Industry pressures encourage and sometimes ‘force’ biopharmaceutical companies to implement process changes throughout a product’s lifecycle, so as to enhance yields, purity, robustness and cost-effectiveness. However, making a change involves technical, regulatory, and clinical risks. Possible changes to a product’s quality mean that all changes must be backed-up either with non-clinical bioequivalence studies or with lengthy and costly clinical trials and approved by regulatory authorities. These hurdles combined with the upfront costs can results in a tendency to avoid changes, whereas they may represent economic opportunity if evaluated holistically. This thesis explores the possibility of creating a systematic evaluation framework that captures the technical and regulatory activities involved in process changes to rapidly gauge the potential cost and risk implications. Fundamentaldrivers and consequencesof making bioprocesses changes were benchmarked in a survey to help create the framework model. Key technical activities were captured, namely development, manufacturing, retrofitting and validation at all stages of development. Impacts of changes were linked to regulatory activities needed to assess comparability. Resulting uncertainties such as the likelihood of repeating clinical trials, market losses, delays to market from retrofit, revalidation, or regulatory approval disruptions, and the costs involved in proving product equivalence were captured. The framework was translated into Microsoft Excel with macros for Monte Carlo simulations to account for the uncertainties. Minor and major change scenarios based on the purification of polyclonal IVIG by means of a blood-plasma fractionation process were used to demonstrate the usefulness of the proposed framework. The impact of ‘forced’ and optional changes were compared at different stages of development. Changes made during late-phase development resulted in market share losses and delays that outweighed any yield improvement modifications. The model predicted that it would be more profitable to make process modifications either during early phase development or post-product approval assuming stockpiling of approved product was feasible. The feasibility of purifying a new product, alpha-1 antitrypsin (AAT) from a waste fraction, Fraction IV precipitate, was another process change scenario explored using scale-down studies. Experimental trials of the preliminary filtration and anion exchange purification steps were carried out, yielding low recoveries of AAT. Ciphergen®’s SELDI-TOF-MS ProteinChip technology was used to investigate the value of using a high throughput optimisation method to improve the isolation of AAT. Quantitative analysis of the protein samples using the Ciphergen® was compared to well-established protein concentration determination methods, eliminating variability in samples and differences in MS intensity by normalising the data. The work in this thesis has demonstrated the usefulness of a combined business, technical and risk approach for evaluating the risks and benefits of implementing process changes.

In order to fully realise the benefits of microscale mammalian cell culture and microbial fermentation systems, a device capable of online sample preparation to enable further investigation of product quality is a key requirement. The aim of this work is to move toward such a device by designing and characterising a microfluidic lysis device and microaffinity chromatography device that are compatible with each other. The resulting microfluidic lysis device is useful for preparatory lysis of microbial cells. It works by mixing a lysis reagent (BugBuster MastermixTM), with microbial culture, using a T-Piece connection. Lysis takes place in a 700µm internal diameter fused silica capillary. The device was able to successfully lyse microbial cells with similar active Glutathione S Transferase release to sonication. The operating flowrate range of the device was 3.207µL min-1 to 6.414 µL min-1 and the device volume was 30µL - 60µL. The microaffinity chromatography column performed well in studies with pure Glutathione S Transferase. It showed good loading and elution behaviour. The breakthrough and elution curves, and quantity of protein eluted per unit bed volume, were similar to lab scale. The difference being as a result of experimental error. The column also performed well with a 100% clarified Escherichia coli lysate containing recombinant Glutathione S Transferase from Schistosoma japonicum. The eluate had a purity of 55% and concentration of 2.24 mg/ml. The column was fabricated from inexpensive fused silica capillary. It had an internal diameter of 700µm, a length of 5cm (the same length as a typical lab scale Glutathione Affinity column), and a bed volume of approximately 19µL. The operating flowrate range for the column was the same as the microlysis device.

The synthesis of two sets of different orthogonally protected lanthionine ready for incorporation into solid phase peptide synthesis to form cyclised peptides is described in this thesis, along with the cyclisation of individual rings D and E and the overlapping rings D and E. Previously developed orthogonally protected lanthionine containing Aloc, allyl, Fmoc and tBu protecting groups was synthesised using published synthetic route developed by Tabor’s group. A novel orthogonally protected lanthionine containing Teoc, TMSE, Fmoc and Tce group derivative has also been synthesised, after carrying several synthetic pathways. Both lanthionine residues contain protecting groups which are orthogonal to each other, which are also orthogonal to the transient Fmoc and permanent Boc/tBu protecting groups which are used in Fmoc based solid phase peptide synthesis. Incorporation of the previously developed lanthionine with Aloc/allyl protecting groups was carried out to form an analogue of ring E of nisin for the first time. Deprotection of the Aloc/allyl protecting groups were carried out with Ph(PPh3)4 using N’,N-dimethyl-barbituric acid (NDMBA). The second orthogonally protected lanthionine was also incorporated into solid phase peptide synthesis to synthesise an analogue of ring D of nisin. This was also to see whether this can be used to synthesise lanthionine-containing thio-ether bridged cyclic peptide by solid phase peptide synthesis. Teoc and TMSE deprotection was carried out in the presence of TBAF without effecting the other side chain and Fmoc protecting groups. Full characterisation of individual rings D and E were obtained. Quadruply orthogonal protecting group strategy was used to synthesise bicyclic peptide with two overlapping lanthionine bridges rings D and E. An effective methodology has been developed for the synthesis of the overlapping rings D and E of nisin by solid phase peptide synthesis.

Electrodialysis (ED) is an established technology used to transport small ions from one solution to another through an ion exchange (IE) membrane under the influence of an applied electric potential difference. This project aimed to develop a novel integrated bioreactor-ED system and to explore its application to a variety of bioprocesses including microbial fermentation and enzymatic bioconversion. A custom ED module was first designed and constructed that enabled the flexible configuration of different IE membranes. In order to establish the performance of the ED module for extraction of charged organic molecules the mass transfer rate of acetate, lactate and malate were first quantified as a function of key operating parameters such as membrane area and current applied. For extraction of acetate mass transfer rates of up to 2.5 g.L-1.h-1 could be achieved. The primary application considered for ED was to overcome inhibition by metabolic acetate by-product formation in fed-batch Escherichia coli (E.coli) fermentation. Conventionally, a controlled substrate feeding strategy would be employed to repress acetate formation but at the expense of bioreactor productivity. With the application of the integrated bioreactor-ED system it was shown that acetate was removed instantly as it was formed during fermentation. For the heterologous expression of the model protein GFP this resulted in enhancement of protein production by up-to four fold. The level of enhancement depended upon the rate of acetate removal, residence time of feed in the ED module and reducing ammonium toxicity. Additionally a novel ED configuration incorporating a bipolar membrane was used to facilitate bipolar electrodialysis (BPED). In addition to normal ED function the BPED module generate hydroxide ions in situ to facilitate pH control without the requirement for extraneous acid/base addition. BPED was shown to achieve the same enhanced levels of GFP production but with a 50% reduction in base addition. The wider application of both ED and BPED technology to enzymatic bioconversions was also investigated for the lipase catalysed hydrolysis of ethyl acetate and fumaric acid. ED and BPED were used in both phases of the bioprocess, which included the initial pH adjustment and in situ removal of inhibitory molecule. In this case result showed a two-folded increase in product yield and base addition was reduced by 60% when ED was applied. Overall this work has shown the wide range of potential applications and benefits of a novel integrated bioreactor-ED technology. It illustrates some of the critical design aspects for larger scale application and also considers the regulatory and commercial potential of the technology.

As the bioprocessing industry advances in response to global pressure of driving down the cost of new and existing therapeutics the limitations of current chromatography systems are being ever more strained. From the cost of resins to the throughput volumes achievable, the development in downstream processing is not matching that of its upstream counterpart leading to chromatography accounting for over 50% of the total manufacturing costs in a typical process today. As a result the research and development of new systems is prominent and crucial to continuing advances in the industry. This study aims to offer one possible solution to the growing limitations of current purification techniques by exploring a novel chromatographic adsorbent from fabrication to optimised performance in a simulated moving bed system. Non-woven nanofibres offer a high surface area material that allows for convective flow operations. With regulatory demands of a commercial product in mind, we have demonstrated the reproducibility of fabrication of the adsorbents using controlled environment conditions. DEAE functionality was used to demonstrate proof of concept based on criteria including flow & mass transfer properties, binding capacity, reproducibility and life-cycle performance. Assembly into bespoke holders allowing for suitable flow distribution resulted in binding capacities of 20% and reproducible operation at flowrates of 50x those associated with beaded systems giving a potential 10-fold productivity increase. Lifetime studies showed that this adsorbent material operated reproducibly with the complex feed material of clarified yeast homogenate and harsh cleaning-in-place conditions over multiple cycles with reduced fouling when compared to membrane adsorbents. To exploit the properties of these novel adsorbents a simulated moving bed device was developed and optimised to demonstrate true productivities and the preferential operation of nanofibre adsorbents. Reproducible performance in this rapid re-use system was demonstrated using a two-component protein solution of BSA and cytochrome c.

Human Mesenchymal Stromal Cells (hMSC) have the potential to differentiate into lineages of mesoderm origin, such as osteogenic, chondrogenic, and adipogenic lineages, presenting a promising potential for autologous regenerative medicine applications. However, one of the major challenges associated with delivering hMSC to the clinic is the propagation of undifferentiated cells in vitro in order to reach the quantity required for therapeutic applications. This thesis investigates the effect of environmental parameters that impact cell growth characteristics of hMSC in microwell plates, for the design of an automated cell expansion process in a robotic platform. As a result of this work, the main parameters that can be used to control the growth rate and differentiation potential of hMSC at all stages of the cell expansion process have been identified. A mathematical model has also been developed to forward predict the cell growth characteristic of hMSC for an individual donor, allowing for a patient specific bioprocess design that will ensure enough cells can be supplied back to the patient in a timely manner while assuring the quality of the final product. Process parameters for the cell expansion process of hMSC in automated microwells have been characterised. Optimum values for inoculation cell density and medium exchange strategies have been proven to reduce the overall time of the cell expansion process for hMSC in an undifferentiated state for their use in regenerative medicine therapies. These parameters were implemented in the cell expansion in an automated platform for the duration of one passage, proving the potential of such technology for the delivery of hMSC to the clinic.

During the bioprocessing of therapeutics, proteins may become damaged leading to modifications and changes in stability. This may, as a consequence, cause serious and potentially fatal side effects when administered to patients making damage assessment crucial. High throughput determination of protein stability has become an important factor in many different areas such as protein engineering, formulation and manufacturing. Microfluidics, defined as micro-scale fluid flow systems, can be used to create high throughput methods to monitor these effects, while reducing reagent consumption without compromising sensitivity. Protein denaturation can be measured in many ways however, fluorescence spectroscopy is thought to be the most adaptable to use with microfluidics. In this thesis the feasibility of using microfluidics to detect protein denaturation using this fluorescence method of analysis adapted from a microplate format assay is examined. Protein unfolding transitions were monitored by detecting tryptophan fluorescence at 340nm upon excitation at 266nm. A laser-excited detection system was optimised to detect minimum concentrations of protein, in both the native and denatured states. The range and limitations of this system were assessed and compared to that of the established microplate reader method. The minimum protein concentration detectable in microfluidics was higher than that of the microplate reader, with a reduction in volume leading to a reduction in reagent consumption (105 molecules) while increasing throughput by 50%. Three representative proteins were assessed in an array of process relevant conditions. The 3D protein response surfaces obtained were characterized by global fitting to provide parameters for assessment of protein stability and assist in the determination of processing conditions.

In biopharmaceutical manufacturing there have been significant increases in drug complexity, risk of clinical failure, regulatory pressures and demand. Compounded with the rise in competition and pressures of maintaining high profit margins this means that manufacturers have to produce more efficient and lower capital intensive processes. More are opting to use simulation tools to perform such revisions and to experiment with various process alternatives, activities which would be time consuming and expensive to carry out within the real system. A review of existing models created for different biopharmaceutical activities using the Extend® (ImagineThat!, CA) platform led to the development of a standard framework to guide the design and construct of a more efficient model. The premise of the framework was that any ‘good’ model should meet five requirement specifications: 1) Intuitive to the user, 2) Short Run-Time, 3) Short Development Time, 4) Relevant and has Ease of Data Input/Output, and 5) Maximised Reusability and Sustainability. Three different case studies were used to test the framework, two biotechnology manufacturing and one fill/finish, with each adding a new layer of understanding and depth to the standard due to the challenges faced. These Included procedures and constraints related to complex resource allocation, multi-product scheduling and complex ‘lookahead’ logic for scheduling activities such as buffer makeup and difficulties surrounding data availability. Subsequently, in order to review the relevance of the models, various analyses were carried out including schedule optimisation, debottlenecking and Monte Carlo simulations, using various data representation tools to deterministically and stochastically answer the different questions within each case study scope. The work in this thesis demonstrated the benefits of using the developed standard as an aid to building decision-making tools for biopharmaceutical manufacturing capacity management, so as to increase the quality and efficiency of decision making to produce less capital intensive processes.

Developing a bioprocess model can not only reduce cost and time in process development, but now also assist the routine manufacturing and guarantee the quality of the final products through Quality by Design (QbD) and Process Analytical Technology (PAT). However, these activities require a model based process design to efficiently direct, identify and execute optimal experiments for the best bioprocess understanding and optimisation. Thus an integrated model based process design methodology is desirable to significantly accelerate bioprocess development. This will help meet current urgent clinical demands and also lower the cost and time required. This thesis examines the feasibility of a model based process design for bioprocess optimisation. A new process design approach has been proposed to achieve such optimal design solutions quickly, and provide an accurate process model to speed up process understanding. The model based process design approach includes bioprocess modelling, model based experimental design and high throughput microwell experimentation. The bioprocess design is based on experimental data and a computational framework with optimisation algorithm. Innovative model based experimental design is a core part in this approach. Directed by the design objectives, the method uses D-optimal design to identify the most information rich experiments. It also employs Random design and Simplex to identify extra experiments to increase the accuracy, and will iteratively improve the process design solutions. The modelling and implementation method by high throughput experimentation was first achieved and applied to an antibody fragment (Fab’) precipitation case study. A new precipitation model based on phase equilibrium has been developed using the data from microwell experimentation, which was further validated by statistical tests to provide high confidence. The precipitation model based on good data accurately describes not only the Fab' solubility but also the solubility of impurities treated as a pseudo-single protein, whilst changing two critical process conditions: salt concentration and pH. The comparison study has shown the model was superior to other published models. The new precipitation model and the Fab' microwell data provided the basis to test the efficiency and robustness of the algorithms in model based process design approach. The optimal design solution with the maximum objective value was found by only 5 iterations (24 designed experimental points). Two parameterised models were obtained in the end of the optimisation, which gave a quantitative understanding of the processes involved. The benefit of this approach was well demonstrated by comparing it with the traditional design of experiments (DoE). The whole model based process design methodology was then applied to the second case study: a monoclonal antibody (mAb) precipitation process. The precipitation model was modified according to experimental results following modelling procedures. The optimal precipitation conditions were successfully found through only 4 iterations, which led to an alternative process design to protein A chromatography in the general mAb purification platform. The optimal precipitation conditions were then investigated at lab scale by incorporating a depth filtration process. The final precipitation based separation process achieved 93.6% (w/w) mAb yield and 98.2 % (w/w) purity, which was comparable to protein A chromatography. Polishing steps after precipitation were investigated in microwell chromatographic experimentation to rapidly select the following chromatography steps and facilitate the whole mAb purification process design. The data generated were also used to evaluate the process cost through process simulations. Both precipitation based and protein A chromatography based processes were analysed by the process model in the commercial software BioSolve under several relevant titre and scale assumptions. The results showed the designed precipitation based processes was superior in terms of process time and cost when facing future process challenges.

Non-viral gene delivery into mammalian cells is widely used in bio-processing for the production of recombinant proteins as well as considered for clinical trials in gene therapy and vaccination. DNA can be delivered through various non-viral methods including polymers, lipids, peptides and entrapment within nanoparticles. Non-viral gene delivery often entails nucleic acids that are bound to a polymer or polycation to form a complex referred to as polyplexes. Various factors may affect the efficiency of polyplex uptake in mammalian cells. One factor is DNA topology, which is important from a regulatory perspective whereby FDA guidelines require the majority of plasmid DNA (pDNA) (>80%) to be in its supercoiled (SC) form. Therefore the motivation of this study was to investigate the impact of DNA topology on non-viral gene delivery. In this study pDNA (6.8kb) was complexed with poly-L-lysine (PLL) (MW, 9600) to form PLL/DNA polyplexes. pDNA of three topologies; SC, open circular (OC) and linear-pDNA were complexed with PLL. Biophysical analyses which included size, surface charge, DNA binding and nuclease resistance assays revealed topology dependent results. For example SCpDNA polyplexes were smaller (<140nm); more efficiently packaged and displayed greater nuclease resistance than OC- and linear-pDNA polyplexes. DNA release from PLL was analysed although such experiments were not a time course study, rather a confirmatory assay to identify PLL-bound DNA. Polyplex uptake in Chinese hamster ovary (CHO), HeLa and dendritic cells (DCs) were studied. Uptake was monitored by fluorescent confocal microscopy, flow cytometry and reporter gene expression assays. Regardless of cell type, complexes containing SC-pDNA displayed greater reporter gene expression than OC- and linear-pDNA polyplexes. In regards to CHO cells confocal image analysis revealed SC-pDNA polyplexes associated most efficiently with host cell nuclei. SC-pDNA polyplexes were smaller and nuclease resistant than its counterparts which may facilitate uptake. Endocytic mechanisms of uptake were analysed in CHO cells. This is important as knowledge of polyplex uptake pathways could be exploited for future gene delivery studies. Polyplex nuclear import was studied in regards to importin-7 (Imp7). Imp7 is key nuclear import receptor identified in previous studies, which was a preselected candidate. Gene expression studies along with qualitative and quantitative confocal microscopy analyses indicated possible exploitation of Imp7. However live cell imaging experiments showed colocalisation between DNA and nuclei fluorescence in Imp7 KD cells which suggests other routes of nuclear import may be employed. Polyplex uptake in DCs was also studied as these are key sentinels of the immune system. SC-pDNA polyplexes displayed the most efficient uptake and gene expression profiles in DCs. Gene expression and ability to induce DC phenotypic changes was dependent on dosage and DNA topology. Therefore this study stresses the importance of DNA topology which impacts on the bio-processing of non-viral gene delivery products.