Data driven approach for DSP development of recombinant glycoproteins from yeast

Abstract

In modern biotechnology, a huge variety of proteins are recombinantly produced for different applications. Today, the yeast Pichia pastoris is a promising recombinant production platform with many advantages compared to bacterial expression hosts, like posttranslational modifications and product secretion. However, P. pastoris hyperglycosylates recombinant proteins rendering the DSP development and purification itself laborious. Developing efficient purification protocols for glycoproteins is a cumbersome task since the offline analytical measurements are time consuming, expensive and prone to operator error. In addition, conventional particle-based chromatography systems that are in standard approaches for screening of different stationary phases and support materials are very time consuming, the throughput is very low and the screenings are based on trial-and-error approaches. Monolithic columns are the fourth generation of chromatography supports that are becoming popular because of their linear scalability, high mass transfer properties and short purification times. In the course of this thesis, we developed a novel toolbox using chromatogram fingerprints and multivariate data analysis (MVDA) tools for fast DSP development of recombinant glycoproteins from yeast. We screened for different columns and purification conditions in a design of experiments (DoE) approach using our toolbox in analytical scale and validate these results in laboratory scale. Based thereon, we implemented this novel toolbox for DSP development of six different recombinant glycoproteins from P. pastoris. Use of this platform toolbox for purification of novel proteins can reduce DSP development time and circumvent cumbersome offline analytics.