Journal of Leukocyte Biology eBioscience full spectrum cell analysis
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

Originally published online as doi:10.1189/jlb.0507272 on September 7, 2007 Originally published online as doi:10.1189/jlb.0507272 on August 28, 2007

Published online before print August 28, 2007
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow All Versions of this Article:
jlb.0507272v1
jlb.0507272v2
82/6/1491    most recent
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Eriksson, J.
Right arrow Articles by Venge, P.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Eriksson, J.
Right arrow Articles by Venge, P.
(Journal of Leukocyte Biology. 2007;82:1491-1500.)
© 2007 by Society for Leukocyte Biology

A SELDI-TOF MS study of the genetic and post-translational molecular heterogeneity of eosinophil cationic protein

Jenny Eriksson*,1, Charlotte Woschnagg*, Eva Fernvik{dagger} and Per Venge*

* Department of Medical Sciences, Clinical Chemistry, Uppsala University, Uppsala, Sweden; and
{dagger} Bio-Rad Laboratories, Sundbyberg, Sweden

1Correspondence: Department of Medical Sciences, Clinical Chemistry, Uppsala Academic Hospital, SE-75185 Uppsala, Sweden. E-mail: jenny.eriksson{at}medsci.uu.se


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
Eosinophil cationic protein (ECP), a secretory protein of the eosinophil granulocyte, is a basic and highly heterogeneous protein. This heterogeneity is dependent on polymorphisms in the ECP gene and post-translational modifications, and it affects the functional properties of the protein in terms of cytotoxicity. The aim of this study was to further investigate the molecular heterogeneity, hence, an affinity capture assay based on an antigen-antibody interaction with the surface-enhanced laser desorption/ionization-time of flight mass spectrometry (SELDI-TOF MS) technique was developed. Of three monoclonal antibodies tested, that is, EG2, 614, and 652, the 614 mab was chosen for the experiments. ECP heterogeneity of single individuals was studied in extracts of purified blood eosinophils, and the presence of ~5 major molecular species was demonstrated in each subject. ECP from subjects with different ECP 434(G>C) genotypes (arg97thr) showed mass differences corresponding to the amino acid shift from arginine to threonine. ECP purified from pooled leukocytes of large numbers of healthy blood donors demonstrated an extensive mass heterogeneity with ~10 major molecular species. By the use of a variety of glucosidases it was shown that this heterogeneity was mainly due to N-linked oligosaccharides on which sialic acid, galactose, and acetylglucosamine was positioned. We conclude that the SELDI-TOF MS technique using specific monoclonal antibodies is a convenient and versatile tool; by means of this technique, we could detect both genetic and post-translational causes of the molecular heterogeneity of the eosinophil cationic protein.

Key Words: eosinophil granulocyte • proteomics • glycosylation • gene polymorphism


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
Protein heterogeneity is common throughout the proteome and manifests itself both as differences in molecular weight and differences in the net charge of the protein. This heterogeneity is a result of several factors, changes in the genome such as single nucleotide polymorphisms (SNPs) or posttranslational modifications, glycosylation, phosphorylation, and others. Glycosylation is one of the most common post-translational modifications, and large mass differences in a protein are usually due to these oligosaccharide substitutions, especially N-linked oligosaccharides.

The eosinophil granulocyte is a leukocyte comprising only a few percentages of the total number of blood leukocytes in a healthy subject [1 ]. It is involved in most inflammatory disorders in the body and plays a significant role in allergy, asthma [2 , 3 ], and tissue remodeling [4 5 6 ]. The eosinophil produces several proteins of which some granule proteins, such as eosinophil cationic protein (ECP), are highly heterogeneous. Western blot analysis and MS/MS analysis have shown that the molecular weight of the protein can differ from 15 to 22 kDa [7 ]. This heterogeneity is known, in part, to be due to glycosylation of the protein; there are three known sites for N-linked glycosylations in the molecule [8 ]. Several authors showed the presence of N-linked glycans on ECP [9 10 11 ], but whether other glycans also are present is still unknown. Several SNPs are present in the ECP gene, of which one is present in the coding region [12 ]. The ECP 434(G>C) polymorphism changes the amino acid arginine to a threonine (arg97thr) [12 , 13 ] with the consequence of alterations of the net charge and mass of ECP, thus contributing to the protein heterogeneity further, but also with an alteration in the cytotoxic activity [14 ].

ECP has multiple functions, and its cytotoxic properties might be the most conspicuous. ECP is cytotoxic to several mammalian cells, to bacteria [15 ], viruses [10 ], and parasites [16 ], and this cytotoxic property makes it an important protein in the reactions of the innate immunity. The antibacterial properties of ECP seem to be dependent on both aromatic and basic amino acids [17 ], the RNase activity of the protein, however, seems to be of no importance for this toxicity [18 ].

The variant alleles of the polymorphic gene have been associated to different pathological conditions. A recent study by our group demonstrated a correlation of ECP 434(G>C) genotype and infection with the S. mansoni parasite, subjects with the 434GG genotype (97arg) were found to be infected by the parasite to a lower extent than subjects with the other ECP genotypes, implicating a role for ECP in the primary defense reaction in the skin [19 ]. Also, the infected subjects carrying the G-allele displayed a higher frequency of peri-portal fibrosis, implicating a role for ECP, and in particular, ECP with arginine at position 97, in development of fibrosis. The 434G-allele has also been correlated to Hodgkin lymphoma [20 ] and the presence of allergic symptoms [13 ].

The coding ECP 434(G>C) polymorphism (arg97thr) not only changes an amino acid in the mature protein, but when this shift takes place, a potential new glycosylation site is created. It could be either O-linked glycosylation where the O-linked glycans are linked to the hydroxyl group of threonine, or N-linked glycosylation where the carbohydrate is attached to the asparagine in the tripeptide sequence Asn-x-Thr. The reason for the differences in function of the two allele products could be either due to the actual amino acid shift, or to differences in glycosylation of the two allele products.

The aim of this study was to further examine the posttranslational heterogeneity of ECP, using an affinity capture assay with the surface-enhanced laser desorption/ionization-time of flight mass spectrometry (SELDI-TOF MS) technology. The heterogeneity was studied both in ECP purified from pooled buffy coats of several blood donors and ECP that was extracted from the eosinophils of single individuals. To try to identify the carbohydrate constituents on ECP, enzymatic deglycosylation was used.


   MATERIALS AND METHODS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS
 DISCUSSION
 REFERENCES
 
Isolation of granules
ECP was purified from buffy coats using the method described by Peterson et al. [21 ] with some modification. Buffy coats from healthy blood donors were mixed with an equal volume of 2% Dextran T-500 (GE Healthcare Biosciences, Uppsala, Sweden) in NaCl/PBS (Invitrogen, Groening, The Netherlands) in a measuring cylinder and the erythrocytes were left to sediment for 60 min in room temperature. The granulocyte-rich plasma was collected from the cylinders, and the cells were washed twice with PBS and once with 0.34 M sucrose (Merck, Darmstadt, Germany), with each wash centrifugation was performed for 10 min at 400 g at room temperature. Finally, the cells were suspended in five times the volume of 0.34 M sucrose. Three hundred milliliter cell suspension was mixed with an equal amount of cold 0.34 M sucrose and was pressurized with N2 in a nitrogen bomb (Parr Instrument Company, Moline, IL, USA) at 750 psi, 4°C with constant stirring for 30 min. The homogenate was then collected in 400 ml 0.34 M sucrose, 0.3 M NaCl, and centrifuged at 450 g for 20 min in 4°C. The supernatant was collected and centrifuged for 20 min at 10,000 g at 4°C. The pellet containing granules was collected and frozen at –70°C until further purification.

Extraction of granules
The granule preparations were thawed and 5 volumes of 50 mM HAc were added, extraction was performed at 4°C for 1 h with constant stirring. The double amount of 0.4 M NaAc (pH 4.0) was added, and the granules were extracted for another 3 h at 4°C with constant stirring. The extract was centrifuged at 12,000 g for 30 min at 4°C, and the supernatant was collected. The granule extract was then concentrated to ~3 ml using YM-10 filters (Amicon Corporation, Lexington, MA, USA).

ECP purification by gel filtration and ion exchange chromatography
Gel filtration chromatography was performed using the Sephadex G-75 superfine column (GE Healthcare Biosciences) calibrated with 0.2 M NaAc (pH 4.5). The eluted fractions were divided into 10 pools of which 2 contained ECP (as shown in Fig. 1A ). The first ECP containing G-75 pool contained high molecular weight ECP (HMW-ECP), and the second contained low molecular weight ECP (LMW-ECP). The eluted fractions were kept at –70°C until further purification. Ion-exchange chromatography was performed using the ÄKTAprime system and a Mono-S column (GE Healthcare Biosciences). The column was equilibrated with 50 mM MES (Merck), 2% Betaine (Sigma-Aldrich, St. Louis, MO, USA), 0.1 M LiCl (Merck) (pH 6.0), and the proteins were eluted with a linear gradient from 0.1 to 2.0 M LiCl (pH 6.0). Eluted fractions were pooled together according to the chromatograms in Fig. 1B 1C .


Figure 1
View larger version (41K):
[in this window]
[in a new window]

  		Figure 1. (A–C) Gel filtration and ion exchange chromatography of the granule extract from buffy coats. (A) Gel filtration of the granule extract (corresponding to ~150–200 buffy coats) on a Sephadex G-75 superfine column. The absorbance of the proteins was measured at 280 nm. Eluted proteins in peaks were pooled together according to the dotted line in the chromatogram. Two pools contained ECP; first pool contained high molecular weight ECP (HMW-ECP), and the second pool low molecular weight ECP (LMW-ECP). (B) Ion exchange chromatography of the HMW-ECP pool on a Mono-S column. The absorbance of the proteins was measured at 280 nm. Proteins in peaks were pooled together according to the dotted lines in the chromatogram (III–VII). The hatched line shows the salt gradient expressed as conductivity. (C) Ion exchange chromatography of the LMW-ECP pool on a Mono-S column. The absorbance of the proteins was measured at 280 nm. Proteins in peaks were pooled together according to the dotted lines in the chromatogram (III–V). The hatched line shows the salt gradient expressed as conductivity. (D) Gel electrophoresis of the HMW-ECP fractions. Samples were analyzed on a NuPAGE® 10% bis-Tris gel (with the SeeBlue® Plus2 Pre-Stained Standard as a molecular weight marker. First lane from the left is the molecular weight marker, lanes to follow show HMW-ECP III–VII in that order. (E) Gel electrophoresis of the LMW-ECP fractions. Samples were analyzed on a NuPAGE® 10% Bis-Tris Gel (Invitrogen) with the SeeBlue Plus2 Pre-Stained Standard (Invitrogen) as a molecular weight marker. First lane from the left is the molecular weight marker, lanes to follow show LMW-ECP III–V in that order.

 
Protein determination and concentration
The purity of ECP in the pools was assessed by analysis on NuPAGE® 10% Bis-Tris Gels (Invitrogen) (Fig. 1D 1E) , and the purity was >95%. The concentration of ECP was determined by the absorbance at 280 nm using the extinction coefficient (E1%.1cm) of ECP of 15.45 [21 ] and by immunoassay in the UniCAP system (Phadia, Uppsala, Sweden). ECP-containing fractions were concentrated using YM-10 filters (Amicon) and buffer change was accomplished using the ÄKTAprime system (GE Healthcare Biosciences). Purified ECP was stored in 0.2 M NaAc buffer (pH 5.5) at –70°C.

Enzymatic deglycosylation of ECP
Five micrograms of ECP in 0.2 M NaAc (pH 5.5) was incubated without or with 5 units of either N-glycosidase F (Roche Diagnostics, Penzberg, Germany), 2.5 mU endoglycosidase H (Roche) or 2.5 mU O-glycosidase (Sigma-Aldrich) at 37°C overnight, according to the instructions of the manufacturer. Deglycosylation was also performed with the E-DEGLY kit from Sigma-Aldrich, according to the instructions from the supplier. The kit contained enzymes PNGase F, O-glycosidase, {alpha} -2(3,6,8,9) neuraminidase, β-1,4-galactosidase and β-N-acetylglucosaminidase. Shortly, 2 µg of ECP was incubated with 35 µl MQ water, 10 µl reaction buffer, and 0.5 µl of each enzyme or different enzyme combinations in 37°C for 24 h.

As a control, 2 µg of the ECP sample was incubated with only 35 µl of MQ water and 10 µl of reaction buffer at 37°C for 24 h.

Preparation of granulocyte suspension and eosinophil isolation
Eosinophils were isolated from 80 ml of peripheral venous blood from each donor using the magnetic cell separation system (MACS) (Miltenyi Biotec, Bergisch-Gladbach, Germany), as described by Hansel et al. [22 ]. Venous heparinized blood was diluted 1:1 in PBS before loaded on 67% Percoll (density 1.085 g/ml) (GE Healthcare Biosciences). After centrifugation at 1000 g for 30 min, mononuclear cells were removed and the erythrocytes in the cell pellet were hypotonically lysed for 60 s. The obtained PMN (polymorphonuclear leukocytes) were incubated for 1 h at 4°C with anti-CD16 beads (Miltenyi Biotec), and the suspension was placed in a MACS column for the final separation. Eosinophil purity was > 95% in all cases. The study was approved by the Medical Ethical Committee at Uppsala Academic Hospital, and all blood donors gave a written informed consent.

Cell counting and quantification
Cells were stained with Türk’s dye and counted under the light microscope. Cell viability was measured using Trypan blue exclusion. Total blood cell counts were performed with a Technicon H1 (Tournai, Belgium) cell counter. Differential counts were obtained using a cytocentrifuge preparation (Cytospin, Shandon Southern Instruments, Sewickley, PA, USA) stained with May Grünewald and Giemsa, then examined under the light microscope.

Release and extraction of ECP from eosinophils
Serums-opsonization of Sephadex G-15 particles (GE Healthcare Biosciences) was performed as described [23 ], with some modification. 42 mg Sephadex G-15 particles were dissolved in 250 µl PBS. NHS (normal human serum) (50%, vol/vol) was added and the particles were incubated at 37°C for 10 min. After centrifugation at 1000 rpm for 10 min, the particles were washed twice in PBS and resuspended in 500 µl Gey’s buffer (pH 7.4).

The eosinophils were pelleted by centrifugation for 10 min at 600 g at 4°C and were then resuspended in Gey’s buffer (pH 7.4). Cell suspension was divided in two parts: unstimulated and stimulated cells. Unstimulated cells were kept on ice, and 200 µl of the serum-opsonized particles were added to the cells to be stimulated. The stimulated cells were incubated for 30 min at 37°C with careful mixing every 10 min. The cells were put on ice to stop the reaction and were then centrifuged for 10 min at 600 g in 4°C. The supernatant was collected, and the granule proteins were extracted from the remaining cell pellet of the stimulated cells and from the unstimulated cells by addition of 100 µl 0.5% CTAB (cetyl-N,N,N,trimetylammoniumbromide) (Merck), and 0.9% NaCl for 1 h in room temperature.

Affinity capture assay with SELDI-TOF MS
QC of antibodies
Three monoclonal anti-ECP antibodies were used in the study, the commercially available EG2 and the 614 and 652 clones provided by Diagnostics Development (Uppsala, Sweden). Quality control of the antibodies was performed on NP20 ProteinChip® arrays (Bio-Rad Laboratories, Hercules, CA, USA) with saturated sinapinic acid (SPA) (Bio-Rad Laboratories) dissolved in 0.5% trifluoroacetic acid (TFA) (Sigma-Aldrich) and 50% acetonitrile (ACN) (Merck, Darmstadt, Germany) as a matrix molecule. Arrays were analyzed in a PBS-IIC system (Bio-Rad Laboratories) to check for intact antibodies.

Affinity Capture assay
PS20 ProteinChip® arrays (Bio-Rad Laboratories) were assembled in a bioprocessor, and 0.6 µg antibody in a 10 µl volume of PBS was added to each spot. The arrays were then incubated in a humidity chamber in room temperature for 2.5 h. Twenty-five microliters of 0.5 M ethanolamine (Merck) in PBS, pH 8.0 was added to each well to block unspecific binding and the arrays were incubated for another 30 min in RT with mild shaking. The bioprocessor was inverted to discard the contents of the wells, and 100 µl of 0.5% Triton X100 (Merck) in PBS was added to each well and then incubated with mild shaking in RT for 5 min. The contents of the wells were discarded, and the Triton wash was repeated twice. One-hundred microliters of PBS was added to each well and incubated with mild shaking for 5 min in RT. The contents of the wells were discarded, and the PBS wash was repeated twice. Samples (purified ECP of 2 or 5 µg or eosinophil extracts of 1–2x106 cells) were added to the wells, and PBS was added to make a total volume of 50 µl, and then the arrays were incubated in 4°C overnight with mild shaking.

The following morning, the contents of the wells were discarded, and the arrays were then washed with 100 µl 0.5% Triton X100 in PBS with mild shaking for 5 min in 4°C. The contents of the wells were discarded, and the Triton wash was repeated two times. The arrays were then washed with 100 µl PBS and incubated in 4°C for 5 min with mild shaking. The PBS wash was repeated twice. Finally, the arrays were washed with 200 µl of 1 mM HEPES buffer. The arrays were taken out of the bioprocessor and were allowed to dry in room temperature for ~15 min, after which 0.6 µl of saturated SPA in 0.5% TFA and 50% ACN was added twice to each spot on the array. The arrays were then analyzed in the PBS-IIC instrument (Bio-Rad Laboratories), and a total of 192 transients were collected from each spot.

Two eosinophil extracts were treated as above mentioned, but instead of SPA, a saturated solution of 50% {alpha}-cyano-4-hydroxycinnamic acid (CHCA) (Bio-Rad Laboratories) was added.

Data analysis
Background was deducted, and all mass spectra were externally calibrated with the all-in-one protein standard (Bio-Rad Laboratories) using three calibrants covering the mass region of interest.

Genotyping of the ECP gene was performed by TaqMan analysis as described earlier [19 ].


   RESULTS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
DISCUSSION
 REFERENCES
 
Development of ECP affinity capture assay
Three different monoclonal anti-ECP antibodies were used for the development of an Affinity Capture assay with the PS20 ProteinChip arrays (Bio-Rad Laboratories); the commercially available mabs EG2, 614 and 652 clones provided by Diagnostics Development (Uppsala, Sweden). Quality control of the mabs was performed to make sure that they were intact. All three antibody suspensions contained intact antibodies with little dissociation into heavy and light chain and few contaminants (data not shown). All three mabs were selected for further use in the development of an ECP affinity capture assay with PS20 arrays.

Eosinophil extracts were analyzed by the three anti-ECP mabs on PS20 arrays, and Fig. 2 shows a representative spectrum of an eosinophil extract captured by the anti-ECP mabs and also the capture of the eosinophil extract by purified bovine IgG (Sigma). The bovine IgG generated no peaks in the 16–20 kDa region. All anti-ECP mabs generated peaks corresponding in mass to ECP, and the same molecular variants of ECP were detected with the three mabs. However, the signal/noise relation was superior with the 614 mab. The signal/noise relation for all three antibodies was 614>652>EG2. Hence, the 614 mab was selected for the affinity capture assay.


Figure 2
View larger version (23K):
[in this window]
[in a new window]

  		Figure 2. Affinity capture of ECP from one eosinophil extract using purified bovine IgG and three monoclonal anti-ECP antibodies; EG2, 652, and 614. Distinct peaks of ECP are evident with all three anti-ECP mabs, but no peaks corresponding to ECP were generated by the bovine IgG. The 614 clone generated peaks of highest intensity and had the highest signal/noise ratio.

 
Affinity capture of ECP from eosinophil extracts with the 614 mab
Eosinophil extracts from 20 blood donors were analyzed with the affinity capture assay. Several different forms of ECP were detected in each eosinophil extract, demonstrating the heterogeneity of ECP in individuals. Also, five supernatants and pellet extracts of serum-opsonized Sepadex particle-stimulated eosinophils were analyzed. Figure 3A displays the spectra of one eosinophil extract; unstimulated cells, supernatant, and pellet extract of stimulated cells. In each eosinophil extract of unstimulated cells, up to five clear peaks were detected with the 614 antibody, with masses ranging from ~15.9 kDa to 17.2 kDa. Two peaks with molecular weights at ~16.1 and 16.3 kDa were the predominant peaks in the extracts of all 20 subjects. The same pattern of peaks was detected in the supernatants and pellets of stimulated cells indicating the absence of molecular processing.


Figure 3
View larger version (28K):
[in this window]
[in a new window]

  		Figure 3. (A) ECP capture with the 614 mab from one eosinophil extract; total extract of unstimulated cells, supernatant, and pellet extract of stimulated cells. Approximately five different peaks corresponding to ECP were detected in eosinophil extract; the same molecular species were found both in the supernatant and pellet extract. (B) ECP capture of an eosinophil extract with two different matrix molecules: SPA and CHCA. Top spectrum displays the eosinophil extract analyzed with SPA matrix; bottom spectrum is the same eosinophil extract analyzed with CHCA matrix. Both matrix molecules gave rise to several peaks corresponding to molecular variants of ECP. The two spectra had similar appearances with the same peaks, but analysis with the CHCA matrix generated a spectrum with lower resolution. Five clear peaks were detected in the analysis with SPA; the same five peaks were evident in the analysis with CHCA, although with much poorer resolution. (C) ECP capture with the 614 mab of three eosinophil extracts, a zoom in on the 16 kDa region illustrates the two-peak pattern depending on ECP 434(G>C) genotype, homozygous subjects have one peak and heterozygous subjects two peaks ~50 Da apart. Top spectrum is from a homozygous GG subject (97arg), middle spectrum from a heterozygous subject (97arg/thr), and bottom spectrum is from a homozygous CC subject (97thr). Please note that the scale of peak intensity (y-axis) differs in the spectra.

 
Two eosinophil extracts of unstimulated cells were analyzed with the alternative matrix molecule CHCA in order to investigate the possibility of cluster ions due to matrix adducts. Figure 3B displays one eosinophil extract analyzed with both SPA and CHCA. Similar peak patterns were observed with the two matrix molecules, but the resolution was poorer with CHCA, and more double charged species were evident in the spectra. Analysis with SPA generated ~5 peaks corresponding to molecular variants of ECP; analysis with CHCA also generated 5 peaks. After a comparison of the SPA and CHCA spectra, the five peaks in each spectrum were estimated to correspond to the same five molecular variants of ECP. Thus, the peak patterns are most likely due to differences in post-translational modifications by carbohydrate molecules and not due to cluster ions by matrix adducts.

In some eosinophil extracts the peak at 16.1 kDa was divided in two peaks ~50 Da apart. A closer look at the peak pattern around 16.1 kDa in all extracts showed that some extracts only had the peak at 16.1 kDa, and some had both peaks ~50 Da apart, and two extracts had only the peak at ~16.05 kDa. A mass difference of ~50 Da in-between the two peaks could correspond to the 55 Da difference in mass between amino acids arginine and threonine, the consequence of the ECP 434(G>C) gene polymorphism (arg97thr). Therefore, all eosinophil donors were genotyped for the ECP 434(G>C) polymorphism by TaqMan analysis, and genotype results were compared with the results from the affinity capture assay. Subjects that were heterozygous for the ECP 434(G>C) polymorphism had two peaks in their mass spectra around 16.1 kDa, subjects that were homozygous GG (97arg) had one peak at 16.1 kDa, and homozygous CC subjects (97thr) had one peak at ~16.05 kDa in the mass analysis. Figure 3C shows a zoom in on the 16.1 kDa area in the mass spectra of eosinophil extracts from one blood donor of each ECP 434(G>C) genotype. The same tendency of a two-peak pattern was found at the 16.3 kDa mass region, where subjects that were homozygous G (97arg) had a peak at ~16300 Da, and subjects that were homozygous C (97thr) had a peak at ~16250 Da.

Affinity capture of ECP from purified ECP fractions using the 614 mab
Native ECP fractions purified from pooled buffy coats of blood donors were analyzed by the ECP affinity capture assay. The eight ECP fractions of highest concentration were selected for the capture experiments; HMW-ECP III-VII and LMW-ECP III-V. Figure 4A shows the ECP contents of HMW-ECP III-VII, and Fig. 4B shows the ECP contents of LMW-ECP III-V. The figures illustrate the great heterogeneity of ECP, in all ~10 different forms of ECP with molecular weights ranging from 15.8 to 17.4 kDa were captured by the antibody. As expected, the largest ECP molecules were found in the first ECP containing G-75 pool. Molecular weight decreased in the later fractions; the most cationic forms of the protein had the lowest mass. Some variants of ECP were present in several fractions and were found in both G-75 pools. Early fractions were more heterogeneous with fraction HMW-ECP III containing five different molecular species as compared with fraction HMW-ECP VII with only two. The same pattern was seen in fractions from the second ECP containing G-75 pool, where fraction LMW-ECP III contained five molecular species of ECP, whereas fraction LMW-ECP V contained only two.


Figure 4
View larger version (22K):
[in this window]
[in a new window]

  		Figure 4. (A) Affinity capture with the 614 mab. Mono-S fractions from the HMW- ECP are shown. The same amount of protein (5 µg) was used for all experiments. Please note that the scale of peak intensity (y-axis) differs in the spectra. (B) Affinity capture with the 614 mab. Mono-S fractions from the LMW-ECP are shown. The same amount of protein (5 µg) was used for all experiments. Please note that the scale of peak intensity (y-axis) differs in the spectra.

 
Affinity capture of ECP from deglycosylated ECP fractions using the 614 mab
To study the post-translational heterogeneity of ECP further, the native ECP fractions from buffy coats were deglycosylated by different combinations of enzymes before MS analysis. Again, fractions HMW-ECP III-VII and LMW-ECP III-V were selected due to high ECP contents. Figure 5 illustrates the HMW-ECP III-VII and LMW-ECP III-V fractions in a gel view; undigested samples are displayed together with samples that have been deglycosylated by the N-glycosidase F enzyme. Deglycosylation with the N-glycosidase F enzyme resulted in a clear shift in mass in all fractions, but HMW-ECP VII and LMW-ECP V. In these two fractions, only weak partial deglycosylation occurred, the major peak was seemingly unaltered, but a small new peak at ~15.9 kDa was generated. The degree of N-linked glycosylation varied greatly between the fractions, with the fractions containing ECP of the highest molecular mass also being the ones most heavily N-glycosylated, up to 1.7 kDa was cleaved from the protein. In all studied fractions but HMW-ECP VII and LMW-ECP V did treatment with the N-glycosidase F enzyme result in a clear peak corresponding to the mass of ECP without any posttranslational modifications (~15.5 kDa) [24 ]; however, there was a hint of this peak in the HMW-ECP VII fraction. In all fractions, there were still also peaks of higher mass than 15.5 kDa after removal of N-linked carbohydrates, indicating there are some other modifications on the protein as well. Enzymes endoglycosidase H (digests N-glycans of high mannose type), and O-glycosidase did not digest any sugars from any ECP fraction, even though treatment of bovine fetuine protein with these enzymes and analysis on SDS-PAGE showed that they were enzymatically active (data not shown).


Figure 5
View larger version (49K):
[in this window]
[in a new window]

  		Figure 5. Affinity capture of deglycosylated ECP fractions using the 614 mab. Gel view of HMW-ECP III–VII and LMW-ECP III–V fractions. Samples are shown pairwise with untreated sample on top of sample deglycosylated by the N-glycosidase F enzyme. Enzyme treatment resulted in a mass shift in all fractions but HMW-ECP VII and LMW-ECP V. Please note that the scale of peak intensity (y-axis) differs in the spectra, resulting in different strength of the gel bands.

 
The HMW-ECP III-VII fractions were further analyzed by deglycosylation with the E-DEGLY kit (Sigma-Aldrich), and the results are summarized in Table 1 . Again, N-linked glycosylation was evident when the fractions were digested by PNGase F, digestion occurred in all fractions but HMW-ECP VII. After deglycosylation the major peak in all fractions but HMW-ECP VII was ~15.7 kDa. Also, a peak corresponding to the mass of ECP without any posttranslational modifications (~15.5 kDa) was generated in all fractions but HMW-ECP VII. The same results were generated when the fractions were treated with all enzymes in the kit at once. Deglycosylation with other enzymes than PNGase F were successful in two fractions; HMW-ECP IV, and HMW-ECP V, as is shown in Table 1 . The deglycosylation of the HMW-ECP IV fraction is shown in Fig. 6A 6B . Figure 6A displays the mass spectra of the stepwise deglycosylation, and Fig. 6B displays the same data in a gel view. Treatment with the {alpha}-2(3,6,8,9)-neuraminidase enzyme resulted in the digestion of 283 Da as a mean, corresponding to sialic acid with a known molecular weight of 291 Da. Treatment with the {alpha}-2(3,6,8,9)-neuraminidase enzyme in combination with β-N-acetylglucosaminidase resulted in no additional cleavage; however the {alpha}-2(3,6,8,9)-Neuraminidase enzyme in combination with β-1,4-galactosidase resulted in the digestion of 446 Da as a mean, which could correspond to the additional cleavage of one molecule of galactose with the known molecular weight of 162 Da. The combination of enzymes {alpha}-2(3,6,8,9)-neuraminidase, β-N-acetylglucosaminidase and β-1,4-galactosidase resulted in additional removal of carbohydrates, a total of 645 Da as a mean was digested off the protein. This could correspond to one molecule of sialic acid, one molecule of galactose, and one molecule of acetylglucosamine with the molecular weight of 203 Da. In the HMW-ECP V fraction, partial deglycosylation was seen with the combination of enzymes {alpha}-2(3,6,8,9)-neuraminidase, β-N-acetylglucosaminidase, and β-1,4-galactosidase, the major peak was still 17.2 kDa after deglycosylation, but a peak at ~16.6 kDa was also apparent. These ~600 Da cleaved off the protein could correspond to any combination of sialic acid, galactose, or acetylglucosamine.


View this table:
[in this window]
[in a new window]

  		Table 1. Summary of MS Analysis of ECP Fractions HMW-ECP III-VII After Enzymatic Deglycosylation With the E-DEGLY Kit and Endo H Enzyme

 

Figure 6
View larger version (35K):
[in this window]
[in a new window]

  		Figure 6. (A) Stepwise deglycosylation of HMW-ECP IV by the enzymes of the E-DEGLY kit. Top spectrum is the glycosylated HMW-ECP IV (untreated sample), and the following spectra are the HMW-ECP IV fraction treated with different enzyme combinations. Please note that the scale of peak intensity (y-axis) differs in the spectra. (B) Stepwise deglycosylation of the HMW-ECP IV fraction by the enzymes of the E-DEGLY kit, mass spectra are displayed in a gel view. Please note that the scale of peak intensity (y-axis) differs in the spectra, resulting in different strength of the gel bands.

 

   DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
REFERENCES
 
In this study, a new method for studying ECP heterogeneity was developed; affinity capture on Protein Chip Arrays, using SELDI-TOF MS technology. Three anti-ECP mabs were tested for the assay, and the 614 mab was selected due to superior signal/noise ratio. The EG2 antibody has been shown previously to react with a lower affinity with glycosylated variants of ECP [25 ], which may explain its inferiority in the SELDI capture assay. It was discovered that the 614 mab used in the assay could distinguish the two various proteins of the ECP 434(G>C) gene polymorphism (arg97thr), with two peaks ~50 Da apart at ~16.1 kDa in heterozygous subjects and either one peak in homozygous subjects. This ~50 Da difference of the two allele products resulting in two almost identical peaks close in mass was clearly seen at the 16.1 kDa peak in all eosinophil extracts, but theoretically this two-peak pattern should be there for all molecular variants of ECP and was indeed also observed at the 16.3 kDa region. However, the somewhat poor resolution rendered it impossible to observe this pattern for all molecular species.

When eosinophils of healthy subjects were treated with serum-opsonized particles to facilitate ECP secretion, MS analysis of the ECP containing supernatant and pellet extract demonstrated that the same molecular variants were excreted as remained in the cell. Although these results suggest that the cell releases unmodified ECP upon activation with complement-coated particles, other mechanisms such as in vivo priming might result in the release of modified molecular species. The question as to why there are so many molecular species of ECP still remains and work is now in progress to assess possible functional differences. Previous results have indeed suggested that different molecular variants of ECP have different functions in terms of cytotoxicity [14 ]. Trulson et al. demonstrated that the recombinant variants of the ECP 434(G>C) polymorphism, ECP97arg and ECP97thr differed in their cytotoxic properties toward a mammalian cell line. The threonine-containing ECP was noncytotoxic; however, when it was deglycosylated it regained the cytotoxic properties, implicating a functional role of the carbohydrates. In a study by Kim et al. [26 ], recombinant glycosylated onconase, an RNase, was produced and the importance of carbohydrates in protein function was again demonstrated; the glycosylated onconase was found to have a 50-fold increase in the cytotoxic activity toward a cancer cell line as compared with the nonglycosylated protein. Other studies on members of the RNase A family also suggest the possibility of functional roles of carbohydrates; in a study by Ye et al. [27 ], a glycosylated form of EPX/EDN was found to be significantly elevated in ovarian cancer patients.

To fully understand the purpose of such a molecular diversity as is the case in ECP, it is important to study the heterogeneity in detail and explore whether a certain molecular variant is linked to a particular trait or function. The affinity capture assay based on SELDI-TOF technology developed in this study provides a useful tool to further explore the molecular heterogeneity of ECP in eosinophils involved in diseases such as allergy and asthma.

Approximately ten different molecular species of ECP were detected when ECP purified from buffy coats was studied, but when ECP of eosinophil extracts of individual subjects were examined, only five different molecular species were detected. Clearly, fractionation of the sample, as is the case with ECP from buffy coats, facilitates the identification of molecular variants of the protein that are close in mass.

A study by Dijkstra et al. [28 ] demonstrated the presence of nonspecific cluster ions when using SPA in SELDI analysis of myoglobin. In their study, a repetitive pattern of peaks 206 Da apart was detected at the 17 kDa region, and the peaks corresponded to myoglobin with 0, 1, 2, or 3 SPA adducts; so-called nonspecific cluster ions. Since the two major peaks that we detected when we analyzed eosinophil extracts with SPA were close in mass (in average 202 Da) and this difference in mass, in theory, could correspond to one SPA adduct, the question of whether these peaks were molecular variants of ECP or possibly nonspecific cluster ions was raised. To examine whether these peaks were, in fact, cluster ions of ECP and SPA, two eosinophil extracts were analyzed with CHCA as matrix (molecular weight of 189.2 Da) instead of SPA. If the major peak at 16.3 kDa observed in the eosinophil extract indeed, was a nonspecific cluster ion of the 16.1-kDa peak, then this peak would disappear in the analysis using CHCA as the matrix. Possibly, if CHCA also forms cluster ions with ECP, the cluster ion would be +189.2 Da (molecular weight of CHCA) from the 16.1-kDa peak. However, the same pattern of peaks was generated when using the CHCA matrix as when using SPA, but a poorer resolution was obtained. The two peaks of highest peak intensity were still the peaks at ~16.1 and 16.3 kDa, differing 202 Da on average. Hence, these two peaks should be molecular variants of ECP and not cluster ions due to SPA. The difference in mass in between adjacent peaks (delta mass) was compared for both spectra in Fig. 3B and found to be the same for both spectra, hence, no peaks corresponded to cluster ions due to matrix adducts.

Deglycosylation experiments to study ECP heterogeneity further showed that N-linked carbohydrates were the most common modification of ECP, but that there were other modifications as well. These findings verify experiments by others in which similar deglycosylation experiments, though under denaturing conditions, have been performed [9 ]. However, another study of ECP purified from peripheral leukocytes where ECP was deglycosylated by PNGase F at denaturing conditions showed that the protein was completely deglycosylated by this treatment [11 ]. Incubation with the enzyme endoglucosidase H, which digests N-glycans of high mannose type did not result in any cleavage at all, alone or in combination with the other enzymes. This again confirms what other authors have found, although those studies were performed at denaturing conditions [9 , 11 ] and on human recombinant ECP [10 ]. Stepwise deglycosylation of ECP with the enzymes of the E-DELGY kit revealed the presence of sialic acid, as suggested before [24 ], and galactose and acetylglucosamine were also found to be present on the ECP molecule. Simultaneous treatment of the ECP fractions with all of the enzymes of the E-DEGLY kit resulted in the same peaks as with treatment with only the PNGase F enzyme. This indicates that all carbohydrate constituents demonstrated by step-wise deglycosylation are present on the same chain, and treatment with the PNGase F enzyme would in fact remove the complete chain. In a few fractions only partial deglycosylation was evident after incubation with the enzymes. This is most likely explained by the fact that these fractions only contained a very small portion of glycosylated ECP rather than lack of enzymes in the experiments, since enzymes were added in excess. No O-linked carbohydrates could be cleaved off the protein by treatment with O-glycosidase of any ECP fraction, which might indicate that no such glycans are present. However, extended incubation might be needed for a noticeable effect, and there is the possibility that there is some steric hindrance simply preventing the O-glycosidase enzyme to reach its substrate. Hence, the possibility that O-linked carbohydrates are present on some ECP variants still exists.

We speculate that there might be an additional site for glycosylation, either O-linked or N-linked, with the 434C-allele and to test this hypothesis ECP needs to be purified from individuals of the different genotypes. This is an ongoing project in our group.

SELDI-TOF technology provides a useful complement to other proteomic tools and the affinity capture method is a simple and relatively high throughput method as compared with other MS techniques. Sample purification is not needed since the assay is based on an antibody-antigen interaction giving an effective capture of the target protein, and also, the sample is studied in its native form making sample preparation easy. We conclude that the SELDI-TOF technology is well suited for studying protein heterogeneity in general. The affinity capture assay designed here provides a simple and accurate tool to study the molecular heterogeneity of ECP in eosinophils from individual subjects after activation and priming of the cells in vitro or in vivo in conditions such as allergy and asthma.


   ACKNOWLEDGEMENTS
 
The expert technical assistance of Ms. Lena Moberg is greatly appreciated, and the authors thank all volunteering blood donors in the study. This study was supported by grants from the Swedish Medical Research Council, the Swedish Heart and Lung Foundation, and the Swedish Association against Asthma and Allergy.

Received May 3, 2007; revised July 12, 2007; accepted July 17, 2007.


   REFERENCES
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 

  1. Wagner, L., Leiferman, K., Gleich, G. J. (2006) Eosinophils Encyclopedia of Life Sciences John Wiley & Sons Hoboken, NJ.
  2. Kay, A. B. (2005) The role of eosinophils in the pathogenesis of asthma Trends Mol. Med. 11,148-152[CrossRef][Medline]
  3. Holgate, S. T., Polosa, R. (2006) The mechanisms, diagnosis, and management of severe asthma in adults Lancet 368,780-793[CrossRef][Medline]
  4. Zagai, U., Skold, C. M., Trulson, A., Venge, P., Lundahl, J. (2004) The effect of eosinophils on collagen gel contraction and implications for tissue remodelling Clin. Exp. Immunol. 135,427-433[CrossRef][Medline]
  5. Pegorier, S., Wagner, L. A., Gleich, G. J., Pretolani, M. (2006) Eosinophil-derived cationic proteins activate the synthesis of remodeling factors by airway epithelial cells J. Immunol. 177,4861-4869[Abstract/Free Full Text]
  6. Humbles, A. A., Lloyd, C. M., McMillan, S. J., Friend, D. S., Xanthou, G., McKenna, E. E., Ghiran, S., Gerard, N. P., Yu, C., Orkin, S. H., et al (2004) A critical role for eosinophils in allergic airways remodeling Science 305,1776-1779[Abstract/Free Full Text]
  7. Venge, P., Bystrom, J., Carlson, M., Hakansson, L., Karawacjzyk, M., Peterson, C., Seveus, L., Trulson, A. (1999) Eosinophil cationic protein (ECP): molecular and biological properties and the use of ECP as a marker of eosinophil activation in disease Clin. Exp. Allergy 29,1172-1186[CrossRef][Medline]
  8. Hamann, K. J., Ten, R. M., Loegering, D. A., Jenkins, R. B., Heise, M. T., Schad, C. R., Pease, L. R., Gleich, G. J., Barker, R. L. (1990) Structure and chromosome localization of the human eosinophil-derived neurotoxin and eosinophil cationic protein genes: evidence for intronless coding sequences in the ribonuclease gene superfamily Genomics 7,535-546[CrossRef][Medline]
  9. Gleich, G. J., Loegering, D. A., Bell, M. P., Checkel, J. L., Ackerman, S. J., McKean, D. J. (1986) Biochemical and functional similarities between human eosinophil-derived neurotoxin and eosinophil cationic protein: homology with ribonuclease Proc. Natl. Acad. Sci. USA 83,3146-3150[Abstract/Free Full Text]
  10. Domachowske, J. B., Dyer, K. D., Adams, A. G., Leto, T. L., Rosenberg, H. F. (1998) Eosinophil cationic protein/RNase 3 is another RNase A-family ribonuclease with direct antiviral activity Nucleic Acids Res. 26,3358-3363[Abstract/Free Full Text]
  11. Tiffany, H. L., Li, F., Rosenberg, H. F. (1995) Hyperglycosylation of eosinophil ribonucleases in a promyelocytic leukemia cell line and in differentiated peripheral blood progenitor cells J. Leukoc. Biol. 58,49-54[Abstract]
  12. Zhang, J., Rosenberg, H. F. (2000) Sequence variation at two eosinophil-associated ribonuclease loci in humans Genetics 156,1949-1958[Abstract/Free Full Text]
  13. Jonsson, U. B., Bystrom, J., Stalenheim, G., Venge, P. (2002) Polymorphism of the eosinophil cationic protein-gene is related to the expression of allergic symptoms Clin. Exp. Allergy 32,1092-1095[CrossRef][Medline]
  14. Trulson, A., Bystrom, J., Engstrom, A., Larsson, R., Venge, P. (2007) The functional heterogeneity of eosinophil cationic protein is determined by a gene polymorphism and post-translational modifications Clin. Exp. Allergy 37,208-218[CrossRef][Medline]
  15. Lehrer, R. I., Szklarek, D., Barton, A., Ganz, T., Hamann, K. J., Gleich, G. J. (1989) Antibacterial properties of eosinophil major basic protein and eosinophil cationic protein J. Immunol. 142,4428-4434[Abstract]
  16. McLaren, D. J., McKean, J. R., Olsson, I., Venges, P., Kay, A. B. (1981) Morphological studies on the killing of schistosomula of Schistosoma mansoni by human eosinophil and neutrophil cationic proteins in vitro Parasite Immunol. 3,359-373[Medline]
  17. Carreras, E., Boix, E., Rosenberg, H. F., Cuchillo, C. M., Nogues, M. V. (2003) Both aromatic and cationic residues contribute to the membrane-lytic and bactericidal activity of eosinophil cationic protein Biochemistry 42,6636-6644[CrossRef][Medline]
  18. Rosenberg, H. F. (1995) Recombinant human eosinophil cationic protein. Ribonuclease activity is not essential for cytotoxicity J. Biol. Chem. 270,7876-7881[Abstract/Free Full Text]
  19. Eriksson, J, Reimert, CM, Kabatereine, NB, Kazibwe, F, Ireri, E, Kadzo, H, Eltahir, HB, Mohamed, AO, Vennervald, BJ, Venge, P. (2007) The 434(G>C) polymorphism within the coding sequence of Eosinophil Cationic Protein (ECP) correlates with the natural course of Schistosoma mansoni infection Int. J. Parasitol. 37,1359-1366[CrossRef][Medline]
  20. Byström, J. (2002) Eosinophil cationic protein, expression levels and polymorphisms. Ph.D. thesis Faculty of Medicine, Uppsala University Uppsala, Sweden.
  21. Peterson, C. G., Jornvall, H., Venge, P. (1988) Purification and characterization of eosinophil cationic protein from normal human eosinophils Eur. J. Haematol. 40,415-423[Medline]
  22. Hansel, T. T., De Vries, I. J., Iff, T., Rihs, S., Wandzilak, M., Betz, S., Blaser, K., Walker, C. (1991) An improved immunomagnetic procedure for the isolation of highly purified human blood eosinophils J Immunol. Methods. 145,105-110[CrossRef][Medline]
  23. Carlson, M., Peterson, C., Venge, P. (1993) The influence of IL-3, IL-5, and GM-CSF on normal human eosinophil and neutrophil C3b-induced degranulation Allergy 48,437-442[Medline]
  24. Barker, R. L., Loegering, D. A., Ten, R. M., Hamann, K. J., Pease, L. R., Gleich, G. J. (1989) Eosinophil cationic protein cDNA. Comparison with other toxic cationic proteins and ribonucleases J. Immunol. 143,952-955[Abstract]
  25. Rosenberg, H. F., Tiffany, H. L. (1994) Characterization of the eosinophil granule proteins recognized by the activation-specific antibody EG2 J. Leukoc. Biol. 56,502-506[Abstract]
  26. Kim, B. M., Kim, H., Raines, R. T., Lee, Y. (2004) Glycosylation of onconase increases its conformational stability and toxicity for cancer cells Biochem. Biophys. Res. Commun. 315,976-983[CrossRef][Medline]
  27. Ye, B., Skates, S., Mok, S. C., Horick, N. K., Rosenberg, H. F., Vitonis, A., Edwards, D., Sluss, P., Han, W. K., Berkowitz, R. S., et al (2006) Proteomic-based discovery and characterization of glycosylated eosinophil-derived neurotoxin and COOH-terminal osteopontin fragments for ovarian cancer in urine Clin. Cancer Res. 12,432-441[Abstract/Free Full Text]
  28. Dijkstra, M., Vonk, R. J., Jansen, R. C. (2007) SELDI-TOF mass spectra: a view on sources of variation J. Chromatogr. B Analyt. Technol. Biomed. Life Sci. 847,12-23[CrossRef][Medline]




This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow All Versions of this Article:
jlb.0507272v1
jlb.0507272v2
82/6/1491    most recent
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Eriksson, J.
Right arrow Articles by Venge, P.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Eriksson, J.
Right arrow Articles by Venge, P.

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS