- Email: [email protected]
Protein chemical characterization of Gc globulin (vitamin D-binding protein) isoforms; Gc-1f, Gc-1s and Gc-2
Biochimica et Biophysica Acta 1774 (2007) 481 – 492 www.elsevier.com/locate/bbapap
Protein chemical characterization of Gc globulin (vitamin D-binding protein) isoforms; Gc-1f, Gc-1s and Gc-2 Maja Christiansen a , Charlotte S. Jørgensen b , Inga Laursen c , Daniel Hirschberg d , Peter Højrup d , Gunnar Houen a,⁎ a b
Department of Autoimmunology, Statens Serum Institut, Artillerivej 5, DK-2300 Copenhagen, Denmark Department of Bacteriology, Mycology and Parasitology, Statens Serum Institut, Copenhagen, Denmark c Department of Clinical Biochemistry, Statens Serum Institut, Copenhagen, Denmark d Institute of Biochemistry and Molecular Biology, University of Southern Denmark, Odense, Denmark Received 10 November 2006; revised 26 January 2007; accepted 31 January 2007 Available online 12 February 2007
Abstract Gc globulin, also called vitamin D-binding protein, is a plasma protein involved in the extracellular actin-scavenger system, vitamin D transport and possibly also other biological activities. Low levels of Gc globulin have been found to correlate with multiple organ failure and nonsurvival of patients with fulminant hepatic failure and trauma. Here, we characterize the dominant isoforms of plasma-derived Gc globulin from Cohn fraction IV paste with respect to amino acid sequence and posttranslational modifications. Gc globulin was purified in large scale and the isoforms separated by ion exchange chromatography. The separated isoforms and several commercial preparations of individual isoforms were characterized by mass spectrometry. This revealed that the major isoforms were non-glycosylated. Compared to the Gc-1f isoform the other dominating isoforms represented an Asp/Glu substitution (Gc-1s) and a Thr/Lys substitution (Gc-2) in agreement with DNA sequencing studies. The commercial preparations were found to represent mainly one or two isoforms. An O-linked glycan with a mass of 656 Da and terminating with a sialic acid residue was detected on a minor proportion of Gc globulin molecules. © 2007 Elsevier B.V. All rights reserved. Keywords: Gc globulin; Vitamin D-binding protein; Isoforms; Glycosylation; Mass spectrometry
1. Introduction Vitamin D-binding protein (DBP), also known as Gc (Group-specific component) globulin, is a plasma protein
Abbreviations: BCIP/NBT, 5-bromo-4-chloro-3-indolyl phosphate/nitro blue tetrazolium; CIE, crossed immunoelectrophoresis; DBP, vitamin D-binding protein; DTT, dithiothreitol; ESI, electro spray ionization; FPLC, fast protein liquid chromatography; Gc, group-specific component; HCCA, alpha-cyano-4hydroxycinnamic acid; HILIC, hydrophilic interaction chromatography; HPLC, high performance liquid chromatography; IEF, isoelectric focusing; LC, liquid chromatography; MAF, macrophage-activating factor; MALDI, matrix assisted laser desorption ionization; MS, mass spectrometry; RP-HPLC, reversed phase HPLC; TFA, trifluoroacetic acid; TNBP, tri-N-butylphosphate; TOF, time of flight; TTN, 50 mM Tris, 0.3 M NaCl, 1% Tween 20, pH 7.5 ⁎ Corresponding author. Tel.: +45 3268 3276; fax: +45 3268 3149. E-mail address: [email protected] (G. Houen). 1570-9639/$ - see front matter © 2007 Elsevier B.V. All rights reserved. doi:10.1016/j.bbapap.2007.01.005
serving two major physiological functions: (1) binding and transport of vitamin D and its metabolites and (2) binding, sequestration and removal of monomeric actin [1–5]. Gc globulin is a member of the albumin family, which also comprises serum albumin, alpha-fetoprotein and afamin [6–12]. The genes for the proteins of the family are located on chromosome 4 in a multigene cluster [13]. Gc globulin is a single chain polypeptide of 458 amino acid residues and, like albumin, it consists of three domains [6–10]. Binding of vitamin D occurs with high affinity at a site located in domain I [14–16]. The binding site for actin is independent of the vitamin D binding site and has been located to a region covering residues 360–372 [15]. However, crystallographic studies have shown that the binding of actin occurs in a large cavity composed of all three domains [17,18]. Depolymerization of filamentous F actin by plasma gelsolin and sequestration of monomeric G actin by Gc globulin is crucial to prevent
482
M. Christiansen et al. / Biochimica et Biophysica Acta 1774 (2007) 481–492
circulatory disturbances in cases of severe cell necrosis or apoptosis, and exhaustion of the extracellular actin scavenger system may lead to deposition of actin, obstruction of blood flow and, eventually, multiple organ failure [4,19–39]. In addition to the two major functions described above, Gc globulin has also been reported to have a co-chemotactic activity for C5a [40,41] and to be a precursor for a potent macrophage-activating factor (MAF) [42–47], which also has anti-tumor and anti-angiogenic properties [46–52]. The conversion of Gc globulin to Gc-MAF is supposed to occur by the action of glycosidases on an O-linked trisaccharide resulting in successive removal of neuraminic acid (by neuraminidase) and galactose (by β-galactosidase) leaving an O-linked N-acetyl-galactosamine (GalNAc) [42–49]. However, except for one study [53] the location and the structure of the Olinked glycan has not been thoroughly characterized and its existence is mainly based on indirect evidence. The concentration of Gc globulin in plasma is in the range of 200–500 mg/L [22,54–60] and it occurs in three major isoforms, which can be separated by isoelectric focusing (IEF) and are denoted Gc-1f, Gc-1s, Gc-2 [59,61–68]. The Gc-1f and Gc-1s isoforms have been reported to be sialylated [69], whereas Gc-2 was reported to be non-glycosylated [53]. However, DNA sequencing studies have associated the isoforms with amino acid substitutions at positions 416 and 420. Gc-1f has aspartic acid at position 416 and threonine at position 420, Gc-1s has glutamic acid and then threonine, and Gc-2 has aspartic acid and lysine at these positions [70,71]. In order to characterize the nature of the different isoforms, we have purified Gc globulin from plasma fraction IV [72] and characterized the three major isoforms by protein chemical methods, including mass spectrometry (MS) in combination with protease digestion. 2. Materials and methods 2.1. Large scale purification Human plasma Gc globulin was purified using fraction IV paste as starting material [72]. Paste IV was precipitated during ethanol fractionation (40% ethanol at pH 5.9) of human plasma as described by Kistler and Nitschmann [73]. The fraction IV paste was dissolved in 20 mM Tris, pH 8.0, by overnight stirring at 4 °C, followed by serial depth and delipid filtration (50 LA, 90 LA, and delipid filters). The filtrate was added NaCl to 0.04 M, and applied onto a QSepharose F.F. column. The column was washed using 20 mM Tris pH 8.0, 0.05 M NaCl, before elution with 20 mM Tris, 0.11 M NaCl, pH 8.0. The eluted material was ultra-filtered to half of the original volume, and dia-filtered against 20 mM phosphate buffer, pH 5.7. After ultra-dia-filtration the material was loaded onto a CM-Sepharose F.F. column, and the run-through fraction was collected, and treated by solvent-detergent (1% Tween 80, 0.3% tri-Nbutylphosphate (TNBP)) overnight at room temperature, followed by a dilution with 20 mM phosphate buffer, 48 mM NaCl, pH 9.4 to a final pH of 7.8. The solution was then applied onto a Q-Sepharose F.F. column. Following washing with 20 mM phosphate buffer, pH 7.8, 0.04 M NaCl, the column was eluted with 20 mM phosphate buffer, pH 7.8, 0.1 M NaCl. The peak fraction was collected and ultra-filtered, followed by gel filtration on a Sephacryl S200 column equilibrated with 8 mM Na2HPO4, 1.4 mM NaH2PO4, 130 mM NaCl, pH 7.0,. The main peak was collected and subjected to nano-filtration through a 20 nm and a 15 nm filter connected in series, and finally stabilized by addition of 5% maltose before sterile filtration and filling.
2.2. Small scale ion exchange chromatography for separation of isoforms 10 mg of human Gc globulin purified from plasma fraction IV was dialyzed against 20 mM sodium acetate pH 4.5, followed by further separation by ion exchange chromatography on a MonoS column connected to a fast protein liquid chromatography (FPLC) system, using a flow rate of 1 mL/min and collecting fractions of 2.5 mL. The column was eluted with 20 mM sodium acetate, pH 4.5, 0.5 M NaCl according to the following setup: 30 min of 0% elution buffer, followed by a 10 min linear increase from 0 to 50%, and then a linear increase from 50 to 80%. Immediately after the chromatography, the collected fractions were dialyzed against 50 mM Tris, pH 7.5 before further analysis.
2.3. SDS-PAGE SDS-PAGE was performed as described [74,75] using precast gels. The PAGE was performed using precast 4–20% Tris-glycine gels following the instructions of the manufacturer, using a variation of the recommended sample buffers. Samples were mixed with an equal volume of sample buffer (nonreduced SDS-PAGE: 50 mM Tris–HCl, pH 6.8, 10% glycerol, 2% SDS, 0.06% pyronin G; reduced SDS-PAGE: as non-reduced but added 0.1 M dithiothreitol (DTT)) before loading into the wells of the gel. After electrophoresis, gels were either stained with Coomassie Brilliant Blue (GelCode) according to the manufacturer's recommendation, or electroblotted to nitrocellulose membranes for subsequent immunoblotting.
2.4. Immunoblotting SDS-PAGE gels were sandwiched with nitrocellulose membranes between 12 sheets of Whatman paper no. 1 soaked in 10 times diluted electrophoresis buffer, and subjected to semi-dry electroblotting overnight at 0.1 mA/cm2. The membranes were blocked for 1 h in 50 mM Tris, 0.3 M NaCl, 1% Tween 20, pH 7.5 (TTN), incubated with monoclonal antibody against Gc globulin diluted 1:100 or rabbit antibodies against human serum proteins diluted 1:1000 in TTN for 1 h, followed by three washes using TTN. Next the membranes were incubated for 1 h with alkaline phosphatase-conjugated goat immunoglobulins against mouse or rabbit immunoglobulins diluted 1:1000 in TTN, followed by three washes in TTN. Finally, bound antibodies were visualized by incubation with 5-bromo-4-chloro-3indolyl phosphate/nitro blue tetrazolium (BCIP/NBT) substrate.
2.5. Isoelectric focusing Isoelectric focusing was performed using precast pH 3–7 gels following the directions of the manufacturer. Samples were mixed with an equal volume of the recommended sample buffer before loading onto the gel. After electrophoresis and fixing, gels were stained with Coomassie Brilliant Blue (GelCode) according to the manufacturer's directions.
2.6. Glycosidase treatment of Gc globulin Samples of approximately 100 μg Gc globulin were vacuum dried and redissolved in 100 μL 50 mM NH4HCO3, pH 6. The samples were treated individually, one with neuraminidase, alpha-mannosidase and beta-galactosidase, another with neuraminidase and O-glycosidase and a control sample was left untreated. 1 mU of each enzyme was utilized and samples were incubated over night at room temperature. In a separate experiment for ESI MS, selected fractions from the cation exchange chromatography and commercial preparations (approximately 100 pmol Gc globulin in 10 μL water) were treated with neuraminidase (3 μL reaction buffer and 0.3 μL sialidase A (GLYCO Prozyme, San Leandro, CA) for 1 h at 37 °C.
2.7. Matrix assisted laser desorption ionization (MALDI)-MS of intact protein A sample of intact Gc globulin was micro-purified on custom made POROS R150 columns and loaded onto a PerSeptive steel-target in a matrix sandwich of
M. Christiansen et al. / Biochimica et Biophysica Acta 1774 (2007) 481–492 sinapic acid (0.5 μL 20 μg/μL sinapic acid in acetone was deposited on the target; the sample was eluted in four droplets on top of this deposit with 0.8 μL of 20 μg/μL sinapic acid in 70% acetonitrile, 0.1% trifluoroacetic acid (TFA)). The samples were subjected to MS analysis on a PerSeptive Voyager Elite MALDI-MS instrument (PerSeptive Biosystems, Framingham, MA) with delayed extraction, operated in the positive linear mode.
2.8. Electrospray ionization (ESI)-MS Gc globulin ion exchange fractions and commercial preparations were analyzed by ESI-MS. The liquid chromatography (LC) system for protein desalting, has been described previously [76]. Solvents were delivered by two Applied Biosystems high performance liquid chromatography (HPLC) syringe pumps (model 140 B), one for desalting and one for elution of proteins. Samples (100 pmol) were dissolved in 100 μL 1% TFA and loaded onto a 200 μL stainless steel loop through a 6-port Rheodyne injection valve (7725i). When the valve was switched to inject position the sample was flushed from the loop to a 10-port two position Valco valve (C2-1000A) equipped with a C18 micro column (bed volume 2 μL). The sample was trapped on the micro column and desalted with a 400 μL/min flow of 0.05% (v/v) TFA for 1 min. The Valco valve switched automatically after the preset desalting time, and the sample was eluted for coupled MS analysis. The protein was eluted from the trap column with a 30 μL/min flow of 80% (v/v) acetonitrile and 0.05% (v/v) TFA. The LC system was coupled to an electrospray ionization quadrupole time-of-flight (TOF) (QToF Ultima, Micromass) mass spectrometer. Capillary voltage was set to 3.5 kV, cone voltage to 55 V and ion source block temperature to 80 °C with a desolvation gas flow of 500 L/h at 200 °C and nebulizing gas flow of 20 L/h at room temperature. The protein charge state envelope was deconvoluted by the MaxEnt 1 algorithm provided with the Masslynx software.
2.9. Bottom up analysis Approximately 200 μg of the large scale Gc globulin preparation was reduced with 4 mM DTT for 30 min at 56 °C, carbamidomethylated with 6 mM
483
iodacetamide for 30 min in the dark, at room temperature, and then digested with 4 μg trypsin overnight, at 37 °C. The sample was separated on an HPLC system (Amersham Pharmacia Biotech) using a Phenomenex Jupiter C18 column 250 × 4.60 mm. The eluted fractions were subjected to MS analysis on a PerSeptive Voyager Elite MALDI-MS instrument (PerSeptive Biosystems, Framingham, MA) with delayed extraction, operated in the positive reflector mode. Samples for MS analysis were micro-purified on custom made POROS R350 or R250 columns, and loaded onto a PerSeptive steel-target using alphacyano-4-hydroxycinnamic acid (HCCA) as matrix, according to a previously described procedure [77] with slight modifications: Samples were loaded onto the micro-columns in 0.1% TFA, washed twice, and eluted with 0.8 μL matrix solution (10 μg/μL in 70% acetonitrile, 0.1% TFA) in four spots. MS/MS spectra were obtained for selected peptides on an Applied Biosystems 4700 proteomics analyzer TOF-TOF MALDI instrument (Applied Biosystems, Foster City, CA), with POROS R250 micro-purification and 5 μg/μL HCCA matrix.
2.10. Mass spectrometric peptide mass mapping Identification of proteins from gels: Protein spots were cut from the gel and subjected to tryptic digestion according to a previously described procedure [78]. After excision of the gel pieces, they were washed 2 × 15 min in 50% acetonitrile and dehydrated in 100% acetonitrile. A volume of 50 μL 10 mM DTT in 100 mM NH4HCO3 (sufficient to cover the gel pieces) was added and the proteins were reduced for 45 min, at 56 °C. After cooling, the DTT solution was replaced with an equivalent volume of 55 mM iodacetamide in 100 mM NH4HCO3 and incubated for 30 min at room temperature in the dark. The washing and dehydrating steps were repeated, and the gel pieces were reswollen in digestion buffer containing 50 mM NH4HCO3 and 6.25 ng/μL trypsin for 45 min on ice. The excess digestion buffer was removed and replaced by 20 μL 50 mM NH4HCO3 without trypsin and left to digest overnight at 37 °C. Samples were purified on POROS R250 micro-columns and loaded onto the target using 5 μg/μL HCCA matrix according to [77], applying the same modifications as above, eluting in two droplets. MS and MS/MS spectra were acquired on an Applied Biosystems 4700 proteomics analyzer TOF-TOF MALDI instrument
Fig. 1. MALDI-MS of intact Gc globulin from large scale purification obtained in linear mode. An average molecular mass of 51.14 kDa is obtained by the average of the singly charged (51.143) and doubly charged (25.597) ions), which is within the estimated accuracy of 100 Da from the theoretical mass of 51.2 kDa (Table 1).
484
M. Christiansen et al. / Biochimica et Biophysica Acta 1774 (2007) 481–492
Table 1 Gc globulin isoform variation Isoform
Position 416
Position 420
Theoretical mass (Da)
pI (84)
Gc-2 Gc-1s Gc-1f
D E D
K T T
51215 51202 51188
5.21 5.16 5.15
Average mass and theoretical pI of the Gc globulin isoforms. The pI's were calculated using the values of Rickard et al. [84]. (Applied Biosystems, Foster City, CA). Proteinase K digestion followed by graphite micro-purification was performed as described [79,80], and hydrophilic interaction chromatography (HILIC) micro-purification was performed as in [81–83].
2.11. Bioinformatics Molecular mass and pI values were calculated using the GPMAW software package (Lighthouse data, Odense, Denmark).
3. Results Gc globulin was purified from plasma fraction IV by a large scale process comprising three ion-exchange chromatography steps (anion → cation → anion exchange) followed by a gelfiltration [72]. This yielded an 85% pure preparation containing minor amounts of alpha-1-antitrypsin, apolipoprotein A1, and haptoglobin as revealed by non-reducing SDS-PAGE and MS. In reducing SDS-PAGE alpha-1-antitrypsin has nearly the same mobility as Gc globulin, whereas the two are clearly separated by non-reducing SDS-PAGE. The actin-binding ability was verified by crossed immunoelectrophoresis (CIE) after the addition of actin in excess relative to Gc globulin (results not shown).
Isoelectric focusing revealed that the purified Gc globulin contained the three major isoforms denoted Gc-1f, Gc-1s, Gc-2 in almost equal amounts. The identity of the individual bands as Gc globulin, was confirmed by mass spectrometric peptide mapping (results not shown). Analysis of the large scale purified Gc globulin by MALDIMS showed an average molecular mass of 51.14 kDa (Fig. 1). In comparison, the theoretical average mass of the Gc globulin calculated by GPMAW is between 51.18 and 51.21 kDa, depending on the isoform (Table 1). An attached glycan of the structure GalNAc(Gal)Neu would contribute with 656 Da and should thus be easily detectable. The Gc globulin preparation was treated with a range of different glycosidases (neuraminidase, O-glycosidase and neuraminidase, beta-galactosidase, alpha-mannosidase) and subjected to analysis, all with similar mass results as the non-treated samples (results not shown) indicating that the major proportion of Gc globulin molecules are not glycosylated. Several techniques were applied to detect a partial glycosylation directly in the large scale Gc globulin preparation. Proteinase K digestion followed by graphite micro-purification, and hydrophilic interaction chromatography (HILIC) micropurification of a tryptic Gc globulin digest was performed, but no glycosylated peptides were found. Two samples of trypsindigested Gc globulin with and without prior glycosidase treatment were analyzed by RP-HPLC and subsequently compared. A few differences appeared, which were investigated by MS, but no glycosylation (neither N- nor O-linked) was identified. PNGase F treatment of the fraction IV extract used for purification of Gc globulin and of purified Gc globulin, with following analysis by one- or two-dimensional PAGE showed no changes for Gc globulin, whereas alpha-1-antitrypsin did
Fig. 2. Chromatogram of trypsin-digested Gc globulin. The separation was performed on a Phenomenex Jupiter C18 column 250 × 4.60 mm. Fractions were collected manually to obtain the best separation possible. All fractions were subjected to MALDI-MS analysis (not shown). Mass spectra of the marked peaks (star and triangle) are shown in Fig. 3A and B, respectively.
M. Christiansen et al. / Biochimica et Biophysica Acta 1774 (2007) 481–492
change its mobility (verifying that the enzyme treatment was effective). Furthermore, magnetic Dynabeads with immobilized lectins from Lens culinaris and Erythrina cristagalli were used, aiming at catching glycosylated peptides from different enzyme digests of Gc globulin, but no glycosylated peptides were isolated. An oxidation of possible sugar residues on Gc globulin with following biotin-hydrazide binding and isolation with streptavidin-coupled magnetic Dynabeads was also attempted, and Gc globulin was cleaved with CNBr and separated by RP-HPLC to isolate larger fragments for further digestion with AspN. However, subsequent analyses did not reveal glycosylation. To further investigate human plasma Gc globulin for the presence of glycosylation, a sample of the protein purified from fraction IV was reduced, carbamidomethylated, digested with trypsin, and subjected to reversed phase (RP)-HPLC to separate the obtained peptides (Fig. 2). All the different fractions eluted were analyzed by MALDI-MS and the peptides containing amino acids 414–420, previously reported to be O-glycosylated, were found to be without signs of glycosylation (Fig. 3). This region is also where the amino acid substitution between the different isoforms is present, hence generating three
485
different peptides, namely a 741.41 Da, a 1155.62 Da and a 1169.64 Da peptide respectively. The tree peptides were sequenced by MALDI-MS/MS and verified as being the varying peptides in the three isoforms (Fig. 4), thus confirming previous results found on the DNA level [70,71]. The sequences of the peptides are as follows: LPDATPTELAK for Gc-1f, LPEATPTELAK for Gc-1s and LPDATPK for Gc-2. The peptides containing the two potential N-glycosylation sites were also identified in the eluate from the RP-HPLC and analyzed by MALDI-MS, but no glycosylation was observed. This, altogether, indicates that the differences between the major Gc globulin isoforms are due to amino acid substitutions. In order to characterize the individual isoforms, we attempted separation of these by cation exchange chromatography (Fig. 5). Analysis of fractions by SDS-PAGE and IEF revealed essentially complete purity of Gc globulin and a partial separation of the isoforms (Fig. 6). Isoelectric focusing of three commercial Gc globulin preparations showed that two of them contained mainly the Gc-1s isoform (ICN and Biodesign) while the third (Biomedical) contained Gc-1s and Gc-2. Immunoblotting of the non-reduced SDS-PAGE gel using both monoclonal and polyclonal antibodies against Gc globulin and alpha-1-
Fig. 3. Characterization of Gc globulin isoforms by MALDI-MS. MS analysis of the two marked fractions from the RP-HPLC analysis in Fig. 2. The peak with the m/z 741.41 (A), accounts for the Gc globulin peptide 414–420 from Gc-2 and the peaks at m/z 1155.62 and 1169.64 (B) corresponds to the related peptides from Gc-1f and Gc-1s. All assignments were verified by MS/MS analysis (Fig. 4).
486
M. Christiansen et al. / Biochimica et Biophysica Acta 1774 (2007) 481–492
antitrypsin, verified that all the tested samples contained Gc globulin, and that alpha-1-antitrypsin was present in the Gc globulin purified from plasma fraction IV (results not shown).
ESI-MS was performed on all samples from Fig. 6. The raw data showed heterogeneity in the individual mass peaks. However, upon smoothing and deconvolution, the mass spectra in
Fig. 4. MALDI-MS/MS spectra of the three peptides m/z 741.41 (A), 1155.6 (B) and 1169.6 (C) from Fig. 3. The sequencing confirms the identity of the peptides as the ones containing amino acid residues 414–420 from the different Gc globulin isoforms.
M. Christiansen et al. / Biochimica et Biophysica Acta 1774 (2007) 481–492
487
Fig. 4 (continued).
Fig. 5. Cation exchange chromatography of large scale purified Gc globulin. 10 mg of Gc globulin was separated on a MonoS column by a gradient of 0.5 M NaCl. 2.5 mL fractions were collected as indicated.
488
M. Christiansen et al. / Biochimica et Biophysica Acta 1774 (2007) 481–492
Fig. 7 could be obtained. The observed heterogeneity is in agreement with the presence of multiple Gc globulin isoforms (Table 1) in all ion exchange fractions and commercial preparations.
For the commercial samples (Biomedicals, BioDesign) the peak around 50810 Da corresponds to the loss of approximately 400 Da relative to the main peak. This corresponds to the loss of the three N-terminal amino acid residues, Leu, Glu and Arg (398.5 Da), which was confirmed by Edman degradation. In fractions 27, 30, and 33 from the ion exchange chromatography a main peak appeared in a position similar to the commercial preparations, while the truncated form was not observed. However, multiple peaks at higher mass values were observed. Adducts of approximately 42 Da were observed, which could be a consequence of acetylation (42 Da) caused by the acetate buffer used for the ion exchange. Adducts around 51535, 51865 and 52190 could be caused by multiple additions of non-enzymatic glycation by maltose (324 Da) added to stabilize the final Gc globulin product. However, the mass at 51865 could also be the result of O-glycosylation. The samples from Biodesign and ICN show a very small peak at m/z 51865 which is approximately 655 above the mass of the intact Gc Globulin. This value is very close to the value of 656 Da for the expected glycosylation (see below). In order to ascertain the presence of the O-glycosylation, selected samples showing the 51865 peak (Biomedical, ICN and #27) were digested with Sialidase A and re-analyzed by ESI-MS (Fig. 8). All fractions showed a loss of the 51865 peak and the appearance of a prominent peak around 51576 consistent with the loss of one unit of sialic acid (291 Da). For fraction 27 the peak at 52190 disappeared with the appearance of a peak at 51897, again consistent with the loss of sialic acid. These results are consistent with an Oglycosylation in combination with up to two units of nonenzymatic glycation by maltose. 4. Discussion In this study we have performed a protein chemical characterization of the three major Gc globulin isoforms, Gc1f, Gc-1s and Gc-2. The isoforms were separated by ion exchange chromatography and characterized by mass spectrometry. The mass values of the isoforms, determined by MALDIMS and ESI-MS were in agreement with the sequence-derived theoretical mass values of the various isoforms. This was true for both Gc globulin purified in large scale from plasma fraction IV and for commercial preparations. Previous DNA sequencing studies have revealed amino acid substitutions at position 416 and 420, giving rise to three different isoforms. These amino acid substitutions have been verified here by peptide sequencing and have been shown to be
Fig. 6. SDS-PAGE and IEF analysis of fractions from cation exchange chromatography of Gc globulin. SDS-PAGE is shown with reducing conditions at the top (A) and non-reducing conditions in the middle (B); and an IEF-PAGE can be seen at the bottom (C). All three gels are loaded in identical manner: lane 1: standard marker for molecular weight or pH, lane 2: starting material (Gc globulin purified from fraction IV), lane 3: fraction 27, lane 4: fraction 30, lane 5: fraction 33, lane 6: Gc globulin from ICN, lane 7: Gc globulin from Biodesign, and lane 8: Gc globulin from Biomedicals. Lanes 3–5 are fractions from the separation in Fig. 2.
M. Christiansen et al. / Biochimica et Biophysica Acta 1774 (2007) 481–492
489
Fig. 7. ESI mass spectra of fractions from cation exchange chromatography (Fig. 5) and commercial Gc globulin preparations. Spectra show from the bottom and up: fraction 27, fraction 30, fraction 33, Gc globulin from ICN, Gc globulin from Biodesign and then Gc globulin from Biomedicals. The estimated accuracy is ± 10 Da.
without glycosylations. When comparing the theoretical pI values of Gc globulin with the different substitutions, only small variations appear (Table 1) [84]; however, as the theoretical pI values are based on a linear sequence, the diversity may be more pronounced between the folded proteins. Thus, it is highly
likely that these amino acid substitutions are the cause of the differences in mobility on an IEF-gel of the three major Gc globulin isoforms. In the commercial Gc globulin preparations, N-terminal truncation of three amino acid residues (LeuGluArg) was
Fig. 8. ESI mass spectra after sialidase treatment of fractions from cation exchange chromatography (Fig. 5) and commercial Gc globulin preparations. Spectra show from the bottom and up: fraction 27, Gc globulin from ICN, and then Gc globulin from Biomedicals. The estimated accuracy is ± 10 Da.
490
M. Christiansen et al. / Biochimica et Biophysica Acta 1774 (2007) 481–492
observed to varying degrees. This is most likely a result of aminopeptidase activity or of a trypsin-like endopeptidase activity on an accessible N-terminus. Non-enzymatic glycation by maltose and acetylation by buffer components was seen to some degree in the large scale Gc globulin preparation after stabilization with maltose and ion exchange chromatography in acetate buffer. Analysis of Gc globulin with respect to glycosylation proved to be a very difficult task. Human Gc globulin contains a potential N-glycosylation site at residues 272–274, but no Nglycosylation has been reported for this site and all our experiments capable of detecting N-glycosylations yielded negative results. Another potential, but rare, N-glycosylation site is present at residues 435–437 (Asp–Cys–Cys), but this site was retrieved in good yield in the non-glycosylated form. According to the literature, Gc globulin carries an O-linked trisaccharide composed of N-acetylgalactosamine with branched galactose and sialic acid [4,27,28,53,69]. However, apart from a single study of the glycosylation [53], the remaining reports are based on circumstantial evidence. Apart from its role in scavenging of actin and transport of vitamin D metabolites, Gc globulin has been found to play a role in macrophage activation [42–47], tumorigenesis/angiogenesis [46–52] and bone metabolism [85–87]. The presence of the glycan structure is based on the observation that Gc globulin attains macrophage-activating and anti-angiogenic properties upon combined sialidase and beta-galatosidase treatment. Several attempts were made at finding and identifying this glycosylation directly in the large scale Gc globulin preparation, but none were successful. We have previously reported that the macrophage activation may be caused by endotoxins present in the commercial glycosidases [72], and a possible explanation for the reported glycosylation could be, that most Gc globulin products are slightly contaminated by alpha-1-antitrypsin, which is known to be glycosylated [26], hence the supposed glycosylation might originate from this contamination. Alpha1-antitrypsin was found in both large scale and affinity purifications of Gc globulin, as well as in commercial products (results not shown). The problem is further complicated by the fact that available polyclonal and monoclonal antibodies against Gc globulin are not specific, but appear to recognize alpha-1antitrypsin as well. In contrast to the many attempts at detecting O-linked glycosylation directly on the Gc globulin preparation, a combination of ion exchange chromatography and ESI MS did reveal the presence of a glycan containing a sialic acid residue and having a composition consistent with the putative glycan described by Viau et al. [53]. This species seemed to be present in largest amounts in fraction 27, eluting early from the cation exchange column, in agreement with an extra negative charge conferred by the sialic acid residue. The sialylated species was also detected in two of the commercial preparations. The question therefore arises, whether the glycosylated species represent glycosylated Gc1f, Gc1s or Gc2. On the basis of the present results, we cannot answer this question. In fact, it may be suspected that all Gc globulin isoforms may experience O-linked glycosylation.
In conclusion, the final product of the Gc globulin large scale purification method applied here, was found to contain all three isoforms of the protein, whereas the commercial products tested mainly contained one or two isoforms. The isoforms were partially separated and characterized, and amino acid substitutions verifying previous results from DNA sequencing studies were found upon peptide sequencing at amino acid residues 416 and 420. These isoforms represent the bulk of the Gc globulin and the electrophoretic differences between the major isoforms are caused by the amino acid substitutions. However, a minor proportion of Gc globulin is, in fact, O-glycosylated with a GalNAc(Gal)Neu glycan. The precise location in the amino acid sequence of this glycan and its cellular origin must await further studies. Acknowledgments Kirsten B. Hansen, Dorthe T. Olsen and Inger Christiansen are thanked for skilful technical assistance. The Oticon foundation is gratefully acknowledged for financial support to Maja Christiansen. References [1] P. White, N. Cooke, The multifunctional properties and characteristics of vitamin D-binding protein, Trends Endocrinol. Metab. 11 (2000) 320–327. [2] R. Ray, Molecular recognition in vitamin D-binding protein, Proc. Soc. Exp. Biol. Med. 212 (1996) 305–312. [3] J.G. Haddad, Plasma vitamin D-binding protein (Gc-globulin): multiple tasks, J. Steroid Biochem. Mol. Biol. 53 (1995) 579–582. [4] W.M. Lee, R.M. Galbraith, The extracellular actin-scavenger system and actin toxicity, N. Engl. J. Med. 326 (1992) 1335–1341. [5] N.E. Cooke, J.G. Haddad, Vitamin D binding protein (Gc-globulin), Endocr. Rev. 10 (1989) 294–307. [6] N.E. Cooke, E.V. David, Serum vitamin D-binding protein is a third member of the albumin and alpha fetoprotein gene family, J. Clin. Invest. 76 (1985) 2420–2424. [7] F. Yang, J.L. Brune, S.L. Naylor, R.L. Cupples, K.H. Naberhaus, B.H. Bowman, Human group-specific component (Gc) is a member of the albumin family, Proc. Natl. Acad. Sci. U.S. A. 82 (1985) 7994–7998. [8] F. Yang, V.J. Luna, R.D. McAnelly, K.H. Naberhaus, R.L. Cupples, B.H. Bowman, Evolutionary and structural relationships among the groupspecific component, albumin and alpha-fetoprotein, Nucleic Acids Res. 13 (1985) 8007–8017. [9] F. Schoentgen, M.H. Metz-Boutigue, J. Jolles, J. Constans, P. Jolles, Homology between the human vitamin D-binding protein (group specific component), alpha-fetoprotein and serum albumin, FEBS Lett. 185 (1985) 47–50. [10] F. Schoentgen, M.H. Metz-Boutigue, J. Jolles, J. Constans, P. Jolles, Complete amino acid sequence of human vitamin D-binding protein (group-specific component): evidence of a three-fold internal homology as in serum albumin and alpha-fetoprotein, Biochim. Biophys. Acta 871 (1986) 189–198. [11] H.S. Lichenstein, D.E. Lyons, M.M. Wurfel, D.A. Johnson, M.D. McGinley, J.C. Leidli, D.B. Trollinger, J.P. Mayer, S.D. Wright, M.M. Zukowski, Afamin is a new member of the albumin, alpha-fetoprotein, and vitamin D-binding protein gene family, J. Biol. Chem. 269 (1994) 18149–18154. [12] H. Nishio, A. Dugaiczyk, Complete structure of the human alpha-albumin gene, a new member of the serum albumin multigene family, Proc. Natl. Acad. Sci. U.S. A. 93 (1996) 7557–7561. [13] Y.H. Song, A.K. Naumova, S.A. Liebhaber, N.E. Cooke, Physical and meiotic mapping of the region of human chromosome 4q11–q13
M. Christiansen et al. / Biochimica et Biophysica Acta 1774 (2007) 481–492
[14]
[15]
[16]
[17]
[18]
[19]
[20]
[21]
[22]
[23]
[24] [25]
[26]
[27]
[28]
[29]
[30]
[31]
[32]
[33]
encompassing the vitamin D binding protein DBP/Gc-globulin and albumin multigene cluster, Genome Res. 9 (1999) 581–587. R. Ray, R. Bouillon, H. Van Baelen, M.F. Holick, Photoaffinity labeling of human serum vitamin D binding protein and chemical cleavages of the labeled protein: identification of an 11.5-kDa peptide containing the putative 25-hydroxyvitamin D3 binding site, Biochemistry 30 (1991) 7638–7642. J.G. Haddad, Y.Z. Hu, M.A. Kowalski, C. Laramore, K. Ray, P. Robzyk, N.E. Cooke, Identification of the sterol- and actin-binding domains of plasma vitamin D binding protein (Gc-globulin), Biochemistry 31 (1992) 7174–7181. C. Verboven, A. Rabijns, M. De Maeyer, H. Van Baelen, R. Bouillon, C. De Ranter, A structural basis for the unique binding features of the human vitamin D-binding protein, Nat. Struct. Biol. 9 (2002) 131–136. L.R. Otterbein, C. Cosio, P. Graceffa, R. Dominguez, Crystal structures of the vitamin D-binding protein and its complex with actin: structural basis of the actin-scavenger system, Proc. Natl. Acad. Sci. U. S. A. 99 (2002) 8003–8008. J.F. Head, N. Swamy, R. Ray, Crystal structure of the complex between actin and human vitamin D-binding protein at 2.5 A resolution, Biochemistry 41 (2002) 9015–9020. J.G. Haddad, K.D. Harper, M. Guoth, G.G. Pietra, J.W. Sanger, Angiopathic consequences of saturating the plasma scavenger system for actin, Proc. Natl. Acad. Sci. U.S. A. 87 (1990) 1381–1385. K.D. Harper, J.F. McLeod, M.A. Kowalski, J.G. Haddad, Vitamin D binding protein sequesters monomeric actin in the circulation of the rat, J. Clin. Invest. 79 (1987) 1365–1370. S. Dueland, M.S. Nenseter, C.A. Drevon, Uptake and degradation of filamentous actin and vitamin D-binding protein in the rat, Biochem. J. 274 (Pt. 1) (1991) 237–241. P.J. Goldschmidt-Clermont, W.M. Lee, R.M. Galbraith, Proportion of circulating Gc (vitamin D-binding protein) in complexed form: relation to clinical outcome in fulminant hepatic necrosis, Gastroenterology 94 (1988) 1454–1458. P.A. Janmey, T.P. Stossel, S.E. Lind, Sequential binding of actin monomers to plasma gelsolin and its inhibition by vitamin D-binding protein, Biochem. Biophys. Res. Commun. 136 (1986) 72–79. P.A. Janmey, S.E. Lind, Capacity of human serum to depolymerize actin filaments, Blood 70 (1987) 524–530. J.F. Mc Leod, M.A. Kowalski, J.G. Haddad Jr., Interactions among serum vitamin D binding protein, monomeric actin, profilin, and profilactin, J. Biol. Chem. 264 (1989) 1260–1267. W.X. Tang, H.M. Bazaraa, H. Magiera, N.E. Cooke, J.G. Haddad, Electrophoretic mobility shift assay identifies vitamin D binding protein (Gc-globulin) in human, rat, and mouse sera, Anal. Biochem. 237 (1996) 245–251. M. Coue, J. Constans, A. Olomucki, Effects of serum vitamin-D-binding protein on actin in the presence of plasma gelsolin, Eur. J. Biochem. 160 (1986) 273–277. S.E. Lind, D.B. Smith, P.A. Janmey, T.P. Stossel, Role of plasma gelsolin and the vitamin D-binding protein in clearing actin from the circulation, J. Clin. Invest. 78 (1986) 736–742. W.M. Lee, D.L. Emerson, P.A. Werner, P. Arnaud, P. GoldschmidtClermont, R.M. Galbraith, Decreased serum group-specific component protein levels and complexes with actin in fulminant hepatic necrosis, Hepatology 5 (1985) 271–275. W.M. Lee, D.L. Emerson, W.O. Young, P.J. Goldschmidt-Clermont, D.J. Jollow, R.M. Galbraith, Diminished serum Gc (vitamin D-binding protein) levels and increased Gc:G-actin complexes in a hamster model of fulminant hepatic necrosis, Hepatology 7 (1987) 825–830. W.M. Lee, D. Reines, G.H. Watt, J.A. Cook, W.C. Wise, P.V. Halushka, R.M. Galbraith, Alterations in Gc levels and complexing in septic shock, Circ. Shock 28 (1989) 249–255. W.M. Lee, R.M. Galbraith, G.H. Watt, R.D. Hughes, D.D. McIntire, B.J. Hoffman, R. Williams, Predicting survival in fulminant hepatic failure using serum Gc protein concentrations, Hepatology 21 (1995) 101–105. F.V. Schiodt, S. Bondesen, N. Tygstrup, Serial measurements of serum Gc-
[34]
[35]
[36]
[37]
[38]
[39]
[40]
[41]
[42]
[43]
[44]
[45]
[46]
[47]
[48]
[49]
[50]
[51]
[52]
491
globulin in acetaminophen intoxication, Eur. J. Gastroenterol. Hepatol. 7 (1995) 635–640. F.V. Schiodt, S. Bondesen, I. Petersen, K. Dalhoff, P. Ott, N. Tygstrup, Admission levels of serum Gc-globulin: predictive value in fulminant hepatic failure, Hepatology 23 (1996) 713–718. F.V. Schiodt, P. Ott, S. Bondesen, N. Tygstrup, Reduced serum Gcglobulin concentrations in patients with fulminant hepatic failure: association with multiple organ failure, Crit. Care Med. 25 (1997) 1366–1370. B. Dahl, F.V. Schiodt, T. Kiaer, P. Ott, S. Bondesen, N. Tygstrup, Serum Gc-globulin in the early course of multiple trauma, Crit.Care Med. 26 (1998) 285–289. B. Dahl, F.V. Schiodt, P. Ott, F. Wians, W.M. Lee, J. Balko, G.E. O'Keefe, Plasma concentration of Gc-globulin is associated with organ dysfunction and sepsis after injury, Crit. Care Med. 31 (2003) 152–156. F.H. Wians Jr., W. Lin, L.P. Brown, F.V. Schiodt, W.M. Lee, Immunonephelometric quantification of group-specific component protein in patients with acute liver failure, Liver Transpl. Surg. 3 (1997) 28–33. W.O. Young, P.J. Goldschmidt-Clermont, D.L. Emerson, W.M. Lee, D.J. Jollow, R.M. Galbraith, Correlation between extent of liver damage in fulminant hepatic necrosis and complexing of circulating group-specific component (vitamin D-binding protein), J. Lab. Clin. Med. 110 (1987) 83–90. H.D. Perez, Gc globulin (vitamin D-binding protein) increases binding of low concentrations of C5a des Arg to human polymorphonuclear leukocytes: an explanation for its cochemotaxin activity, Inflammation 18 (1994) 215–220. R.R. Kew, J.A. Fisher, R.O. Webster, Co-chemotactic effect of Gc-globulin (vitamin D binding protein) for C5a. Transient conversion into an active co-chemotaxin by neutrophils, J. Immunol. 155 (1995) 5369–5374. N. Yamamoto, V.R. Naraparaju, Vitamin D3-binding protein as a precursor for macrophage activating factor in the inflammation-primed macrophage activation cascade in rats, Cell Immunol. 170 (1996) 161–167. N. Yamamoto, S. Homma, Vitamin D3 binding protein (group-specific component) is a precursor for the macrophage-activating signal factor from lysophosphatidylcholine-treated lymphocytes, Proc. Natl. Acad. Sci. U.S. A. 88 (1991) 8539–8543. N. Yamamoto, R. Kumashiro, Conversion of vitamin D3 binding protein (group-specific component) to a macrophage activating factor by the stepwise action of beta-galactosidase of B cells and sialidase of T cells, J. Immunol. 151 (1993) 2794–2802. H. Nagasawa, H. Sasaki, Y. Uto, S. Kubo, H. Hori, Association of the macrophage activating factor (MAF) precursor activity with polymorphism in vitamin D-binding protein, Anticancer Res. 24 (2004) 3361–3366. H. Nagasawa, Y. Uto, H. Sasaki, N. Okamura, A. Murakami, S. Kubo, K.L. Kirk, H. Hori, Gc protein (vitamin D-binding protein): Gc genotyping and GcMAF precursor activity, Anticancer Res. 25 (2005) 3689–3695. S.B. Mohamad, H. Nagasawa, Y. Uto, H. Hori, Preparation of Gc proteinderived macrophage activating factor (GcMAF) and its structural characterization and biological activities, Anticancer Res. 22 (2002) 4297–4300. S.B. Mohamad, H. Hori, H. Nagasawa, K. Usui, Y. Uto, Characterization of human Gc protein-derived macrophage activation factor (GcMAF) and its functional role in macrophage tumoricidal activity, Adv. Exp. Med. Biol. 510 (2003) 77–82. S.B. Mohamad, H. Nagasawa, H. Sasaki, Y. Uto, Y. Nakagawa, K. Kawashima, H. Hori, Gc protein-derived macrophage activating factor (GcMAF): isoelectric focusing pattern and tumoricidal activity, Anticancer Res. 23 (2003) 4451–4457. Y. Koga, V.R. Naraparaju, N. Yamamoto, Antitumor effect of vitamin Dbinding protein-derived macrophage activating factor on Ehrlich ascites tumor-bearing mice, Proc. Soc. Exp. Biol. Med. 220 (1999) 20–26. S. Kanda, Y. Mochizuki, Y. Miyata, H. Kanetake, N. Yamamoto, Effects of vitamin D(3)-binding protein-derived macrophage activating factor (GcMAF) on angiogenesis, J. Natl. Cancer Inst. 94 (2002) 1311–1319. O. Kisker, S. Onizuka, C.M. Becker, M. Fannon, E. Flynn, R. D'Amato, B. Zetter, J. Folkman, R. Ray, N. Swamy, S. Pirie-Shepherd, Vitamin D
492
[53]
[54]
[55]
[56]
[57]
[58]
[59]
[60]
[61] [62] [63]
[64] [65]
[66]
[67]
[68]
[69]
[70]
M. Christiansen et al. / Biochimica et Biophysica Acta 1774 (2007) 481–492 binding protein-macrophage activating factor (DBP-maf) inhibits angiogenesis and tumor growth in mice, Neoplasia 5 (2003) 32–40. M. Viau, J. Constans, H. Debray, J. Montreuil, Isolation and characterization of the O-glycan chain of the human vitamin-D binding protein, Biochem. Biophys. Res. Commun. 117 (1983) 324–331. R. Bouillon, H. Van Baelen, P. de Moor, The measurement of the vitamin D-binding protein in human serum, J. Clin. Endocrinol. Metab. 45 (1977) 225–231. J.M. Barragry, D. Corless, J. Auton, N.D. Carter, R.G. Long, J.D. Maxwell, S. Switala, Plasma vitamin D-binding globulin in vitamin D deficiency, pregnancy and chronic liver disease, Clin. Chim. Acta 87 (1978) 359–365. P.G. Walsh, J.G. Haddad, “Rocket” immunoelectrophoresis assay of vitamin D-binding protein (Gc globulin) in human serum, Clin. Chem. 28 (1982) 1781–1783. R. Bouillon, J. Auwerx, L. Dekeyser, J. Fevery, W. Lissens, P. de Moor, Serum vitamin D metabolites and their binding protein in patients with liver cirrhosis, J. Clin. Endocrinol. Metab. 59 (1984) 86–89. P.J. Goldschmidt-Clermont, R.M. Galbraith, D.L. Emerson, P.A. Werner, A.E. Nel, W.M. Lee, Accurate quantitation of native Gc in serum and estimation of endogenous Gc: G-actin complexes by rocket immunoelectrophoresis, Clin. Chim. Acta 148 (1985) 173–183. M. Ogata, I. Nakasono, M. Iwasaki, S. Kubo, H. Suyama, Comparative immunological and electrophoretic analysis of Gc protein in sera from various animals, Comp. Biochem. Physiol. B 90 (1988) 193–199. C.S. Jorgensen, M. Christiansen, B. Norgaard-Pedersen, E. Ostergaard, F. V. Schiodt, I. Laursen, G. Houen, Gc globulin (vitamin D-binding protein) levels: an inhibition ELISA assay for determination of the total concentration of Gc globulin in plasma and serum, Scand. J. Clin. Lab. Invest. 64 (2004) 157–166. R.E. Arndt, Group specific component: isoelectric focusing subtyping and immunoblot detection, J. Forensic Sci. 34 (1989) 1318–1322. H. Van Baelen, R. Bouillon, P. de Moor, The heterogeneity of human Gcglobulin, J. Biol. Chem. 253 (1978) 6344–6345. H. Cleve, J. Constans, S. Berg, B. Hoste, G. Ishimoto, H. Matsumoto, E.K. Spees, W. Weber, Gc revisited: six further Gc-phenotypes delineated by isoelectric focusing and by polyacrylamide gel electrophoresis, Hum. Genet. 57 (1981) 312–316. H. Cleve, J. Constans, The mutants of the vitamin-D-binding protein: more than 120 variants of the GC/DBP system, Vox Sang. 54 (1988) 215–225. D.H. Coppenhaver, N.P. Sollenne, B.H. Bowman, Post-translational heterogeneity of the human vitamin D-binding protein (group-specific component), Arch. Biochem. Biophys. 226 (1983) 218–223. J. Svasti, A. Kurosky, A. Bennett, B.H. Bowman, Molecular basis for the three major forms of human serum vitamin D binding protein (groupspecific component), Biochemistry 18 (1979) 1611–1617. J. Lizana, I. Olsson, High resolution isoelectric focusing in a narrow pH interval for the phenotyping of vitamin D-binding protein (Gc-globulin), J. Clin. Chem. Clin. Biochem. 22 (1984) 545–549. S.A. Westwood, D.J. Werrett, Group-specific component: a review of the isoelectric focusing methods and auxiliary methods available for the separation of its phenotypes, Forensic Sci. Int. 32 (1986) 135–150. J. Svasti, B.H. Bowman, Human group-specific component. Changes in electrophoretic mobility resulting from vitamin D binding and from neuraminidase digestion, J. Biol. Chem. 253 (1978) 4188–4194. A. Braun, R. Bichlmaier, H. Cleve, Molecular analysis of the gene for the human vitamin-D-binding protein (group-specific component): allelic
[71] [72]
[73]
[74] [75] [76]
[77]
[78]
[79]
[80]
[81]
[82]
[83] [84]
[85]
[86]
[87]
differences of the common genetic GC types, Hum. Genet. 89 (1992) 401–406. R.L. Reynolds, G.F. Sensabaugh, Adv. Forensic Haemog. (1991) 158–161. C.S. Jørgensen, M. Christiansen, I. Laursen, L.B. Krogsøe, P. Højrup, L. Blou, G. Houen, Large scale purification and characterisation of nonglycosylated Gc globulin (vitamin D-binding protein) from plasma fraction IV, Biotecnol. Appl. Biochem. 44 (2006) 35–44. P. Kistler, H. Nitschmann, Large scale production of human plasma fractions. Eight years experience with the alcohol fractionation procedure of Nitschmann, Kistler and Lergier, Vox Sang. 7 (1962) 414–424. U.K. Laemmli, Cleavage of structural proteins during the assembly of the head of bacteriophage T4, Nature 227 (1970) 680–685. F.W. Studier, Analysis of bacteriophage T7 early RNAs and proteins on slab gels, J. Mol. Biol. 79 (1973) 237–248. W. Rist, M.P. Mayer, J.S. Andersen, P. Roepstorff, T.J. Jorgensen, Rapid desalting of protein samples for on-line microflow electrospray ionization mass spectrometry, Anal. Biochem. 342 (2005) 160–162. J. Gobom, E. Nordhoff, E. Mirgorodskaya, R. Ekman, P. Roepstorff, Sample purification and preparation technique based on nano-scale reversed-phase columns for the sensitive analysis of complex peptide mixtures by matrix-assisted laser desorption/ionization mass spectrometry, J. Mass Spectrom. 34 (1999) 105–116. A. Shevchenko, M. Wilm, O. Vorm, M. Mann, Mass spectrometric sequencing of proteins silver-stained polyacrylamide gels, Anal. Chem. 68 (1996) 850–858. M.R. Larsen, P. Hojrup, P. Roepstorff, Characterization of gel-separated glycoproteins using two-step proteolytic digestion combined with sequential microcolumns and mass spectrometry, Mol. Cell Proteomics. 4 (2005) 107–119. M.R. Larsen, S.J. Cordwell, P. Roepstorff, Graphite powder as an alternative or supplement to reversed-phase material for desalting and concentration of peptide mixtures prior to matrix-assisted laser desorption/ionization-mass spectrometry, Proteomics 2 (2002) 1277–1287. P. Hagglund, J. Bunkenborg, F. Elortza, O.N. Jensen, P. Roepstorff, A new strategy for identification of N-glycosylated proteins and unambiguous assignment of their glycosylation sites using HILIC enrichment and partial deglycosylation, J. Proteome Res. 3 (2004) 556–566. Y.Q. Yu, M. Gilar, J. Kaska, J.C. Gebler, A rapid sample preparation method for mass spectrometric characterization of N-linked glycans, Rapid Commun. Mass Spectrom. 19 (2005) 2331–2336. T. Yoshida, Peptide separation by Hydrophilic-Interaction Chromatography: a review, J. Biochem. Biophys. Methods 60 (2004) 265–280. E.C. Rickard, M.M. Strohl, R.G. Nielsen, Correlation of electrophoretic mobilities from capillary electrophoresis with physicochemical properties of proteins and peptides, Anal. Biochem. 197 (1991) 197–207. K.A. Benis, G.B. Schneider, The effects of vitamin D binding proteinmacrophage activating factor and colony-stimulating factor-1 on hematopoietic cells in normal and osteopetrotic rats, Blood 88 (1996) 2898–2905. N. Swamy, S. Ghosh, G.B. Schneider, R. Ray, Baculovirus-expressed vitamin B-binding protein-macrophage activating factor (DBP-maf) activates osteoclasts and binding of 25-hydroxyvitamin D3 does not influence this activity, J. Cell. Biochem. 81 (2001) 535–546. H.K. Datta, D.B. Cook, R.M. Kanan, Evidence for continuous basal generation of Gc-MAF; absence in infantile osteopetrosis and restoration after bone marrow transplant, Blood 93 (1999) 4026–4027.
Protein chemical characterization of Gc globulin (vitamin D-binding protein) isoforms; Gc-1f, Gc-1s and Gc-2 Maja Christiansen a , Charlotte S. Jørgensen b , Inga Laursen c , Daniel Hirschberg d , Peter Højrup d , Gunnar Houen a,⁎ a b
Department of Autoimmunology, Statens Serum Institut, Artillerivej 5, DK-2300 Copenhagen, Denmark Department of Bacteriology, Mycology and Parasitology, Statens Serum Institut, Copenhagen, Denmark c Department of Clinical Biochemistry, Statens Serum Institut, Copenhagen, Denmark d Institute of Biochemistry and Molecular Biology, University of Southern Denmark, Odense, Denmark Received 10 November 2006; revised 26 January 2007; accepted 31 January 2007 Available online 12 February 2007
Abstract Gc globulin, also called vitamin D-binding protein, is a plasma protein involved in the extracellular actin-scavenger system, vitamin D transport and possibly also other biological activities. Low levels of Gc globulin have been found to correlate with multiple organ failure and nonsurvival of patients with fulminant hepatic failure and trauma. Here, we characterize the dominant isoforms of plasma-derived Gc globulin from Cohn fraction IV paste with respect to amino acid sequence and posttranslational modifications. Gc globulin was purified in large scale and the isoforms separated by ion exchange chromatography. The separated isoforms and several commercial preparations of individual isoforms were characterized by mass spectrometry. This revealed that the major isoforms were non-glycosylated. Compared to the Gc-1f isoform the other dominating isoforms represented an Asp/Glu substitution (Gc-1s) and a Thr/Lys substitution (Gc-2) in agreement with DNA sequencing studies. The commercial preparations were found to represent mainly one or two isoforms. An O-linked glycan with a mass of 656 Da and terminating with a sialic acid residue was detected on a minor proportion of Gc globulin molecules. © 2007 Elsevier B.V. All rights reserved. Keywords: Gc globulin; Vitamin D-binding protein; Isoforms; Glycosylation; Mass spectrometry
1. Introduction Vitamin D-binding protein (DBP), also known as Gc (Group-specific component) globulin, is a plasma protein
Abbreviations: BCIP/NBT, 5-bromo-4-chloro-3-indolyl phosphate/nitro blue tetrazolium; CIE, crossed immunoelectrophoresis; DBP, vitamin D-binding protein; DTT, dithiothreitol; ESI, electro spray ionization; FPLC, fast protein liquid chromatography; Gc, group-specific component; HCCA, alpha-cyano-4hydroxycinnamic acid; HILIC, hydrophilic interaction chromatography; HPLC, high performance liquid chromatography; IEF, isoelectric focusing; LC, liquid chromatography; MAF, macrophage-activating factor; MALDI, matrix assisted laser desorption ionization; MS, mass spectrometry; RP-HPLC, reversed phase HPLC; TFA, trifluoroacetic acid; TNBP, tri-N-butylphosphate; TOF, time of flight; TTN, 50 mM Tris, 0.3 M NaCl, 1% Tween 20, pH 7.5 ⁎ Corresponding author. Tel.: +45 3268 3276; fax: +45 3268 3149. E-mail address: [email protected] (G. Houen). 1570-9639/$ - see front matter © 2007 Elsevier B.V. All rights reserved. doi:10.1016/j.bbapap.2007.01.005
serving two major physiological functions: (1) binding and transport of vitamin D and its metabolites and (2) binding, sequestration and removal of monomeric actin [1–5]. Gc globulin is a member of the albumin family, which also comprises serum albumin, alpha-fetoprotein and afamin [6–12]. The genes for the proteins of the family are located on chromosome 4 in a multigene cluster [13]. Gc globulin is a single chain polypeptide of 458 amino acid residues and, like albumin, it consists of three domains [6–10]. Binding of vitamin D occurs with high affinity at a site located in domain I [14–16]. The binding site for actin is independent of the vitamin D binding site and has been located to a region covering residues 360–372 [15]. However, crystallographic studies have shown that the binding of actin occurs in a large cavity composed of all three domains [17,18]. Depolymerization of filamentous F actin by plasma gelsolin and sequestration of monomeric G actin by Gc globulin is crucial to prevent
482
M. Christiansen et al. / Biochimica et Biophysica Acta 1774 (2007) 481–492
circulatory disturbances in cases of severe cell necrosis or apoptosis, and exhaustion of the extracellular actin scavenger system may lead to deposition of actin, obstruction of blood flow and, eventually, multiple organ failure [4,19–39]. In addition to the two major functions described above, Gc globulin has also been reported to have a co-chemotactic activity for C5a [40,41] and to be a precursor for a potent macrophage-activating factor (MAF) [42–47], which also has anti-tumor and anti-angiogenic properties [46–52]. The conversion of Gc globulin to Gc-MAF is supposed to occur by the action of glycosidases on an O-linked trisaccharide resulting in successive removal of neuraminic acid (by neuraminidase) and galactose (by β-galactosidase) leaving an O-linked N-acetyl-galactosamine (GalNAc) [42–49]. However, except for one study [53] the location and the structure of the Olinked glycan has not been thoroughly characterized and its existence is mainly based on indirect evidence. The concentration of Gc globulin in plasma is in the range of 200–500 mg/L [22,54–60] and it occurs in three major isoforms, which can be separated by isoelectric focusing (IEF) and are denoted Gc-1f, Gc-1s, Gc-2 [59,61–68]. The Gc-1f and Gc-1s isoforms have been reported to be sialylated [69], whereas Gc-2 was reported to be non-glycosylated [53]. However, DNA sequencing studies have associated the isoforms with amino acid substitutions at positions 416 and 420. Gc-1f has aspartic acid at position 416 and threonine at position 420, Gc-1s has glutamic acid and then threonine, and Gc-2 has aspartic acid and lysine at these positions [70,71]. In order to characterize the nature of the different isoforms, we have purified Gc globulin from plasma fraction IV [72] and characterized the three major isoforms by protein chemical methods, including mass spectrometry (MS) in combination with protease digestion. 2. Materials and methods 2.1. Large scale purification Human plasma Gc globulin was purified using fraction IV paste as starting material [72]. Paste IV was precipitated during ethanol fractionation (40% ethanol at pH 5.9) of human plasma as described by Kistler and Nitschmann [73]. The fraction IV paste was dissolved in 20 mM Tris, pH 8.0, by overnight stirring at 4 °C, followed by serial depth and delipid filtration (50 LA, 90 LA, and delipid filters). The filtrate was added NaCl to 0.04 M, and applied onto a QSepharose F.F. column. The column was washed using 20 mM Tris pH 8.0, 0.05 M NaCl, before elution with 20 mM Tris, 0.11 M NaCl, pH 8.0. The eluted material was ultra-filtered to half of the original volume, and dia-filtered against 20 mM phosphate buffer, pH 5.7. After ultra-dia-filtration the material was loaded onto a CM-Sepharose F.F. column, and the run-through fraction was collected, and treated by solvent-detergent (1% Tween 80, 0.3% tri-Nbutylphosphate (TNBP)) overnight at room temperature, followed by a dilution with 20 mM phosphate buffer, 48 mM NaCl, pH 9.4 to a final pH of 7.8. The solution was then applied onto a Q-Sepharose F.F. column. Following washing with 20 mM phosphate buffer, pH 7.8, 0.04 M NaCl, the column was eluted with 20 mM phosphate buffer, pH 7.8, 0.1 M NaCl. The peak fraction was collected and ultra-filtered, followed by gel filtration on a Sephacryl S200 column equilibrated with 8 mM Na2HPO4, 1.4 mM NaH2PO4, 130 mM NaCl, pH 7.0,. The main peak was collected and subjected to nano-filtration through a 20 nm and a 15 nm filter connected in series, and finally stabilized by addition of 5% maltose before sterile filtration and filling.
2.2. Small scale ion exchange chromatography for separation of isoforms 10 mg of human Gc globulin purified from plasma fraction IV was dialyzed against 20 mM sodium acetate pH 4.5, followed by further separation by ion exchange chromatography on a MonoS column connected to a fast protein liquid chromatography (FPLC) system, using a flow rate of 1 mL/min and collecting fractions of 2.5 mL. The column was eluted with 20 mM sodium acetate, pH 4.5, 0.5 M NaCl according to the following setup: 30 min of 0% elution buffer, followed by a 10 min linear increase from 0 to 50%, and then a linear increase from 50 to 80%. Immediately after the chromatography, the collected fractions were dialyzed against 50 mM Tris, pH 7.5 before further analysis.
2.3. SDS-PAGE SDS-PAGE was performed as described [74,75] using precast gels. The PAGE was performed using precast 4–20% Tris-glycine gels following the instructions of the manufacturer, using a variation of the recommended sample buffers. Samples were mixed with an equal volume of sample buffer (nonreduced SDS-PAGE: 50 mM Tris–HCl, pH 6.8, 10% glycerol, 2% SDS, 0.06% pyronin G; reduced SDS-PAGE: as non-reduced but added 0.1 M dithiothreitol (DTT)) before loading into the wells of the gel. After electrophoresis, gels were either stained with Coomassie Brilliant Blue (GelCode) according to the manufacturer's recommendation, or electroblotted to nitrocellulose membranes for subsequent immunoblotting.
2.4. Immunoblotting SDS-PAGE gels were sandwiched with nitrocellulose membranes between 12 sheets of Whatman paper no. 1 soaked in 10 times diluted electrophoresis buffer, and subjected to semi-dry electroblotting overnight at 0.1 mA/cm2. The membranes were blocked for 1 h in 50 mM Tris, 0.3 M NaCl, 1% Tween 20, pH 7.5 (TTN), incubated with monoclonal antibody against Gc globulin diluted 1:100 or rabbit antibodies against human serum proteins diluted 1:1000 in TTN for 1 h, followed by three washes using TTN. Next the membranes were incubated for 1 h with alkaline phosphatase-conjugated goat immunoglobulins against mouse or rabbit immunoglobulins diluted 1:1000 in TTN, followed by three washes in TTN. Finally, bound antibodies were visualized by incubation with 5-bromo-4-chloro-3indolyl phosphate/nitro blue tetrazolium (BCIP/NBT) substrate.
2.5. Isoelectric focusing Isoelectric focusing was performed using precast pH 3–7 gels following the directions of the manufacturer. Samples were mixed with an equal volume of the recommended sample buffer before loading onto the gel. After electrophoresis and fixing, gels were stained with Coomassie Brilliant Blue (GelCode) according to the manufacturer's directions.
2.6. Glycosidase treatment of Gc globulin Samples of approximately 100 μg Gc globulin were vacuum dried and redissolved in 100 μL 50 mM NH4HCO3, pH 6. The samples were treated individually, one with neuraminidase, alpha-mannosidase and beta-galactosidase, another with neuraminidase and O-glycosidase and a control sample was left untreated. 1 mU of each enzyme was utilized and samples were incubated over night at room temperature. In a separate experiment for ESI MS, selected fractions from the cation exchange chromatography and commercial preparations (approximately 100 pmol Gc globulin in 10 μL water) were treated with neuraminidase (3 μL reaction buffer and 0.3 μL sialidase A (GLYCO Prozyme, San Leandro, CA) for 1 h at 37 °C.
2.7. Matrix assisted laser desorption ionization (MALDI)-MS of intact protein A sample of intact Gc globulin was micro-purified on custom made POROS R150 columns and loaded onto a PerSeptive steel-target in a matrix sandwich of
M. Christiansen et al. / Biochimica et Biophysica Acta 1774 (2007) 481–492 sinapic acid (0.5 μL 20 μg/μL sinapic acid in acetone was deposited on the target; the sample was eluted in four droplets on top of this deposit with 0.8 μL of 20 μg/μL sinapic acid in 70% acetonitrile, 0.1% trifluoroacetic acid (TFA)). The samples were subjected to MS analysis on a PerSeptive Voyager Elite MALDI-MS instrument (PerSeptive Biosystems, Framingham, MA) with delayed extraction, operated in the positive linear mode.
2.8. Electrospray ionization (ESI)-MS Gc globulin ion exchange fractions and commercial preparations were analyzed by ESI-MS. The liquid chromatography (LC) system for protein desalting, has been described previously [76]. Solvents were delivered by two Applied Biosystems high performance liquid chromatography (HPLC) syringe pumps (model 140 B), one for desalting and one for elution of proteins. Samples (100 pmol) were dissolved in 100 μL 1% TFA and loaded onto a 200 μL stainless steel loop through a 6-port Rheodyne injection valve (7725i). When the valve was switched to inject position the sample was flushed from the loop to a 10-port two position Valco valve (C2-1000A) equipped with a C18 micro column (bed volume 2 μL). The sample was trapped on the micro column and desalted with a 400 μL/min flow of 0.05% (v/v) TFA for 1 min. The Valco valve switched automatically after the preset desalting time, and the sample was eluted for coupled MS analysis. The protein was eluted from the trap column with a 30 μL/min flow of 80% (v/v) acetonitrile and 0.05% (v/v) TFA. The LC system was coupled to an electrospray ionization quadrupole time-of-flight (TOF) (QToF Ultima, Micromass) mass spectrometer. Capillary voltage was set to 3.5 kV, cone voltage to 55 V and ion source block temperature to 80 °C with a desolvation gas flow of 500 L/h at 200 °C and nebulizing gas flow of 20 L/h at room temperature. The protein charge state envelope was deconvoluted by the MaxEnt 1 algorithm provided with the Masslynx software.
2.9. Bottom up analysis Approximately 200 μg of the large scale Gc globulin preparation was reduced with 4 mM DTT for 30 min at 56 °C, carbamidomethylated with 6 mM
483
iodacetamide for 30 min in the dark, at room temperature, and then digested with 4 μg trypsin overnight, at 37 °C. The sample was separated on an HPLC system (Amersham Pharmacia Biotech) using a Phenomenex Jupiter C18 column 250 × 4.60 mm. The eluted fractions were subjected to MS analysis on a PerSeptive Voyager Elite MALDI-MS instrument (PerSeptive Biosystems, Framingham, MA) with delayed extraction, operated in the positive reflector mode. Samples for MS analysis were micro-purified on custom made POROS R350 or R250 columns, and loaded onto a PerSeptive steel-target using alphacyano-4-hydroxycinnamic acid (HCCA) as matrix, according to a previously described procedure [77] with slight modifications: Samples were loaded onto the micro-columns in 0.1% TFA, washed twice, and eluted with 0.8 μL matrix solution (10 μg/μL in 70% acetonitrile, 0.1% TFA) in four spots. MS/MS spectra were obtained for selected peptides on an Applied Biosystems 4700 proteomics analyzer TOF-TOF MALDI instrument (Applied Biosystems, Foster City, CA), with POROS R250 micro-purification and 5 μg/μL HCCA matrix.
2.10. Mass spectrometric peptide mass mapping Identification of proteins from gels: Protein spots were cut from the gel and subjected to tryptic digestion according to a previously described procedure [78]. After excision of the gel pieces, they were washed 2 × 15 min in 50% acetonitrile and dehydrated in 100% acetonitrile. A volume of 50 μL 10 mM DTT in 100 mM NH4HCO3 (sufficient to cover the gel pieces) was added and the proteins were reduced for 45 min, at 56 °C. After cooling, the DTT solution was replaced with an equivalent volume of 55 mM iodacetamide in 100 mM NH4HCO3 and incubated for 30 min at room temperature in the dark. The washing and dehydrating steps were repeated, and the gel pieces were reswollen in digestion buffer containing 50 mM NH4HCO3 and 6.25 ng/μL trypsin for 45 min on ice. The excess digestion buffer was removed and replaced by 20 μL 50 mM NH4HCO3 without trypsin and left to digest overnight at 37 °C. Samples were purified on POROS R250 micro-columns and loaded onto the target using 5 μg/μL HCCA matrix according to [77], applying the same modifications as above, eluting in two droplets. MS and MS/MS spectra were acquired on an Applied Biosystems 4700 proteomics analyzer TOF-TOF MALDI instrument
Fig. 1. MALDI-MS of intact Gc globulin from large scale purification obtained in linear mode. An average molecular mass of 51.14 kDa is obtained by the average of the singly charged (51.143) and doubly charged (25.597) ions), which is within the estimated accuracy of 100 Da from the theoretical mass of 51.2 kDa (Table 1).
484
M. Christiansen et al. / Biochimica et Biophysica Acta 1774 (2007) 481–492
Table 1 Gc globulin isoform variation Isoform
Position 416
Position 420
Theoretical mass (Da)
pI (84)
Gc-2 Gc-1s Gc-1f
D E D
K T T
51215 51202 51188
5.21 5.16 5.15
Average mass and theoretical pI of the Gc globulin isoforms. The pI's were calculated using the values of Rickard et al. [84]. (Applied Biosystems, Foster City, CA). Proteinase K digestion followed by graphite micro-purification was performed as described [79,80], and hydrophilic interaction chromatography (HILIC) micro-purification was performed as in [81–83].
2.11. Bioinformatics Molecular mass and pI values were calculated using the GPMAW software package (Lighthouse data, Odense, Denmark).
3. Results Gc globulin was purified from plasma fraction IV by a large scale process comprising three ion-exchange chromatography steps (anion → cation → anion exchange) followed by a gelfiltration [72]. This yielded an 85% pure preparation containing minor amounts of alpha-1-antitrypsin, apolipoprotein A1, and haptoglobin as revealed by non-reducing SDS-PAGE and MS. In reducing SDS-PAGE alpha-1-antitrypsin has nearly the same mobility as Gc globulin, whereas the two are clearly separated by non-reducing SDS-PAGE. The actin-binding ability was verified by crossed immunoelectrophoresis (CIE) after the addition of actin in excess relative to Gc globulin (results not shown).
Isoelectric focusing revealed that the purified Gc globulin contained the three major isoforms denoted Gc-1f, Gc-1s, Gc-2 in almost equal amounts. The identity of the individual bands as Gc globulin, was confirmed by mass spectrometric peptide mapping (results not shown). Analysis of the large scale purified Gc globulin by MALDIMS showed an average molecular mass of 51.14 kDa (Fig. 1). In comparison, the theoretical average mass of the Gc globulin calculated by GPMAW is between 51.18 and 51.21 kDa, depending on the isoform (Table 1). An attached glycan of the structure GalNAc(Gal)Neu would contribute with 656 Da and should thus be easily detectable. The Gc globulin preparation was treated with a range of different glycosidases (neuraminidase, O-glycosidase and neuraminidase, beta-galactosidase, alpha-mannosidase) and subjected to analysis, all with similar mass results as the non-treated samples (results not shown) indicating that the major proportion of Gc globulin molecules are not glycosylated. Several techniques were applied to detect a partial glycosylation directly in the large scale Gc globulin preparation. Proteinase K digestion followed by graphite micro-purification, and hydrophilic interaction chromatography (HILIC) micropurification of a tryptic Gc globulin digest was performed, but no glycosylated peptides were found. Two samples of trypsindigested Gc globulin with and without prior glycosidase treatment were analyzed by RP-HPLC and subsequently compared. A few differences appeared, which were investigated by MS, but no glycosylation (neither N- nor O-linked) was identified. PNGase F treatment of the fraction IV extract used for purification of Gc globulin and of purified Gc globulin, with following analysis by one- or two-dimensional PAGE showed no changes for Gc globulin, whereas alpha-1-antitrypsin did
Fig. 2. Chromatogram of trypsin-digested Gc globulin. The separation was performed on a Phenomenex Jupiter C18 column 250 × 4.60 mm. Fractions were collected manually to obtain the best separation possible. All fractions were subjected to MALDI-MS analysis (not shown). Mass spectra of the marked peaks (star and triangle) are shown in Fig. 3A and B, respectively.
M. Christiansen et al. / Biochimica et Biophysica Acta 1774 (2007) 481–492
change its mobility (verifying that the enzyme treatment was effective). Furthermore, magnetic Dynabeads with immobilized lectins from Lens culinaris and Erythrina cristagalli were used, aiming at catching glycosylated peptides from different enzyme digests of Gc globulin, but no glycosylated peptides were isolated. An oxidation of possible sugar residues on Gc globulin with following biotin-hydrazide binding and isolation with streptavidin-coupled magnetic Dynabeads was also attempted, and Gc globulin was cleaved with CNBr and separated by RP-HPLC to isolate larger fragments for further digestion with AspN. However, subsequent analyses did not reveal glycosylation. To further investigate human plasma Gc globulin for the presence of glycosylation, a sample of the protein purified from fraction IV was reduced, carbamidomethylated, digested with trypsin, and subjected to reversed phase (RP)-HPLC to separate the obtained peptides (Fig. 2). All the different fractions eluted were analyzed by MALDI-MS and the peptides containing amino acids 414–420, previously reported to be O-glycosylated, were found to be without signs of glycosylation (Fig. 3). This region is also where the amino acid substitution between the different isoforms is present, hence generating three
485
different peptides, namely a 741.41 Da, a 1155.62 Da and a 1169.64 Da peptide respectively. The tree peptides were sequenced by MALDI-MS/MS and verified as being the varying peptides in the three isoforms (Fig. 4), thus confirming previous results found on the DNA level [70,71]. The sequences of the peptides are as follows: LPDATPTELAK for Gc-1f, LPEATPTELAK for Gc-1s and LPDATPK for Gc-2. The peptides containing the two potential N-glycosylation sites were also identified in the eluate from the RP-HPLC and analyzed by MALDI-MS, but no glycosylation was observed. This, altogether, indicates that the differences between the major Gc globulin isoforms are due to amino acid substitutions. In order to characterize the individual isoforms, we attempted separation of these by cation exchange chromatography (Fig. 5). Analysis of fractions by SDS-PAGE and IEF revealed essentially complete purity of Gc globulin and a partial separation of the isoforms (Fig. 6). Isoelectric focusing of three commercial Gc globulin preparations showed that two of them contained mainly the Gc-1s isoform (ICN and Biodesign) while the third (Biomedical) contained Gc-1s and Gc-2. Immunoblotting of the non-reduced SDS-PAGE gel using both monoclonal and polyclonal antibodies against Gc globulin and alpha-1-
Fig. 3. Characterization of Gc globulin isoforms by MALDI-MS. MS analysis of the two marked fractions from the RP-HPLC analysis in Fig. 2. The peak with the m/z 741.41 (A), accounts for the Gc globulin peptide 414–420 from Gc-2 and the peaks at m/z 1155.62 and 1169.64 (B) corresponds to the related peptides from Gc-1f and Gc-1s. All assignments were verified by MS/MS analysis (Fig. 4).
486
M. Christiansen et al. / Biochimica et Biophysica Acta 1774 (2007) 481–492
antitrypsin, verified that all the tested samples contained Gc globulin, and that alpha-1-antitrypsin was present in the Gc globulin purified from plasma fraction IV (results not shown).
ESI-MS was performed on all samples from Fig. 6. The raw data showed heterogeneity in the individual mass peaks. However, upon smoothing and deconvolution, the mass spectra in
Fig. 4. MALDI-MS/MS spectra of the three peptides m/z 741.41 (A), 1155.6 (B) and 1169.6 (C) from Fig. 3. The sequencing confirms the identity of the peptides as the ones containing amino acid residues 414–420 from the different Gc globulin isoforms.
M. Christiansen et al. / Biochimica et Biophysica Acta 1774 (2007) 481–492
487
Fig. 4 (continued).
Fig. 5. Cation exchange chromatography of large scale purified Gc globulin. 10 mg of Gc globulin was separated on a MonoS column by a gradient of 0.5 M NaCl. 2.5 mL fractions were collected as indicated.
488
M. Christiansen et al. / Biochimica et Biophysica Acta 1774 (2007) 481–492
Fig. 7 could be obtained. The observed heterogeneity is in agreement with the presence of multiple Gc globulin isoforms (Table 1) in all ion exchange fractions and commercial preparations.
For the commercial samples (Biomedicals, BioDesign) the peak around 50810 Da corresponds to the loss of approximately 400 Da relative to the main peak. This corresponds to the loss of the three N-terminal amino acid residues, Leu, Glu and Arg (398.5 Da), which was confirmed by Edman degradation. In fractions 27, 30, and 33 from the ion exchange chromatography a main peak appeared in a position similar to the commercial preparations, while the truncated form was not observed. However, multiple peaks at higher mass values were observed. Adducts of approximately 42 Da were observed, which could be a consequence of acetylation (42 Da) caused by the acetate buffer used for the ion exchange. Adducts around 51535, 51865 and 52190 could be caused by multiple additions of non-enzymatic glycation by maltose (324 Da) added to stabilize the final Gc globulin product. However, the mass at 51865 could also be the result of O-glycosylation. The samples from Biodesign and ICN show a very small peak at m/z 51865 which is approximately 655 above the mass of the intact Gc Globulin. This value is very close to the value of 656 Da for the expected glycosylation (see below). In order to ascertain the presence of the O-glycosylation, selected samples showing the 51865 peak (Biomedical, ICN and #27) were digested with Sialidase A and re-analyzed by ESI-MS (Fig. 8). All fractions showed a loss of the 51865 peak and the appearance of a prominent peak around 51576 consistent with the loss of one unit of sialic acid (291 Da). For fraction 27 the peak at 52190 disappeared with the appearance of a peak at 51897, again consistent with the loss of sialic acid. These results are consistent with an Oglycosylation in combination with up to two units of nonenzymatic glycation by maltose. 4. Discussion In this study we have performed a protein chemical characterization of the three major Gc globulin isoforms, Gc1f, Gc-1s and Gc-2. The isoforms were separated by ion exchange chromatography and characterized by mass spectrometry. The mass values of the isoforms, determined by MALDIMS and ESI-MS were in agreement with the sequence-derived theoretical mass values of the various isoforms. This was true for both Gc globulin purified in large scale from plasma fraction IV and for commercial preparations. Previous DNA sequencing studies have revealed amino acid substitutions at position 416 and 420, giving rise to three different isoforms. These amino acid substitutions have been verified here by peptide sequencing and have been shown to be
Fig. 6. SDS-PAGE and IEF analysis of fractions from cation exchange chromatography of Gc globulin. SDS-PAGE is shown with reducing conditions at the top (A) and non-reducing conditions in the middle (B); and an IEF-PAGE can be seen at the bottom (C). All three gels are loaded in identical manner: lane 1: standard marker for molecular weight or pH, lane 2: starting material (Gc globulin purified from fraction IV), lane 3: fraction 27, lane 4: fraction 30, lane 5: fraction 33, lane 6: Gc globulin from ICN, lane 7: Gc globulin from Biodesign, and lane 8: Gc globulin from Biomedicals. Lanes 3–5 are fractions from the separation in Fig. 2.
M. Christiansen et al. / Biochimica et Biophysica Acta 1774 (2007) 481–492
489
Fig. 7. ESI mass spectra of fractions from cation exchange chromatography (Fig. 5) and commercial Gc globulin preparations. Spectra show from the bottom and up: fraction 27, fraction 30, fraction 33, Gc globulin from ICN, Gc globulin from Biodesign and then Gc globulin from Biomedicals. The estimated accuracy is ± 10 Da.
without glycosylations. When comparing the theoretical pI values of Gc globulin with the different substitutions, only small variations appear (Table 1) [84]; however, as the theoretical pI values are based on a linear sequence, the diversity may be more pronounced between the folded proteins. Thus, it is highly
likely that these amino acid substitutions are the cause of the differences in mobility on an IEF-gel of the three major Gc globulin isoforms. In the commercial Gc globulin preparations, N-terminal truncation of three amino acid residues (LeuGluArg) was
Fig. 8. ESI mass spectra after sialidase treatment of fractions from cation exchange chromatography (Fig. 5) and commercial Gc globulin preparations. Spectra show from the bottom and up: fraction 27, Gc globulin from ICN, and then Gc globulin from Biomedicals. The estimated accuracy is ± 10 Da.
490
M. Christiansen et al. / Biochimica et Biophysica Acta 1774 (2007) 481–492
observed to varying degrees. This is most likely a result of aminopeptidase activity or of a trypsin-like endopeptidase activity on an accessible N-terminus. Non-enzymatic glycation by maltose and acetylation by buffer components was seen to some degree in the large scale Gc globulin preparation after stabilization with maltose and ion exchange chromatography in acetate buffer. Analysis of Gc globulin with respect to glycosylation proved to be a very difficult task. Human Gc globulin contains a potential N-glycosylation site at residues 272–274, but no Nglycosylation has been reported for this site and all our experiments capable of detecting N-glycosylations yielded negative results. Another potential, but rare, N-glycosylation site is present at residues 435–437 (Asp–Cys–Cys), but this site was retrieved in good yield in the non-glycosylated form. According to the literature, Gc globulin carries an O-linked trisaccharide composed of N-acetylgalactosamine with branched galactose and sialic acid [4,27,28,53,69]. However, apart from a single study of the glycosylation [53], the remaining reports are based on circumstantial evidence. Apart from its role in scavenging of actin and transport of vitamin D metabolites, Gc globulin has been found to play a role in macrophage activation [42–47], tumorigenesis/angiogenesis [46–52] and bone metabolism [85–87]. The presence of the glycan structure is based on the observation that Gc globulin attains macrophage-activating and anti-angiogenic properties upon combined sialidase and beta-galatosidase treatment. Several attempts were made at finding and identifying this glycosylation directly in the large scale Gc globulin preparation, but none were successful. We have previously reported that the macrophage activation may be caused by endotoxins present in the commercial glycosidases [72], and a possible explanation for the reported glycosylation could be, that most Gc globulin products are slightly contaminated by alpha-1-antitrypsin, which is known to be glycosylated [26], hence the supposed glycosylation might originate from this contamination. Alpha1-antitrypsin was found in both large scale and affinity purifications of Gc globulin, as well as in commercial products (results not shown). The problem is further complicated by the fact that available polyclonal and monoclonal antibodies against Gc globulin are not specific, but appear to recognize alpha-1antitrypsin as well. In contrast to the many attempts at detecting O-linked glycosylation directly on the Gc globulin preparation, a combination of ion exchange chromatography and ESI MS did reveal the presence of a glycan containing a sialic acid residue and having a composition consistent with the putative glycan described by Viau et al. [53]. This species seemed to be present in largest amounts in fraction 27, eluting early from the cation exchange column, in agreement with an extra negative charge conferred by the sialic acid residue. The sialylated species was also detected in two of the commercial preparations. The question therefore arises, whether the glycosylated species represent glycosylated Gc1f, Gc1s or Gc2. On the basis of the present results, we cannot answer this question. In fact, it may be suspected that all Gc globulin isoforms may experience O-linked glycosylation.
In conclusion, the final product of the Gc globulin large scale purification method applied here, was found to contain all three isoforms of the protein, whereas the commercial products tested mainly contained one or two isoforms. The isoforms were partially separated and characterized, and amino acid substitutions verifying previous results from DNA sequencing studies were found upon peptide sequencing at amino acid residues 416 and 420. These isoforms represent the bulk of the Gc globulin and the electrophoretic differences between the major isoforms are caused by the amino acid substitutions. However, a minor proportion of Gc globulin is, in fact, O-glycosylated with a GalNAc(Gal)Neu glycan. The precise location in the amino acid sequence of this glycan and its cellular origin must await further studies. Acknowledgments Kirsten B. Hansen, Dorthe T. Olsen and Inger Christiansen are thanked for skilful technical assistance. The Oticon foundation is gratefully acknowledged for financial support to Maja Christiansen. References [1] P. White, N. Cooke, The multifunctional properties and characteristics of vitamin D-binding protein, Trends Endocrinol. Metab. 11 (2000) 320–327. [2] R. Ray, Molecular recognition in vitamin D-binding protein, Proc. Soc. Exp. Biol. Med. 212 (1996) 305–312. [3] J.G. Haddad, Plasma vitamin D-binding protein (Gc-globulin): multiple tasks, J. Steroid Biochem. Mol. Biol. 53 (1995) 579–582. [4] W.M. Lee, R.M. Galbraith, The extracellular actin-scavenger system and actin toxicity, N. Engl. J. Med. 326 (1992) 1335–1341. [5] N.E. Cooke, J.G. Haddad, Vitamin D binding protein (Gc-globulin), Endocr. Rev. 10 (1989) 294–307. [6] N.E. Cooke, E.V. David, Serum vitamin D-binding protein is a third member of the albumin and alpha fetoprotein gene family, J. Clin. Invest. 76 (1985) 2420–2424. [7] F. Yang, J.L. Brune, S.L. Naylor, R.L. Cupples, K.H. Naberhaus, B.H. Bowman, Human group-specific component (Gc) is a member of the albumin family, Proc. Natl. Acad. Sci. U.S. A. 82 (1985) 7994–7998. [8] F. Yang, V.J. Luna, R.D. McAnelly, K.H. Naberhaus, R.L. Cupples, B.H. Bowman, Evolutionary and structural relationships among the groupspecific component, albumin and alpha-fetoprotein, Nucleic Acids Res. 13 (1985) 8007–8017. [9] F. Schoentgen, M.H. Metz-Boutigue, J. Jolles, J. Constans, P. Jolles, Homology between the human vitamin D-binding protein (group specific component), alpha-fetoprotein and serum albumin, FEBS Lett. 185 (1985) 47–50. [10] F. Schoentgen, M.H. Metz-Boutigue, J. Jolles, J. Constans, P. Jolles, Complete amino acid sequence of human vitamin D-binding protein (group-specific component): evidence of a three-fold internal homology as in serum albumin and alpha-fetoprotein, Biochim. Biophys. Acta 871 (1986) 189–198. [11] H.S. Lichenstein, D.E. Lyons, M.M. Wurfel, D.A. Johnson, M.D. McGinley, J.C. Leidli, D.B. Trollinger, J.P. Mayer, S.D. Wright, M.M. Zukowski, Afamin is a new member of the albumin, alpha-fetoprotein, and vitamin D-binding protein gene family, J. Biol. Chem. 269 (1994) 18149–18154. [12] H. Nishio, A. Dugaiczyk, Complete structure of the human alpha-albumin gene, a new member of the serum albumin multigene family, Proc. Natl. Acad. Sci. U.S. A. 93 (1996) 7557–7561. [13] Y.H. Song, A.K. Naumova, S.A. Liebhaber, N.E. Cooke, Physical and meiotic mapping of the region of human chromosome 4q11–q13
M. Christiansen et al. / Biochimica et Biophysica Acta 1774 (2007) 481–492
[14]
[15]
[16]
[17]
[18]
[19]
[20]
[21]
[22]
[23]
[24] [25]
[26]
[27]
[28]
[29]
[30]
[31]
[32]
[33]
encompassing the vitamin D binding protein DBP/Gc-globulin and albumin multigene cluster, Genome Res. 9 (1999) 581–587. R. Ray, R. Bouillon, H. Van Baelen, M.F. Holick, Photoaffinity labeling of human serum vitamin D binding protein and chemical cleavages of the labeled protein: identification of an 11.5-kDa peptide containing the putative 25-hydroxyvitamin D3 binding site, Biochemistry 30 (1991) 7638–7642. J.G. Haddad, Y.Z. Hu, M.A. Kowalski, C. Laramore, K. Ray, P. Robzyk, N.E. Cooke, Identification of the sterol- and actin-binding domains of plasma vitamin D binding protein (Gc-globulin), Biochemistry 31 (1992) 7174–7181. C. Verboven, A. Rabijns, M. De Maeyer, H. Van Baelen, R. Bouillon, C. De Ranter, A structural basis for the unique binding features of the human vitamin D-binding protein, Nat. Struct. Biol. 9 (2002) 131–136. L.R. Otterbein, C. Cosio, P. Graceffa, R. Dominguez, Crystal structures of the vitamin D-binding protein and its complex with actin: structural basis of the actin-scavenger system, Proc. Natl. Acad. Sci. U. S. A. 99 (2002) 8003–8008. J.F. Head, N. Swamy, R. Ray, Crystal structure of the complex between actin and human vitamin D-binding protein at 2.5 A resolution, Biochemistry 41 (2002) 9015–9020. J.G. Haddad, K.D. Harper, M. Guoth, G.G. Pietra, J.W. Sanger, Angiopathic consequences of saturating the plasma scavenger system for actin, Proc. Natl. Acad. Sci. U.S. A. 87 (1990) 1381–1385. K.D. Harper, J.F. McLeod, M.A. Kowalski, J.G. Haddad, Vitamin D binding protein sequesters monomeric actin in the circulation of the rat, J. Clin. Invest. 79 (1987) 1365–1370. S. Dueland, M.S. Nenseter, C.A. Drevon, Uptake and degradation of filamentous actin and vitamin D-binding protein in the rat, Biochem. J. 274 (Pt. 1) (1991) 237–241. P.J. Goldschmidt-Clermont, W.M. Lee, R.M. Galbraith, Proportion of circulating Gc (vitamin D-binding protein) in complexed form: relation to clinical outcome in fulminant hepatic necrosis, Gastroenterology 94 (1988) 1454–1458. P.A. Janmey, T.P. Stossel, S.E. Lind, Sequential binding of actin monomers to plasma gelsolin and its inhibition by vitamin D-binding protein, Biochem. Biophys. Res. Commun. 136 (1986) 72–79. P.A. Janmey, S.E. Lind, Capacity of human serum to depolymerize actin filaments, Blood 70 (1987) 524–530. J.F. Mc Leod, M.A. Kowalski, J.G. Haddad Jr., Interactions among serum vitamin D binding protein, monomeric actin, profilin, and profilactin, J. Biol. Chem. 264 (1989) 1260–1267. W.X. Tang, H.M. Bazaraa, H. Magiera, N.E. Cooke, J.G. Haddad, Electrophoretic mobility shift assay identifies vitamin D binding protein (Gc-globulin) in human, rat, and mouse sera, Anal. Biochem. 237 (1996) 245–251. M. Coue, J. Constans, A. Olomucki, Effects of serum vitamin-D-binding protein on actin in the presence of plasma gelsolin, Eur. J. Biochem. 160 (1986) 273–277. S.E. Lind, D.B. Smith, P.A. Janmey, T.P. Stossel, Role of plasma gelsolin and the vitamin D-binding protein in clearing actin from the circulation, J. Clin. Invest. 78 (1986) 736–742. W.M. Lee, D.L. Emerson, P.A. Werner, P. Arnaud, P. GoldschmidtClermont, R.M. Galbraith, Decreased serum group-specific component protein levels and complexes with actin in fulminant hepatic necrosis, Hepatology 5 (1985) 271–275. W.M. Lee, D.L. Emerson, W.O. Young, P.J. Goldschmidt-Clermont, D.J. Jollow, R.M. Galbraith, Diminished serum Gc (vitamin D-binding protein) levels and increased Gc:G-actin complexes in a hamster model of fulminant hepatic necrosis, Hepatology 7 (1987) 825–830. W.M. Lee, D. Reines, G.H. Watt, J.A. Cook, W.C. Wise, P.V. Halushka, R.M. Galbraith, Alterations in Gc levels and complexing in septic shock, Circ. Shock 28 (1989) 249–255. W.M. Lee, R.M. Galbraith, G.H. Watt, R.D. Hughes, D.D. McIntire, B.J. Hoffman, R. Williams, Predicting survival in fulminant hepatic failure using serum Gc protein concentrations, Hepatology 21 (1995) 101–105. F.V. Schiodt, S. Bondesen, N. Tygstrup, Serial measurements of serum Gc-
[34]
[35]
[36]
[37]
[38]
[39]
[40]
[41]
[42]
[43]
[44]
[45]
[46]
[47]
[48]
[49]
[50]
[51]
[52]
491
globulin in acetaminophen intoxication, Eur. J. Gastroenterol. Hepatol. 7 (1995) 635–640. F.V. Schiodt, S. Bondesen, I. Petersen, K. Dalhoff, P. Ott, N. Tygstrup, Admission levels of serum Gc-globulin: predictive value in fulminant hepatic failure, Hepatology 23 (1996) 713–718. F.V. Schiodt, P. Ott, S. Bondesen, N. Tygstrup, Reduced serum Gcglobulin concentrations in patients with fulminant hepatic failure: association with multiple organ failure, Crit. Care Med. 25 (1997) 1366–1370. B. Dahl, F.V. Schiodt, T. Kiaer, P. Ott, S. Bondesen, N. Tygstrup, Serum Gc-globulin in the early course of multiple trauma, Crit.Care Med. 26 (1998) 285–289. B. Dahl, F.V. Schiodt, P. Ott, F. Wians, W.M. Lee, J. Balko, G.E. O'Keefe, Plasma concentration of Gc-globulin is associated with organ dysfunction and sepsis after injury, Crit. Care Med. 31 (2003) 152–156. F.H. Wians Jr., W. Lin, L.P. Brown, F.V. Schiodt, W.M. Lee, Immunonephelometric quantification of group-specific component protein in patients with acute liver failure, Liver Transpl. Surg. 3 (1997) 28–33. W.O. Young, P.J. Goldschmidt-Clermont, D.L. Emerson, W.M. Lee, D.J. Jollow, R.M. Galbraith, Correlation between extent of liver damage in fulminant hepatic necrosis and complexing of circulating group-specific component (vitamin D-binding protein), J. Lab. Clin. Med. 110 (1987) 83–90. H.D. Perez, Gc globulin (vitamin D-binding protein) increases binding of low concentrations of C5a des Arg to human polymorphonuclear leukocytes: an explanation for its cochemotaxin activity, Inflammation 18 (1994) 215–220. R.R. Kew, J.A. Fisher, R.O. Webster, Co-chemotactic effect of Gc-globulin (vitamin D binding protein) for C5a. Transient conversion into an active co-chemotaxin by neutrophils, J. Immunol. 155 (1995) 5369–5374. N. Yamamoto, V.R. Naraparaju, Vitamin D3-binding protein as a precursor for macrophage activating factor in the inflammation-primed macrophage activation cascade in rats, Cell Immunol. 170 (1996) 161–167. N. Yamamoto, S. Homma, Vitamin D3 binding protein (group-specific component) is a precursor for the macrophage-activating signal factor from lysophosphatidylcholine-treated lymphocytes, Proc. Natl. Acad. Sci. U.S. A. 88 (1991) 8539–8543. N. Yamamoto, R. Kumashiro, Conversion of vitamin D3 binding protein (group-specific component) to a macrophage activating factor by the stepwise action of beta-galactosidase of B cells and sialidase of T cells, J. Immunol. 151 (1993) 2794–2802. H. Nagasawa, H. Sasaki, Y. Uto, S. Kubo, H. Hori, Association of the macrophage activating factor (MAF) precursor activity with polymorphism in vitamin D-binding protein, Anticancer Res. 24 (2004) 3361–3366. H. Nagasawa, Y. Uto, H. Sasaki, N. Okamura, A. Murakami, S. Kubo, K.L. Kirk, H. Hori, Gc protein (vitamin D-binding protein): Gc genotyping and GcMAF precursor activity, Anticancer Res. 25 (2005) 3689–3695. S.B. Mohamad, H. Nagasawa, Y. Uto, H. Hori, Preparation of Gc proteinderived macrophage activating factor (GcMAF) and its structural characterization and biological activities, Anticancer Res. 22 (2002) 4297–4300. S.B. Mohamad, H. Hori, H. Nagasawa, K. Usui, Y. Uto, Characterization of human Gc protein-derived macrophage activation factor (GcMAF) and its functional role in macrophage tumoricidal activity, Adv. Exp. Med. Biol. 510 (2003) 77–82. S.B. Mohamad, H. Nagasawa, H. Sasaki, Y. Uto, Y. Nakagawa, K. Kawashima, H. Hori, Gc protein-derived macrophage activating factor (GcMAF): isoelectric focusing pattern and tumoricidal activity, Anticancer Res. 23 (2003) 4451–4457. Y. Koga, V.R. Naraparaju, N. Yamamoto, Antitumor effect of vitamin Dbinding protein-derived macrophage activating factor on Ehrlich ascites tumor-bearing mice, Proc. Soc. Exp. Biol. Med. 220 (1999) 20–26. S. Kanda, Y. Mochizuki, Y. Miyata, H. Kanetake, N. Yamamoto, Effects of vitamin D(3)-binding protein-derived macrophage activating factor (GcMAF) on angiogenesis, J. Natl. Cancer Inst. 94 (2002) 1311–1319. O. Kisker, S. Onizuka, C.M. Becker, M. Fannon, E. Flynn, R. D'Amato, B. Zetter, J. Folkman, R. Ray, N. Swamy, S. Pirie-Shepherd, Vitamin D
492
[53]
[54]
[55]
[56]
[57]
[58]
[59]
[60]
[61] [62] [63]
[64] [65]
[66]
[67]
[68]
[69]
[70]
M. Christiansen et al. / Biochimica et Biophysica Acta 1774 (2007) 481–492 binding protein-macrophage activating factor (DBP-maf) inhibits angiogenesis and tumor growth in mice, Neoplasia 5 (2003) 32–40. M. Viau, J. Constans, H. Debray, J. Montreuil, Isolation and characterization of the O-glycan chain of the human vitamin-D binding protein, Biochem. Biophys. Res. Commun. 117 (1983) 324–331. R. Bouillon, H. Van Baelen, P. de Moor, The measurement of the vitamin D-binding protein in human serum, J. Clin. Endocrinol. Metab. 45 (1977) 225–231. J.M. Barragry, D. Corless, J. Auton, N.D. Carter, R.G. Long, J.D. Maxwell, S. Switala, Plasma vitamin D-binding globulin in vitamin D deficiency, pregnancy and chronic liver disease, Clin. Chim. Acta 87 (1978) 359–365. P.G. Walsh, J.G. Haddad, “Rocket” immunoelectrophoresis assay of vitamin D-binding protein (Gc globulin) in human serum, Clin. Chem. 28 (1982) 1781–1783. R. Bouillon, J. Auwerx, L. Dekeyser, J. Fevery, W. Lissens, P. de Moor, Serum vitamin D metabolites and their binding protein in patients with liver cirrhosis, J. Clin. Endocrinol. Metab. 59 (1984) 86–89. P.J. Goldschmidt-Clermont, R.M. Galbraith, D.L. Emerson, P.A. Werner, A.E. Nel, W.M. Lee, Accurate quantitation of native Gc in serum and estimation of endogenous Gc: G-actin complexes by rocket immunoelectrophoresis, Clin. Chim. Acta 148 (1985) 173–183. M. Ogata, I. Nakasono, M. Iwasaki, S. Kubo, H. Suyama, Comparative immunological and electrophoretic analysis of Gc protein in sera from various animals, Comp. Biochem. Physiol. B 90 (1988) 193–199. C.S. Jorgensen, M. Christiansen, B. Norgaard-Pedersen, E. Ostergaard, F. V. Schiodt, I. Laursen, G. Houen, Gc globulin (vitamin D-binding protein) levels: an inhibition ELISA assay for determination of the total concentration of Gc globulin in plasma and serum, Scand. J. Clin. Lab. Invest. 64 (2004) 157–166. R.E. Arndt, Group specific component: isoelectric focusing subtyping and immunoblot detection, J. Forensic Sci. 34 (1989) 1318–1322. H. Van Baelen, R. Bouillon, P. de Moor, The heterogeneity of human Gcglobulin, J. Biol. Chem. 253 (1978) 6344–6345. H. Cleve, J. Constans, S. Berg, B. Hoste, G. Ishimoto, H. Matsumoto, E.K. Spees, W. Weber, Gc revisited: six further Gc-phenotypes delineated by isoelectric focusing and by polyacrylamide gel electrophoresis, Hum. Genet. 57 (1981) 312–316. H. Cleve, J. Constans, The mutants of the vitamin-D-binding protein: more than 120 variants of the GC/DBP system, Vox Sang. 54 (1988) 215–225. D.H. Coppenhaver, N.P. Sollenne, B.H. Bowman, Post-translational heterogeneity of the human vitamin D-binding protein (group-specific component), Arch. Biochem. Biophys. 226 (1983) 218–223. J. Svasti, A. Kurosky, A. Bennett, B.H. Bowman, Molecular basis for the three major forms of human serum vitamin D binding protein (groupspecific component), Biochemistry 18 (1979) 1611–1617. J. Lizana, I. Olsson, High resolution isoelectric focusing in a narrow pH interval for the phenotyping of vitamin D-binding protein (Gc-globulin), J. Clin. Chem. Clin. Biochem. 22 (1984) 545–549. S.A. Westwood, D.J. Werrett, Group-specific component: a review of the isoelectric focusing methods and auxiliary methods available for the separation of its phenotypes, Forensic Sci. Int. 32 (1986) 135–150. J. Svasti, B.H. Bowman, Human group-specific component. Changes in electrophoretic mobility resulting from vitamin D binding and from neuraminidase digestion, J. Biol. Chem. 253 (1978) 4188–4194. A. Braun, R. Bichlmaier, H. Cleve, Molecular analysis of the gene for the human vitamin-D-binding protein (group-specific component): allelic
[71] [72]
[73]
[74] [75] [76]
[77]
[78]
[79]
[80]
[81]
[82]
[83] [84]
[85]
[86]
[87]
differences of the common genetic GC types, Hum. Genet. 89 (1992) 401–406. R.L. Reynolds, G.F. Sensabaugh, Adv. Forensic Haemog. (1991) 158–161. C.S. Jørgensen, M. Christiansen, I. Laursen, L.B. Krogsøe, P. Højrup, L. Blou, G. Houen, Large scale purification and characterisation of nonglycosylated Gc globulin (vitamin D-binding protein) from plasma fraction IV, Biotecnol. Appl. Biochem. 44 (2006) 35–44. P. Kistler, H. Nitschmann, Large scale production of human plasma fractions. Eight years experience with the alcohol fractionation procedure of Nitschmann, Kistler and Lergier, Vox Sang. 7 (1962) 414–424. U.K. Laemmli, Cleavage of structural proteins during the assembly of the head of bacteriophage T4, Nature 227 (1970) 680–685. F.W. Studier, Analysis of bacteriophage T7 early RNAs and proteins on slab gels, J. Mol. Biol. 79 (1973) 237–248. W. Rist, M.P. Mayer, J.S. Andersen, P. Roepstorff, T.J. Jorgensen, Rapid desalting of protein samples for on-line microflow electrospray ionization mass spectrometry, Anal. Biochem. 342 (2005) 160–162. J. Gobom, E. Nordhoff, E. Mirgorodskaya, R. Ekman, P. Roepstorff, Sample purification and preparation technique based on nano-scale reversed-phase columns for the sensitive analysis of complex peptide mixtures by matrix-assisted laser desorption/ionization mass spectrometry, J. Mass Spectrom. 34 (1999) 105–116. A. Shevchenko, M. Wilm, O. Vorm, M. Mann, Mass spectrometric sequencing of proteins silver-stained polyacrylamide gels, Anal. Chem. 68 (1996) 850–858. M.R. Larsen, P. Hojrup, P. Roepstorff, Characterization of gel-separated glycoproteins using two-step proteolytic digestion combined with sequential microcolumns and mass spectrometry, Mol. Cell Proteomics. 4 (2005) 107–119. M.R. Larsen, S.J. Cordwell, P. Roepstorff, Graphite powder as an alternative or supplement to reversed-phase material for desalting and concentration of peptide mixtures prior to matrix-assisted laser desorption/ionization-mass spectrometry, Proteomics 2 (2002) 1277–1287. P. Hagglund, J. Bunkenborg, F. Elortza, O.N. Jensen, P. Roepstorff, A new strategy for identification of N-glycosylated proteins and unambiguous assignment of their glycosylation sites using HILIC enrichment and partial deglycosylation, J. Proteome Res. 3 (2004) 556–566. Y.Q. Yu, M. Gilar, J. Kaska, J.C. Gebler, A rapid sample preparation method for mass spectrometric characterization of N-linked glycans, Rapid Commun. Mass Spectrom. 19 (2005) 2331–2336. T. Yoshida, Peptide separation by Hydrophilic-Interaction Chromatography: a review, J. Biochem. Biophys. Methods 60 (2004) 265–280. E.C. Rickard, M.M. Strohl, R.G. Nielsen, Correlation of electrophoretic mobilities from capillary electrophoresis with physicochemical properties of proteins and peptides, Anal. Biochem. 197 (1991) 197–207. K.A. Benis, G.B. Schneider, The effects of vitamin D binding proteinmacrophage activating factor and colony-stimulating factor-1 on hematopoietic cells in normal and osteopetrotic rats, Blood 88 (1996) 2898–2905. N. Swamy, S. Ghosh, G.B. Schneider, R. Ray, Baculovirus-expressed vitamin B-binding protein-macrophage activating factor (DBP-maf) activates osteoclasts and binding of 25-hydroxyvitamin D3 does not influence this activity, J. Cell. Biochem. 81 (2001) 535–546. H.K. Datta, D.B. Cook, R.M. Kanan, Evidence for continuous basal generation of Gc-MAF; absence in infantile osteopetrosis and restoration after bone marrow transplant, Blood 93 (1999) 4026–4027.