The Interactions Between Fermentation and Protein Recovery

Mathematical modelling of expanded bed adsorption - a perspective on in silico process design

Expanded bed adsorption (EBA) emerged in the early 1990s in an attempt to integrate the clarification, capture and initial product concentration/purification process. Several mathematical models have been put forward to describe its operation. However, none of the models developed specifically for EBA allows simultaneous prediction of bed hydrodynamics, mass transfer/adsorption and (unwanted) interactions and fouling. This currently limits the development and early optimization of EBA-based separation processes. In multiphase reactor engineering, the use of multiphase computational fluid dynamics has been shown to improve fundamental understanding of fluidized beds. To advance EBA technology, a combination of particle, equipment and process scale models should be used. By employing a cascade of multiscale simulations, the various challenges EBA currently faces can be addressed. This allows for optimal design and selection of equipment, materials and process conditions, and reduces risks and development times of downstream processes involving EBA. © 2018 The Authors. Journal of Chemical Technology & Biotechnology published by John Wiley & Sons Ltd on behalf of Society of Chemical Industry.

Application of CFD in Bioprocessing: Separation of mammalian cells using disc stack centrifuge during production of biotherapeutics

Centrifugation continues to be one of the most commonly used unit operations for achieving efficient harvest of the product from the mammalian cell culture broth during production of therapeutic monoclonal antibodies (mAbs). Since the mammalian cells are known to be shear sensitive, optimal performance of the centrifuge requires a balance between productivity and shear. In this study, Computational Fluid Dynamics (CFD) has been successfully used as a tool to facilitate efficient optimization. Multiphase Eulerian-Eulerian model coupled with Gidaspow drag model along with Eulerian-Eulerian k-ε mixture turbulence model have been used to quantify the complex hydrodynamics of the centrifuge and thus evaluate the turbulent stresses generated by the centrifugal forces. An empirical model has been developed by statistical analysis of experimentally observed cell lysis data as a function of turbulent stresses. An operating window that offers the optimal balance between high productivity, high separation efficiency, and low cell damage has been identified by use of CFD modeling.

Artificial Intelligence vs. Statistical Modeling and Optimization of Continuous Bead Milling Process for Bacterial Cell Lysis

For a commercially viable recombinant intracellular protein production process, efficient cell lysis and protein release is a major bottleneck. The recovery of recombinant protein, cholesterol oxidase (COD) was studied in a continuous bead milling process. A full factorial response surface methodology (RSM) design was employed and compared to artificial neural networks coupled with genetic algorithm (ANN-GA). Significant process variables, cell slurry feed rate (A), bead load (B), cell load (C), and run time (D), were investigated and optimized for maximizing COD recovery. RSM predicted an optimum of feed rate of 310.73 mL/h, bead loading of 79.9% (v/v), cell loading OD600nm of 74, and run time of 29.9 min with a recovery of ~3.2 g/L. ANN-GA predicted a maximum COD recovery of ~3.5 g/L at an optimum feed rate (mL/h): 258.08, bead loading (%, v/v): 80%, cell loading (OD600nm): 73.99, and run time of 32 min. An overall 3.7-fold increase in productivity is obtained when compared to a batch process. Optimization and comparison of statistical vs. artificial intelligence techniques in continuous bead milling process has been attempted for the very first time in our study. We were able to successfully represent the complex non-linear multivariable dependence of enzyme recovery on bead milling parameters. The quadratic second order response functions are not flexible enough to represent such complex non-linear dependence. ANN being a summation function of multiple layers are capable to represent complex non-linear dependence of variables in this case; enzyme recovery as a function of bead milling parameters. Since GA can even optimize discontinuous functions present study cites a perfect example of using machine learning (ANN) in combination with evolutionary optimization (GA) for representing undefined biological functions which is the case for common industrial processes involving biological moieties.

In situ magnetic separation of antibody fragments from Escherichia coli in complex media

Background

In situ magnetic separation (ISMS) has emerged as a powerful tool to overcome process constraints such as product degradation or inhibition of target production. In the present work, an integrated ISMS process was established for the production of his-tagged single chain fragment variable (scFv) D1.3 antibodies (“D1.3”) produced by E. coli in complex media. This study investigates the impact of ISMS on the overall product yield as well as its biocompatibility with the bioprocess when metal-chelate and triazine-functionalized magnetic beads were used.

Results

Both particle systems are well suited for separation of D1.3 during cultivation. While the triazine beads did not negatively impact the bioprocess, the application of metal-chelate particles caused leakage of divalent copper ions in the medium. After the ISMS step, elevated copper concentrations above 120 mg/L in the medium negatively influenced D1.3 production. Due to the stable nature of the model protein scFv D1.3 in the biosuspension, the application of ISMS could not increase the overall D1.3 yield as was shown by simulation and experiments.

Conclusions

We could demonstrate that triazine-functionalized beads are a suitable low-cost alternative to selectively adsorb D1.3 fragments, and measured maximum loads of 0.08 g D1.3 per g of beads. Although copper-loaded metal-chelate beads did adsorb his-tagged D1.3 well during cultivation, this particle system must be optimized by minimizing metal leakage from the beads in order to avoid negative inhibitory effects on growth of the microorganisms and target production. Hereby, other types of metal chelate complexes should be tested to demonstrate biocompatibility. Such optimized particle systems can be regarded as ISMS platform technology, especially for the production of antibodies and their fragments with low stability in the medium. The proposed model can be applied to design future ISMS experiments in order to maximize the overall product yield while the amount of particles being used is minimized as well as the number of required ISMS steps.

Semi-continuous in situ magnetic separation for enhanced extracellular protease production-modeling and experimental validation

In modern biotechnology proteases play a major role as detergent ingredients. Especially the production of extracellular protease by Bacillus species facilitates downstream processing because the protease can be directly harvested from the biosuspension. In situ magnetic separation (ISMS) constitutes an excellent adsorptive method for efficient extracellular protease removal during cultivation. In this work, the impact of semi-continuous ISMS on the overall protease yield has been investigated. Results reveal significant removal of the protease from Bacillus licheniformis cultivations. Bacitracin-functionalized magnetic particles were successfully applied, regenerated and reused up to 30 times. Immediate reproduction of the protease after ISMS proved the biocompatibility of this integrated approach. Six subsequent ISMS steps significantly increased the overall protease yield up to 98% because proteolytic degradation and potential inhibition of the protease in the medium could be minimized. Furthermore, integration of semi-continuous ISMS increased the overall process efficiency due to reduction of the medium consumption. Process simulation revealed a deeper insight into protease production, and was used to optimize ISMS steps to obtain the maximum overall protease yield.

Enhanced cell disruption strategy in the release of recombinant hepatitis B surface antigen from Pichia pastoris using response surface methodology

BACKGROUND

Cell disruption strategies by high pressure homogenizer for the release of recombinant Hepatitis B surface antigen (HBsAg) from Pichia pastoris expression cells were optimized using response surface methodology (RSM) based on the central composite design (CCD). The factors studied include number of passes, biomass concentration and pulse pressure. Polynomial models were used to correlate the above mentioned factors to project the cell disruption capability and specific protein release of HBsAg from P. pastoris cells.

RESULTS

The proposed cell disruption strategy consisted of a number of passes set at 20 times, biomass concentration of 7.70 g/L of dry cell weight (DCW) and pulse pressure at 1,029 bar. The optimized cell disruption strategy was shown to increase cell disruption efficiency by 2-fold and 4-fold for specific protein release of HBsAg when compared to glass bead method yielding 75.68% cell disruption rate (CDR) and HBsAg concentration of 29.20 mg/L respectively.

CONCLUSIONS

The model equation generated from RSM on cell disruption of P. pastoris was found adequate to determine the significant factors and its interactions among the process variables and the optimum conditions in releasing HBsAg when validated against a glass bead cell disruption method. The findings from the study can open up a promising strategy for better recovery of HBsAg recombinant protein during downstream processing.

Bacterial cell disruption: a key unit operation in the recovery of intracellular products

The need for microbial cell disruption has hindered the large scale production of commercial biotechnological products of intracellular derivation. The intracellular nature of many recombinant products and the potential use of the bacterial storage product, PHB as a commodity thermoplastic have renewed interest in the improvement of this unit operation. This paper provides a review of processes of a mechanical, physical, chemical or biological nature used for cell disruption on both the laboratory and large scale. Applicability of the techniques to large scale operation is discussed. Modification of existing processes is suggested for the reduction of energy requirements and improved process economics. The requirements for the liberation of granular intracellular products such as inclusion bodies and virus-like yeast particles are distinguished from those for the liberation of soluble products, mainly proteinaceous in nature. The integrated nature of the process with both upstream and downstream processes is addressed. Finally, the recent approach of selective liberation of soluble products of interest is reviewed.

Hydrophobic interaction ligand selection and scale-up of an expanded bed separation of an intracellular enzyme from Saccharomyces cerevisiae

A prototype Streamline-Phenyl matrix was evaluated in a hydrophobic interaction mode for the direct recovery of alcohol dehydrogenase (ADH) from yeast cell homogenate. At 5% breakthrough of ADH, a yield of 100% was obtained for a dynamic expanded bed capacity of 240 U(ADH)/ml matrix with a purification factor of 9.2. This compared with a dynamic capacity of 3013 U(ADH)/ml matrix for the packed bed equivalent and a purification factor of 18. In both systems the purification factor was found to increase simultaneously with a decrease in yield as the load of homogenate or breakthrough of ADH was increased. The expanded bed mode of operation conferred considerable robustness with respect to process fouling. No loss in yield was seen over five cycles of repeat loading with an unclarified homogenate. By contrast the packed bed media showed a decrease in yield from 86 to 56% over the same period. Successful scale up of the expanded bed protocol for a 20% breakthrough was demonstrated over a fourfold increase in column diameter. The application of hydrophobic interaction chromatography mediated expanded bed adsorption and its scale-up is discussed in the context of large-scale operations.

Immobilised metal ion affinity chromatography purification of alcohol dehydrogenase from baker's yeast using an expanded bed adsorption system

Alcohol dehydrogenase (ADH) from solutions of homogenised packed bakers' yeast has been successfully purified using immobilised metal-ion affinity chromatography in an expanded bed. Method scouting carried out using pure ADH solutions loaded onto 5-ml HiTrap columns charged with Zn2+, Ni2+ and Cu2+ and eluted using 0-50 mM EDTA gradient found that charging with Zn2+ gave the highest recovery and the lowest EDTA concentration required for elution. These results were used to develop a protocol for the expanded bed system and further tested using clarified yeast homogenate loaded onto XK16/20 packed beds (approximately 30 ml) packed with Chelating Sepharose FastFlow matrix in order to determine the optimum elution conditions using EDTA. The ADH was found to elute at 5 mM EDTA and the dynamic and total binding capacities of Streamline chelating for ADH were found to be 235 U/ml and 1075 U/ml matrix, respectively. Expanded bed work based on a step EDTA elution protocol demonstrated that ADH could be successfully eluted from unclarified homogenised bakers' yeast diluted to 10 mg/ml total protein content with a recovery of 80-100% that was maintained over five consecutive runs with a vigorous clean-in-place procedure between each run.

Process-scale disruption of microorganisms

Common hosts for the large-scale manufacture of biological products, such as Escherichia coli and Saccharomyces cerevisiae, do not excrete products to the medium. Effective techniques for cell disruption are therefore required. These include physical, chemical, enzymatic and mechanical methods. Mechanical methods such as bead milling, high-pressure homogenization, and microfluidization are preferred. However, gentler, specific methods are receiving increasing attention particularly when used in combination to synergistically exploit their different specificities. Benefits can also be derived by integrating product release and recovery. In all cases it is essential to consider the interaction of the disruption operation with downstream units and to clearly demonstrate the cost benefits of alternative strategies.

Downstream processing of proteins: Recent advances

This review on the downstream processing of proteins describes innovations that have occurred in the field since 1983. Several areas have seen particularly high levels of achievement, and are accorded expanded coverage relative to our previous review [1]. As an example, the increasing integration of downstream operations with upstream technologies, such as molecular biology and fermentation, has led to the development of some very powerful processes. The degree to which organizations understand that there needs to be one unified process, rather than the independent steps of cloning, fermentation and recovery, seems directly related to the ultimate speed and success of the development effort. In 1983 one of the most active development areas was chromatography, especially affinity chromatography. This is still true today, and this topic has been expanded to include biospecific adsorptions that would not traditionally be classified as chromatography. With more proteins being developed for human administration, there has been an increased emphasis on all aspects of process hygiene. In addition, there has been much discussion about the impact of regulatory demands on the design and development of the manufacturing processes. Therefore, a section has been added which covers several of the regulatory issues that have been raised for products of the new biotechnology. Finally, as some of the early process development achievements are now beginning to bear fruit in the form of patents, we have increased our citation of this area of the literature.