(571b) Proteomics Approach for Predicting Separation Behavior of a Mixture of Proteins During Downstream Purification

Swanson, R. K. - Presenter, Iowa State University
Xu, R. - Presenter, Iowa State University
Nettleton, D. S. - Presenter, Iowa State University

approach for predicting separation behavior of a mixture of proteins during downstream

Ryan Swanson1, Ruo Xu2,
Dan Nettleton2 and Charles E. Glatz1

1Department of Chemical and
Biological Engineering, Iowa State University, Ames, IA 50011

2Department of Statistics,
Iowa State University, Ames, IA 50011

Recombinant protein production has many applications spanning a
wide range of scientific fields, with downstream purification being the most
costly part of the overall process. One reason for this stems from a lack of
knowledge of the selected host cell protein's (HCP) separation behavior during
downstream purification. The process of selecting the downstream purification
method(s) as well as the host cell can benefit from an accurate prediction of
the HCP separation behavior, thereby reducing the resources needed to
investigate both. The HCP separation behavior was characterized using a method where
three of the most common protein properties involved in downstream purification
were obtained by aqueous two-phase partitioning or ATPS (surface
hydrophobicity) and 2-dimensional electrophoresis or 2DE (pI or isoelectric
point and molecular weight). Once the proteins had been characterized or
?mapped-out? by the three properties, a multivariate model was generated with
each characterization property used as a predictor variable and separation
behavior as the response variables. The accuracy of predicting a proteins separation
behavior using this 3-dimensional characterization method (ATPS + 2DE) had been
previously verified for ion exchange chromatography using a set of known model
proteins. Here results for hydrophobic interaction chromatography will be added,
again using a set of known proteins. This talk will additionally discuss the extension
of alternative statistical models for predicting the separation behavior of
selected downstream purification methods from 3D characterization of a complex
HCP extract (transgenic corn).