Processing tools and methods to enable continuous biomanufacturing for recombinat protein production with E.coli

Abstract

Fed-batch based cultivations are still state of the art in microbial pharmaceutical production, as high product titers can be achieved within short cultivation times. However, the trend in industry leads towards long-term cultivations, using equipment at optimum capacity. Continuous cultivations achieving stable productivity, would not only increase time-space yield in upstream, but also enable more robust downstream processing. However, the main challenge in microbial biomanufacturing is the instability of recombinant protein expression over time. Hence, the goal of this thesis was to (i) create basic knowledge upon regulation of continuous processes with the host E. coli and (ii) to establish an industrial relevant process design for robust continuous upstream for the production of recombinant proteins. This thesis is structured into two main parts, namely an enabler section and a second section treating continuous processing itself. In order to establish a knowledge platform to facilitate continuous cultivations with E. coli, we investigated our target strains in classical fed-batch approaches. As quantification of microbial proteins could be challenging, we designed an analytical method for proper quantification of produced target proteins via reversed phase liquid chromatography. Recombinant protein formation is known to stress host cells. In order to reduce stress onto host cells, we switched to the utilization of a soft inducer lactose and characterized our used E. coli strains in terms of physiological performance and the stress of induction agent. Even though, process parameters could have a severe impact onto measured process variables, many cultivations lack in proper process control. We established a mechanistic understanding using multivariate design spaces based on the process parameters temperature, specific feeding rate, pH and induction time.However, when switching to continuous cultivation, time-dependent variation in process variables could be observed. Therefore, we tried to (i) get an understanding of physiological effects, causing the drop in productivity, and (ii) to establish a cultivation device being capable of achieving time-independent productivity.In order to solve the first issue, we employed an online flow cytometry device using an E. coli BL21(DE3) strain expressing a green fluorescent protein as a model protein. By exploiting this new technique, we are able to detect population heterogeneities regarding productivity within microbial chemostat cultivations. The formation of these different subpopulations might be the reason for the drop in recombinant protein production over time. Hence, we are able to present a process analytical technology tool, being capable of measuring subpopulations. Despite our established knowledge on subpopulation appearance, we were not able to avoid time-dependent productivity using common cultivation techniques. To overcome these problems, we implemented a continuous cascaded system consisting of two sequentially operated continuous reactors. This allowed us to spatially separate biomass growth from recombinant protein production. Using our preliminary results, we were able to (i) declare the influence of the inducer in continuous cultivations (ii) target a promising continuous cultivation mode for recombinant protein formation and (iii) find out critical process parameters in cascaded continuous cultivation. The cascaded continuous cultivation device thus was investigated in a multivariate design of experiment approach, varying the inducer concentration and the overall dilution rate. By optimizing this continuous cascaded cultivation mode, we controlled the continuous process at stable productivity, whilst achieving industrially relevant titers. As continuous cultivations with bacterial hosts are known to suffer from time-dependent productivity, cultivation modes and analytical methods investigated in this thesis could help to realize continuous biomanufacturing with bacterial hosts.