Published online before print May 10, 2004
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
* Rigshospitalet, Department of Haematology, Copenhagen, Denmark;
DakoCytomation Corporation, Carpinteria, California; and
The Netherlands Cancer Institute, Division of Cell Biology, Amsterdam, The Netherlands
1Correspondence: Rigshospitalet, Department of Haematology, Granulocytlaboratoriet, Building 9322, Blegdamsvej 9, DK-2100, Copenhagen, Denmark. E-mail: niemann{at}dadlnet.dk
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Key Words: granules • differentiation • myeloid • proteoglycan 1, secretory granule • PRG1
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
1.3 kb size and reveals a protein core of molecular weight 17,600 [2
] with 42% homology to rat serglycin at the amino acid level [3
] with the highest conservation in the N terminus [4
]. The glycosaminoglycan side-chains are attached to serines in the serine glycine repeat region, which covers 18 amino acids in the middle of the core protein (amino acids 94–111 of the 161 amino acids core protein) [5
]. The types of glycosaminoglycans attached to serglycin varies between and within cell types, e.g., heparin sulfate and chondroitin sulfate in different subsets of mast cells [6
, 7
]. Serglycin mRNA has been detected in several hematopoietic cells and cell lines including myeloid HL-60 cells and primary myeloid cells [2
, 8
9
10
11
12
13
14
15
16
]. Serglycin proteoglycan has been detected in monocytic THP-1 cell medium [17
], and in cytotoxic CTLL-2 cells [10
], serglycin immunoreactivity has been detected in cell supernatants from EL-4 (lymphoma), X63 (myeloma), BAF/BO3 (Pro-B), WEHI3 (myelomonocytic), and P815 (mastocytoma) cells [18
] and by immunoprecipitational [5
] and immunohistochemical [19
] methods in HL-60 cells. It has been suggested that serglycin plays a major role in genesis of granules and in packaging and reversible inactivation of proteases in granules of neutrophil granulocytes [20 , 21 ], but there is no evidence to support this notion. In cytotoxic lymphocytes, serglycin has been shown to form complexes with granzyme B and to be delivered to target cells by perforin as a macromolecular complex with granzyme B [22 , 23 ]. In macrophages, pro-matrix metalloproteinase 9 appears to be covalently linked to a proteoglycan core protein, presumably serglycin, but the significance of this remains undetermined [24 ]. Serglycin proteoglycan from cell supernatants has been identified as a ligand for CD44 and has been shown to cause cell aggregation of CD44-positive cells in vitro [18 ]. In exocrine pancreas, Kleene and his colleagues [25 , 26 ] found evidence that intact serglycin proteoglycan was implicated in granule formation. In connective tissue-type mast cells, inhibition of sulfation of heparin sulfate on serglycin has been shown to cause morphological changes and decrease the amount of granule proteins [27 , 28 ].
The neutrophil granulocyte is the most abundant of the mobile phagocytes and constitutes the majority of leukocytes in peripheral blood. It contains at least four subsets of granules with different contents of matrix and membrane proteins, which are released in a regulated and hierarchical manner [29 ]. The different content of different granule subsets is explained by targeting as a result of timing of synthesis of different proteins during differentiation of neutrophil granulocytes, i.e., that proteins synthesized at the same time will localize to the same granule subset and that differences in granule constituents reflect differences in time of biosynthesis of the granule proteins [30 ]. The mechanisms of granule assembly and packaging of granule matrix proteins and membrane proteins from the trans Golgi network into granules in neutrophil granulocytes are still poorly understood [31 ].
The reports on serglycin mRNA expression in neutrophil granulocytes and their precursors, the localization of glycosaminoglycans to granules of these cells [32 33 34 ], and the reports on the implications for serglycin in granule formation and in complexing with granule proteins, combined with lack of direct evidence that serglycin is present in granules of neutrophil granulocytes, prompted us to undertake the present study. Here, we show the presence and the localization of serglycin immunoreactivity at the cellular and subcellular level in neutrophil granulocytes and their precursors.
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Escherichia coli (XL1-blue; Stratagene, La Jolla, CA), expressing the His-tagged recombinant serglycin, was incubated overnight (o/n) at 37°C in Luria-Bertani (LB) medium (Bie and Berntsen, Rødovre, Denmark). The culture (50 mL) was inoculated in 450 mL LB medium and incubated with shaking at 37°C for 2 h. Thereafter, the expression of recombinant protein was induced with 0.1 mM isopropyl-ß-D-thiogalactopyranoside (Boehringer Mannheim, Mannheim, Germany), and the culture was further incubated at 37°C for 4 h. The pellet of bacteria was resuspended in 50 mL extraction buffer (50 mM sodium phosphate, 500 mM NaCl, pH 7.0) and incubated with 0.75 mg/mL lysozyme (Sigma, Steinheim, Germany) for 30 min at room temperature (RT). The bacteria were sonicated three times, 15 s on ice at max level, in a Soniprep 150 ultrasonic disintegrator (Sanyo, Leicester, UK) and centrifuged at 30,000 g for 20 min. The His-tagged recombinant serglycin was extracted by adding 3 mL TALON metal affinity resin (Clontech Laboratories, Palo Alto, CA) equilibrated in extraction buffer and incubating for 2 h at RT. The resin was washed three times in extraction buffer and two times in extraction buffer with 10 mM imidazole (Sigma). The His-tagged recombinant serglycin was eluted with 100 mM imidazole in extraction buffer. The buffer was changed over a PD-10 column (Amersham Biosciences, Uppsala, Sweden) to 25 mM Tris-HCl, 25 mM NaCl, pH 8.5, and the solution was fractionated by anion-exchange chromatography on a mono-Q column (Amersham Biosciences) with increasing salt concentration. The size of the eluted recombinant serglycin was visualized by sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) [35 ], and the identity was confirmed by mass spectrometry following tryptic digestion (kindly performed by Anders H. Johnsen, Department of Clinical Biochemistry, Rigshospitalet, Denmark).
Antibody generation and purification
A rabbit was immunized by Dako (Glostrup, Denmark) with the purified recombinant serglycin (nonglycosylated). The titer of the antiserum was measured against the recombinant serglycin used for immunization. When the titer reached a plateau, the animal was exsanguinated. The immunoglobulin G (IgG) fraction was isolated on a HiTrap Protein A HP column (Amersham Biosciences) following instructions from the manufacturer; elution was with 3 M KSCN, followed by dialysis against phosphate-buffered saline (PBS; 136 mM NaCl, 2.7 mM KCl, 8 mM Na2HPO4, 1.46 mM KH2PO4, pH7.4).
The IgG fraction was further purified by affinity chromatography on a column with 1.5 mg recombinant serglycin coupled to Sepharose 4B (Amersham Biosciences), as described by the manufacturer. The IgG fraction in PBS, 1 M NaCl, was applied to the column, which was first equilibrated in the same buffer, followed by washing with 10 column vol of this buffer and elution with 3 M KSCN. The eluted Ig was dialyzed against PBS, and preserved with 0.1% NaN3.
A portion of the IgG fraction was biotinylated as described previously [36 ], dialyzed against PBS, and preserved with 0.1% NaN3.
Isolation of bone marrow cells
Bone marrow aspirates and blood samples were obtained from healthy volunteers, who gave informed consent according to the permission and guidelines from the local ethics committee. Cells from different stages of myeloid differentiation were isolated as described previously [37
, 38
]. In brief, the erythrocytes were sedimented with dextran, and the cells were separated by density centrifugation on discontinuous Percoll gradients of 1.065 g/mL and 1.080 g/mL. Three different bands of cells, numbered in order of decreasing density, were harvested: Band 1 primarily containing segmented neutrophils (SC) and band cells (BC); band 2 primarily containing metamyelocytes (MM) and myelocytes (MC); band 3 primarily containing promyelocytes (PM) and myeloblasts (MB). Neutrophil granulocytes (PMN) from peripheral blood were isolated as described previously [39
]. For Northern blotting, the cells from the different bands were further depleted for contaminating non-neutrophil hematopoietic cells by magnetic cell sorting (MACS) with mouse antibodies against surface epitopes of non-neutrophil cells, as described previously [40
].
Subcellular fractionation of PMN from peripheral blood and exocytosis stimulation
PMN were obtained from buffy coats of peripheral blood isolated from healthy donors by standard procedures [39
]. The subcellular fractionation was performed as described previously [39
]. In brief, the cells were disrupted by nitrogen cavitation, after which nuclei and unbroken cells were pelleted. The supernatant, containing granules, secretory vesicles, plasma membrane, Golgi apparatus, and cytosol, was separated by density centrifugation on discontinuous Percoll gradients of 1.050 g/mL and 1.120 g/mL. Three bands, referred to as 
, ß, and 
, in order of decreasing density, were collected: The -band, containing primary granules; the ß-band, containing secondary and tertiary granules; and the -band, containing secretory vesicles, plasma membrane, and Golgi apparatus. The cytosol was collected above the gradient. Percoll was removed from the subcellular fractions by centrifugation before further analysis. Exocytosis stimulation was performed with 5 µg/mL phorbol 12-myristate 13-acetate (PMA; Sigma) or 1 µM ionomycin (Sigma) at a cell concentration of 108 cells/mL as described previously [41
].
Cell culture
HL-60 cells were obtained from American Type Culture Collection (Manassas, VA). The cells were cultured in RPMI-1640 medium (Invitrogen) with 10% fetal calf serum (Invitrogen), 100 U/mL penicillin, and 100 µg/mL streptomycin in a humidified incubator with 5% CO2 at 37°C. Conditioned medium was obtained by pelleting the cells. The medium was concentrated, and the buffer was changed to ABC buffer (see below) using Centriprep YM-10 columns (Millipore, Bedford, MA) following instructions from the manufacturer.
Northern blotting
Total RNA was extracted using the Trizol reagent (Invitrogen). Northern blotting was performed as described previously [40
]. The full-length serglycin probe (kindly provided by Dr. Urban Gullberg, University of Lund, Sweden) was -32P-labeled with the Random Primers DNA labeling system (Invitrogen). Hybridization conditions were as described previously [16
].
Immunocytochemistry
Cytospins of HL-60 cells, PMN, MB + PM, MC + MM, and BC + SC were fixed in 4% formaldehyde in Tris-buffered saline (TBS; 50 mM Tris-HCl, 150 mM NaCl, pH 7.6) and permeabilized in 1% Triton X-100 (AppliChem, Darmstadt, Germany) in TBS with 1% bovine serum albumin (BSA; Sigma). Unspecific binding was blocked with 1% BSA in TBS. The cytospins were incubated with affinity-purified anti-serglycin Ig or control normal rabbit Ig (Dako, X0903; diluted to the same protein concentration as the anti-serglycin Ig) for 60 min. The slides were washed in TBS between applications of antibodies. Detection of immunoreactivity was with mouse anti-rabbit Ig (Dako, M0737) and the APAAP KIT System 40 (Dako, K0670). Fast red was used as enzyme substrate; hematoxylin (Bie and Berntsen) was used for counterstaining.
Immunoelectron microscopy
Human bone marrow cells, separated as described above, were fixed for 1 h in a mixture of 0.5% glutaraldehyde and 4% paraformaldehyde in 120 mM PIPES, 50 mM HEPES, 8 mM MgCl2, 40 mM EGTA, pH 6.9, and then processed for ultrathin cryosectioning as described previously [42
]. Cryosections (45 nm thick) were cut at –125°C using diamond knives (Drukker Cuijk, The Netherlands) in an ultracryomicrotome (Leica Aktiengesellschaft, Vienna, Austria) and transferred with a mixture of sucrose and methylcellulose onto formvar-coated copper grids. The grids were placed on 35-mm petri dishes containing 2% gelatine. For immunolabeling, the sections were incubated with affinity-purified rabbit anti-serglycin Ig for 45 min, followed by 30 min incubation with 10 nm protein-A-conjugated colloidal gold (EM Laboratory, Utrecht University, The Netherlands). After immunolabeling, the cryosections were embedded in a mixture of methylcellulose and uranyl acetate and were examined with a Philips CM 10 electron microscope (Eindhoven, The Netherlands). For controls, the primary antibody was replaced by a nonrelevant rabbit Ig.
Anion exchange chromatography
The subcellular fractions were lysed in 8 M urea, 50 mM Tris-HCl, pH 8.0, 2% Triton X-100 with protease inhibitors [1 mM phenylmethylsulfonyl fluoride (PMSF; Sigma), 200 KIE/mL trasylol (Bayer AG, Leverkusen, Germany), and 100 µg/mL leupeptin (Sigma)] at 4°C o/n. The samples were centrifuged at 180,000 g for 20 min, and the supernatants were subjected to anion exchange chromatography by a mono Q column (Amersham Biosciences) equilibrated in 50 mM Tris-HCl, pH 8.0, 2% Triton X-100. Elution was with stepwise increase in NaCl from 0 to 1.5 M (0.15 M steps).
Separation of granule proteins by gel filtration on Superose 12 column
The -band and the ß-band from a subcellular fractionation of 1010 PMN were solubilized in 1 mL 40 mM n-octyl-ß-D-glucopyranoside (NOG; Calbiochem, La Jolla, CA) in PBS with the addition of protease inhibitors (1 mM PMSF, 200 KIE/mL trasylol, 100 µg/mL leupeptin) o/n at 4°C. The samples were diluted by PBS to 20 mM NOG and centrifuged at 15,000 g for 15 min; 200 µL of the supernatants was subjected to gel filtration on a Superose 12 column (Amersham Biosciences) equilibrated in the same buffer.
Enzyme-linked immunosorbent assay (ELISA)
Immunoplates (96-well flat-bottom; Nunc, Roskilde, Denmark) were incubated o/n at RT with affinity-purified anti-serglycin Ig in 50 mM Na2CO3/NaHCO3, pH 9.6. The plates were subsequently washed in buffer A (500 mM NaCl, 3 mM KCl, 8 mM Na2HPO4/KH2PO4, 1% Triton X-100, pH 7.2) with a SkanWasher (Skatron, Roskilde, Denmark) and blocked with buffer B (500 mM NaCl, 3 mM KCl, 8 mM Na2HPO4/KH2PO4, 1% BSA, 1% Triton X-100, pH 7.2). Buffer B was discarded, and the plates were incubated for 1 h with the recombinant serglycin standard or relevant samples diluted in buffer B, washed as above, and incubated with biotinylated anti-serglycin Ig for 1 h. After an additional wash in buffer A, the wells were incubated for 1 h with horseradish peroxidase (HRP)-conjugated avidin (Dako, P0347) in buffer B. After washing, the color reaction was developed with Ortho-Phenylene-Diamine tablets (KEM-EN-TEC, Copenhagen, Denmark), solubilized in buffer C (100 mM Na2HPO4, 100 mM citric acid, pH 5.0) with the addition of 0.006% H2O2 immediately before use. The color reaction was stopped with 1 M H2SO4, and the absorbance at 492 nm was measured with a multiscan ascent ELISA reader (Labsystems, Helsinki, Finland). The conditions for myeloperoxidase [43
], lysozyme [44
], neutrophil gelatinase-associated lipocalin (NGAL) [45
], lactoferrin [46
], gelatinase [47
], and human cationic antimicrobial peptide 18 (hCAP18) [48
] ELISA were as described previously.
Chondroitinase (cABC) digestion
Samples were digested with 0.2 U/mL cABC (Sigma) in 250 mM Tris-HCl, 300 mM Na-acetate, pH 8.0, 0.1% BSA, 0.25% Triton X-100 (cABC buffer) with protease inhibitors (1 mM PMSF, 200 KIE/mL trasylol, 100 µg/mL leupeptin) at 37°C for 2 h to release the core protein.
Western blotting
SDS-PAGE and immunoblotting [49
] were performed according to standard procedures. A 12% gel with 3% stacking gel was used; the samples were reduced with mercaptoethanol in Laemmli buffer [35
]. Protein-transfer to nitrocellulose was performed in 10 mM 3-(cyclohexylamino)-1-propanesulfonic acid, pH 11.0, 10% methanol. Blots were blocked in 5% skim milk powder in PBS for 1 h and incubated o/n with affinity-purified rabbit anti-serglycin Ig. Detection was by HRP-conjugated swine anti-rabbit Ig (Dako, P0217) and 1,4-diaminobutane/metal concentrate in stable peroxide substrate buffer (Pierce, Rockford, IL).
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
-32P-labeled full-length serglycin probe. The hybridization signals are shown in Figure 1A
. The blot was reprobed with 18S rRNA for internal loading control (Fig. 1B)
. The Northern blot clearly showed the decreasing expression of serglycin mRNA during neutrophil maturation, as schematically illustrated in Figure 1C
, where hybridization signals are normalized to 18S rRNA signals. This tendency would be even more pronounced if the hybridization signals were normalized to cell number instead of 18S rRNA, as the mature neutrophil granulocytes in the G0-phase of cell cycle have less RNA/cell compared with their cycling precursors [50
]. This finding is illustrated in Figure 1D
. Here, the hybridization signals were normalized to the number of cells, from which the amount of total RNA in each lane was isolated.
 View larger version (26K): [in a new window] |
Figure 1. Northern blotting of total RNA from myeloid cells of human bone marrow and peripheral blood, separated according to maturational stage. (A) Northern blot probed with a full-length -32P-serglycin probe; 4.6 µg total RNA loaded in each lane. (B) The same blot, stripped and reprobed with an -32P-18S probe. (C) Schematic representation of hybridization intensities, normalized to 18S for each lane. (D) Schematic representation of hybridization intensities, normalized to cell number for each lane.
|
 View larger version (45K): [in a new window] |
Figure 2. (A and B) Immunocytochemistry of HL-60 cells. (A) Rabbit anti-serglycin Ig. (B) Negative control with rabbit Ig from nonimmunized rabbits as primary antibody. Primary antibody preabsorbed with excess recombinant serglycin gave the same result. (C) Western blotting with rabbit anti-serglycin Ig. HL-60-conditioned medium. Lane 1, Pretreated with cABC before electrophoresis; lane 2, untreated. MW, Molecular weight; markers are indicated.
|
1% positive cells. A single PMN with central immunoreactivity is shown to the left in Figure 3G
, and a few PMN and SC, with more faint immunoreactivity, are seen, as illustrated in Figure 3E
and 3G
. No staining was seen with nonimmunized rabbit Ig as primary antibody (data not shown). As another control, the affinity-purified anti-serglycin Ig was preincubated with excess recombinant serglycin before immunostaining on parallel cytospins. No staining was observed (Fig. 3B
3D
3F and 3H)
.
 View larger version (65K): [in a new window] |
Figure 3. Immunocytochemistry of cytospins of myeloid cells from human bone marrow and peripheral blood. (A, C, E, G) Rabbit anti-serglycin Ig as primary antibody. (B, D, F, H) Negative controls; primary antibody was preabsorbed with excess recombinant serglycin. Negative controls with rabbit Ig from nonimmunized rabbits as primary antibody were also blank (data not shown). (A, B) MB and PM, some contaminating lymphoid cells with scarce staining and a few more mature myeloid cells that are negative. (C, D) MC and MM, few contaminating BC are negative. (E, F) BC and SC. (G, H) PMN.
|
 View larger version (202K): [in a new window] |
Figure 4. Immunoelectron microscopy with rabbit antiserglycin Ig. (A–C) Cryosections of myeloid precursor cells. (A) Labeling for serglycin on the Golgi stacks (G), the trans Golgi network (TGN), and on the matrix of a few, presumably immature, granules (arrows). N, Nucleus. (B) Ellipsoid Golgi with a multilayered shell of cisternae, which is highly labeled for serglycin. (C) Slightly labeled endoplasmic reticulum (er) and denser labeling of two granules with dispersed matrix (arrows); most granules are unlabeled. (D) Cryosection of a mature neutrophil granulocyte labeled for serglycin. The nucleus (N) is multilobed and with condensed chromatin. The Golgi (G) apparatus is weakly labeled. Original bars = 200 nm.
|
-band containing primary granules; the ß-band containing secondary and tertiary granules; and the -band containing secretory vesicles, plasma membrane, and Golgi apparatus [39
]. Anion exchange chromatography of each band was performed to isolate proteoglycans from the other granule proteins. The results without cABC digestion are shown in Figure 5
. ELISA for serglycin immunoreactivity on subcellular fractions without anion exchange chromatography gave the same results with predominant reactivity in the -band; predigestion with cABC did not change the immunoreactivity (data not shown). Rabbit Ig from nonimmunized animals and Ig that did not bind to recombinant serglycin during affinity purification of the anti-serglycin Ig were used as negative controls. No immunoreactivity was seen for recombinant serglycin or the -band in the negative controls. The cytosolic fraction and the fraction containing nuclei and undisrupted cells were also tested without significant immunoreactivity for serglycin (data not shown).
 View larger version (10K): [in a new window] |
Figure 5. ELISA for serglycin immunoreactivity in anion exchange chromatography fractions of subcellular fractions from neutrophil granulocytes. Elution was with stepwise, increased concentrations of NaCl. Molar concentrations of NaCl are indicated for each tested fraction.
|
Western blotting localized serglycin to the Golgi-containing fraction of PMN subcellular fractions
Serglycin proteoglycan cannot be blotted by normal techniques, probably as a result of the size and negative charge of the sulfated glycosaminoglycan side-chains [24
]. We therefore digested fractions from anion exchange chromatography of subcellular fractions of neutrophil granulocytes with cABC to obtain the serglycin core protein before gel electrophoresis and Western blotting. Western blottings are shown in Figure 6
. The figure represents -, ß-, and -band fractions from the same number of cells. Immunoreactivity for serglycin was restricted to the 0.45 M NaCl fraction of the -band (Fig. 6C
, lane 3) containing the Golgi apparatus, plasma membrane, and secretory vesicles. This is in accordance with the rare but clear immunoreactivity for serglycin seen in the Golgi area of neutrophil granulocytes (Fig. 3G)
and the immunoreactivity for serglycin predominantly in the -band by ELISA (Fig. 5)
. The size of the band (22 kDa) is in agreement with a previous report on the size of the serglycin core protein from the -band of neutrophil granulocytes [53
]. Figure 6D
shows the size of the recombinant serglycin (nonglycosylated) by Western blotting for comparison.
 View larger version (65K): [in a new window] |
Figure 6. Western blotting with rabbit anti-serglycin Ig as primary antibody. Subcellular fractions from neutrophil granulocytes subjected to anion exchange chromatography. Elution was with stepwise increased concentrations of NaCl. Molar concentrations of NaCl are indicated below. (A) -Band: Primary granules. (B) ß-Band: Secondary granules and tertiary granules. (C) -Band: Secretory vesicles, plasma membrane, and Golgi apparatus. All fractions were subjected to cABC digestion before gel electrophoresis to relieve the core protein. Molecular weight markers are indicated. (D) Recombinant serglycin (nonglycosylated).
|
-band and ß-band from a subcelluar fractionation of neutrophil granulocytes to gel filtration by Superose 12. The subcellular fractions were solubilized by the non-ionic detergent NOG [54
] in PBS thereby preserving probable ionic interactions. As shown in Figure 7
, the tested granule proteins eluted as distinct peaks at the positions expected by their molecular weights and known dimer forms and subunits. Furthermore, the known binding of a minor part of NGAL to gelatinase [57
] was detected as a high molecular weight peak by this method. This indicates that granule proteins of neutrophil granulocytes are stored as separate entities and not as a macromolecular complex with serglycin.
 View larger version (20K): [in a new window] |
Figure 7. Gel filtration by Superose 12. An -band and a ß-band from a subcellular fractionation of PMN were subjected to gel filtration. Fractions were tested by ELISA for granule proteins. V0, Void volume; MPO, myeloperoxidase, 80 kDa; Lys, lysozyme, 14 kDa; NGAL, 25 kDa (monomer, homodimer, and a minor part linked to gelatinase, representing the high molecular weight peak); Gela, gelatinase, 92 kDa; LF, lactoferrin, 78 kDa; hCAP18, 19 kDa [55
, 56
].
|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Previously, the expression of serglycin in neutrophil granulocytes and their precursors has been studied at the mRNA level. Stellrecht et al. [15 ] have demonstrated the distribution of serglycin mRNA by in situ hybridization on cytospins from whole human bone marrow and peripheral blood. A decreasing expression during neutrophil maturation was shown. Our group has previously studied the expression of serglycin mRNA by Northern blotting with bone marrow cells representing different maturational stages of myeloid development [16 ]. A biphasic expression of serglycin mRNA with the highest expression in MM/MC and in peripheral blood neutrophil granulocytes was seen. The results presented here are a refinement of the technique for separation of bone marrow cells supplemented with MACS for depletion of non-neutrophil cells. Using this refined method, we showed a decreasing expression of serglycin mRNA from the highest level in MB/PM toward the lowest expression in mature neutrophil granulocytes (Fig. 1C) . The decreasing serglycin mRNA expression during neutrophil differentiation was even more pronounced, when the hybridization signals were normalized to cell numbers instead of 18S hybridization signals (Fig. 1D) . This is in accordance with the results reported by Stellrecth et al. [15 ]. The divergence between our previous and present results can be explained by contamination of mature neutrophil granulocytes with eosinophils and basophils in the previous report, as these cells express serglycin mRNA at a level seven times higher than neutrophil granulocytes [13 , 15 ]. Also, by the former method of preparation, the MB and PM in band 3 were contaminated with lymphocytes, which express serglycin mRNA at a very low level [15 ]. We have previously shown that MACS depletion reduced contaminating cells in band 3 from 68% to 8% and that eosinophils were reduced from 4% to 0.7% in neutrophil granulocytes [40 ].
In this study, we have demonstrated a clear immunoreactivity for serglycin in the Golgi area of PM and MC at the level of light microscopy. In addition to the Golgi signal, we have found a fainter granular staining in some cells. These findings are in agreement with a previous report on serglycin immunoreactivity in myeloid HL-60 cells [19 ]. In more mature precursors and in neutrophil granulocytes, the immunoreactivity for serglycin was scarce. The decreasing immunoreactivity for serglycin with more differentiated cells is in full accordance with our Northern blotting results presented here.
Our immunoelectron microscopic studies further confirmed and expanded the light microscopic findings. The Golgi stacks and trans Golgi network of PM and MC demonstrated serglycin immunoreactivity localized close to the Golgi cisternae. A few, presumably immature, granules with disperse matrix also showed immunoreactivity for serglycin. No mature, electron-dense granules showed immunoreactivity for serglycin, and the majority of granules at all maturational stages of neutrophil development was negative for serglycin immunoreactivity. In mature neutrophil granulocytes, only a very faint signal was localized to the Golgi apparatus in some cells. Parmley et al. [58 ] have studied the ultrastructural localization of glycosaminoglycans in neutrophil granulocytes and precursors. They found evidence for glycosaminoglycan reactivity localized to the Golgi apparatus, immature primary granules, and tertiary granules by cytochemical methods. The reactivity in the Golgi apparatus and in immature granules may well be representative of serglycin proteoglycan; it was stated in their report that the glycosaminoglycans must be attached to a protein core, as they were not extracted during the staining procedure. In another work from the same group, 35SO4 was initially (2 h) localized to the Golgi apparatus and immature granules, and the radioactivity was predominantly localized to mature granules after a longer chase (18 h) [34 ]. Yang et al. [33 ] localized glycosaminoglycans to the trans Golgi network and immature electron-lucent primary and secondary granules of PM and MC by electron microscopy with no reactivity for glycosaminoglycan in more mature electron-dense granules and no reactivity in BC or mature neutrophil granulocytes. Their findings for glycosaminoglycans exactly parallel our localization of serglycin at the electron microscopic level. Payne and Ackerman [59 ] reported 35SO4 uptake primarily by PM and MM but not by MB and MC. This uptake may represent incorporation into serglycin proteoglycan, as 35SO4 is incorporated into the glycosaminoglycan side-chains, and the autoradiography located the radioactivity to the Golgi apparatus and nascent granules. This is also in accordance with our Northern blotting data, showing that the majority of serglycin mRNA is localized to PM and MB.
The failure to detect serglycin in mature granules could be explained by degradation of the serglycin proteoglycan by enzymes in the harsh microenvironment of maturing granules in neutrophil precursors. The degradation products might be free glycosaminoglycans or a glycosaminoglycan-bearing, short peptide. A total degradation of the core protein and glycosaminoglycan side-chains might also occur. It has previously been stated that serglycin proteoglycan is protease-resistant [6 , 60 ], but it is likely that these reports concern an N- and C-terminally truncated form, mostly consisting of the serine glycine repeat region with attached glycosaminoglycans [20 ]. The N-terminal part of serglycin proteoglycan is described to be rapidly degraded in HL-60 myeloid cells [5 ], and it seems probable that most of the HL-60 intracellulary serglycin is truncated at the C- and N-terminus [61 ]. This was supported by the recent findings by Lemansky et al. [62 ], who reported a half-life of 15 min for sulfate-labeled material in granules of myeloid HL-60 cells, which were precipitated by trichloroacetic acid (TCA), whereas the half-life of precipitated glycosaminoglycan-bearing molecules were several hours. This indicates that the core protein of these molecules, which might well be serglycin, is readily truncated or degraded and therefore, cannot be precipitated by TCA, whereas the glycosaminoglycans are degraded at a slower rate.
Another explanation for the lack of serglycin immunoreactivity in mature granules might be that binding partners inside the granules mask the epitopes, as it has been proposed previously [34 ]. To rule out whether granule proteins inside granules are binding to macromolecules, we subjected subcellular fractions of neutrophil granulocytes to gel filtration by a Superose 12 column. All the tested granule proteins eluted as distinct peaks in agreement with their known molecular weights, as seen in Figure 7 . The binding of a minor part of NGAL to gelatinase as well as the homodimer of NGAL was preserved during solubilization with the nonionic detergent NOG, as demonstrated by the three elution peaks for NGAL. The solubilization was performed with NOG in PBS to resemble the ionic strength and composition inside the phagocytic vacuole, described by Segal and colleagues [63 ]. We could not find evidence for the presence of lactoferrin and myeloperoxidase in a macromolecular form under these conditions, as suggested in their work [63 ]. These findings make a binding of granule proteins to a macromolecule inside the mature granules very unlikely and hereby, also preclude the speculations that the low or absent immunoreactivity for serglycin in primary and secondary granules might be a result of epitope masking by binding partners. Furthermore, we were unable to detect serglycin immunoreactivity in the secreted granule contents after exocytosis stimulation with PMA or ionomycin. Thereby, our results cannot support the assumption [63 ] that granule proteins are released from serglycin proteoglycan upon excretion.
Our findings presented here, taken together with the findings by Yang et al. [33 ] that glycosaminoglycan reactivity is restricted to the trans Golgi network and immature primary and secondary granules, seem to clarify that serglycin proteoglycan is not present in mature granules of neutrophil granulocytes.
Faint serglycin immunoreactivity was seen in the Golgi area of some mature neutrophil granulocytes by imunocytochemistry (Fig. 3G)
and by immunoelectron microscopy (Fig. 4C)
in our study. ELISA supported the presence of serglycin in the Golgi apparatus from some neutrophil granulocytes by revealing serglycin immunoreactivity restricted to the Golgi-containing -band from subcellular fractionation of neutrophil granulocytes.
The ELISA showed no significant difference in immunoreactivity whether the subcellular fractions were pretreated with cABC or not. Furthermore, our ELISA recognized serglycin in HL-60 medium without prior cABC digestion, where no serglycin could be detected by Western blotting (Fig. 2C , lane 2). This serglycin immunoreactivity must originate from the serglycin proteoglycan, which cannot be blotted [24 ]. These findings indicate that the anti-serglycin Ig can recognize serglycin proteoglycan and the core protein. This is in contrast to a previous finding, where a polyclonal rabbit Ig, raised against a synthetic 20-amino acid peptide from the mouse serglycin sequence juxtaposed to the serine glycine repeat region, could only recognize the cABC-treated core protein [18 ]. Our antiserglycin Ig was raised against recombinant serglycin, encompassing the entire sequence of the serglycin core protein, which explains the capacity to recognize serglycin proteoglycan and the core protein, as epitopes away from the glycosylated part of the proteoglycan will be available for recognition.
By Western blotting, the serglycin core protein was recognized solely in the 0.45 M NaCl anion exchange chromatography fraction of a -band from a subcellular fractionation of neutrophil granulocytes. The size of the band was as previously reported from a neutrophil granulocyte -band [53
]. These findings further support the localization of serglycin to the Golgi apparatus in a minor part of mature neutrophil granulocytes.
Previous studies on the functions of serglycin have given insight to different functions depending on cell type and type of glycosaminoglycan side-chains. In connective tissue-type mast cells, which synthesize serglycin with heparan sulfate, inhibition of sulfation of heparin confers morphological changes and a severe decrease in the amount of stored granule proteins [27 , 28 ]. In cytotoxic cells, serglycin is forming a complex with granzyme B and perforin, inside the granules and during delivery to target cells as a macromolecular complex [22 ]. In exocrine pancreas, serglycin has been shown to be implicated in granule formation and allocation of granule proteins to granules [25 ].
The immunocytochemical localization of serglycin during neutrophil differentiation reported here can shed light on possible functions for serglycin during neutrophil development. The immunoreactivity for serglycin was predominantly localized to the Golgi apparatus in PM and MC, which are known to produce primary granules and secondary granules, respectively [29 ]. A function for serglycin in assembling granules and granule contents from the trans Golgi network is thus in agreement with these findings. The finding that serglycin is localized to the Golgi cisternae and the trans Golgi network as well as a few, presumably immature, granules at the electron microscopic level supports this. It seems more unlikely that serglycin is implicated in neutralizing charges and decreasing osmotic strength of granule proteins, as we showed that granule proteins are stored as separate entities and that serglycin is not present in the granules of the mature neutrophil granulocyte.
Received October 24, 2003; revised March 16, 2004; accepted April 8, 2004.
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
in a subpopulation of cytoplasmic granules Blood 90,1255-1266This article has been cited by other articles:
 |
|
 |
|
  D. S. Woulfe, J. K. Lilliendahl, S. August, L. Rauova, M. A. Kowalska, M. Abrink, G. Pejler, J. G. White, and B. P. Schick Serglycin proteoglycan deletion induces defects in platelet aggregation and thrombus formation in mice Blood, April 1, 2008; 111(7): 3458 - 3467. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. U. Niemann, M. Abrink, G. Pejler, R. L. Fischer, E. I. Christensen, S. D. Knight, and N. Borregaard Neutrophil elastase depends on serglycin proteoglycan for localization in granules Blood, May 15, 2007; 109(10): 4478 - 4486. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 P. Lemansky, I. Fester, E. Smolenova, C. Uhlander, and A. Hasilik The cation-independent mannose 6-phosphate receptor is involved in lysosomal delivery of serglycin J. Leukoc. Biol., April 1, 2007; 81(4): 1149 - 1158. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
O. Ramilo, W. Allman, W. Chung, A. Mejias, M. Ardura, C. Glaser, K. M. Wittkowski, B. Piqueras, J. Banchereau, A. K. Palucka, et al. Gene expression patterns in blood leukocytes discriminate patients with acute infections Blood, March 1, 2007; 109(5): 2066 - 2077. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
N. Borregaard, K. Theilgaard-Monch, J. B. Cowland, M. Stahle, and O. E. Sorensen Neutrophils and keratinocytes in innate immunity--cooperative actions to provide antimicrobial defense at the right time and place J. Leukoc. Biol., April 1, 2005; 77(4): 439 - 443. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
