Differences in the glycosylation profile of a monoclonal antibody produced by hybridomas cultured in serum-supplemented, serum-free or chemically defined media.

SFM (serum-free medium) is preferred to media containing animal-derived components when culturing mammalian cells for the production of therapeutic recombinant proteins and mAbs (monoclonal antibodies). Nonetheless, eliminating animal-derived components from media can strongly modify culture performance and alter protein glycosylation. In the present study, mAb glycosylation profiles, extracellular exoglycosidase activities, hybridoma growth and mAb production in traditional medium containing 10% (v/v) FBS (fetal bovine serum) [DMEM (Dulbecco's modified Eagle's medium)/FBS] were compared with those obtained in either SFM or CDM (chemically defined medium). SFM and CDM supported higher cell and mAb concentrations than did DMEM/FBS; however, CE (capillary electrophoresis) analyses revealed important changes in mAb glycosylation patterns. Glycosylation patterns showed a broad microheterogeneity in all the media, ranging from complex to high-mannose and paucimannosidic glycans. mAb produced in DMEM/FBS presented 26 glycan structures, whereas a lower glycan microheterogeneity was found for cultures in CDM or SFM, which presented 24 and 22 structures respectively. In DMEM/FBS and CDM, complex glycans without terminal galactose (G0) represented 28 and 32% of the total glycans respectively and 42 and 46% corresponded to galactosylated structures (G1 plus G2) respectively. In contrast, G0 glycans in SFM accounted for 58%, whereas only 28% corresponded to G1 and G2 structures. Extracellular beta-galactosidase activity increased approx. 3-fold in SFM, which can explain the higher G0 content compared with cultures in the other two media. A desirable decrease in sialylated structures, but an undesirable increase in fucosylated forms, was observed in mAb produced in SFM and CDM media. Approxi. 80% of potential mAb glycosylation sites were occupied, regardless of the culture medium used.

Heterogeneous conditions in dissolved oxygen affect N-glycosylation but not productivity of a monoclonal antibody in hybridoma cultures.

It is known that heterogeneous conditions exist in large-scale animal cell cultures. However, little is known about how heterogeneities affect cells, productivities, and product quality. To study the effect of non-constant dissolved oxygen tension (DOT), hybridomas were subjected to sinusoidal DOT oscillations in a one-compartment scale-down simulator. Oscillations were forced by manipulating the inlet oxygen partial pressure through a feedback control algorithm in a 220-mL bioreactor maintained at a constant agitation. Such temporal DOT oscillations simulate spatial DOT gradients that can occur in large scales. Different oscillation periods, in the range of 800 to 12,800 s (axis of 7% (air saturation) and amplitude of 7%), were tested and compared to constant DOT (10%) control cultures. Oscillating DOT decreased maximum cell concentrations, cell growth rates, and viability indexes. Cultures at oscillating DOT had an increased glycolytic metabolism that was evidenced by a decrease in yield of cells on glucose and an increase in lactate yield. DOT gradients, even several orders of magnitude higher than those expected under practical large-scale conditions, did not significantly affect the maximum concentration of an IgG(1) monoclonal antibody (MAb). The glycosylation profile of the MAb produced at a constant DOT of 10% was similar to that reported in the literature. However, MAb produced under oscillating culture conditions had a higher amount of triantennary and sialylated glycans, which can interfere with effector functions of the antibody. It was shown that transient excursions of hybridomas to limiting DOT, as occurs in deficiently mixed large-scale bioreactors, is important to culture performance as the oscillation period, and thus the time cells spent at low DOT, affected cell growth, metabolism, and the glycosylation pattern of MAb. Such results underline the importance of monitoring protein characteristics for the development of large-scale processes.