Sequence Variants and Sequence Variant Analysis in Biotherapeutic Proteins

A “sequence variant” is a surrogate term covering any unintentional amino acid substitutions, omissions, or insertions during protein biosynthesis. Production of biotherapeutic proteins by living organisms is governed by biological processes responsible for protein production, including DNA replication, RNA transcription, and protein translation steps. Each of the biosynthesis steps has finite fidelity, with error rates ranging from 10 -9 per base pair for DNA replication to 10 -4 to 10 -5 per codon for protein translation. The occurrence of sequence variants contributes to heterogeneity of recombinant protein therapeutics. Establishing a sequence variant profile of a biotherapeutic product is essential in providing proof of its structure, its manufacturing consistency, and the stability of the producing cell line. DNA mutations can arise during generation of stable producing cell lines and be amplified by selection pressures used to establish high-producing clones. Once a clone with genetic variant is produced, this mutation will propagate though sub-cloning and can be affected by a cell age. For that reason, the prime goal of cell line (upstream) development is to screen for a possibility of DNA mutations early on during upstream development. In contrast, naturally abundant translational errors can further amplify due to nutrient and amino acid deprivation in production medium or a non-optimal codon usage in the construct. These errors usually occur at low levels and affect certain amino acids in multiple positions along the protein sequence, and can be both codon-dependent and independent. Although there is no standardized method for detection of sequence variants, this chapter discusses analytical methodologies for their screening. Peptide mapping with liquid chromatography-mass spectrometry (LC-MS) is currently viewed as a technology of choice for detection of sequence variants in recombinant proteins, with potential for detecting amino acid substitutions at low levels. On the example of sequence variant analysis in the NIST monoclonal antibody (mAb), performed by two independent laboratories, this chapter discusses strategy, capabilities, and limitations of peptide mapping with LC-MS for this application. Comparison of the intra-laboratory data illustrates reasonable agreement between the results, exhibiting a significant overlap in the specific sequence variants detected. Despite the excellent capabilities of peptide mapping method for detecting sequence variants in most cases at low or trace levels, there is no guarantee that such analysis is comprehensive, as discussed in this chapter. The use of multiple and orthogonal analytical methods often assists with the detection of variants missed by one or another method.