The therapeutic efficacy of a biopharmaceutical protein is extremely sensitive to the production conditions because they influence post-translation modifications. Therefore, any new concept for production of such proteins used in humans requires thorough characterization of its impact on the integrity of the product. The proposed study will investigate the author's hypothesis that slow growing perfusion cultures produce more therapeutically relevant glycoproteins compared to those produced in faster growing batch cultures. This proposal addresses an issue that has profound consequences from the practical perspective of advancing the development of an important new class of therapeutic agents (i.e., recombinant glycoproteins). Success in this area will have a significant financial payoff. In addition, from the perspective of "advancing knowledge" this proposal also has as one of its aims the goal of better understanding the underlying biological processes that determine glycosylation patterns.