Work package 3

Project Title: Automating experimentation in miniaturized reactors (WP3)

ESR 9    

Host: DTU, Denmark

Objectives: To develop next generation high throughput screening methodology. Initially a standard DoE will be carried out and tested in a miniaturized reactor system, where a proposed series of experiments will be analysed with automated data interpretation before the next experiment is carried out. Subsequently Optimal experimental design (OED), usually based on mechanistic models,  and enabling automated experimentation in microscale, will be developed on a recombinant yeast model system and then validated on industrial case studies from FUJI and SAN.

Expected Results: Optimal DoE methodology, based on first principle models, automating the high throughput experimentation data analysis formulated in a model recombinant yeast system and then validated on the industrial case studies of the beneficiaries. The resulting methodology will form a crucial part of the developed rapid bioprocess development framework.



Project Title: Effect of scale up strategies on population heterogeneity (WP3)

ESR 14  

Host: TUB, Germany

Objectives: Investigation of the effects of substrate heterogeneities, as they occur in large-scale bioreactors, on the population heterogeneity of recombinant yeast processes. To develop models which describe the population heterogeneity in dependent on substrate oscillations and to use these models in model based experimental designs for process optimisation and final validation in a scale down bioreactor. Studies in a scale-down more-compartment bioreactor will be complemented by flow cytometric analyses for population analysis. Population heterogeneity balance models will be performed on the basis of an earlier constructed multi-scale model for a model system S. cerevisiae consisting of a two-stage one-dimensional population balance model (PBM) describing cell size and cell cycle distributions connected to an unstructured kinetic model describing macroscopic variables will be expanded to describe the changes of the population as a function of time, when the cells are transported from one compartment to the other in a multi-compartment reactor (simulating different zones in a large-scale reactor). One of the most significant heterogeneity reporter properties, e.g. stress level or glucose availability will be selected and knowledge about population distribution of this property will be integrated in the model. The resulting PBMs will then be used to simulate a larger number of experimental conditions in a parallel minifermenter system (2mag, which has been set up at TUB) and to optimise the process. Finally, the optimal conditions will tested in the two- and/or three-compartment bioreactor.

Expected Results: Methodology of yeast process development n mini-fermenter and scale down bioreactor systems. Deeper knowledge on population balance in fed-batch yeast cultivations. A population balance model which is applicable and tested fed-batch fermentations, which compromise well mixed bioreactors and inhomogeneous reactor systems. An improved yeast process for the model applied in the project.



Project Title: Scale-up methodology for recombinant protein production in Escherichia coli (WP3)

ESR 15  

Host: TUB, Germany

Objectives: To develop the methodology which can estimate the effect of oscillations, as they occur in large-scale industrial reactors, on the quality of recombinant proteins. The examples include (i) a disulphide bond connected pharmaceutical protein which is produced in inclusion bodies (e.g. interleukin or proinsulin). We will investigate in detail the effect of process relevant oscillations on the quantity of the target product, its refoldability from inclusion bodies, and its co-and posttranslational modifications of the protein.

Expected Results: Comparison of glucose-limited fed-batch cultivations in an STR and in a two- and three-CR system for the recombinant model protein. Analysis of the general fermentation parameters and of the product (amount, amino acid composition by MS, composition of IBs, refoldability). Model formation, reduction and application for phase recognition, implementation of population balances. Model guided robustness analysis and process optimisation in a parallel mnibioreactos system under special consideration of intermittent feed strategies to simulate large scale heterogeneities.  Validation of optimised strategy in the scale down reactor system.



Project Title: Minifermenter based scale down strategies (WP3)

ESR 13  

Host: Sanofi, Germany

Objectives: Development of strategies which allow the simulation of the large-scale behaviour of E. coli and yeast processes in mini-fermenter systems. The specific aim is to use a pall minifermenter system for process development and robustness analysis. Therefore aside from the fed-batch process with continuous feeding, which is implemented by the EnBase technology, pulse based feed scenarios are tested to simulate scaling effects. The experiments will be connected to DoE and model based approaches (model based experimental design) to achieve the evaluation of the experimental space and to select robust process points. The results are evaluated and compared in relation to industrial process schemes. Also it is intended that the scale down reactor at TU Berlin may be used as a scale down simulator.

Expected Results: Implementation of high throughput technologies, DoE and model based approaches in a combined way for a faster process development. New miniscale fermentation devices are important tools and quality by design is reached through combining these instruments with model based computational approaches with the aim to determine the key model parameters in parallel experiments.



Project Title: Consistent scale-up of downstream processing unit (WP3)

ESR 7    

Host: Chr Hansen AS, Denmark

Objectives: This project is concerned about establishing models for the efficient and reliable transfer of freeze drying profiles from laboratory/pilot (1-5 kg) scale dryers to production scale (2000 kg). This will also decrease the need for process validation in large scale, along with improving understanding of the influence of the interrelation between the drying process (drying profiles), the drying equipment and the material to be dried – on the final product quality. The project will combine the use of freeze drying theory, physical chemistry, theoretical physics, cryo-biology and microbiology. The focus will be on selected lactic acid bacteria (including Lactococcus lactis, Streptococcus thermophiles, Lactobacillus acidophilus, Bifidobacterium). The analytical approach will include drying evaluation based on drying time, dryness/residual water content, structure analysis (collaps, meringue formation), porosity, microscopy as well as evaluation of product performance (acidification activity, colony forming units pr. gram) along with storage stability testing. The physical characteristics of the freeze dryers will also influence energy and mass transport during drying and a model will be build using computational fluid dynamics to integrate dryer characteristics and to support transfer of lab-scale results to large scale.

Expected Results: Formulation of a model based downstream processing scale up methodology for freeze-drying of a range of microorganisms and its validation on an industrial case study. The resulting methodology will form part of the final rapid bioprocess scale up framework.