Glycobiology Advance Access originally published online on December 29, 2004
Glycobiology 2005 15(6):571-583; doi:10.1093/glycob/cwi037
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Glycobiology vol. 15 no. 6 © Oxford University Press 2004; all rights reserved.
Anti-pig antibody adsorption efficacy of 
-Gal carrying recombinant P-selectin glycoprotein ligand-1/immunoglobulin chimeras increases with core 2 ß1, 6-N-acetylglucosaminyltransferase expression
3 Division of Clinical Immunology, Karolinska Institutet, Karolinska University Hospital, S-141 86 Stockholm, Sweden; 4 Department of Clinical Chemistry, Sahlgrenska University Hospital, S-413 45 Göteborg, Sweden; 5 Department of Surgery, Sahlgrenska University Hospital, S-413 45 Göteborg, Sweden; 6 Clinical Research Centre, Karolinska Institutet, Karolinska University Hospital, S-141 86 Stockholm, Sweden; and 7 University College of South Stockholm, S-141 89 Stockholm, Sweden
1 These authors contributed equally to this work.
2 To whom correspondence should be addressed; e-mail: jan.holgersson{at}labmed.ki.se
Received on July 2, 2003; revised on September 13, 2004; accepted on December 23, 2004
 |
Abstract |
|---|
 Top Abstract IntroductionResultsDiscussionMaterials and methodsReferences |
|---|
We have previously described the construction of a P-selectin glycoprotein ligand-1—mouse immunoglobulin Fc fusion protein, which when transiently coexpressed with the porcine
1,3 galactosyltransferase in COS cells becomes a very efficient adsorber of xenoreactive, anti-pig antibodies. To relate the adsorption capacity with the glycan expression of individual fusion proteins produced in different cell lines, stable CHO-K1, COS, and 293T cells producing this fusion protein have been engineered. On 1,3 galactosyltransferase coexpression, high-affinity adsorbers were produced by both COS and 293T cells, whereas an adsorber of lower affinity was derived from CHO-K1 cells. Stable coexpression of a core 2 ß1,6 N-acetylglucosaminyltransferase in CHO-K1 cells led to increased -Gal epitope density and improved anti-pig antibody adsorption efficacy. ESI-MS/MS of O-glycans released from PSGL-1/mIgG2b produced in an 1,3 galactosyl- and core 2 ß1,6 N-acetylglucosaminyltransferase expressing CHO-K1 cell clone revealed a number of structures with carbohydrate sequences consistent with terminal Gal-Gal. In contrast, no O-glycan structures with terminal Gal-Gal were identified on the fusion protein when expressed alone or in combination with the 1,3 galactosyltransferase in CHO-K1 cells. In conclusion, the density of -Gal epitopes on PSGL-1/mIgG2b was dependent on the expression of O-linked glycans with core 2 structures and lactosamine extensions. The structural complexity of the terminal Gal-Gal expressing O-glycans with both neutral as well as sialic acid–containing structures is likely to contribute to the high adsorption efficacy. Key words: glycoconjugate / mass spectrometry / mucin / natural antibodies / xenotransplantation
|
Introduction |
|---|
|
TopAbstractIntroductionResultsDiscussionMaterials and methodsReferences |
|---|
Polyvalent interactions are ubiquitous in biology and are characterized by the simultaneous binding of multiple ligands on one biological entity, for example, a molecule or a cell surface, to multiple receptors on another. Examples of polyvalent interactions include the binding of microbes (viruses, bacteria, or bacterial toxins) to a cell surface, cell–cell binding, and binding of polyvalent molecules such as antibodies to a cell surface. Monovalent protein–carbohydrate interactions are usually of low affinity, and the binding strength in carbohydrate-dependent cell–cell and cell–microbe adhesions, or soluble protein–carbohydrate interactions, is a result of multivalent binding (Lee and Lee, 2000
; Mammen et al., 1998).
Inhibition of polyvalent interactions with monovalent inhibitors is usually ineffective even if the binding activity of the inhibitor has been structurally optimized. Polyvalent inhibitors have been designed by covalently linking two or more ligands via a spacer, by tethering the ligands to a polymer backbone, the head groups of a dendrimer, or a small protein (Kitov et al., 2000; Lindhorst et al., 1998; Mammen et al., 1998; Renkonen et al., 1997; Totani et al., 2003). An alternative approach involves noncovalent association of the ligand with the head groups in liposomes, membranes, or other surfaces (Mammen et al., 1998). We have developed a system that relies on the recombinant production of mucin-type proteins, which act as scaffolds for multivalent attachment of biologically active carbohydrates. The carbohydrate substitution is determined by the host cell, which is further engineered to express specific glycosyltransferases involved in the biosynthesis of a particular carbohydrate epitope (Liu et al., 1997, 2003; Lofling et al., 2002).
The mucin-type glycoprotein P-selectin glycoprotein ligand-1 (PSGL-1) on leukocytes is a high-affinity ligand for P-selectin on activated blood vessel endothelial cells (McEver et al., 1995). It is a homodimeric glycoprotein with two disulfide-bonded 120-kDa subunits (Moore et al., 1994) of type 1 transmembrane topology, each containing 402 amino acids (Sako et al., 1995). In the extracellular domain there are 15 repeats of a 10-amino-acid consensus sequence A Q(M) T T P(Q) P(LT) A A(PG) T(M) E that contains 3 or 4 potential sites for addition of O-linked oligosaccharides (Sako et al., 1993). Theoretically, PSGL-1 is predicted to have more than 53 sites for O-linked glycosylation and 3 sites for N-linked glycosylation in each monomer (Moore et al., 1994; Sako et al., 1993).
The chronic lack of donor organs for allotransplantation may be solved if transplantation of organs or tissues from animal to human, that is, xenotransplantation, could be performed. The pig is considered to be the most suitable donor species for human xenotransplantations, but several problems need to be solved before it can be used on a routine basis (Auchincloss and Sachs, 1998). The initial immunological barrier is caused by xenoreactive natural antibodies (XNAbs) in humans, apes, and Old world monkeys that are specific for a carbohydrate epitope, Gal1,3Gal (-Gal), present in other mammals, including pigs. The binding of XNAbs to -Gal epitopes on pig endothelial cells initiates a series of events that leads to graft loss within minutes to hours (Galili, 1993; Oriol et al., 1993). To prevent this, several methods, including plasmapheresis (Cairns et al., 1991) and adsorption on columns carrying anti-human IgM (Pascher et al., 1997), have been used to remove anti-pig antibodies from human serum. More selective adsorbers based on the -Gal carbohydrate determinant have also been developed to specifically remove anti-pig antibodies, leaving other antibody specificities untouched (Neethling et al., 1994).
Previously, we have used transient transfections to produce a recombinant mucin-type glycoprotein, PSGL-1/mIgG2b, that became heavily substituted with -Gal epitopes following coexpression with the porcine 1,3galactosyltransferase (1,3GalT) in COS cells (Liu et al., 1997). Using gas chromatography mass spectrometry (GC-MS), we estimated the number of terminal -Gal on the mucin to be 140 per dimer (Liu et al., 2003). The XNAb adsorption efficiency of PSGL-1/mIgG2b were on a carbohydrate molar basis 20 times higher compared to pig thyroglobulin immobilized on agarose beads, and 5000 and 30,000 times higher than Gal1,3Gal-conjugated agarose and macroporous glass beads, respectively (Liu et al., 2003).
To investigate the importance of the host cell for -Gal epitope density on and anti-pig antibody adsorption efficacy of PSGL-1/mIgG2b, the protein, together with the porcine 1,3GalT, was stably expressed in Chinese hamster ovary (CHO), COS, and 293T cells. Here we show that the level of -Gal substitution on PSGL-1/mIgG2b and its anti-pig antibody adsorption capacity were dependent on the host cell and correlated to the expression of a core 2 ß1,6 N-acetylglucosaminyltransferase (C2 GnTI). PSGL-1/mIgG2b expressed in CHO cells together with the porcine 1,3GalT and the C2 GnTI carried three different O-glycans having sequences consistent with terminal Gal-Gal.
|
Results |
|---|
|
TopAbstractIntroductionResultsDiscussionMaterials and methodsReferences |
|---|
Stable expression of
-Gal substituted PSGL-1/mIgG2b in different host cellsFollowing 15–20 days of culture in selection medium supplemented with puromycin, differently sized colonies of CHO-K1, COS7m6, and 293T cells were identified by phase contrast microscopy. Under the microscope, 192 colonies of each cell type were picked by pipette and transferred to two 96-well plates for further expansion under selection. An Ig sandwich enzyme-linked immunosorbent assay (ELISA) was used to assess fusion protein concentration in the supernatants of individual clones, and 31 CHO-K1, 8 COS7m6, and 36 293T colonies were anti-mouse IgG Fc positive. The top five secreting colonies from each cell line were moved to a 24-well plate and further expanded. The best expressing CHO-K1, COS7m6, and 293T clones were transfected with the
1,3GalT-encoding plasmid carrying the hygromycin B resistance gene. PSGL-1/mIgG2b-expressing cells that had stably integrated the 1,3GalT gene were selected using both puromycin and hygromycin. Twenty-seven CHO-K1, 3 COS7m6, and 31 293T colonies were selected. Colonies to be expanded were chosen based on the concentration of fusion protein and its relative level of -Gal epitope substitution as determined in anti-mouse IgG and Griffonia simplicifolia (GSA) I IB4 lectin ELISAs.
Immunoaffinity isolated PSGL-1/mIgG2b expressed in CHO-K1 (clone CHO-P and CHO-PG, respectively), COS7m6 (clone COS-P and COS-PG, respectively), and 293T (clone 293T-P and 293T-PG, respectively) cells with or without the porcine 1,3GalT was characterized by sodium dodecyl sulfate–polyacrylamide gel electrophoresis (SDS–PAGE) and western blotting (Figure 1). All cell lines produced an anti-mouse IgG Fc-reactive protein of 
300 kDa under nonreducing conditions (Figure 1A). In accordance with previous observations (Liu et al., 1997, 2003), PSGL-1/mIgG2b was produced as a dimer as indicated by the reduction to half the size on reduction (compare Figures 1A and 1B). The presence of -Gal epitopes on the fusion protein made in the different cell types was detected using the GSA I IB4 lectin (Figure 1B). Coexpression of the 1,3GalT in CHO-K1 (CHO-PG), COS7m6 (COS-PG), and 293T (293T-PG) led to expression of -Gal epitopes on the fusion protein as detected by the lectin. The lectin reactivity of PSGL-1/mIgG2b made in 293T cells without the 1,3GalT (293T-P) was unexpected and indicates the presence of -Gal residues other than the Galili antigen on that fusion protein (Figure 1B). The fusion protein produced in clones COS-PG and 293T-PG contained glycoforms of bigger size than the fusion protein produced in clone CHO-PG (Figure 1B). Additional bands of lower molecular weight seen in lane 1 and 2 of Figure 1A most likely represent fusion protein breakdown products.
|
The
-Gal epitope density on PSGL-1/mIgG2b is dependent on the host cell used for its productionThe relative
-Gal epitope density on PSGL-1/mIgG2b made in CHO, COS, and 293T cells was determined by ELISA (Figure 2). PSGL-1/mIgG2b made in COS-PG exhibited a 5.3-fold increase in the relative OD (GSA-reactivity/anti-mouse IgG reactivity) compared to PSGL-1/mIgG2b made in COS-P (Figure 2). For 293T cells there was a 3.1-fold increase in the relative OD, and for CHO cells there was just a 1.8-fold increase (Figure 2). The ELISA results were in agreement with the relative GSA lectin staining seen in the western blot experiments of immunoaffinity purified PSGL-1/mIgG2b (Figure 1B).
|
The anti-pig antibody adsorption efficacy of PSGL-1/mIgG2b made in different host cells correlates to its degree of -Gal substitution
Porcine aortic endothelial cell (PAEC) cytotoxicity was used to evaluate the ability of PSGL-1/mIgG2b produced in CHO-PG, COS-PG, and 293T-PG cells to adsorb the anti-pig reactive antibodies of human blood group AB serum (Figure 3). To reduce the PAEC cytotoxicity of human AB serum to 40% of maximum, 9.1 µg of CHO cell-made PSGL-1/mIgG2b were needed (Figure 3). For COS and 293T cell-made PSGL-1/mIgG2b, 16 and 4 times less, respectively, were needed to reduce the PAEC cytotoxicity to the same level. Furthermore, PSGL-1/mIgG2b made in CHO-PG cells could not reduce the PAEC cytotoxicity of blood group AB serum below 36%, whereas a reduction to 12% and 25% was seen with PSGL-1/mIgG2b made in COS-PG and 293T-PG cells, respectively, even under nonsaturating conditions.
|
Co- expression of a core 2 ß1,6 GlcNAc transferase in CHO cells improves PSGL-1/mIgG2b -Gal epitope density and anti-pig antibody adsorption efficacy
In an attempt to increase the number of -Gal epitopes on CHO-K1 cell-secreted mucin/Igs, CHO-K1 cells stably expressing PSGL-1/mIgG2b, 1,3GalT, and C2 GnTI were established, and PSGL-1/mIgG2b secreted by those cells were analyzed by ELISA, SDS–PAGE, and western blot using the anti-mouse IgG antibody and GSA I IB4. The apparent molecular weight of PSGL-1/mIgG2b increased following stable expression of the core 2 enzyme, indicating more complex glycans on the fusion protein (Figure 4). The -Gal epitope density on PSGL-1/mIgG2b showed a 13.0-fold increase compared to PSGL-1/mIgG2b made in clone CHO-P and a 7.4-fold increase compared to PSGL-1/mIgG2b made in clone CHO-PG (Figure 2). Furthermore, the anti-pig antibody adsorption efficacy of PSGL-1/mIgG2b produced in CHO-K1 cells stably expressing 1,3GalT and the C2 GnTI (CHO-PGC) was similar to the adsorption efficacy of PSGL-1/mIgG2b produced in 293T-PG and COS-PG (Figure 3), with 10 times less fusion protein needed to reduce the PAEC cytotoxicity of human AB serum to 40% of maximum compared to PSGL-1/mIgG2b made in CHO-PG.
|
Terminal -Gal epitope density on CHO-P-, CHO-PG- and CHO-PGC-derived PSGL-1/mIgG2b as determined by GC-MS
To verify the increased -Gal epitope density on PSGL-1/mIgG2b derived from CHO-PGC as compared to CHO-P and CHO-PG, a second analysis was performed using -galactosidase and GC-MS to quantify released galactose. PSGL-1/mIgG2b produced in CHO-PGC carries 20 mol terminal -Gal per mol protein, whereas PSGL-1/mIgG2b produced in CHO-PG and CHO-P carries 14 and 10 mol terminal -Gal, respectively. This confirms the ELISA results showing that CHO-PGC-derived PSGL-1/mIgG2b has a higher content of terminal -Gal.
Purification of recombinant PSGL-1/mIgG2b for structural characterization of its O-linked glycans
Recombinant PSGL-1/mIgG2b was purified from 1 L stirred flask cultures of stably transfected CHO-K1 cells expressing PSGL-1/mIgG2b alone (CHO-P), in combination with the porcine 1,3GalT (CHO-PG) or in combination with the 1,3GalT and the C2 GnTI (CHO-PGC). A two-step purification process, involving anti-mouse IgG affinity chromatography and gel filtration, was set up to fully remove contaminating glycosylated proteins that could interfere with the O-glycan structural analysis. Affinity purification of 2 L of cell supernatant from each cell clone resulted in 2.2 mg, 1.2 mg, and 0.95 mg of PSGL-1/mIgG2b from CHO-P, -PG, and -PGC, respectively, as assessed by ELISA. Further purification on a gel filtration column resulted in a final PSGL-1/mIgG2b yield of 0.22 mg, 0.19 mg, and 0.29 mg, respectively. The fractions eluted from the affinity and gel filtration columns were analyzed by SDS–PAGE and western blotting (shown here for CHO-P).
A glycoprotein staining kit was used in combination with Ruby to detect glycosylated as well as nonglycosylated proteins (Figure 5A and B), and an anti-PSGL-1 antibody confirmed the presence of PSGL-1/mIgG2b (Figure 5C). This antibody bound strongly to a band of 300 kDa (Figure 5C, lanes 2 and 4–9) representing the PSGL-1/mIgG2b dimer. A band of 150 kDa is also seen (lanes 4–6), derived from the fusion protein in its reduced form, as well as a weak band of 60–70 kDa (lanes 7–9) most likely representing fusion protein breakdown products. In Fig. 5 A and B, a 300 kDa band not stained by the anti PSGL-1 antibody can be seen also in lanes 1 and 3, most likely representing a protein derived from the cell culture medium. This is also supported by its presence in the affinity-purified supernatant (lane 3), which indicates that it is not adsorbed on the anti-IgG affinity column. However, a glycosylated band with a molecular weight of 50–60 kDa, not stained by the anti PSGL-1 antibody, can be seen in the affinity purified fraction (Figure 5A, lane 4). This protein is probably also derived from the cell culture medium and is adsorbed on the affinity column together with the fusion protein. This protein was removed by gel filtration, during which it eluted later (Figure 5A and B, lanes 7–9) than the fusion protein (Figure 5A, B, and C, lanes 5–6). Additional nonglycosylated proteins of 50–70 kDa were removed by gel filtration (Figure 5B, compare lane 4 with lanes 7–9). For each clone, the gel filtration fraction containing the highest amount of fusion protein was chosen for oligosaccharide release. As shown for CHO-P (Figure 5A and B, lane 5), this fraction did not contain any significant amounts of contaminating proteins, glycosylated or nonglycosylated.
|
MS of permethylated oligosaccharides released from purified, recombinant PSGL-1/mIgG2b
The permethylated oligosaccharides released from PSGL-1/mIgG2b produced in clones CHO-P and -PG gave similar MS spectra with two predominant groups of peaks around m/z 895.4 and 1256.5 (Figure 6A), whereas the mass spectrum of O-glycans released from PSGL-1/mIgG2b produced in CHO-PGC showed a more complex pattern (Figure 6B). The oligosaccharide sequences of the ions in the eletrospray ionization mass spectrometry (ESI-MS) spectra were deduced by tandem MS (MS/MS). The sequences and tentative structures thus obtained are shown in Table I. We will describe the results of the MS/MS analyses of the O-glycans on PSGL-1/mIgG2b produced in CHO-PGC cells shortly. All ions in the MS and MS/MS spectra were detected as sodiated ions.
|
|
MS/MS analyses of CHO-PGC
The most intense peak in the mother spectra is a sodiated molecular ion ([M+Na]+) at m/z 1548.7 representing a NeuAc-Hex-HexNol-HexN-Hex-Hex structure as assessed by MS/MS in sequential steps (Figure 7). MS2 of this ion gave two major fragment ions at m/z 951.3 ([M–NeuAc-Hex-O+Na]+) and 1173.5 ([M–NeuAc+Na]+) and several minor at m/z 506.1 ([M–Hex-Hex-HexN–NeuAc+Na]+), 620.2 ([NeuAc-Hex-O+Na]+), 690.2 ([Hex-Hex-HexN+Na]+), 751.3 ([M–Hex-Hex–NeuAc+Na]+ or [M–Hex–NeuAc-Hex+Na]+), 881.3 ([M–Hex-Hex-HexN+Na]+), 969.5 ([M–NeuAc-Hex+Na]+), and 1330.7 ([M–Hex+Na]+). The fragment ion at m/z 1173.5 was isolated and analyzed by MS3, resulting in fragment ions at m/z 951.4, 506.2, 690.3, and 751.5. The major peak, 951.4, was further analyzed by MS4 and gave rise to fragment ions at m/z 445.3 ([Hex-Hex+Na]+), 463.0 ([Hex-Hex-O+Na]+), 690.3, and 733.6 ([M–Hex–NeuAc-Hex-O+Na]+). Finally, the dominant fragment ion in the MS4 analysis (690.3) was analyzed by MS5. This resulted in sequence ions at m/z 415.1 and 445.3 representing a terminal Hex-Hex, the former ion having lost one oxygen and its methyl group. A Hex-Hex-O structure was also found (463.0). Furthermore, internal Hex-HexN structures were seen, with (472.2) and without (454.0) one oxygen linked to the hexose. Losses of O-Me (660.1), C-O-Me (648.2), and N-C-O-Me (619.4) from the Hex-Hex-HexN structure was also seen, where the last one probably represents loss of the N-acetyl group from the internal HexN. A major fragment ion at m/z 533.2 was also seen in the MS5 spectra. This ion corresponds to a cross-ring fragment of the innermost HexN (Figure 7) and indicates that the hexose is linked to the HexN in a 1-4 linkage. This sequence is most likely consistent with a sialidated core 2 elongated with a type 2 structure and a terminal Gal.
|
Apart from the ion at m/z 1548.7, two other sodiated molecular ions possibly terminating with Gal1,3Gal was found in the ESI-MS spectra of clone CHO-CPG at m/z 1578.7 and 1187.6. The ion at m/z 1578.7 was isolated for MS2 analysis. The result indicates a NeuGc-Hex-HexNol-HexN-Hex-Hex structure, with fragment ions at m/z 1173.5 ([M–NeuGc+Na]+), 951.5 ([M–NeuGc-Hex-O+Na]+), 911.3 ([M–Hex-Hex-HexN+Na]+), 690.3 ([Hex-Hex-HexN+Na]+), and 676.3 ([Hex-Hex-HexN–Me+Na]+). MS3 analysis of the 1173.5 ion resulted in a major fragment ion at m/z 951.5 and several minor at m/z 506.1 ([M–Hex-Hex-HexN–NeuAc+Na]+), 690.2, and 969.3 ([M–NeuAc-Hex+Na]+). The fragment ion at m/z 951.5 was analyzed by MS/MS in a fourth step, giving one major fragment ion at m/z 690.4 and a minor one at m/z 658.2 (690.4, O-Me). However, in the MS2 spectra of the ion at m/z 1578.7, unidentified fragment ions at m/z 981.5 and 1203.4 were observed. MS3 and MS4 analyses of the ion at m/z 1203.4 resulted in fragment ions at m/z 981.4, 720.4, 690.1, 506.1, and 688.3 (720.1, O-Me). The ion at m/z 720.4, seen in both the MS3 and MS4 spectra, is 30 mass units more than the characteristic fragment ion at m/z 690.1, representing a Hex-Hex-HexN sequence. Unfortunately, further MS/MS analysis of the ion at m/z 720.4 was not possible.
The other sodiated molecular ion in the ESI-MS spectra (Figure 6B) with a possible terminal Gal1-3Gal was observed at m/z 1187.6. MS2 experiment of this ion resulted in fragment ions at m/z 969.5 ([M–Hex+Na]+), 951.4 ([M–Hex-O+Na]+), 756.2 ([M–Hex-Hex+Na]+), 690.3 ([Hex-Hex-HexN+Na]+), 520.3 ([M–Hex-Hex-HexN+Na]+), and 445.1 ([Hex-Hex+Na]+), consistent with a Hex-HexNol-HexN-Hex-Hex or a core 2 with a type 2 elongation and a terminal Gal (Table I). Hence, both neutral and sialylated oligosaccharides potentially expressing terminal Gal1,3Gal is produced by clone CHO-CPG, although the sialidated (NeuAc) structure seem to be the most abundant one. In addition to this, several sialidated oligosaccharides without terminal Hex-Hex (Gal1-3Gal) can be seen, but at a lower relative abundance (Figure 6B and Table I).
|
Discussion |
|---|
|
TopAbstractIntroductionResultsDiscussionMaterials and methodsReferences |
|---|
The expression of carbohydrate chains on different proteins is dependent on several factors. Except for the host cell and hence the glycosyltransferase phenotype of that cell, the location of the individual transferases in the endoplasmic reticulum/Golgi, substrate and acceptor availability, as well as the transfer rate of individual proteins through the endoplasmic reticulum/Golgi, will affect the types of carbohydrate chains that will be produced (Bhatia and Mukhopadhyay, 1998
; Schachter, 2000). Also, the structure of the protein backbone can affect the number and kinds of glycans (Silverman et al., 2003; Ten Hagen et al., 2003). Hence, the performance of a recombinant protein whose therapeutic value is dependent on the glycans attached will thus vary with the host cell used for its production. The result of the present study show that the mucin/Ig chimeras produced in 1,3GT-expressing CHO-K1 cells had a lower -Gal epitope density and also a lower adsorption capacity than the mucin/Ig chimeras produced in 1,3GT-expressing COS and 293T cells (Figures 2 and 3). These two latter cell lines, in contrast to CHO cells that make exclusively simple core 1 O-glycans (Dennis, 1993), are known to express a ß1,6GlcNAcT responsible for core 2 structure biosynthesis (Holgersson, unpublished data) (Bierhuizen and Fukuda, 1992). This structure can be elongated with a galactose in a ß1,4 linkage, thus creating a type 2 chain (lactosamine). Even though lactosamine units are preferred precursors for the 1,3GalT enzyme (Joziasse et al., 1991), the -Gal epitope has been identified on different core saccharide chains (Rydberg et al., 1999). However, using MS/MS we were unable to detect any terminal Hex-O-Hex sequences on O-glycans of PSGL-1/mIgG2b produced in CHO-K1 cells without coexpression of the C2 GnTI (Figure 6A and Table I). This indicates that the core 1 structure is not a suitable acceptor for the 1,3GT.
In accordance with our results, previous studies on O-glycans expressed on recombinant proteins in CHO cells have shown the exclusive presence of mono- and disialylated core 1 structures (Hokke et al., 1995; Itoh et al., 2002). The fact that PSGL-1/mIgG2b produced in CHO-PG was a slightly better adsorber than the fusion protein produced in CHO-P might be explained by terminal -Gal on N-glycans of the mucin/Ig chimera. The 2,6 sialyltransferase in CHO cells responsible for the addition of sialic acid to O-glycans has been claimed to require the NeuNAc2,3Galß1,3GalNAc structure as an acceptor, because no monosialylated structures with 2-6 linked sialic acid were detected (Skrincosky et al., 1997). In contrast, we found a small peak in the MS2 spectra corresponding to a NeuNAc-HexNAcol fragment, indicating the presence of a Galß1,3(NeuNAc2,6)GalNAc structure.
To see if mucin–Ig fusion proteins produced in CHO cells could be modified to express -Gal epitopes at a level comparable to that on fusion proteins produced in COS and 293T cells, the 1,3GalT-expressing CHO cell line was stably transfected with the C2 GnTI. The introduction of this enzyme into CHO cells has previously been shown to increase core 2 branching and the number of lactosamine extensions on O-glycans (Bierhuizen and Fukuda, 1992; Kumar et al., 1996; Mitoma et al., 2003; Yeh et al., 1999). As can be seen (Figure 2), the level of -Gal epitopes on the fusion protein produced in CHO-PGC was strikingly increased, even exceeding the -Gal epitope levels on the fusion protein made in COS-PG and 293T-PG. PSGL-1/mIgG2b produced in CHO-PGC, CHO-PG, and CHO-P carried 20, 14, and 10 mol of terminal Gal per mol of protein, respectively. The fact that the ELISA and the -galactosidase cleavage resulted in different relative amounts of terminal -Gal on PSGL-1/mIgG2b produced in CHO-PGC, CHO-PG, and CHO-P may be explained by differences in the carbohydrate specificity of the lectin as compared to the -galactosidase, as well as their ability to access the -Gal epitopes on the protein. More relevant, though, is the fact that the anti-pig antibody adsorption capacity of the CHO-PGC-derived fusion protein was also increased (Figure 3). As revealed by MS (Figures 6B and 7, Table I), the increased -Gal epitope density was indeed due to core 2 branching and lactosamine extensions on O-glycans of PSGL-1/mIgG2b made in such engineered CHO cells.
The binding of an antibody to a carbohydrate epitope is dependent on the conformation of the presented saccharide. Histo-blood group antigens have been shown to adopt different conformations in solution (Imberty et al., 1995) and anti–blood group antibodies to recognize different areas of the carbohydrate determinant, the so-called micro-epitopes, that were conformation-dependent (Imberty et al., 1996). Studies have also shown that a receptor can induce a less energetically favorable conformation when bound to the ligand (Imberty et al., 1993). Studies on two xenoantigens, the -Gal-LacNAc and -Gal-Lewis x, revealed that even though the fucose residue of the latter epitope induced a conformational restraint on the lactosamine structure, the terminal Gal1,3Gal disaccharide was rather flexible (Corzana et al., 2002). Accordingly, histo-blood group antigens as well as xenoantigens, both recognized by so-called natural antibodies, can adopt several different conformations, and it might be difficult to predict the conformation of the epitope that these antibodies recognize.
Furthermore, the inner core structure of an oligosaccharide has been shown to influence the binding affinity of the protein–carbohydrate complex, although it was most likely not directly involved in the binding (Maaheimo et al., 1995). The structural analysis of the O-glycans expressed on CHO cells coexpressing the 1,3GalT and the C2 GnTI showed that the -Gal epitope was expressed on three different oligosaccharides (Figure 6B and Table I). It is tempting to speculate that apart from the effect of the core 2 and lactosamine structure, the N-acetyl- and N-glycolyl neuraminic acid situated on the core 1 branch in two of these oligosaccharides might influence the conformations adopted by these xenoantigens. Also, the sialic acids might in fact be a part of the binding epitope. If so, it is possible that a portion of the xenoreactive antibody repertoire recognizes these branched epitopes with a different binding specificity than the ones recognizing the Galili antigen (Galili et al., 1985). Furthermore, N-glycolylneuraminic acid is expressed in pigs (Bouhours et al., 1996; Malykh et al., 2001, 2003) but not in humans (Varki, 2001), and has been shown to be recognized by human xenoreactive antibodies (Zhu and Hurst, 2002). It is therefore possible that glycans containing N-glycolyl neuraminic acid can bind yet another group of xenoreactive antibodies.
The anti-pig antibody adsorption capacity of the mucin/Ig chimeras produced in COS and 293T equalled the adsorption capacity seen for the CHO-PGC clone derived mucin/Ig chimera. Several studies describing the structural characterization of glycans expressed on recombinant proteins have been performed. In most cases, CHO was used as the producing cell line, although a few studies have also been done on recombinant proteins expressed in other cell lines. Furthermore, most investigations were on N-glycans (Gervais et al., 2003; Sheeley et al., 1997; Van den Nieuwenhof et al., 2000; Yan et al., 1993). A structural characterization of the O-glycans expressed on the mucin/Ig chimeras made in COS and 293T cells might reveal more subtle differences in glycan expression between the different cell lines—differences that might influence the anti-pig antibody adsorption capacity and that can now only be speculated on. This study is currently under way.
Recently, pigs homozygous for the deletion of the 1,3GalT gene were produced (Phelps et al., 2003). The use of organs from such pigs in xenotransplantation may prove removal of anti-Gal antibodies unnecessary. However, it remains to be seen whether the absence of this glycosyltransferase will result in pigs completely devoid of -Gal (Sharma et al., 2003) or in pigs producing other carbohydrate epitopes to which humans produce antibodies, for example, the Thomsen-Friedenreich and the pk antigens (Cairns et al., 1996). Also, the removal of the 1,3GalT might increase the terminal addition of sialic acids, including the N-glycolyl neuraminic acid. Thus, antibodies recognizing these epitopes might still need to be removed for pig-to-human xenotransplantation to be successful. Furthermore, studies in mice have indicated that knocking out the 1,3GalT gene is not sufficient to prevent endothelial cell antibody binding, complement activation and hyperacute rejection (Miyata and Platt, 2003).
In conclusion, we have produced an efficient adsorber of anti-pig antibodies. The efficiency of this adsorber might be related to (1) the multivalent expression of O-glycans on the mucin/Ig chimera, (2) the presence of several different -Gal-expressing glycans on the mucin/Ig chimera as shown for clone CHO-PGC, and (3) the spatial presentation of these -Gal epitopes that comes from using a highly glycosylated mucin-type protein as glycan carrier.
|
Materials and methods |
|---|
|
TopAbstractIntroductionResultsDiscussionMaterials and methodsReferences |
|---|
Cell culture
CHO-K1, COS7m6, 293T, and the PAEC line PEC-A (Khodadoust et al., 1995)
were cultured in Dulbecco’s modified Eagle’s medium with 10% fetal bovine serum and 25 µg/ml gentamicin sulfate. The selection media contained puromycin (Sigma, St. Louis, MO), hygromycin (Calbiochem, La Jolla, CA), and G418 (Sigma) as indicated next.
Construction of expression plasmids
The porcine 1,3GalT (Gustafsson et al., 1994) and PSGL-1/mIgG2b expression plasmids were constructed as described (Liu et al., 1997). The C2 GnTI cDNA was amplified by polymerase chain reaction from an HL60 cDNA library using cgcgggctcgagaccatgctgaggacgttgctg and cgcgggcggccgctcagtgttttaatgtctc as forward and reverse primers, respectively. The vectors used to generate stable transfectants were bidirectional having the EF1 promoter upstream of a polylinker, a splice donor and acceptor site, and the bidirectional poly(A) addition signal of SV40; opposite in orientation to this transcription unit, and using the poly(A) signals from the opposite direction, was a second transcription unit consisting of the HSV TK promoter followed by the coding sequences for puromycin acetyltransferase (EF1/PAC), the hygromycin b (EF1/Hyg), and the neomycin (EF1/Neo) resistance genes (Chiu et al., unpublished data). The cDNAs of porcine 1,3GalT and PSGL-1/mIgG2b were swapped into the EF1/Hyg and EF1/PAC vectors, respectively, using Hind III and Not I. The gene of C2GnTI was swapped into EF1/Neo using Xho I and Not I.
DNA transfection and clonal selection
Adherent CHO-K1, COS7m6, and 293T cells were seeded in 75 cm2 T-flasks and were transfected 24 h later at a cell confluency of 70–80%. A modified polyethylenimine transfection method was used for transfection (Boussif et al., 1995; He et al., 2001). Twenty-four hours after transfection, cells in each T-flask were split into five 100-mm petri dishes and incubated in selection medium. The concentration of puromycin in the selection medium was 6.0, 1.5, and 1.0 µg/ml respectively, for CHO-K1, COS7m6, and 293T cells. A hygromycin b concentration of 550, 50, and 100 µg/ml was used for CHO-K1, COS7m6, and 293T cells, respectively, and a G418 concentration of 900 µg/ml was used for CHO-K1 cells. The selection medium was changed every third day. The drug resistant clones formed after 2 weeks. Clones were identified under the microscope and handpicked using a pipetman. Selected colonies were cultured in 96-well plates in the presence of selection drugs for another 2 weeks. Cell culture supernatants were harvested when the cells had reached 80–90% confluency, and the concentration of PSGL-1/mIgG2b was assessed by ELISA using a goat anti-mouse IgG Fc antibody. The CHO-K1, COS7m6, and 293T clones with the highest PSGL-1/mIgG2b expression were transfected with the porcine 1,3GalT-encoding plasmid and selected in hygromycin-containing medium. Resistant clones were isolated and characterized by ELISA, SDS–PAGE, and western blot using both a goat anti-mouse IgG Fc antibody and the GSA I IB4-lectin recognizing terminal -Gal. Two CHO clones with a high relative -Gal expression on PSGL-1/mIgG2b were further transfected with the C2 GnTI and selected in G418-containing medium. Expression of C2 GnTI was verified by an increase in size of PSGL-1/mIgG2b indicating more complex O-glycans.
SDS–PAGE and western blotting
SDS–PAGE was run by the method of Laemmli (1970) with 5% stacking gels and 8% resolving gels using a vertical Mini-Protean II electrophoresis system (Bio-Rad, Hercules, CA). Samples were electrophoretically run under reducing and nonreducing conditions. To increase the resolution, 4–15% gradient gels (Bio-Rad), or 4–12% gradient gels (Invitrogen, Carlsbad, CA) were used in some experiments. The latter gels were used in combination with the MES buffer (nvitrogen). A precision protein standard (Amersham Biosciences, Uppsala, Sweden) was applied as a reference for protein molecular weight determination. Protein gels were stained using the Pro Q Emerald 300 Glycoprotein detection kit in combination with Ruby (Molecular Probes, Leiden, Netherlands). These gels were visualized in a Flour-S Max MultiImager carrying a CCD camera. Separated proteins were also electrophoretically blotted onto Hybond C extra membranes (Amersham Biosciences), or nitrocellulose membranes (Invitrogen) using a Mini TransBlot (Bio-Rad) electrophoretic transfer cell (Towbin et al., 1979). Following blocking for 1 h in 3% bovine serum albumin (BSA) in phosphate buffered saline (PBS) with 0.2% Tween 20, the membranes were probed for 1 h at room temperature with peroxidase-conjugated GSA I IB4-lectin (Sigma) diluted to a concentration of 1 µg/ml, peroxidase-conjugated goat anti-mouse IgG Fc antibodies (Sigma) diluted 1:1000, and a mouse anti-PSGL-1 antibody (clone KPL-1, BD PharMingen, San Diego, CA) diluted 1:1000. Secondary antibody was a peroxidase-conjugated goat anti-mouse IgG F(ab)'2 (Sigma) diluted 1:50,000. All dilutions were done in blocking buffer. The membranes were washed three times with PBS containing 0.2% Tween 20 between and after incubations. Bound lectins and antibodies were visualized by chemiluminescence using the ECL kit according to the manufacturer’s instructions (Amersham Biosciences).
-Gal epitope density on and quantification of PSGL-1/mIgG2b using ELISA
The concentration of recombinant PSGL-1/mIgG2b in cell culture supernatants and its relative -Gal epitope densitywas determined by a two-antibody sandwich ELISA. The 96-well ELISA plate was coated overnight at 4°C with an affinity-purified, polyclonal goat anti-mouse IgG Fc antibody (Cappel/Organon Teknika, Durham, NC) at a concentration of 20 µg/ml. The plate was blocked with 1% BSA in PBS for 1 h. The supernatant containing PSGL-1/mIgG2b was incubated for 4 h and then washed three times with PBS containing 0.5% (v/v) Tween 20. After washing, the plate was incubated with a peroxidase-conjugated, anti-mouse IgG Fc antibody (Sigma) in a 1:3000 dilution or with peroxidase-conjugated GSA I IB4-lectin (Sigma) diluted 1:2000 for 2 h. Bound peroxidase-conjugated antibody or peroxidase-conjugated GSA-lectin was visualized with 3,3',5,5'-tetramethylbenzidine dihydrochloride (Sigma, Sweden). The reaction was stopped by 2 M H2SO4 and the plates read at 450 nm. The PSGL-1/mIgG2b concentration was estimated using for calibration a dilution series of purified mouse IgG Fc fragments (Jackson ImmunoResearch Labs, West Grove, PA) resuspended in the medium used for fusion protein production or in PBS containing 1% BSA. The -Gal epitope density was determined by comparing the relative OD from the two ELISAs (GSA-reactivity/anti-mouse IgG reactivity).
-Gal epitope density on PSGL-1/mIgG2b as determined by -galactosidase and GC-MS
The -Gal epitope density on PSGL1-mIgG2b produced in CHO-P, CHO-PG, and CHO-PGC was determined by enzymatic release of terminal -Gal and subsequent GC-MS analysis as described previously (Liu et al., 2003).
PAEC cytotoxicity assay
The PAEC cytotoxicity assay was performed as described elsewhere (Liu et al., 2003). The amount of PSGL-1/mIgG2b needed from each cell clone to reduce cell cytotoxicity to 40% of maximum (y = 0.4) was deduced from the formula describing the curve obtained after linear regression of the measured values for each fusion protein, and thereafter calculating the corresponding x-value (microgram adsorber).
Stirred flask batch cultures of CHO clones
Each batch culture was started with 6.0 x 107 cells (representing 10 175-cm2 T-flasks with cells of 90–100% confluency). After digestion with trypsin (0.5 mg/ml)· ethylenediamine tetra-acetic acid (0.2 mg/ml), cells were resuspended in a small volume of medium and centrifuged at 200 x g for 5 min to remove excess of trypsin. The cell density was determined by counting the cells in a Bürker chamber, and medium was added to a final concentration of 3.0 x 105 cells/ml. The cell suspension was transferred to 1-L stirred flasks, and a cell spin device (Integra Biosciences, Wallisellen, Switzerland) was used to stir the cultures at a speed of 60 rpm. PSGL-1/mIgG2b secreting CHO-K1 cells expressing 1,3GalT alone or in combination with C2 GnTI were cultured in the presence of puromycin (200 µg/ml) or puromycin (200 µg/ml) and G418 (500 µg/ml), respectively. The cells were counted every second day. When the cell density reached 5.0 x 105 cells/ml, new medium was added so that the cell density once again equaled 3.0 x 105 cells/ml. This was repeated until the cell suspension volume reached 1000 ml. Cells were then continuously cultured until cell viability was reduced to 50%.
Purification of recombinant PSGL-1/mIgG2b
The supernatants were cleared from debris by centrifugation at 1420 x g for 20 min. Cleared supernatants were passed through a column containing 10 ml goat anti-mouse IgG (whole molecule)-agarose (Sigma) at a flow rate of 0.5 ml/min. Following washing with 120 ml PBS, bound fusion protein was eluted with 120 ml 3 M NaSCN. The contents of the tubes containing the fusion protein was pooled following analysis by SDS–PAGE and western blotting using anti-mouse IgG for detection. The fraction with PSGL-1/mIgG2b was dialyzed against distilled water, lyophilised, and resuspended in 1–2 ml distilled H2O. The concentration of the fusion protein was determined by ELISA. To remove low-molecular-weight contaminants, the fusion protein was further purified by gel filtration on a HiPrep 16/60 Sephacryl S-200 HR column (Amersham Biosciences) eluted with PBS at a flow rate of 0.5 ml/min using a FPLC (Pharmacia Biotech, Sweden). Five-milliliter fractions were collected, and tubes containing protein were identified by UV spectrophotometry at 280 nm. Pooled fractions were again analyzed by SDS–PAGE and western blotting, pooled, dialyzed, and resuspended in distilled water.
Chemical release and permethylation of O-linked glycans from purified PSGL-1/mIgG2b
Oligosaccharides were released by ß-elimination as described (Carlstedt et al., 1993). Released oligosaccharides were evaporated under a stream of nitrogen at 45°C and permethylated according to Ciucanu and Kerek (1984), with slight modifications as described (Hansson and Karlsson, 1993).
MS
ESI-MS in positive-ion mode was performed using an LCQ ion-trap mass spectrometer (ThermoFinnigan, San Jose, CA). The sample was dissolved in methanol:water (1:1) and introduced into the mass spectrometer at a flow rate of 5–10 µl/min. Nitrogen was used as sheath gas and the needle voltage set to 4.0 kV. The temperature of the heated capillary was set to 200°C. A total of 10–20 spectra were summed to yield the ESI-MS and ESI-MS/MS spectra.
|
Acknowledgements |
|---|
This work was supported by the Swedish Research Council (K2002-06X-13031-04A and 11621), the Swedish Agency for Innovation Systems, and AbSorber AB. J.H. holds a position within the program Glycoconjugates in Biological Systems, financed by the Swedish Foundation for Strategic Research. We thank professor Andrej Weintraub for his assistance.
|
Abbreviations |
|---|
BSA, bovine serum albumin; C2 GnTI, core 2 ß1,6 N-acetylglucosaminyltransferase; CHO, Chinese hamster ovary; ELISA, enzyme-linked immunosorbent assay; ESI-MS, electrospray ionization mass spectrometry; GC-MS, gas chromatography mass spectrometry; GSA, Griffonia simplicifolia agglutinin; PAEC, porcine aortic endothelial cell; PBS, phosphate buffered saline; PSGL-1, P-selectin glycoprotein ligand-1; SDS–PAGE, sodium dodecyl sulfate–polyacrylamide gel electrophoresis; XNAb, xenoreactive natural antibody
|
References |
|---|
|
TopAbstractIntroductionResultsDiscussionMaterials and methodsReferences |
|---|
Auchincloss, H. Jr. and Sachs, D.H. (1998) Xenogeneic transplantation. Annu. Rev. Immunol., 16, 433–470.
Bhatia, P.K. and Mukhopadhyay, A. (1998) Protein glycosyation: implications for in vivo functions and therapeutic applications. Adv. Biochem. Eng. Biotechnol., 64, 155–201.
Bierhuizen, M.F. and Fukuda, M. (1992) Expression cloning of a cDNA encoding UDP-GlcNAc:Gal beta 1-3-GalNAc-R (GlcNAc to GalNAc) beta 1-6GlcNAc transferase by gene transfer into CHO cells expressing polyoma large tumor antigen. Proc. Natl Acad. Sci. USA, 89, 9326–9330.
Bouhours, D., Pourcel, C., and Bouhours, J.E. (1996) Simultaneous expression by porcine aorta endothelial cells of glycosphingolipids bearing the major epitope for human xenoreactive antibodies (Gal alpha 1-3Gal), blood group H determinant and N-glycolylneuraminic acid. Glycoconj. J., 13, 947–953.
Boussif, O., Lezoualc’h, F., Zanta, M.A., Mergny, M.D., Scherman, D., Demeneix, B., and Behr, J.P. (1995) A versatile vector for gene and oligonucleotide transfer into cells in culture and in vivo: polyethylenimine. Proc. Natl Acad. Sci. USA, 92, 7297–7301.
Cairns, T.D., Taube, D.H., Stevens, N., Binns, R., and Welsh, K.I. (1991) Xenografts—future prospects for clinical transplantation. Immunol. Lett., 29, 167–170.
Cairns, T., Lee, J., Goldberg, L.C., Hakim, N., Cook, T., Rydberg, L., Samuelsson, B., and Taube, D. (1996) Thomsen-Friedenreich and PK antigens in pig-to-human xenotransplantation. Transplant. Proc., 28, 795–796.
Carlstedt, I., Herrmann, A., Karlsson, H., Sheehan, J., Fransson, L.A., and Hansson, G.C. (1993) Characterization of two different glycosylated domains from the insoluble mucin complex of rat small intestine. J. Biol. Chem., 268, 18771–18781.
Ciucanu, I. and Kerek, F. (1984) A simple and rapid method for the permethylation of carbohydrates. Carbohydr. Res., 131, 209–217.
Corzana, F., Bettler, E., Herve du Penhoat, C., Tyrtysh, T.V., Bovin, N.V., and Imberty, A. (2002) Solution structure of two xenoantigens: alpha Gal-LacNAc and alpha Gal-Lewis X. Glycobiology, 12, 241–250.
Dennis, J.W. (1993) Core 2 GlcNAc-transferase and polylactosamine expression in O-glycans. Glycobiology, 3, 91–93.
Galili, U. (1993) Interaction of the natural anti-Gal antibody with alpha-galactosyl epitopes: a major obstacle for xenotransplantation in humans. Immunol. Today, 14, 480–482.
Galili, U., Macher, B.A., Buehler, J., and Shohet, S.B. (1985) Human natural anti-alpha-galactosyl IgG. II. The specific recognition of alpha (1-3)-linked galactose residues. J. Exp. Med., 162, 573–582.
Gervais, A., Hammel, Y.A., Pelloux, S., Lepage, P., Baer, G., Carte, N., Sorokine, O., Strub, J.M., Koerner, R., Leize, E., and Van Dorsselaer, A. (2003) Glycosylation of human recombinant gonadotrophins: characterization and batch-to-batch consistency. Glycobiology, 13, 179–189.
Gustafsson, K., Strahan, K., and Preece, A. (1994) Alpha 1,3 galactosyltransferase: a target for in vivo genetic manipulation in xenotransplantation. Immunol. Rev., 141, 59–70.
Hansson, G.C. and Karlsson, H. (1993) Gas chromatography and gas chromatography-mass spectrometry of glycoprotein oligosaccharides. Methods Mol. Biol., 14, 47–54.
He, Z., She, R., Sumitran-Holgersson, S., Blomberg, P., Islam, K.B., and Holgersson, J. (2001) The in vitro activity and specificity of human endothelial cell-specific promoters in porcine cells. Xenotransplantation, 8, 202–212.
Hokke, C.H., Bergwerff, A.A., Van Dedem, G.W., Kamerling, J.P., and Vliegenthart, J.F. (1995) Structural analysis of the sialylated N- and O-linked carbohydrate chains of recombinant human erythropoietin expressed in Chinese hamster ovary cells. Sialylation patterns and branch location of dimeric N-acetyllactosamine units. Eur. J. Biochem., 228, 981–1008.
Imberty, A., Perez, S., Hricovini, M., Shah, R.N., and Carver, J.P. (1993) Flexibility in a tetrasaccharide fragment from the high mannose type of N-linked oligosaccharides. Int. J. Biol. Macromol., 15, 17–23.
Imberty, A., Mikros, E., Koca, J., Mollicone, R., Oriol, R., and Perez, S. (1995) Computer simulation of histo-blood group oligosaccharides: energy maps of all constituting disaccharides and potential energy surfaces of 14 ABH and Lewis carbohydrate antigens. Glycoconj. J., 12, 331–349.
Imberty, A., Mollicone, R., Mikros, E., Carrupt, P. A., Perez, S., and Oriol, R. (1996) How do antibodies and lectins recognize histo-blood group antigens? A 3D-QSAR study by comparative molecular field analysis (CoMFA). Bioorg. Med. Chem., 4, 1979–1988.
Itoh, S., Kawasaki, N., Ohta, M., and Hayakawa, T. (2002). Structural analysis of a glycoprotein by liquid chromatography-mass spectrometry and liquid chromatography with tandem mass spectrometry. Application to recombinant human thrombomodulin. J. Chromatogr. A, 978, 141–152.
Joziasse, D.H., Shaper, J.H., Jabs, E.W., and Shaper, N.L. (1991) Characterization of an alpha 1-3-galactosyltransferase homologue on human chromosome 12 that is organized as a processed pseudogene. J. Biol. Chem., 266, 6991–6998.
Khodadoust, M.M., Candal, F.J., and Maher, S.E. (1995) PEC-A: an immortalized porcine aortic endothelial cell. Xenotransplantation, 2, 79.
Kitov, P.I., Sadowska, J.M., Mulvey, G., Armstrong, G.D., Ling, H., Pannu, N.S., Read, R.J., and Bundle, D.R. (2000) Shiga-like toxins are neutralized by tailored multivalent carbohydrate ligands. Nature, 403, 669–672.
Kumar, R., Camphausen, R.T., Sullivan, F.X., and Cumming, D.A. (1996) Core2 beta-1,6-N-acetylglucosaminyltransferase enzyme activity is critical for P-selectin glycoprotein ligand-1 binding to P-selectin. Blood, 88, 3872–3879.
Laemmli, U.K. (1970) Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature, 227, 680–685.
Lee, R.T. and Lee, Y.C. (2000) Affinity enhancement by multivalent lectin-carbohydrate interaction. Glycoconj. J., 17, 543–551.
Lindhorst, T.K., Kieburg, C., and Krallmann-Wenzel, U. (1998) Inhibition of the type 1 fimbriae-mediated adhesion of Escherichia coli to erythrocytes by multiantennary alpha-mannosyl clusters: the effect of multivalency. Glycoconj. J., 15, 605–613.
Liu, J., Qian, Y., and Holgersson, J. (1997) Removal of xenoreactive human anti-pig antibodies by absorption on recombinant mucin-containing glycoproteins carrying the Gal alpha1,3Gal epitope. Transplantation, 63, 1673–1682.
Liu, J., Weintraub, A., and Holgersson, J. (2003) Multivalent Galalpha1,3 Gal-substitution makes recombinant mucin-immunoglobulins efficient absorbers of anti-pig antibodies. Xenotransplantation, 10, 149–163.
Lofling, J.C., Hauzenberger, E., and Holgersson, J. (2002) Absorption of anti-blood group A antibodies on P-selectin glycoprotein ligand-1/immunoglobulin chimeras carrying blood group A determinants: core saccharide chain specificity of the Se and H gene encoded alpha1,2 fucosyltransferases in different host cells. Glycobiology, 12, 173–182.
Maaheimo, H., Renkonen, R., Turunen, J.P., Penttila, L., and Renkonen, O. (1995) Synthesis of a divalent sialyl Lewis x O-glycan, a potent inhibitor of lymphocyte-endothelium adhesion. Evidence that multivalency enhances the saccharide binding to L-selectin. Eur. J. Biochem., 234, 616–625.
Malykh, Y.N., King, T.P., Logan, E., Kelly, D., Schauer, R., and Shaw, L. (2003) Regulation of N-glycolylneuraminic acid biosynthesis in developing pig small intestine. Biochem. J., 370, 601–607.
Malykh, Y.N., Krisch, B., Shaw, L., Warner, T.G., Sinicropi, D., Smith, R., Chang, J., and Schauer, R. (2001) Distribution and localization of CMP-N-acetylneuraminic acid hydroxylase and N-glycolylneuraminic acid-containing glycoconjugates in porcine lymph node and peripheral blood lymphocytes. Eur. J. Cell Biol., 80, 48–58.
Mammen, M., Choi, S., and Whitesides, G. (1998) Polyvalent interactions in biological systems:implications for design and use of multivalent ligands and inhibitors. Angew. Chem. Int. Ed., 37, 2754–2794.
McEver, R.P., Moore, K.L., and Cummings, R.D. (1995) Leukocyte trafficking mediated by selectin-carbohydrate interactions. J. Biol. Chem., 270, 11025–11028.
Mitoma, J., Petryniak, B., Hiraoka, N., Yeh, J.C., Lowe, J.B., and Fukuda, M. (2003) Extended core 1 and core 2 branched O-glycans differentially modulate sialyl Lewis X-type L-selectin ligand activity. J. Biol. Chem., 278, 9953–9961.
Miyata, Y. and Platt, J.L. (2003) Xeno-still stuck without alphaGal. Nat. Biotechnol., 21, 359–360.
Moore, K.L., Eaton, S.F., Lyons, D.E., Lichenstein, H.S., Cummings, R.D., and McEver, R.P. (1994) The P-selectin glycoprotein ligand from human neutrophils displays sialylated, fucosylated, O-linked poly-N-acetyllactosamine. J. Biol. Chem., 269, 23318–23327.
Neethling, F.A., Koren, E., Oriol, R., Richards, S.V., Ye, Y., Kujundzic, M., and Cooper, D.K. (1994) Immunoadsorption of natural antibodies from human serum by affinity chromatography using specific carbohydrates protects pig cells from cytotoxic destruction. Transplant. Proc., 26, 1378.[Medline]
Oriol, R., Ye, Y., Koren, E., and Cooper, D.K. (1993) Carbohydrate antigens of pig tissues reacting with human natural antibodies as potential targets for hyperacute vascular rejection in pig-to-man organ xenotransplantation. Transplantation, 56, 1433–1442.
Pascher, A., Poehlein, C., Stangl, M., Thiery, J., Mueller-Derlich, J., and Hammer, C. (1997) Immunoapheresis, an advanced technique for depleting human anti-porcine antibodies, delays hyperacute rejection of xenogeneic perfused pig livers. Transplant. Proc., 29, 962–963.
Phelps, C.J., Koike, C., Vaught, T.D., Boone, J., Wells, K.D., Chen, S.H., Ball, S., Specht, S.M., Polejaeva, I.A., Monahan, J.A., and others. (2003) Production of alpha 1,3-galactosyltransferase-deficient pigs. Science, 299, 411–414.
Renkonen, O., Toppila, S., Penttila, L., Salminen, H., Helin, J., Maaheimo, H., Costello, C.E., Turunen, J.P., and Renkonen, R. (1997) Synthesis of a new nanomolar saccharide inhibitor of lymphocyte adhesion: different polylactosamine backbones present multiple sialyl Lewis x determinants to L-selectin in high-affinity mode. Glycobiology, 7, 453–461.
Rydberg, L., Holgersson, J., Samuelsson, B.E., and Breimer, M.E. (1999) alpha-Gal epitopes in animal tissue glycoproteins and glycolipids. Subcell. Biochem., 32, 107–125.
Sako, D., Chang, X.J., Barone, K.M., Vachino, G., White, H.M., Shaw, G., Veldman, G.M., Bean, K.M., Ahern, T.J., Furie, B., and others. (1993) Expression cloning of a functional glycoprotein ligand for P-selectin. Cell, 75, 1179–1186.
Sako, D., Comess, K.M., Barone, K.M., Camphausen, R.T., Cumming, D.A., and Shaw, G.D. (1995) A sulfated peptide segment at the amino terminus of PSGL-1 is critical for P-selectin binding. Cell, 83, 323–331.
Schachter, H. (2000) The joys of HexNAc. The synthesis and function of N- and O-glycan branches. Glycoconj. J., 17, 465–483.
Sharma, A., Naziruddin, B., Cui, C., Martin, M.J., Xu, H., Wan, H., Lei, Y., Harrison, C., Yin, J., Okabe, J., and others. (2003) Pig cells that lack the gene for alpha1-3 galactosyltransferase express low levels of the gal antigen. Transplantation, 75, 430–436.
Sheeley, D.M., Merrill, B.M., and Taylor, L.C. (1997) Characterization of monoclonal antibody glycosylation: comparison of expression systems and identification of terminal alpha-linked galactose. Anal. Biochem., 247, 102–110.
Silverman, H.S., Sutton-Smith, M., McDermott, K., Heal, P., Leir, S.H., Morris, H.R., Hollingsworth, M.A., Dell, A., and Harris, A. (2003) The contribution of tandem repeat number to the O-glycosylation of mucins. Glycobiology, 13, 265–277.
Skrincosky, D., Kain, R., El-Battari, A., Exner, M., Kerjaschki, D., and Fukuda, M. (1997) Altered Golgi localization of core 2 beta-1, 6-N-acetylglucosaminyltransferase leads to decreased synthesis of branched O-glycans. J. Biol. Chem., 272, 22695–22702.
Ten Hagen, K.G., Fritz, T.A., and Tabak, L.A. (2003) All in the family: the UDP-GalNAc:polypeptide N-acetylgalactosaminyltransferases. Glycobiology, 13, 1–16.
Totani, K., Kubota, T., Kuroda, T., Murata, T., Hidari, K.I., Suzuki, T., Suzuki, Y., Kobayashi, K., Ashida, H., Yamamoto, K., and Usui, T. (2003) Chemoenzymatic synthesis and application of glycopolymers containing multivalent sialyloligosaccharides with a poly(L-glutamic acid) backbone for inhibition of infection by influenza viruses. Glycobiology, 13, 315–326.
Towbin, H., Staehelin, T., and Gordon, J. (1979) Electrophoretic transfer of proteins from polyacrylamide gels to nitrocellulose sheets: procedure and some applications. Proc. Natl Acad. Sci. USA, 76, 4350–4354.
Van den Nieuwenhof, I.M., Koistinen, H., Easton, R.L., Koistinen, R., Kamarainen, M., Morris, H.R., Van Die, I., Seppala, M., Dell, A., and Van den Eijnden, D.H. (2000) Recombinant glycodelin carrying the same type of glycan structures as contraceptive glycodelin-A can be produced in human kidney 293 cells but not in chinese hamster ovary cells. Eur. J. Biochem., 267, 4753–4762.
Varki, A. (2001) N-glycolylneuraminic acid deficiency in humans. Biochimie, 83, 615–622.[Medline]
Yan, S.B., Chao, Y.B., and van Halbeek, H. (1993) Novel Asn-linked oligosaccharides terminating in GalNAc beta (1®4)[Fuc alpha (1®3)]GlcNAc beta (1®.) are present in recombinant human protein C expressed in human kidney 293 cells. Glycobiology, 3, 597–608.
Yeh, J.C., Ong, E., and Fukuda, M. (1999) Molecular cloning and expression of a novel beta-1, 6-N-acetylglucosaminyltransferase that forms core 2, core 4, and I branches. J. Biol. Chem., 274, 3215–3221.
Zhu, A. and Hurst, R. (2002) Anti-N-glycolylneuraminic acid antibodies identified in healthy human serum. Xenotransplantation, 9, 376–381.
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  J. Lofling, M. Diswall, S. Eriksson, T. Boren, M. E Breimer, and J. Holgersson Studies of Lewis antigens and H. pylori adhesion in CHO cell lines engineered to express Lewis b determinants Glycobiology, July 1, 2008; 18(7): 494 - 501. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M.-O. Sadoulet, C. Franceschi, M. Aubert, F. Silvy, J.-P. Bernard, D. Lombardo, and E. Mas Glycoengineering of {alpha}Gal xenoantigen on recombinant peptide bearing the J28 pancreatic oncofetal glycotope Glycobiology, June 1, 2007; 17(6): 620 - 630. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. Holgersson and J. Lofling Glycosyltransferases involved in type 1 chain and Lewis antigen biosynthesis exhibit glycan and core chain specificity Glycobiology, July 1, 2006; 16(7): 584 - 593. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||