunofluorescence and ELISA tests, we
have identified three populations of anti-CD16 autoantibodies,
recognizing either PMN-bound CD16, soluble CD16, or both
(14). PMN-binding FcRIIIb autoantibodies might possibly
be implicated in the fate of PMNs. This concept is supported by our
preliminary results that show that, among anti-CD16 autoantibodies,
some with specificity for PMN-bound FcRIIIb have the capacity to
rescue senescent PMNs from spontaneous apoptosis (15).
Intriguingly, soluble FcRIIIb produces a similar effect
(16). These observations imply that CD16 may perpetuate
inflammation, following autoantibody cross-linking, through the
production of PMN-derived cytokines (reviewed in Ref. 17).
However, relatively little is known about the induction and the
mechanism(s) of such processes. Although FcRIIIb is GPI anchored to
the cell membrane (18), several groups have established
surprisingly that it can transduce signals by itself
(19, 20, 21). In contrast, others have stated that FcRIIIb
must cooperate with one or several neighboring transmembrane partners.
This latter view is strengthened by the demonstration that FcRIIIb
works in concert with FcRIIa to activate PMNs (22, 23, 24).
Alternatively, FcRIIIb can interact with 2
integrins, most notably complement receptor (CR) type 3, as indicated
by cocapping experiments (25). This major adhesion
molecule of PMNs (26) is formed by an -chain (CD11b)
noncovalently linked to a -chain (Mac-1, CD18), which is shared by
other integrins, CD11a/CD18 (LFA-1), CD11c/CD18 (CR4), and CD11d/CD18.
One practical problem is that a proportion of anti-FcRIIIb
autoantibody-containing sera is not uniquely specific for FcRIIIb,
inasmuch as autoantibodies from some of them react with FcRIIIb, but
also FcRII and/or FcRI (27). Therefore, it remains
unclear whether FcRIIIb autoantibody-containing sera inhibit PMN
apoptosis either through FcRIIIb cross-linking alone, through
FcRII ligation alone, or through collaboration between these two
FcR.
In the present study, we have used anti-CD16 mAb F(ab')2 to eliminate several artifacts, e.g., one population of autoantibodies cross-reacting with CD16 and CD32, or the coexistence in a given serum of one group of anti-CD16 autoantibodies and a second of anti-CD32 autoantibodies. We confirm that cross-linking of CD16 retards apoptosis of PMNs, and show that the proportion of CD11bdim PMNs is increased, relative to aged PMNs. The key mechanism of the sustained longevity of PMNs seems to be the CD16-induced production of cytokines. For the first time, we provide evidence for the anti-CD16 Ab-triggered transcription of mRNA for G-CSF and GM-CSF, followed by the synthesis and resulting release of these cytokines. Such a contention was confirmed by the blocking effect of anti-G-CSF and anti-GM-CSF mAbs. Another consequence of the CD16 stimulation was the lowered level of Bax (a proapoptotic member of the Bcl-eg2 family), and the subsequent reduction in the caspase-3 activity. Thus, a resolution of the inflammation might, at least in part, be modulated by CD16.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
FITC mAbs to CD16 (clone 3G8), CD11b (clone Bear 1), CD4 (clone
13B8.2), CD8 (clone B9.11), CD15 (clone 80H5), CD18 (clone 7E4), CD19
(clone J4.119), CD23 (clone 9P25), CD32 (clone 2E1), CD35 (clone
J3.D3), CD45 (clone J33), CD48 (clone J4-57), CD56 (clone N901), CD64
(clone 22), CD66b (clone 80H3), and to monomorphic determinants in HLA
class I molecules (clone B9.12.1); PE mAbs to CD11a (clone 25.3.1),
CD11b (clone Bear 1), CD11c (clone BU-15), and CD18, as well as
unconjugated mAbs to CD16, CD11b (clone Bear 1), HLA-I, CD3 (clone
UCHT1), CD4, CD8, CD18, CD19, CD23, CD56, and CD64 were all obtained
from Beckman Coulter (Villepinte, France). FITC mAb to CD53 (clone MEM
53) was purchased from Interchim (Montluçon, France), and FITC
mAbs to CD55 (clone BRIC 110) and to CD59 (clone MEM-43) from Serotec
(Oxford, U.K.). Additional PE anti-CD11b mAbs, clones 44 and
2LPM19c, were obtained from BD PharMingen (San Diego, CA) and Dakopatts
(Glostrup, Denmark), respectively. FITC anti-CD62 ligand
(CD62L) (clone TQ1) mAb, FITC and PE IgG1, IgG2a, and IgM
isotype controls were purchased from Beckman Coulter. FITC-labeled and
unconjugated sheep F(ab')2 anti-mouse IgG
F(ab')2 were purchased from Jackson
ImmunoResearch Laboratories (West Grove, PA). Rabbit polyclonal
anti-Bax F(ab')2 was obtained from
Santa Cruz Biotechnology (Santa Cruz, CA), and revealed by FITC
goat F(ab')2 anti-rabbit
F(ab')2 (Sigma, St. Louis, MO). Unconjugated
anti-TNF-, anti-G-CSF, and anti-GM-CSF goat polyclonal
Abs were obtained from R&D Systems (Minneapolis, MN). Anti-CD16,
anti-CD18, anti-HLA-I, and anti-CD11b
F(ab')2 were prepared by pepsin digestion.
Cell preparation
Blood samples were drawn from healthy nonsmoking volunteers into
heparinized Vacutainer tubes (BD Biosciences, Franklin Lakes, NJ).
Five-milliliter aliquots were collected to study unmanipulated cells.
These were immediately cooled to 4°C, washed three times in 125 mM
NaCl, 10 mM phosphate, 5 mM KCl, 5 mM glucose, 1.09 mM
CaCl2 and 1.62 mM MgCl2, pH
7.35, and resuspended to the original volume. PMNs were isolated from
the rest of the blood samples by Dextran T500 (Pharmacia, Uppsala,
Sweden) sedimentation, followed by Ficoll-Hypaque density gradient
centrifugation (Eurobio, Les Ulis, France). Residual erythrocytes were
lysed with hypotonic buffer. To reduce macrophage (M
), eosinophil,
T, B, and NK cell contamination below 1% (28), the cell
suspension was incubated for 20 min in the presence of a mixture of
anti-CD3, anti-CD4, anti-CD8, anti-CD19, anti-CD23,
anti-CD56, and anti-CD64 mAbs. After three washes in PBS
supplemented with 5% BSA, a negative selection was performed using
goat anti-mouse Ig Ab-coated magnetic microbeads (BioAdvance,
Emerainville, France).
M were prepared from 100 ml of blood obtained from the same donors
as the PMNs. PBMCs were allowed to adhere to polystyrene for 2 h
at 37°C. Nonadherent cells were washed away, whereas the M were
harvested with 0.05% trypsin/0.02% EDTA, washed three times, and
cultured for another 4 h in the absence or the presence of 500
pg/ml IL-1 (Genzyme, Cambridge, MA). B cells from a patient with
chronic lymphocytic leukemia were also purified by negative selection
with magnetic beads, following treatment of the PBL with a mixture of
mAbs, as described above.
Cell culture
To cross-link mAb binding to PMNs, 96-well microtiter plates were coated with 30 µg/ml sheep F(ab')2 anti-mouse F(ab')2 by an initial 30-min incubation at 37°C, and a second at 4°C overnight, followed by three washes with RPMI 1640 medium. PMNs were then dispensed at 5 x 105 cells/well, and cultured in 200 µl RPMI 1640 medium (Life Technologies, Paisley, Scotland) supplemented with 2.5% FCS, 2 mM L-glutamine (bio-Mérieux, Lyon, France), 200 U/ml penicillin, and 500 µg/ml streptomycin. CD16 mAb F(ab')2 was added at a final concentration of 5 µg/ml, unless otherwise indicated. In selected experiments, the effect of anti-CD16 F(ab')2 was evaluated in the absence of the second-layer Ab. It was essential (6, 28) to ensure that all the reagents used for PMN isolation and culture were LPS free, as judged by a quantitative chromogenic Limulus amebocyte lysate assay (BioWhittaker, Walkersville, MD). As an additional precaution, polymyxin B (Sigma) was added at 100 IU/ml to neutralize undetectable endotoxin trace amounts.
FACS analysis
Staining consisted of incubating 5 x 105 PMNs for 30 min on ice with 10 µl appropriate mAbs at previously determined concentrations. After three washes in PBS supplemented with 2% BSA and 0.1% sodium azide, PMNs were analyzed on an Epics Elite flow cytometer (Coulter, Hialeah, FL), along with isotype controls. The PMNs in whole blood were identified by forward and side scatters. Purified PMNs were also gated to avoid unwanted debris. As for the anti-CD11b mAb, the binding of Bear 1 was not blocked by either of the two I domain-specific CD11b mAbs, 44 and 2LPM19c (data not shown). The percentages of positive cells and mean fluorescence intensities (MFI) were compared with isotype controls in all experiments. In a series of experiments, saturation by Abs was investigated by incubating 105 freshly isolated PMNs with increasing amounts of unconjugated anti-CD16, anti-CD18, or anti-HLA-I F(ab')2, and developing with FITC-labeled sheep F(ab')2 anti-mouse IgG F(ab')2. The MFI of the second-layer reagent alone was systematically subtracted from that obtained with previous incubation of anti-CD16, anti-CD18, or anti-HLA-I F(ab')2. In pilot experiments, the number of molecules per cell was quantified by determining the amount of Ab binding to the cells at saturating concentrations, using the Quantum Simply Cellular kit (Flow Cytometry Standards, San Juan, PR). According to the manufacturer’s instructions, 50 µl of microbeads was added to 20 µl of Ab. Ab-binding capacities (ABC), derived from the MFI and accounting for the numbers of molecules recognized by these Abs, were expressed as arbitrary units.
Assessment of apoptosis
To determine apoptosis, 5 x 105 PMNs were stained with FITC-annexin (29) and propidium iodide (PI), according to the manufacturer’s instructions (Beckman Coulter). After a 10-min incubation in the dark, cells were analyzed by FACS. Early apoptotic PMNs were defined as those annexin V+/PI-, and necrotic PMNs as PI+ nonpermeabilized cells.
Independent assessment of apoptosis was evaluated using the method described by Nicoletti et al. (30). Briefly, after 18 h in culture, PMNs were washed in citrate buffer (0.1 M sodium citrate, 0.1% Triton X-100) and incubated in 250 µl of citrate buffer overnight in the dark at 4°C. Apoptosis was expressed as the percentage of hypoploid PMNs in each flow cytometric profile.
To confirm DNA fragmentation, PMNs were washed three times in HBSS (Eurobio) and once in lysis buffer (250 mM sucrose, 50 mM Tris, pH 7.5, 25 mM KCl, and 5 mM MgCl2). Cell pellets were resuspended and incubated for 8 min on ice in a solution of 0.25% Triton X-100 (Sigma) added to 500 µl of lysis buffer. After centrifugation for 5 min at 500 x g and 4°C, nuclei were resuspended in 500 ml of lysis buffer. This was supplemented with 25 µl of 0.5 M EDTA, 70 µl of 10% SDS, and 0.2 mg of proteinase K (all obtained from Sigma), then incubated for 3 h at 37°C. The DNA was extracted with phenol/chloroform/isoamyl alcohol (25:24:1), followed by two extractions with chloroform (v/v). After 5 min of washing at 400 x g and 4°C, a 1:10 volume of sodium acetate, followed by 2 vol of absolute ethanol was added. DNA was kept for 12 h at 4°C and centrifuged for 10 min at 450 x g. Two volumes of absolute ethanol were then added before addition of 70% ethanol. DNA was dissolved in 10 mM Tris, pH 7.5, and 1 mM EDTA, pH 8, for 12 h after evaporation of the ethanol. The DNA was loaded into wells of a 1% agarose gel and electrophoresed at 75 mV using 100 mM Tris, 100 mM boric acid, and 0.2 mM EDTA as running buffer. DNA was visualized by ethidium bromide staining.
Cell morphology of 105 fresh and apoptotic PMNs was studied. These were centrifuged for 5 min at 200 x g on microscope slides. The slides were air dried for 10 min, and the cells were stained in a May-Grünwald solution (Merck, Darmstadt, Germany) for 3 min, washed in water for 1 min, stained in a Giemsa solution (Merck) 7% in distilled water, and finally rinsed in water. Morphologic changes characteristic of apoptosis, such as nuclear condensation and vacuolation, were subsequently analyzed.
Identification of survival factors (SFs)
To address the question as to whether SFs are released by PMNs in response to CD16 cross-linking, 5 x 105 PMNs were treated with anti-CD16, anti-HLA-I, or anti-CD18 F(ab')2 at final concentrations of 1–20 µg/ml. After an 18-h incubation, supernatants were collected. These were incubated with magnetic beads coated with anti-mouse F(ab')2 Ab for 30 min to remove any residual F(ab')2. Aliquots were then harvested to identify SF(s), and the remaining 200 µl, diluted 1/2 in fresh medium, was added to triplicate wells of 5 x 105 freshly isolated PMNs. Incubation lasted a further 18 h, until the time when annexin V+/PI- cells were enumerated in these secondary cultures. Supernatant-induced variation of PCD was calculated according to the formula: (control apoptotic PMNs - supernatant apoptotic PMNs/control apoptotic PMNs) x 100.
RT-PCR of mRNA for SFs
Concomitantly, five different suspensions of
107 fresh PMNs each were cultured in the presence
of 10 µg/ml anti-CD16 F(ab')2, and cells
were collected at 0, 3, 6, 9, and 12 h for RNA isolation. B cells,
and resting or IL-1-activated M served as control cells in the
RT-PCR experiments. For the preparation of RNA, PMNs, M, or B cells
were first washed twice with HBSS. Total RNA was isolated by the
guanidine isothiocyanate method (31) using RNABle
(Eurobio).
In the reverse-transcriptase step, 1 µg of total RNA was used in a total incubation volume of 20 µl. Oligo(dT) primers, deoxynucleotides mix, and 200 U of reverse transcriptase from Moloney murine leukemia virus were supplied (Life Technologies). After incubation at 42°C for 50 min, 2 U of RNase H was added and incubated for 20 min at 37°C. Two microliters from this incubation were used in each of the subsequent PCR amplification steps.
Five pairs of primers were selected (32):
5'-CTCTGGACAGTGCAGGAAGCCACC-3' plus 5'-GCTGGGCAAGGTGGCGTAGAACGC-3'
for G-CSF; 5'-GCAGCCCTGGGAGCATGTGAATGC-3' plus
5'-ATGCCTGTATCAGGGTCAGTGTGC-3' for GM-CSF;
5'-AACTCCTTCTCCACAAGCGCCTTC-3' plus 5'-TGGACTGCAGGAACTCCTTAAAGC-3'
for IL-6; 5'-ACCGCCATGGAGGAAGGTCAATATTCAG-3' plus
5'-TATCCATGCTCAAGAGTGGAGAGGGGAG-3' for CD23; and
5'-GAAGATCAAGATCATTGCTCCTCC-3' plus
5'-CTGGTCTCAAGTCAGTGTACAGG-3' for -actin. DNA fragments of
533 bp (G-CSF), 497 bp (GMCSF), 593 bp (IL-6), 981 bp (CD23), and
707 bp (-actin) were obtained.
The following program was used. After the initial template denaturation
for 5 min at 94°C, 2.5 U Taq polymerase (Genaxis
Biotechnology, Saint-Cloud, France) was added. Standard cycle
conditions were: 30 s at 94°C, 60 s at 55°C, and 60
s at 72°C. Thirty-five cycles had to be conducted, except for
-actin, in which 30 cycles were sufficient. The resulting products
were run on a 3% agarose gel (Nusieve 3.1; FMC, Rockland, ME) and
stained with 0.5 µg/ml ethidium bromide. In addition, PCR products
were checked by digestion with various restriction enzymes. Expected
digestion patterns were obtained in each case. -actin mRNA was
measured in RNA samples at each point with the same cDNA as that
analyzed for cytokine transcripts. PCR band densities were determined
using Molecular Analyst software (Bio-Rad, Hercules, CA), and the mean
density for each point was normalized with -actin.
Synthesis and release of G-CSF and GM-CSF
G-CSF (sensitivity: 12 pg/ml) and GM-CSF (sensitivity: 5 pg/ml)
were assayed in the aliquots of CD16-conditioned medium, using
commercial ELISA kits (R&D Systems and Beckman Coulter, respectively).
In inhibition experiments designed to confirm the nature of SFs, human
rG-CSF and rGM-CSF (R&D Systems) were used to establish the specificity
of the related Abs by Western blotting (data not shown). Then, 200 µl
of the primary culture supernatants was preincubated with 2 µg of
anti-TNF-, 2 µg of anti-G-CSF, 2 µg of anti-GM-CSF,
or 1 µg of anti-G-CSF plus 1 µg of anti-GM-CSF Abs for 90
min at 37°C, before being added to fresh PMNs. Variation of
spontaneous apoptosis was calculated as above.
Bcl-2 family member expression and caspase-3 activity
To identify intracellular Bax, PMNs were first labeled with PE anti-CD11b mAb, and washed three times. They were then permeabilized in 1 ml of 0.1% saponin for 12 min in the dark. After two washes, the cells were incubated with rabbit anti-Bax F(ab')2 for 30 min at 4°C, followed by another three washes. This unconjugated Ab was revealed by a 30-min incubation with FITC goat F(ab')2 anti-rabbit F(ab')2 at 4°C and washed again. PMNs incubated with the second-layer Ab alone served as a negative control. The preparations were fixed in 5% paraformaldehyde buffer before FACS analysis. A colorimetric assay was then used to measure caspase-3 activity. As recommended by the manufacturer (Tebu, Le Perray-en-Yvelines, France), 2 x 107 PMNs were lysed in 1 ml of lysis buffer (50 mM HEPES, pH 7.4, 0.1% CHAPS, 1 mM DTT, 0.1 mM EDTA) for 5 min on ice. Cell extracts were mixed in assay buffer with or without Ac-DEVD-CHO, a specific inhibitor for caspase-3 activity, in the presence of Ac-DEVD-pNA substrate. Caspase-3 activity was measured at 405 nm in an automatic plate reader (Multiskan Labsystems, Helsinki, Finland).
Statistics
Means and SEM were calculated from a minimum of three independent experiments, each run in triplicate (the results were averaged). Comparisons were made using the unpaired Mann-Whitney U test and the Wilcoxon’s test for paired data.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The final cell suspension consisted of 
99% PMNs, as determined
by CD4, CD8, CD19, CD23, CD56, and CD64 staining (Fig. 1
, upper panels). The cell
preparations contained <0.1 µg/ml endotoxin, and PMNs had not been
activated by the isolation procedure or hypotonic lysis, as documented
by the absence of reciprocal changes in expression of CD11b, CD66b, and
CD45, which were not up-regulated (33, 34, 35), whereas CD62L,
CD53, and CD16 were not down-regulated (36, 37, 38) in
purified PMNs (Fig. 1, lower panels), compared with
unmanipulated PMNs in whole blood. Our PMN preparations were extremely
pure, LPS free, and nonactivated, thus fulfilling the three criteria
for evaluating their behavior in culture.
|
A). This
finding agrees with previous reports (39, 40). The
proportion of necrotic cells was <5% for CD16-treated, CD18-treated,
and untreated aged PMNs, indicating that CD16 cross-linking did not
induce necrosis after a 12-h incubation (Fig. 2, inset).
|
. annexin
V binding is presented for comparison (Fig. 3A). Hypoploid
cells became detectable after 18 h in culture, and there was a
37.7 ± 5.3% reduction (p < 0.04) in
CD16-treated PMNs, compared with CD18 (Fig. 3B, histograms 3
and 4). DNA fragmentation in PMNs cultured alone or in the presence of
anti-CD18, but not yet in that of anti-CD16, was confirmed
(Fig. 3C, lanes 2, 4, and 3). There
were May-Grünwald-Giemsa-stained PMNs from the same cultures
displaying apoptotic morphology in medium alone or following incubation
with anti-CD18, then treated with anti-CD16 (Fig. 3D, samples 2, 3, and 4).
|
B). Supporting this view is the fact that the doses
required to saturate surface CD16 and CD18 (Fig. 4) were exactly the same (5–10 µg/ml),
even though this does not necessarily imply that such doses produced
maximal activity.
|
2 integrins
The expression of adhesion molecules was evaluated in the context
of CD16-delayed apoptosis. Virtually all freshly isolated PMNs
expressed CD11b with an MFI of 7.1 ± 0.3 (mean ± SEM of
five independent experiments). After a 12-h incubation in medium, two
peaks became distinctly apparent for CD11b (Table I), in that the MFI raised to 11.5
± 0.7 in a 66.6 ± 2.5% fraction of the PMNs (this
CD11bbright population will be referred to as
such), whereas it declined to 2.3 ± 0.2 in the remaining
22.4 ± 1.4% fraction (CD11dim population).
After a 12-h CD16 stimulation, the CD11bbright
PMN subpopulation accounted for as few as 36 ± 2.5% of the PMNs
(p < 0.01, compared with untreated aged PMNs),
whereas 51.6 ± 1.8% of PMNs became
CD11bdim (p < 0.01,
compared with untreated PMNs). A representative example is shown in
Fig. 5A. Because stimulation
induces conformational changes of CR3 and alters the accessibility of
some epitopes on CD11b/CD18, the decrease in
CD11bbright PMN may be due to the burying of
CD11b epitopes, rather than a genuine reduction in the number of CD11b
molecules. However (Table II), similar
results were obtained, irrespective of which of the three PE mAb (Bear
1, 44, or 2LPM19C) was used to identify CD11b.
|
|
|
A), or GPI linked, such as CD48, CD55,
CD59, and CD66b. Anti-CD16 treatment produced a lowered expression of
CD18, the -chain of CR3. This effect was restricted to the
CD11bdim subpopulation. The same holds true for
the alternative -chains of the dimer, CD11a and CD11c, albeit to a
lesser degree. Representative dual fluorescence histograms are shown in
Fig. 5B, based on three independent experiments for mAbs
defining CD11b and CD18, CD11a, or CD11c. A time-course study (Fig. 6A) revealed that the
expansion in the dimmer CD11b subpopulation correlated with the length
of time the CD16 had been cross-linked by anti-FcRIIIb
F(ab')2. After 16 h of CD16 stimulation,
this treatment prevented 55.2 ± 2.6% (mean ± SEM of three
separate experiments) of the PMNs to become
CD11bbright. The reduction in the
CD11bright population was also related to the
amount of anti-CD16 F(ab')2 added to the
culture (Fig. 6B).
|
A.
|
). A mixture of 5 µg/ml anti-CD16
and 5 µg/ml anti-CD11b showed additive effects (71.4 ±
1.5%, three experiments).
|
We next investigated whether SFs were induced in response to CD16
cross-linking. Following an 18-h incubation of 2 x
106 PMNs in medium or in the presence of
increasing amounts of F(ab')2 specific for CD16,
CD18, or HLA-I, the PMN-conditioned supernatants of these primary
cultures were collected and used as the culture medium for 5 x
105 freshly explanted PMNs. CD18 mAb was taken as
a negative control, and, because ligation of HLA-I molecules retards
somewhat apoptosis (47), anti-HLA-I mAb served
as a positive control. Annexin V-binding fresh PMNs were enumerated
after another 18 h in culture, and the supernatant-induced
reduction of apoptosis was calculated. These PMNs were rescued from
spontaneous apoptosis by CD16-treated PMN supernatants in a
dose-dependent manner, whereas, as expected, there was a 25%
inhibition in cells treated with anti-HLA-I mAb, and
CD18-conditioned supernatants produced no effect on apoptosis of fresh
PMNs (Fig. 9A). Furthermore,
PMN-PMN interactions do not seem to be involved in the production of
SFs. To rule out this possibility, PMNs were cultured at a variety of
different cell densities in medium alone or in the presence of 5
µg/ml anti-CD16 or anti-CD18. As above, the supernatants were
added to a constant number of fresh PMNs (0.5 x
106), which were examined 18 h later (Fig. 9B). The PCD-inhibiting effect of the supernatants was the
same, as indicated by similar percentages of Annexin V-binding cells,
irrespective of the number of PMNs present in the primary cultures.
|
G-CSF and GM-CSF emerged as credible agents in the CD16-mediated
reduction of apoptosis (6, 17 , and our pilot experiments).
Increasing amounts of G-CSF, GM-CSF, and half G-CSF plus half GM-CSF
were added to PMNs without CD16 engagement (Fig. 10). The antiapoptotic effect was
calculated after 18 h in culture, compared with PMNs in medium
alone. At a dose of 10 ng/ml, G-CSF, GM-CSF, and G-CSF plus GM-CSF
reduced Annexin V binding by 62.5 ± 0.9, 62.7 ± 0.1, and
86.2 ± 0.4%, respectively (mean ± SEM of three
experiments).
|
and B
cells served as positive controls for IL-6 and CD23 transcripts, and
provided additional evidence (Fig. 10A) for the lack of
contaminating M and eosinophils in the preparations of fresh PMNs,
in which the message for CD23 and IL-6 was undetectable. In contrast,
the former was expressed in B cells, and the latter in resting M,
IL-1-activated M, and B cells. As shown in Fig. 11B, when cultured in the
presence of anti-CD16 F(ab')2, transcription
of mRNA for G-CSF was evident by 3 h (G-CSF/-actin: 30),
reached a maximum at 6 h (G-CSF/-actin: 59.6), and declined at
12 h (G-CSF/-actin: 5.1). It was also found that, to a lesser
degree, mRNA for GM-CSF peaked at 6 h (GM-CSF/-actin: 24.2),
and plateaued until at least 12 h.
|
). Substantial
quantities of G-CSF, and lower quantities of GM-CSF were detected in
the PMN supernatant. The levels of both cytokines paralleled the
amounts of CD16 mAb used to stimulate the cells.
|
A prerequisite was to verify that anti-G-CSF and
anti-GM-CSF mAbs were strictly specific. After SDS-PAGE,
recombinant human G-CSF (5 µg/ml) and GM-CSF (1 µg/ml) were
transferred onto polyvinylidene difluoride membrane and probed with
both Abs. Neither of these Abs exhibited cross-reactivity (data not
shown). To confirm the functional activity of G-CSF and GM-CSF released
by CD16-stimulated PMNs in primary culture, fresh PMNs were incubated
SF(s)-containing supernatants in the absence or in the presence of
neutralizing Abs. Annexin V binding was measured 18 h later, and
variations were calculated by comparison with PMNs purified from the
same samples and cultured without Ab for the same period of time. As
shown in Fig. 12, CD16-conditioned
supernatants diminished the percentages of annexin V-positive cells by
52.7 ± 1.9% (mean ± SEM of five separate experiments).
Although anti-TNF- Ab did not influence the antiapoptotic effect
of SFs (still reduction of apoptosis was 52.4 ± 1.7%),
anti-G-CSF and anti-GM-CSF generated 31.5 ± 2.7 and
33.1 ± 1% abrogation of the antiapoptotic effect produced by the
primary culture supernatants (p < 0.04,
compared with CD16-conditioned supernatant alone). The capacity of SFs
to inhibit apoptosis was almost completely blocked (5.9 ± 1.4%
reduction, p < 0.04) with a mixture of anti-G-CSF
and anti-GM-CSF Abs, suggesting that these two cytokines produced
an additive effect, and accounted for most of the protection induced by
treatment of PMNs with anti-CD16.
|
Additional experiments were performed to refine the understanding
of the mechanism that underlies the antiapoptotic effect of G-CSF and
GM-CSF. Direct evidence has been provided recently (42)
for a key role of Bax as a proapoptotic molecule in PMNs.
This finding was applied to our model. PMN membrane staining of CD11b,
coupled with intracellular staining of Bax, was performed in
PMNs cultured for 18 h either in primary nonconditioned or in
CD16-stimulated supernatants. After another 18 h in culture, the
percentage and MFI of Bax-containing cells were
down-regulated (Table IV and Fig. 7B) by CD16 cross-linking in PMNs. Interestingly, although
significant also in the CD11bdim population, this
reduction predominated in the CD11bbright
population.
|
The involvement of caspase-3 in spontaneous apoptosis of PMNs
(9) prompted us to determine the effect of this protease,
following incubation of the cells with anti-CD16
F(ab')2. Although it was inactive in untreated
fresh cells (Fig. 13), aged PMNs
developed a strong caspase-3 activity, which was retained in the
presence of anti-CD18 F(ab')2 used as a
negative control. This was completely abolished by Ac-DEVD-CHO, an
inhibitor for caspase-3, demonstrating the specificity of the used
assay. Incubation of PMNs in the presence of anti-FcRIIIb
F(ab')2 resulted in a 26 ± 2.7%
reduction (mean ± SEM of four separate experiments) of
caspase-3 activity (p < 0.05, compared
with anti-CD18).
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
RIIIb long-term
cross-linking generates the production of G-CSF and GM-CSF, and thereby
retards spontaneous apoptosis of senescent PMNs. Obviously, it cannot
be inferred from this finding that other apoptotic processes of PMNs,
such as Fas-mediated apoptosis, steroid stimulation, or radiation
exposure, would also be affected by CD16 stimulation. After 18 h
in culture, the cell population has not yet started to exhibit
significant evidence of necrosis, as previously highlighted by Savill
et al. (1). In this study, it was essential to rule out
that contaminating LPS, an apoptosis-delaying stimulus (6, 28) possibly present in the reagents, was responsible for the
PMN survival or worked in synergy with CD16 ligation to cause the
effect.
In this study, F(ab')2 of the CD16 mAb 3G8 were
incubated with the cells and cross-linked with a second Ab.
Furthermore, albeit to a much lower extent, the same tendency was
observed in the absence of the second-layer Ab. Thus, in contrast to
the 10-min culture, in which an extensive cross-linking of surface
FcRIIIb is required to transduce signals (43), pairing
of these molecules seems to be sufficient in an 18-h culture to trigger
the earliest events of transduction. Such a phenomenon might perhaps be
assigned to the extreme abundance of FcRIIIb molecules on the PMN
surface (Ref. 44 , and this study). Our finding, which is
not this surprising, is consistent with the demonstration that
FcRIIIb cross-linking is not an absolute requirement for signaling
functions of PMNs (45).
Given the homeostatic role attributed to the 2
integrins in the programmed elimination of PMNs (39, 46),
the effect of CD16 activation on the expression of CD11b/CD18 was then
investigated. Spontaneously apoptotic PMNs have been shown to express
CD11b/CD18 and CD11c/CD18 at increased levels (46), and a
20-min cross-linking of CD16 demonstrated to promote this enhancement
(47). We found that, following this initial up-regulation
of several 2 integrins secondary to a rapid
translocation of the - and the -chains from the intracellular
pool to the cell surface (33), CD16 induced the
persistence of significantly less CD11bbright
PMNs in 12-h cultured cells, compared with PMNs maintained in medium.
Conformational changes of CD11b caused by CD16 long-term engagement and
masking certain epitopes are unlikely to account for this reduction,
because three different anti-CD11b mAbs produced similar results.
However, it is interesting to note that the MFI of the dimmer CD11b
subpopulation was 10-fold higher with the 44 than the Bear 1 mAb,
suggesting that the I domain epitope recognized by 44 and the 2LPM19c
mAbs remained more accessible than the Bear 1 epitope during CD16
activation (48). Nonetheless, treatment of PMNs with
anti-CD11b F(ab')2 delayed apoptosis of aging
cells. This finding confirms and extends our previous report that
soluble FcRIIIb prolongs the survival of PMNs in vitro
(16), presumably through CRs (49). The
effects of anti-CD16 and anti-CD11b were additive. Ligation of
CD16 may favor the binding of its polysaccharides to the lectin site of
CD11b (50), whereas the anti-CD11b mAb Bear 1 used in
our experiments is specific for this part of CD11b (51).
Such data are consistent with the conclusion that CD16 polysaccharides
and Bear 1 share overlapping, but not identical binding sites on the
CD11b lectin site. Our finding that the anti-CD11b, along with the
anti-CD18 mAb treatment, exerted additive effects on PMN survival
is consistent with, but does not prove, the model of functional
interactions between these receptors. Furthermore, the fact that
anti-CD18 had no effect on survival argues against such
interactions. However, similar results have been reported by Watson et
al. (40). Using the mAbs that we used in our own study,
these investigators showed that cross-linking of the -chain of the
2 integrins, CD11a and CD11b, significantly
delayed the apoptosis of resting PMNs, whereas cross-linking of the
common -chain, CD18, failed to alter the apoptotic rate of the
cells. In theory, CD16 might however cooperate with CD11b in
apoptosis-delaying signal transduction. In addition, although C is not
involved, CD11b/CD18 and CD16 cooperate in adhesiveness, phagocytosis,
and the generation of respiratory burst (39, 52, 53, 54). As a
result, the CD16-induced augmentation in the proportion of
CD11bdim should generate a functional alteration
of PMNs, and this was actually the case (15 and data
not presented).
There appeared that the CD11bdim population
became apoptotic, as reported by Jones and Morgan (55). It
follows that long-term engagement of CD16 seems to put more cells at
risk of becoming apoptotic. In support of this view is the
demonstration that increased PMN accumulation in CD11b/CD18-null mice
in thioglycolate-induced peritonitis is accompanied by decreased
apoptosis of extravasated PMNs (45). Interestingly,
although predominating in the CD11bbright
population, there was a down-regulation of Bax in the
CD11bdim population. CR3 does not initiate
apoptosis by itself, but potentiates TNF--mediated
(39), and possibly spontaneous apoptosis of PMNs. We
failed to establish the involvement of PMN-PMN interactions in the
prolonged survival, inasmuch as the conditioned supernatants of
increasing numbers of cells produced almost the same PCD-inhibiting
effect when incubated with a constant number of PMNs for a further
18 h.
G-CSF and GM-CSF are present at sites of neutrophilic inflammation. In
this respect, our major finding was that PMNs release substantial
quantities of cytokines in response to CD16 cross-linking, explaining
in part the delayed apoptotic process (56). It is well
established that, among other agents, these growth factors exert a
potent effect on the regulation of apoptosis in PMNs (57, 58). The physiologic control of G-CSF and GM-CSF production
remains only partially understood. G-CSF (59) and GM-CSF
(60) are produced by a variety of cell types, following
appropriate stimulation; in this study, we provide some unanticipated
evidence for the CD16-initiated transcription of their mRNAs in PMNs.
This is not constitutive (17), as confirmed by the absence
of transcripts in fresh PMNs. The proteins became detectable after
18 h of culture in the presence of cross-linked anti-CD16
F(ab')2. To ensure that the proteins measured in
the culture supernatants were not released by contaminating M, it
was essential to yield pure populations of PMNs. This concern was
clearly addressed, because our cell suspension was comprised of 99%
PMNs. The absence of contaminating M and eosinophils was confirmed
by the fact that mRNA for IL-6 and CD23 were undetectable in fresh
PMNs. Neutralization of G-CSF biologic activity by specific antiserum
abrogated G-CSF-mediated inhibition of PCD. In addition,
immunodepleting GM-CSF blocked the effect of GM-CSF. These data suggest
that G-CSF and GM-CSF produce distinct, but additive antiapoptotic
effects. Comparable results have been reported by Matute-Bello et al.
(61) in bronchoalveolar lavage fluid from patients with
the acute respiratory distress syndrome. It is known that GM-CSF can
induce the secretion of G-CSF by PMNs (62). It would thus
be interesting to study the kinetics of GM-CSF and G-CSF production in
the CD16-stimulated PMN model, to address the issue as to whether the
G-CSF expression results from that of GM-CSF. It is surprising that 5
µg/ml F(ab')2 anti-CD16 was sufficient for
maximal activity of conditioned supernatants (see Fig. 6), although
doses as high as 20 µg/ml still induced the production of larger
amounts of G-CSF and GM-CSF (see Table III). As previously described
(63), this discrepancy suggests that subsaturating dosages
of mAb bind uniformly to all the cells, but a mAb dose-dependent
proportion of those mAb-bound PMNs responded with a release of G-CSF
and GM-CSF.
The precise mechanism by which GM-CSF and G-CSF extend the survival of PMNs warrants further analysis. To begin to address this issue, the level of the death promoter Bax was measured and found to be reduced by CD16 long-term engagement. This is consistent with the recent report that, in neutrophilic inflammatory diseases, the delay in PMN apoptosis was associated with markedly decreased levels of Bax, and normalized by stimulation with G-CSF and GM-CSF in vitro (45).
Caspase-3 has been reported to be pivotal in spontaneous
apoptosis of PMNs (9). In the present study,
CD16-stimulated PMNs reduced, but did not abolish activity of this
cysteine protease. Such a finding is consistent with the observation
that impaired apoptotic death signaling in inflammatory lung PMNs is
associated with decreased, but not totally inhibited expression of
caspases (40). One possibility is that, due to
insufficient activation by caspase-3, protein kinase C
does not
acquire the capacity to stimulate phospholipid scramblase
(64), and to induce the ensuing translocation of
phosphatidylserine to the outer leaflet of the plasma membrane. There
appear to be at least two explanations. Caspase-3 might be one of
several different molecular pathways involved in PMN spontaneous
apoptosis, so that it would be of no surprise that the CD16-induced
inhibition of apoptosis remains partial. It has nevertheless been
established that PCD may occur through apparently caspase-3-independent
ways (65). Although it has not been clearly defined in
humans, PML nuclear bodies, which are nuclear matrix-associated
structures inducing a caspase-independent death process
(66), might follow another pathway in PMN spontaneous
apoptosis. Alternatively, this spontaneous apoptosis might be
exclusively mediated by caspase-3. In the latter scenario, it is
possible that caspase-3 becomes accessible to CD16-induced signals in
only a fraction of the cells, presumably depending on their age.
Indeed, terminally differentiated PMNs undergo major functional changes
as they age in the circulation (67), resulting inevitably
in a heterogeneous population of cells. Relatively low levels of
reactive oxygen species generated in this model (68) do
not prevent caspases from functioning, and offer another explanation
for the incomplete inhibition of caspase-3 following CD16
engagement.
In conclusion, CD16 long-term stimulation significantly prolonged the
survival of PMN. This phenomenon was associated with induction of mRNA
and protein synthesis of G-CSF and GM-CSF. PMN-binding
anti-FcRIIIb autoantibodies might produce the same effect,
accompanied by the absence of up-regulation of
2 integrins. CD16 and the related
autoantibodies may thus be influential in the control of inflammatory
responses, and thereby facilitate the development of autoimmune
states.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
2 Address correspondence and reprint requests to Dr. Pierre Youinou, Laboratory of Immunology, Brest University, Medical School, 5 av. Foch, F 29 609 Brest Cedex, France. E-mail address: youinou{at}univ-brest.fr
3 Abbreviations used in this paper: PMN, polymorphonuclear neutrophil; ABC, Ab-binding capacity; CR, complement receptor; M, macrophage; MFI, mean fluorescence intensity; PCD, programmed cell death; PI, propidium iodide; SF, survival factor; CD62L, CD62 ligand.
Received for publication December 27, 2000. Accepted for publication July 19, 2001.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
. Int. Immunol. 5:691.-mediated regulation of human neutrophils via caspase-3. Blood 93:3106. receptor antibodies. J. Exp. Med. 171:1581.R are found in sera of autoimmune patients. J. Immunol. 150:2018.[Abstract]
RIIIb autoantibodies on polymorphonuclear neutrophil apoptosis and function. J. Leukocyte Biol. 69:233. receptor IIIb alters the function of polymorphonuclear neutrophils but extends their survival. Eur. J. Immunol. 31:1952.[Medline]
RIIIpmn mediates transmembrane signaling events distinct from FcRII. J. Exp. Med. 171:1239.RII and FcRIII in immune complex stimulation of human neutrophils. J. Biol. Chem. 29:20659.
receptor III represents the dominant receptor structure for immune complex activation of neutrophils. Eur. J. Immunol. 22:811.[Medline]
receptor (FcRII) on human neutrophils is essential for the IgG-induced respiratory burst and IgG-induced phagocytosis. J. Immunol. 142:2365.[Abstract]
receptor classes on human leukocytes. J. Exp. Med. 171:1333. receptor III on human neutrophils. J. Immunol. 150:3030.[Abstract]
-subunits and a common -subunit: the lymphocyte function-associated antigen (LFA-1), the C3bi complement receptor (OKM1/Mac-1), and the p150,95 molecule. J. Exp. Med. 158:1785. receptor autoantibodies from patients with Sjögren’s syndrome do not react with native receptor on human polymorphonuclear leukocytes. J. Autoimmun. 9:181.[Medline]
2-integrins (CD11b/CD18) promote apoptosis of human neutrophils. FASEB J. 11:1177.[Abstract]
receptor IIa and Fc receptor IIIb induces different proadhesive phenotypes on human neutrophils. J. Immunol. 159:3940.[Abstract]
2 integrin CD11b/CD18 in neutrophil apoptosis: a homeostatic mechanism in inflammation. Immunity 5:653.[Medline]
receptor type III (FcRIII, CD16) triggers cell activation through interaction with complement receptors. J. Immunol. 157:1184.[Abstract]
-glucan-binding lectin site of complement receptor type 3 (CD11b/CD18). J. Immunol. 156:1235.[Abstract]
-chain (CD11b) potentially involved in intramembrane complex formation with glycophosphatidylinositol-anchored FcRIIIB (CD16) molecules. J. Immunol. 19:5452.
receptor IIIB and the complement receptor type 3. J. Immunol. 153:1769.[Abstract]
M2, CD11b/CD18) and FcRIII cooperate in generation of a neutrophil respiratory burst: requirement for FcRII and tyrosine phosphorylation. J. Cell Biol. 125:1407.. J. Biol. Chem. 275:23065.This article has been cited by other articles:
|
|
 |
|
  L. Dyugovskaya, A. Polyakov, P. Lavie, and L. Lavie Delayed Neutrophil Apoptosis in Patients with Sleep Apnea Am. J. Respir. Crit. Care Med., March 1, 2008; 177(5): 544 - 554. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. G. Wann, Y.-H. Hsu, C.-C. Yang, C.-S. Lin, D. W. Tai, J.-S. Chen, C.-W. Hsiao, and C.-F. Chen Neutrophils in acidotic haemodialysed patients have lower intracellular pH and inflamed state Nephrol. Dial. Transplant., September 1, 2007; 22(9): 2613 - 2622. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. Ortiz-Stern and C. Rosales Fc{gamma}RIIIB stimulation promotes {beta}1 integrin activation in human neutrophils J. Leukoc. Biol., May 1, 2005; 77(5): 787 - 799. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. Kawakami, S. Ohashi, Y. Kawa, H. Takahama, M. Ito, Y. Soma, and M. Mizoguchi Elevated Serum Granulocyte Colony-Stimulating Factor Levels in Patients With Active Phase of Sweet Syndrome and Patients With Active Behcet Disease: Implication in Neutrophil Apoptosis Dysfunction Arch Dermatol, May 1, 2004; 140(5): 570 - 574. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 G. van Zandbergen, J. Gieffers, H. Kothe, J. Rupp, A. Bollinger, E. Aga, M. Klinger, H. Brade, K. Dalhoff, M. Maass, et al. Chlamydia pneumoniae Multiply in Neutrophil Granulocytes and Delay Their Spontaneous Apoptosis J. Immunol., February 1, 2004; 172(3): 1768 - 1776. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. Schettini, G. Salamone, A. Trevani, S. Raiden, R. Gamberale, M. Vermeulen, M. Giordano, and J. R. Geffner Stimulation of neutrophil apoptosis by immobilized IgA J. Leukoc. Biol., October 1, 2002; 72(4): 685 - 691. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
L. Ottonello, G. Frumento, N. Arduino, M. Bertolotto, P. Dapino, M. Mancini, and F. Dallegri Differential regulation of spontaneous and immune complex-induced neutrophil apoptosis by proinflammatory cytokines. Role of oxidants, Bax and caspase-3 J. Leukoc. Biol., July 1, 2002; 72(1): 125 - 132. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
