Improving Downstream Process Related Manufacturability Based on Protein Engineering-A Feasibility Study.

While bioactivity and a favorable safety profile for biotherapeutics is of utmost importance, manufacturability is also worth of consideration to ease the manufacturing process. Manufacturability in the scientific literature is mostly related to stability of formulated drug substances, with limited focus on downstream process-related manufacturability, that is, how easily can a protein be purified. Process-related impurities or biological impurities like viruses and host cell proteins (HCP) are present in the harvest which have mostly acid isoelectric points and need to be removed to ensure patient safety. Therefore, during molecule design, the surface charge of the target molecule should preferably differ sufficiently from the surface charge of the impurities to enable an efficient purification strategy. In this feasibility study, we evaluated the possibility of improving manufacturability by adapting the surface charge of the target protein. We generated several variants of a GLP1-receptor-agonist-Fc-domain-FGF21-fusion protein and demonstrated proof of concept exemplarily for an anion exchange chromatography step which then can be operated at high pH values with maximal product recovery allowing removal of HCP and viruses. Altering the surface charge distribution of biotherapeutic proteins can thus be useful allowing for an efficient manufacturing process for removing HCP and viruses, thereby reducing manufacturing costs.

Functional studies with IgM and IgA immunoglobulins: binding to pIgR, FcαµR, FcµR, and CDC activities.

IgMs are the first antibodies produced by the immune system upon encounter of a possible pathogen and are one of five antibody subclasses in humans. For IgG, the most intensively studied antibody class, the N-linked glycosylation site located in the Fc-domain is directly involved in high affinity binding to the respective receptors and initiation of corresponding immune response. IgM molecules have five N-glycosylation sites and one N-glycosylation site in the J-chain, which can be incorporated in IgM or IgA molecules. There is only limited knowledge available concerning the function of these N-glycosylations in IgMs. To address this question, we produced IgM molecules lacking a particular N-glycosylation site and tested these variants as well as IgA molecules for binding to the known receptors: the polymeric immunoglobulin receptor (pIgR), the dual receptor for IgA and IgM, FcαµR, and the specific receptor for IgM, FcµR. The single glycosylation sites did not show an impact on expression and multimerization, except for variant N402Q, which could not be expressed. In SPR measurements, no major impact on the binding to the receptors by particular glycosylation sites could be detected. In cellular assays, deglycosylated variants showed some alterations in induction of CDC activity. Most strikingly, we observed also binding of IgA to the FcµR in the same affinity range as IgM, suggesting that this might have a physiological role. To further substantiate the binding of IgA to FcµR we used IgA from different origins and were able to confirm binding of IgA preparations to the FcµR.

A multivalent antibody assembled from different building blocks using tag/catcher systems: a case study.

The field of therapeutic antibodies and, especially bi- or multispecific antibodies, is growing rapidly. Especially for treating cancers, multispecific antibodies are very promising, as there are multiple pathways involved and multispecific antibodies offer the possibility to interfere at two or more sites. Besides being used as therapeutic, multispecific antibodies can be helpful tools in basic research. However, the design and choice of the most appropriate multispecific antibody format are far from trivial. The generation of multispecific antibodies starts with the generation of antibodies directed against the desired targets and then combining the different antigen-binding sites in one molecule. This is a time-consuming and laborious approach since the most suitable geometry cannot be predicted. The SpyTag technology is based on a split-protein system, where a small peptide of said protein, the SpyTag, can bind to the remaining protein, the SpyCatcher. An irreversible isopeptide bond between the SpyTag and the SpyCatcher is formed. A related Tag-Catcher system is the SnoopTag-SnoopCatcher. These systems offer the opportunity to separately produce proteins fused to the tag-peptides and to the catcher-domains and assemble them in vitro. Our goal was to design and produce different antibody fragments, Fab domains and Fc-containing domains, with different tags and/or catchers as building blocks for the assembly of different multivalent antibodies. We have shown that large multivalent antibodies consisting of up to seven building blocks can be prepared. Binding experiments demonstrated that all binding sites in such a large molecule retained their accessibility to their corresponding antigens.

Impact of lipopolysaccharides on cultivation and recombinant protein expression in human embryonal kidney (HEK-293) cells.

The human embryonal kidney 293 cell (HEK-293) is a widely used expression host for transient gene expression. The genes or plasmids used for the transient transfections are usually propagated and extracted from the gram-negative bacterium Escherichia coli, the workhorse for molecular biologists. As a gram-negative bacterium E. coli has an outer membrane (OM) containing lipopolysaccharides (LPS) or endotoxins. LPS are very potent inducers of inflammatory cytokines in the body. In early research phases DNA intended for transient transfections is not routinely checked for LPS-levels. In this study we addressed the question whether LPS has an impact on the cultivation and production of a recombinant antibody. At high concentrations the presence of LPS has a detrimental impact on cell viability and recombinant protein expression. But low LPS concentrations are tolerated and might even enhance protein expression levels.

Production of a novel heterodimeric two-chain insulin-Fc fusion protein.

Insulin is a peptide hormone produced by the pancreas. The physiological role of insulin is the regulation of glucose metabolism. Under certain pathological conditions the insulin levels can be reduced leading to the metabolic disorder diabetes mellitus (DM). For type 1 DM and, dependent on the disease progression for type 2 DM, insulin substitution becomes indispensable. To relieve insulin substitution therapy for patients, novel insulin analogs with pharmacokinetic and pharmacodynamic profiles aiming for long-lasting or fast-acting insulins have been developed. The next step in the evolution of novel insulins should be insulin analogs with a time action profile beyond 1-2 days, preferable up to 1 week. Nowadays, insulin is produced in a recombinant manner. This approach facilitates the design and production of further insulin-analogs or insulin-fusion proteins. The usage of the Fc-domain from immunoglobulin as a fusion partner for therapeutic proteins and peptides is widely used to extend their plasma half-life. Insulin consists of two chains, the A- and B-chain, which are connected by two disulfide-bridges. To produce a novel kind of Fc-fusion protein we have fused the A-chain as well as the B-chain to Fc-fragments containing either 'knob' or 'hole' mutations. The 'knob-into-hole' technique is frequently used to force heterodimerization of the Fc-domain. Using this approach, we were able to produce different variants of two-chain-insulin-Fc-protein (tcI-Fc-protein) variants. The tcI-Fc-fusion variants retained activity as shown in in vitro assays. Finally, prolonged blood glucose lowering activity was demonstrated in normoglycemic rats. Overall, we describe here the production of novel insulin-Fc-fusion proteins with prolonged times of action.

A fully automated three-step protein purification procedure for up to five samples using the NGC chromatography system.

The drug discovery process in the biopharmaceutical industry usually starts with the generation of plasmids coding for certain proteins. Due to advances in cloning techniques the generation of thousands of different plasmids is not a limiting factor anymore. The next step is the expression and evaluation of the proteins. In recent years significant progress has been made in the miniaturization of protein expression and purification. These processes have been adapted to robotic platforms and hundreds of proteins can be expressed and purified in parallel. As a consequence of miniaturization, the protein purification is restricted to a one-step process. In addition the amount of purified protein is usually in the µg-range. This might be suitable if a sensitive initial screening assay is available. However, when larger amounts of proteins are required robotic platforms are no longer appropriate. In addition, a one-step purification procedure is often not sufficient to obtain pure protein preparations. To address this topic we have used the NGC chromatography system for automated purification of up to five samples using a three-step purification procedure. The first chromatographic step is the capture step followed by a desalting step. The final purification was done using size exclusion chromatography. This set-up reduces the overall-time needed for protein production, needs minimal operator invention, is easy to handle and thus increases the throughput.