| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
* Department of Pathology, University of Michigan Medical School, Ann Arbor, MI 48109; and
DNAX, Palo Alto, CA 94304
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The oxygen-dependent antimicrobial arsenal in neutrophils is generated by the multicomponent enzyme, NADPH oxidase (17, 18). This oxidase system, which catalyzes the reduction of oxygen to O2-, is dormant in unstimulated cells, but in stimulated neutrophils undergoes activation by phosphorylation and translocation to the cell membrane of a major cytosolic component, p47phox, also accompanied by translocation of p67phox, p21rac, and p40phox to the integral membrane protein, flavocytochrome b558 (19, 20, 21). After assembly, electron transfer from NADPH to oxygen occurs, generating O2-, which undergoes dismutation to H2O2, a key product for phagocyte killing of ingested bacteria (17, 18). Little is known about the status of NADPH oxidase during sepsis. Increased levels of cytosolic p47phox in neutrophils have been reported in humans with sepsis occurring after severe trauma (22). In vitro exposure of neutrophils to bacterial LPS does not activate NADPH-oxidase, but primes neutrophils for enhanced assembly of NADPH oxidase when cells are subsequently stimulated with the bacterial chemotactic peptide, fMLP (23). To elucidate the linkage between complement activation and neutrophil dysfunction during sepsis, we evaluated the status of NADPH oxidase in blood neutrophils from CLP rats treated with either preimmune IgG or anti-C5a IgG, and also used in vitro approaches in which neutrophils were incubated with C5a. Treatment of CLP rats with anti-C5a has been shown to be protective against the lethal consequences of sepsis (12). The current studies define the molecular basis for C5a-induced loss of the respiratory burst in neutrophils during sepsis.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Unless otherwise indicated, all reagents were purchased from Sigma-Aldrich (St. Louis, MO).
Experimental sepsis induced by CLP
Male Long-Evans-specific pathogen-free rats (275–300 g in body weight; Harlan, Indianapolis, IN) were anesthetized by i.p. administration of ketamine (20 mg/100 g body weight). Via a 2-cm abdominal midline incision, the cecum was tightly ligated below the ileocecal valve, carefully avoiding bowel obstruction. The cecum was then punctured through and through with a 21-gauge needle. After repositioning the bowel, the abdominal incision was closed in layers (4-0 silk suture and skin clips; Ethicon, Somerville, NJ). Immediately after CLP (time 0), animals received either 400 µg preimmune rabbit IgG or 400 µg rabbit anti-C5a peptide IgG i.v., the latter generated against the middle peptide region of rat C5a (corresponding to amino residues 17–36). This Ab is described elsewhere (10). Before and after surgery, rats had unrestricted access to food and water.
Neutrophil isolation
Human or rat neutrophils were isolated from whole blood by the traditional techniques of Ficoll-Paque gradient centrifugation (Pharmacia Biotech, Uppsala, Sweden) and dextran sedimentation, followed by hypotonic lysis of residual RBCs. Blood was drawn using 10% anticoagulant citrate dextrose (Baxter, Deerfield, IL).
Phagocytosis assay
Rat neutrophils (50 µl of 106 cells/ml HBSS; Life Technologies, Bethesda, MD) were layered onto adhesion reaction fields on glass slides (Bio-Rad, Munich, Germany) and allowed to settle down. After washing, the adherent cells were then incubated in absence or presence of different amounts of C5a (0.01–100 nM) for 1 h at 37°C, followed by a 30-min incubation step with IgG-coated zymosan particles (100/polymorphonuclear cells (PMN)). After additional washing, cells were stained with May-Gruenwald stain and trypan blue. Phagocytotic uptake was evaluated by light microscopy by scoring random fields.
Measurement of respiratory burst
O2- generation was measured by the superoxide dismutase (SOD)-inhibitable reduction of ferricytochrome c. Neutrophils (5 x 106 cells/ml) were preincubated with ferricytochrome c (80 µM) ± SOD (24 µg/ml) in the presence or absence of human C5a (huC5a) for 60 min at 37°C. As indicated, cells were subsequently stimulated by PMA (100 ng/ml for 10 min). Then absorbance of the supernatant fluids was measured at 550 nm, and the amount of O2- calculated was based on the amount of reduced ferricytochrome c in the presence or absence of SOD. Hydrogen peroxide (H2O2) generation was determined in the presence of 1 mM sodium azide. As indicated, neutrophils (2 x 106 cells/ml) were pretreated with various concentrations of huC5a (1.0–100 µM) for 60 min at 37°C. To stimulate neutrophils, cells were incubated with various concentrations of PMA (10–1000 ng/ml) for additional 10 min. The reaction was stopped by addition of TCA (50% v/v). Then, ferrous ammonium sulfate (1.5 mM) and potassium thiocyanate (0.25 M) were added to the supernatant fluid. The absorbance of the ferrithiocyanate complex was measured at 480 nm and compared with a standard curve generated from dilutions of reference solutions of H2O2.
Translocation of p47phox and immunoblotting
After stimulation, neutrophil lysates were centrifuged (400 x g, 5 min) to remove nuclei and unbroken cells. The resulting supernatant fluids were further centrifuged (15,000 x g, 45 min) to isolate the crude cytosol-derived fraction (supernatant) and the crude membrane fraction (pellet). Equal amounts of protein of the crude subcellular fractions were separated under reducing conditions by 10% SDS-PAGE and transferred onto polyvinylidene fluoride membrane. The membrane was blocked in 5% milk (in PBS) for 1 h and then incubated with rabbit anti-p47phox IgG Ab at a concentration of 1 µg/ml (Genentech, San Francisco, CA). As a secondary Ab, alkaline phosphatase-conjugated goat anti-rabbit IgG (1:1000; Jackson ImmunoResearch Laboratories, West Grove, PA) was added, and the blot was developed using alkaline phosphatase color development system (Bio-Rad, Hercules, CA).
For determination of the phosphorylation state of p47phox, anti-p47phox mAb was used to immunoprecipitate from equal amount of cell lysates (100 µg). After Western blotting, anti-phosphoserine Ab (Zymed Laboratories, San Francisco, CA) was used for probing the membranes, followed by ECL (Amersham, Piscataway, NJ).
Mitogen-activated protein kinase (MAPK) activity
MAPK activity (p42/p44 MAPK or extracellular signal-regulated kinase (Erk) 1/2) of human neutrophils was determined by Western blot analysis, as per manufacturer’s instructions, for p42/p44 protein and phospho-p42/p44 protein (NEB, Beverly, MA). Image analysis of the signal on the films obtained by the chemiluminescence procedure was performed using Adobe System software (San Jose, CA).
Analysis of C5aR on neutrophils
Flow cytometric analysis was conducted immediately after blood collection. A total of 10 µl anti-C5aR serum (Research Diagnostics, Flanders, NJ) in 100 µl staining buffer (PBS with 0.1% sodium azide and 1% FBS) was incubated with 100 µl human whole blood for 30 min. After washing, the cells were suspended in FITC-conjugated rat anti-rabbit IgG (Biosource International, Thousand Oaks, CA) diluted 1/50 in staining buffer. Erythrocytes were lysed for 10 min by addition of 1x FACS lysing solution (BD PharMingen, San Diego, CA). After washing, the leukocytes were reconstituted in a fixing solution of 1% paraformaldehyde prepared in PBS plus 0.1% sodium azide and analyzed on a flow cytometer (Coulter, Miami, FL). Granulocytes were gated by the typical forward and side light scatter profiles. The same gated population in rat blood cells has been confirmed to be granulocytes by staining of whole blood with a rat granulocyte marker, HIS48 (BD PharMingen). More than 90% of gated cells were found to be granulocytes.
Statistical analysis
All values were expressed as mean ± SEM. Data sets were analyzed with one-way ANOVA; differences in the mean values among experimental groups were then compared using the Tukey multiple comparison test. Results were considered statistically significant where p < 0.05.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
During sepsis, plasma levels of complement-derived anaphylatoxins
are increased (see above) and alteration of the respiratory burst
(production of O2- and
H2O2) occurs, resulting in
impaired bacterial killing (9, 10, 11, 12, 13, 14). In CLP-induced
experimental sepsis, we have recently shown the
H2O2 response of
neutrophils to be defective in a manner that is C5a dependent
(12). We have now evaluated in blood neutrophils (from
normal rats or from rats 24 h after CLP) the status of the
assembly (activation) of NADPH oxidase, which requires translocation of
cytosolic p47phox to the cell membrane after
stimulation with PMA. CLP animals were treated i.v. at time 0 with 400
µg preimmune IgG or affinity-purified anti-rat C5a. The ability
of p47phox to translocate to the cell membrane
of neutrophils stimulated with PMA was evaluated in blood neutrophils
obtained 24 h after CLP. As shown in Fig. 1
, Western blot analysis of neutrophil
crude membrane fractions from control rats without sepsis revealed some
cell membrane-associated p47phox (arrow) in
nonstimulated (ctrl) neutrophils (lane 2) and a more
intense band in the membrane fraction of PMA-stimulated neutrophils
(lane 3). p47phox was not
detectable in membrane fractions if these neutrophils had been first
exposed to C5a (10 nM), followed by stimulation with PMA (100 ng/ml)
(Fig. 1, lane 4). Blood neutrophils from CLP rats (treated
with preimmune IgG) had only faint evidence of
p47phox in the cell membrane fraction (Fig. 1, lane 5). After stimulation of these cells with PMA, no
change in cell membrane-associated p47phox band
was found (Fig. 1, lane 6). If blood neutrophils from these
animals were first exposed in vitro to rat C5a (10 nM) followed by PMA,
the membrane-associated band vanished altogether (Fig. 1, lane
7). In CLP rats treated with anti-C5a, blood neutrophils that
were not otherwise treated in vitro showed a cell membrane-associated
p47phox band in the expected position in the gel
(Fig. 1, lane 8), and, as in control cells stimulated with
PMA (lane 3), this band was much more intense if the
cells were stimulated in vitro with PMA (lane 9).
When these neutrophils were first exposed in vitro to C5a, followed by
addition of PMA, the p47phox band in the cell
membrane fraction was undetectable (Fig. 1, lane 10). These
data indicate that PMA-induced translocation of
p47phox is defective in neutrophils from CLP
rats, that the in vivo acquisition of this defect is C5a dependent, and
that this defect can be reproduced by in vitro exposure of normal blood
neutrophils to C5a.
|
To evaluate the status of phagocytic function in neutrophils, two
separate experiments were conducted. In the first, blood neutrophils
were obtained from normal rats (ctrl) or from CLP rats (at 24 h)
that received at time 0 either 400 µg preimmune IgG or 400 µg
affinity-purified anti-rat C5a IgG. The results are shown in Fig. 2A. Neutrophils were incubated
with IgG-opsonized zymosan particles (100 particles/PMN, 30 min at
37°C), and the number of intracellular particles in neutrophils was
determined after this period of incubation. Neutrophils from normal
rats contained 3.0 ± 0.30 particles/cell, while neutrophils from
CLP rats pretreated with preimmune IgG showed a significantly lower
level of particle ingestion (1.2 ± 0.31 particles/cell),
indicating sepsis-induced dysfunction. In CLP rats treated with
anti-C5a, uptake of particles was the same as in neutrophils from
normal rats (Fig. 2A). These results indicate that the
phagocytic defect acquired by neutrophils during CLP-induced sepsis is
C5a related. In companion experiments (B), in vitro exposure
of blood neutrophils from normal rats to increasing amounts (0.01–100
nM) of rat C5a (at 37°C for 60 min), followed by incubation of cells
with particles (for 30 min at 37°C), led to a progressive reduction
in numbers of zymosan particles taken up into neutrophils as a function
of the concentration of C5a. The IC50 for C5a was
0.1 nM C5a.
|
For these and all subsequent experiments, human neutrophils were
exposed to human rC5a, and the effects on C5aR and on signaling
pathways were evaluated. For the first series of experiments, human
neutrophils were exposed to 100 nM C5a at different periods of time and
at different temperatures, and then the cells were evaluated for C5aR
content by flow cytometry. As shown in Fig. 3A, there was a measurable
reduction in detectable C5aR content on neutrophils exposed to C5a at
25°C for 30 min. In Fig. 3B, as can be seen in the
upper panel, there was virtually no difference in C5aR
content (mean channel fluorescence) over a 1-h period when neutrophils
were or were not exposed to 100 nM C5a, indicating that presence of C5a
does not affect detection of C5aR at 4°C. In the lower
panel, there was a striking reduction (nearly 40%) in C5aR
content on neutrophils exposed to C5a at either 25 or 37°C. By 15
min, the reduction in detectable C5aR was complete, at least over a 1-h
period. These data are consistent with other studies that suggest the
C5a/C5aR complex is partially internalized.
|
As shown in Fig. 4, H2O2 production was used as
an indicator for activation of NADPH oxidase. In A, human
blood neutrophils were incubated with 0.01–100 nM huC5a for 60 min at
37°C and the amount of
H2O2 generated was
determined. At doses of 1.0, 10, and 100 nM C5a, small but
statistically significant increases in
H2O2 were noted. In
B, a dose response for PMA-induced generation of
H2O2 revealed sharp
increases between 10 and 1000 ng PMA/ml, with no further increase at
1.0 µg/ml. In C, neutrophils were exposed to C5a
(0.01–100 nM) for 60 min at 37°C, and the subsequent amount of
H2O2 generated after
addition of PMA (100 ng/ml, for 10 min) was determined. When exposed to
0.1–100 nM C5a (60 min at 37°C), neutrophils showed a progressive
loss in production of H2O2
generation. Finally, exposure of neutrophils to 10 nM C5a for 0–60 min
revealed a time-dependent loss in
H2O2 production by these
neutrophils after addition of PMA (Fig. 4D).
|
, neutrophils were exposed in
vitro to buffer alone or 10 nM C5a for 60 min at 37°C, followed by
addition (where indicated) of PMA (100 ng/ml, for 10 min at 37°C). As
shown in Fig. 5A, C5a induced a modest, but significant
increase in O2- production in neutrophils that
were not otherwise treated. Addition of PMA to neutrophils not exposed
to C5a caused the expected robust increase in
O2-. In C5a-exposed neutrophils that were then
stimulated with PMA, the production of
H2O2 was reduced by nearly
60%. When companion studies were done with the same batch of
neutrophils, effects of the same manipulations on
H2O2 generation were
similar (Fig. 5B), with pre-exposure of neutrophils to C5a
(10 nM) causing an 81% reduction in
H2O2 after stimulation by
PMA. Thus, the effect of C5a exposure to neutrophils causes similar
inhibitory effects on O2- and
H2O2 generation following
cell stimulation with PMA.
|
. In normal (ctrl) neutrophils, a strong
cytosolic band for p47phox was found, together
with a faint band in the cell membrane fraction (lower
panel, lane 2). When normal neutrophils were exposed to
PMA, cytosolic p47phox was detected, but there
was also a strong band detected in the membrane fraction
(lane 5). When neutrophils were exposed only to C5a,
the cytosolic band for p47phox remained, but no
membrane band was detected (lane 3). In neutrophils
exposed to C5a followed by PMA, the intensity of the cytosolic band for
p47phox diminished in intensity and no
translocation to the cell membrane could be detected (lane
4). As shown in this and in other blots, fragmentation products of
p47phox were not detected, and supernatant
fluids failed to contain detectable p47phox
(data not shown). The defect in translocation of
p47phox after PMA stimulation could be detected
after an exposure time of cells to C5a of only 10 min (data not shown).
Thus, in vitro exposure of neutrophils to C5a under the conditions
described induces a defect in the ability of
p47phox to translocate to the cell membrane
after addition of PMA.
|
Studies were conducted using naturally occurring mediators to
determine whether the effects of C5a were mediator specific. The
mediators used and effects on
H2O2 responses are shown in
Table I. The concentrations of various
mediators used were based on preliminary data in which a dose of each
mediator that elicited a robust
H2O2 response in
neutrophils was determined (data not shown). In experiment A, the
suppressive effects of C5a on the PMA-induced
H2O2 response of human
neutrophil were confirmed (79% suppression). It was also shown that
the fMLP-induced H2O2
response could be reduced by 97% in cells that had been pre-exposed to
C5a. In experiment B, exposure of neutrophils to C5a suppressed the
TNF-
-induced response by 88%. Prior cell exposure to
platelet-activating factor resulted in a 92% reduction in the
PMA-induced H2O2 response.
Finally, prior cell exposure to platelet-activating factor reduced the
H2O2 response
to fMLP by 93%. These data indicate that, when neutrophils are exposed
to a variety of naturally occurring mediators, the
H2O2 responses induced by other naturally
occurring mediators are also suppressed. The importance of the potent
ability of C5a to suppress the respiratory burst in
neutrophils may relate to the fact that high concentrations (10 nM) of
C5a have been found in sera of humans with sepsis (4, 7).
|
To determine the possible basis for the failure of
p47phox to translocate in stimulated neutrophils
exposed to C5a, experiments were conducted to assess by Western blots
the status of phosphorylation of p47phox, using
Ab to phosphoserine after immunoprecipitation using mAb to
p47phox. Immunoprecipitation was performed using
equivalent amounts of cell lysates (100 µg protein). The results are
shown in Fig. 7. Faint evidence for
phosphorylation of p47phox was found in
unstimulated (ctrl) neutrophils (lane 1). When
neutrophils were exposed to 10 nM C5a for 60 min at 37°C, no increase
in phosphorylation of p47phox occurred
(lane 2), in striking contrast to
p47phox that was immunoprecipitated from PMA
(100 ng/ml, 37°C, 10 min)-stimulated neutrophils, in which strong
evidence of phosphorylation was found (lane 3). When
neutrophils were first exposed to C5a, followed by stimulation with
PMA, there was no evidence of increased phosphorylation of
p47phox (lane 4). Thus, C5a
blocks both phosphorylation of p47phox as well
as its translocation (Fig. 6
) to the cell membrane in PMA-stimulated
neutrophils.
|
Recently, extracellular signal-regulated kinase 1/2 (p42/p44 MAPK,
Erk 1/2) has been shown to be required for phosphorylation of
p47phox in neutrophils exposed to the bacterial
chemotactic peptide, fMLP (24). In addition, PD98059 has
been described to be a selective inhibitor of MEK-1 (25),
which immediately precedes Erk 1/2 in the signaling pathways of
neutrophils. Therefore, two separate experiments were conducted. In the
first, neutrophils were exposed to PD98059 (50 µM for 1 h at
37°C); these cells were then assessed for
H2O2 production in the
presence or absence of PMA (100 ng/ml). The results are shown in Fig. 8A. Cells exposed to PD98059
alone showed a small increase in
H2O2 when compared with
otherwise untreated cells (ctrl). When PMA was used as a stimulus, the
prior exposure of cells to PD98059 greatly suppressed (by 72%)
H2O2 production, when
compared with cells that were not exposed to PD98059, suggesting that
MEK-1 activation is required for PMA-induced generation of
H2O2 in neutrophils.
|
B). As shown, when anti-phospho-Ab was used,
neutrophils at 0 time point showed little evidence of phosphorylation
of p42/44 MAPK (Fig. 8B, upper panel). However,
when cells were exposed to C5a, there was an increase in
phosphorylation by 2 min, with a decline by 15 min and no signal by 60
min. Similarly, an increase in phosphorylation was observed when cells
were treated with PMA. But, when neutrophils were first exposed to 10
nM C5a, followed by PMA, all evidence of phospho-p42/44 disappears
(Fig. 8B, upper panel), indicating that
neutrophil contact with C5a inhibits phosphorylation of Erk (p42/44
MAPK) or triggers dephosphorylation of Erk. Proposed pathways leading to neutrophil dysfunction during sepsis
Fig. 9 depicts proposed mechanisms
of C5a-induced neutrophil dysfunction during sepsis, focusing on
intracellular signal transduction pathways involved in NADPH oxidase
activation. In nonstimulated (resting) neutrophils, the multicomponent
NADPH oxidase is dormant in its unassembled state and consists of
cytoplasmatic components (p21rac,
p40phox, and p47phox), or
cytoskeletal component (p67phox), and the
integral membrane protein, flavocytochrome
b558 (Fig. 9A). The
response of neutrophils to C5a (and to other agonists such as fMLP or
TNF-) involves interaction with transmembrane receptors. Separate
intracellular signaling pathways (especially involving protein kinase C
(PKC) and MAPK) are known to be activated, leading within minutes to
phosphorylation and activation of numerous target proteins. To simulate
cell activation in vitro, PMA was used as an in vitro stimulus of PKC
and MAPK. Whereas phosphorylation and translocation of
p47phox are considered key for assembly of NADPH
oxidase, generation of arachidonic acid also appears to play a key role
in activation events, which mainly involve the PKC and MAPK pathways
(Fig. 9B). In an attempt to simulate conditions of sepsis,
exposure of neutrophils to amounts of C5a (10 nM) that have been found
in serum during sepsis (3, 4, 7) paralyzed the process,
leading to assembly (activation) of NADPH oxidase. Using PMA-stimulated
neutrophils, C5a-induced defects were associated with the appearance of
at least three different events: greatly reduced phosphorylation of Erk
1/2, reduced phosphorylation of p47phox, and
failure of p47phox to translocate to the cell
membrane. It is also possible that C5a-induced dysfunction in
neutrophils might be linked to production of cAMP, which could activate
protein kinase A (PKA). In turn, PKA would inhibit both Raf-1 and B-Raf
(and reduce intracellular levels of extracellular
Ca2+). Under these conditions, it would also be
predicted that downstream activation of Erk 1/2 (p42/p44 MAPK) and
phosphorylation and translocation of p47phox
would be suppressed. It is also possible that the generation of
phospholipase A2-dependent arachidonic acid in
neutrophils exposed to C5a was inhibited by neutrophil exposure to C5a,
resulting in suppressed activation of NADPH oxidase. Collectively,
these events would lead to defective bacterial killing. To what extent
the suppressive effects of C5a are tied to the cAMP and PKA pathways or
have direct effects on the Raf-1/B-Raf pathway remains to be
determined. Such an outcome could occur by generation of cAMP and
PKA-induced inhibition of phospholipase C (PLC), PKC, and Raf-1/B-Raf;
inhibition of p42/p44 MAPK (Erk 1/2) phosphorylation; blockade of the
p47phox phosphorylation and translocation; and,
possibly, inhibition of phospholipase
A2-dependent arachidonic acid, finally resulting
in the inability to assemble NADPH oxidase and generate an effective
respiratory burst, resulting in impaired bacterial killing (Fig. 9C).
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
In the present study, experimental sepsis was induced by CLP. This
model, which closely simulates events in human sepsis (5, 6), was used to investigate mechanisms of neutrophil dysfunction
during sepsis. Two aspects of innate immunity were found to be
seriously compromised in blood neutrophils: phagocytosis and the
ability to mount a respiratory burst. We have recently demonstrated
that in the CLP model of sepsis, a third component of innate immunity
is defective in neutrophils: chemotactic responsiveness
(16). All three acquired defects have now been shown to be
C5a dependent. Phagocytotic activity (ingestion of IgG-opsonized
zymosan particles) was suppressed in neutrophils from CLP animals, but
this response was completely restored when complement activation
product, C5a, was blocked in vivo by anti-C5a Abs (Fig. 2). When
normal neutrophils were exposed in vitro to C5a at levels found during
sepsis, there was a dose-dependent impairment of in vitro phagocytosis
(Fig. 2). Recently, it has been shown during phagocytosis that the
onset of the respiratory burst (featuring generation of reactive oxygen
species O2-,
H2O2) is linked to
phosphorylation of the NADPH oxidase cytosolic subunit,
p47phox, which is then translocated to the cell
membrane of the phagosome (27). A dysfunction of the
NADPH-dependent respiratory burst during sepsis has been described in
clinical (8, 9, 13, 14, 22) and in experimental settings
(12, 16). To elucidate the mechanisms involved,
neutrophils were activated in vitro by PMA, which bypasses conventional
membrane receptors often defective during sepsis (3, 16)
and which leads to direct activation of intracellular PKC. As expected,
activation of normal neutrophils by PMA induced robust
O2- and
H2O2 production. Exposure
of neutrophils to C5a induced the expected reduction in C5aR content
(Fig. 3), presumably due to partial internalization of C5a/C5aR
complexes. Our recent report of greatly reduced binding of rat C5a to
blood neutrophils obtained from CLP rats (16) appears to
be due to both receptor occupancy as well as loss of receptors due to
internalization. After in vitro exposure of neutrophils to C5a,
O2- and
H2O2 production in
PMA-activated cells was greatly inhibited in a dose- and time-dependent
manner, indicating that exposure to C5a in vitro or in vivo can result
in impairment of the respiratory burst (Fig. 4). Exposure of
neutrophils in vitro to C5a alone caused very modest production of
O2- and
H2O2. These responses are
known to be greatly magnified in the presence of cytochalasin B, which
amplifies the oxidative response by inhibiting internalization of NADPH
oxidase subunits (28). Similar to our demonstration of a
C5a-induced defect in the PMA-induced respiratory burst in neutrophils,
a recent study has demonstrated that C-reactive protein, which
activates the classical complement cascade, inhibits chemotaxis and the
PMA-induced respiratory burst by affecting PKC-dependent translocation
and phosphorylation of p47phox
(29). Phosphorylation of at least two serine residues on
p47phox and translocation of this molecule from
the cytosol to the membrane (which takes place after phosphorylation)
are required to initiate assembly and activation of NADPH oxidase
(21, 30). The proline-rich region of cytochrome
b558 is considered to interact with
the Src homology 3 domains of p47phox and
p67phox, enabling the docking process (19, 20). The subsequent electron transfer from NADPH to
O2 generates O2- and is
charge compensated by the opening of a proton channel in the cell
membrane of the neutrophil (19, 31). In the current study,
the presence of small amounts of p47phox in the
membrane fraction of nonstimulated neutrophils suggests the presence of
basal oxidase activity, in accordance with earlier reports (19, 20, 24). This could represent an artifact of neutrophil
isolation. After PMA stimulation, enhanced
p47phox translocation was observed, which was
completely abolished in neutrophils that had been pre-exposed to C5a
(Fig. 6), indicating a C5a-induced defect in NADPH oxidase activation
via blockade of p47phox translocation.
LPS, which is often present during sepsis, primes neutrophils for an
enhanced respiratory burst after subsequent stimulation of these cells
(27). The MAPK pathway has recently been shown in
neutrophils to be involved in the fMLP-induced phosphorylation of
p47phox (24). Because both
phosphorylation and translocation of p47phox
alone are not sufficient enough to fully activate NADPH oxidase
(32, 33, 34), it was of interest to determine whether the MAPK
pathway was also involved in activation of NADPH oxidase. Using the
potent and selective MEK-1 inhibitor, PD 98059 (35), the
H2O2 response of
neutrophils after PMA stimulation was significantly suppressed,
indicating that MEK-1 in the MAPK pathway is involved in activation of
NADPH oxidase (Fig. 8). Some studies have suggested that fMLP and C5a
may activate the MAPK pathway via tyrosine phosphorylation of p42/p44
MAPK (35, 36, 37), which has been shown to occur in human
astrocytes incubated with C5a. This process is time dependent, with a
peak at 15 min (37). Exposure of neutrophils to C5a
virtually abrogated all signs of PMA-induced
p47phox phosphorylation as well as p42/p44 MAPK
(Erk 1/2) phosphorylation (Figs. 7 and 8). Whereas transient
stimulation of neutrophils with C5a may activate the MAPK pathway, the
present data suggest that a persistent presence of C5a or the presence
of relatively high concentrations (10 nM or higher) of C5a results in
an inhibition of the MAPK pathway. This is consistent with other
findings that C5a causes the activation of Raf-1 and B-Raf, which are
upstream stimulators of MEK-1 in human neutrophils, with a peak of
activation at 5 min, followed by suppressed activation thereafter
(38). Activated p42/p44 MAPK (Erk 1/2) is known to
phosphorylate p47phox (24) and,
thereby, to initiate NADPH oxidase assembly in a PKC-independent way
(34). When assembled, p47phox
undergoes a continuous cycle of phosphorylation and dephosphorylation
throughout the period of O2- release, with the
phosphorylation reaction predominating (20). Termination
of oxidase activity correlates with dissociation of
p47phox/p67phox from
flavocytochrome b558 (20, 27). It is possible that events that occur during
internalization of C5a/C5aR and receptor recycling perturb the MAPK
pathway, interfering with the ability of PMA to induce generation of
H2O2.
Whereas many in vitro studies of NADPH oxidase exist, little is known
about alterations of NADPH oxidase occurring during sepsis. In the
present study, we demonstrated in later stages (12 h) of CLP-induced
sepsis nearly complete loss of p47phox
translocation after activation of neutrophils by PMA (Fig. 1). The loss
of the respiratory burst, which is essential for bacterial killing,
could be reversed in animals that had been treated with an anti-C5a
Ab preparation. In our recent studies, such treatment has greatly
improved survival rates during CLP-induced sepsis and has reduced the
accompanying multiorgan failure (12, 16). These in vivo
data strongly suggest that in the sepsis model, C5a is a major
contributor to NADPH oxidase dysfunction, and that this defect can be
reversed by treatment of animals with a blocking Ab to C5a. The current
studies establish that C5a blocks three critical steps in the pathway
leading to activation of NADPH oxidase: phosphorylation of the p42 MAPK
(Erk 2); phosphorylation of p47phox; and
translocation of p47phox to the cell
membrane.
Possible pathways involved in C5a-induced neutrophil dysfunction are
demonstrated in a hypothetical model (Fig. 9). Despite the persistent
presence of complement activation products (7) and
inflammatory mediators during sepsis (3, 4), the
intracellular signaling events are often transient (Ras, Raf-1, B-Raf,
Erk-1, PLC, etc.), with activation of these factors occurring within
minutes (37, 38, 39, 40). The brief duration of these events may
result from the loss of receptors through their phosphorylation and
rapid internalization (41). Lack of phosphorylation of
C5aR (42) was found to result not only in sustained
Ca2+ mobilization and MAPK activity, but also in
enhancement of the respiratory burst (43, 44), suggesting
that phosphorylation of C5aR may be a regulatory mechanism for NADPH
oxidase activity. Additionally, C5a has been shown to activate the
adenylate cyclase-cAMP pathway in neutrophils (40, 45).
Stimulation of adenylate cyclase by forskolin (1 h, 100 nM) reduced the
respiratory response of PMA-stimulated cells by >50% (data not
shown), suggesting involvement of cAMP and PKA in regulation of the
respiratory burst. Sustained elevation of cAMP in neutrophils has been
reported to suppress the O2- release
(40, 43, 46). If cAMP levels were elevated, the MAPK
response was also inhibited (47, 48, 49), which may represent
an additional mechanism of C5a-induced neutrophil dysfunction.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
2 Address correspondence and reprint requests to Dr. Peter A. Ward, Department of Pathology, University of Michigan Medical School, 1301 Catherine Road, Ann Arbor, MI 48109-0602. E-mail address: pward{at}umich.edu
3 Abbreviations used in this paper: CLP, cecal ligation/puncture; Erk, extracellular signal-regulated kinase; huC5a, human C5a; MAPK, mitogen-activated protein kinase; MEK, mitogen-activated protein/Erk kinase; PKA, protein kinase A; PKC, protein kinase C; PLC, phospholipase C; PMN, polymorphonuclear cell; SOD, superoxide dismutase.
Received for publication March 11, 2002. Accepted for publication July 8, 2002.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
|
|
 |
|
  A. Conway Morris, K. Kefala, T. S. Wilkinson, K. Dhaliwal, L. Farrell, T. Walsh, S. J. Mackenzie, H. Reid, D. J. Davidson, C. Haslett, et al. C5a Mediates Peripheral Blood Neutrophil Dysfunction in Critically Ill Patients Am. J. Respir. Crit. Care Med., July 1, 2009; 180(1): 19 - 28. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. A. Flierl, D. Rittirsch, B. A. Nadeau, D. E. Day, F. S. Zetoune, J. V. Sarma, M. S. Huber-Lang, and P. A. Ward Functions of the complement components C3 and C5 during sepsis FASEB J, October 1, 2008; 22(10): 3483 - 3490. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 H. Nitta, T. Imamura, Y. Wada, A. Irie, H. Kobayashi, K. Okamoto, and H. Baba Production of C5a by ASP, a Serine Protease Released from Aeromonas sobria J. Immunol., September 1, 2008; 181(5): 3602 - 3608. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. A. Flierl, D. Rittirsch, H. Gao, L. M. Hoesel, B. A. Nadeau, D. E. Day, F. S. Zetoune, J. V. Sarma, M. S. Huber-Lang, J. L. M. Ferrara, et al. Adverse functions of IL-17A in experimental sepsis FASEB J, July 1, 2008; 22(7): 2198 - 2205. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 P. A. Ward Role of the complement in experimental sepsis J. Leukoc. Biol., March 1, 2008; 83(3): 467 - 470. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. J. Anand, S. C. Gribar, J. Li, J. W. Kohler, M. F. Branca, T. Dubowski, C. P. Sodhi, and D. J. Hackam Hypoxia causes an increase in phagocytosis by macrophages in a HIF-1{alpha}-dependent manner J. Leukoc. Biol., November 1, 2007; 82(5): 1257 - 1265. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
D. Liu, F. Lu, G. Qin, S. M. Fernandes, J. Li, and A. E. Davis III C1 Inhibitor-Mediated Protection from Sepsis J. Immunol., September 15, 2007; 179(6): 3966 - 3972. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. Garantziotis, J. W. Hollingsworth, R. B. Ghanayem, S. Timberlake, L. Zhuo, K. Kimata, and D. A. Schwartz Inter-{alpha}-Trypsin Inhibitor Attenuates Complement Activation and Complement-Induced Lung Injury J. Immunol., September 15, 2007; 179(6): 4187 - 4192. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
O.-L. Brekke, D. Christiansen, H. Fure, M. Fung, and T. E. Mollnes The role of complement C3 opsonization, C5a receptor, and CD14 in E. coli-induced up-regulation of granulocyte and monocyte CD11b/CD18 (CR3), phagocytosis, and oxidative burst in human whole blood J. Leukoc. Biol., June 1, 2007; 81(6): 1404 - 1413. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. D. Wrann, N. A. Tabriz, T. Barkhausen, A. Klos, M. van Griensven, H. C. Pape, D. O. Kendoff, R. Guo, P. A. Ward, C. Krettek, et al. The Phosphatidylinositol 3-Kinase Signaling Pathway Exerts Protective Effects during Sepsis by Controlling C5a-Mediated Activation of Innate Immune Functions J. Immunol., May 1, 2007; 178(9): 5940 - 5948. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R.-F. Guo, L. Sun, H. Gao, K. X. Shi, D. Rittirsch, V. J. Sarma, F. S. Zetoune, and P. A. Ward In vivo regulation of neutrophil apoptosis by C5a during sepsis J. Leukoc. Biol., December 1, 2006; 80(6): 1575 - 1583. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R.-F. Guo, N. C. Riedemann, L. Sun, H. Gao, K. X. Shi, J. S. Reuben, V. J. Sarma, F. S. Zetoune, and P. A. Ward Divergent Signaling Pathways in Phagocytic Cells during Sepsis J. Immunol., July 15, 2006; 177(2): 1306 - 1313. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
N. C. Riedemann, R.-F. Guo, and P. A. Ward A key role of C5a/C5aR activation for the development of sepsis J. Leukoc. Biol., December 1, 2003; 74(6): 966 - 970. [Abstract] [Full Text]
|
|
|
|
 |
|
 W. C. Aird The Hematologic System as a Marker of Organ Dysfunction in Sepsis Mayo Clin. Proc., July 1, 2003; 78(7): 869 - 881. [Abstract] [PDF]
|
|
|
|
|
|
M. S. Huber-Lang, J. V. Sarma, S. R. McGuire, K. T. Lu, V. A. Padgaonkar, E. M. Younkin, R. F. Guo, C. H. Weber, E. R. Zuiderweg, F. S. Zetoune, et al. Structure-Function Relationships of Human C5a and C5aR J. Immunol., June 15, 2003; 170(12): 6115 - 6124. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
