James C. Jamieson

Dissolved oxygen concentration in serum-free continuous culture affects N-linked glycosylation of a monoclonal antibody

The murine B-lymphocyte hybridoma, CC9C10, was grown at steady state in serum-free continuous culture at dissolved oxygen (DO) concentrations of 10, 50, and 100% of air saturation. The secreted mAb, an IgG1, was purified and subjected to both enzymatic deglycosylation using PNGase F and chemical deglycosylation by hydrazinolysis. Both methods resulted in complete removal of N-linked oligosaccharide chains. Isolated N-glycan pools were analyzed by fluorophore-assisted carbohydrate electrophoresis (FACE) and high pH anion exchange chromatography with pulsed amperometric detection (HPAEC-PAD). The FACE profiles and corresponding HPAEC-PAD chromatograms of N-linked oligosaccharides obtained by PNGase F digestion and hydrazinolysis provided complementary and corroborating information. The predominant N-linked structures were core-fucosylated asialo biantennary chains with varying galactosylation. There were also minor amounts of monosialylated, and trace amounts of afucosyl, oligosaccharides. A definite shift towards decreased galactosylation of glycan chains was observed as DO concentration in continuous culture was reduced. The vast majority of N-linked glycosylation occurred on the heavy chain. There was no evidence for N-linked glycosylation of the light chain or for O-linked glycosylation of the mAb.

Comparisons of the glycosylation of a monoclonal antibody produced under nominally identical cell culture conditions in two different bioreactors

The murine B‐lymphocyte hybridoma cell line, CC9C10, was grown in serum‐free continuous culture at steady‐state dissolved oxygen (DO) concentrations of 10%, 50%, and 100% of air saturation in both LH Series 210 (LH) and New Brunswick Scientific (NBS) CelliGen bioreactors. All culture parameters were monitored and controlled and were nominally identical at steady state in the two bioreactors. The secreted monoclonal antibody (mAb), an immunoglobulin G1, was purified and subjected to enzymatic deglycosylation using peptide N‐glycosidase F (PNGase F). Asparagine‐linked (N‐linked) oligosaccharide pools released from mAb samples cultured in each bioreactor at each of the three DO setpoints were analyzed by high‐pH anion‐exchange chromatography with pulsed amperometric detection (HPAEC‐PAD). The predominant N‐linked structures were core‐fucosylated asialo biantennary chains with varying galactosylation. There were also minor amounts of monosialyl oligosaccharides and trace amounts of afucosyl oligosaccharides. The level of DO affects the glycosylation of this mAb. A definite reduction in the level of galactosylation of N‐glycan chains was observed at lower DO in both bioreactors, as evidenced by prominent increases in the relative amounts of agalactosyl chains and decreases in the relative amounts of digalactosyl chainswith the relative amounts of monogalactosyl chains being comparatively constant. However, the quantitative results are not precise matches between the two bioreactors. The effect of DO on galactosylation is less pronounced in the NBS bioreactor than in the LH bioreactor, particularly the shift between the relative amounts of agalactosyl and digalactosyl chains in 10% and 50% DO. There are also perceptibly higher levels of sialylation of the mAb glycans in the NBS bioreactor than in the LH bioreactor at all three DO setpoints. The results indicate that the DO effect is not bioreactor specific and that nominally identical steady‐state conditions in different chemostat bioreactors may still lead to some incongruities in glycosylation, possibly due to the particular architectures of the bioreactors and the design of their respective monitoring and control systems. The observed differences in N‐linked glycosylation of the mAb secreted by the hybridoma grown in the LH and NBS bioreactors may be explained by the differences in oxygen supply and control strategies between the two bioreactors.

Large-scale expression of recombinant sialyltransferases and comparison of their kinetic properties with native enzymes

Values ofKm were determined for three purified sialyltransferases and the corresponding recombinant enzymes. The enzymes were Galβ1-4GlcNAc α2-6sialyltransferase and Galβ1-3(4)GlcNAc α2-3sialyltransferase from rat liver; these enzymes are responsible for the attachment of sialic acid to N-linked oligosaccharide chains; and the Galβ1-3GalNAc α2-3sialyltransferase from porcine submaxillary gland that is responsible for the attachment of sialic acid to O-linked glycoproteins and glycolipids. A procedure for the large scale expression of active sialyltransferases from recombinant baculovirus-infected insect cells is described. For the liver enzymes values ofKm were determined using rat and human asialoα1 acid glycoprotein andN-acetyllactosamine as variable substrates; lacto-N-tetraose was also used with the Galβ1-3(4)GlcNAc α2-3sialyltransferase. Antifreeze glycorprotein was used as the macromolecular acceptor for the porcine enzyme. Values forKm were also determined using CMP-NeuAc as the variable substrate.

The role of the carbohydrate chains of Galβ-1,4-GlcNAcα2,6-sialyltransferase for enzyme eactivity

Abstract Galβ-1,4-GlcNAcα2,6-sialyltransferase (CMP-N-acetylneuraminate : β-galactosideα2,6 sialyltransferase, EC 2.4.99.1) is a glycoprotein containing carbohydrate chains of the complex type (Jamieson, J.C. (1989) Life Sci. 43, 691–697). The carbohydrate chains may be important for controlling the expression of sialyltransferase catalytic activity during transit of the enzyme from the rough endoplasmic reticulum to the Golgi complex where it is active as a membrane bound enzyme anchored to the luminal face. To study the role of the carbohydrate chains of sialyltransferase for enzyme activity, conditions were established in which the native enzyme was deglycosylated with N-Glycanase and endo F. It was found that Glycanase removed the carbohydrate chains from native sialyltransferase, but methanol had to be present for rapid and complete deglycosylation. Presence of methanol or ethanol were not essential for removal of carbohydrate chains with endo F. There was a correlation between the loss of catalytic activity of sialyltransferase with increased deglycosylation. After deglycosylation with Glycanase for 18 h catalytic activity was largely eliminated and there was a reduction in molecular mass of about 5 kDa compared to the untreated enzyme when examined by immunoblot analysis; this reduction was identical to that found when the denatured enzyme was deglycosylated with Glycanase. At shorter times of incubation partially deglycosylated forms of the enzyme were detected. Complete deglycosylation of native or denatured sialyltransferase with endo F could not be achieved. However, incubation with endo F for 24 h resulted in a loss of catalytic activity of about 60%. Immunoblot analysis showed the presence of three forms of the enzyme corresponding in molecular mass to the native and deglycosylated enzyme and a third form corresponding to a partially deglycosylated enzyme. Sialyltransferase was also subjected to sequential treatment with exoglycosidases. Removal of NeuAc and Gal had little effect on catalytic activity, but subsequent removal of GlcNAc resulted in a significant loss in catalytic activity suggesting that the presence of the trimannose core with GlcNAc attached is important for the expression of catalytic activity. The presence of organic solvents during deglycosylation with Glycanase may be a useful method that can be applied to other glycoproteins.

Studies on the effect of glycoprotein processing inhibitors on fusion of L6 myoblast cell lines

The effect of oligosaccharide processing inhibitors on the fusion of L6 myoblasts was studied. The glucosidase inhibitors, castanospermine, 1-deoxynojirimycin and N-methyl-deoxynojirimycin were potent inhibitors of myoblast fusion, as was the mannosidase II inhibitor, swainsonine. Inhibition of fusion was reversed when inhibitors were removed. However, the mannosidase I inhibitor, 1-deoxymannojirimycin did not inhibit fusion. Changes in cell membrane oligosaccharide structure were followed by monitoring the binding of concanavalin A (conA) and wheat germ agglutinin (WGA) to cell surface membranes in cells treated with processing inhibitors. All the processing inhibitors resulted in increased binding of conA and decreased binding of WGA; this is consistent with the known mechanisms of inhibition of the inhibitors used in the study. Inhibition of fusion by the processing inhibitors also resulted in reduced activities of creatine phosphokinase, an enzyme used as a marker for biochemical differentiation during fusion. Treatment of a non-differentiating conA-resistant cell line with processing inhibitors did not induce fusion, but the cells did show altered lectin-binding properties. The main conclusion drawn from these studies is that cell surface glycoproteins probably containing the mannose (Man)9 structure are important for the fusion reaction.

Effect of nonionic detergent on fractionation of proteins by isoelectric focusing

Abstract The nonionic detergent Brij 35 was found to eliminate precipitation of protein during isoelectric focusing. Proteins were recovered after isoelectric focusing by chromatographic or electrophoretic methods. Recovered proteins were identical with unfractionated proteins and with corresponding proteins recovered after isoelectric focusing in absence of detergent when examined by electrophoresis on starch gel. Examination of proteins by infrared spectroscopy did not reveal the presence of any protein-bound detergent. The technique of isoelectric focusing in presence of Brij 35 appears to be useful for fractionating proteins when there is a possibility of precipitation occurring.

Effect of 1-phenyl-3-methyl-5-pyrazolone labeling on the fragmentation behavior of asialo and sialylated N-linked glycans under electrospray ionization conditions.

The advantages of labeling free N-linked oligosaccharides with 1-phenyl-3-methyl-5-pyrazolone (PMP), for high performance liquid chromatography (HPLC) and electrospray ionization mass spectrometry (ESI-MS) are discussed. The study focuses on some asialo and sialylated sugars, and compares the HPLC and ESI-MS behaviors of the PMP-labeled substances vs. the native compounds. It is pointed out that native free N-linked carbohydrates have very low affinities for the C18 reversed phases commonly used in HPLC. Native asialo oligosaccharides yield good ESI-MS sensitivity, although they are very susceptible to in-source collision-induced dissociation (CID), and the fragments are produced from any of the branches of the molecules, i.e. do not give specific structural information. Native N-linked standards bearing one sialic acid residue yield a 10-fold loss of ESI-MS sensitivity vs. asialo compounds, and native sugars with two sialic acid moieties were not detectable. The PMP labeling of asialo and sialylated sugars yielded higher affinities for HPLC C18 columns and, even at the early stages of method development, it was possible to separate three PMP-labeled standards to a useful extent. In ESI-MS, PMP-asialo sugars did not yield a significant increase in sensitivity vs. the native species; however, fragmentation produced by in-source CID was more directed as all predominant fragment ions contained the bis-PMP label. This feature is particularly useful when structural determination of an unknown sugar is required. PMP-sialylated sugars gave rise to very clean and informative ESI mass spectra. The monosialo sugar yielded a 100-fold sensitivity improvement vs. its native analog and, in the case of the disialylated compound, a 100% improvement was obtained in the positive mode. Most fragment ions were informative and contained the reducing end on the molecules, thus facilitating spectral interpretation. The combination of PMP derivatization with on-line HPLC/ESI-MS is a promising method for the analysis of asialo and sialylated carbohydrate mixtures.

Studies on the effect of the hepatocyte-stimulating factor on galactose-β1 → 4-N-acetylglucosamine α2 → 6-sialyltransferase in cultured hepatocytes

Abstract Rat hepatic Galβ1 → 4GlcNAcα2 → 6 sialyltransferase is released into the blood at elevated levels following an inflammatory challenge: this is a typical response of the group of plasma proteins known as acute-phase reactants. In the present study, primary cultures of liver parenchymal cells are used to demonstrate that the same hepatic cell type that produces plasma proteins such as fibrinogen also produces and releases sialyltransferase. Hepatic production of sialyltransferase is stimulated by a major regulator of hepatic acute-phase reactant production, the hepatocyte-stimulating factor (HSF), while another monokine, interleukin-1, does not affect hepatocyte sialyltransferase production. The maximum increase in sialyltransferase occurs 48 h after exposure to HSF which is considerably later than the fibrinogen response. The sialyltransferase that is stimulated by HSF is the Galβ1 → 4GlcNAcα2 → 6 isozyme.

Evidence for reduced uptake of asialo-α1-acid glycoprotein during the acute phase response to inflammation

Abstract Survival times of rat α 1 -acid glycoprotein were increased in the circulation of rats suffering from inflammation when compared with controls. Hepatic plasma membranes from rats suffering inflammation also had reduced capacities to bind asialo-α 1 -glycoprotein, but binding capacities increased to nearly normal values after washing with 0.01 M EDTA. One suggestion to explain the results is that there are elevated levels of circulating asialo glycoproteins during the acute phase response to inflammation.

Rat corticotropin, insulin and thyroid hormone levels during the acute phase response to inflammation.

Circulating levels of corticotropin, thyroid hormones and insulin were measured in rats at various times after turpentine-induced inflammation. Corticotropin increased rapidly showing a biphasic response with a four-fold increase at about 6-8 hr after inflammation and a 10-fold increase at 10 hr after inflammation. The response of insulin to inflammation was slower than corticotropin and the magnitude of the increase was smaller. Insulin increased by three-fold at 20 hr after inflammation. Thyroid hormone levels were depressed by turpentine inflammation. Levels fell after 4 hr and remained at low levels throughout. Administration of a cytokine preparation to rats also caused depressed thyroxine levels at short intervals after administration. However, levels increased at longer intervals after administration. This suggests that, like corticotropin and insulin, thyroid hormone levels could be under the control of immunotransmitters during the acute phase response.