Hemolytically inactive C5b67 complex: an agonist of polymorphonuclear leukocytes

From www.bloodjournal.org by guest on November 18, 2014. For personal use only.
1995 85: 2570-2578
Hemolytically inactive C5b67 complex: an agonist of
polymorphonuclear leukocytes
C Wang, S Barbashov, RM Jack, T Barrett, PF Weller and A Nicholson-Weller
Updated information and services can be found at:
http://www.bloodjournal.org/content/85/9/2570.full.html
Articles on similar topics can be found in the following Blood collections
Information about reproducing this article in parts or in its entirety may be found online at:
http://www.bloodjournal.org/site/misc/rights.xhtml#repub_requests
Information about ordering reprints may be found online at:
http://www.bloodjournal.org/site/misc/rights.xhtml#reprints
Information about subscriptions and ASH membership may be found online at:
http://www.bloodjournal.org/site/subscriptions/index.xhtml
Blood (print ISSN 0006-4971, online ISSN 1528-0020), is published weekly by the American
Society of Hematology, 2021 L St, NW, Suite 900, Washington DC 20036.
Copyright 2011 by The American Society of Hematology; all rights reserved.
From www.bloodjournal.org by guest on November 18, 2014. For personal use only.
Hemolytically Inactive C5b67 Complex: An Agonist of
Polymorphonuclear Leukocytes
By Ce Wang, Sergei Barbashov, Richard M. Jack, Tonya Barrett, Peter F. Weller, and A n n e Nicholson-Weller
The activity of hemolytically inactive C5b67, designated
iC5b67, was evaluated as an agonist for functional responses
of human polymorphonuclear leukocytes (PMN). C5b67 was
formed from purified human complement components and
decayed in phosphate-buffered saline (PBS) until it had no
lytic activity for sheep erythrocytes in a standard assay.
iC5b67, at nanomolar concentrations, stimulated PMN chemotaxis and Ca2+ fluxes, but inhibited superoxide production and failed to upregulate CR1 and CR3. There was no
significant contamination of the iC5b67 with C5a to explain
these results. Neither isolated C5b6 nor C7 alone exhibited
the activities of iC5b67, while insolubilized anti-C7 could remove the PMNagonist activity from theiC5b67 preparation.
Binding studies to define a specific receptor for iC5b67 on
PMN were hampered by the very hydrophobic nature of the
ligand. '251-iC5b67. by contrast to hemolytically active '251C5b67, was unable to insert in erythrocytes, suggesting that
iC5b67 need not insert in the PMN membrane to induce
signaling. Two lines ofevidencesuggest that iC5b67 and
C5a and FMLP sharecommon steps in intracellular signaling
(1) pretreatment of PMN with iC5b67 deactivates PMN for
C5a- and FMLP-inducedchemotaxis; and (2) pretreatment of
PMN with pertussis toxin inhibits iC5b67-induced chemotaxis. Thus, iC5b67 has important effects on the activity of
PMN and G-proteins and Ca2+are involved in thesignaling.
0 1995 by The American Societyof Hematology.
T
defect. The second reason that studies of the chemotactic
activity of iC5b67 were not pursued related to the discovery
of C5a as a potent chemotaxin.' With the chromatographic
and immunologic reagents available at the time, it would
have been difficult to rule out the possibility of C5a contamination of iCSb67.Withnewer
chromatographic methods,
specific antisera, and better assays, it is now possible to
define the composition of this ligand. Therefore, we reexamined the signaling potential of iC5b67 for human polymorphonuclear leukocytes (PMN). Our findings indicate that
iC5b67 is potent in nanomolar concentrations for inhibiting
superoxide production, while stimulating chemotaxis and
Ca2* fluxes. The ability of iC5b67 to deactivate the chemotactic activity of C5a and FMLP, as well as the demonstration of pertussis toxin (PTX) inhibition of iC5b67 activity,
are consistent with iC5b67 sharing elements of the signaling
pathway used by C5a and FMLP. When complement is activated in the fluid phase, such as would happen in the presence of soluble lipopolysaccharide (LPS) or immune cornplexes, a large fraction of the C5b67 formed may decay to
iC5b67, and thus iC5b67 has the potential to be an important
modulator of PMN activation. However, a ligand with the
potency of iC5b67 might be expected to have a very short
half life in vivo, and iC5b67 binding proteins, such as
vitronectin and clusterin, may exert regulatory roles by inactivating iC5b67.
HE FORMATION OF C5b67 is initiated by the limited
proteolysis of complement C5 to form C5b, which
binds C6 to form a stable C5b6 complex. The addition of
C7 to C5b6 forms the C5b67 complex, which transiently
rearranges to expose the hydrophobic domains necessary for
membrane insertion.'.' There are three possible fates for the
nascent C5b67: (1) insert into membrane lipids and act as a
signaling ligand3.'; (2) insert into membrane lipids and, by
providing binding sites for C8 and C9, become part of a
transmembrane channel (hemolytically active); or (3) decay
to a hemolytically inactive complex in thefluidphase
(iC5b67).5 The iC5b67 complex can react with components
C8 and C9 andwith vitronectin (S-protein) and clusterin,
but no biologic functions have been described for anyof
these nonhemolytic terminal complement complexes.
In retrospect, over 2 decades ago a fluid phase complex
comprised of C5, C6, and C7, likely what we refer to as
iC5b67, was reported to have chemotactic a~tivity.~.'
There
were at least two reasons why this work was not pursued:
first, critics at that time held that fluid phase C5b67 could
not be important as a chemotactic factor if C6 deficient
rabbits had normal chemotactic function.' We now have a
growing appreciation of the redundancy of crucial mediators,
even a deficiency of C5, which precludes the generation of
both C5a and iC5b67, does not lead to a global immune
From theCharles A. Dana Research Institute and Harvard-Thorndike Laboratory of Beth Israel Hospital; Department of Medicine,
Harvard Medical School and Beth Israel Hospital, Boston; and the
Department of Rheumatology and Immunology, Brighum and Women's Hospital andthe Department of Medicine, Harvurti Medical
School, Boston, MA.
Submitted September 8, 1994; accepted December 12, 1994.
Supported by National Institutes of Health Grant MO. HL.33768
to A.M.W., Grant No. AI20241 to P.F. W., and Grant No. AI26292
to R.M.J.
Address reprint requests to Anne Nicholson-Weller, MD, Division
of Infectious Diseases, BethIsrael Hospital, 330 Brookline Ave,
Boston, MA 022 15.
The publication costs of this article were defrayed in part by page
charge payment. This article must therefore be hereby marked
"advertisement" in accorclance with 18 U.S.C. section 1734 solely to
indicate this fact.
0 1995 by The American Society of Hematology.
0ao6-497l/95/8509-0008$3.00/0
2570
MATERIALS AND METHODS
Complement Complexes and Assays
CSb6 was made frompurified CS, C6, cobra venom factor (CVF),
factor B (all from Quidel, San Diego, CA), and recombinant mouse
factor D (adipsin; gift of Dr Ty White, Scios, Mountain View, CA)
C5b6 was fractionwith modifications"'of the original method.'' The
ated by highperformanceliquidchromatography(HPLC)ona
DEAE column (AP-1 Protein Pak 8HR; Waters Associates, Milford,
MA) asdescribed."The
peak CSb6containingfractionswere
pooled, and the pool titered using human erythrocytes. One unit of
C5b6 was defined as the amount of C5b6 required to produce SO%
lysis of 2.5 x IO' human E when incubated in a total volume of
300 pL with C7 (0.1 pg), C8 (0.5 pg), and C9 (0.5 pg) (all from
Quidel)."' To f o m CSb67,30 pg of CSb6andatwofoldmolar
excess of C7(20pg)weremixedandincubatedat
37°C for IS
minutes, then 4°C overnight. Although iC5b67 may aggregate," we
have used the monomeric molecular weight (mw) of 400 kD in the
calculation of molarity.
Blood, Vol 85,No 9 (May l ) , '1995:pp 2570-2578
From www.bloodjournal.org by guest on November 18, 2014. For personal use only.
AGONIST
COMPLEMENT iC5b67: AN
FOR PMN
Antibody Methods
For the iC5b67 enzyme-linked immunosorbent assay (ELISA),
wells (Immulon 2; Dynatech, Chantilly, VA) were coated with polyclonal goat antihuman C7 (U200 dilution; Quidel) overnight at 4°C.
All subsequent steps were performed at m m temperature. Phosphatebuffered saline (PBS) 0.1% Tween was used for all washes and dilutions, except as noted; blocking was performed withSuperblockbuffer
Dilutions of sample, as well as dilutions of an
(Pierce, Rockford, L).
iC5b67 standard (16 to 130 ng), were added to wells for 30 minutes.
Indicator monoclonal antibody (MoAb) anti-C6 (Quidel; 111,OOO dilutionin0.5%TweenPBS)wasadded
for 30 minutes followed by
the sequential additions of the following reagents: biotinylated goatantimouse IgG(H+L) (1/1,ooO,Kirkegaard & Perry, Gaithersburg,
MD) for 30 minutes; streptavidin-horseradish peroxide (111,OOO;
Pierce) for 30 minutes, and finally 100 pL of the substrate solution
(TMB Microwell Peroxidase Substrate System; Kirkegaard & Perry).
After blue color developed, the reaction was terminated by addition
of 1 0 0 pL of 250 mmoVL phosphoric acid. The absorbance unit (AU)
at 450 nm of the wells were read within 1 hour using a Thermomax
Microplate Reader (Molecular Devices, Menlo Park, CA). The assay
was linear over a range of 16 to 130 ng.
IgG fractions of goat antihuman C7 (Quidel) and of goat antihuman factor H (Quidel) were prepared by adsorption to protein G
Superose (HR 10/2; Pharmacia, Piscataway, NJ) and elution with
0.1 molL glycine-HC1 buffer, pH 2.7. The IgG peaks were neutralized, and reabsorbed to protein G agarose beads (P-7700; Sigma, St
Louis, MO) for use in the iC5b67 absorption experiments.
Protein Methods
Protein was assayed using a modified Fohn assay (BCA, enhanced, Pierce), using BSA as a standard. Sodium dodecyl sulfatepolyacrylamide gel electrophoresis (SDS-PAGE) was performed using a 7% to 10% gradient slab gel with discontinuous buffers." To
radiolabel C5b6, approximately 40 pg of C5b6 was mixed with 40
pCi carrier free sodium IZ5I (New England Nuclear, Boston, MA)
in Iodogen (Pierce)-coated tubes according to the manufacturer's
specifications. A PD-10 sizing column (Pharmacia) wasusedto
remove free Iz5I.Ninety-six percent of the cpm of the "'I-C5b6 pool
were precipitable with 10% trichloroacetic acid (TCA), yielding a
specific activity of 5 X lo3 c p d p g . The IZ5I-C5b6 wasfully functional in terms of forming C5b67 and iC5b67.
Preparation of PMN
PMN were isolated at room temperature from 20 mL of venous
blood from normal volunteers. Blood was anticoagulated with 5 mL
acidified citrate, mixed with 5 mL dextran (6% Dextran 70 in 0.9%
NaC1; Kendall McGaw Laboratories, Inc, Irvine, CA) and the E
allowed to sediment for 60 minutes at room temperature. The leukocyte-rich supernatant was under-layered with LO mL Ficoll-Paque
(Phannacia) and centrifuged for 20 minutes at 500g. The PMN
fractions were recovered, and contaminating E lysed with hypotonic
saline. PMN were washed 2X, quantified by hemocytometer, and
resuspended in Hanks' Balanced Salt Solution (HBSS), without Caz+
or Mg2+but with 0.1% bovine serum albumin (HBSA). The isolated
cells were greater than 95% to 98% neutrophils (2% to 5% eosinophils) and greater than 98% viable as assessed by Wright-Giemsa
and trypan blue staining, respectively. PMN were used within 2
hours of venipuncture.
2571
taxis chamber (Neuro Probe, Cabin John, MD). The apparatus was
placed in a 37"C, humidified, 5% CO, incubator for a specified time
to allow migration. The first assay to assess PMN chemotaxis utilized
5 pm pore polyvinylpyrrolidone-free polycarbonate filters (Nucleopore, Neuro Probe #NFB5).I4 Each dilution of chemoattractant was
set up in quadruplicate. After a 60-minute incubation, the filters
were stained with Wright-Giemsa (Diff Quik; Baxter Scientific,
McGaw Park, IL) and the cells on the bottom surface of the filter
were enumerated by an individual blinded to the protocol. For each
well, four to five fields on the filter were analyzed using a 63X
objective. The second chemotactic assay utilized 5 pm pore nitrocellulose mesh filters (Toyo, Neuro Probe #TCB5). Each dilution of
chemoattractant was added to triplicate wells. After a 60-minute
incubation, the filters were fixedand stained with Congo red.''.'6
An Optomax V image analyzing system (Analytical Instruments,
Shaffron Walden, Essex, England) enumerated the cells in four 25X
fields at 20-pm intervals, beginning 20 pm from the upper surface.
The third chemotactic assay measured the ability of PMN that were
loaded with BCECF-AM (Molecular Probes, Eugene, OR) (2 pmol/
L for 20 minutes) to migrate through a 3-pm pore polyvinylpymolidone-free polycarbonate filter (Neuro Probe #PFD3). An incubation
time of 45 minutes was sufficient time to allow cells responding to
chemoattractants to migrate through the filter, and minimized the
chance the PMN would fall off the filter, as detected by the fluorescence in the lower chamber. Fluorescence was quantified with a
fluorescence plate reader (Cytofluor, Millipore, Bedford, MA) using
excitation at 485 nm and emission at 530 nm. The ability of PTX
(Sigma) (2 pg/mL for an hour at 37"C), to inhibit the chemotactic
response was assessed using the first chemotactic assay.
Superoxide Release Assay
Superoxide release was measured as the superoxide dismutase
(SOD)-inhibitable reduction of fenicytochrome c. In the routine
assay, PMN suspensions (8 X lo5 cells/mL) were incubated with
fenicytochrome c (1 mg/mL; Sigma) and either buffer or buffer
plus agonist. Replicate tubes contained SOD (21 pglmL; Sigma) in
addition. In the experiment depicted (see Fig 3), the reaction tubes
were preincubated with a 0.5% gelatin solution, which eliminated
the release of superoxide by resting PMN. After 15 minutes at 37"C,
the reaction was stopped by chilling to 4°C and centrifugation. The
optical density (OD) 550 nm of the supernatants was determined in
a spectrophotometer (DU series 60; Beckman Instruments, Fullerton,
CA). In the microassay, each component was reduced to 114 volume
of the regular assay, and 210 pL of the supernatant was transferred
to a microtiter plate welland the OD 550 nm quantified with a
ThermoMax microtiter plate reader (Molecular Devices, Menlo Park,
CA). All reactions were performed in duplicate or triplicate for the
regular assay and in triplicate for the microassay. Results are expressed asthemean
nanomoles superoxide/number of cells/l5
minutes.17
To test if iC5b67 might itself inhibit the detection of superoxide
anion, superoxide was generated in a cell free system." Purine (10
mmoVL, Sigma), cytochrome c (1 mg/mL, Sigma) and either iC5b67
mom), or buffer, or SOD were mixed, followed by the
(3 X
addition of xanthine oxidase 15 pL (0.05 U; Worthington Biochemical Corp, Freehold, NJ). The reaction mixtures were incubated for
15 minutes at 37°C then immediately put on ice. Cold PBS (750
pL) was added to each tube and the OD 550 nm was determined.
Ca2+Flux Assay
Chemotaxis
Three different chemotactic assay procedures were used. Freshly
isolated PMN (2.5 x IO') were placed in the upper wells and dilutions of chemoattractants (28 @L),or control buffer (HBSS" with
0.1% ovalbumin) were added to the bottom wells of a microchemo-
PMN (1 X 107/mLin HBSA++)were incubated with Indo-l AM
(5 pmoVL; Molecular Probes, Eugene, OR) at 37°C for 7 minutes.
The cell suspension was diluted 5X with HBSA2+,and incubated a
further 10 minutes at 37°C. The Indo-loaded cells were pelleted at
500g for 10 minutes, and resuspended in HBSA at 107/mLat room
From www.bloodjournal.org by guest on November 18, 2014. For personal use only.
WANG ET AL
2572
0
l
I
2
6
4
I
I
I
8
10
I
12
16
I
I
14
Preincubation time (min)
Fig 1. C5b67 decays in the absence of serum proteins. Purified
C5b6 ( l 0 0 pL, 9 ngl was aliquoted, and purified C7 (l00 pL, 6 ng)
was added to the
tubes sequentially at the indicated timesincubafor
tion at 37°C. At zero time, 100 p L human E (2.5 x 10') were then
added t o each tube and the reaction continued for the hemolytic
assay for C5b67, as described in Materials and Methods. The results
were expressed as 2, the average hemolytic sites per cell." A timedependent loss of hemolytic activity was observed. The initial halflife was approximately
30 seconds.This experiment is representative
of t w o performed.
fication and characterization of the components of the complex and the method of forming the complex at that time
differed from our procedures. Because the characterization
of iCSb67 iscentral to the thesis of this work, we reevaluated
the time course of the decay of C5b67 to iCSb67. One hundred microliters of diluted C5b6 (9 ng)wasaliquotedto
tubes and a twofold molar excess of C7 (6 ng) was added
to different tubes at -30, - 15, -5, -2, - 1, and -0.5 minUtes for incubation at 37°C. At zero time, 1 0 0 yL human E
was added to each tube with incubation at 37°C for 20 minutes followed by addition of 0.5 yg C8 and 0.5 yg C9 and
a further incubation at37°C for 60 minutes. The maximal
hemolytic potential of the C5b6 was assessed by adding
C5b6 and C7 to a tube already containing E, and continuation
of the reaction as noted above. All reactions were stopped
by the addition of 2 mL 0.15 molL saline-EDTA to each
tube, centrifugation of the samples, and determination of the
hemoglobin in the supernatant (OD 541 nm). The initial halflife of this second order decay reaction was approximately
30 seconds (Fig l ) . Although this value is longer than the
< 0.1 second half-life previously reported when different
experimental conditions were used,' our results do confirm
1
temperature. One hundred microliters o f the cell suspensionwas
diluted with 400 pL prewarmed HBSA'+. and 50 pL dilutions o f
iC5b67 were added to the cells. The cell fluorescence was analyzed
by flow cytometry (FACStar
Plus; BecktonDickinson,Mountain
View. CA). The ratio o f bound/free Ca" (fluorescence intensity at
405 nmlfluorescence intensity at 485 nm) was determined (Chronys
Software; Becton Dickinson). Ca" ionophore (A23187, 3 pmol/L;
Sigma) and FMLP (10"' mol/L) were used as positive controls;
HBSS buffer was used as a negative control.
Assa.yfor CRI and CR3 Expression
PMN (200 pL at 2 X 10'/mL HBSA) were incubated at 37°C for
30 minutes with iC5b67 (10"' mol/L), buffer alone (negative control), and either C5a (IO-' mol/L) or FMLP ( IO" mol/L), as positive
controls. Fifty microliters o f the above reaction suspensionswas
aliquoted into four tubes, and further manipulations weredoneat
4°C. Each aliquot o f cells was reacted with saturating doses o f one
o f the following MoAb: anti-HLA class I (W6/32).'" anti-CRI(YZl),"' anti-CR3 (OKM-l)?' or an equivalent amount o f an irrelevant
M o A b (TIB157. antihuman I g A chain). After addition o f F I T C goat-antimouse IgG (Jackson Labs, West Grove, PA), the fluorescence intensity o f washed cells was measuredby FACS(BD FacSort;
Becton Dickinson).
Statistical Analysis
Data are presented as mean 5 SEM. The Student's t-test was used
to compare two samples.
RESULTS
Confirmation That C5h67 Decays to a Hemolytically
Inactive Form
Although the decay of C5b67 to a hemolytically inactive
form was previously experimentally documented,' the puri-
Incubationtime (min)
'%
Fig 2. Differential trypsin susceptibility of lnl-iC5b67-E and
C5b67-E. Erythrocyte-complement complexes were formed by the
addition of 1 mL of human E (2.5 x 10'lmL) t o 2 mL buffer (GVB")
containing '"I-C5b6, or preformed '251-iC5b67, or lnl-C5b6 plus C7
t o generate nascent '''l-C5b67. Incubation (37°C x 20 minutes) was
terminated by washing the cells and resuspending them in 3 mL
buffer. Three hundred microliters of
each reaction was aliquoted into
t w o sets of tubes, one set containing 6 pLltube trypsin(6Yo wtlvol)
(-1 and the other set of tubes containinga comparable volume of
buffer ( - 4 . Each reaction was run
in duplicate. After incubation(37°C)
for 0.5,20, 40, and 60 minutes, the tubes were put intoan ice bath
and 10 p L of 9% soybean trypsin inhibitor (wtlvoll was added t o
stop thedigestion. The pellets of E were washed and counted (Cobra
Auto-Gamma; Packard Instrument CO, Meriden, CT). The percentage
of cpm remaining on
E was calculated as (mean cpm of samplehean
cpm of buffer treatedcells at zero time) x 100. Cpm at zero time for
'251-C5b6,preformed '%C5b67, or nascent '251-C5b67were 137,322,
and 521 respectively. Inset: Net percent of cpm removed after a 20minute incubation with trypsin. This experiment was performed six
times with similarresults.
From www.bloodjournal.org by guest on November 18, 2014. For personal use only.
2573
COMPLEMENTiC5b67: AN AGONIST FOR PMN
.-c
E
m
0.6
T-
buffer
C5a
C5a+
FMLP FMLPt
iC5b67 iC5b67
t
0.5
addition of trypsin caused the time-dependent loss of cell
associated cpm that differed for CSb67 and iCSb67. Consistent with its known insertion in membranes after 20 minutes
of trypsin treatment, only 34% of the nonbuffer elutable
c m associated with EC5b67 were susceptible totrypsin
cleavage; whereas for EC5b6, which is known not to insert
v)
2
X
Q!
0.4
v)
a,
3
0.3
2
a,
2
0.2
v)
v)
a,
z
0.1
c
Fig 3. iC5b67 inhibits C5a- and FMLP-induced superoxide production by PMN. Additions of buffer, or iC5b67 (lo-' mol/L) were made
t o PMN (2 x lo5),which weresuspended in ferricytochrome c( l m g l
mL), and the reactants were incubated at 37°C. After 15 minutes,
C5a (lo-'mol/L) or FMLP
mol/L)were added for an additional
incubation of 15 minutes at37°C. Reactions were set up in sextuplicate with half thesamples containing SOD 121 pg/mLI. The reaction
tubes had been precoated with gelatin 10.50101 for 1 hour at room
temperature before the additionof PMN t o decrease contact activation of the cells. A microassay was performedas described in Materials and Methods. The nanomoles of superoxide produced per 2 x lo5
PMN per 15 minutes werecalculated from thedifferences in the OD5=
of the supernatants from samples with SOD and without SOD. The
results are the mean 2 SEM of triplicate values. "Compared with
C5a or FMLP alone,P < .05. This experiment isrepresentative of two
performed.
I
0.2
l
I
l
0.3
0.5
0.4
l
B
Resting
**
that there is a relativelyrapid decay of CSb67 hemolytic
function.
Accessihiliry of Erythrocyte-Bound '251-iC5h67to Trypsin
Although iC5b67 was not active in lysing erythrocytes, it
was possible that it inserted into the plasma membrane. Recently published data indicate that hemolytically competent
terminal complement complexes, including CSb67 and CSb8, as well as C5b-9, can directly interact with G proteins
and thereby mediate signaling by a receptor independent
mechanism: In the early studies of the biology of the terminal complement complexes, evidence that hemolytically
competent CSb67 was physically inserted into the plasma
membrane of target E came from measuring the accessibility
of radioiodinated components of the complex to trypsin digestion.' When EC5b6 was formed using radiolabeled C6,
virtually all the cpm were solubilized when the target cells
were treated with trypsin. In contrast, whenECSb67 was
formed using either radiolabeled C6 or radiolabeled C7, approximately 50% of the original cpm remained cell associated after trypsin treatment. We used a similar experimental
design to test the relative trypsin susceptibility of bound
EiCSb67 and hemolytically competent ECSb67, which were
formed from Iz5I-C5b6and C7 (Fig 2). Only about 10%
of the cpm associated with either erythrocyte-complement
complex eluted in buffer during the l-hour incubation. The
I
0.6
iC5b67
* iC5b67+a C7
m
I
0.2
0.30.5
**
I
0.4
iC5b67+a H
I
I
0.6
nmoles 0, / 15 min / 10 PMN
Fig 4. The specificity of the iC5b67 ligand. (A) iC5b67 activity is
not reproduced by isolatedC5b6 or C7. C5b6 and C7 used for making
iC5b67 were assayed for superoxide production byPMN as described
in Materials and Methods. The reaction tubes were not precoated
with gelatin, which explains the high constitutiverelease of resting
cells. C5b6 (0.18 pg) and C7 (0.8 pgl did not affect the superoxide
production of PMN. The results represent the mean f SEM of triplicate values. (B) lnsolubilized antLC7 IgG removes iC5b67 activity.
Polyclonal anti-C7 IgG and anti-H IgG, as control IgG, were each purified from respective antiserum by adsorption and elution from Protein G. The IgG peaks were separately reabsorbed t o protein Gagarose, and iC5b67 (15pg) was mixed with
each type ofbead overnight
at 4°C. Aliquots from supernatants of the reaction mixtures were
tested in the superoxide assay. The results represent the mean 2
SEM of triplicatevalues. Compared with resting, " P > .05; ""P < .05.
The supernatant from anti C7-beads had no reactivity by ELSAand
failed t o inhibit superoxide production by PMN.
From www.bloodjournal.org by guest on November 18, 2014. For personal use only.
WANG ET AL
2574
in the membrane, 75% of the cpm were trypsin susceptible
(Fig 2 inset). Eighty-one percent of the EiC5b67 associated
cpmweretrypsin
sensitive. indicating thatiCSb67, like
CSb6, was extracellular.
iCSh67 Inhibits CSa- and FMLP-Induced Superoxide
Production
Superoxide production is a well-characterized response of
PMN to agonists such as C5a and FMLP. We useda standard
assay based on the reduction of cytochrome c to test whether
iC5b67 might havean effect on the generation of superoxide.
Unexpectedly, iC5b67 not only failed to stimulate superoxide production (data not shown). it inhibited superoxide production elicited by CSa (IO-’ mol/L) and FMLP (IO-” mol/
L) (Fig 3).
To rule out the possibility that iC5b67 might be inhibiting
the assay for superoxide. superoxide was generated in a cell
free system using purine as the substrate for xanthine oxidase
and cytochrome c as the electronic acceptor and indicator
of color change.’’ When additions of PUS, iC5b67 (3 X IO“)
mol/L), or SOD (21 pg/mL) were made to the reaction, the
mean OD 550 was 0.012, 0.01 1, and 0.0.respectively (n =
5). Thus, the iC5b67 must be inhibiting the ability of PMN
to generate superoxide, because the iCSb67 has no significant
effect on the detection of chemically generated superoxide
by this assay.
Additional experiments were performed to confirm that
the biologic activity we were noting was really caused by
the iC5b67 and was not caused by one of its constituents,
or a contaminant. In these experiments the reaction tubes
were not precoated with gelatin, and there is substantial
superoxide produced by the resting cells, presumably because of contact activation. First, C5b6 (0.18 pg) alone
and C7 (0.8 pg) alone were tested and neither affected the
ability of PMN to produce superoxide (Fig 4A). Second,
the ability of insolubilized anti-C7 to remove the biologic
activity of the iC5b67 was tested. iC5b67 complex was
mixed with anti-C7 IgG- (or, anticomplement factorH
IgG-) protein G-beads for an indicated time. After the
beads were pelleted, the supernatants were assayed for
iC5b67 by ELISA. The supernatant from anti-C7 protein
G-beads were ELISA-negative and failed to inhibit PMN
superoxide production, whereas the supernatantsfrom antiH protein G-beads retained inhibitory biologic activity (Fig
4B).
Chemotaxis and Chemokinesis
To screen for an effect of iC5b67 on PMN mobility, dilutions of iC5b67 (IO”’ to IO” mol/L) were tested for chemotactic activity using a Boyden chamber with a polyvinylpyrrolidone-free polycarbonate filter. There was distinct
chemotactic activity, maximal activity at 10“) mol/L,with
apparent high dose inhibition at IO-’ and IO” mol/L (Fig
6). Effects of iC5b67 on random migration were evaluated.
To perform this analysis. three chambers with nitrocellulose
filters were used:one chamber contained dilutions of iCSb67
in the lower chamber to test chemotaxis; the second contained dilutions of iC5b67 in the upper chamber to assess
chemokinesis; and the third contained dilutions of iCSb67
in both the upper and lower, which is also a test of chemokinesis. Each chamber also contained positive controls for
the migratory capacity of the PMN: PMN in the upper wells,
with FMLP (IO-’ mol/L) in the lower wells. The results
(Table I ) again show a peak of iC5b67 chemotactic activity
at IO-’ mol& and no dose-related effect on chemokinesis.
In fact, when iC5b67 was in the upper wells with the cells,
there was frequently less migration into the filter than when
the cells were in medium alone, which is indicated by negative numbers in Table 1. In this assay, cells were counted
at multiple depths within the filter, which specifically would
allow the identification of distinct subpopulations of migrating cells. The distribution of cells migrating to iC5b67 was
1400
t
1000
Y2
800
600
400
200
0
iCSb67 Complex Does Not Upregulate the Expression of
CRI and CR3
iC5b67 was compared with FMLP and CSa for its ability
to affect the expression of class I MHC, CRI (CD3S), and
CR3 (CD1 1b, CD 18). None of the agonists significantly affected the expression of class I MHC. FMLP and CSa both
upregulated CRI and CR3, as previously noted.”.” By contrast, iC5b67 (1 0”’ mol/L) did not affect the expression of
these complement receptors (Fig 5). In another experiment,
seven twofold dilutions of iC5b67 from 6.4 X IO-’ mol/L
to IO-’ mol/L had no effect on class I, CRI, or CR3 expression (data not shown).
Control W6/32
I!@=
YZ-l
OKM-1
Fig 5. Comparison of the effects of iC5b67and C5a onthe expression of CR1 and CR3 by PMN. PMN (200 pL at 2 x 10’lmLI were
incubated at 37°C for 30 minutes with buffer alone, FMLP (10” moll
L), iC5b67
mollLI, or
C5a (lo-’ mollL), respectively.
Aliquots
(50 pL) were reacted with the following MoAb control MoAb, antiHLA class I (W6/32), anti-CR1 (YZ-l) or anti-CR3 (OKM-l), followed
by FITC-conjugated second antibody, as described in Materials and
Methods. The cells were analyzed by flow cytometry and the mean
fluorescentchannel (MFC) determined. Nanomolar iC5b67did not
affect the expression of either CR1 or CR3, which were shown to be
upregulated by C5a and FMLP. This experiment was representative
of t w o performed.
From www.bloodjournal.org by guest on November 18, 2014. For personal use only.
COMPLEMENTiC5b67: AN AGONIST FOR PMN
2575
100
T
I
Medium
C5a
FM LP
I
0
Log Concentration(M) of iC5b67
Fig 6. iC5b67 stimulates chemotaxis of PMN. Different doses of
iC5b67 (lo”* -10” mol/L, 28 pL) were testedin chemotaxis assays
using polycarbonatefilters, as described in Materials and Methods.
The filter was fixed and stained by
Quick.
Diff The mean PMN number
per field was calculated. iC5b67 significantly stimulated chemotaxis
of PMN and theeffects were dose-dependent. The error bars indicate
standards errors of quadruplicate samples. This experiment isrepresentative of fourperformed.
like that of FMLP and provided no evidence that iC5b67
stimulated only a subpopulation of PMN.
I
I
I
I
I
I
500 o
l
o
0 15002000250030003500
I
4oo0
4 x)
Relative Fluorescence Unit
Fig 7. iC5b67 deactivates C5a and FMLP-induced chemotaxis.
PMN 12.5 x lo5)were preincubatedwith iC5b67 (lo-’ mol/L) or buffer
alone at 37°C for 60 minutes and then added to the upper wells of
Boyden chambers for the assessment of chemotaxis toward either
rC5a (10” mol/L) or FMLP [lo-’ mol/L), which were in the lower
used, and the cells were enumerated
wells. A nitrocellulose filter was
and the migration index calculated as described in Materials and
Methods. Results are expressed as the mean 2 SEM, n = 9. Three
fields were counted
and reactions were setupin triplicate. Preincubation of PMN with iC5b67 significantly diminished their Chemotactic
response t o C5a and FMLP. *Compared with C5a or FMLP alone, P
c .05.
Deactivation of CSa- or FMLP-Induced Chemotaxis by
iCSb67
Both C5a and FMLP are well characterized chemotactic
factors for PMN. Although each has a distinct G protein
Table 1. Effect of iC5b67 on PMN Mobility: Chemotactic and
Chemokinetic Analysis
Net Cell No. at 40 pm Depth.
Chamber No.
Presence of
iC5b67
10”’ mol/L
10”’ mol/L
1OP mol/L
mol/L
1
Top-/Bottom+
42.2 2 12.5
57.7 ? 8.4
72.3 2 10.3
15.9
52.3 -C -34.0
2
Top+/Bonom+
Top+/Bottom-
0.6 -c 7.7
-14.9 -t 10.4
-17.6 -C 9.0
-t
20.9
6.7
-41.8 -C 6.5
2.3 2 10.7
ND
2 6.7
3
Each chemotactic chamber had its own controls to test for PMN
responsiveness. Medium alone was a negative control and FMLP at
10.’ mol/L in the bottom well only was the positive
control. The net
cell number migrating toward FMLP for chambers 1, 2 and 3 were
72.3,58.6, and 65.1, respectively. + with iC5b67added: - without
iC5b67.
Abbreviation: ND, no data.
Net cell number = total cells migrating into the filter at 40 pg
depth - cells migrating to the same depth in response to medium
alone. Results are expressed as the mean, n = 9 fields (three fields
counted from triplicate wells). Negative numbers indicate that there
was more cell mobilityin the medium control well than
in the sample
well.
coupled receptor, exposure to one factor diminishes the response to the second, by a process known as cross-deactivation. Experimental evidence indicates that cross-deactivation
results from the fact that boththe C5a receptor and the FMLP
receptor share the same signaling pathway. To test if iC5b67
might also share the same pathway, BCECF-loaded PMN
that had been pretreated with iC5b67 (IO-’ mol/L at 37°C
X 1 hour), were compared with control cells incubated with
buffer alone, for their migratory response to C5a (IO-’ mol/
L) or FMLP (IO-’ mol/L) in the bottom wells. After a 45minute incubation, the fluorescence of the cells that reached
the bottom of the filter was quantified. The results (Fig 7)
indicate that previous exposure to iC5b67 inhibited the chemotactic response of PMN to both C5a and FMLP.
P7X Inhibition of iCSb67 Stimulated Chemotaxis
The receptors for both C5a and FMLP signal through PTX
inhibitable Gi proteins. We tested the PTX susceptibility of
iC5b67 signaling by preincubating the PMN with the toxin
( 2 ,ug/mL for 1 hour at 37°C). Subsequently, the PMN were
assayed for their chemotactic response to iC5b67 ( 1 O-’ mol/
L) and FMLP (IO-’ mol&). Compared with PMN that had
been preincubated in buffer, PMN that were preincubated
with PTX had a modest inhibition of random motility (Fig
8). FTX completely inhibited the PMN chemotactic response
From www.bloodjournal.org by guest on November 18, 2014. For personal use only.
WANG ET AL
2576
70
60
50
40
30
20
10
0
Fig 8. Effect of PTX on the PMN response t o iC5b67. PMN were
exposed t o PTX (2 pg/mL for1 hour at 37°C). The cells were assayed
for their chemotactic
response t o iC5b67 (lo-* mol/L) andFMLP (lo-*
mol/L) using a polyvinylpyrrolidone-free polycarbonate filter.
The filter was stained and cells
the that had migrated
to the bottom
surface
were enumerated as described in Materials andMethods. Each reaction mixture wasassayed in triplicate. The results are expressed as
the mean f SEM.PTX completely inhibited the PMN chemotactic
response t o iC5b67. Comparing the responses of control cells with
PTX treated cells: *P > .05 and * * P < .05. In two other experiments
PTX treatment inhibited the PMN chemotactic response t o iC5b67
by 50% and 100%.
We performed ligand binding assays using "%iCSb67 to
establish if there were saturable binding, which would provide evidence for a specific receptor on the PMN. Replicate
samples did not yield reproducible binding (data not shown),
a fact we attribute to aggregation of this hydrophobic ligand.
Aggregation of iCSb67 has been noted previously." To determine if iCSb67 were acting on the surface of the cell, the
accessibility of erythrocyte bound "'I-iCSb67 to trypsin was
assessed. Weused erythrocytes for these studies because
they had been used to compare the trypsin accessibility of
ECSb6 and EC5b67.' Additionally, complement-induced release of intracellular proteases is less of a consideration with
erythrocytes thanwith PMN. In thepublished study, 90%
of the cpm associated with bound '"I-CSb6 were released
by trypsin digestion, while only about SO% the cpm associated with bound '251-CSb67 were releasedby trypsin.' These
data were interpreted as evidence that the CSb6 was on the
surface of the cell, while theCSb67wasinserted
in the
plasma membrane and partially shielded from proteolysis by
trypsin. In terms of its susceptibility to trypsin digestion, the
iCSb67 behaved like CSb6 (Fig 2), which is outside the cell.
Thus, our results s*.ronglysuggest that iCSb67 binds to the
cell surface, perhaps to a specific cell receptor. The degree
of proteolysis of CSb67 was comparable with that previously
reported. In retrospect, the reported cpm released by trypsin
from E-'2sI-CSb67,as well as our current result, may overestimate the accessibility of the complex. We now know that
when the CSb67 is formed, some may insert, and the iCSb67
1.6
to iCSb67, and significantly inhibited the PMN response to
FMLP. Thus, iCSb67, like CSa and FMLP mustuse a G
protein for signaling.
iC5b67 Induces the CO'+ Flux of PMN
Ca'+ fluxes are an integral part of many signaling pathways used by the PMN and other cells. Indo-I -loaded PMN
were tested for their ability to generate a Ca'+ flux when
challenged with iCSb67. The Ca" flux was determined by
the ratio of fluorescent intensity (405/48S)as measured by
flow cytometry (Fig 9). Over the 1.4 X 10"' to 2.8 X IO"'
mol/L range tested, iCSb67induced a dose-dependent increase in the Ca'+ flux of PMN.
h
m
CO
2
Ln
0
d
v
2
1.4
v)
C
Q)
c.
c
c
C
Q)
0
v)
g
1.2
3
G
.c
0
DISCUSSION
.-0
c.
(d
We have confirmed that newly formed C5b67 decays rapidly to a hemolytically inactive form, designated iC5b67
(Fig l ) . Using a different experimental method, the half-life
previously reported was calculated to be less than 0.1 second.' Because the formation of CS67 is rate-limiting, experimental conditions that would favor the formation of CSb67
would shorten the apparent half-life of nascent CSb67.I' The
important consideration for the interpretation of our results
is the confirmation that the CSb67 complex we formed using
a twofold molar excess of C7 is labile with respect to its
hemolytic activity. Although the decay rate we measured
was at 37°C and we routinely stored our C5b67 at 4"C, we
confirmed that ligand, which was active for PMN,hadno
hemolytic activity.
U
1.o
10-1
1o4
l og
Doses of iC5b67 (M)
Fig 9. iC5b67 induces a Ca" flux in PMN. PMN (1 x lO'/rnL
HBSA2*) were incubated with Indo-l AM (5 pmol/Ll at 37°C for 7
minutes. The cells were treatedas described in Materials and Methods. PMN suspension (100 pL) wasadded t o prewarmed HBSS" and
50 p L various doses of iC5b67 were added t o PMN. The change of
fluorescent intensity within thePMN of the fluorochrome was monitored for 5 minutes by
FACS. The fluorescent intensity ratio ofCa2+
bound/Caz' free (405/485), which reflects Ca2' flux, was calculated
from the curves. iC5b67 induced a Ca" flux, and the magnitude of
the positive flux was
dose-dependent. The experiment is representative of three performed.
From www.bloodjournal.org by guest on November 18, 2014. For personal use only.
2577
COMPLEMENT iC5b67: A N AGONIST FOR PMN
which is also formed may associate with the cell surface and
be accessible to trypsin degradation.
The iC5b67 ligand exhibited divergent biologic activities.
iC5b67 did not stimulate PMN superoxide production, rather
it inhibited it. This was surprising because C5b-9 has been
reported to stimulate superoxide prod~ction.’~
By contrast,
we found that a preparation of C5b-9 made from purified
C5b6 and terminal components also inhibited superoxide
formation, but the active moiety was iC5b67 (C. Wang and
A. Nicholson-Weller, unpublished results, July 1994). We
propose that the differences in activity may be due to our use
of purified components, compared with the use of terminal
complexes made from deficient sera. Many ligands can lead
to the stimulation of superoxide production, including the
chemotactic factors C5a and FMLP, which act through specific receptors, as
iC5b67 was able to inhibit
C5a- and FMLP-induced superoxide production (Fig 3). Neither C5b6, nor C7 could reproduce the activity of iC5b67
(Fig 4A), while anti-C7 could remove the activity (Fig 4B).
All these data support the conclusion that the active ligand
is iC5b67. Receptor pathways, which inhibit superoxide are
not known, although the intracellular accumulation of CAMP
is known to inhibit PMN superoxide prod~ction.~’
iC5b67,
unlike C5a and FMLP22.23
did not cause the upregulation of
CR1 and CR3 (Fig 5), which is a second example of this
complement complex evoking a different response than C5a
and FMLP.
iC5b67 did stimulate chemotaxis (Fig 6) with maximal
activity at
m o a , and displayed inhibition at higher
doses. An assay to test for directed versus random migration
confirmed that iC5b67 primarily stimulated directed migration (Table l). These results are important for two reasons:
first, they confirm the original reports of iC5b67 chemotactic
activity6,’; second, they indicate that the ability of iC5b67
to inhibit superoxide production and its failure to stimulate
complement receptor upregulation are not the result of a
general suppression of all cell functions. The ability of
iC5b67 to inhibit C5a and FMLP directed chemotaxis (Fig
7) may be via activation of the same pathway that inhibited
C5a and FMLP-induced superoxide production, or it may
reflect cross-deactivation. Cross-deactivation occurs when
the same signaling machinery is used by two receptors, and
the first receptor activated preempts the ability of the second
ligand to signal, as reviewed.26The FTX susceptibility of
iC5b67 induced chemotaxis (Fig 8) indicates that this ligand
uses G, proteins for signaling. However, the finding of
iC5b67 chemotactic activity in vitro may not mean that this
is the principal function of this ligand in vivo. For example,
c-kit is chemotactic for mast cells, but its primary function
is apparently as a growtwdifferentiation factor.28
Indo- 1-loaded PMN showed a iC5b67 dose-dependent
Ca2+flux (Fig 9). Ca2+fluxes are common to many signaling
pathway^,'^ and further studies will be necessary to define if
the Ca2+is coming from intracellular or extracellular stores.
We have shown in these studies that iC5b67 is active in
signaling PMN. Further studies will be necessary to define
if iC5b67 occurs in vivo, andif the binding of vitronectin (Sprotein) or the later complement components affects iC5b67
signaling. The fact that iC5b67 is potent at nanomolar concentrations suggests that it, like other potent complement
ligands including C3a and C5a, may have a short half-life.
To date, active terminal complement complexes have included membrane inserted C5b67, C5b-8, and C5b-9 complexes. Now we have also reconfirmed that iCSb67 has biologic activity, and that unlike the other terminal complement
complexes, it apparently signals from outside the cell, rather
than by insertion through the bilayer. We do not know if it
is acting through a specific receptor. It is interesting that the
changes in light polarization could be inhibited by PTX,
making iC5b67 similar to signaling by C5a and FMLP, yet in
the assays of superoxide and CRlKR3 upregulation, iC5b67
evoked different responses. Preliminary data indicate that
multiple signaling pathways are triggered by this ligand (C.
Wang and A. Nicholson-Weller, manuscript in preparation).
Finally, the fact that iC5b67 is large, hydrophobic, and has
some proinflammatory andsome antiinflammatory activities,
while C5a is small, hydrophilic, and is only proinflammatory,
suggests that these two ligands have differing roles in the
initiation and regulation of inflammation.
REFERENCES
1. Gotze 0, Muller-Eberhard HJ: Lysis of erythrocytes by complement in the absence of antibody. J Exp Med 132:898, 1970
2. Hammer CH, Nicholson A, Mayer MM: On the mechanism of
cytolysis by complement: Evidence on the insertion of the C5b and
C7 subunits of the C5b,6,7 complex into the phospholipid bilayer
of the erythrocyte membrane. Proc NatlAcad Sci USA 725076,
1975
3. Carney DF, Lang TJ, Shin ML: Multiple signal messengers
generated by terminal complement complexes and their role in terminal complement complex elimination. J Immunol 145:623, 1990
4. Niculescu F, Rus H, Shin ML: Receptor-independent activation
of guanine nucleotide-binding regulatory proteins by terminal complement complexes. J Biol Chem 2694417, 1994
5. Nicholson-Weller A, Halperin JA: Membrane signaling by
complement C5b-9, the membrane attack complex. Immunol Res
12:244, 1993
6. Ward PA, Cocbrane CG, Muller-Eberhard HJ: Further studies
on the chemotactic factor of complement and its formation in vivo.
Immunology 11 :141, 1966
7. Lacbmann PJ, Kay AB, Thompson RA: The chemotactic activity for neutrophil and eosinophil leucocytes of the bimolecular complex of the fifth, sixth and seventh components of human complement (C567) prepared in free solution by the ‘reactive lysis’
procedure. Immunology 19395, 1970
8. Stecher VJ, Sorkin E: Studies on chemotaxis. XII. Generation
of chemotactic activity for polymorphonuclear leucocytes in sera
with complement deficiencies. Immunology 16:231, 1969
9. Shin HS, Snyderman R, Friedman E, Mellors A, Mayer MM:
Chemotactic and anaphylatoxic fragment cleaved from the fifthcomponent of guinea pig complement. Science 162:361, 1968
10. Halperin JA, Taratuska A, Rynkiewicz M, Nicholson-Weller
A: Transient changes in erythrocyte membrane permeability are induced by sublytic amounts of the complement membrane attack
complex (C5b-9). Blood 81:200, 1993
1 I . DiScipio RG, Smith CA, Muller-Eberhard HJ, Hugh T: The
activation of human complement component C5 by a fluid phase C5
convertase. J Biol Chem 258:10629, 1983
12. Podack ER, Biesecker G , Kolb W, Muller-Eberhard W :The
C5b-6 complex: Reaction with C7, C8, C9. J Immunol 121:484,
1978
13. Laemmli UK: Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature 227:680, 1970
14. Harvath L, Falk W, Leonard EJ: Rapid quantitation of neutro-
From www.bloodjournal.org by guest on November 18, 2014. For personal use only.
2578
phi1 chemotaxis: Use of a polyvinylpyrrolidone-free polycarbonate
membrane in a multiwell assembly. J Immunol Methods 37:39, 1980
15. McCrone EL, Lucey DR, Weller PF: Fluorescent staining for
leukocyte chemotaxis. Eosinophil-specific fluorescence with aniline
blue. J Immunol Methods 114:79, 1988
16. Rand TH, Cruikshank WW, Center DM, Weller PF: CD4mediated stimulation of human eosinophils: Lymphocyte chemoattractant factor and other CD4-binding ligands elicit eosinophil migration. J Exp Med 173:1521, 1991
17. Metcalf JA, Gallin J1, Nauseef WM, Root RK: Laboratory
Manual of Neutrophil Function. New York, NY, Raven, 1986 p 109
18. Parks DA, Granger DN: Xanthine oxidase: biochemistry, distribution and physiology. Acta Physiol Scand 548s:87, 1986
19. Barnstable CJ, Bodmer WF, Brown G, Galfre G, Milstein C,
Williams AF, Ziegler A: Production of monoclonal antibodies to
group A erythrocytes, HLA and other human cell surface antigens:
New tools for genetic analysis. Cell 14:9, 1979
20. Changelian PS, Jack RM, Collins LA, Fearon DT: PMA induces the ligand-independent internalization of CRl on human neutrophils. J Immunol 134:1851, 1985
2 I . Wright SD, Rao PE, Van Voorhis WC, Craigmyle LS, Iida
K, Talle A, Westberg EF, Goldstein G, Silverstein SC: Identification
of the C3bi receptor on human monocytes and macrophages by using
monoclonal antibodies. Proc Natl Acad Sci USA 805699, 1983
22. Fearon DT, Collins LA: Increased expression of C3b receptors on polymorphonuclear leukocytes by chemotactic factors and
by purification procedures. J Immunol 130:370, 1983
WANG ET AL
23. Berger M, O’Shea J, Cross AS, Folks TM, Chused TM,
Brown EJ, Frank MM:Human neutrophils increase expression of
C3bi as well asC3b receptors upon activation. J Clin Invest 74:1.566.
1984
24. Hallett MB, Luizo JP, Campbell AK: Stimulation of Ca+2dependent chemiluminescence in rat polymorphonuclear leukocytes
by polystyrene beads and the non-lytic action of complement. Immunology 44:569, 1981
25. Gerard C, Gerard NP: C5a anaphylatoxin and its seven trdnsmembrane-segment receptor. AnnRev Immunol 12:775, 1994
26. Snyderman R, Uhing RJ:Chemoattractant Stimulus-Response
Coupling, in Gallin J1, Goldstein I, Snyderman R (eds): Infammation, Basic Principles and Clinical Correlates (ed 2). New York, NY,
Raven, 1992, p 42 I
27. Nielson CP, Bayer C, Hodson S, Hadjokas N: Regulation
of the respiratory burst by cyclic 3’,5’-AMP, an association with
inhibition of arachidonic acid release. J lmmunol 149:4036,
1992
28. Meninger CJ, Yano H, Rottapel R, Bernstein A, Zsebo KM.
Zetter BR: The c-kit receptor ligand functions as a mast cell chemoattractant. Blood 79:958, 1992
29. Berridge MJ: Inositol triphosphate and calcium signalling.
Nature 361:315, 1993
30. Mayer MM: Complement and Complement Fixation, in Kabat
E, MayerMM: Immunochemistry. Springfield, IL, Thomas, 1961,
p 13.5