Published online before print January 12, 2005
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
1-acid glycoprotein, which is transiently deposited in human myocardial infarction
,
,
,,,,||
* Departments of Molecular Cell Biology & Immunology, Glycoimmunology Group,
Pathology, and
Clinical Chemistry, and
Institute for Cardiovascular Research, VU University Medical Centre, Amsterdam, the Netherlands;
¶ Division of Cell Biology, the Netherlands Cancer Institute, Amsterdam; and
|| Department of Immunopathology, Sanquin Research at CLB and Laboratory for Experimental and Clinical Immunology, Academic Medical Center, University of Amsterdam, the Netherlands
1 Correspondence: Glycoimmunology Group, Department of Molecular Cell Biology & Immunology, VU University Medical Center, Van der Boechorststraat 7, P.O. Box 7057, 1007 MB Amsterdam, the Netherlands. E-mail: w.vandijk{at}vumc.nl
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
1-Acid glycoprotein (AGP) is a major acute-phase protein present in human plasma as well as in polymorphonuclear leukocytes (PMN). In this report, we show that PMN synthesize a specific glycoform of AGP, which is stored in the specific and azurophilic granules. Activation of PMN results in the rapid release of soluble AGP. PMN AGP exhibits a substantially higher apparent molecular weight than plasma AGP (50–60 kD vs. 40–43 kD), owing to the presence of strongly fucosylated and sialylated polylactosamine units on its five N-linked glycans. PMN AGP is also released in vivo from activated PMN, as appeared from studies using well-characterized myocard slices of patients that had died within 2 weeks after an acute myocardial infarction. AGP was found deposited transiently on damaged cardiomyocytes in areas with infiltrating PMN only. It is interesting that this was inversely related to the deposition of activated complement C3. Strongly fucosylated and sialylated AGP glycoforms have the ability to bind to E-selectin and to inhibit complement activation. We suggest that AGP glycoforms in PMN provide an endogenous feedback-inhibitory response to excessive inflammation.
Key Words: activated complement C3 • molecular weight • orosomucoid • subcellular localization • secretion • specific granules • azurophilic granules
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
1-Acid glycoprotein (AGP), also known as orosomucoid, is one of the major acute-phase glycoproteins. Human plasma AGP contains five N-linked complex-type glycans, which are strongly sialylated and comprise about half of its molecular mass [10
, 11
]. The hepatic acute-phase response induces an increased 3-fucosylation of these glycans, resulting in the occurrence of AGP glycoforms rich in sialyl Lewisx groups in plasma during inflammation [12
]. It is interesting that these AGP glycoforms, by binding to E-selectin [13
] and by amelioration of complement-induced damage [14
], possibly counteract the influx of PMN in inflamed tissue. Most acute-phase proteins are produced in the liver. It is remarkable that a number of acute-phase proteins are also present in resting PMN [15 16 17 18 ]. AGP is one of them; a membrane-associated as well as an intracellular-localized form has been described for PMN [15 , 17 ]. The glycosylation process in PMN gives rise to polyfucosylated and sialyl Lewisx-expressing glycans on its cell-surface glycoproteins, which may constitute specific ligands for E- and P-selectin [19 , 20 ]. AGP produced by PMN would contain similar glycans [21 ] and hence, may serve as an endogenous feedback inhibitor of excessive recruitment of PMN by activated complement factors and the selectin-dependent processes.
In this study, we provide biochemical and electron-microscopical evidence that PMN have AGP, which is secreted rapidly upon activation in vitro as well as in vivo in patients that had died of acute myocardial infarction (AMI) [22 ].
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Patients and processing of tissue specimens
Patients, referred to the Department of Pathology, VU University Medical Centre (Amsterdam, the Netherlands) for autopsy, were included in this study when at autopsy, performed within 24 h following death, they showed signs of a recently developed AMI, as has been described in more detail in earlier studies [22
, 26
]. Permission to use material left over (including fetal material), after the usual diagnostic process has been completed, is part of the standard patient contract in our hospital. Tissue specimens were obtained from the infarcted area (i.e., the area with decreased lactate dehydrogenase staining) as well as from the adjacent, noninfarcted myocardium (internal control) of 27 patients. Infarct age, assessed by microscopical criteria [27
, 28
], varied from less than 12 h up to 2 weeks: <12 h (n=2), 12–24 h (n=5), 1–3 days (n=10), 5–9 days (n=3), and 9–14 days (n=2). The further characteristics of the patients are supplied in supplementary data to this paper. Before being prepared as cryosections, the tissue specimens were stored at –196°C (liquid N2).
Localization of AGP, di- and/or trimeric Lex structures, activated C3, and CD66b
Four µm-thick frozen sections were mounted onto glass slides, dried for 1 h by exposure to air, and fixed in acetone ("Baker Analyzed Reagent," Mallinckrodt Baker BV, Deventer, Holland). The slides were incubated at room temperature for 10 min with normal rabbit serum (Dakopatts A/S), 1:50 diluted in phosphate-buffered saline (PBS) containing 1% (w/v) bovine serum albumin (BSA) after a rinse in PBS. Incubation of the slides with specific antibody solutions was performed for 60 min at room temperature (G9 diluted 1:300 in PBS-BSA; C3-9 diluted 1:1000 in PBS-BSA; CD66b diluted 1:1000 in PBS-BSA; 6H3 diluted 1:300 in PBS-BSA). As control, irrelevant mAb IgG1 and mouse myeloma protein were tested as well. After washing with PBS, slides were incubated with horseradish peroxidase-conjugated rabbit anti-mouse Ig (Dakopatts A/S), 1:25 diluted in PBS-BSA. Thereafter, the slides were washed again in PBS and incubated for 4 min in 0.5 mg/ml 3,3'-diaminobenzidine-tetrahydrochloride (Sigma Chemical Co., St. Louis, MO) in PBS (pH 7.4), containing 0.01% (v/v) H2O2, washed again, counterstained with hematoxylin for 40 s, dehydrated, cleared, and finally, mounted. Microscopic criteria were used to estimate infarct duration in all myocardial tissue specimens [27
, 28
], as described in more detail in earlier studies [22
, 26
]. Two independent investigators (H. W. M. Niessen and R. Nijmeyer) performed these microscopic studies; for the final scoring results, two investigators achieved consensus.
Quantification of the immunohistochemical deposition of complement C3, CD66b, and AGP and statistical treatment
The extent of depositions of C3, CD66b, and AGP was determined as follows: The slides were analyzed at magnification 25x and divided in at least four parts. For each part of the slide, the percentage of positivity for the specific antibodies was analyzed. Subsequently, the average of positivity of the whole slide was calculated.
Data were analyzed with the statistical program SPSS for Windows, version 9.0 (SPSS Inc., Chicago, IL). The Mann-Whitney U-test was used to evaluate the significance of differences. A P value of <0.05 was considered to represent a significant difference.
Processing of myocardial tissue specimens for analysis of AGP mRNA
Myocardial tissue specimens from a patient, dead from a cause not related to myocardial infarction, were obtained within 12 h after decease. Specimens, taken from the left ventricle of the heart, were trimmed free of other tissue, minced into smaller pieces, and subsequently, incubated for 30 min at 37°C in a collagenase buffer (118 mM NaCl, 2.6 mM KCl, 1.2 mM KH2PO4, 1.2 mM MgSO4, 25 mM NaHCO3, 11.1 mM glucose) containing 4.7 µM CaCl2 and collagenase (Worthington type II, 121 U/ml). Thereafter, cells were strained through a cell dissociation mesh size 100 (Sigma Chemical Co.). Cells recovered consisted of >95% cardiomyocytes. These cells were stored at –80°C.
Processing of tissue specimens for isolation of AGP from infarcted and noninfarcted myocardial tissue
Specimens from infarcted and noninfarcted myocardial areas were homogenized using a Dysmembranator in Veronal buffer (0.01 M CaCl2, 0.02 M MgCl2, 0.01 M sodium diethylbarbital, 0.65 M NaCl, pH 7.4). The homogenate was centrifuged (10,000 g, 20 min), and the supernatant was used for isolation of AGP. All procedures were performed at 4°C.
AGP mRNA expression
Total RNA was extracted from PMN, cardiomyocytes, and IHH, as described by Chomczynski and Sacchi [29
]. For polymerase chain reaction (PCR), cDNA was synthesized from 4 µg total RNA by using the 3' rapid amplification of 3' cDNA ends system from Gibco, Life Technologies (Grand Island, NY). cDNA corresponding to 80 ng RNA template was used for PCR analysis. PCR reactions were performed in a final volume of 50 µl containing: Taq buffer (Promega, Madison, WI), 1.5 mM MgCl2, 0.2 mM deoxy-unspecfied nucleoside 5'-triphosphate each (Boehringer Mannheim, Mannheim, Germany), 1.25 units TaqBeads Hot Start polymerase (Promega Corp., Madison, WI), and 50 pmol each primer (Invitrogen, Carlsbad, CA). Human plasma AGP is encoded by three genes: AGP-A, AGP-B, and AGP-B'; the latter two are identical and are further referred to as AGP-B/B' [30
]. The sense primer sequences defined for AGP-A and AGP-B/B' were GACCAGTGCATCTATAACACCACC and TGATGTTTGGTTCCTACCTGGAC, respectively; the antisense primer sequence GTTCCAAACACAGAAGCTTTATTG was used for AGP-A and AGP-B/B' [30
]. For actin, the sense primer TGACGGGGTTCACCCACACTGTGCCCATCT and the antisense primer AGTCATAGTCCGCCTAGAAGCATTTGC were used as a positive control. Amplifications were carried out for 40 cycles using the following conditions: 1 min at 60°C, 1.5 min at 72°C, and 1 min at 94°C. PCR products were fractionated by electrophoresis in 1.5% (w/v) agarose gels. Quantitative real-time PCR reactions were performed with the SYBR Green method in an ABI 7900HT sequence detection system (Applied Biosystems, Foster City, CA) as detailed elsewhere [31
] using glyceraldehyde 3-phosphate dehydrogenase (GAPDH) as a reference gene. The thermal profile for all the reactions was 2 min at 50°C, followed by 10 min at 95°C, and then 40 cycles of 15 s at 95°C and 1 min at 60°C. The fluorescence monitoring occurred at the end of each cycle. Specific primers were designed for these reactions by using the computer software Primer Express 2.0 (Applied Biosystems) [31
]. The sense primers used for AGP-A, AGP-B/B', and GAPDH were CGTGCCTCCTGGTCTCAGTATG, AGAGTACCAGACCCGCCAGAAC, and CCATGTTCGTCATGGGTGTG, respectively; the antisense primers were AAACCACTTGCCAGTGATCCG, CCAGGTAGGAACCAAACATCA, and GGTGCTAAGCAGTTGGTGGTG, respectively. Data were expressed as the normalized amount of the AGP-A and AGP-B/B' gene (Nt) with respect to the expression of the endogenous reference GAPDH gene, using the formula Nt = 2–(CtAGP–CtGAPDH) [32
]. The Ct value is defined as the number of PCR cycles where the fluorescence signal exceeds the detection threshold value, which was fixed as 10 times the standard deviation of the fluorescence during the first 15 cycles and typically corresponds to 0.2 relative fluorescence units.
PMN isolation and stimulation
PMN were purified from the pooled buffy coats of blood [anticoagulated with 0.4% (w/v) trisodium citrate (pH 7.4)] obtained from 20 healthy volunteers by density gradient centrifugation in isotonic Percoll and lysis of the contaminating erythrocytes, by incubation with ice-cold isotonic ammonium chloride [33
]. Prior to the preparation of PMN cDNA, the cells were further purified by cell sorting for CD15+/CD16– cells as well as on scatter. Stimulation experiments with PMN were performed as described by Kuijpers et al. [33
]. In short, PMN (10x106/ml in Hepes medium) were preincubated in a shaking water bath at 37°C for 5 min, before incubation in the presence of cytochalasin B (CytoB; 5 µg/ml, 2 min) or platelet-activating factor (PAF; 1 µM, 5 min) and further stimulation with f-Met-Leu-Phe (fMLP; 1 µM, 15 min) at 37°C. Reactions were stopped by centrifugation (5 min, 800 g), and pellets and supernatants were stored at –80°C until further analysis.
Cryoimmunogold electron microscopy
Bone marrow cells and PMN from peripheral blood were prepared and fixed as described earlier by Kuijpers et al. [33
]. Ultrathin, frozen sections were incubated at room temperature with the indicated antibodies and 10- and/or 15-nm protein-A-gold [34
]. After immunolabeling, cryosections were embedded in a mixture of methylcellulose and uranyl acetate and were examined with a CM 10 electron microscope (Philips Eindhoven, the Netherlands). As a control, the primary antibody was replaced by an irrelevant murine or rabbit antiserum.
AGP isolation and molecular weight analysis
PMN (400x106) were incubated for 10 min at 0°C in 5 ml 50 mM Tris-HCl (pH 8), supplemented with a protease inhibitor mixture (Boehringer Mannheim) and 1% (w/v) Triton X-100, followed by centrifugation (14,000 g; Eppendorff centrifuge) for 30 min at 4°C. The obtained cell extract was stored at –20°C until further use. Extracts of 3.6 x 106 isolated human cardiomyocytes were obtained in the same way. AGP was isolated from the various sources by affinity chromatography on a 1-ml mAb-anti-AGP HiTrap column [25
]. AGP concentrations were determined by single radial immunodiffusion according to Mancini et al. [35
], using polyclonal rabbit anti-human AGP IgG for detection. Sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE; 10%) of 50 ng AGP, followed by blotting on nitrocellulose and detection of AGP by immunostaining with polyclonal rabbit anti-human AGP IgG [12
], was used to assess the apparent molecular weights.
Studies toward the nature, soluble and/or membrane-bound, of AGP in PMN
Preparations of 108 PMN of various donors were subjected to seven freeze-thaw cycles in a final volume of 5 ml PBS containing a protease-inhibitor mixture. Between each cycle, the cells were passed through a 0.15-mm needle. The final preparation was centrifuged for 1 h at 100,000 g in a Pegasus 65 centrifuge (MSE Scientific Instruments, Crawley, UK) at 4°C, and pellets and supernatant were collected. The pellets were extracted for 30 min with 5 ml 1% 3-[(3-cholamidopropyl)dimethylammonio]-1-propane sulfonate in PBS supplemented with protease inhibitors. AGP was isolated from the extract and the 100,000 g supernatant, by addition of a 0.1-ml suspension of Sepharose beads containing immobilized mAb-AGP C9 in PBS (1:1, v/v), followed by incubation in a rotating device overnight at 4°C. AGP bound to the beads was detected by 10% SDS-PAGE subsequent blotting and AGP detection, by polyclonal rabbit anti-human AGP IgG after incubation of the beads for 10 min at 100°C in 0.1 ml Laemmli sample buffer supplemented with ß-mercaptoethanol [12
].
N-Glycosidase F (PNGase F) treatment and monosaccharide analysis
Approximately 100 µg AGP, isolated in triplicate from PMN or normal human plasma, was treated with PNGase F to release the N-linked glycans. Incubations were carried out under reducing conditions in 100 µl 50 mM sodium phosphate buffer, 1% (w/v) Nonidet P-40, and 1000 U PNGase F for 1 h at 37°C. Short incubation times with PNGase F (t=0.5, 1, and 5 min) were used to obtain partial release of glycans to assess the number of glycans per molecule. Aliqots (50 ng) of AGP were treated and subjected to SDS-PAGE gels with subsequent blotting as described above. Monosaccharide analysis of acid hydrolysates of PNGase F-released glycans was performed by high pH anion-exchange chromatography with pulsed amperometric detection (HPAEC-PAD) [25
].
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
 View larger version (62K): [in a new window] |
Figure 1. Biochemical characterization and stimulus-dependent secretion of AGP present in PMN. Western blots of 10% SDS-PAGE gels are presented of AGP preparations obtained by immunoprecipitation with mAb-anti-AGP C9 of PMN extracts, 100,000 g supernatants, and secretion media. AGP was detected on the blots by staining with polyclonal anti-AGP (see Materials and Methods for details; M=prestained marker proteins). (a) Deglycosylation by PNGase F. Aliquots (50 ng) of plasma AGP and PMN AGP after treatment for various times between 0 and 5 min with PNGase F. (b) Detection of di- and/or trimeric LewisX groups. Plasma AGP isolated from healthy blood donors (lanes 1 and 4) or from patients suffering from acute trauma (lanes 2 and 5) and of PMN AGP (lanes 3 and 6) were immunostained with rabbit anti-human AGP IgG or the di- and/or trimeric LewisX-specific antibody 6H3. Lane 7, Trimeric LewisX-albumin conjugate, a positive control for 6H3 [23
]. (c) AGP secretion by activated PMN. AGP present in PMN extracts and in extracellular medium without and with stimulation by CytoB + fMLP, PAF + fMLP, or fMLP alone, as indicated in the table
and a; immunoprecipitated AGP from human plasma was used as a control. (d) Distribution of AGP over 100,000 g supernatant (Sup) and extract of the residual pellet of PMN and monocytes after lysis by repeated freezing and thawing. The extract of the pellet was set at the same volume as the supernatant prior to immunoprecipitation. Immunoprecipitated AGP from human plasma was used as a control.
|
|
View this table: [in a new window] |
Table 1. Monosaccharide Composition of PMN and Plasma AGP
|
Release of AGP from PMN by degranulation
Nonstimulated PMN did not release AGP spontaneously, as it was not detectable in the extracellular medium of the nonstimulated PMN (Fig. 1c)
. Degranulation of specific and azurophilic granules of PMN in vitro takes place in the order: CytoB + fMLP (both type of granules) > PAF + fMLP (specific granules) > fMLP alone [33
]. Stimulation of PMN with CytoB + fMLP resulted in a maximal and total release of soluble AGP into the medium (Fig. 1c) . Stimulation with PAF + fMLP resulted in a diminished release and with fMLP alone, in a minimal release of AGP into the extracellular medium (Fig. 1c)
.
Subcellular localization of AGP in PMN
Immunoelectron microscopy showed that AGP was localized in electron-dense granules of PMN and not on the cell membrane (Fig. 2a
). To characterize the AGP-positive granules, cryosections were double-labeled with anti-AGP IgG and antibodies against lactoferrin, a marker for PMN-specific granules (Fig. 2b)
, MPO, a marker for PMN azurophilic granules (Fig. 2c)
, and albumin, a marker for secretory vesicles and multivesicular bodies (Fig. 2d)
[6
]. Almost all AGP-positive granules were also lactoferrin-labeled (Fig. 2b)
, whereas only a few AGP-positive granules were MPO-positive (Fig. 2c) . Colocalization of AGP was also found with albumin in the secretory vesicles.
 View larger version (200K): [in a new window] |
Figure 2. Localization of AGP in PMN. Ultrathin cryosections of neutrophils from peripheral blood were immunogold-labeled with antibodies against AGP or double-labeled with anti-AGP and, respectively, antilactoferrin (anti-Lf), anti-MPO, or antialbumin (anti-Alb), as indicated. (a) The majority of the granules is labeled. (b) Colabeling of AGP with lactoferrin (large arrows), a marker for specific granules; a granule with only AGP (small arrow), and a small vesicle labeled with AGP (arrowhead) are shown; n, nucleus. (c) A granule is shown in which AGP and MPO (large arrows), a marker for azurophilic granules, are labeled and as well, granules labeled with AGP alone (small arrow). (d) Colabeling of AGP and albumin, a marker for secretory vesicles, is shown in vesicles (upper panel, arrows) and in a multivesicular body (lower panel, arrow). Original bars = 200 nm.
|
 View larger version (185K): [in a new window] |
Figure 3. Localization of AGP in PMN myelocytes. Cryosections of bone marrow were double-labeled for AGP and lactoferrin as indicated. Colabeling of AGP and lactoferrin is shown on granules (arrows), Golgi stacks (G), and endoplasmic reticulum (er). Nucleus (n). Original bar = 200 nm.
|
 View larger version (99K): [in a new window] |
Figure 4. Immunohistochemical localization of activated C3, CD66b, and AGP in infarcted myocardium. Tissue specimens were from a patient with an infarct duration of 1–3 days. Frozen tissue sections shown were from the same myocardial infarction site (original magnification: 250x). (a) AGP; (b) CD66b; (c) activated C3 (act C3; see Materials and Methods for details of the mAb used). n = Normal tissue; b = border-zone infarction side; c = center-zone infarction side. (d) Quantification of the time-dependent immunohistochemical deposition of AGP, activated C3 and CD66b, in infarcted myocardium of 27 patients (see Materials and Methods for details and number of patients per time-point). ( ) AGP; ( ) activated C3; (hatched squares) CD66b. *, P value <0.05, comparing infarcted versus control area.
|
 View larger version (157K): [in a new window] |
Figure 5. Colocalization and deposition of AGP and di- and/or trimeric Lex structures in the same myocardial infarction site. Immunohistochemical localization of AGP (a and c) and di- and/or trimeric Lex structures (Lex; b and d). (a and c) Sequential frozen tissue sections that were obtained from the same myocardial infarction site [original magnification: (a, b) 250x; (c, d) 400x]. Tissue specimens were from a patient with an infarct duration of 1–3 days. See Materials and Methods for details of the mAb used. (d) White arrows, Neutrophilic granulocytes.
|
Different molecular weight forms of AGP in infarcted and noninfarcted myocardial tissue
AGP was isolated from homogenates of infarcted and noninfarcted myocardial tissue and subjected to SDS-PAGE and blotting. Plasma AGP was present in homogenates of both sources of tissue (Fig. 6a
, lanes 1–3, heavy band). An additional, higher molecular weight AGP band was present in the homogenate of the infarcted area, most probably representing PMN AGP (Fig. 6a
, lanes 1 and 4). This band was not detectable in the noninfarcted tissue (lane 2) and in normal plasma (lane 3).
 View larger version (52K): [in a new window] |
Figure 6. AGP and AGP mRNA in normal and infarcted myocardial tissue. (a) Western blot of AGP isolated from homogenized myocardial tissue originating from infarcted (lane 1) and noninfarcted tissue (lane 2). AGP isolated from normal plasma (lane 3) and PMN lysate (lane 4) were used as controls. See Materials and Methods for further details. (b) PCR analysis of cDNA isolated from cardiomyocytes (lanes 1–3) and as a positive control, IHH (lanes 4–6). Lanes 1 and 4, AGP-A; lanes 2 and 5, AGP-B/B'; lanes 3 and 6, actin. M = Phage  marker. PCR products were visualized on a 1.5% agarose gel containing ethidium bromide. See Materials and Methods for further details.
|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
AGP was mainly present in specific granules but also in secretory granules and in some azurophilic granules of PMN, as was clear from electron microscopic observations in PMN and PMN progenitors from bone marrow (Figs. 2 and 3) . This is supported biochemically by the difference in effects of CytoB and PAF on the secretion of AGP by fMLP-stimulated PMN (Fig. 1c) . Gahmberg and Andersson [15 ] have reported that AGP was present as a membrane-bound form on PMN. Our studies show that AGP was present in PMN as a soluble molecule (Fig. 1d) and was recovered as such from the secretion media after activation (Fig. 1c) . A possible explanation for this discrepancy may be that the specific sialylated and fucosylated glycan structures of PMN AGP can bind to cell-surface L-selectin molecules. When so, stimulation of PMN will result in the rapid proteolytic release of L-selectin and then also will release bound AGP. Alternatively, the G9 mAb that was used in the present study, which is directed against the C-terminal part of AGP, may not recognize a membrane-bound isoform of AGP. De novo synthesis of AGP by PMN was indicated by its presence in the rough endoplasmic reticulum and Golgi stacks of PMN progenitors from bone marrow (Fig. 3) . This was further supported by the expression of the two AGP mRNAs in PMN, and AGP-A was more strongly expressed than AGP-B/B', as in human hepatocytes [30 ].
The molecular weight of PMN AGP was higher than that of plasma AGP, 50–60 kD versus 40–43 kD. Only a minor fraction of AGP in PMN might have originated from plasma, as is indicated by the colocalization with albumin in secretory vesicles (Fig. 2d) as well as by the weak appearance of an AGP band at the same molecular weight as plasma AGP in Figure 1 . The difference in molecular weight between PMN and plasma AGP appeared to reside in the presence of strongly fucosylated polylactosamine units on the five N-linked glycans of PMN AGP (Fig. 1a and 1b , Table 1 ). This further supports synthesis of AGP by PMN, as the type of glycosylation is typical for PMN [19 , 37 ]. Like in the studies of Fukuda and co-workers [19 , 37 ], the glycans appeared to be undersialylated, as indicated by the lower sialic acid/mannose ratio (Table 1) as well as the observed lower electrophoretic mobility of PMN AGP in relation to normal plasma AGP (not shown). Most probably, the glycans were of the tri- or tetra-antennary type, as was indicated by the observed lack of concanavalin A reactivity of PMN AGP (not shown).
Presumably, PMN release the specifically fucosylated AGP glycoforms also in vivo, as was suggested by the immunohistochemical studies in the patients that had died from AMI. In the first place, positive staining for AGP of myocardial tissue with the mAb-anti-AGP G9 was only detected when infiltrating PMN were found to be present, i.e., only during the first few days after onset of the infarct (Fig. 4d)
. AGP staining colocalized with PMN in the center zone of the infarct and was found deposited on jeopardized cardiomyocytes as well as on PMN (Fig. 4)
. Second, colocalization was found on AGP and di- and/or trimeric LewisX structures (Fig. 5) . The colocalization on cardiomyocytes strongly suggests that the mAb 6H3 recognized deposited PMN AGP and not plasma AGP, as the latter does not express di- and/or trimeric LewisX-substituted glycans (Fig. 1b
, Table 1
) [10
, 11
, 36
]. The intense staining of PMN by di- and/or trimeric LeX is in accordance with the abundant occurrence of these structures on these cells [19
]. In the third place, in tissue from an AMI, 41–43 kD and 
60 kD AGP was present in the myocardial infarcts, whereas in normal myocardial tissue, only 41–43 kD AGP was found (Fig. 6a)
. The latter AGP most probably originated from plasma, as no AGP could be detected in purified cardiomyocytes (data not shown). Siegel et al. [38
] have reported a localization of AGP at the cell surface of the cardiomyocytes in the region of the sarcolemma of normal myocardium and diminished amounts in necrotic tissue in acute myocardial infarcts. These studies were performed using polyclonal antibodies for detection. We also found abundant staining of the myofibers with commercial polyclonal antibodies against AGP (data not shown). However, we concluded that this was the result of cross-reactivity, e.g., with acid glycoproteins that are known to be present in the region of the sarcolemma [38
]. This conclusion was based on the finding that cytoplasmic as well as membrane-staining with mAb-anti-AGP G9 for AGP were only found in AMI and not in normal myocardial tissue. In addition, PCR analysis of cDNA prepared from isolated and purified cardiomyocytes did not show detectable expression of AGP mRNA. Although it cannot be excluded that during AMI the conditions of hypoxia and/or ischemia have induced the biosynthesis of AGP by cardiomyocytes, we do not think that this would have resulted in the specific AGP molecules detected in the present study on those cells. On the contrary, we think that the cytoplasmic staining of AGP can be explained by the leakage of the membranes of the ischemic-jeopardized and complement-positive cardiomyocytes [26
]. AGP appeared to be decorated with the typical PMN type of glycans expressing di- and/or trimeric LeX groups; no indications are presented in the literature pointing to the presence of such a type of glycans on myocardial cells.
A major stimulus for ongoing inflammation in AMI is presumably the activation of complement, resulting in an extension of the inflamed zone as well as further attraction of PMN [26 ]. When accumulating at high numbers, PMN may cause significant tissue damage [22 ]. It is interesting that the intensity of complement-staining in the AGP-positive areas during an AMI was less than that in the border zone of the infarct area, in which AGP was negative (Fig. 4) . It is tempting to speculate that less-complement depositions were a result of release of PMN AGP, as specifically fucosylated AGP glycoforms have been reported to inhibit complement activity in vitro and possibly also in vivo [14 ]. PMN granules contain other complement-inhibiting substances as well, such as defensins [39 ].
It is not likely that the loss of complement positiveness in the center of the infarction area can be attributed to a low flow state in the reperfused heart. In a recent study in which all patients of the present study were included, we found that reperfusion therapy indeed induces larger areas of complement positiveness compared with nonreperfused hearts [40 ]. However, also in these reperfused hearts, the center of the infarction area showed less-complement depositions similarly, as in the hearts of patients with infarctions not treated with reperfusion therapy. Furthermore, in none of the patients was an obstructive thrombus found after 1–3 days infarction, which makes it hard to believe that no reperfusion would have taken place in the infarcted area of patients who did not receive reperfusion therapy. Next to this, we and others have shown that in these infarcted areas also, complement is produced locally by endothelium, which makes it very likely that a source of complement in infarcted areas is also present, even in less-perfused areas [26 , 40 , 41 ].
The fucosylated structures, such as expressed on PMN AGP, are suitable ligands for E-selectin [4 , 20 , 42 ]. So, PMN AGP released into the inflamed area and diffused into the blood plasma may also antagonize adhesion of newly arriving PMN directly, by neutralizing E-selectin on the inflamed endothelium. For plasma AGP, it has been reported that it is able to diffuse through the endothelium (see e.g., ref. [43 ]). A masking effect of selectins was already proposed for the strongly fucosylated and sialyl LewisX-containing AGP glycoform present in acute-phase plasma [12 , 13 ]. Indeed, studies in an ischemia-reperfusion model in rats support this notion [14 ]. Amelioration of PMN-induced damage was observed by injecting recombinant human AGP prior to reperfusion; a sialyl-Lewisx-containing AGP glycoform is more effective in that regard than a nonfucosylated AGP glycoform.
The exact function of AGP is not known. In addition to the above-described effects on complement and E-selectin, other anti-inflammatory effects have been reported for total AGP or specific AGP glycoforms, for example, inhibition of chemotactic responses of PMN to fMLP and C5a [8 , 13 , 44 45 46 47 ]. The demonstration in this report that PMN harbor endogenously synthesized, soluble AGP, which is released upon activation, strongly suggests that PMN are equipped with a feedback-response mechanism to excessive inflammation.
Received October 6, 2004; accepted December 9, 2004.
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
1-Acid glycoprotein: an acute phase protein with inflammatory and immunomodulating properties Cytokine Growth Factor Rev. 14,25-34[CrossRef][Medline]1-acid glycoprotein Biochim. Biophys. Acta 492,291-3021-acid glycoproteins J. Biol. Chem. 256,8476-84841-acid glycoprotein (orosomucoid) in human sera J. Exp. Med. 177,657-6661-acid glycoprotein (orosomucoid) in health and disease: occurrence, regulation and possible functional implications Trends Glycosci. Glycotechnol. 10,235-2451-Acid glycoprotein reduces local and remote injuries after intestinal ischemia in the rat Am. J. Physiol. 273,G1031-G10351-acid glycoprotein (orosomucoid) J. Exp. Med. 148,507-5211-acid glycoprotein) Microbiol. Immunol. 28,99-111[Medline]1-Acid glycoprotein expression in human leucocytes: possible correlation between 1-acid glycoprotein and inflammation cytokines in rheumatoid arthritis Inflammation 17,33-45[CrossRef][Medline]
T cells J. Immunol. 159,3574-3583[Abstract]1-acid glycoprotein in human seminal plasma is associated with high branching and expression of Lewis(a) groups on its glycans: supporting evidence for a prostatic origin Prostate 52,34-42[CrossRef][Medline]1-acid glycoprotein EMBO J. 6,2289-2296[Medline] in a subpopulation of cytoplasmic granules Blood 90,1255-12661-acid glycoprotein having different interactions with concanavalin A. Variations in the occurrence of di-, tri-, and tetraantennary glycans and the degree of sialylation Eur. J. Biochem. 175,387-394[Medline]1-acid glycoproteins in human myocardium Lab. Invest. 52,107-112[Medline]1-acid glycoprotein in the continuous microvascular endothelium Proc. Natl. Acad. Sci. USA 95,6175-61801-acid glycoprotein and 1-antitrypsin against ischemia/reperfusion injury by preventing apoptosis and inflammation Circulation 102,1420-14261-acid glycoprotein Inflammation 14,1-9[CrossRef][Medline]1-acid glycoprotein on human polymorphonuclear neutrophils: influence of glycan heterogeneity Clin. Chim. Acta 224,65-71[CrossRef][Medline]This article has been cited by other articles:
 |
|
 |
|
  R. T. Clements, G. Smejkal, N. R. Sodha, A. R. Ivanov, J. M. Asara, J. Feng, A. Lazarev, S. Gautam, V. Senthilnathan, K. R. Khabbaz, et al. Pilot Proteomic Profile of Differentially Regulated Proteins in Right Atrial Appendage Before and After Cardiac Surgery Using Cardioplegia and Cardiopulmonary Bypass Circulation, September 30, 2008; 118(14_suppl_1): S24 - S31. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
