Glycobiology Advance Access originally published online on March 24, 2004
Glycobiology 2004 14(7):609-619; doi:10.1093/glycob/cwh063
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Glycobiology vol. 14 no. 7 © Oxford University Press 2004; all rights reserved.
Structural variability of BM-40/SPARC/osteonectin glycosylation: implications for collagen affinity
2 Center for Biochemistry, Medical Faculty, University of Cologne, Joseph-Stelzmann-Str. 52, D-50931 Cologne, Germany; and 3 Center for Molecular Medicine, Medical Faculty, University of Cologne, Joseph-Stelzmann-Str. 52, D-50931 Cologne, Germany
Received on November 26, 2003; revised on February 12, 2004; accepted on February 14, 2004
 |
Abstract |
|---|
 Top Abstract IntroductionResultsDiscussionMaterials and methodsReferences |
|---|
We performed a detailed investigation of N-glycan structures on BM-40 purified from different sources including human bone, human platelets, mouse Engelbreth-Holm-Swarm (EHS) tumor, and human BM-40 recombinantly expressed in 293 and osteosarcoma cells. These preparations were digested with endoglycosidases and N-glycans were further characterized by sequential exoglycosidase digestion and high-performance liquid chromatography (HPLC) analyses. Bone BM-40 carries high-mannose structures as well as biantennary complex type N-glycans, whereas the protein from platelets and 293 cells has exclusively bi- and triantennary complex type structures. BM-40 derived from the EHS tumor carries biantennary complex type and additional hybrid structures. Using the osteosarcoma-derived MHH-ES1 cell line we successfully expressed a recombinant BM-40 that bears at least in part the bone-specific high-mannose N-glycosylation in addition to complex type and hybrid structures. Using chromatography on Concanavalin-A Sepharose, we further purified a fraction enriched in high-mannose structures. This array of differentially glycosylated BM-40 proteins was assayed by surface plasmon resonance measurements to investigate the binding to collagen I. BM-40 carrying high-mannose structures binds collagen I with higher affinity, suggesting that differentially glycosylated forms may have different functional roles in vivo.
Key words: bone / complex type / high mannose / MHH-ES1 cells / platelets
|
Introduction |
|---|
|
TopAbstractIntroductionResultsDiscussionMaterials and methodsReferences |
|---|
BM-40 (also known as secreted protein, acidic, rich in cysteines [SPARC] or osteonectin) is an extracellular glycoprotein with affinity for calcium ions and collagens (Maurer et al., 1996
). Among the numerous functions described for BM-40 are regulation of extracellular matrix organization, binding of growth factors, and inhibition of cell adhesion and spreading (Bradshaw and Sage, 2001). BM-40 has a modular structure made up from an N-terminal domain containing about 50 amino acid residues, out of which 18 are negatively charged, followed by a follistatin-like domain with 10 cysteines in a typical pattern and an extracellular calcium-binding domain with two EF-hand calcium binding motifs, each with a bound calcium in the X-ray structure (Hohenester et al., 1996; Maurer et al., 1995). BM-40 forms the prototype of a protein family characterized by containing both extracellular calcium-binding and follistatin-like domains and often sharing functional features, such as calcium and collagen affinity. Other members of this family are SC1/hevin/QR1 (Johnston et al., 1990), tsc36/Flik/FRP (Shibanuma et al., 1993), testican-1 (Alliel et al., 1993), testican-2 (Vannahme et al., 1999), testican-3 (Hartmann and Maurer, 2001), SMOC-1 (Vannahme et al., 2002), and SMOC-2 (Vannahme et al., 2003).
The glycoprotein nature of BM-40 was recognized directly on discovery of the protein (Termine et al., 1981), and analysis of the bovine and human protein sequences yielded two consensus sequences for potential N-glycosylation at Asn71 and Asn99 (Bolander et al., 1988; Villareal et al., 1989), but there was no evidence for the presence of O-linked oligosaccharides. Later studies indicated that only the N-glycan acceptor site at Asn99 is used (Hohenester et al., 1997; Xie and Long, 1995). This site is widely conserved among species, often with the sequence Asn-Lys-Thr followed by either Phe or Tyr, for example, in chicken (Bassuk et al., 1993), rabbit (Bluteau et al., 2000), mouse (Lankat-Buttgereit et al., 1988; Mason et al., 1986), and Xenopus laevis (Damjanovski et al., 1992). A potential N-glycosylation site in a similar position but with another sequence is found in BM-40 from rainbow trout (Tang and McKeown, 1995) and brine shrimp (Tanaka et al., 2001), and BM-40 from Caenorhabditis elegans contains a site at Asn80 (Schwarbauer and Spencer, 1993).
BM-40 from bone and platelets migrate differently in sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE), and enzymatic digestion with N-glycosidase F and endoglycosidase H showed that the difference is due to variable N-glycosylation with the bone-derived protein carrying predominantly high-mannose oligosaccharides, whereas those on platelet BM-40 are mainly of the complex type (Kelm and Mann, 1991). The difference in type of N-glycans appeared to have functional consequences, as the bone form of the protein bound to collagen I, III, and V in solid phase assays, whereas the platelet form did not. The differences were confirmed in independent studies on BM-40 binding to collagen V, and it was also shown that removal of N-linked oligosaccharides by N-glycosidase F treatment increased the affinity of BM-40 from both bone and platelets to collagen V to reach equal levels. Site- directed mutagenesis was used to show that only the glycosylation of Asn99 affects collagen binding (Xie and Long, 1995).
The collagen affinity of BM-40 is of great functional importance. The collagen I-deficient Mov-13 mouse line does not retain BM-40 in the extracellular matrix (Iruela-Arispe et al., 1996) and dermal collagen fibrils in BM- 40-null mice have a decreased diameter and a reduced tensile strength (Bradshaw et al., 2003). Not only glycosylation but other forms of posttranslational modification of BM-40 modulate the collagen affinity. Limited proteolytic cleavage of BM-40 occurs in tissues, and treatment with matrix metalloproteinases increases its affinity to collagens 7–20-fold due to a cleavage in helix 
C in the extracellular calcium-binding domain (Sasaki et al., 1997). Indeed, deletion of helix C in recombinant BM-40 gave a similar increase in binding affinity, and by X-ray crystallography this could be shown to be due to the removal of a steric constraint on the binding site, which was mapped to a loop between the two EF-hands in the extracellular calcium-binding domain (Sasaki et al., 1998).
Variable glycosylation provides an opportunity to vastly increase the information of a concise genome, and it has been demonstrated previously that the function of a protein can be modulated by its glycosylation pattern. In experimental tumor models the most common aberrant N-glycosylations are an increase in terminal sialylation and a shift to more highly branched N-linked oligosaccharides; it has been shown that the metastatic potential of tumor cells correlates with these changes (Hakomori, 1996; Varki, 1993; Yamamoto et al., 2000). Glycosylation has also been implicated in the regulation of CD44-mediated cell binding of hyaluronan. Interestingly, N-linked oligosaccharides can both enhance and reduce the CD44 affinity for hyaluronan depending on the specific structure of the glycan (Skelton et al., 1998).
In the present work we analyze the structure of the N-glycans in BM-40 derived from bone, platelets, and a variety of osteosarcoma cell lines. By expressing human BM-40 recombinantly in different cell lines, we were able to prepare BM-40 forms of variable glycosylation and could show that BM-40 carrying high-mannose N-glycans indeed binds collagen I with higher affinity than other forms. Our work shows how tissue-specific variation in glycosylation of a given protein may modulate its biological function.
|
Results |
|---|
|
TopAbstractIntroductionResultsDiscussionMaterials and methodsReferences |
|---|
N-glycosylation of well-characterized forms of BM-40
In a first step we performed a detailed characterization of N-glycosylation of BM-40 purified from human bone, human platelets, the mouse Engelbreth-Holm-Swarm sarcoma (EHS) tumor, and human BM-40 recombinantly expressed in human embryonic kidney 293 cells. These are the most commonly studied preparations of BM-40, and their analysis provides information on the spectrum of N-glycan variants encountered in different mammalian tissues. Initially, highly purified BM-40 preparations were subjected to digestion with N-glycosidase F or endoglycosidase H followed by SDS–PAGE to assess the overall extent of N-glycosylation and to identify BM-40 forms carrying endoglycosidase H–sensitive high-mannose glycans (Figure 1). In each case digestion with N-glycosidase F resulted in a slight shift in electrophoretic mobility, corresponding to a mass loss of
2 kDa. Only the BM-40 purified from human bone was sensitive to endoglycosidase H, indicating the presence of high-mannose glycans (Figure 1A). For BM-40 from the mouse EHS tumor, a degradation band was observed, in addition to the intact protein (Figure 1C). This well-characterized BM-40 fragment results from cleavage by a protease present in the EHS tumor tissue (Maurer et al., 1997).
|
The same BM-40 samples were also analyzed by matrix-assisted laser desorption/ionization time-of-flight mass spectrometry (MALDI-TOF MS), without and with prior digestion with N-glycosidase F (Figure 2). Again a shift was seen for each sample, showing that all carried N-glycans. For the recombinant human BM-40 expressed in 293 cells the peak had a leading shoulder, both before and after deglycosylation, indicating a microheterogeneity that is not due to the N-glycan portion. The mass lost by N-glycosidase F digestion was between 1.7 kDa (bone BM-40) and 2.7 kDa (platelet BM-40), in good agreement with the expected substitution with a single N-glycan. The protein molecular mass, measured after removal of N-glycans, was identical to the theoretical mass calculated from the amino acid sequence for BM-40 from the EHS tumor and for the main portion of the recombinant BM-40 sample. BM-40 from bone and platelets and
40% of the recombinant BM-40 molecules from 293 cells gave masses after deglycosylation that were 0.4 and 0.7 kDa higher than the theoretical one, indicating that additional posttranslational modifications may be present.
|
N-glycans were cleaved from the various BM-40 samples with N-glycosidase F and labeled at their reducing ends with the fluorescent dye 2-aminobenzamide. The presence and number of terminal sialic acid residues was determined by anion-exchange chromatography on a Q Hyper D10 high-performance liquid chromatography (HPLC) column and comparison of the elution profiles with those of standard glycans (Figure 3, Table I). The extent of sialic acid substitution varied markedly between BM-40 forms, with BM-40 from bone and recombinant BM-40 from 293 cells often carrying neutral N-glycans, whereas more than half of the N-glycans from platelet BM-40 carried up to three sialic acid residues (Table I). The highly variable sialic acid content of various BM-40 preparations is likely to modulate charge-dependent interactions.
|
|
In a next step, sialic acid residues were removed by digestion with neuraminidase and the asialo-N-glycans separated according to size and hydrophilic properties by HPLC on a calibrated normal phase column (Figure 4). The N-glycan structures were determined by sequential digestion with different exoglycosidases (ß-galactosidase, ß-N-acetylhexosaminidase, and
-fucosidase) and the retention times were compared with commercially available carbohydrate standards (Novatec Analytical, Basel, Switzerland).
|
|
The analysis suggests that although about half of all N-glycans in both bone and platelet BM-40 are of the complex biantennary type, the other half are of the high-mannose type, and a corresponding portion in platelet BM-40 have a complex triantennary structure (Table II). The N-glycans from EHS tumor BM-40 are to 80% of complex biantennary type, with the rest being of hybrid structure, and the bulk of those on the recombinant BM-40 from 293 cells are of the complex triantennary type with a small contribution of complex biantennary structures. The latter sample showed a marked undergalactosylation, with the N-glycans carrying no or at the most one galactose residue, even in the triantennary structures. Accordingly, the BM-40 N-glycan structure varies markedly dependent on tissue or cell source and the N-glycosylation on recombinant BM-40 from 293 cells does not closely reflect any of the natural forms studied.
Recombinant production of BM-40 with a bone-specific N-glycosylation
The observation that BM-40 N-glycosylation is highly tissue-specific caused us to attempt the recombinant production of a BM-40 with a high-mannose type N-glycosylation similar to that found in human bone. For this purpose we constructed an expression vector based on the episomal eucaryotic expression vector pCEP-Pu (Kohfeldt et al., 1997), containing the human BM-40 cDNA, including the signal peptide, and in addition coding for an N-terminal extension of six histidines followed by a myc sequence and a Factor X cleavage site (Wuttke et al., 2001). To identify a cell line expressing a set of glycosyl transferases similar to that of a human osteoblast in situ, we screened the human osteosarcoma cell lines SaOS-2, MG-63, U2-OS, HOS, and MHH-ES-1 as well as the rat osteosarcoma cell line UMR-108 for the N-glycosylation pattern of their endogenously produced BM-40 by parallel digestion with N-glycosidase F and endoglycosidase H (data not shown). Among all these cell lines, only the MHH-ES-1 cells produced BM-40 with a partial susceptibility to endoglycosidase H cleavage, indicating the presence of high-mannose type N-glycans (Figure 5). These cells were then transfected with the expression plasmid, recombinant his-myc tagged BM-40 harvested from the serum-free medium, and purified by affinity chromatography on a matrix carrying immobilized cobalt.
|
As endogenous BM-40 secreted by the MHH-ES-1 cells was only partially susceptible to digestion with endoglycosidase H, we expected recombinant BM-40 produced by these cells to be a mixture of different glycoforms. To enrich BM-40 with a high-mannose type glycosylation we therefore fractionated the material by chromatography on a Concanavalin A (Con A) Sepharose column. The lectin Con A binds tightly to oligosaccharides carrying terminal mannose or glucose residues. After sample application and wash with a sugar-free buffer, weakly bound glycoforms were eluted with 15 mM methyl
-glucoside (fraction 1) and tightly bound forms with 500 mM methyl -glucoside (fraction 2). SDS–PAGE showed that recombinant BM-40 was contained in the flow-through and in both eluted fractions and that only the material from fraction 2 was sensitive to endoglycosidase H (Figure 6A). The fractions eluted with methyl -glucoside were analysed further by digestion with N-glycosidase F or endoglycosidase H followed by MALDI-TOF MS (Figure 6B, 6C). All BM-40 molecules eluted in fraction 1 were sensitive to digestion with N- glycosidase F but not to endoglycosidase H (Figure 6B), whereas those in fraction 2 showed marked shifts with both enzymes, indicating the presence of high-mannose N-glycans.
|
The N-glycans on recombinant BM-40 produced in MHH-ES-1 cells were analyzed for sialic acid content and oligosaccharide structure in the same manner as described for other BM-40 forms. Anion-exchange chromatography showed that 47.5% were neutral, and 30.5% carried one sialic acid residue, 15.8% two, and 6.2% three. Sequential exoglycosidase digestion followed by normal-phase HPLC showed the sample to be a mixture of complex bi- or triantennary N-glycans as well as hybrid and high-mannose types (Figure 7A, Table III). Parallel analysis of fraction 2, the BM-40 fraction that bound tightly to Con A-Sepharose, showed this material to be enriched in molecules carrying high-mannose oligosaccharides and in this regard more similar to bone-derived BM-40 than any other fraction (Figure 7B, Table III).
|
|
Affinity of different BM-40 glycoforms to collagen I
To test the suggestion that the nature of the N-glycans influences the collagen affinity of BM-40 (Kelm and Mann, 1991
), we determined the binding of four different BM-40 preparations to immobilized collagen I in a Biacore apparatus (Figure 8).
|
First we used increasing concentrations of BM-40 from 293 cells ranging from 2.4 to 4.5 µM to validate the measurements. The resulting curves were aligned and clearly demonstrate a concentration-dependent increase in response units (Figure 8A). Compared to the EBNA-293 derived BM-40 the preparations from bone and fraction 2 from MHH-ES-1 cells showed a markedly slower dissociation (Figure 8B). The binding and dissociation curves obtained allowed the calculation of approximate KD values (Table IV). Bone-derived BM-40 showed the highest affinity, with a KD of 15.5 nM, followed by recombinant BM-40 from MHH-ES-1 cells that had been enriched for high-mannose N-glycan structures by chromatography on Con A-Sepharose (fraction 2), with KD 130 nM. Fraction 1 from the Con A-Sepharose chromatography, as well as recombinant BM-40 from 293 cells, predominantly carrying complex type bi- and triantennary N-glycans, showed lower affinity for collagen I, with KD values of 448 and 1.520 nM, respectively. MHH-ES-1 derived BM-40 recovered in the flow-through of the Con A-Sepharose column and devoid of high-mannose N-glycan did not bind to the collagen I coupled chip (data not shown).
|
|
Discussion |
|---|
|
TopAbstractIntroductionResultsDiscussionMaterials and methodsReferences |
|---|
Analysis by MALDI-TOF MS before and after digestion with N-glycosidase F did for all BM-40 variants gave a mass difference between 1.7 and 2.7 kDa, in good agreement with a substitution with a single N-glycan. Our experiments did not address the position of this oligosaccharide, but earlier studies indicate a preferential or unique substitution of Asn99 (Hohenester et al., 1997
; Xie and Long, 1995). Comparison of the molecular masses obtained after digestion with N-glycosidase F with the theoretical one, calculated from the amino acid sequence, showed that BM-40 from the MHH-ES-1 cells and from the EHS tumor carry no posttranslational modifications in addition to N-glycans, whereas BM-40 from bone and platelets gave masses after deglycosylation that were 0.4 and 0.7 kDa higher than the theoretical ones. The recombinant BM-40 preparation from 293 cells contained both molecules with additional mass after deglycosylation and such without any more modifications.
Even though our study was not primarily aimed at determining the nature of these adducts, we determined the incorporation of 32P-phosphate and 35S-sulfate into BM-40 produced by 293, MHH-ES-1, and SaOS-2 cells, in each case with negative results (data not shown). Even so, phosphorylation has been demonstrated for bone-derived BM-40 (Fisher et al., 1987; Kelm and Mann, 1991; Romberg et al., 1985) and tyrosine sulfation has been suggested (Maillard et al., 1992). There is no evidence from other studies for the presence of O-glycans, even though the software Net-O-Glyc indicates the presence of two potential O-glycosylation sites (data not shown). BM-40 is a substrate for tissue transglutaminase both in vivo (Aeschlimann et al., 1995) and in vitro (Hohenadl et al., 1995), and it is possible that additional mass is contributed by incorporation of low-molecular-mass amines. The presence and potential functions of BM-40 adducts other than N-glycans needs further study.
A starting point for our work were the results from Kelm and Mann (1991) who, through a combination of endoglycosidase digestion and lectin binding studies, proposed that bone BM-40 carries high-mannose N-glycans and platelet BM-40 complex type ones. Our results partially confirm theirs but show a greater variety of N-glycans in BM-40 from each source. Indeed, our experiments showed a rather pronounced sensitivity of bone derived BM-40 to endoglycosidase H when the digests were analyzed by SDS–PAGE (Figure 1), but a detailed analysis of the glycans released by N-glycosidase F showed the presence of about equal amounts of high-mannose and biantennary complex type N-glycans in human bone BM-40 (Figure 4 and Table II). This also demonstrates that endoglycosidase digestion followed by SDS–PAGE provides basic information but does not allow definitive conclusions on the N-glycan structures present. Human platelet BM-40, on the other hand, carries bi- and triantennary complex type oligosaccharides in equal proportions. In agreement with earlier analysis (Nischt et al., 1991) our results show that the much studied recombinant BM-40 produced by 293 cells is similar to platelet BM-40 in that it carries a mixture of bi- and triantennary complex type N-glycans, but the proportion of biantennary structures is much lower and both forms are markedly undergalactosylated. Therefore 293 cell-derived BM-40 differs from both bone- and platelet-derived BM-40 in its glycosylation and presumably also in such biological features that are affected by glycosylation. BM-40 from the mouse EHS tumor was unique in that it carries hybrid N-glycans in addition to biantennary complex type ones.
The highly variable N-glycosylation of BM-40 derived from different sources indicates that the type of glycosylation is dependent on the glycosyl transferase repertoire of the producing cell type, rather than being directed by the acceptor protein. However, in earlier studies we analyzed the glycosylation of bone sialoprotein expressed in the same 293-cell system as the BM-40 studied in the present work. The N-glycans synthesized onto bone sialoprotein by 293 cells clearly differed from those on BM-40 produced in the same cells and were mainly tetraantennary complex type structures (Wuttke et al., 2001). The glycosylation pattern is obviously determined by both the acceptor protein and by the cell type used for expression.
One purpose of our study was to recombinantly produce a BM-40 variant that as closely as possible mimics the BM-40 found in human bone. On the basis of the arguments above, we expected to be able to do so by adapting our expression system to use in an osteoblast-like cell line. For this purpose we selected a number of well-characterized osteosarcoma cell lines. We were surprised to find that among those only one, MHH-ES-1, produced BM-40 with high-mannose type oligosaccharides. Apparently, the pattern of N-glycans produced is highly sensitive to transformation and/or dedifferentiation. Even in MHH-ES-1 cells a broader variety of N-glycans were produced than found on BM-40 in human bone (Table II and Table III) and chromatography on Con A–Sepharose was needed to prepare a fraction of BM-40 molecules enriched in high-mannose N-glycans that allowed us to test the hypothesis that this kind of glycosylation favors interactions with collagen I. Our Biacore binding studies indeed supported this assumption, showing a higher affinity of the high-mannose N-glycan-containing BM-40 variants from human bone and from the Con A-binding fraction of the MHH-ES-1 derived molecules to collagen I than of other forms, even though the binding by bone-derived BM-40 was for unknown reasons stronger than that of the MHH-ES-1 BM-40. Both preparations showed similar characteristics in circular dichroism spectroscopy, indicating a similar and native fold (data not shown). Possibly the minor additional posttranslational modifications found in bone BM-40 have a favorable influence on collagen affinity. The effect of N-glycans on the binding is likely to be mediated through steric or charge influences on the neighboring collagen-binding site, with the oligosaccharide hindering or promoting the docking of the BM-40 molecule onto the collagen triple helix.
|
Materials and methods |
|---|
|
TopAbstractIntroductionResultsDiscussionMaterials and methodsReferences |
|---|
Purification of tissue-derived forms of BM-40
Human bone powder (300 g) was stirred at 4°C for 48 h in 1 L 4 M guanidine hydrochloride, 50 mM Tris–HCl, pH 7.4, to remove residues of blood and bone marrow. The residue was washed twice for 24 h with 1.5 l 50 mM Tris–HCl, pH 7.4, and extracted three times for 24 h with 1.5 L 0.5 M ethylenediamine tetra-acetic acid (EDTA), 50 mM Tris–HCl, pH 7.4, containing the protease inhibitors 0.1 M 6-aminohexanoic acid, 5 mM benzamidine hydrochloride, 0.5 mM phenylmethylsulfonyl fluoride, and 5 mM N-ethylmaleimide. The three extracts were pooled, dialyzed against 50 mM Tris–HCl, pH 8.6, and loaded onto a column of DEAE-Sepharose Fast Flow (Amersham Biosciences, Little Chalfont, U.K.). BM-40 eluted at about 0.2 M NaCl in a linear gradient from 0 to 1 M. Pooled BM-40-containing fractions were rechromatographed on a ResourceQ (Amersham Biosciences) column under the same conditions and eluted at 0.25–0.3 M NaCl. BM-40-containing fractions were dialyzed against 50 mM Tris–HCl, pH 7.0, and applied to Hitrap Heparin column. BM-40 bound weakly and was eluted at 0.1–0.15 M NaCl.
Mouse EHS tumor tissue was sequentially extracted with Tris-buffered saline (TBS) and TBS containing 10 mM EDTA and BM-40 chromatographically purified from the EDTA extract as previously described (Mann et al., 1987). Purified human platelet osteonectin (BM-40) was obtained commercially from Calbiochem (San Diego, CA).
Expression and purification of recombinant forms of BM-40
For cloning of the expression vector pCEP-Pu/BM-40, the full-length human BM-40 cDNA including its signal peptide was cut out of the plasmid huBM-40blue (Nischt et al., 1991) using Not I and Xho I restriction sites and inserted between the same restriction sites of the episomal eucaryotic expression vector pCEP-Pu (Kohfeldt et al., 1997). For expression of a His6-Myc-Factor X–tagged fusion protein, the human BM-40 cDNA sequence without signal peptide was amplified using the huBM-40blue plasmid as a template and cloned into a N-terminal His6-Myc-Factor X–tagged expression vector based on the pCEP-Pu (Wuttke et al., 2001). The correct reading frame was confirmed by sequencing both strands using the ABI Prism 377 Automated Sequencer (PE Biosystems, Foster City, CA). Human embryonic kidney cells EBNA-293 (Invitrogen, Carlsbad, CA) were transfected with the pCEP-Pu/BM-40 plasmid using the Fugene transfection reagent (Roche, Indianapolis, IN). Transfected cells were selected with puromycin as described (Kohfeldt et al., 1997). Serum-free culture supernatants were collected, dialyszd against 50 mM Tris-HCl, pH 8.6, passed over a DEAE-Sepharose Fast Flow column, and rechromatographed on a ResourceQ column as described for the bone-derived protein.
MHH-ES-1 cells (German Collection of Microorganisms and Cell Cultures) were transfected with the expression plasmid encoding the N-terminal His6-Myc-Factor X–tagged BM-40 using Fugene transfection reagent. Transfected cells were selected with 1 µg/ml puromycin in RPMI 1640 medium containing 10% fetal calf serum and 2 mM L-glutamine. The fusion protein was purified from serum-free culture supernatants by using metal-affinity chromatography. The supernatant was dialyzed against 50 mM sodium phosphate, pH 7.0, containing 0.3 M NaCl and loaded on a column of Talon metal affinity resin (Clontech, Palo Alto, CA) with a flow rate of 0.5 ml/min. After washing with five column volumes of the same buffer containing 2.5 mM imidazol, the protein was eluted with a linear gradient between 2.5 and 250 mM imidazol. BM-40-containing fractions were pooled, dialyzed against 50 mM Tris–HCl, pH 8.6, and loaded onto a ResourceQ (Amersham Biosciences) column. BM-40 eluted at about 0.3 M NaCl in a linear gradient from 0 to 1 M.
SDS–PAGE and western blotting
SDS–PAGE was performed according to Laemmli (1970) and gels either stained with Coomassie brilliant blue R250 or electrophoretically transferred to nitrocellulose. BM-40 was detected with polyclonal rabbit antisera against either human or mouse BM-40 followed by a peroxidase-conjugated antibody against rabbit IgG (Dako, Highwycombe, U.K.) and the reaction developed with the enhanced chemoluminescence method (Amersham Biosciences).
Analytical endoglycosidase digestions
Prior to SDS–PAGE analysis, glycoprotein samples (2–3 µg) were heat-denaturated for 15 min at 100°C and incubated over night at 37°C with 0.2 U N-glycosidase F (Roche) in 0.15 M NaCl, pH 7.4, containing 0.1% SDS and 0.5% Nonidet P-40 or with 2 mU endoglycosidase H (Roche) in 0.1 M sodium acetate, pH 5.8, containing 0.02% SDS and 0.1% ß-mercaptoethanol.
N-glycosidase F release of N-glycans, sequential exoglycosidase digestions, and HPLC analyses of 2-aminobenzamide-labeled oligosaccharides
A previously described procedure was followed (Wuttke et al., 2001). Briefly, 200 µg ethanol-precipitated BM-40 was incubated in 2 µl 1% SDS, 0.5% ß-mercaptoethanol, 0.1 M EDTA for 30 min at room temperature. After the addition of 40 ml 0.2 M sodium phosphate, pH 8.5, followed by a 5-min denaturation at 100°C, 5 µl 7.5% Nonidet P-40 was added to the mixture. The sample was incubated with 1 U N-glycosidase F (Roche) for 18–20 h at 37°C and then applied to a 150-mg carbon column (Carbograph SPE, Alltech, Unterhaching, Germany) according to the method of Packer et al. (1998). The column was eluted with water and then with 2 ml 25% acetonitrile in 0.05% trifluoroacetic acid to elute the oligosaccharides, which were dried in a SpeedVac evaporator.
The glycans were labeled with the fluorescent dye 2- aminobenzamide according to the method described by Bigge et al. (1995). The dried glycans were resuspended in 2 µl 1 M 2-aminobenzamide in 100% acetic acid and 3 µl 2 M sodium cyanoborohydride in dimethyl sulfoxide. After a 2-h labeling reaction at 60°C, the samples were dotted onto chromatography paper, and excessive labeling reagents were separated by chromatography in n-butanol:ethanol:water (4:1:1). The labeled glycans do not migrate under these conditions. After chromatography the application points were cut out, and the labeled glycans eluted with 200–500 µl water by using centrifugal microfiltration tubes and stored at –20°C.
To determine the monosaccharide sequence of the N-linked oligosaccharides, the 2-aminobenzamide-labeled N-glycans were digested with specific exoglycosidases. Twenty microliters of each sample were incubated sequentially with neuraminidase (New England Biolabs, Beverly, MA), ß-N-acetylhexosaminidase (Glyko, Novato, CA), ß-galactosidase (Glyko), and -fucosidase (Glyko) for a total of 18–20 h at 37°C in 50 mM sodium citrate, pH 4.5. The digests were dried in a SpeedVac evaporator, dissolved in 75% acetonitrile in water, and used for HPLC analysis.
For analysis a Beckman System Gold HPLC station was used together with a Shimadzu RF-10A XL fluorescence detector and the Beckman Gold Noveau software. The excitation wavelength was 330 nm and the emission wavelength 420 nm. For anion-exchange HPLC a Q HyperD10 column (10 µm, 4.6 x 100 mm, Beckman, Palo Alto, CA) was used at a flow rate of 1 ml/min. Twenty microliters of the 2-aminobenzamide-labeled oligosaccharides were loaded to the column. Elution buffers were water and 0.5 M ammonium formate, pH 9.0 (buffer B). Sialo-N-glycans were eluted for 1 min with 0% B, for 12 min with 0–5% B, for 13 min with 5–21% B, for 25 min with 25–80% B, and for 4 min with 80–100% B.
Normal-phase HPLC was performed with a polymer-based aminophase column (astec NH2 polymer, 5 µm, 4.6 x 250 mm). The 2-aminobenzamide-labeled oligosaccharides (20 µl) were loaded onto the column dissolved in 75% acetonitrile in water. The linear elution gradient for N-glycans started at 68% acetonitrile and 32% 50 mM ammonium formate, pH 4.4, went over 60 min to 50% acetonitrile and 50% ammonium formate, pH 4.4, and in 3 min to 100% ammonium formate at a flow rate of 0.5 ml/min. Standard glycans (e.g., derived from bovine fetuin) and glucose ladders were used to identify the different structures.
MALDI-TOF MS
Samples of BM-40 (3 µg) were analyzed either in intact form or after digestion overnight at 37°C with 0.2 U N-glycosidase F (Roche) in 20 mM Tris–HCl, pH 7.4, or with 2 mU endoglycosidase H (Roche) in 30 mM sodium acetate, pH 5.8. For MALDI-TOF MS analysis, the samples were dissolved in 5 ml 0.1% aqueous trifluoroacetic acid. One milliliter of the sample solution was placed on the target, and 1 ml of a freshly prepared saturated solution of sinapinic acid in acetonitrile/H2O (2:1) with 0.1% trifluoroacetic acid was added. The spot was then recrystallized by addition of another 1 ml acetonitrile/H2O (2:1), which resulted in a fine crystalline matrix. MALDI-MS was carried out in linear mode on a Bruker Reflex IV equipped with a video system, a nitrogen UV laser (
max = 337 nm), and a HiMass detector. For recording of the spectra an acceleration voltage of 20 kV was used, and the detector voltage was adjusted to 1.9 kV. Three hundred to four hundred single laser shots were summed into an accumulated spectrum. Calibration was carried out using the single and doubly protonated ion signal of bovine serum albumin for external calibration.
Biacore studies of BM-40-collagen I interactions
Assays were performed using a Biacore 2000 (BIAcore AB, Uppsala, Sweden). Coupling of collagen I (native calf skin, IBFB) to the CM5 chip was performed in 50 mM sodium acetate, pH 4.0, at a flow rate of 5 µl/min. To activate the surface, a 7-min pulse of 0.05 mM N-hydroxy-succinimide/0.2 M N-ethyl-N'-dimethylaminopropyl carbodiimide was used. Collagen I (60 µl, 0.2 mg/ml) was injected until the desired amount was coupled (5000–7000 RU), and subsequently excess reactive groups were deactivated by a 7-min pulse of 1 M ethanolamine hydrochloride, pH 8.5. Measurements were carried out in HBS (20 mM HEPES, 150 mm NaCl, 0.005% P20, pH 7.4) containing 2 mM CaCl2 at a flow of 25 µl/min. The injection of 100 µl BM-40 solution (0.9–4.5 µM) was followed by a 400-s dissociation. The data were analyzed with BIAevaluation software 3.0 according to the Langmuir model for 1:1 binding and association and dissociation rate constants were determined and KD values were calculated. All binding curves could be fitted with an accuracy of 
2 < 0.2.
|
Acknowledgements |
|---|
This work was supported by the Bundesministerium für Bildung und Forschungî (Grant 0311443), the Deutsche Forschungsgemeinschaft (MA 1932/2 and PA 660/6) and the Köln Fortune program of the Medical Faculty of the University of Cologne.
|
Footnotes |
|---|
1 To whom correspondence should be addressed; e-mail: frank.zaucke{at}uni-koeln.de
|
Abbreviations |
|---|
Con A, Concanavalin A; EDTA, ethylenediamine tetra-acetic acid; EHS, Engelbreth-Holm-Swarm; HPLC, high-performance liquid chromatography; MALDI-TOF MS, matrix-assisted laser desorption/ionization time-of-flight mass spectrometry; SDS-PAGE, sodium dodecyl sulfate-polyacrylamide gel electrophoresis; TBS, Tris-buffered saline
|
References |
|---|
|
TopAbstractIntroductionResultsDiscussionMaterials and methodsReferences |
|---|
Aeschlimann, D., Kaupp, O., and Paulsson, M. (1995) Transglutaminase-catalyzed matrix cross-linking in differentiating cartilage: Identification of osteonectin as a major glutaminyl substrate. J. Cell Biol., 129, 881–892.
Alliel, P.M., Perin, J.P., Jolles, P., and Bonnet, F.J. (1993) Testican, a multidomain testicular proteoglycan resembling modulators of cell social behaviour. Eur. J. Biochem., 214, 347–350.[Web of Science][Medline]
Bassuk, J.A., Iruela-Arispe, M.L., Lane, T.F., Benson, J.M., Berg, R.A., and Sage, E.H. (1993) Molecular analysis of chicken embryo SPARC (osteonectin). Eur. J. Biochem., 218, 117–127.[Web of Science][Medline]
Bigge, J.C., Patel, T.P., Bruce, J.A., Goulding, P.N., Charles, P.N., and Parekh, R. (1995) Nonselective and efficient fluorescent labeling of glycans using 2-amino benzamide and anthranilic acid. Anal. Biochem., 230, 229–238.[CrossRef][Web of Science][Medline]
Bluteau, G., Mathieu, P., Conrozier, T., Vignon, E., Herbage, D., and Mallein-Gerin, F. (2000) Sequence deposited in SWISS-PROT. Accession number P36233.
Bolander, M.E., Young, M.F., Fisher, L.W., Yamada, Y., and Termine, J.D. (1988) Osteonectin cDNA sequence reveals potential binding regions for calcium and hydroxyapatite and shows homologies with both a basement membrane protein (SPARC) and a serine protease inhibitor (ovomucoid). Proc. Natl Acad. Sci. USA, 85, 2919–2923.
Bradshaw, A.D. and Sage, E.H. (2001) SPARC, a matricellular protein that functions in cellular differentiation and tissue response to injury. J. Clin. Invest., 107, 1049–1054.[Web of Science][Medline]
Bradshaw, A.D., Puolakkainen, P., Dasgupta, J., Davidson, J.M., Wight, T.N., and Sage, E.H. (2003) SPARC-null mice display abnormalities in the dermis characterized by decreased collagen fibril diameter and reduced tensile strength. J. Invest. Dermatol., 120, 949–955.[CrossRef][Web of Science][Medline]
Damjanovski, S., Liu, F., and Ringuette, M. (1992) Molecular analysis of Xenopus laevis SPARC (secreted protein, acidic, rich in cysteine). A highly conserved acidic calcium-binding extracellular-matrix protein. Biochem. J., 281, 513–517.[Medline]
Fisher, L.W., Hawkins, G.R., Tuross, N., and Termine, J.D. (1987) Purification and partial characterization of small proteoglycans I and II, bone sialoproteins I and II, and osteonectin from the mineral compartment of developing human bone. J. Biol. Chem., 262, 9702–9708.
Hakomori, S. (1996). Tumor malignancy defined by aberrant glycosylation and sphingo(glyco)lipid metabolism. Cancer Res., 56, 5309–5318.
Hartmann, U. and Maurer, P. (2001) Proteoglycans in the nervous system—the quest for functional roles in vivo. Matrix Biol., 20, 23–35.[CrossRef][Web of Science][Medline]
Hohenadl, C., Mann, K., Mayer, U., Timpl, R., Paulsson, M., and Aeschlimann, D. (1995) Two adjacent N-terminal glutamines of BM-40 (osteonectin, SPARC) act as amine acceptor sites in transglutaminase-catalyzed modification. J. Biol. Chem., 270, 23415–23420.
Hohenester, E., Maurer, P., Hohenadl, C., Timpl, R., Jansonius, J.N., and Engel, J. (1996) Structure of the extracellular Ca2+-binding module in BM-40. Nat. Struct. Biol., 3, 67–73.[CrossRef][Web of Science][Medline]
Hohenester, E., Maurer, P., and Timpl, R. (1997) Crystal structure of a pair of follistatin-like and EF-hand calcium-binding domains in BM-40. EMBO J., 16, 3778–3786.[CrossRef][Web of Science][Medline]
Iruela-Arispe, M.L., Vernon, R.B., Wu, H., Jaenisch, R., and Sage, E.H. (1996) Type I collagen-deficient Mov-13 mice do not retain SPARC in the extracellular matrix: implications for fibroblast function. Dev. Dyn., 207, 171–183.[CrossRef][Web of Science][Medline]
Johnston, I.G., Paladino, T., Gurd, J.W., and Brown, I.R. (1990) Molecular cloning of SC1: a putative brain extracellular matrix glycoprotein showing partial similarity to osteonectin/BM40/SPARC. Neuron, 2, 165–176.
Kelm, R.J. and Mann, K.G. (1991) The collagen binding specificity of bone and platelet osteonectin is related to differences in glycosylation. J. Biol. Chem., 266, 9632–9639.
Kohfeldt, E., Maurer, P., Vannahme, C., and Timpl, R. (1997) Properties of the extracellular calcium-binding module of the proteoglycan testican. FEBS Lett., 414, 557–561.[CrossRef][Web of Science][Medline]
Laemmli, U.K. (1970) Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature, 227, 680–685.[CrossRef][Medline]
Lankat-Buttgereit, B., Mann, K., Deutzmann, R., Timpl, R., and Krieg, T. (1988) Cloning and complete amino acid sequences of human and murine basement membrane protein BM-40 (SPARC, osteonectin). FEBS Lett., 236, 352–356.[CrossRef][Web of Science][Medline]
Maillard, C., Malaval, L., and Delmas, P.D. (1992) Immunological screening of SPARC/osteonectin in nonmineralized tissues. Bone, 13, 257–264.[Medline]
Mann, K., Deutzmann, R., Paulsson, M., and Timpl, R. (1987) Solubilization of protein BM-40 from a basement membrane tumor with chelating agents and evidence for its identity with osteonectin and SPARC. FEBS Lett., 218, 167–172.[CrossRef][Web of Science][Medline]
Mason, I.J., Taylor, A., Williams, J.G., Sage, H., and Hogan, B.L.M. (1986) Evidence from molecular cloning that SPARC, a major product of mouse embryo parietal endoderm, is related to an endothelial cell "culture shock" glycoprotein of Mr = 43,000. EMBO J., 5, 1465–1472.[Web of Science][Medline]
Maurer, P., Hohenadl, C., Hohenester, E., Göhring, W., Timpl, R., and Engel, J. (1995) The C-terminal portion of BM-40 (SPARC/osteonectin) is an autonomously folding and crystallisable domain that binds calcium and collagen IV. J. Mol. Biol., 253, 347–357.[CrossRef][Web of Science][Medline]
Maurer, P., Göhring, W., Sasaki, T., Mann, K., Timpl, R., and Nischt, R. (1997) Recombinant and tissue-derived mouse BM-40 bind to several collagen types and have increased affinities after proteolytic activation. Cell. Mol. Life Sci., 53, 478–484.[CrossRef][Web of Science][Medline]
Maurer, P., Hohenester, E., and Engel, J. (1996) Extracellular calcium-binding proteins. Curr. Opin. Cell Biol., 8, 609–617.[CrossRef][Web of Science][Medline]
Nischt, R., Pottgiesser, J., Krieg, T., Mayer, U., Aumailley, M., and Timpl, R. (1991) Recombinant expression and properties of the human calcium-binding extracellular matrix protein BM-40. Eur. J. Biochem., 200, 529–536.[Web of Science][Medline]
Packer, N.H., Lawson, M.A., Jardine, D.R., and Redmond, J.W. (1998) A general approach to desalting oligosaccharides released from glycoproteins. Glycoconj. J., 15, 737–747.[CrossRef][Web of Science][Medline]
Romberg, R.W., Werness, P.G., Lollar, P., Riggs, B.L., and Mann, K.G. (1985) Isolation and characterization of native adult osteonectin. J. Biol. Chem., 260, 2728–2736.
Sasaki, T., Göhring, W., Mann, K., Maurer, P., Hohenester, E., Knäuper, V., Murphy, G., and Timpl, R. (1997) Limited cleavage of extracellular matrix protein BM-40 by matrix metalloproteinases increases its affinity for collagens. J. Biol. Chem., 272, 9237–9243.
Sasaki, T., Hohenester, E., Göhring, W., and Timpl, R. (1998) Crystal structure and mapping by site-directed mutagenesis of the collagen-binding epitope of an activated form of BM-40/SPARC/osteonectin. EMBO J., 17, 1625–1634.[CrossRef][Web of Science][Medline]
Schwarzbauer, J.E. and Spencer, C.S. (1993) The Caenorhabditis elegans homologue of the extracellular calcium binding protein SPARC/osteonectin affects nematode body morphology and mobility. Mol. Biol. Cell, 4, 941–952.[Abstract]
Shibanuma, M., Mashimo, J., Mita, A., Kuroki, T., and Nose, K. (1993) Cloning from a mouse osteoblastic cell line of a set of transforming-growth-factor-beta 1-regulated genes, one of which seems to encode a follistatin-related polypeptide. Eur. J. Biochem., 217, 13–19.[Web of Science][Medline]
Skelton, T. P., Zeng, C., Nocks, A., and Stamenkovic, I. (1998) Glycosylation provides both stimulatory and inhibitory effects on cell surface and soluble CD44 binding to hyaluronan. J. Cell Biol., 140, 431–446.
Tanaka, S., Nambu, F., and Nambu, Z. (2001) Isolation of a cDNA encoding a putative SPARC from the brine shrimp, Artemia franciscana. Gene, 268, 53–58.[CrossRef][Web of Science][Medline]
Tang, S. and McKeown, B.A. (1995) Sequence deposited in the NCBI data base. Accession number AAC99813.
Termine, J.D., Kleinman, H.K., Whitson, S.W., Conn, K.M., McGarvey, M.L., and Martin, G.R. (1981) Osteonectin, a bone specific protein linking mineral to collagen. Cell, 26, 99–105.[CrossRef][Web of Science][Medline]
Vannahme, C., Gösling, S., Paulsson, M., Maurer, P., and Hartmann, U. (2003) Characterization of SMOC-2, a modular extracellular calcium-binding protein. Biochem. J., 373, 805–814.[CrossRef][Web of Science][Medline]
Vannahme, C., Schübel, S., Herud, M., Gösling, S., Hülsmann, H., Paulsson, M., Hartmann, U., and Maurer, P. (1999) Molecular cloning of testican-2: defining a novel calcium-binding proteoglycan family expressed in brain. J. Neurochem., 73, 12–20.[CrossRef][Web of Science][Medline]
Vannahme, C., Smyth, N., Miosge, N., Gösling, S., Frie, C., Paulsson, M., Maurer, P., and Hartmann, U. (2002) Characterization of SMOC-1, a novel modular calcium-binding protein in basement membranes. J. Biol. Chem., 277, 37977–37986.
Varki, A. (1993) Biological roles of oligosaccharides: all of the theories are correct. Glycobiology, 3, 97–130.
Villareal, X.S., Mann, K.G., and Long, G.L. (1989) Structure of human osteonectin based upon analysis of cDNA and genomic sequences. Biochemistry, 28, 6483–6491.[CrossRef][Medline]
Wuttke, M., Müller, S., Nitsche, P., Paulsson, M., Hanisch, F.-G., and Maurer, P. (2001) Structural characterization of human recombinant and bone-derived bone sialoprotein (BSP). Functional implications for cell attachment and hydroxyapatite binding. J. Biol. Chem., 276, 36839–36848.
Xie, R.-L. and Long, G.L. (1995) Role of N-linked glycosylation in human osteonectin. Effect of carbohydrate removal by N-glycanase and site-directed mutagenesis on structure and binding of type V collagen. J. Biol. Chem., 270, 23212–23217.
Yamamoto, H., Swoger, J., Greene, S., Saito, T., Hurh, J., Sweeley, C., Leestma, J., Mkrdichian, E., Cerullo, L., Nishikawa, A., and others. (2000) ß1,6-N-acetylglucosamine-bearing N-glycans in human gliomas: implications for a role in regulating invasivity. Cancer Res., 60, 134–142.
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  T. Reding, U. Wagner, A. B. Silva, L-K. Sun, M. Bain, S.-Y. Kim, D. Bimmler, and R. Graf Inflammation-dependent expression of SPARC during development of chronic pancreatitis in WBN/Kob rats and a microarray gene expression analysis Physiol Genomics, July 9, 2009; 38(2): 196 - 204. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. W.M. Schellings, D. Vanhoutte, M. Swinnen, J. P. Cleutjens, J. Debets, R. E.W. van Leeuwen, J. d'Hooge, F. Van de Werf, P. Carmeliet, Y. M. Pinto, et al. Absence of SPARC results in increased cardiac rupture and dysfunction after acute myocardial infarction J. Exp. Med., January 16, 2009; 206(1): 113 - 123. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 E. Hohenester, T. Sasaki, C. Giudici, R. W. Farndale, and H. P. Bachinger Structural basis of sequence-specific collagen recognition by SPARC PNAS, November 25, 2008; 105(47): 18273 - 18277. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. Giudici, N. Raynal, H. Wiedemann, W. A. Cabral, J. C. Marini, R. Timpl, H. P. Bachinger, R. W. Farndale, T. Sasaki, and R. Tenni Mapping of SPARC/BM-40/Osteonectin-binding Sites on Fibrillar Collagens J. Biol. Chem., July 11, 2008; 283(28): 19551 - 19560. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. Stork, K. A. Zettlitz, D. Muller, M. Rether, F.-G. Hanisch, and R. E. Kontermann N-Glycosylation as Novel Strategy to Improve Pharmacokinetic Properties of Bispecific Single-chain Diabodies J. Biol. Chem., March 21, 2008; 283(12): 7804 - 7812. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 N. Martinek, J. Shahab, J. Sodek, and M. Ringuette Is SPARC an Evolutionarily Conserved Collagen Chaperone? Journal of Dental Research, April 1, 2007; 86(4): 296 - 305. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Parry, F. G. Hanisch, S.-H. Leir, M. Sutton-Smith, H. R. Morris, A. Dell, and A. Harris N-Glycosylation of the MUC1 mucin in epithelial cells and secretions Glycobiology, July 1, 2006; 16(7): 623 - 634. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
Q. Shi, S. Bao, J. A. Maxwell, E. D. Reese, H. S. Friedman, D. D. Bigner, X.-F. Wang, and J. N. Rich Secreted Protein Acidic, Rich in Cysteine (SPARC), Mediates Cellular Survival of Gliomas through AKT Activation J. Biol. Chem., December 10, 2004; 279(50): 52200 - 52209. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
G. Ge, N.-S. Seo, X. Liang, D. R. Hopkins, M. Hook, and D. S. Greenspan Bone Morphogenetic Protein-1/Tolloid-related Metalloproteinases Process Osteoglycin and Enhance Its Ability to Regulate Collagen Fibrillogenesis J. Biol. Chem., October 1, 2004; 279(40): 41626 - 41633. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||