S46: Troubleshooting and mitigating the severe impact of a raw material lot-to-lot variability on the performance of a commercial fermentation process used for producing therapeutic proteins

Tuesday, August 14, 2012: 8:00 AM
Jefferson East, Concourse Level (Washington Hilton)
Hedieh Barkhordarian1, Kirsten Lewus2 and Xiaoming (Jerry) Yang1, (1)Cell Science and Technology, Amgen Inc., Thousand Oaks, CA, (2)Cell Science and Technology, Amgen Inc., Longmont, CO
Nearly 50% of currently-approved bio-pharmaceuticals are produced in the microbial systems. Of the microbial expression systems in use, E. coli is a major bacterial platform for expressing therapeutic proteins. Many different types of raw materials can be used for E. coli fermentations, ranging from simple chemicals to complex nutrients.  To promote cell growth and recombinant protein production, batch media and feeds consisting of concentrated sugar solutions such as glucose and yeast extract are used as a carbon source and a complex nitrogen source respectively.

The variability of the aforementioned complex raw material, yeast extract, can have a significant impact on process performance. The characterization of complex raw materials and understanding their impact on process and product quality are a necessity for the development of a well-understood manufacturing process that consistently produces high quality products.   A key task during commercial development process studies is to identify and set specifications for raw materials used for manufacturing scheme.   

In this study, lot-to-lot variations of yeast extract were investigated for an industrial bacterial fermentation process using specific orthogonal spectroscopic techniques. It was determined that the production of a recombinant protein in E. coli was affected by the variability of yeast extract and was linked to elevated acetate production, decreased cell growth and deficient product quality. These techniques proved to be a simple and low-cost strategy to ensure that raw material variations were identified early during the process development cycle. Subsequent strategies to mitigate the variability issues as identified by spectroscopic techniques will also be discussed.