Event
ChBE Seminar Series: Maria Klapa
Friday, November 1, 2013
10:00 a.m.-11:00 a.m.
Room 2108, Chemical and Nuclear Enginering Bldg.
Professor Jeffery Klauda
jbklauda@umd.edu
Metabolomics and Network Biology for Sensitive Monitoring in Cell Culture Engineering
Maria Klapa
Principal Researcher, Metabolic Engineering and Systems Biology Laboratory, FORTH/ICE-HT, Greece; and
Adjunct Professor
Dept. of Chemical and Biomolecular Engineering
University of Maryland
Mammalian cell cultures have been increasingly used for the production of complex biopharmaceuticals. As these products are of significance for human health, there is a crucial need for the development of robust processes that consistently produce material of high quality. Therefore, the industry is in need of tools for sensitive, high-resolution monitoring of the cell culture physiology throughout the entire process. Conventionally, the monitoring of mammalian cell cultures has been based on a small set of prime variables (i.e. growth rate, cell density and viability, product quality, substrate consumption and lactate production) and the relevant specific rates). However, this monitoring approach has limitations more prominently apparent when changes in product quality are observed despite no evident changes in the prime variable measurements. In addition, prime variables alone cannot provide extensive information about the culture physiology to allow for better understanding and optimization of the culture process based on experiments from process development activities, such as bioreactor scale-down model qualification and design space studies.
On the other hand, metabolic profiling can inherently provide a more extensive perspective of the cellular metabolic state than the set of the prime variables. In collaboration with Bayer Healthcare LLC, Berkeley, CA, we presented the first metabolic profiling analysis of industrial scale perfusion cultures of baby hamster kidney (BHK) cells, using Gas Chromatography–Mass Spectrometry (GC-MS), which provided strong evidence that metabolic profiles could sense subtle metabolic changes due to cell age. Based on a second specifically designed study, our group built on those results and used metabolomics and metabolic network analysis to determine characteristic metabolic patterns for the different phases of both laboratory and manufacturing scale perfusion systems. Finally, we used metabolomics to analyze the modifications in the physiology of cell cultures when subjected to various combinations of changes in pH, dissolved oxygen (DO), temperature (T) and cell specific perfusion rate (CSPR) at various points of the culture time course in the currently most extensive industrial-scale design of space (DOE) experiment. These studies validated that metabolomics analysis can indeed enhance the prime variable dataset for the monitoring of mammalian cell perfusion cultures, by providing a higher resolution view of the metabolic phenotype. In this context, metabolic profiling could be integrated into the monitoring of cell physiology in perfusion cultures. Additionally, metabolic profiling and network analysis can increase the information content of process development experiments by helping better understand the impact of changes in bioreactor operating conditions on cell physiology.
