Glycobiology, 2000, Vol. 10, No. 5 477-486
© 2000 Oxford University Press
Species-specific variation in glycosylation of IgG: evidence for the species-specific sialylation and branch-specific galactosylation and importance for engineering recombinant glycoprotein therapeutics
Analytical Chemistry, Genentech Inc., 1 DNA Way, South San Francisco, CA 94080, USA
Received on August 12, 1999; revised on November 7, 1999; accepted on December 8, 1999.
 |
Abstract |
|---|
 Top Abstract IntroductionResultsDiscussionMaterials and methodsAcknowledgmentsAbbreviationsReferences |
|---|
Immunoglobulins (IgG) are soluble serum glycoproteins in which the oligosaccharides play significant roles in the bioactivity and pharmacokinetics. Recombinant immunoglobulins (rIgG) produced in different host cells by recombinant DNA technology are becoming major therapeutic agents to treat life threatening diseases such as cancer. Since glycosylation is cell type specific, rIgGs produced in different host cells contain different patterns of oligosaccharides which could affect the biological functions. In order to determine the extent of this variation N-linked oligosaccharide structures present in the IgGs of different animal species were characterized. IgGs of human, rhesus, dog, cow, guinea pig, sheep, goat, horse, rat, mouse, rabbit, cat, and chicken were treated with peptide-N-glycosidase-F (PNGase F) and the oligosaccharides analyzed by matrix-assisted laser desorption/ionization time-of-flight mass spectrometry (MALDI-TOF-MS) for neutral and acidic oligosaccharides, in positive and negative ion modes, respectively. The data show that for neutral oligosaccharides, the proportions of terminal Gal, core Fuc and/or bisecting GlcNAc containing oligosaccharides vary from species to species; for sialylated oligosaccharides in the negative mode MALDI-TOF-MS show that human and chicken IgG contain oligosaccharides with N-acetylneuraminic acid (NANA), whereas rhesus, cow, sheep, goat, horse, and mouse IgGs contain oligosaccharides with N-glycolylneuraminic acid (NGNA). In contrast, IgGs from dog, guinea pig, rat, and rabbit contain both NANA and NGNA. Further, the PNGase F released oligosaccharides were derivatized with 9-aminopyrene 1,4,6-trisulfonic acid (APTS) and analyzed by capillary electrophoresis with laser induced fluorescence detection (CE-LIF). The CE-LIF results indicate that the proportion of the two isomers of monogalactosylated, biantennary, complex oligosaccharides vary significantly, suggesting that the branch specificity of ß1,4-galactosyltransferase might be different in different species. These results show that the glycosylation of IgGs is species-specific, and reveal the necessity for appropriate cell line selection to express rIgGs for human therapy. The results of this study are useful for people working in the transgenic area.
Key words: immunoglobulins/glycoproteins/oligosaccharides/carbohydrates/sialic acid/glycosyltransferases/mass spectrometry/capillary electrophoresis/laser induced fluorescence detection
|
Introduction |
|---|
|
TopAbstractIntroductionResultsDiscussionMaterials and methodsAcknowledgmentsAbbreviationsReferences |
|---|
Immunoglobulins (IgG) are soluble serum glycoproteins involved in the humoral immune response, binding to antigens to inactivate them or triggering an inflammatory response which results in their clearance. On the basis of non-cross-reacting antigenic determinants in the regions of highly conserved amino acid sequences in the constant domains of the heavy chains, five classes of immunoglobulins (IgG, IgA, IgD, IgE, and IgM) have been distinguished. IgG contains a conserved N-glycosylation site in the CH2 domain of each heavy chain of the Fc region (Sutton and Phillips, 1983
). The oligosaccharides present at this site are highly heterogeneous, and are known to affect the biological, pharmacological and physicochemical properties of IgGs (Mizuochi et al., 1982; Fujii et al., 1990). The biological functions affected by the oligosaccharides include resistance to proteases, binding to monocyte Fc receptors, interaction with complement component C1q, feedback immunosupression of IgG synthesis, and circulatory half-life in in vivo (Leatherbarrow et al., 1985; Tao and Morrison, 1989; Walker et al., 1989; Leader et al., 1991). Changes in N-glycosylation of IgG are associated with the status in diseases such as rheumatoid arthritis (Parekh et al., 1985, 1988; Rademachar et al., 1994; Malhotra et al., 1995).
The heterogeneous array of oligosaccharides on nascent proteins is synthesized in the Golgi compartments, where glycosyltransferases are localized (Kornfeld and Kornfeld, 1985). These glycosyltransferases are developmentally regulated and differentially expressed (Schachter, 1986; Stanley et al., 1996). Also, the expression of glycosyltransferases and protein glycosylation is cell type specific and varies with cell culture conditions (Stanley, 1984; Patel et al., 1992; Kumpel et al., 1994; Wright and Morrison, 1998). For example, normal CHO cells do not express N-acetylglucosaminyltransferase-III (GlcNAcT-III), the enzyme responsible for the biosynthesis of bisecting GlcNAc containing oligosaccharides. However, a mutant cell line isolated by the mutagenesis of parent CHO cells has been shown to express GlcNAcT-III (Campbell and Stanley, 1984). Oligosaccharides containing the bisecting GlcNAc are found in human and chicken IgG. Hamako et al. (1993) analyzed the asparagine-linked oligosaccharides of IgGs from 11 different animal species and demonstrated that the proportion of galactosylated, bisecting GlcNAc containing and/or core fucosylated oligosaccharides vary among species. These authors used hydrazinolysis to release oligosaccharides from the IgGs. Hydrazinolysis causes sialic acids (NGNA and NANA) and amino sugars to undergo N-deacetylation (in the case of NGNA, N-deglycolylation), which results in a loss of information regarding sialylated oligosaccharides (Patel and Parekh, 1994). Further, the experiments of Hamako et al. (1993) did not provide information on the diversity of branching pattern of monogalactosylated oligosaccharides in IgGs although significant amounts of these oligosaccharides were observed in IgGs preparations.
Recombinant IgGs (rIgGs), produced by recombinant DNA technology and/or by transgenic technology, are becoming major therapeutic agents in the treatment of cancer and other life threatening diseases. Recently, the FDA (United States Food and Drug Administration) approved two monoclonal antibodies, a chimeric monoclonal antibody for treating non-Hodgkin’s lymphoma and a humanized monoclonal antibody for treating metastatic breast cancer. These monoclonal antibodies are produced in Chinese hamster ovary cells (CHO). Different biopharmaceutical companies produce rIgGs in different host cell lines in which the glycosylation machinery might be different. In order to understand these issues and also to obtain information on the nature of sialylated oligosaccharides and the branching pattern of monogalactosylated oligosaccharides, we undertook a detailed structural study of N-linked oligosaccharides present in the IgGs of 13 different animal species. The N-linked oligosaccharides of IgGs from different animal species were released by PNGase F and analyzed by MALDI-TOF-MS in the positive ion mode and negative ion mode, and by CE-LIF after derivatizing with APTS. In this paper, we show that IgG glycosylation, particularly terminal sialylation and galactosylation of IgGs, is species-specific suggesting that a careful selection of host cell lines to produce rIgGs is necessary to avoid potentially immunogenic carbohydrate epitopes.
|
Results |
|---|
|
TopAbstractIntroductionResultsDiscussionMaterials and methodsAcknowledgmentsAbbreviationsReferences |
|---|
Initial estimation of neutral sugar content by phenol-sulfuric acid method (Dubois et al., 1956
) showed that the values of neutral sugars/protein ranged from ~13 µg/mg for rabbit, goat and cat IgGs to 32 µg/mg for chicken IgG, with human, cow, sheep, rhesus, rat, mouse, horse, guinea pig, and dog IgGs being intermediate (Table I). Determination of NANA and NGNA content by an RP-HPLC method (Anumula, 1995) showed that human and chicken IgG contain only NANA, whereas IgGs from rhesus, cow, sheep, goat, horse, and mouse contain only NGNA. However the IgGs from dog, guinea pig, rat, and rabbit contain both NANA and NGNA (see Table I), showing that the distribution of sialic acid residues is different in IgGs of different animal species.
|
Variations in neutral oligosaccharides
The N-linked oligosaccharides of IgGs were released by PNGase F and analyzed by MALDI-TOF-MS in the positive ion mode using a DHB matrix containing NaCl (Papac et al., 1998
). Consequently, quasimolecular ions of the oligosaccharides (M + Na)+, were observed. The (M + Na)+ ions of the neutral oligosaccharides found in the IgGs and their composition are provided in Table II. The proposed oligosaccharide structures for the respective (M + Na)+ ions are shown in Schemes 1 and 2.
|
Analysis of human IgG derived N-linked oligosaccharides by MALDI-TOF-MS in the positive ion mode afforded (M + Na)+ ions at m/z 1486.3, 1648.4, 1810.5, 1689.4, 1851.5, and 2013.6 (Table III, Figure 1). Since these oligosaccharides were released by PNGase F, the ions must be due to N-linked oligosaccharides (Patel and Parekh, 1994
). N-Linked oligosaccharides contain a common core region consisting of three Man and two GlcNAc residues. The core region of a complex N-linked oligosaccharide often contains a Fuc (dHex) residue linked to the reducing terminal GlcNAc residue of the free N-linked oligosaccharides. Therefore, the additional Hex and HexNAc residues of each human IgG oligosaccharide beyond Hex3HexNAc2dHex1 sequence found in the fucosylated core region must be in the oligosaccharide antennae. For example, in a glycan with composition Hex3HexNAc4dHex1 (m/z 1486.3) all but two HexNAc residues can be attributed to residues found in the fucosylated core sequence. Hence, the two HexNAc residues would reside in the antennae of an N-linked oligosaccharide. Since previous studies suggest that HexNAc residues found in human IgG are GlcNAc residues, the Hex3HexNAc4dHex1 glycan was assigned to structure 9 (Scheme 2). Similarly, the ions at m/z 1648.4 and 1810.5 have masses consistent with the presence of additional 1 and 2 hexose residues, respectively, beyond those observed in the oligosaccharide assigned to structure 9. Since biantennary N-linked oligosaccharides contain 1 or 2 Gal residues, the ions at m/z 1648.4 and 1810.5 were assigned to structures 19 and/or 20 (structures 19 and 20 are the branch isomers of monogalactosylated biantennary structures), and 21 respectively. The (M + Na)+ ion at m/z 1689.4 contains one additional HexNAc residue compared to the (M + Na)+ ion at m/z 1486.3. This ion might be due to a triantennary structure consisting of three terminal GlcNAc residues or due to a biantennary oligosaccharide containing a bisecting GlcNAc residue. Mizuochi et al. (1982) characterized human IgG derived oligosaccharides and found no evidence for tri- and tetraantennary structures. Further, these authors reported the presence of a bisecting GlcNAc residue in some of the oligosaccharides in human IgG (Mizuochi et al., 1982). Furthermore, if human IgG contained any triantennary structures, we expect to have observed a (M + Na)+ ion at m/z 2175.4 due to 3 Gal residues. The positive ion mode MALDI-TOF-MS analysis of human IgG derived oligosaccharides showed no evidence for such structures. Hence, the (M + Na)+ ion at m/z 1689.4 was assigned to the structure 32 (Scheme 2). Similarly, the (M + Na)+ ion at m/z 1851.5 was assigned to structures 33 and/or 34, and the ion at m/z 2013.6 was assigned to structure 35.
|
|
The positive ion mode MALDI-TOF-MS analysis of the PNGase F released oligosaccharides from chicken IgG gave (M + Na)+ ions at m/z from 1258.1 to 2013.6 (Figure 1, Table III). The ion at m/z 1258.1 corresponds to the glycosyl composition of Hex5HexNAc2. From this ion, 3Hex and 2HexNAc residues could be accounted for the core region of an N-linked oligosaccharide. The remaining two Hex residues could not be accounted for by Gal residues because there are no outer core GlcNAc residues to which Gal residues could be linked. A high mannose type N-linked oligosaccharide containing 5 Man and 2 GlcNAc residues could account for this ion (ion at m/z 1258.1). Hence the (M + Na)+ ion at m/z 1258.1 was assigned to structure 1 (see Scheme 1). Similarly the (M + Na)+ ions at m/z 1420.2, 1582.3, 1744.4, 1906.4, and 2068.5 were assigned to the structures 2A-C, 3A-C, 4A-C, 5 and 6, respectively. In structures 2A-C, 3A-C, and 4A-C, the symbols A-C are assigned to the positional isomers which are not distinguishable by MS because they produce isobaric ions. Structure 6 was assigned to the (M + Na)+ ion at m/z 2068.5 because, in the biosynthetic pathway of N-linked oligosaccharides, a high mannose type N-linked oligosaccharides can contain maximum 9 Man residues. The additional 1 Hex residue is interpreted as a Glc residue. According to the biosynthesis of N-linked oligosaccharides, dolichol linked Glc3Man9GlcNAc2 moiety is transferred en bloc to Asn by oligosaccharyltransferase (Kornfeld and Kornfeld, 1985
). After the initial oligosaccharide transfer reaction, trimming by glucosidases and mannosidases takes place. During this trimming reaction, various intermediates of Glc3Man9GlcNAc2 like Glc2Man9GlcNAc2, Glc1Man9GlcNAc2, Man9GlcNAc2 etc. are produced. Further, Ohta et al. (1991) reported the presence of a monoglucosylated Man9GlcNAc2 structure in chicken IgG along with an appreciable amount of high mannose type oligosaccharides. These observations support the structural assignments to (M + Na)+ ions at m/z 1258.1, 1420.2, 1582.3, 1744.4, 1906.4, and 2068.5. The identification of structures 1–6 suggests that the processing of N-linked oligosaccharides in chicken IgG is incomplete. The other ions observed from the positive ion mode MALDI-TOF-MS analysis of chicken IgG derived N-linked oligosaccharides and their structural assignments are shown in Table III and Figure 1, which suggest that, in addition to high mannose type structures, complex biantennary structures with or without core Fuc, terminal Gal and bisecting GlcNAc residues are also present. These data indicate that the N-linked oligosaccharides of chicken IgG are more heterogeneous than the N-linked oligosaccharides of human IgG. The N-linked oligosaccharides from other IgGs were also released by PNGase F and analyzed by MALDI-TOF-MS in the positive ion mode. The observed (M + Na)+ ions and their structural assignments are shown in Table III (see also Figure 1). It is evident that these IgGs also contain complex biantennary structures with or without core Fuc residues and that the proportion of bisecting GlcNAc containing oligosaccharides varies among species. For example, ~67% of sheep IgG oligosaccharides contain a bisecting GlcNAc residue compared to ~53% of chicken IgG oligosaccharides. However, the N-linked oligosaccharides of dog, horse and cat IgGs contain no detectable bisecting GlcNAc residue. All other IgGs contain an appreciable proportion of oligosaccharides with terminal GlcNAc residues which vary from 2% to 40% (see Table III). Similarly, the variation in the galactosylated oligosaccharides is also evident (Table III, Figure 1). Rat, horse, and dog IgGs contain very few galactosylated oligosaccharides whereas ~90% of the sheep IgG oligosaccharides are galactosylated.
Variations in acidic oligosaccharides
The PNGase F released oligosaccharides of different IgGs were also analyzed by MALDI-TOF-MS in the negative ion mode using THAP as matrix (Papac et al., 1996). The data are shown in Table IV and Figure 2. The oligosaccharides released from human IgG gave molecular ions, (M-H)– ions, at m/z 1915.7, 2077.8, 2369.1, 2281.0, and 2572.3 (see Figure 2). The glycosyl composition of these ions and the corresponding structural assignments are shown in Tables II and IV. Only NANA-containing oligosaccharides were detected in human IgG. The (M-H)– ions corresponding to oligosaccharides with NGNA (i.e., at m/z values of +16 amu compared to their NANA counterparts) were not detected in human IgG.
|
|
The (M-H)– ions observed for oligosaccharides of other IgGs and their structural assignments are shown in Table IV. Like human IgG, chicken IgG also contained oligosaccharides with only NANA. However, mouse, horse, sheep, rhesus, cow, and goat IgG-derived oligosaccharides contain only NGNA. In contrast, the oligosaccharides derived from rat, guinea pig, rabbit, cat, and dog contain both NANA and NGNA but in different proportions (see Table IV). These data show that sialylation of IgG oligosaccharides varies from species to species. The identification and quantitation of NANA- and NGNA-containing oligosaccharides of these IgGs by MALDI-TOF-MS is in good agreement with the data obtained by RP-HPLC method (Table I). The data in Tables I and IV also suggest that the amount of sialic acids present in IgGs is significantly less than the amount of neutral sugars indicating that the sialylated oligosaccharides are not the major species.
Evidence for branch-specific galactosylation
The oligosaccharides released by PNGase F were labeled with 9-aminopyrene-1,4,6-trisulfonic acid by reductive amination and analyzed by CE-LIF. Figure 3A shows the electropherogram of human IgG oligosaccharides. In Figure 3A, peaks I and IV were identified by comparing the migration time of standard oligosaccharides and assigned to structures 18 and 21, respectively (data not shown). Peaks II, III, and IV migrated at the position of peak I after ß-galactosidase digestion (see Figure 3B), and hence the former peaks are galactosylated biantennary structures with core Fuc residue. Since peak IV was identified as Structure 21 which contains 2 Gal residues, peaks II and III must be the two isomers of monogalactosylated biantennary structures (structures 19 and 20). Jefferis et al. (1990) reported that human IgG contains two monogalactosylated core fucosylated biantennary structures in which structure 19 is the predominant species. Based on these observations, peaks II and III were assigned to structures 19 and 20, respectively. This assignment was independently confirmed by in vitro galactosylation of structure 18 with ß1,4-galactosyltransferase (ß1,4GT) and UDP-Gal (Raju et al., unpublished observations) which was in good agreement with the results described by Paquet et al. (1984).
|
The electropherogram of APTS-labeled oligosaccharides derived from cow IgG is shown in Figure 3C in which the proportion of peaks II and III is different compared to the proportion observed for human IgG (see Table V). In this case also, peaks II, III, and IV migrated to the position of peak I after ß-galactosidase treatment (Figure 3D). The data for peaks II and III (structures 19 and 20, respectively), obtained from CE-LIF data for other IgGs is summarized in Table V. The data in Table V show that the proportions of peaks II and III vary, showing that branch-specific galactosylation of IgGs varies from species to species.
|
Discussion |
|---|
|
TopAbstractIntroductionResultsDiscussionMaterials and methodsAcknowledgmentsAbbreviationsReferences |
|---|
Serum IgGs are soluble glycoproteins in which the covalently bound oligosaccharide moieties present in the Fc domain play significant roles in their bioactivity and serum half life. Although some glycosylation can occur outside the Fc domain, this is usually a rare event for IgG (Parekh et al., 1985
). The N-linked oligosaccharides of human IgG have been extensively characterized. The Fc derived oligosaccharides of human IgG are mainly complex biantennary type with heterogeneity in core fucosylation, terminal sialylation, and galactosylation. Heterogeneity of these oligosaccharides also arises due to the presence or absence of a bisecting GlcNAc (Mizuochi et al., 1982). However, the structures of oligosaccharides of IgGs from other animal species have not been studied thoroughly. Because of their biological importance, and to further our understanding of the structure-function relationships, we determined the structures of N-glycans of IgGs from 13 different animal species. The results of these studies will enhance our understanding of recombinant glycoproteins of therapeutic interest. The IgGs studied are representative of respective animal species. The IgGs were affinity-purified before analysis and the oligosaccharides released by PNGase F from the intact molecules were analyzed. The majority of structural assignments were based on masses (determined by mass spectrometry) which provide only general monosaccharide compositions of hexose, N-acetylhexosamines, sialic acids, etc. (see, for example, Table II). This information, together with the knowledge base of the structures which have been previously determined (Mizuochi et al., 1982; Nose and Wigzell, 1983; Leatherbarrow et al., 1985; Fujii et al., 1990), provided the basis for structural assignments in this work. In some cases, additional information was derived from the use of specific enzymes such as glycosidases and glycosyltransferases, and CE-LIF analysis.
The MALDI-TOF-MS analysis of PNGase F released oligosaccharides show that IgGs from 13 different animal species contain a heterogeneous array of biantennary complex type oligosaccharides. However, there seems to be species-specific variation in core fucosylation and terminal galactosylation (see Figure 4A,B). For example, the oligosaccharides derived from human IgG contain mostly core fucosylated oligosaccharides, whereas the oligosaccharides derived from rabbit IgG contain mostly nonfucosylated oligosaccharides (Figure 4A). Similarly, about 90% of sheep IgG oligosaccharides are galactosylated, whereas only ~10% of rat IgG oligosaccharides contain galactose residues (Figure 4B). The MALDI-TOF-MS analyses also show that IgGs from human, rhesus, cow, guinea pig, sheep, goat, rat, rabbit, and chicken contain biantennary oligosaccharides with one additional HexNAc residue (see Tables II and III). Based on the observations made by Mizuochi et al. (1982) on human IgG derived oligosaccharides, these were considered as bisecting GlcNAc containing oligosaccharides. The oligosaccharides from dog, horse, and mouse did not contain detectable bisecting GlcNAc residue (Figure 4C). These observations confirm the species-specific variation in core fucosylation, terminal galactosylation and the presence of a bisecting GlcNAc residue in IgGs from different animal species. The presence of oligosaccharides containing bisecting GlcNAc residues in human, rhesus, cow, guinea pig, sheep, goat, rat, rabbit, and chicken IgGs suggests that these animal species express N-acetylglucosaminyltransferase-III, the enzyme responsible for the biosynthesis of bisecting GlcNAc containing oligosaccharides (Campbell and Stanley, 1984). The absence of hybrid and complex tri- and tetraantennary structures suggests that the glycosylation observed in this study is consistent with the restriction of Fc glycosylation to biantennary structures (Parekh et al., 1985). In all of the species studied here, with the exception of chicken (which additionally contains high-mannose structures), the predominant structures are also biantennary complex oligosaccharides. The data presented here therefore probably describe only the glycosylation site of the Fc domains.
|
Yamada et al. (1997)
reported that the galactosylation of human IgG varies with age and is gender specific. Parekh et al. (1988) showed that the terminal galactosylation of human IgG is an indication of disease status in rheumatoid arthritis patients. Further, Patel et al. (1992) observed that the galactosylation of IgG varies with cell culture conditions. The CE-LIF data suggest that the galactosylation of IgGs is also branch- specific (Table V, Figure 3). The MALDI-TOF-MS data (Table III) suggest that IgGs from different animal species contain significant amount of monogalactosylated biantennary structures (for example structures 19 and 20). Using MS techniques it was not possible to distinguish structures 19 and 20. Hence, we used CE-LIF to determine the proportion of structures 19 and 20 in IgGs. The data in Figure 3 and Table V show that the proportion of structures 19 and 20 varies from species to species. In human and rhesus IgGs the proportion of structures 19 and 20 is very similar. However, the proportions observed in cow and mouse IgGs is opposite to each other. These data (Table V ) suggest that the tendency for one branch to be galactosylated over another branch is species-specific. The branch and species-specific galactosylation of IgGs implies species-specific variation in the activity of ß1,4GT. The enzyme, ß1,4GT is a constitutively expressed, trans-Golgi resident, type II membrane-bound glycoprotein that catalyzes the transfer of Gal to GlcNAc from UDP-Gal (Beyer and Hill, 1968). ß1,4GT enzymatic activity is widely distributed in the vertebrate kingdom, in both mammals and nonmammals, including avians and amphibians (Shaper et al., 1997). Recently, Lo et al. (Lo et al., 1998) reported the presence of a family of ß1,4GT genes. Our in vitro galactosylation of agalactosylated biantennary complex oligosaccharide with commercially available human and cow ß1,4GT suggest that the enzyme preferentially adds Gal to GlcNAc linked to 
1,3-Man arm (Raju et al., unpublished observations). These observations also suggest that the activity of ß1,4GT is species- and branch-specific.
Sialic acid determinations indicate that the IgGs from different animal species contain an appreciable amount of sialylated oligosaccharides (see Table I). Data in Tables I and IV and Figure 2 provide additional information on the variability of NANA and NGNA content from species to species and the structures in which they are found. For example, human and chicken IgG-derived acidic oligosaccharides contain exclusively NANA (Figure 4D), whereas sheep, goat, rhesus, cow, horse, and mouse derived acidic oligosaccharides contain only NGNA. Interestingly, the acidic oligosaccharides derived from dog, guinea pig, rat, rabbit, and cat IgGs contain both NANA and NGNA residues in different proportions, confirming significant variations in the sialylated oligosaccharides of IgGs. Muchmore et al. (Muchmore et al., 1998) reported that NGNA is essentially undetectable on human plasma proteins and erythrocytes, but is a major component in chimpanzee, bonobo, gorilla, and orangutan. In our study also, NGNA is essentially undetectable in human IgG (see Table I and VI). Biosynthetically, NGNA arises from the action of a hydroxylase that converts the nucleotide donor sugar, CMP-NANA to CMP-NGNA (Chou et al., 1998). The activity of CMP-NANA hydroxylase is reported to be present in chimpanzee cells but not in human cells (Varki, 1992; Chou et al., 1998; Muchmore et al., 1998). As major terminal structures on cell surfaces, sialic acids are involved in intercellular cross-talk and microbe–host recognition. The level of NGNA is known to positively or negatively affect several of these endogenous and exogenous interactions (Varki, 1993, 1994, 1997, 1998; Varki and Marth, 1995). Hence there are potential functional consequences of this structural change which affect the cell surface functions (Muchmore et al., 1998).
rIgGs are being developed by many biopharmaceutical companies as therapeutic agents to treat human diseases. Two rIgGs, produced in CHO cells were approved by regulatory agencies around the world, one to treat non-Hodgkin’s lymphoma and the other to treat metastatic breast cancer overexpressing HER-2 oncogene. Different biopharmaceutical firms use different cell lines and different cell culture conditions to produce the rIgGs. However, different cell lines have different glycosylation machinery. Our data on the glycosylation of IgGs from different species suggest that the glycosylation varies from species to species. This in turn suggests that the glycosylation of rIgGs produced in different cell lines might be significantly different. For example, monoclonal antibodies produced in mouse cell lines might contain NGNA residues which are potentially immunogenic to humans (Cho et al., 1996; Noguchi, 1995). The data obtained by the analysis of oligosaccharides of IgGs from 13 different animal species clearly suggests that a careful selection of cell line is a prerequisite to produce rIgGs for human therapy. The data is also helpful to understand the effect of glycosylation on protein therapeutics produced by transgenic technology. There is a growing interest to express protein therapeutics using transgenic animals such as goat, cow, sheep, etc. The data presented here clearly suggest that IgGs of goat, cow, and sheep have different glycosylation pattern which might influence their biological and pharmacological functions. Further, Cabanes-Macheteau et al. expressed a mouse IgG in transgenic tobacco plants and shown that they contain unusual carbohydrates which might be immunogenic to humans (Cabanes-Macheteau et al., 1999). Such studies clearly show the importance of our results on species-specific glycosylation in the areas of biotechnology, immunology, glycobiology, and others working on glycoprotein therapeutics.
|
Materials and methods |
|---|
|
TopAbstractIntroductionResultsDiscussionMaterials and methodsAcknowledgmentsAbbreviationsReferences |
|---|
Materials
Human, rhesus, dog, cow, guinea pig, sheep, goat, horse, rat, mouse, rabbit, cat, and chicken IgGs were obtained from Sigma Chemical Co. (St. Louis, MO) and purified on protein A/G columns before use. PNGase-F and human ß1,4-galactosyltransferase were obtained from Boehringer Mannheim (Germany) or from Oxford GlycoSciences (London, UK). The cow ß1,4-galactosyltransferase and polyvinylpyrrolidone were from Sigma Chemical Co. Immobilized Protein A and Protein G cartridges were obtained from Pharmacia and used as per the manufacturers protocol. The matrices 2',4',6'-trihydroxyacetophenone monohydrate and 2,5-Dihydroxybenzoic acid were purchased from Aldrich (Milwaukee, WI). 9-Aminopyrene-1,4,6-trisulfonic acid was purchased from Beckman (Jackson, 1990).
Analytical methods
Neutral hexoses were quantitated by phenol-sulfuric acid assay (Dubois et al., 1956), and the sialic acids were measured by RP-HPLC (Anumula, 1995).
Release of N-linked oligosaccharides by PNGase-F
The N-linked oligosaccharides from IgGs (at least two different batches of IgGs from each species were used in the study) released by PNGase-F using a high-throughput microscale method as described by Papac et al. (1998). The PVDF membrane wells of a MultiScreen-IP plate (pore size 0.45 µm, Millipore) were preconditioned by washing with 1 x 100 µl methanol, 3 x 100 µl water and 1 x 50 µl reduction and carboxymethylation (RCM) buffer (8 M urea containing 360 mM Tris, pH 8.6, and 3.2 mM EDTA). About 25–40 µg of glycoprotein was loaded into wells containing 10 µl RCM buffer and the solution was brought to 50 µl by adding additional RCM buffer. Reduction of protein was performed in the presence of 50 µl of 0.1 M dithiothreitol in RCM buffer for 1 h at 37°C followed by washing with water (3 x 300 µl). Carboxymethylation was accomplished by adding 50 µl of 0.1 M iodoacetic acid in RCM buffer and incubating at room temperature for 30 min in the dark. Following carboxymethylation, the wells were washed with water (3 x 300 µl) and the membranes were blocked with 1% aqueous polyvinylpyrrolidone 360 (100 µl) at room temperature for 60 min. The wells were washed with water (3 x 300 µl) to remove blocking agent and incubated with 1.25 U of PNGase-F (Oxford GlycoSciences) in 50 µl of 10 mM Tris-acetate buffer (pH 8.3) at 37°C for 3 h in plates covered with Parafilm to prevent evaporative loss of the digestion buffer. The solution containing released oligosaccharide, buffer and enzyme was transferred to an Eppendorf tube and treated with 150 mM acetic acid for 3 h at room temperature. The solution containing the released oligosaccharides was passed through a 0.6 ml of cation-exchange resin (AG50W-X8 resin, H+ form, 100–200 mesh, Bio-Rad, Hercules, CA) to remove salt and protein contaminants prior to analysis by mass spectrometry.
Matrix-assisted laser desorption/ionization time-of-flight mass spectrometry (MALDI-TOF-MS)
MALDI-TOF-MS was performed on a Voyager DE Biospectrometry Work Station (Perseptive Biosystems, Framingham, MA) equipped with delayed extraction. A nitrogen laser was used to irradiate samples with ultraviolet light (337 nm), and an average of 240 scans was taken. The instrument was operated in linear configuration (1.2 m flight path), and an acceleration voltage of 20 kV was used to propel ions down the flight tube after a 60 ns delay. Samples (0.5 µl) were applied to a polished stainless steel target to which 0.3 µl of matrix was added and dried under vacuum (50 x 10–3 Torr). Oligosaccharide standards were used to achieve a two-point external calibration for mass assignment of ions (Papac et al., 1996). 2,5-Dihydroxybenzoic acid (DHB) and 2,4,6-trihydroxyacetophenone (THAP) matrices were used in the analysis of neutral and acidic oligosaccharides, respectively (Papac et al., 1996, 1998).
Analysis of oligosaccharides by CE-LIF
The IgG samples (400–500 µg, at least two different batches of IgGs from each species were used in the study) were treated with PNGase-F (50 U/mg) in 20 mM phosphate buffer (pH 8.2, 200 µl) containing 50 mM EDTA and 0.02% (w/v) sodium azide at 37°C for 24 h. Released oligosaccharides were separated from protein and enzyme by ethanol precipitation and/or by heat denaturation. The supernatant was evaporated to dryness using a Savant Speed Vac. The samples were fluorescently labeled by adding 15 µl of a 19 mM solution of 9-aminopyrene-1,4,6-trisulfonic acid (APTS, Beckman) in 15% acetic acid, and 5 µl of 1 M sodium cyanoborohydride in tetrahydrofuran. The labeling reaction was carried out for two h at 55°C, followed by ~25-fold dilution with water prior to capillary electrophoretic (CE) analysis. CE analysis of the labeled oligosaccharides was performed on a P/ACE 5000 CE system (Beckman) with reversed polarity, using a 50-µm internal diameter coated capillary and 20 cm effective length (eCAP, N-CHO coated capillary, Beckman). The samples were introduced by pressure injection at 0.5 psi for 2–4 s, and the separation was carried out at a constant voltage of 20 kV. The temperature of the capillary was maintained at 20°C. The separations were monitored on-column with a Beckman laser-induced fluorescence detection system using a 3 mW argon ion laser with an excitation wavelength of 488 nm and emission bandpass filter at 520 ± 10 nm.
ß-Galactosidase digestion
Human and cow IgGs (~2 mg each) in 100 mM citrate-phosphate buffer, pH 6.4 were treated with ß-D-galactosidase (Diplococcus pneumoniae, 40 mU/ mg protein, Boehringer Mannheim) at 37°C for 24 h. The antibodies were purified using a HiTrap Protein A cartridge (Pharmacia) as described by the manufacturer. The modified antibodies were treated with PNGase F to release the oligosaccharides which were analyzed by CE-LIF as described above.
In vitro galactosylation with ß1,4-galactosyltransferase
Oligosaccharide samples (20 µg) in 50 mM sodium cacodylate buffer, pH 6.7 (in a final volume of 100 µl) were treated with UDP-Gal (5 µmol, Sigma Chemical Co.) and ß1,4-galactosyltransferase (25 mU, human milk or cow, Boehringer Mannheim or Sigma) at 37°C for 0–4 h. The reaction was stopped by adding ethanol (~1.0 ml). The samples were centrifuged and the centrifugate was dried using a Speed Vac. The oligosaccharides were labeled with APTS and analyzed by CE-LIF as described above.
|
Acknowledgments |
|---|
|
TopAbstractIntroductionResultsDiscussionMaterials and methodsAcknowledgmentsAbbreviationsReferences |
|---|
We thank Drs. John O’Connor and Pamela Stanley (Albert Einstein College of Medicine, New York) for helpful discussions. We greatly acknowledge Mr. Michael Wilks for sialic acids analysis by RP-HPLC.
|
Abbreviations |
|---|
|
TopAbstractIntroductionResultsDiscussionMaterials and methodsAcknowledgmentsAbbreviationsReferences |
|---|
CHO cells, Chinese hamster ovary cells; IgG, immunoglobulin G; rIgG, recombinant IgG; MALDI-TOF-MS, matrix-assisted laser/desorption ionization time-of-flight mass spectrometry; ESI-MS, electrospray ionization-mass spectrometry; RP-HPLC, reverse phase high-performance liquid chromatography; HPAEC-PAD, high-performance anion exchange chromatography with pulsed amperometric detection; CE-LIF, capillary electrophoresis with laser induced fluorescence detection; RCM, reduction and carboxymethylation; NANA, N-acetyl neuraminic acid; NGNA, N-glycolyl neuraminic acid: APTS, 9-aminopyrene 1,4,6-trisulfonic acid; PNGase F, Peptide-N-glycosidase-F; GlcNAc, N-acetyl D-glucosamine; Gal, D-galactose; Fuc, L-fucose; ß1,4GT, ß1,4-galactosyltransferase; GlcNAcT-III, N-acetylglucosaminyltransferase-III.
|
|
Footnotes |
|---|
1 To whom correspondence should be addressed
|
References |
|---|
|
TopAbstractIntroductionResultsDiscussionMaterials and methodsAcknowledgmentsAbbreviationsReferences |
|---|
Anumula,K.R. (1995) Rapid quantitative determination of sialic acids in glycoproteins by high-performance liquid chromatography with a sensitive fluorescence detection. Anal. Biochem., 230, 24–30.[ISI][Medline]
Beyer,T.A. and Hill,R.L. (1968) The Glycoconjugates, Vol. III. Academic Press, New York, pp. 25–45.
Cabanes-Macheteau,M., Fitchette-Laine,A.-C., Loutelier-Bourhis,C., Lange,C., Vine,N.D., Ma,J.K.C., Lerouge,P. and Faye,L. (1999) N-Glycosylation of a mouse IgG expressed in transgenic tobacco plants. Glycobiology, 9, 365–372.
Campbell,C. and Stanley,P. (1984) The Chinese hamster ovary glycosylation mutants LEC11 and LEC12 express two novel GDP-fucose:N-acetylglucosaminide 3--L-fucosyltransferase enzymes. J. Biol. Chem., 259, 13370–13378.
Cho,S.K., Yeh,J.-C., Cho,M. and Cummings,R.D. (1996) Transcriptional regulation of 1,3-galactosyltransferase in embryonal carcinoma cells by retinoic acid. Masking of Lewis x antigens by alpha-galactosylation. J. Biol. Chem., 271, 2338–3246.
Chou,H.-H., Takematsu,H., Diaz,S., Iber,J., Nickerson,E., Wright,K.L., Muchmore,E.A., Nelson,D.L., Warren,S.T. and Varki,A. (1998) A mutation in human CMP-sialic acid hydroxylase occurred after the Homo-Pan divergence. Proc. Natl. Acad. Sci. USA, 95, 11751–11756.
Dubois,M., Gills,K.A., Hamilton,J.K., Rebers,P.A. and Smith,F. (1956) Colorimetric method for determination of sugars and related substances. Anal. Chem., 28, 350–356.
Fujii,S., Nishiura,T., Nishikawa,A., Miura,R. and Taniguchi,N. (1990) Structural heterogeneity of sugar chains in immunoglobulin G. Conformation of immunoglobulin G molecule and substrate specificities of glycosyltransferases. J. Biol. Chem., 265, 6009–6018.
Hamako,J., Matsui,T., Ozeki,Y., Mizuochi,T. and Titani,K. (1993) Comparative studies of asparagine-linked sugar chains of immunoglobulin G from eleven mammalian species. Comp. Biochem. Physiol., 106B, 949–954.
Jackson,P. (1990) The use of polyacrylamide-gel electrophoresis for the high-resolution separation of reducing saccharides labelled with the fluorophore an imaging system based on a cooled charge-coupled device. Biochem. J., 270, 705–713.[ISI][Medline]
Jefferis,R., Lund,J., Mizutani,H., Nakagawa,H., Kawazoe,Y., Arata,Y., and Takahashi,N. (1990). A comparative study of the N-linked oligosaccharide structures of human IgG subclass proteins. Biochem. J., 268, 529–537.[ISI][Medline]
Kornfeld,R. and Kornfeld,S. (1985) Assembly of asparagine-linked oligosaccharides. Annu. Rev. Biochem., 54, 631–664.[ISI][Medline]
Kumpel,B.M., Rademacher,T.W., Rook,G.A.W., Williams,P.J. and Wilson,I.B.H. (1994) Galactosylation of human IgG monoclonal anti-D produced by EBV-transformed B-lymphoblastoid cell lines is dependent on culture method and affects Fc receptor-mediated functional activity. Hum. Antibod. Hybridomas, 5, 143–151.[Medline]
Leader,K.A., Kumpel,B.M., Hadley,A.G. and Bradley,B.A. (1991) Functional interactions of aglycosylated monoclonal anti-D with Fc gamma RI+ and Fc gamma RIII+ cells. Immunology, 72, 481–485.[ISI][Medline]
Leatherbarrow,R.J., Rademacher,T.W., Dwek,R.A., Woof,J.M., Clark,A., Burton,D.R., Richardson,N. and Feinstein,A. (1985) Effector functions of a monoclonal aglycosylated mouse IgG2a: binding and activation of complement component C1 and interaction with human monocyte Fc receptor. Mol. Immunol., 22, 407–415.[ISI][Medline]
Lo,N.-W., Shaper,J.H., Pevsner,J. and Shaper,N.L. (1998) The expanding ß 4-galactosyltransferase gene family: messages from the databanks. Glycobiology, 8, 517–526.
Malhotra,R., Wormald,M.R., Rudd,P.M., Fischer,P.B., Dwek,R.A. and Sim,R.B. (1995) Glycosylation changes of IgG associated with rheumatoid arthritis can activate complement via the mannose-binding protein. Nature Med., 1, 237–241.[ISI][Medline]
Mizuochi,T., Taniguchi,T., Shimizu,A. and Kobata,A. (1982) Structural and numerical variations of the carbohydrate moiety of immunoglobulin G. J. Immunol., 129, 2016–20.[Abstract]
Muchmore,E.A., Diaz,S. and Varki, A. (1998) A structural difference between the cell surfaces of humans and the great apes. Am. J. Phys. Anthropol., 107, 187–198.[ISI][Medline]
Noguchi,A., Mukuria,C.J., Suzuki,E. and Naiki,M. (1995) Immunogenicity of N-glycolylneuraminic acid-containing carbohydrate chains of recombinant human erythropoietin expressed in Chinese hamster ovary cells. J. Biochem., 117, 59–62.
Nose,M. and Wigzell,H. (1983) Biological significance of carbohydrate chains on monoclonal antibodies. Proc. Natl. Acad. Sci. USA, 80, 6632–6636.
Ohta,M., Hamako,J., Yamamoto,S., Hatta,H., Kim,M., Yamamoto,T., Oka,S., Mizuochi,T. and Matsuura,F. (1991) Structures of asparagine-linked oligosaccharides from hen egg-yolk antibody (IgY). Occurrence of unusual glucosylated oligo-mannose type oligosaccharides in a mature glycoprotein. Glycoconjugate J., 8, 400–413.[ISI][Medline]
Papac,D.I., Briggs,J.B., Chin,E.T. and Jones,A.J.S. (1998) A high-throughput microscale method to release N-linked oligosaccharides from glycoproteins for matrix-assisted laser desorption/ionization time-of-flight mass spectrometric analysis. Glycobiology, 8, 445–454.
Papac,D.I., Wong,A. and Jones,A.J.S. (1996) Analysis of acidic oligosaccharides and glycopeptides by matrix-assisted laser desorption/ionization time-of-flight mass spectrometry. Anal. Chem., 68, 3215–3223.[Medline]
Parekh,R.B., Dwek,R.A., Sutton,B.J., Fernandes,D.L., Leung, A, Stanworth,D., Rademacher,T.W., Mizuochi,T., Taniguchi,T., Matsuta,K., Takeuchi,R., Nagano,Y., Miyamoto,T. and Kobata,A. (1985) Association of rheumatoid arthritis and primary osteoarthritis with changes in the glycosylation pattern of total serum IgG. Nature, 316, 452–457.[Medline]
Parekh,R.B., Dwek,R.A. and Rademacher,T.W. (1988) Rheumatoid arthritis as a glycosylation disorder. Br. J. Rheumatol., 27 (Suppl. II), 162–169.
Patel,T.P. and Parekh,R.B. (1994) Release of oligosaccharides from glycoproteins by hydrazinolysis. Methods Enzymol., 230, 57–66.[ISI][Medline]
Patel,T.P., Parekh,R.B., Moellering,B.J. and Prior,C.P. (1992) Different culture methods lead to differences in glycosylation of a murine IgG monoclonal antibody. Biochem. J., 285, 839–845.
Rademacher,T.W., Williams,P. and Dwek,R.A. (1994) Agalactosyl glycoforms of IgG autoantibodies are pathogenic. Proc. Natl. Acad. Sci. USA, 91, 6123–6127.
Schachter,H. (1986) Biosynthetic controls that determine the branching and microheterogeneity of protein-bound oligosaccharides. Biochem. Cell Biol., 64, 163–181.[ISI][Medline]
Paquet,M.R., Narasimhan,S., Schachter,H. and Moscarello,M.A. (1984) Branch specificity of purified rat liver Golgi UDP-galactose: N-acetylglucosamine ß-1,4-galactosyltransferase. Preferential transfer of of galactose on the GlcNAc ß 1,2-Man 1,3-branch of a complex biantennary Asn-linked oligosaccharide. J. Biol. Chem., 259, 4716–4721.
Shaper,N.L., Meurer,J.A., Joziasse,J.H., Chou,T.-D.D., Smith,E.J., Schnaar,R.L. and Shaper,J.H. (1997) The chicken genome contains two functional nonallelic beta1,4-galactosyltransferase genes. Chromosomal assignment to syntenic regions tracks fate of the two gene lineages in the human genome. J. Biol. Chem., 272, 31389–31399.
Stanley,P. (1984) Glycosylation mutants of animal cells. Annu. Rev. Genet., 18, 525–552.[ISI][Medline]
Stanley,P., Raju,T.S. and Bhaumik,M. (1996) CHO cells provide access to novel N-glycans and developmentally regulated glycosyltransferases. Glycobiology, 6, 695–699.
Sutton,B.J. and Phillips,D.C. (1983) The three-dimensional structure of the carbohydrate within the Fc fragment of immunoglobulin G. Biochem. Soc. Trans., 11, 130–132.
Tao,M.H. and Morrison,S.L. (1989) Studies of aglycosylated chimeric mouse-human IgG. Role of carbohydrate in the structure and effector functions mediated by the human IgG constant region. J. Immunol., 143, 2595–2601.[Abstract]
Varki,A. (1992) Diversity in the sialic acids. Glycobiology, 2, 25–40.
Varki,A. (1993) Biological roles of oligosaccharides: all of the theories are correct. Glycobiology, 3, 97–130.
Varki,A. (1994) Selectin ligands. Proc. Natl Acad. Sci. USA, 91, 7390–7397
Varki,A. (1997) Sialic acids as ligands in recognition phenomena. FASEB J., 11, 248–255.[Abstract]
Varki,A. (1998) Factors controlling the glycosylation potential of the Golgi apparatus. Trends Cell Biol., 8, 34–40.[ISI][Medline]
Varki,A. and Marth,J. (1995) Oligosaccharides in vertebrate development. Semin. Dev. Biol., 6, 127–138.
Walker,M.R., Lund,J., Thomson,K.M. and Jefferis,R. (1989) Aglycosylation of human IgG1 and IgG3 monoclonal antibodies can eliminate recognition by human cells expressing Fc gamma RI and/or Fc gamma RII receptors. Biochem. J., 259, 347–353.[ISI][Medline]
Wright,A. and Morrison,S.L. (1998) Effect of C2-associated carbohydrate structure on Ig effector function: studies with chimeric mouse-human IgG1 antibodies in glycosylation mutants of Chinese hamster ovary cells. J. Immunol., 160, 3393–3402.
Yamada,E., Tsukamoto,Y., Sasaki,R., Yagyu,K. and Takahashi,N. (1997) Structural changes of immunoglobulin G oligosaccharides with age in healthy human serum. Glycoconjugate J., 14, 401–405.[ISI][Medline]
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  B. C. Koo, M. S. Kwon, B. R. Choi, J.-H. Kim, S.-K. Cho, S. H. Sohn, E. J. Cho, H. T. Lee, W. Chang, I. Jeon, et al. Production of germline transgenic chickens expressing enhanced green fluorescent protein using a MoMLV-based retrovirus vector FASEB J, November 1, 2006; 20(13): 2251 - 2260. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 Z. Yasukawa, C. Sato, K. Sano, H. Ogawa, and K. Kitajima Identification of disialic acid-containing glycoproteins in mouse serum: a novel modification of immunoglobulin light chains, vitronectin, and plasminogen Glycobiology, July 1, 2006; 16(7): 651 - 665. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Kamihira, K.-i. Ono, K. Esaka, K.-i. Nishijima, R. Kigaku, H. Komatsu, T. Yamashita, K. Kyogoku, and S. Iijima High-Level Expression of Single-Chain Fv-Fc Fusion Protein in Serum and Egg White of Genetically Manipulated Chickens by Using a Retroviral Vector J. Virol., September 1, 2005; 79(17): 10864 - 10874. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D. E. Akiyoshi, C. M. Rich, S. O'Sullivan-Murphy, L. Richard, J. Dilo, A. Donohue-Rolfe, A. S. Sheoran, S. Chapman-Bonofiglio, and S. Tzipori Characterization of a Human Monoclonal Antibody against Shiga Toxin 2 Expressed in Chinese Hamster Ovary Cells Infect. Immun., July 1, 2005; 73(7): 4054 - 4061. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
K. Nakajima, M. Kinoshita, Y. Oda, T. Masuko, H. Kaku, N. Shibuya, and K. Kakehi Screening method of carbohydrate-binding proteins in biological sources by capillary affinity electrophoresis and its application to determination of Tulipa gesneriana agglutinin in tulip bulbs Glycobiology, September 1, 2004; 14(9): 793 - 804. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 |