1. Ingeniería

From www.bloodjournal.org by guest on February 6, 2015. For personal use only.
Purification and Characterization of a C5a-Inactivating Enzyme From
Human Peritoneal Fluid
By Suhail K. Ayesh, Yehudith Azar, lyad I. Barghouti, Julie M. Ruedi, Bernard M. Babior, and Yaacov Matzner
than the crude material. On Westernblot, an inhibitory antiEarlier work has suggested that familial Mediterranean feat thesame mover, aninherited disorder characterized by sporadic episodes body recognized a single antigenic species
lecular weight. The enzyme had no activity against denaof inflammation involvingthe pleural and peritoneal cavities
tured bovineserumalbumin.
With recombinant C5aas
and the joints, is caused by the lackof a C5a inactivator
substrate, the K,, and V,,, were 3.4 pmol/L and 52 nmol C5al
normally found in serosal fluid.We have purified this inactimin/mg protein, respectively.
vator from ascites fluid and obtained a protein of molecular
0 1995 by The American Societyof Hematology.
weight 53 to 56 kD with a specificactivity 10,000-fold greater
Ascites fluid was heated at 56°C to inactivate complement;
EUTROPHILS ARE initially attracted to a site of in(NH&S04 precipitates were dissolved in 10 mmoUL Tris . HC1
flammation by chemotactic factors, released firstby the
causative agent (eg, invading microorganisms) and then by the(pH 7.4) at an Azsoof 6.6 and then desalted by dialysis against
neutrophils themselves. A major chemotactic factor produced the same buffer. The C5a-inactivating enzyme in the samples was
determined by measuring the chemotaxis of fresh human neutrophils
during inflammation is the complement fragmentC5a, which
as previously de~cribed,~
using a mixture of 10% (voYvol) sample
is released not only through complement activation but
also
and either 1% (voVvol) zymosan-activated serum or 1 nrnoVL rC5a
through the actionof a protease released from neutrophil spe- as chemoattractant. Random migration was determined by measuring
cific granules.’.’ C5a-md1ated chemotaxis is thus a self-ammigration toward phosphate-buffered saline (PBS)/0.6% BSA/O.l%
plifying process in which low concentrations of C5a attract
glucose. Chemotaxis was calculated by subtracting random migraneutrophils to sites of inflammation, where higher local C5a
tion from the distance traveled toward rC5a.
concentrationsprovoke the releaseof the CS-splittingenzyme
Myeloperoxidase release. C5a-induced myeloperoxidase release
from neutrophils’ was used tolocate the C5a-inactivating enzyme in
from the neutrophil-specific granules, resultingin the produccolumn fractions. Fifty microliters of 10 nmoUL rC5a in suspending
tion of more C5a that attracts and activates more neutrophils.
buffer (Hank’s Balanced Salt Solution [HBSS] containing 25 mmoU
Because of self-amplification, the accidental release of even
L HEPES [pH 7.41 and 0.25% BSA) and an equal volume of the
minimal amounts of C5a could cause an inappropriate fullfraction to be assayed were loaded in triplicate into wells in a 96scale inflammatoryresponse,unless
some countervailing
well microtiter plate. The plate also contained a negative control
mechanism existed to prevent such an occurrence.
consisting of 50 p L of suspending buffer plus 50 pL of column
In seeking such a countervailing mechanism, we discoveluting buffer and a positive control containing 5 nmolL rC5a in
ered a diisopropyl fluorophosphate (Dm)-inhibitable enthe same buffer mixture used for the negative control. The plate was
zyme in serosal fluids that neutralizes C5a, possibly by limincubated at 37°C for the indicated period of time (usually 30 minited
The activity of this C5a inactivating
utes). Twenty-five microliters of neutrophil suspension (4 X 10’
enzyme was greatly reduced in serosal fluids from patients
celldmL in suspending buffer) that had been incubated for 10 minwith familial Mediterranean fever (FMF), an inherited disutes at37°Cwith
5 pg/mL cytochalasin B (added from a stock
ease characterized by episodes of unprovoked inflammation
solution containing 1 mg cytochalasin B/mL in dimethylsulfoxide)
were added to each well, and degranulation was allowed to proceed
involving serosal spaces.6-sThese findings suggested that the
for 10 minutes at 37°C. Myeloperoxidase release was then measured
C5a-inactivating enzyme functions to prevent inappropriate
by adding to each well 50 pL sodium phosphate buffer, pH 6.8,
inflammation in serosal tissues and that its deficiency may
followed by 25 pL of o-phenylenediamine/H,Ozsolution (see below)
explain the attacks of inflammation characteristic of FMF.
and incubating for 5 to 20 minutes at room temperature, determining
In this report, we describe the purification of this enzyme
the incubation time by watching the color development in the posifrom ascites fluid.
N
tive control well. When the color in the positive control well had
MATERIALS AND METHODS
Materials
Recombinant C5a (rCSa), cytochalasin B, o-phenylenediamine,
goat antirabbit IgG conjugated to alkaline phosphatase, p-nitroblue
tetrazolium . HCI, 5-bromo-4-chloro-3-indolyl phosphate, benzamidine, Blue Sepharose CL-6B, and arginine agarose were purchased
from Sigma (St Louis, MO). The rC5a was dissolved in distilled
water containing bovine serum albumin (BSA) at 2.5 mg/mL.
[‘Z51]albumin
(2.7 pCi/pg) was obtained from Amersham (Arlington
Heights, IL). Freund’s adjuvants were from Difco Lab (Detroit, MI).
Diethyl aminoethyl (DEAE) cellulose (DE-52) was obtained from
Whatman Biosystems (Maidstone, Kent, UK), Sephadex (3-10 and
G-l00 from Pharmacia (Uppsala, Sweden), and nitrocellulose membranes from Schleicher and Schuell (Germany). All other chemicals
were of reagent grade and were purchased from Sigma.
Methods
Assays of the CSa Inactivating Enzyme
Neutrophil chemotaxis. Chemotaxis was used to measure the
C5a-inactivating enzyme in ascites fluid and (NH4)2S04precipitates.
Blood, Vol 85,No 12 (June 15).
1995:
pp 3503-3509
From the Hematology Unit, Hadassah University Hospital, Mount
Scopus, Jerusalem, Israel; and the Department of Molecular and
Experimental Medicine, The Scripps Research Institute, La Jolla,
CA.
Submitted November 8, 1994; accepted January 25, 1995.
Supported in part by Grant No. 89-00217from the United StatesIsrael Binational Science Foundation, US Public Health Services
Grant No. AI-24227, a grant from the Kovshar Foundation, and a
Wolfson Research Award administered by the Israel Academy of
Sciences and Humanities.
Address reprint requests to Yaacov Matzner, MD, Hematology
Unit, Hadassah University Hospital, Mount Scopus, Jerusalem, Israel 91240.
The publication costs of this article were defrayed in part by page
charge payment. This article must therefore be hereby marked
“advertisement” in accordance with 18 U.S.C. section 1734 sole1.y to
indicate this fact.
0 1995 by The American Society of Hematology.
0006-4971/95/8512-0029$3.00/0
3503
From www.bloodjournal.org by guest on February 6, 2015. For personal use only.
3504
AYESH ET AL
..................................................
the marker were pooled and used as the
proteolysis substrate. To
measure proteolysis, inactivator at the concentrations indicated was
added to 3.5 nmol of protein (C5a or denatured ['2sI]albumin)in 0.5
mL of PBS. Duplicate incubations were performed for 15 minutes
at 37°C. Loss of C5a activity was measured by the myeloperoxidase
assay. Proteolysis of albumin was measured by sodium dodecyl
sulfate-polyacrylamide gel electrophoresis (SDS-PAGE)/autoradiography and by the release of trichloroacetic acid-soluble counts.
SDS-PAGE was performed on 20-pL portions of the reaction mixtures using a 10% gel, and radioactive bands were detected by
exposing the gel to Agfa XP 18/24 film (Curix, Mortsel. Belgium).
C5a only
lnactkvator only
~.~~*~~~~~.~.~...~....~.......~.~..~......~*..~..~.~-...~~.~..~~.
&.I.......
0
4
8
12
16
T~me(mm)
Fig1. Spectrophotometric determination ofC5a inactivation by
the purified C5a inactivator. Incubations were performed as described in Materials and Methods, with omissions asnoted. The concentrations of C5a and inactivator used in these experiments were
0.72 pmoilL and 3.5 pglmL (0.066 pmollL. assuming a molecular
weight of 53 kD [see text]), respectively. Absorbances at zero time
were set arbitrarily; their absolute values were 0.337 for C5a only,
0.137 for inactivator only, and 0.148 for both.
Preparation of Polyclonal Antibody Against the C5a
Inhibitor
Rabbits were immunized with 100 to 200 pg ofpurified C5a
inactivator in 1 mL of complete Freund's adjuvant (0.25 mL in each
footpad). The animals were boosted 2 weeks later byan identical
procedure and then three more times by subcutaneous injections of
100 to 200 pg of C5a inactivator in 1 mL of incomplete Freund's
adjuvant. Serum was collected 3 weeks after the last injection, decomplemented by heating at 56°C for 30 minutes, and then assayed
for activity against the C5a inactivator using the chemotaxis and the
myeloperoxidase methods.
Gel Electrophoresis and Western Blotting
reached a satisfactory intensity, the reactions were stopped with 3
N HCl and A,,, was read in a microtiter plate reader.
In conducting this assay, particular attention hadtobepaid
to
two points. First, neutrophil quality had to be optimum; the cells
had to beas fresh as possible and completely free of clumps. Second,
the o-phenylenediamine/H,02 solution had to be prepared just before
use by dissolving a 10-mg o-phenylenediamine pill (Sigma) in 2.5
mL citrate-phosphate buffer (25.7 mL 0.2 mom NaZHPO, + 24.3
mL 0.1 m o m citric acid + 50 mL distilled water) and then adding
to this solution 7.5 pL 30% HzOz (Sigma). Any delay between the
preparation of the o-diphenyleneamine/H20z solution and its use
resulted in an unsatisfactory assay. Results were corrected for myeloperoxidase release in the absence of rC5a.
Spectrophotometry. The inactivation of C5a by the C5a inactivator could be followed spectrophotometrically at 254 nm. Reaction
mixtures containing various concentrations of rC5a together with
0.2-mL portions of purified inactivator in 1 mL of 10 mmol/L Tris *
HC1,pH 7.4, were incubated at room temperature in a recording
spectrophotometer. Rates were calculated from the differences in
absorbance at 1 and 5 minutes. Figure 1 shows that a loss in absorbance of 0.01 U corresponds to the inactivation of 0.72 nmol
C5dmL reaction volume. Loss of absorbance was not observed in
incubations containing C5a alone, purified C5a inactivator alone,
or a mixture of C5a and the inactivator together with 1 mmom
phenylmethylsulfonyl fluoride (PMSF; PMSF experiment not
shown). The spectrophotometric results were confirmed by chemotaxis measurements, which showed that no chemotactic activity remained after 15 minutes of incubation.
Nonspecific Proteolysis
Nonspecific proteolysis was determined by comparing C5a inactivation with the hydrolysis of denatured ['LSI]albuminunder identical
experimental conditions. For denaturation, 25 pL of ['2sI]albumin
(0.4 pg/pL) was incubated with 200 pL of 4 m o m guanidine .
HCV50mmoVL dithiothreitol for 5 minutes. Cyanocobalamin was
added as marker and the mixture was desalted over a 10-mL Sephadex G-25 column equilibrated with PBS, eluting with PBS, and
collecting 0.5-mL fractions. "'I-containing fractions eluting before
SDS-PAGE was performed by the method of Laemmli"' using
12% running gels. The gels were stained with either Coomassie Blue
or silver stain. Nondenaturing PAGE was performed under the same
conditions except for the omission of SDS and dithiothreitol. For
Western blotting, the proteins were transferred electrophoretically
to a nitrocellulose membrane that was blocked for 2 hours at room
temperature with PBS/5% BSA andthen incubated for 1 hourat
room temperature with a 1:lOO dilution of the rabbit antiserum.
Bands were visualized with goat antirabbit IgG conjugated to alkaline phosphatase followed by treatment with a solution containing
p-nitroblue tetrazolium HCl (0.33 mg/mL) and 5-bromo-4-chloro3-indolyl phosphate (0.165 mg/mL)."
RESULTS
Purification of the C5a-Inactivating Enzyme
Some ascites fluids from patients with benign hepatic cirrhosis were found to contain the C5a inactivator at the concentration previously found in normal peritoneal fl~ids.~.'.'
Because ascites can be obtained in liter amounts,
we used
this fluid as the source of the inactivator for purposes of
purification.
Ascites was obtainedfrom patients with alcoholicor posthepatic cirrhosis of the liver and assayed to make sure the
inactivator was present in the material (the inactivator tended
to be absent from ascites obtained from patients with hepatitis B or malignant ascites). Ascites that contained the inactivatorwascentrifuged
(1,500g for 10 minutes at4°C)to
remove cells and debris and then divided into50-mL aliquots
andstoredat
-20°C untiluse. The purificationwasperformed at 4°C.
The C5a inactivator was precipitated from50 mL of peritonealfluidbybringingthefluidto
35% saturationwith
(NH4)?S04. After 4 hours,themixturewascentrifuged
(15,OOOg for 20 minutes) and the supernatant was discarded.
The pellet was suspended in 5 mL of 0.01 m o m Tris-HC1
(pH 7.4) and desalted over a Sephadex G-l0 column (2.4 X
From www.bloodjournal.org by guest on February 6, 2015. For personal use only.
3505
C~A-INACTIVATINGENZYME
Table 1. Purification of the Serosal Fluid C5a lnactivator
Total Protein
7
5
9
l1
13
15
l?
Peritoneal fluid
Ammonium sulfate
precipitation
2300
100 0.00003
800
110
0.00009
104
3
18 0.00054
86
88
67
37
333
7
9,990
IS
2.50
100
2.00
80
1S O
60
1.oo
40
0.50
20
5
10
15
25
20
30
35
0
40
Fraction Number
0.04
0.03
0.02
0.01
0.00
2
3
4
5
6
7
8
55
4.6
0.016
0.3
(fold)
1
The experiments were performed as described in the text. All assay
mixtures contained 3.3 nmol/LC5a except the assay mixture used
with crude peritoneal fluid, which contained 1 nmol/L C5a. Because
these assays were performed using C5a concentrations far below the
K, of the enzyme, the specific activities are expressed as nmol C5a/
min/mg inactivator/nmol/L C5a.
Fraction number
0.00
Purification
(mg)
Affinity
chromatography
3
Yield
(96)
Purification Step
DE-52
Blue Sepharose
0.0011
0.01
Sephadex G-100
1
Specific
Activity (nrnol
CBa/rnin/rng
inactivator/
nrnol/L C5a)
9 1 0 1 1 1 2
Fraction Number
Fig 2. Column chromatography ofthe C5a inactivator. This figure
shows the results from a representative purification. C5a
activity was
measured as inhibition of rC5a-induced release of myeloperoxidase
from neutrophils as describedin Materials and Methods. IO) Inhibition of myeloperoxidase release;(01, protein concentration l&m). (A)
DEAE-wllulose. & in the control incubationwas 0.92. Fractions 14
through 18 were pooled for further purification. (1) Start of elution
with 0.25 mollL NaCI. (B) Sephadex 6-100. & in the control incubation was 0.65. Fractions26 through 32 were pooled for further purification. IC) L-arginine agarose.& in the control incubationwas 0.65.
Fractions 8 through 11 were pooled and concentrated for further
analysis.
25 cm) equilibrated with the same buffer. Fractions containing the C5a inactivator were then pooled and applied to
a DEAE-cellulose column (DE-52, 3 X 6 cm) equilibrated
with 0.01 m o m Tris-HC1 (pH 7.4). The active material was
eluted with 0.25 m o m NaCl in the same buffer, collecting
5-mL fractions (Fig 2A). Fractions containing inhibitory activity were pooled, desalted over Sephadex G-10, adjusted
to a final buffer composition of 0.05 m o m Tris HCl (pH
7.0)/0.1 m o m KCl, and applied to a Blue Sepharose CL6B column (1.6 X 10 cm) equilibrated with the same buffer,
eluting with the same buffer at 0.7 ml/min. The C5a inactivator eluted with the pass-through. Fractions containing the
C5a inactivator were pooled, concentrated to 4 mL by ultrafiltration with a PM-l0 membrane (Amicon Corp, Lexington,
MA), and applied to a Sephadex G-l00 column (1.5 X 48
cm) equilibrated with 0.04 m o m Tris * HCl (pH 7.4). The
column was eluted at 0.3 mL/minwiththe
same buffer,
collecting 3-mL fractions (Fig 2B). The active fractions*
were analyzed by SDS-PAGE, and those with the fewest
protein bands were pooled. The pooled active fractions were
applied to an L-arginine agarose affinity column (1.2 X 22
cm) equilibrated with 0.04 m o m Tris HCl (pH 8.5). The
inhibitor activity was eluted at 0.45 mL/min with the same
buffer containing 0.2 m o m NaCl, and the fractions (3 mL
each) with the highest specific activity were pooled (Fig
2C). For SDS-PAGE analysis, the purified inactivator was
concentrated using a Speedvac concentrator (SVC 100H;
Savant, Fanningdale, NY). The purification is summarized
in Table 1.
* C5a activity was also inhibited by certain high molecular weight
fractions, suggesting either that some of the CSa-inactivating protease was associated with one or more high molecular weight proteins
or that the preparation applied to the G-l00 column contained one
or more additional C5a antagonists. Selection of the low molecular
weight fractions from the G100 column was based on earlier gel
filtration studies showing that themolecular weight of the C5a inactivator of interest was 25 to 40 k D . 3 , 4
From www.bloodjournal.org by guest on February 6, 2015. For personal use only.
AYESH ET AL
3506
B
A
A
..-.,
B
CY.
I.
y"
T"
67
43
I
Arginine Column
Gel Filtration
Fig 3. SDS-PAGE of pooled Sephadex G-l00 fractions that were
applied to theL-arginine agarose column (left)and pooled fractions
from the L-arginine agarose column (the purified enzyme; right).
Samples from thepooled fractions weresubjected t o SDS-PAGE as
described in Materials andMethods. Lane A, stained with Coomassie
Blue; lane B, immunoblot using anti-C5a inactivator antiserum (see
Materials and Methods). BSA (molecular weight, 67 kD) and ovalbumin (molecular weight, 43 kD) were used as standards.
Characterization of the C5a lnactivator
About 3 pg of pure CSa inactivator were obtained from
SO mL of ascites. SDS-PAGE of this materialshowed a
single major band at a molecular weight of S3 to 56 kD (Fig
3 right, lane A). The affinity-purifiedenzyme was essentially
free of the impurities present in the material applied to the
affinity column (Fig 3 left, lane A). On immunoblots of both
the preaffinity andpostaffinity column material, an inhibitory
antibody raised to the purified CSa inactivator (see below)
recognized the inactivator as a 53- to 56-kD antigen (Fig 3,
lanes B).
In earlier work, the CSa inactivator was found toinactivate
CSabutnot the chemotactic peptide F-met-leu-phe.' Later
work showed that CSa inactivated by partially purified CSa
inactivator ranon SDS-PAGE slightly morerapidlythan
C%.' Figure 4 shows that the same product was generated
by the pure enzyme.' In addition, pure CSa inactivator, like
the partially purified inactivator.' was inactivated by PMSF
and inhibited by benzamidine, but was only slightly affected
by pepstatin (Table 2). Taken together, these findings support
the ideathat the CSa inactivator is a serine protease that
performs limited proteolysis on a well-defined group of substrates, although the possibility that it inactivates chemotactic peptides by catalyzing some other reaction involving an
amide bond (eg, hydrolysis of a glutamine or asparagine
to the corresponding acid or rearrangement of the peptide
backbone) cannot be ruled out. To date, the group of substrates for the inactivator has been found to consist of two
chemotactic peptides: CSa, released by complement activation. and interleukin-8,'' a polypeptide produced by cells in
response to inflammatory cytokines."."
The specificity of the CSa inactivator was further examined usingdenatured ['"I]albumin. Whereas trypsin wasable
to degrade the labelled albumin to peptides (Fig S ; the aggregated material at the top of the gel was particularly susceptible totryptic digestion), the CSa inactivator could do neither,
even though it completely inactivated a similar (in molar
terms) quantity of CSa under the experimental conditions
(Lw
in the myeloperoxidase assay was 0.324 for the starting
material, 0.106 for the inactivated CSa, and 0.138 for the
CSa-free blank). The results therefore confirm that the CSainactivating enzyme is not a general protease.
An antiserum wasraised in rabbits against the purified
CSa inactivator. This antiserum recognized the 53- to 56-kD
.
m
Fig 4. Change in the electrophoretic mobility of rC5aafter
treatment with C5a inactivator.
Incubationmixturescontained
1.25 p g rC5a in 50 p L of 10
mmol/L Tris HCI, pH 7.4, with1 p g of purified
out (left) or with
C5a inactivator (right).Reactions
were performed for 4 hours at
37°C. Samples werethen analyzed by SDS-PAGE on a 20%
gel, visualizing with silver stain.
The arrowhead indicates the position of the14.5-kD marker.
.
From www.bloodjournal.org by guest on February 6, 2015. For personal use only.
C5A-INACTIVATINGENZYME
3507
Table 2. Effect of Protease Inhibitors on theC5a lnactivator
Concentration
Protease
-2.7
Serine2.0
Aspartyl
Inhibitor
lmmol/L)
PMSF
Benzamidine
Pepstatin
89.5
Residual Activity of
C5a Inactivator (%)
5 1.2
0.8 t 13.8
t 6.1
1.o
0.05
Protease inhibitors at the concentrations indicated were added to
30-pL portions of purifiedC5a inactivator (0.66 nmol/L). The mixtures
were then incubated with 1 nmol/L rC5a (final volume, 300 pL) for
10 minutes at 37°C. Chemotaxis was then measured as described in
Materials and Methods. At the concentrations used, the inhibitorshad
no effect on chemotaxis. Other protease inhibitors (EDTA, dimercaptopropanol, N-ethylmaleimide, and N-tosyl-L-phenylalanine chloromethyl ketone) could notbe examined because of interference with
chemotaxis. Chemotaxis was calculated as the distance travelled in
response to rC5a minus random migration of48.0 2 3.2 pm. Chemotaxis in response to rC5a in the absence of the C5a inactivator was
39.1 t 0.7 pm. Results are expressed as the mean t SE of three
experiments.
m
0
'ij
In
40
.->
%
L
c
20
0
750
250
500
1000
Antibody dilution
Fig 6. Neutralization of C5a inactivator activity by an antiserum
raised against the inactivator.The antiserum wasprepared and heatinactivated as described in Materials andMethods. Its abilityt o neutralize the C5a inactivator was evaluated as follows. Twenty-fivemicroliter portions of heat-inactivated ascites fluid were incubated
for 30 minutes with25 pL of heat-inactivated anti-inactivatorantiserum or normal rabbitserum at the dilutionsindicated. The mixtures
were then incubated with 5 nmol/L rC5a, and myeloperoxidase renonimlease was measured as described. ( 0 ) Immune serum; (0)
mune serum.
band on a Western blot of partially purified C5a inactivator
(Fig 3). whereas normalrabbit serum recognized nothing.
The antiserum did not recognize the protein on a Western
blot of unfractionated peritoneal fluid, presumably because
the amount wastoolow to be detected. However, it did
inhibit C5a inactivation by such fluids, whereas normal rabbit serum had little effect (Fig 6). Treatmentof normal peritoneal fluids with the antiserum reduced C5a inactivator activity to the levels found in peritoneal fluids from patients with
FMF (Table 3).
Table 3. Inhibition of rC5a-Induced Neutrophil Chemotaxis
by Various Peritoneal Fluids
0
172 86 43
pmol enzyme
66
21
Fig 5. Digestion of denatured['"llalbumin by theC5a inactivator
and by trypsin. The experiment was conducted as described in the
text. The identities and concentrations of the enzymes used in the
various reaction mixtures are indicated in the figure.
Inhibition of
Chemotaxis 1%)
Peritoneal Fluid
N
None
Normal, untreated
Normal, nonimmune
serum-treated
Normal, antiserum-treated
FMF
7
8
41.5 t 3.8
24.153.5
2 6.5
0
3
3
7
24.1 2 4.2
46.1 t 2.0
39.4 t 10.6
53.5 2 4.6
-12.7 2 6.7
5.0 t 6.8
Chemotaxis (pm)
2 7.4
Chemotaxis was measured as described in Materials and Methods,
using as chemoattractant 1% (vol/vol) zymosan-activated serum in
the presence and absence of 10% (vol/vol) of the fluid to be
studied.
The mixture was preincubated for 10 minutes at 37°C before addition
to the Boyden chamber. These conditions were chosen to achieve
intermediate levels of C5a inactivation. The antiserum and the nonimmune normal serum were heat-inactivated (56°C for 30 minutes), diluted 1:lOO with PES, and then incubated with an equal volume of
the peritoneal fluid for 30 minutes at 37°C before the chemotaxis
assay was performed. Random migration was 35.0 t 3.2 pm. Results
are expressed as the mean 2 SE. N indicates the number of experiments, each using a different heat-inactivated peritoneal fluid.
From www.bloodjournal.org by guest on February 6, 2015. For personal use only.
3508
AYESH ET AL
The kinetic behavior of the C5a inactivator was evaluated
by the spectrophotometric assay, which allowed measurements to be made of initial rates of C5a inactivation. By this
method, C5a inactivation was proportional to the inactivator
concentration (Fig 7). The purified inactivator obeyed saturation kinetics as the concentration of C5a was varied (Fig 8),
showing K,,, values for C5a of 3.6 and 3.2 ymol/L and V,,,,,
values at 25°C of 60 and 46 nmol CSa/min/mg protein, respectively, in two separate experiments.
0'30
3
DISCUSSION
In defining the physiologic function of the C5ainactivator,
its substrate specificity is a key consideration. Earlier experiments showed that even after inactivation by this enzyme,
C5a wasrecognized by a polyclonal anti-C5a antibody.' This
finding plus others discussed above strongly suggested that
the range of substrates for this inactivator was narrow. Our
experiments with[1251]albumin provide additional support
for this idea andfurther suggest that the activity of the inactivator maybe limited to C5a and possibly to a few other
proteins containing sequences related to the sequence around
the C5a cleavage site.
The C5a inactivator described here was found to be deficient in serosal fluids of patients with FMF, a disorder characterized by inappropriate attacks of inflammation.The clinical features of FMF suggest that the function of the CSa
inactivator is to prevent the development of unprovoked inflammatory reactions by counteracting the effects of small
amounts of C5a that may be accidentally released into the
serosal
At its concentration in peritoneal fluid (=S
pg/mL, based on the peritoneal fluid protein concentration
and the degree of purification neededto obtain homogeneous
enzyme from unfractionated peritoneal fluid), the CSa inactivator will inactivate C5a released by such an accident with
a half-time of about 2 minutes at25°C (and probably an
even shorter half-time at 37"C), a rate that would appear to
be fast enough to prevent the C5a from provoking a full-
?
-2
0
2
4
llC5e
6
8
10
(PM.')
Fig 8. Michaelis constant for the inactivation of rC5a by the C5a
inactivator. Reaction mixtures contained 3.5 p g (66pmol) C5a inactivator and rC5a at theconcentrations indicated.C5a inactivator activity was measured by spectrophotometry at 254 nm as described in
Materials and Methods. The curverepresents a nonlinearleast
squares fit of the data to the Michaelis-Menten e q ~ a t i o n . 'K,,,
~ and
Vmaxvalues were determined from this fit.
scale inflammatory response. We postulate that the attacks
typical of FMF occur because without the protection of the
C5a inactivator, a bolus of C5a released accidentally into a
serosal space will sometimes persist long enough to incite
an inappropriate inflammatory reaction.
At some point it will be important to demonstrate in patients with FMF an abnormality of this C5a-inactivating protein at a molecular level. This demonstration could be accomplished by immunoblotting if the EMF mutation results
in the loss of this protein or the production of a protein of
abnormal size, and if an antibody becomes available that
can detect the protein on a blot of normal unfractionated
ascites. However, a missense mutation may only be detectable as a base replacement in a coding region of the gene
encoding this protein. In principle, this could be determined
by sequencing a cDNA from an FMF patient, but obtaining
the mRNAneeded for cloning such a cDNA will be extremely difficult. It is therefore likely that in patients with
FMF, the detection of mutations in this protein will have to
be performed by exon sequencing at the genomic level.
ACKNOWLEDGMENT
X
We areindebted to Drs Daniel Shouval and Yafa Ashur (Liver
Unit, Hadassah Hospital, Jerusalem) and to Dr Yaacov Yarchovski
(Departmentof Medicine B, Hillel Yave Hospital, Hadera) for kindly
providing us with many of the ascites fluids used in this study.
REFERENCES
30
1 , Ward PA, Hill JH: C5 chemotactic fragments produced by an
enzyme in lysosomal granules of neutrophils. J Immunol 104:535,
Fig 7 . Inactivation of rC5a as a function of C5a inactivator concentration. Reaction mixtures contained 3.3 pmollL rC5a and C5a inactivator at the concentrations indicated. C5a inactivator activity was
measured by spectrophotometryat 254 nm as described in Materials
and Methods.
1970
2. Wright DG, Callin JI: A functional differentiation of human
neutrophil granules: Generation of C5a by a specific (secondary)
granule product andinactivation of C5a by azurophil (primary) granule products. J Immunol 119:1068, 1977
3. Matzner Y, Partridge REH, Babior BM: A chemotactic inhibitor of synovial fluid. Immunology 49:131, 1983
20
0
15 5
10
25
C5a inactivator concentration (nM)
From www.bloodjournal.org by guest on February 6, 2015. For personal use only.
C~A-INACTIVATINGENZYME
4. Matzner Y,Brzezinski A: A C5a inhibitor in peritoneal fluid.
J Lab Clin Med 103:227, 1984
5. Ayesh SK, Feme M, Flechner I, Babior BM, Matzner Y: Partial
characterization of a C5a-inhibitor in peritoneal fluid. J Immunol
144:3066, 1990
6. Matzner Y,Partridge REH, Levy M, Babior BM: Diminished
activity of a chemotactic inhibitor in synovial fluids from patients
with familial Mediterranean fever. Blood 63:629, 1984
7. Matzner Y, Brzezinski A: C5a-inhibitor deficiency in peritoneal fluids from patients with familial Mediterranean fever. N Engl
J Med 311:287, 1984
8. Matzner Y, Ayesh SK, Hochner-Celniker D, Ackerman Z,
Feme M: Proposed mechanism of the inflammatory attacks in famillial Mediterranean fever. Arch Intern Med 150:1289, 1990
9. Gerard N P , Hodges MK, Drazen JM, Weller PF, Gerard C:
Characterization of a receptor for C5a anaphylatoxin on human eosinophils. J Biol Chem 264:1760, 1989
3509
IO. Laemmli U K Cleavage of structural proteins during the assembly of the head of bacteriophage T4.Nature 227:680, 1970
11. Bestango M, Cerino A, Riva S , Astaldi Ricotti GCB: Improvements of Westem blotting to detect monoclonal antibodies.
Biochem Biophys Res Commun 146:1509, 1987
12. Ayesh SK, Azar Y,Babior BM, Matzner Y: Inactivation of
interleukin-8 by the C5a-inactivating protease from serosal fluid.
Blood 81:1424, 1993
13. Baggiolini M, Walz A, Kunkel SL: Neutrophil-activating peptide-lhnterleukin 8, a novel cytokine that activates neutrophils. J
Clin Invest 84:1045, 1989
14. Oppenheim JJ, Zachariae COC, Mukaida N. Matsushima K:
Properties of the novel proinflammatory supergene “intercrine” cytokine family. Annu Rev Immunol 9:617, 1991
15. Leatherbarrow RJ: Enzfitter. A Non Linear Regression Data
Analysis Program. Cambridge, UK, Elsevier Biosoft, 1987
From www.bloodjournal.org by guest on February 6, 2015. For personal use only.
1995 85: 3503-3509
Purification and characterization of a C5a-inactivating enzyme from
human peritoneal fluid
SK Ayesh, Y Azar, II Barghouti, JM Ruedi, BM Babior and Y Matzner
Updated information and services can be found at:
http://www.bloodjournal.org/content/85/12/3503.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.