Search

The past few years have seen an interesting debate on the extent to which DNA-sequencing of expression systems is appropriate as a basis for assessing the quality and consistency of rDNA-derived products. While there is a need to improve and validate the ability of end-product testing procedures to detect low levels of variant proteins, there are several reasons why it may not be justified to require extensive DNA sequence analysis of large numbers of individual clones of producer cells. These include difficulties in interpreting the resultant data due to artefacts introduced by PCR and the fact that other events, such as translational errors and chemical and enzymic alterations during production and downstream processing, also lead to potential heterogeneity of pharmaceutical proteins. The evaluation of these products can only be made at the protein level. Some DNA sequencing will be required during product development, but in practice the quality and consistency of rDNA-derived products are best determined by a range of in-process and end-product testing procedures, including the use of biological assays and standardization.

Retroviruses have received attention as vectors for gene transfer because they can infect a variety of cell types, they integrate stably into the genome of the target cell and they have simple, well-defined structures. Insertion of the gene of interest into a retroviral vector results in the formation of replication-defective particles that can be produced in "packaging" cell lines. Due to the nature of the present generation of vectors and packaging cell lines, multiple recombination events would be needed to produce replication-competent particles. However, the concern still exists that replication-competent virus could arise either in the production of retroviral vector stocks or following their introduction into target cells. The current use of retrovirus vectors and the challenge they present to testing programmes are discussed.

A variety of different fermentation processes has been successfully employed to produce consistent protein-based biopharmaceuticals from genetically engineered animal cells. Chinese hamster ovary (CHO) cells were genetically modified to produce recombinant human soluble CD4, tissue plasminogen activator (tPA) or erythropoietin (EPO). Soluble CD4 was collected from extended perfused fermentations of several months' duration, during which some quantitative loss of DNA copy level, mRNA expression level, and fermentation titer were observed. In one extended run, a novel contaminant appeared in intermediates purified from later harvests. However, in all cases, the final soluble CD4 product was consistent in terms of purity and potency. Evaluation of genetic stability for tPA examined both biological traits at the cellular level as well as potency, purity and structure of product derived from cells at various levels of in vitro age; no significant cell age effects were observed. Similarly, evaluation of the EPO product showed that genetically-determined and process-determined traits such as potency, tryptic peptide mapping, and sialylation were consistent from lot to lot. These data exemplified how process design, process validation, and in-process and quality control assays can be used effectively to ensure the consistency of recombinant products derived from cell culture fermentations.

The genetic stability of biotechnology production systems is of importance in evaluating the safety and efficacy of protein products. The analysis methods currently available for assessing the genetic stability of such systems are limited by the complexity of the eukaryotic genome and by their inability to predict the effects of post-transcriptional and post-translational events, such as glycosylation or host cell protein level changes, on the protein produced. Therefore, it is important to focus attention on the protein product itself in evaluating product consistency rather than on nucleotide sequence analysis for genetic stability of the production systems involved. Peptide mapping has arisen as the best method for the analysis of proteins as a surrogate measure of genetic stability. This method uses known chemical reactions or enzymes to cleave the protein at specific sites, resulting in discrete peptides whose structure and composition can be readily compared to that of the theoretical protein coded for by the DNA sequence of its gene. Peptide mapping, which is capable of detecting single amino acid changes in a molecule, can also be used for determining glycosylation sites. Peptide mapping requires sophisticated chromatographic techniques and equipment, and is limited to molecules of m.w. of c. 100,000. The method also requires careful validation to limit the potential for misinterpretation of artifactual peaks. Despite its limitations, no better method currently exists to evaluate product consistency and, thereby, the genetic stability of recombinant products.