Increased demand for biotherapeutics calls for faster and less expensive production methods. In mammalian cells, recombinant protein therapeutics are normally generated using stable gene expression (SGE), but establishing a productive host cell line is resource intensive and can take many months. Transient gene expression (TGE) has emerged as a promising method to speed up the process and reduce costs, providing high protein yields in just a few days or weeks following transfection. However, TGE brings too many concerns about potential product quality and batch-to-batch variation to be widely adopted in large-scale production. One of the most obvious batch-to-batch variations seen in cell culture is in post-translational modifications (PTM) of the protein.
Post-translational modifications such as glycosylation have profound effects on a protein’s stability and function and are susceptible to the cell culture conditions under which the protein is produced. Thus, methods to determine that a protein’s glycosylation pattern is “correct” are important in the production of protein therapeutics, and high-resolution mass spectrometry has proven to be highly suited for this purpose.
Intact protein and bottom-up MS have been used in conjunction to characterize batch-to-batch variability of N-glycosylation of recombinant monoclonal IgG produced by TGE. IgG samples from ten batches of transiently transfected HEK-293E cells grown using a basic culturing system were purified and compared with an SGE-produced control IgG. Glycosylation of the intact purified IgG was qualitatively measured by on-line LC–electrospray ionization–quadrupole time-of-flight (LC–ESI–qTOF) MS. The results indicated that the glycosylation pattern was conserved from batch to batch.
A bottom-up approach using LC–Fourier transform ion cyclotron resonance (LC–FT-ICR) MS was employed to further characterize the glycosylation profiles of the IgG from each batch. The purified samples were digested with trypsin, and the resulting glycopeptides were identified and quantified. The method enabled identification of glycopeptides containing mannose and complex oligosaccharides, but none with sialic acid, possibly due to low ionization efficiency of these glycopeptides. Relative abundance of the four major glycans was consistent over nine of the ten batches, and the variation observed in the exceptional batch was minimal.
The same bottom-up approach was also used to assess the glycosylation pattern within a given batch at three, five, seven, and ten days post transfection. The relative glycan abundances were consistent at these time points, suggesting that the N-linked glycans remain stable in the cultures, even though the viability of the cells decreases over time.
These experiments show that recombinant IgG can be produced using TGE with reproducible yield and glycosylation in a simple, non-instrumented culture system. Naturally, reproducibility in production-scale bioreactors and means to analyze sialic acid-containing glycopeptides will require further investigation. Nevertheless, the detailed structural information provided by mass spectrometry methods will most certainly enable such assessments and will support the implementation of improved production methods for other biotherapeutics.
Read the details of this study.