Departament de Bioquímica i Biologia Molecular, Facultat de Ciències, Universitat Autònoma de Barcelona, 08193-Bellaterra, Spain
Correspondence: M. Victòria Nogués, Departament de Bioquímica i Biologia Molecular, Facultat de Ciències, Universitat Autònoma de Barcelona, 08193-Bellaterra, Spain. E-mail: victoria.nogues{at}uab.es
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Key Words: immunoassay • granulocytes • plasma • sputum
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
The ECP sequence shows a high homology with the sequence of the eosinophil-derived neurotoxin (EDN), another protein of the eosinophil granule matrix. Both ECP [also known as ribonuclease (RNase) 3] and EDN (also known as RNase 2) are ribonucleolytic enzymes belonging to the pancreatic RNase (EC 3.1.27.5) superfamily. ECP has three potential glycosylation sites for asparagine-linked oligosaccharides, and three glycosylated forms of native ECP (with 18-, 20-, and 22-kDa molecular masses) have been described [7 , 8 ]. Recent studies suggest that the differential glycosylation of ECP may be critical for the regulation of its biological function [2 ].
ECP levels in serum, sputum, and other fluids are currently used as markers for the diagnosis and monitoring of the therapeutic efficacy in acute- and chronic-inflammation diseases, as well as other hypereosinophilia syndromes [2 ]. Several antibodies designed to determine the ECP levels in biological samples have been described. Monoclonal antibodies EG1 and EG2 have been used in clinical studies for the detection and quantification of ECP [9 ]. These antibodies were raised against either eosinophil extracts or the eosinophil granule products, and their recognition capabilities are discussed below. An enzyme-linked immunosorbent assay (ELISA) using polyclonal antibodies against purified native ECP has also been described [10 ]. Pharmacia & Upjohn (Uppsala, Sweden) developed an improved immunoassay for ECP detection (Pharmacia CAP SystemTM) [11 ], and a fully automated assay that combines immunoglobulin (Ig) E with other allergy markers is currently being used for the monitoring of allergy immunotherapy (UniCAPTM, Pharmacia & Upjohn) [12 ]. However, the lack of information on the epitope involved in the immune detection system and the lack of specificity of the assays reduce some of the potential applications of these methods for basic research studies.
We have developed a specific antibody against a defined region of ECP. This antibody does not cross-react with either EDN or the other proteins of the pancreatic RNase superfamily, but it detects both the glycosylated and nonglycosylated forms of ECP, either native or denatured. The selected epitopewas chosen according to a prediction model of the ECP three-dimensional structure [13 ] based on the EDN three-dimensional structure [14 ]. The sequence D115–Y122 corresponds to a unique exposed region, which is a specific large insertion loop in eosinophil-associated RNases and in RNase K6 [15 ] (Fig. 1A and B ). The epitope was further characterized by an analysis of the X-ray crystal structure of ECP (Fig. 1A) [16 ]. Rabbit polyclonal antibodies were obtained with an ECP synthetic peptide, and antibodies were further purified by affinity chromatography using immobilized recombinant ECP (rECP). We have characterized the antibody by ELISA, dot blotting, sodium dodecyl sulfate (SDS)-polyacrylamide gel electrophoresis (SDS-PAGE), and Western blotting. Plasma, granulocytes, and sputum samples have been analyzed with this antibody.
 View larger version (56K): [in a new window] |
Figure 1. (A) Ribbon schematic representation of the three-dimensional structure
of ECP [16] using the program Molscript [31]. All nonhydrogen atoms
are shown for residues D115–Y122. The structure of the loop used as
epitope and also the N- and C-terminal ends are shown in color. (B)
Amino acid sequence alignment for ECP and its human homologues of the
pancreatic RNase family, RNase 1 (pancreatic RNase), EDN (RNase 2), ECP
(RNase 3), RNase 4, Ang (RNase 5), and RNase K6 (RNase 6). The
alignment was performed with the PileUp program (version 8, Wisconsin
Package; Genetics Computer Group). The numbering corresponds to the
human ECP sequence. Conserved residues in all the sequences are marked
in white with a black background, and conserved residues on either five
and four of the sequences are marked in white with a dark-gray
background and in black with a light-gray background respectively. The
potential glycosylation sites are indicated with arrows. The selected
epitope region is indicated with asterisks. _art>
|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Peptide sequence design
The peptide for immunization was selected by comparison of the
primary sequences of both human ECP and EDN in the primate family
[15
] and by comparison of eosinophil-associated RNases
ECP and EDN with other members of the pancreatic RNase superfamily
[17
]. Superposition of the main chain backbone of the
human ECP primary sequence on the EDN three-dimensional structure
[13
] and structure prediction indicated that the loop
region corresponding to D115–Y122 is the region where the main chain
backbones of the two proteins differ most [14
]. Indeed,
the sequence corresponds to an exposed loop in the ECP
three-dimensional structure—as confirmed recently by X-ray
crystallography [16
] (Fig. 1A)
—containing a unique
insertion found in eosinophil RNases (EDN and ECP) and in the
phylogenetically close relative RNase K6 (Fig. 1B)
. A homology search
in the protein data bank by the BLAST program (version 8, Wisconsin
Package; Genetics Computer Group, Madison, WI) indicated that the
selected region has significant homology only with other mammalian ECP
sequences [15
].
The selected peptide sequence used for immunization (KGDNRDPRDSPRYP) included the loop 115–122 and some extra residues at both sides of the ECP primary sequence (D112–P123). Finally, the KG dipeptide was added at the N terminus, and this K residue was linked to a carrier protein (ovalbumin) to increase the immunogenic response.
Antibody production and partial purification using protein
A-Sepharose
The synthetic peptide (KGDNRDPRDSPRYP) conjugated to a carrier
protein (ovalbumin) was used as an immunogen, and rabbit polyclonal
antibodies were obtained by Neosystem Laboratoire according to their
standard immunization protocol. Total antibodies were purified from
plasma by means of protein A-Sepharose CL 6B chromatography on a 2-mL
column. The column was equilibrated with 50 mM Tris-HCl, pH 7.0, and 2
mL of rabbit serum were run twice through the column. Antibodies were
eluted with 0.1 M glycine, pH 3.0; 1 M Tris-HCl, pH 8.0, was added to
the collected fractions to neutralize the pH.
Antibody purification by column affinity chromatography with
immobilized ECP
rECP, previously obtained in our laboratory using a prokaryotic
expression system [14
], was used as antigen for the
purification of specific antibodies. Following the manufacturer’s
recommendations, 2 mg of rECP and a 2-mL Aminolink were both
equilibrated with phosphate-buffered saline (PBS; 7.5 mM
Na2HPO4, 2.5 mM
NaH2PO4, 0.14 M NaCl, pH 7.2). Forty
microliters of 5 M sodium cyanoborate were added as a reducing agent,
and the mixture was left at 25°C for 6 h and then washed with
PBS. Free reactive groups were then blocked with 1 M Tris-HCl, pH 7.4,
and the column was washed with 1 M NaCl and equilibrated again in PBS.
Two milliliters of immunized rabbit serum were applied to the affinity
column, and the column was washed again with PBS and 1 M NaCl. Bound
antibodies were eluted with 0.1 M glycine, pH 3.0, and mixed with 1 M
Tris-HCl, pH 8.0, to neutralize the pH. Elution was monitored by
measuring the absorbance at 280 nm (A280). Peak
fractions were pooled, and the final antibody concentration was
determined spectrophotometrically
(A2800.1% = 0.75). The
purified fraction of epitope-specific antibodies was 
10% of the
protein A-Sepharose-derived total antibodies.
Preparation of granulocytes
Blood samples were collected from healthy volunteers. Five
milliliters of blood samples were mixed with 0.5 mL of 5% EDTA, and
1.375 mL of 3.7% dextran (average relative molecular weight,
250,000) was added. After a 20-min incubation (37°C) at a 45°
inclination and 5 min in a vertical position, the leukocyte-rich plasma
fraction was separated and diluted with PBS (1:1, v/v); and then the
sample was poured over a Lymphoprep solution (3:1, v/v). After
centrifugation of the samples at 800 g for 30 min, four
layers were formed. The upper plasma fraction was decanted and kept at
-20°C until analysis. The pellet, which contained erythrocytes and
granulocytes, was further processed as follows. Red blood cells were
lysed by adding 0.4 M NH4Cl to the tubes, which were kept
at 0°C for 5 min. The remaining granulocytes were washed twice with
PBS, sonicated, and centrifuged before analysis.
Partial purification of plasma by ion-exchange chromatography
Plasma was desalted by gel filtration chromatography with a PD10
column previously equilibrated in a pH 8.0 solution consisting of 15 mM
Tris-HCl and 2 mM EDTA. Samples were then loaded onto a HiTrap Q
anion-exchange prepacked column that had been equilibrated with 15 mM
Tris-HCl–2 mM EDTA, pH 8.0. The nonretained fraction, which contained
plasma basic proteins, was loaded onto a cation-exchange column (HiTrap
SP) equilibrated in the same buffer. Proteins retained by the
cation-exchange column were then eluted with a linear salt gradient of
0–1 M NaCl. Aliquots (0.5 mL) were collected, and the peak fractions
were further concentrated and dialyzed against distilled water. Eluted
fractions were assayed for activity with zymogram staining gels and
were characterized by immunoblotting.
Sputum processing
Healthy volunteers and asthmatic patients were induced to
produce sputum via hypertonic saline inhalation. The sputum samples
were processed as described by Gibson et al. [18
] with
the following modifications: sputum plugs were isolated from the
sample, diluted (1:4, v/v) with 0.1% dithiothreitol, mixed by
vortexing, and kept at room temperature for 15 min to disperse the
cells. The samples were diluted 1:4 with PBS, filtered through a
50-µm nylon mesh filter, and centrifuged at 4°C for 10 min at 700
g. The supernatant was removed, stored at -20°C, and
thawed immediately before assay. The samples were concentrated if
necessary and dialyzed against distilled water. The ECP levels in
sputum were found to be stable during sputum processing and storage
[18
].
ELISA
The reactivity of D112–P123 Ab against both the original
synthetic peptide conjugated to ovalbumin and rECP was tested by ELISA
[19
]. The ELISA method was as follows. Wells were coated
overnight at room temperature with 100 µL of antigen at different
concentrations in a 0.1 M carbonate-bicarbonate buffer, pH 9.6. Wells
were washed three times with 0.05% (v/v) Tween 20 in PBS and blocked
with 1% bovine serum albumin (BSA) in 0.05% Tween 20 in PBS for
1 h at 37°C. The wells were washed again with the same solution
and incubated at 37°C for 2 h with 100 µL of the purified
D112–P123 Ab (20 ng/mL–5 µg/mL concentration range) in 1% BSA and
0.05% Tween 20 in PBS. After the wells were washed with 0.05% Tween
20 in PBS, 100 µL of a 1:3,000 dilution of horseradish
peroxidase-linked goat anti-rabbit IgG in PBS–1% BSA–0.05% Tween 20
were added per well. Finally, after the wells were washed again with
0.05% Tween 20 in PBS, 100 µL of substrate solution containing 40 mM
3-dimethylaminobenzoic acid, 0.8 mM 3-methyl-2-benzothiazoline
hidrazone, and 3 mM H2O2 in 0.1 M
phosphate buffer, pH 7.0, were added to each well. After a 20-min
incubation, the reaction was stopped with 50 µL of 2 M
H2SO4. The A620 was
measured with a plate reader (Anthos reader 2001, Anthos Labtec
Instruments, Cultek, Spain).
Immunoblotting analysis: Western blot and dot blot
Samples were mixed with loading buffer [60 mM Tris-HCl, 10%
(v/v) glycerol, 0.015% (w/v) bromophenol blue, 3% SDS, pH 6.8, with
or without 5% ß-mercaptoethanol] and loaded onto an SDS–15%
polyacrylamide gel. After electrophoresis, proteins were transferred to
a prewetted Immobilon P membrane using an optimized transfer buffer for
highly cationic proteins [39 mM glycine, 0.1% (w/v) SDS, 20% (v/v)
methanol, pH 9.0]. The transfer was performed at 100 V for 25 min. The
Immobilon P membranes were washed with Tris-buffered saline (TBS) (25
mM Tris-HCl, 140 mM NaCl, 3 mM KCl, pH 8.0) and blocked with blocking
buffer [TBS containing 0.05% Tween 20 and 3% (w/v) BSA]. Detection
was carried out with purified D112–P123 Ab as the primary antibody,
and the optimal dilution from an initial stock solution of 0.1 mg/mL
was previously assessed. Blots were then incubated with the secondary
antibody in TBS containing 0.05% Tween 20 and 1% (w/v) BSA, using a
1:3,000 dilution of the biotinylated goat anti-rabbit IgG antibody and
a 1:3,000 dilution of the streptavidin-biotinylated alkaline
phosphatase complex. Blots were finally developed in a pH 9.5 solution
consisting of 100 mM Tris-HCl and 0.5 mM MgCl2 and
containing 0.1 mg/mL of 5-bromo-4-chloro-3-indolyl phosphate and 0.3
mg/mL of nitroblue tetrazolium. The reaction was stopped by adding
distilled water. Alternatively, blot membranes were developed with a
SuperSignal Ultra chemiluminescence kit, using a 1:200,000 dilution of
the horseradish peroxidase-linked goat anti-rabbit IgG antibody and the
fluorescent substrate luminol. Bands were detected as film signals
(Hyperfilm-MP high-performance autoradiography film). For the testing
of the monoclonal antibody EG1, a 1:20 dilution was used; and detection
was carried out by using a 1:3,000 dilution of alkaline
phosphatase-labeled goat anti-mouse antibody. In all cases, bands were
scanned (Imaging Densitometer, model GS-700; Bio-Rad Laboratories) and
quantified using the software Multi Analyst (version 1.1) (Bio-Rad
Laboratories).
For comparison with immunoblotting results, total protein staining was performed with Coomassie blue R-250 dye in 7% acetic acid and 12% methanol. The destaining solution was 7% acetic acid and 20% methanol.
The dot blotting was performed on nitrocellulose membranes with a blotting instrument connected to a vacuum pump (Bio-Dot Microfiltration Apparatus, Bio-Rad Laboratories). Antigen samples were mixed with transfer buffer (39 mM glycine, 0.1% SDS, 20% methanol, pH 9.0) and transferred onto a nitrocellulose membrane. Immunodetection was carried out using purified D112–P123 Ab as the primary antibody and the biotinylated goat anti-rabbit IgG as the secondary antibody according to the procedure described above.
Activity-staining gels (zymogram)
Samples obtained from the partial purification of human plasma
and from granulocytes and sputum fractions were mixed with nonreducing
loading buffer (60 mM Tris-HCl, 10% glycerol, 0.015% bromophenol
blue, 3% SDS, pH 6.8) and were analyzed for RNase activity by zymogram
on SDS–15% polyacrylamide gels containing 0.6 mg/mL of either poly(U)
or poly(C) as substrate [20
]. After elimination of SDS
by incubation with a pH 7.5 solution consisting of 10 mM Tris-HCl and
20% isopropanol, the gels were incubated at 25°C for 90 min in 100
mM Tris-HCl, pH 7.5. The gels were stained with 0.2% toluidine blue in
10 mM Tris-HCl, pH 8.0. The relative intensity of the areas showing
substrate degradation was analyzed by densitometry.
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
10% of the total antibody
fraction previously obtained by the protein A-Sepharose chromatography.
Antibody titration and cross-reactivity
Purified antibodies were assayed by ELISA against the conjugated
synthetic peptide (0.1–9 µg) and rECP (0.2–37 µg). The antibodies
were serially diluted from a 0.1-mg/mL stock solution to working
concentrations ranging from 20 ng/mL to 5 µg/mL and added to
microtiter plates coated with either rECP or the conjugated peptide.
The amount of antibody binding was measured colorimetrically at 620 nm
(Fig. 2
).
 View larger version (16K): [in a new window] |
Figure 2. Analysis of the antibody working range. The relationship between the
D112–P123 Ab concentration and the detection signal
(A620) using rECP (•) and the conjugated
peptide ( ) is shown. The analysis was carried out with the ELISA,
using as the secondary antibody the horseradish peroxidase-linked goat
anti-rabbit IgG at a 1:3,000 dilution. Solutions of D112–P123 Ab
purified by affinity chromatography with immobilized rECP were made
ranging from 20 ng/mL to 5 µg/mL, and 200 ng (127 nM) of rECP and 100
ng (19.5 nM) of the conjugated peptide were used as antigens.
|
75 ng of rECP (Fig. 3A
) and that the assay could detect as little as 1 ng of rECP (Fig. 3B)
. To check the reproducibility of the method, the relationship
between the ECP level (in nanograms) and the integration of the
corresponding band densitogram (in optical density units per square
millimeter) was obtained from four independent experiments, and the
regression line was determined (y = 0.0755x;
r2 = 0.9999). Immunodetection with a SuperSignal Ultra
chemiluminescence kit gave similar results (Fig. 3C)
, although an
increase in the detection levels could be achieved with longer film
exposure times. A faint upper band corresponding to the formation of
small amounts of dimer was observed when large quantities of purified
rECP were loaded on the SDS-polyacrylamide gel. This band was also
observed when the SDS-polyacrylamide gel was stained with Coomassie
blue for detection of total protein. Lack of cross-reactivity against 3
µg of EDN and RNase A was demonstrated by SDS–15% PAGE and Western
blot analysis (Fig. 4A
).
 View larger version (57K): [in a new window] |
Figure 3. Analysis of D112–P123 Ab immunoreactivity by Western blot analysis
using rECP as antigen. (A) Relationship between rECP (ng) and
integrated densitogram of the detected band. After SDS–15% PAGE, the
Western blot was developed with the primary antibody D112–P123 Ab at
0.2 µg/mL using biotinylated goat anti-rabbit IgG Ab at a 1:3,000
dilution and streptavidin-biotinylated alkaline phosphatase complex at
a 1:3,000 dilution. For each quantity of rECP, four independent
measurements were made, and the regression line was determined
(y = 0.0755x; r2 = 0.9999). (B)
SDS–15% PAGE and Western blot analysis developed by the
streptavidin-biotinylated alkaline phosphatase complex. Lane 1, 100 ng
of rECP; lane 2, 75 ng of rECP; lane 3, 50 ng of rECP; lane 4, 25 ng of
rECP; lane 5, 10 ng of rECP; lane 6, 5 ng of rECP; lane 7, 1 ng of
rECP; and lane 8, 0.5 ng of rECP. (C) SDS–15% PAGE and Western blot
analysis of rECP developed using the SuperSignal Ultra
chemiluminescence kit, luminol as substrate, and horseradish peroxidase
goat anti-rabbit IgG secondary antibody diluted to 1:200,000. Lane 1,
50 ng of rECP; lane 2, 10 ng of rECP.
|
 View larger version (41K): [in a new window] |
Figure 4. SDS–15% PAGE and Western blot of rECP and ECP in biological samples.
Given the different ECP level in each biological sample, volumes were
adjusted to obtain the best detection signal. The primary antibody,
D112–P123 Ab, was used at a concentration of 0.2 µg/mL. Blots were
developed using biotinylated goat anti-rabbit IgG Ab at a 1:3,000
dilution and streptavidin-biotinylated alkaline phosphatase complex at
a 1:3,000 dilution (except in lanes M and 6 of panel B). (A) Lane 1, 50
ng of rECP; lane 2, 3 µg of RNase A; lane 3, 3 µg of EDN. (B)
Molecular mass markers stained with Coomassie blue R-250 (bovine serum
albumin, ovalbumin, carbonic anhydrase, trypsin inhibitor, and
lysozyme). Lane 1, 50 ng of rECP; lane 2, 17 µL of the initial
fraction of the ECP elution peak from HiTrap SP chromatography; lane 3,
17 µL of the maximum of the ECP elution peak from HiTrap SP
chromatography; lane 4, 15 µL of processed sputum, corresponding to
0.5 mL of initial collected sputum; lane 5, 8 µL of purified
granulocytes from 120 µL of total blood; and lane 6, 17 µL of the
same sample that was used in lane 2, the incubation step with
D112–P123 Ab was omitted, and the blot was developed using a
SuperSignal Ultra chemiluminescence kit.
|
To check D112–P123 Ab recognition capability in cells and biological fluids, samples of plasma, granulocytes, and sputum were also assayed by Western blotting and immunodetection (Fig. 4B) . In both sputum and granulocyte fractions, we could identify a lower and predominant band corresponding to the nonglycosylated form, with a molecular mass equivalent to that of rECP. For plasma samples, only the nonglycosylated band was detected.
Plasma samples
For plasma samples, detection of ECP in Western blotting was
feasible only if plasma was previously concentrated. ECP levels in
serum ranged from 1 to 20 µg/L in healthy subjects, increasing up
to 200–500 µg/L in subjects with inflammatory diseases
[2
, 15
]. Sensitivity was increased by
partially purifying plasma samples by ion-exchange chromatography as
described in Materials and Methods. SDS-PAGE and total protein staining
with Coomassie blue indicated that most of the plasma proteins had been
eliminated during the partial purification of the sample (results not
shown), although different bands with RNase activity were observed by
the zymogram technique (Fig. 5B
, lane 2). After SDS-PAGE and Western blotting of plasma samples
and immunodetection with the streptavidin-biotinylated alkaline
phosphatase complex, only a band corresponding to the nonglycosylated
form of ECP was detected in the maximum of the elution peak (Fig. 4B
,
lane 3). This was likely due to the low concentration of ECP in plasma
within the limits of the sensitivity of the assay. However, in the
initial fraction of the elution peak of ECP from ion-exchange
chromatography, two additional bands of 25 and 50 kDa were detected
(Fig. 4B
, lane 2). The molecular mass of each band was estimated by
comparison with standard molecular mass markers. These bands were also
detected when the Western blot development was performed without the
primary ECP-specific antibody, using only the commercial anti-rabbit
IgG secondary antibody (Fig. 4B
, lane 6). These bands may correspond to
the light and heavy chains of human IgG. Transfer to an Immobilon
membrane and N-terminal sequencing of these bands showed that the
25-kDa band corresponded to the light chain of human IgG. In the case
of the 50-kDa band, the sequence of the heavy chain was not obtained,
probably due to blocking of the N terminus of the protein. N-terminal
sequencing indicated the presence of albumin or apolipoprotein H. In
conclusion, the bands, identified as human IgG, eluted from the cation
exchange column at high ionic strength, close to the maximum peak
fraction for ECP, during the partial purification from plasma and did
not cross-react with the ECP-specific primary antibody.
 View larger version (52K): [in a new window] |
Figure 5. RNase activity staining on SDS–15% polyacrylamide gels containing
either poly(C) (top) or poly(U) (bottom) as substrate. Because the
detection signal depends on both the amount of protein and the specific
activity of each RNase species, the volumes loaded were accordingly
adjusted to obtain the best detection signal. (A) Lane 1, 10 ng of
native EDN; lane 2, 170 ng of rECP; lane 3, 150 pg of RNase A. (B) Lane
1, 100 ng of rECP; lane 2, 5 µL of HiTrap SP peak fraction. (C) Lane
1, 100 ng of rECP; lane 2, 6 µL of purified granulocytes
corresponding to a volume of 60 µL of total blood; lane 3, 3 µL of
processed sputum, corresponding to 0.1 mL of initial collected sputum.
The differences in the activities of ECP bands depended on the
incubation times.
|
The ECP content in granulocytes includes the quantities detected in
eosinophils, neutrophils, and basophils. Granulocytes are mainly
composed of neutrophils (95.8%) but also contain smaller proportions
of eosinophils (3.5%) and basophils (0.6%), with the ECP content of
neutrophils being 30- to 100-fold lower than that of the eosinophils
[21
, 22
]. Abughazaleh et al.
[21
] reported ECP concentrations of 5
µg/106 cells in eosinophils, 50 ng/106 cells
in neutrophils, and 80 ng/106 cells in basophils; and
recent immunofluorescence studies indicated ECP contents of 3.4
µg/106 cells in eosinophils and 150 ng/106
cells in neutrophils [22
]. Very recent studies suggest
that the ECP content detected in neutrophils results from an active
uptake by the neutrophils of the protein secreted by the eosinophils
[2
].
Sputum samples
Sputum analysis is one of the selective noninvasive markers of
airway inflammation. Measurement of ECP levels in sputum seems to
reflect inflammation and bronchial obstruction better than that of ECP
levels in blood [23
]. Processing of sputum samples was
carried out as described in Material and Methods. Treatment with 0.1%
dithiothreitol ensures cell dispersion, and the assayed fraction
corresponds to the sputum supernatant [18
]. The sputum
supernatant reflects only the ECP released by eosinophil degranulation,
although the possibility of a small percentage of eosinophil lysis due
to sample handling cannot be ruled out. Western blot analysis allows
detection of a predominantly nonglycosylated band as well as some
additional bands of higher molecular mass in the sputum supernatant
(Fig. 4B
, lane 4), although a more detailed study is necessary to
quantify and identify the different glycosylated forms. ECP reference
values in the literature range from 20- to 600-µg/L; values of
>400–600 µg/L are found in asthmatic patients [18
].
RNase activity was also analyzed by the zymogram technique (Fig. 5C
,
lane 3).
We used the sputum samples to check the recognition capability of D112–P123 Ab in relation to the values determined by the UniCAPTM system. Both systems yielded comparable results (Table 1 ). The Western blot method was used to determine the ECP levels and the relationship between the integrated densitogram of each band, and the ECP quantity was correlated using the regression line obtained from Figure 3A (y = 0.0755x; r2 = 0.9999). In each assay, an rECP sample (75 ng) was used as an internal standard. Treated asthmatic patients showed control values, except in case number 2; these samples were used only as a reference to compare the two methods.
|
View this table: [in a new window] |
Table 1. Determination of ECP Levels in Sputum Samples by Immunoreactivity
of the D112–P123 Ab and by the UniCAPTM System
|
RNase A, EDN, and ECP show different cleavage rates for the poly(U) and poly(C) substrates (Fig. 5A) . While pancreatic RNase 1 shows a preference for the poly(C) substrate, eosinophil RNases prefer poly(U), although the poly(U)/poly(C) ratio is much higher for EDN than for ECP. The zymogram is a very sensitive technique that can detect as little as 50–100 pg of RNase A [29 ], 1–5 ng of native EDN [19 ], and 10–15 ng of rECP [14 ]. In Figures 5B and 5C , the bands detected by poly(U) and poly(C) activity-staining electrophoresis of granulocytes, plasma, and sputum samples are compared. Comparison of the plasma fraction (Fig. 5B , lane 2) with the granulocyte sample (Fig. 5C , lane 2) shows a predominance of eosinophil RNases in the granulocyte fraction, with a clear preference for poly(U).
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
We have developed an epitope-specific polyclonal antibody for the specific immunodetection of ECP. A specific epitope was selected according to a prediction of ECP’s three-dimensional structure [13 ]. The region that corresponds to an exposed loop in ECP, where the main chain adopts a different conformation in EDN than in ECP, was selected. The recently resolved crystal structure of ECP [16 ] confirmed the chosen region as a putative antigenic epitope (Fig. 1A) .
Polyclonal antibodies were raised using a conjugated synthetic peptide corresponding to the sequence D112–P123, which includes the identified loop region D115–Y122. Specific purification of the antibodies on an affinity column using immobilized rECP as an antigen eliminated any possible background due to the rabbit immunogenic reaction to the carrier protein or to any other secondary structure adopted by the synthetic peptide that did not correspond to its conformation in the three-dimensional native structure.
D112–P123 Ab reacted with the nonglycosylated rECP, expressed in a prokaryote system, and with the native human ECP present in plasma, granulocytes, and sputum. In the Western blot, a lower band corresponding to the nonglycosylated protein and additional bands of higher molecular mass, which would correspond to the ECP glycosylated forms, were detected by the antibody. The antibody had a high sensitivity, detecting as little as 1 ng of rECP; it could detect ECP in both its reduced and nonreduced forms, and this antibody did not cross-react with either EDN or RNase A.
Immunoblotting results were always compared with those obtained by Coomassie blue total protein staining and by the zymogram technique. The activity-staining method had a very high sensitivity but could not discriminate between the different proteins with RNase activity and similar molecular masses or between the corresponding glycosylated isoforms present in biological samples.
D112–P123 Ab recognition capability can be compared with that of other ECP immunodetection assays. However, both the source of the antigen and the antibody specificity have to be taken into account. Monoclonal antibodies EG1 and EG2 raised against eosinophil extracts and eosinophil granule products [9 ] have been suggested to be useful for the monitoring of eosinophil degranulation. EG1 recognizes ECP in both its storage and secreted forms, and EG2 recognizes both ECP and EDN, albeit only in their secreted forms. EG1 can be used only in nonreducing conditions, and EG2 detects only one glycosylated form of ECP [7 ]. Detection with EG2 has been used in many clinical studies as a tool to distinguish between resting and actively secreting eosinophils, and such a method is useful for monitoring and diagnosing inflammatory diseases. However, the physiological significance of stored and secreted forms of eosinophil RNases is still controversial, and the structural differences are unknown. Moreover, Jahnsen et al. [28 ] have shown that the EG2 reactivity does not discriminate between resting and activated eosinophils. These authors concluded that EG2 antibody could not be used as an activation marker in immunohistochemistry. Their results suggested that the reported differences in staining were likely caused by the granule protein leaking from the eosinophils as a result of the fixation technique rather than as a result of cellular secretion. Recently, Nakajima et al. [27 ] tested the EG1 and EG2 reactivity by radioimmunoassay and Western blot analysis and concluded that their differential reactivity was dependent on both the method of fixation and the antibody concentration.
We have compared the binding affinities and specificities of D112–P123 Ab and EG1. Immunoblotting in nonreducing conditions using a 1:10 dilution of EG1 antibody could detect rECP as well as plasma and granulocyte ECP; there was no cross-reaction with other proteins, but there was a slightly lower sensitivity than that of D112–P123 Ab at a 1:500 dilution. Some basic research studies have also been carried out using polyclonal antibodies against purified native ECP [10 , 29 ].
Most of the current clinical studies use the newly developed Pharmacia
CAP SystemTM for the monitoring of ECP levels, with a detection limit
of 2 ng/mL [11
]. Unfortunately, there is not much
literature available on either its specificity or the epitope
recognized by the antibody. In sputum samples, a good correlation has
been obtained between the automated assay of the Pharmacia CAP SystemTM
(UniCAPTM) and D112–P123 Ab recognition (Table 1)
.
D112–P123 Ab should prove useful for the analysis of different ECP isoforms in biological tissues and fluids. The antibody does not cross-react with EDN, can be used in both reducing and nonreducing conditions, and has a sensitivity detection limit of 1 ng. In the present study, a preliminary analysis of ECP isoforms in plasma, granulocytes, and sputum samples was carried out using the ECP epitope-specific D112–P123 Ab. Further analyses will be carried out to characterize and quantify the glycosylated forms of ECP in plasma, eosinophils, and sputum. It has been reported that the several ECP isoforms display different biological properties [2 ]. Glycosylation is regarded as a way of fine-tuning to modulate the functionality of the secreted protein and as a target-cell recognition marker [30 ]. The specific identification and quantification of the ECP glycosylated forms will help us understand the process of ECP secretion and regulation. To assess the putative involvement of this domain in ECP biological properties, functional studies of the neutralizing effects of the epitope-specific D112–P123 Ab will also be performed.
The authors are grateful to Dr. Franchek Drobnic (Centre d’Alt Rendiment Esportiu, Sant Cugat del Vallès, Spain) for providing blood and sputum samples and for his advice on the sample processing protocols and to Dr. José Belda (Servei de Pneumologia, Hospital de la Santa Creu i Sant Pau, Barcelona, Spain) for providing sputum samples and for determining the ECP levels using the UniCapTM system. We thank Dr. G. J. Gleich (Department of Immunology, Mayo Clinic and Mayo Foundation, Rochester, MN) for providing native EDN. We also thank Dámaso Torres for his contribution to the ICM modeling of ECP, Dr. K. R. Acharya for his help in the analysis of the X-ray crystal structure of ECP, and Dr. Gennady Moiseyev for helpful discussions.
Received March 24, 2000; revised January 29, 2001; accepted January 31, 2001.
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
 |
|
 |
|
  T.-W. Pai, M. D.-T. Chang, W.-S. Tzou, B.-H. Su, P.-C. Wu, H.-T. Chang, and W.-I Chou REMUS: a tool for identification of unique peptide segments as epitopes. Nucleic Acids Res., July 1, 2006; 34(Web Server issue): W198 - W201. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
