(Received for publication, November 21, 1996, and in revised form, March 12, 1997)
From the School of Biological Sciences, Life Sciences Building, University of Liverpool, P. O. Box 147, Liverpool L69 3BX, United Kingdom
Soluble immune complexes bind to unprimed
neutrophils and generate intracellular Ca2+
transients but fail to activate the NADPH oxidase. Following priming of
the neutrophils with either tumor necrosis factor or
granulocyte-macrophage colony-stimulating factor, stimulation of the
cells with the soluble immune complexes leads to an enhanced Ca2+ signal and significant secretion of reactive oxidants.
The enhanced Ca2+ signal observed in primed neutrophils
results from the influx of Ca2+ from the external
environment and is partly sensitive to tyrosine kinase inhibitors. This
is in contrast to the Ca2+ signal observed in unprimed
neutrophils, which arises from the mobilization of intracellular
stores. When the surface expression of Fc
RIIIb on primed neutrophils
was decreased either through incubation with Pronase or
phosphoinositide-specific phospholipase C, the extra enhanced
Ca2+ mobilization seen in primed cells was significantly
lowered, while the initial rise in intracellular Ca2+ was
unaffected. Depletion of Fc
RIIIb had no significant effect on the
Ca2+ transients in unprimed neutrophils. Cross-linking
Fc
RII, but not Fc
RIIIb, induced increases in intracellular
Ca2+ in unprimed neutrophils, while cross-linking either of
these receptors increased Ca2+ levels in primed
neutrophils. The Fc
RII-dependent intracellular Ca2+ rise in primed cells was unaffected by incubation in
Ca2+-free medium, whereas the
Fc
RIIIb-dependent transient was significantly decreased
when Ca2+ influx was prevented in Ca2+-free
medium supplemented with EGTA. Cross-linking either Fc
RII or
Fc
RIIIb in primed or unprimed cells failed to stimulate substantial levels of inositol 1,4,5-trisphosphate production. These results indicate that following stimulation of primed neutrophils with soluble
immune complexes the enhanced Ca2+ mobilization observed is
the result of a functional activation of the
glycosylphosphatidylinositol-linked Fc
RIIIb.
Neutrophils play a major role in host defense via the phagocytosis
and destruction of pathogens during acute inflammation. The binding of
opsonized bacteria or immune complexes to neutrophil immunoglobulin
receptors (FcR) can activate a number of processes such as
phagocytosis, degranulation and activation of the NADPH oxidase (1, 2).
However, in addition to this protective role, inappropriate activation
of neutrophils to release oxygen metabolites and granule enzymes within
tissues can result in tissue damage in inflammatory conditions such as
rheumatoid arthritis (1, 2). Furthermore, immune complex deposition
followed by subsequent neutrophil activation is important in the
pathogenesis of serum sickness, the Arthus reaction, acute
glomerulonephritis, and other idiopathic inflammatory diseases (3, 4).
Neutrophils are activated within the synovial joints of rheumatoid
patients (5-10), and we have shown that synovial fluid from patients
with rheumatoid arthritis contains both soluble and insoluble
IgG-containing immune complexes, which are capable of activating
neutrophils (11-13). However, the soluble immune complexes only
activate neutrophils that have been previously primed in
vivo or in vitro by GM-CSF1
or
-interferon and this activation leads to the secretion
of reactive oxidants and granule enzymes (11-13). In contrast,
insoluble immune complexes activate both primed and unprimed
neutrophils with nearly equal efficacy but the oxidants generated are
largely retained intracellularly, i.e. within
phagolysosomes. Thus, soluble immune complexes may be of greater
importance than insoluble complexes in the destructive processes that
occur during inflammatory diseases because they activate the secretion
of large quantities of cytotoxic products. The identification of the
molecular processes that occur during neutrophil priming to allow
activation of oxidant secretion in response to soluble complexes may
therefore be of value in identifying new targets for therapeutic
intervention.
IgG-containing immune complexes stimulate neutrophils via Fc
receptors, two types of which are expressed on freshly isolated control
blood cells (14). Fc
RII (CD32) is a 40-kDa transmembrane-spanning molecule with a cytoplasmic tail that allows its interaction with G-proteins, whereas Fc
RIIIb is a heavily glycosylated molecule of
50-70 kDa, linked to the membrane via a glycosylphosphatidylinositol anchor (15). Thus, although Fc
RIIIb has the potential for rapid lateral mobility within the membrane, it is unable to couple directly to either G-proteins or to the cytoskeleton, and it is unclear how
ligation to this receptor might lead to the transduction of intracellular signals. Expression of Fc
RIIIb on the plasma membrane (100,000-200,000 molecules/cell) is approximately 10-15-fold greater than the expression of Fc
RII (7000-15,000 molecules/cell).
There is much debate as to the independent and co-operative roles of
FcRII and Fc
RIIIb in neutrophil activation. Neither receptor
binds monomeric IgG, but they bind dimers, trimers, immune complexes,
and opsonized particles (16, 17). Although reports in the literature
are conflicting, it is currently thought that occupancy of Fc
RII
leads to activation of reactive oxidant production, degranulation, and
phagocytosis (18-22). While there have been a few reports of
Fc
RIIIb occupancy inducing functional responses, its main role is
thought to enhance the binding of immune complexes and to augment the
function of Fc
RII. Similarly, the intracellular signaling molecules
generated via Fc
RII and Fc
RIIIb ligation are undefined. Although
it is not clear how Fc
RIIIb, which lacks a transmembrane or
cytoplasmic domain can generate intracellular signals, it has recently
been reported that occupancy of either Fc
RII or Fc
RIIIb can
activate Src family non-receptor tyrosine kinases (23-27). Occupancy
of Fc
RII is associated with the activation and translocation of Fgr
to the Triton-insoluble cytoskeleton, while Fc
RIIIb is associated
with Hck activation. Ligation of either Fc
RII or Fc
RIIIb can
mediate a cytosolic Ca2+ increase that has been reported to
come primarily from the release of intracellular stores, and it has
been suggested that occupancy of Fc
RII and Fc
RIIIb can lead to
the generation of both independent and synergistic Ca2+
fluxes (28-30). However, the mechanisms by which Fc
R engagement leads to an increase in intracellular Ca2+ are unclear;
reports indicate that that it may be pertussis toxin-insensitive and
largely independent of inositol 1,4,5-trisphosphate (IP3) accumulation (31, 32). Most observations have, however, been made
following stimulation with large insoluble immune complexes or by
cross-linking individual Fc
receptors. Few reports have examined
changes in Fc
R function during priming (33), and the method used to
isolate neutrophils may influence the results obtained because some
separation procedures may result in inadvertent priming of the
neutrophil (34, 35).
In this study, we have investigated Ca2+ mobilization in
neutrophils stimulated with soluble immune complexes in both primed and
unprimed neutrophils. We present evidence that the initial cytosolic
Ca2+ increase observed following stimulation with soluble
complexes is the result of ligation of FcRII and that this
interaction is insufficient to activate the respiratory burst. When
soluble complexes are added to primed neutrophils oxidant secretion is initiated and an enhanced mobilization of Ca2+ results,
mainly via influx from extracellular sources. It would appear that this
extra intracellular Ca2+ signal results from functional
activation of Fc
RIIIb during priming. These studies add new insights
into the mechanisms by which cellular priming alters the functional
responsiveness of neutrophils during inflammatory activation.
Neutrophil isolation medium was from Cardinal
Associates Inc., RPMI 1640 medium from Flow Laboratories; Fluo-3-AM,
erbstatin, and genistein were from Calbiochem. rGM-CSF was a
non-glycosylated peptide from Glaxo and had an activity of 1.5
milliunits/mg of protein in the AML-193 proliferation assay. TNF
was
from National Institute for Biological Standards and Controls, Potters
Bar, United Kingdom. F(ab
)2 fragments of monoclonal
antibody 3G8 (recognizing Fc
RIIIb) and Fab fragments of IV.3
(recognizing Fc
RII) were from Medarex, Inc. PI-PLC was from
Boehringer Mannheim. All other specialist reagents were from Sigma.
Neutrophils were isolated from the venous blood of healthy volunteers by centrifugation on neutrophil isolation medium for 15 min at 400 × g (36). After removal of contaminating erythrocytes by hypotonic lysis, purified neutrophils were suspended in RPMI 1640 medium and counted. Neutrophil purity (assessed by Wright's staining) and viability (assessed by trypan blue exclusion) were >97% and >95%, respectively.
Neutrophil PrimingNeutrophils were primed using either
GM-CSF or TNF. Cells (107/ml) were incubated in the
presence and absence of GM-CSF (50 units/ml) for 1 h at 37 °C,
whereas TNF
(50 ng/ml) was added 10 min prior to stimulation.
Synthetic immune complexes were made from human serum albumin and rabbit anti-human serum albumin antibodies as described previously (37). The antigen was titrated against constant antibody concentration and the A450 nm measured to identify equivalence. The soluble complexes were formed at 6 × antigen equivalence and were briefly centrifuged (2 min at 13,000 × g in a microcentrifuge) to remove any contaminating insoluble immune complexes that may have been present. Soluble complexes were formed at 180 µg/ml antigen and 125 µg/ml antibody. A 10% (v/v) solution of complexes was used routinely for neutrophil stimulation.
Reactive Oxidant ProductionChemiluminescence was measured at 37 °C in neutrophil suspensions (5 × 105/ml) in RPMI 1640 medium that was supplemented with 10 µM Luminol using an LKB 1251 luminometer (38, 39). For measurements of superoxide secretion, neutrophils were suspended in RPMI 1640 supplemented with 75 µM cytochrome c (40). Absorption changes were recorded at 550 nm using a Bio-Rad 3550 kinetic plate reader. Reference wells contained 30 µg/ml superoxide dismutase.
Intracellular Free Ca2+Changes in intracellular Ca2+ concentration, [Ca2+]i, were monitored with the fluorescent probe Fluo-3. Neutrophils in RPMI 1640 (2 × 107/ml) were loaded by incubation at 37 °C for 30 min with 2 µM Fluo-3 AM. The cells were then washed twice and resuspended at 2 × 106/ml in Ca2+-free medium (145 mM NaCl, 1 mM Na2HPO4·2H2O, 0.5 mM MgSO4·7H2O, 5 mM glucose, 20 mM Hepes, pH 7.4) to which 1 mM CaCl2 was added as required. In some experiments, 1 mM EGTA was added to media devoid of Ca2+. Fluorescence was then measured at 505 nm excitation and 530 nm emission. Calibration of changes in intracellular Ca2+ levels was as described in Ref. 41 using a Kd for Fluo-3 of 864 nM at 37 °C.
Receptor Cross-linking and mAb Blocking StudiesFor
measurements of intracellular Ca2+ following the
cross-linking of receptors, Fluo-3-loaded neutrophils (2 × 106/ml) were incubated with 1 µg/ml
Fab/F(ab)2 fragments of appropriate monoclonal antibodies
at 37 °C prior to cross-linking with 40 µg/ml F(ab
)2
fragments of goat anti-mouse IgG. For mAb blocking studies, 3 × 106 neutrophils were incubated with 3 µg of
Fab/F(ab
)2 fragments in a volume of 250 µl for 10 min at
37 °C prior to dilution (to 5 × 105 cells/ml) and
stimulation of cells with soluble immune complexes.
Neutrophils were
incubated for 45 min in the presence (primed) and absence (control) of
50 units/ml GM-CSF. FcRII and Fc
RIIIb were then cross-linked (as
described above), and samples were removed at time intervals.
IP3 was then extracted by rapidly mixing samples with 0.2 volume of ice-cold 20% (v/v) perchloric acid. After neutralizing
extracts, IP3 was quantified using a competitive binding
assay (Amersham, UK) using D-myo-inositol
1,4,5-trisphosphate as a calibration standard.
Addition of synthetic soluble IgG immune complexes to
freshly isolated neutrophils failed to activate reactive oxidant
generation (Fig. 1), as detected by Luminol
chemiluminescence. However, if the cells were primed by prior addition
of GM-CSF (50 units/ml for 1 h) or TNF (50 ng/ml for 10 min),
then the soluble complexes activated a rapid and transient burst of
oxidase activity that peaked 2-3 min after addition (Fig. 1,
A and C). Luminol-dependent chemiluminescence detects both intra- and extracellular oxidant production (39), but the use of cytochrome c, which is a
large basic molecule that cannot permeate the cell, specifically
measures O
2 secretion (38). The addition of soluble immune
complexes to unprimed cells similarly did not activate O
2
secretion but did lead to a rapid and significant secretion of
O
2 in cells that had been primed either with TNF
or with
GM-CSF (Fig. 1, B and D).
Intracellular Ca2+ Mobilization
Resting levels of
intracellular Ca2+ were approximately 140 nM
(n = 18) in unstimulated neutrophils. When soluble
immune complexes were added to unprimed cells, there was a rapid and
transient monophasic increase in the level of intracellular
Ca2+, which reached a maximal level of 0.9-1.0
µM Ca2+ approximately 30-45 s following
stimulation (Fig. 2). This increase in intracellular
Ca2+ was comparable to that induced by addition of 0.1 µM fMet-Leu-Phe (data not shown). When the neutrophils
were primed by exposure to TNF prior to stimulation with the soluble
complexes, there was again a rapid rise in intracellular
Ca2+ levels (Fig. 2A), but the kinetics of the
increase were altered. The initial increase in intracellular
Ca2+ seen in unprimed neutrophils was unaffected, but this
was followed by a second, more sustained increase in intracellular
Ca2+. Similarly, when the neutrophils were primed with
GM-CSF (Fig. 2B), the intracellular Ca2+
mobilization induced by the soluble immune complexes was enhanced and
prolonged in comparison to that observed in unprimed cells. Increasing
the concentrations of soluble immune complexes (to 20% and 30%, v/v)
did not change either the kinetics or the magnitude of the
intracellular Ca2+ transients seen in unprimed cells. Thus,
priming does not affect the affinity of binding of soluble immune
complexes under these conditions.
Effect of EGTA on Intracellular Ca2+ Mobilization
To determine the source of the increased
intracellular Ca2+ levels in both unprimed and primed
neutrophils, incubations in Ca2+-free medium in the
presence of 1 mM EGTA were performed. Fig. 3A shows that when unprimed cells were
stimulated with soluble immune complexes in Ca2+-free
buffer, there was no significant difference in the kinetics of
intracellular Ca2+ changes compared with those obtained in
Ca2+-containing buffer. Thus, in unprimed neutrophils, the
observed increases in intracellular Ca2+ arise from the
mobilization of intracellular stores. In contrast, when primed cells
were stimulated with soluble immune complexes Ca2+-free
buffer, the initial intracellular Ca2+ increase was
unaffected, but the extended "extra" Ca2+ signal only
seen in primed cells, was not observed (Fig. 3, B and
C). These experiments thus indicate that the enhanced and prolonged increase in intracellular Ca2+ observed only in
primed cells is due to Ca2+ influx from external
sources.
Role of Tyrosine Kinases in Ca2+ Mobilization
When unprimed neutrophils were incubated with the
tyrosine kinase inhibitor, erbstatin (1 µg/ml), the intracellular
Ca2+ response obtained following stimulation with the
soluble complexes was inhibited by 30% (± 8%, n = 6). When neutrophils were primed with either TNF (Fig.
4B) or GM-CSF (Fig. 4C), the
enhanced intracellular Ca2+ signal seen only in primed
neutrophils was similarly decreased. This concentration of erbstatin
decreased reactive oxidant generation stimulated by soluble immune
complexes in primed neutrophils by >95% (data not shown). Similar
results were obtained following addition of another tyrosine kinase
inhibitor genistein at 100 µM (data not shown). Addition
of erbstatin at 2 µg/ml inhibited the unprimed Ca2+
signal by over 50%, but did not result in greater inhibition of the
primed response (data not shown).
To determine whether erbstatin exhibited nonspecific toxic effects at
these concentrations, the experiments shown in Fig. 5
were performed. After addition of soluble immune complexes to primed
neutrophils incubated in the presence and absence of erbstatin, fMet-Leu-Phe was added. The intracellular Ca2+ transients
stimulated by fMet-Leu-Phe were unaffected by the inhibitor; this
fMet-Leu-Phe-induced response was identical to that observed following
its addition to suspensions that had not previously been stimulated by
immune complexes.
Role of Fc
When neutrophils were incubated with PI-PLC (at 0.25 units/ml for 30 min), the surface expression of FcRIIIb, which is
glycosylphosphatidylinositol-linked (14, 15), detected by
fluorescent-activated cell sorting was decreased to levels that were
indistinguishable from nonspecific antibody binding (Fig.
6A, i). This treatment had no effect on the
expression of Fc
RII (Fig. 6A, ii) or on the expression of CD11b (data not shown). Similarly, incubation with Pronase (0.05 mg/ml
for 30 min) also resulted in the decreased surface expression of
Fc
RIIIb but had no effect on expression of Fc
RII or CD11b (data
not shown).
When FcRIIIb expression on the surface of primed neutrophils was
depleted by treatment with either Pronase or PI-PLC, the soluble immune
complexes failed to activate the respiratory burst (Fig. 6,
inset). Neutrophils treated in this way can, however, still
generate reactive oxidants in response to large, insoluble immune
complexes or PMA (13).
When unprimed neutrophils were depleted of surface FcRIIIb by
treatment with PI-PLC, there was very little effect on the intracellular Ca2+ rise stimulated by soluble immune
complexes (Fig. 6B, i). This result indicates
that expression of Fc
RIIIb is not required for the
transient, intracellular Ca2+ increase that arises from the
mobilization of intracellular stores following the binding of soluble
immune complexes to the surface of unprimed cells. However, in primed
cells, Pronase treatment had a marked effect on the intracellular
Ca2+ signals (Fig. 6B, ii). In primed
cells depleted of Fc
RIIIb, the initial intracellular
Ca2+ transient was unaffected, but the later, sustained
intracellular Ca2+ signal that arises from Ca2+
influx was not observed. Similar results were obtained when the neutrophils were primed with TNF
and when Fc
RIIIb was depleted from the cell surface by incubation with Pronase (data not shown). These data thus indicate distinct roles for individual Fc
R in the
control of these intracellular Ca2+ signals in primed and
unprimed neutrophils. Thus, the initial intracellular Ca2+
transient that is seen in both unprimed and primed neutrophils appears
to be due to signals generated via occupancy of Fc
RII; this
intracellular Ca2+ signal arises from the mobilization of
intracellular stores. In contrast, the extra intracellular
Ca2+ signal that is only seen in primed cells requires
Fc
RIIIb and arises from Ca2+ influx.
To
confirm the above conclusions on the roles of FcRII and Fc
RIIIb
in the differential intracellular Ca2+ signals,
cross-linking Fab/F(ab
)2 fragments of specific anti-Fc
R mAbs was employed. Incubation of unprimed neutrophils with Fab fragments of IV3 (anti-Fc
RII), followed by cross-linking with goat
anti-mouse F(ab
)2 fragments, resulted in a monophasic
increase in intracellular Ca2+ (Fig.
7A). The kinetics of this intracellular
Ca2+ increase were unaltered when the cells were primed
with either GM-CSF (Fig. 7A) or TNF
(data not shown), but
a greater intracellular Ca2+ transient was observed in
primed cells. The kinetics of these increases in intracellular
Ca2+ resemble those observed during incubation of unprimed
neutrophils with soluble immune complexes (Fig. 2). However, when
unprimed neutrophils were incubated with F(ab
)2 fragments
of 3G8 (anti-Fc
RIIIb) prior to cross-linking, no increases in
intracellular Ca2+ were observed (Fig. 7B).
Thus, ligation of Fc
RIIIb in unprimed neutrophils fails to generate
increases in intracellular Ca2+. However, if the cells were
primed with either GM-CSF (Fig. 7B) or TNF
(data not
shown), then cross-linking of 3G8 resulted in increases in
intracellular Ca2+. The kinetics of these increases in
intracellular Ca2+ were significantly slower than those
observed following ligation of Fc
RII and very similar to the
kinetics of the extra intracellular Ca2+ signal only seen
in primed cells (Fig. 2). These results thus clearly show that ligation
of Fc
RII can lead to rapid intracellular Ca2+ transients
in either primed or unprimed neutrophils and the priming process does
not alter the kinetics of these transients. In contrast, ligation of
Fc
RIIIb results in elevations in intracellular Ca2+, but
only in cells that are primed. Thus, priming leads to a functional activation of Fc
RIIIb that results in its
ability to generate intracellular Ca2+ transients via
Ca2+ influx.
When intracellular Ca2+ levels were measured by
cross-linking FcRII in primed neutrophils similar responses were
observed in Ca2+-containing and in Ca2+-free
medium (containing 1 mM EGTA) (Fig.
8A). This result indicates that the
Ca2+ signal generated following ligation of this receptor
arises largely from the mobilization of intracellular stores. In
complete contrast, the Fc
RIIIb-dependent intracellular
Ca2+ signal was greatly decreased when the cells were
suspended in Ca2+-free medium (Fig. 8B),
indicating that Ca2+ influx is largely responsible for the
intracellular Ca2+ signal following ligation of this
receptor.
The Fab/F(ab)2 fragments were then used to confirm the
roles of these Fc
Rs in the generation of intracellular
Ca2+ signals in response to soluble immune complexes.
Addition of 3G8 F(ab
)2 fragments (anti-Fc
RIIIb) prior
to the addition of soluble immune complexes had no effect on the
intracellular Ca2+ transients generated in unprimed cells
(Fig. 9A), but significantly decreased the
extra Ca2+ signal observed in primed cells (Fig.
9B). However, addition of IV3 Fab fragments (anti-Fc
RII)
prior to the addition of soluble immune complexes inconsistently
affected the intracellular Ca2+ transients. It may be that
mAb IV3 is readily displaced from Fc
RII by the soluble immune
complexes. These experiments, however, confirm a role for Fc
RIIIb in
the extra intracellular signal seen in primed cells.
IP3 Production following Neutrophil Activation
Stimulation of neutrophils with fMet-Leu-Phe resulted
in a rapid and transient elevation in the levels of IP3
(Fig. 10A), which peaked within 10-15 s and
then declined. However, ligation of either FcRII (Fig.
10B) or Fc
RIIIb (Fig. 10C) failed to activate the production of significant levels of IP3 in either
primed or unprimed cells.
In this study we have investigated the mechanisms by which soluble
immune complexes activate neutrophils via Fc receptors. Defining
this interaction is important because several immune conditions are
associated with neutrophil activation via immune complexes (1-3).
These soluble immune complexes fail to activate the respiratory burst
when added to suspensions of unprimed cells. However, if the cells are
primed (as in our study using the cytokines GM-CSF and TNF
), then
soluble immune complex binding results in the secretion of
large quantities of reactive oxygen metabolites. Because neutrophil
function in vivo is invariably regulated by exposure to
priming agents and because these soluble immune complexes activate the
secretion of potentially tissue-damaging components, these results are
important in understanding the molecular pathology of certain
neutrophil-mediated conditions. In our study, we have used a neutrophil
isolation procedure that induces minimal priming during cell isolation.
Many commonly used isolation methods (33, 34, 42), however,
inadvertently prime both receptor number/function and NADPH oxidase
activity and hence experiments with cells isolated by such methods
cannot clearly distinguish between neutrophil responses that are
restricted to either the primed or the unprimed state.
Whereas the soluble immune complexes failed to activate the NADPH
oxidase in unprimed cells, the use of fluorescein
isothiocyanate-labeled complexes showed that they clearly bind to the
cell surface (data not shown). These fluorescein isothiocyanate-labeled
complexes activated neutrophils in identical ways to the unlabeled
complexes, indicating that their molecular properties were unaltered
during the labeling process. We have previously shown that the
synthetic soluble immune complexes used in these studies and those
isolated from rheumatoid synovial fluid, activate primed and unprimed
neutrophils by analogous mechanisms (12, 43). The fact that these
complexes activated increases in intracellular Ca2+ in
unprimed cells is indicative that they bind to functionally active
receptors on the cell surface. Hence, we conclude that the key event
that occurs during priming that allows the primed cell to generate
reactive oxidants in response to soluble immune complexes is
not increased complex binding. Indeed, priming results in
only small changes in the level of expression of either FcRII or
Fc
RIIIb, whereas surface expression of CR3 is greatly up-regulated (44). We thus searched for events downstream of receptor/ligand binding
that may account for the altered functional responses of primed
cells.
In unprimed cells, the soluble immune complexes activated a transient increase in intracellular Ca2+ that was due to mobilization of intracellular stores and was partly sensitive to inhibition with tyrosine kinase inhibitors. Several other reports have shown that, while intracellular Ca2+ transients are necessary for NADPH oxidase activation, they are in themselves not sufficient to induce activation (45, 46). When the cells were primed, however, an extra intracellular Ca2+ transient was generated. This arose from influx of Ca2+ from external sources and was also partly sensitive to inhibition by tyrosine kinase inhibitors. These data indicate that this extra intracellular Ca2+ signal may be due to signals generated via receptors that become "activated" during the priming process. We thus set out to identify these putative receptors.
Circulating blood neutrophils possess two types of receptors that
recognize IgG-containing immune complexes, namely FcRII and
Fc
RIIIb (14). Fc
RIIIb is linked to the plasma membrane via a
glycosylphosphatidylinositol linkage that can be easily cleaved either
in vivo or in vitro with Pronase or PI-PLC (15, 17). Our first experiments thus used Pronase and PI-PLC to deplete the
cell surface of Fc
RIIIb while not affecting expression of Fc
RII.
Fc
RIIIb-depleted (unprimed) neutrophils could generate intracellular
Ca2+ transients in response to soluble immune complexes,
and the kinetics of these transients were identical with those observed
in untreated (Fc
RIIIb-expressing) cells. However, the extra
intracellular Ca2+ signal normally observed during priming
was not observed after Fc
RIIIb depletion. Furthermore, this extra
intracellular Ca2+ signal was not observed when primed
cells were incubated F(ab
)2 fragments of 3G8
(anti-Fc
RIIIb) prior to addition of soluble immune complexes. Taken
together, our experiments indicate that in unprimed neutrophils,
soluble immune complexes bind to the cell surface via binding to
Fc
RII and this results in an increase in intracellular
Ca2+: this interaction fails to activate the NADPH oxidase.
Upon priming, the soluble complexes now generate an extra intracellular
Ca2+ signal that arises via the functional
activation of Fc
RIIIb. This extra Ca2+ signal
arises from Ca2+ influx and requires protein-tyrosine
phosphorylation. Experiments cross-linking individual receptors confirm
these conclusions that ligation of Fc
RII can elevate intracellular
Ca2+ via internal store mobilization in either primed or
unprimed cells, whereas ligation of Fc
RIIIb can only elevate
intracellular Ca2+ via Ca2+ influx in primed
cells. If the traces of the intracellular Ca2+ transients
generated via ligation of Fc
RII and Fc
RIIIb are combined, they
closely resemble the kinetics of intracellular Ca2+
transients observed when primed and unprimed neutrophils are stimulated
with soluble immune complexes. It was noteworthy that ligation of
either Fc
RII or Fc
RIIIb failed to activate the synthesis of
substantial levels of IP3, in agreement with previous
reports (31, 32).
There are numerous reports in the literature regarding the interaction
of immune complexes with neutrophil Fc receptors, but none have
addressed the interaction of soluble complexes with primed and unprimed
cells. In our experiments, the intracellular Ca2+
transients generated by FcRII ligation were partly sensitive to
inhibition with tyrosine kinase inhibitors, confirming the recently
reported association of Fc
RII with tyrosine kinase activity (23, 24,
26, 27). Our data also confirm the importance of tyrosine kinase
activity and Fc
RIIIb function (27), and we propose that tyrosine
kinase activity becomes associated with Fc
RIIIb during priming and
regulates Ca2+ influx. We speculate that during priming a
Src-like tyrosine kinase (perhaps Hck; Ref. 27) becomes associated with
Fc
RIIIb. This could then result in the assembly of a new signaling
cassette linking Fc
RIIIb ligation to cell activation. Several
reports have shown that Fc
RIIIb can transduce signals, including
intracellular Ca2+ transients independently of Fc
RII
(28-30). A synergistic response of Fc
RIIIb and CR3 has previously
been reported, whereby CR3 immobilizes Fc
RII to the plasma membrane
via a cytoskeletal-dependent mechanism and then ligation of
Fc
RIIIb induces tyrosine activation in the proximity of Fc
RII
(25). In our experiments using synthetic soluble immune complexes,
anti-CR3 antibodies do not affect the ability of primed neutrophils to
activate a respiratory burst. Thus, the dependence of the response on
CR3 may be diminished in primed neutrophils.
These studies shed new insights into the mechanisms by which
soluble immune complexes activate neutrophils in pathological conditions that may result in tissue damage. There is much interest in
the pharmacological down-regulation of neutrophil function in
inflammatory conditions, but any therapy designed to interfere with
neutrophil-mediated responses must be selective; such therapy must
specifically block the harmful consequences of neutrophil activation
that lead to tissue damage (e.g. result in secretion of
cytotoxic products) but not compromise opsono-phagocytosis and host
defense against infections. It is interesting to note that individuals
with FcRIIIb deficiency are often asymptomatic and do not generally
suffer recurrent infections (47, 48). Therefore, we predict that
neutrophils from these individuals will not be able to secrete reactive
oxygen metabolites in response to soluble immune complexes, and thus it
will be necessary to test this hypothesis using neutrophils from these
individuals. If correct, then our data offer the possibility that
selective inhibition of the signaling pathway from Fc
RIIIb/soluble
immune complex ligation to activation of the NADPH oxidase could
provide the basis for a novel therapeutic strategy for the treatment of immune disorders involving soluble immune complexes and primed neutrophils.