Lentiviral vector determinants of anion-exchange chromatography elution heterogeneity.

The demand for Lentiviral Vector (LV) drug substance is increasing. However, primary capture using convective anion-exchange chromatography remains a significant manufacturing challenge. This stems from a poor understanding of the complex adsorption behaviors linked to LVs intricate and variable structure, such as high binding heterogeneity which is typically characterized by a gradient elution profile consisting of two peaks. Understanding which LV structural components drive these phenomena is therefore crucial for rational process design. This work identifies the key LV envelope components responsible for binding to quaternary-amine membrane adsorbents. Eliminating the pseudotype protein (Vesicular Stomatitis Virus G glycoprotein [VSV-G]) did not impact the heterogenous two-peak elution profile, suggesting it is not a major binding species. Digestion of envelope glycosaminoglycans (GAGs), present on proteoglycans, leads to a dramatic reduction in the proportion of vector eluted in peak 2, decreasing from 50% to 3.1%, and a threefold increase in peak 1 maximum. Data from reinjection experiments point towards interparticle envelope heterogeneity from discrete LV populations, where the two-peak profile emerges from a subpopulation of LVs interacting via highly charged GAGs (peak 2) along with a weaker binding population likely interacting through the phospholipid membrane and envelope protein (peak 1).

Integrated micro-scale protein a chromatography and Low pH viral inactivation unit operations on an automated platform.

High throughput process development (HTPD) is established for time- and resource- efficient chromatographic process development. However, integration with non-chromatographic operations within a monoclonal antibody (mAb) purification train is less developed. An area of importance is the development of low pH viral inactivation (VI) that follows protein A chromatography. However, the lack of pH measurement devices at the micro-scale represents a barrier to implementation, which prevents integration with the surrounding unit operations, limiting overall process knowledge. This study is based upon the design and testing of a HTPD platform for integration of the protein A and low pH VI operations. This was achieved by using a design and simulation software before execution on an automated liquid handler. The operations were successfully translated to the micro-scale, as assessed by analysis of recoveries and molecular weight content. The integrated platform was then used as a tool to assess the effect of pH on HMWC during low pH hold. The laboratory-scale and micro-scale elution pools showed comparable HMWC across the pH range 3.2-3.7. The investigative power of the platform is highlighted by evaluating the resources required to conduct a hypothetical experiment. This results in lower resource demands and increased labor efficiency relative to the laboratory-scale. For example, the experiment can be conducted in 7 h, compared to 105 h, translating to labor hours, 3 h and 28 h for the micro-scale and laboratory-scale, respectively. This presents the opportunity for further integration beyond chromatographic operations within the purification sequence, to establish a fit-to-platform assessment tool for mAb process development.

Practical deployment of automation to expedite aqueous two-phase extraction.

The feasibility of bioprocess development relies heavily on the successful application of primary recovery and purification techniques. Aqueous two-phase extraction (ATPE) disrupts the definition of "unit operation" by serving as an integrative and intensive technique that combines different objectives such as the removal of biomass and integrated recovery and purification of the product of interest. The relative simplicity of processing large samples renders this technique an attractive alternative for industrial bioprocessing applications. However, process development is hindered by the lack of easily predictable partition behaviours, the elucidation of which necessitates a large number of experiments to be conducted. Liquid handling devices can assist to address this problem; however, they are configured to operate using low viscosity fluids such as water and water-based solutions as opposed to highly viscous polymeric solutions, which are typically required in ATPE. In this work, an automated high throughput ATPE process development framework is presented by constructing phase diagrams and identifying the binodal curves for PEG6000, PEG3000, and PEG2000. Models were built to determine viscosity- and volume-independent transfer parameters. The framework provided an appropriate strategy to develop a very precise and accurate operation by exploiting the relationship between different liquid transfer parameters and process error. Process accuracy, measured by mean absolute error, and device precision, evaluated by the coefficient of variation, were both shown to be affected by the mechanical properties, particularly viscosity, of the fluids employed. For PEG6000, the mean absolute error improved by six-fold (from 4.82% to 0.75%) and the coefficient of variation improved by three-fold (from 0.027 to 0.008) upon optimisation of the liquid transfer parameters accounting for the viscosity effect on the PEG-salt buffer utilising ATPE operations. As demonstrated here, automated liquid handling devices can serve to streamline process development for APTE enabling wide adoption of this technique in large scale bioprocess applications.

Machine Learning Integrating Protein Structure, Sequence, and Dynamics to Predict the Enzyme Activity of Bovine Enterokinase Variants.

Despite recent advances in computational protein science, the dynamic behavior of proteins, which directly governs their biological activity, cannot be gleaned from sequence information alone. To overcome this challenge, we propose a framework that integrates the peptide sequence, protein structure, and protein dynamics descriptors into machine learning algorithms to enhance their predictive capabilities and achieve improved prediction of the protein variant function. The resulting machine learning pipeline integrates traditional sequence and structure information with molecular dynamics simulation data to predict the effects of multiple point mutations on the fold improvement of the activity of bovine enterokinase variants. This study highlights how the combination of structural and dynamic data can provide predictive insights into protein functionality and address protein engineering challenges in industrial contexts.

Time-dependent sorption behavior of lentiviral vectors during anion-exchange chromatography.

Use of lentiviral vectors (LVs) in clinical Cell and Gene Therapy applications is growing. However, functional product loss during capture chromatography, typically anion-exchange (AIEX), remains a significant unresolved challenge for the design of economic processes. Despite AIEX's extensive use, variable performance and generally low recovery is reported. This poor understanding of product loss mechanisms highlights a significant gap in our knowledge of LV adsorption and other types of vector delivery systems. This work demonstrates HIV-1-LV recovery over quaternary-amine membrane adsorbents is a function of time in the adsorbed state. Kinetic data for product loss in the column bound state was generated. Fitting a second order-like rate model, we observed a rapid drop in functional recovery due to increased irreversible binding for vectors encoding two separate transgenes ( t Y 1 / 2 ${t}_{{Y}_{1/2}}$ = 12.7 and 18.7 min). Upon gradient elution, a two-peak elution profile implicating the presence of two distinct binding subpopulations is observed. Characterizing the loss kinetics of these two subpopulations showed a higher rate of vector loss in the weaker binding peak. This work highlights time spent in the adsorbed state as a critical factor impacting LV product loss and the need for consideration in LV AIEX process development workflows.

Machine learning and metabolic modelling assisted implementation of a novel process analytical technology in cell and gene therapy manufacturing.

Process analytical technology (PAT) has demonstrated huge potential to enable the development of improved biopharmaceutical manufacturing processes by ensuring the reliable provision of quality products. However, the complexities associated with the manufacture of advanced therapy medicinal products have resulted in a slow adoption of PAT tools into industrial bioprocessing operations, particularly in the manufacture of cell and gene therapy products. Here we describe the applicability of a novel refractometry-based PAT system (Ranger system), which was used to monitor the metabolic activity of HEK293T cell cultures during lentiviral vector (LVV) production processes in real time. The PAT system was able to rapidly identify a relationship between bioreactor pH and culture metabolic activity and this was used to devise a pH operating strategy that resulted in a 1.8-fold increase in metabolic activity compared to an unoptimised bioprocess in a minimal number of bioreactor experiments; this was achieved using both pre-programmed and autonomous pH control strategies. The increased metabolic activity of the cultures, achieved via the implementation of the PAT technology, was not associated with increased LVV production. We employed a metabolic modelling strategy to elucidate the relationship between these bioprocess level events and HEK293T cell metabolism. The modelling showed that culturing of HEK293T cells in a low pH (pH 6.40) environment directly impacted the intracellular maintenance of pH and the intracellular availability of oxygen. We provide evidence that the elevated metabolic activity was a response to cope with the stress associated with low pH to maintain the favourable intracellular conditions, rather than being indicative of a superior active state of the HEK293T cell culture resulting in enhanced LVV production. Forecasting strategies were used to construct data models which identified that the novel PAT system not only had a direct relationship with process pH but also with oxygen availability; the interaction and interdependencies between these two parameters had a direct effect on the responses observed at the bioprocess level. We present data which indicate that process control and intervention using this novel refractometry-based PAT system has the potential to facilitate the fine tuning and rapid optimisation of the production environment and enable adaptive process control for enhanced process performance and robustness.

Recombinant expression of insoluble enzymes in Escherichia coli: a systematic review of experimental design and its manufacturing implications.

Recombinant enzyme expression in Escherichia coli is one of the most popular methods to produce bulk concentrations of protein product. However, this method is often limited by the inadvertent formation of inclusion bodies. Our analysis systematically reviews literature from 2010 to 2021 and details the methods and strategies researchers have utilized for expression of difficult to express (DtE), industrially relevant recombinant enzymes in E. coli expression strains. Our review identifies an absence of a coherent strategy with disparate practices being used to promote solubility. We discuss the potential to approach recombinant expression systematically, with the aid of modern bioinformatics, modelling, and 'omics' based systems-level analysis techniques to provide a structured, holistic approach. Our analysis also identifies potential gaps in the methods used to report metadata in publications and the impact on the reproducibility and growth of the research in this field.

Computer-Aided Strategies for Determining the Amino Acid Composition of Medium for Chinese Hamster Ovary Cell-Based Biomanufacturing Platforms.

Chinese hamster ovary (CHO) cells are used for the production of the majority of biopharmaceutical drugs, and thus have remained the standard industry host for the past three decades. The amino acid composition of the medium plays a key role in commercial scale biologics manufacturing, as amino acids constitute the building blocks of both endogenous and heterologous proteins, are involved in metabolic and non-metabolic pathways, and can act as main sources of nitrogen and carbon under certain conditions. As biomanufactured proteins become increasingly complex, the adoption of model-based approaches become ever more popular in complementing the challenging task of medium development. The extensively studied amino acid metabolism is exceptionally suitable for such model-driven analyses, and although still limited in practice, the development of these strategies is gaining attention, particularly in this domain. This paper provides a review of recent efforts. We first provide an overview of the widely adopted practice, and move on to describe the model-driven approaches employed for the improvement and optimization of the external amino acid supply in light of cellular amino acid demand. We conclude by proposing the likely prevalent direction the field is heading towards, providing a critical evaluation of the current state and the future challenges and considerations.

Rational Design and Methods of Analysis for the Study of Short- and Long-Term Dynamic Responses of Eukaryotic Systems.

The dynamics of eukaryotic systems provide us with a signature of their response to stress, perturbations, or sustained, cyclic, or periodic variations and fluctuations. Studying the dynamic behavior of such systems is therefore elemental in achieving a mechanistic understanding of cellular behavior. This conceptual chapter discusses some of the key aspects that need to be considered in the study of dynamic responses of eukaryotic systems, in particular of eukaryotic networks. However, it does not aim to provide an exhaustive evaluation of the existing methodologies. The discussions in the chapter primarily relate to the cellular networks of eukaryotes and essentially leave higher dynamic community structures such as social networks, epidemic spreading, or ecological networks out of the scope of this argument.

Metabolic response to Parkinson's disease recapitulated by the haploinsufficient diploid yeast cells hemizygous for the adrenodoxin reductase gene.

Adrenodoxin reductase, a widely conserved mitochondrial P450 protein, catalyses essential steps in steroid hormone biosynthesis and is highly expressed in the adrenal cortex. The yeast adrenodoxin reductase homolog, Arh1p, is involved in cytoplasmic and mitochondrial iron homeostasis and is required for activity of enzymes containing an Fe-S cluster. In this paper, we investigated the response of yeast to the loss of a single copy of ARH1, an oxidoreductase of the mitochondrial inner membrane, which is among the few mitochondrial proteins that is essential for viability in yeast. The phenotypic, transcriptional, proteomic, and metabolic landscape indicated that Saccharomyces cerevisiae successfully adapted to this loss, displaying an apparently dosage-insensitive cellular response. However, a considered investigation of transcriptional regulation in ARH1-impaired yeast highlighted that a significant hierarchical reorganisation occurred, involving the iron assimilation and tyrosine biosynthetic processes. The interconnected roles of the iron and tyrosine pathways, coupled with oxidative processes, are of interest beyond yeast since they are involved in dopaminergic neurodegeneration associated with Parkinson's disease. The identification of similar responses in yeast, albeit preliminary, suggests that this simple eukaryote could have potential as a model system for investigating the regulatory mechanisms leading to the initiation and progression of early disease responses in humans.

Metaheuristic approaches in biopharmaceutical process development data analysis.

There is a growing interest in mining and handling of big data, which has been rapidly accumulating in the repositories of bioprocess industries. Biopharmaceutical industries are no exception; the implementation of advanced process control strategies based on multivariate monitoring techniques in biopharmaceutical production gave rise to the generation of large amounts of data. Real-time measurements of critical quality and performance attributes collected during production can be highly useful to understand and model biopharmaceutical processes. Data mining can facilitate the extraction of meaningful relationships pertaining to these bioprocesses, and predict the performance of future cultures. This review evaluates the suitability of various metaheuristic methods available for data pre-processing, which would involve the handling of missing data, the visualisation of the data, and dimension reduction; and for data processing, which would focus on modelling of the data and the optimisation of these models in the context of biopharmaceutical process development. The advantages and the associated challenges of employing different methodologies in pre-processing and processing of the data are discussed. In light of these evaluations, a summary guideline is proposed for handling and analysis of the data generated in biopharmaceutical process development.

Identification of Novel Components of Target-of-Rapamycin Signaling Pathway by Network-Based Multi-Omics Integrative Analysis.

Target of rapamycin (TOR) is a major signaling pathway and regulator of cell growth. TOR serves as a hub of many signaling routes, and is implicated in the pathophysiology of numerous human diseases, including cancer, diabetes, and neurodegeneration. Therefore, elucidation of unknown components of TOR signaling that could serve as potential biomarkers and drug targets has a great clinical importance. In this study, our aim is to integrate transcriptomics, interactomics, and regulomics data in Saccharomyces cerevisiae using a network-based multiomics approach to enlighten previously unidentified, potential components of TOR signaling. We constructed the TOR-signaling protein interaction network, which was used as a template to search for TOR-mediated rapamycin and caffeine signaling paths. We scored the paths passing from at least one component of TOR Complex 1 or 2 (TORC1/TORC2) using the co-expression levels of the genes in the transcriptome data of the cells grown in the presence of rapamycin or caffeine. The resultant network revealed seven hitherto unannotated proteins, namely, Atg14p, Rim20p, Ret2p, Spt21p, Ylr257wp, Ymr295cp, and Ygr017wp, as potential components of TOR-mediated rapamycin and caffeine signaling in yeast. Among these proteins, we suggest further deciphering of the role of Ylr257wp will be particularly informative in the future because it was the only protein whose removal from the constructed network hindered the signal transduction to the TORC1 effector kinase Npr1p. In conclusion, this study underlines the value of network-based multiomics integrative data analysis in discovering previously unidentified components of the signaling networks by revealing potential components of TOR signaling for future experimental validation.

Dynamic modelling of the killing mechanism of action by virus-infected yeasts.

Killer yeasts are microorganisms, which can produce and secrete proteinaceous toxins, a characteristic gained via infection by a virus. These toxins are able to kill sensitive cells of the same or a related species. From a biotechnological perspective, killer yeasts are beneficial due to their antifungal/antimicrobial activity, but also regarded as problematic for large-scale fermentation processes, whereby those yeasts would kill starter cultures species and lead to stuck fermentations. Here, we propose a mechanistic model of the toxin-binding kinetics pertaining to the killer population coupled with the toxin-induced death kinetics of the sensitive population to study toxic action. The dynamic model captured the transient toxic activity starting from the introduction of killer cells into the culture at the time of inoculation through to induced cell death. The kinetics of K1/K2 activity via its primary pathway of toxicity was 5.5 times faster than its activity at low concentration inducing the apoptotic pathway in sensitive cells. Conversely, we showed that the primary pathway for K28 was approximately three times slower than its equivalent apoptotic pathway, indicating the particular relevance of K28 in biotechnological applications where the toxin concentration is rarely above those limits to trigger the primary pathway of killer activity.

Transcriptional regulation of the genes involved in protein metabolism and processing in Saccharomyces cerevisiae.

Topological analysis of large networks, which focus on a specific biological process or on related biological processes, where functional coherence exists among the interacting members, may provide a wealth of insight into cellular functionality. This work presents an unbiased systems approach to analyze genetic, transcriptional regulatory and physical interaction networks of yeast genes possessing such functional coherence to gain novel biological insight. The present analysis identified only a few transcriptional regulators amongst a large gene cohort associated with the protein metabolism and processing in yeast. These transcription factors are not functionally required for the maintenance of these tasks in growing cells. Rather, they are involved in rewiring gene transcription in response to such major challenges as starvation, hypoxia, DNA damage, heat shock or the accumulation of unfolded proteins. Indeed, only a subset of these proteins were captured empirically in the nuclear-enriched fraction of non-stressed yeast cells, suggesting that the transcriptional regulation of protein metabolism and processing in yeast is primarily concerned with maintaining cellular robustness in the face of threat by either internal or external stressors.

Enhancing the functionality of a microscale bioreactor system as an industrial process development tool for mammalian perfusion culture.

Without a scale-down model for perfusion, high resource demand makes cell line screening or process development challenging, therefore, potentially successful cell lines or perfusion processes are unrealized and their ability untapped. We present here the refunctioning of a high-capacity microscale system that is typically used in fed-batch process development to allow perfusion operation utilizing in situ gravity settling and automated sampling. In this low resource setting, which involved routine perturbations in mixing, pH and dissolved oxygen concentrations, the specific productivity and the maximum cell concentration were higher than 3.0 × 106 mg/cell/day and 7 × 10 7 cells/ml, respectively, across replicate microscale perfusion runs conducted at one vessel volume exchange per day. A comparative analysis was conducted at bench scale with vessels operated in perfusion mode utilizing a cell retention device. Neither specific productivity nor product quality indicated by product aggregation (6%) was significantly different across scales 19 days after inoculation, thus demonstrating this setup to be a suitable and reliable platform for evaluating the performance of cell lines and the effect of process parameters, relevant to perfusion mode of culturing.

A heuristic approach to handling missing data in biologics manufacturing databases.

The biologics sector has amassed a wealth of data in the past three decades, in line with the bioprocess development and manufacturing guidelines, and analysis of these data with precision is expected to reveal behavioural patterns in cell populations that can be used for making predictions on how future culture processes might behave. The historical bioprocessing data likely comprise experiments conducted using different cell lines, to produce different products and may be years apart; the situation causing inter-batch variability and missing data points to human- and instrument-associated technical oversights. These unavoidable complications necessitate the introduction of a pre-processing step prior to data mining. This study investigated the efficiency of mean imputation and multivariate regression for filling in the missing information in historical bio-manufacturing datasets, and evaluated their performance by symbolic regression models and Bayesian non-parametric models in subsequent data processing. Mean substitution was shown to be a simple and efficient imputation method for relatively smooth, non-dynamical datasets, and regression imputation was effective whilst maintaining the existing standard deviation and shape of the distribution in dynamical datasets with less than 30% missing data. The nature of the missing information, whether Missing Completely At Random, Missing At Random or Missing Not At Random, emerged as the key feature for selecting the imputation method.

Extension of the yeast metabolic model to include iron metabolism and its use to estimate global levels of iron-recruiting enzyme abundance from cofactor requirements.

Metabolic networks adapt to changes in their environment by modulating the activity of their enzymes and transporters; often by changing their abundance. Understanding such quantitative changes can shed light onto how metabolic adaptation works, or how it can fail and lead to a metabolically dysfunctional state. We propose a strategy to quantify metabolic protein requirements for cofactor-utilising enzymes and transporters through constraint-based modelling. The first eukaryotic genome-scale metabolic model to comprehensively represent iron metabolism was constructed, extending the most recent community model of the Saccharomyces cerevisiae metabolic network. Partial functional impairment of the genes involved in the maturation of iron-sulphur (Fe-S) proteins was investigated employing the model and the in silico analysis revealed extensive rewiring of the fluxes in response to this functional impairment, despite its marginal phenotypic effect. The optimal turnover rate of enzymes bearing ion cofactors can be determined via this novel approach; yeast metabolism, at steady state, was determined to employ a constant turnover of its iron-recruiting enzyme at a rate of 3.02 × 10 -11 mmol·(g biomass) -1 ·h -1 .

Saccharomyces cerevisiae adapted to grow in the presence of low-dose rapamycin exhibit altered amino acid metabolism.

BACKGROUND: Rapamycin is a potent inhibitor of the highly conserved TOR kinase, the nutrient-sensitive controller of growth and aging. It has been utilised as a chemotherapeutic agent due to its anti-proliferative properties and as an immunosuppressive drug, and is also known to extend lifespan in a range of eukaryotes from yeast to mammals. However, the mechanisms through which eukaryotic cells adapt to sustained exposure to rapamycin have not yet been thoroughly investigated. METHODS: Here, S. cerevisiae response to long-term rapamycin exposure was investigated by identifying the physiological, transcriptomic and metabolic differences observed for yeast populations inoculated into low-dose rapamycin-containing environment. The effect of oxygen availability and acidity of extracellular environment on this response was further deliberated by controlling or monitoring the dissolved oxygen level and pH of the culture. RESULTS: Yeast populations grown in the presence of rapamycin reached higher cell densities complemented by an increase in their chronological lifespan, and these physiological adaptations were associated with a rewiring of the amino acid metabolism, particularly that of arginine. The ability to synthesise amino acids emerges as the key factor leading to the major mechanistic differences between mammalian and microbial TOR signalling pathways in relation to nutrient recognition. CONCLUSION: Oxygen levels and extracellular acidity of the culture were observed to conjointly affect yeast populations, virtually acting as coupled physiological effectors; cells were best adapted when maximal oxygenation of the culture was maintained in slightly acidic pH, any deviation necessitated more extensive readjustment to additional stress factors.

Handling variability and incompleteness of biological data by flexible nets: a case study for Wilson disease.

Mathematical models that combine predictive accuracy with explanatory power are central to the progress of systems and synthetic biology, but the heterogeneity and incompleteness of biological data impede our ability to construct such models. Furthermore, the robustness displayed by many biological systems means that they have the flexibility to operate under a range of physiological conditions and this is difficult for many modeling formalisms to handle. Flexible nets (FNs) address these challenges and represent a paradigm shift in model-based analysis of biological systems. FNs can: (i) handle uncertainties, ranges and missing information in concentrations, stoichiometry, network topology, and transition rates without having to resort to statistical approaches; (ii) accommodate different types of data in a unified model that integrates various cellular mechanisms; and (iii) be employed for system optimization and model predictive control. We present FNs and illustrate their capabilities by modeling a well-established system, the dynamics of glucose consumption by a microbial population. We further demonstrate the ability of FNs to take control actions in response to genetic or metabolic perturbations. Having bench-marked the system, we then construct the first quantitative model for Wilson disease-a rare genetic disorder that impairs copper utilization in the liver. We used this model to investigate the feasibility of using vitamin E supplementation therapy for symptomatic improvement. Our results indicate that hepatocytic inflammation caused by copper accumulation was not aggravated by limitations on endogenous antioxidant supplies, which means that treating patients with antioxidants is unlikely to be effective.

Metabolic modeling to identify engineering targets for Komagataella phaffii: The effect of biomass composition on gene target identification.

Genome-scale metabolic models are valuable tools for the design of novel strains of industrial microorganisms, such as Komagataella phaffii (syn. Pichia pastoris). However, as is the case for many industrial microbes, there is no executable metabolic model for K. phaffiii that confirms to current standards by providing the metabolite and reactions IDs, to facilitate model extension and reuse, and gene-reaction associations to enable identification of targets for genetic manipulation. In order to remedy this deficiency, we decided to reconstruct the genome-scale metabolic model of K. phaffii by reconciling the extant models and performing extensive manual curation in order to construct an executable model (Kp.1.0) that conforms to current standards. We then used this model to study the effect of biomass composition on the predictive success of the model. Twelve different biomass compositions obtained from published empirical data obtained under a range of growth conditions were employed in this investigation. We found that the success of Kp1.0 in predicting both gene essentiality and growth characteristics was relatively unaffected by biomass composition. However, we found that biomass composition had a profound effect on the distribution of the fluxes involved in lipid, DNA, and steroid biosynthetic processes, cellular alcohol metabolic process, and oxidation-reduction process. Furthermore, we investigated the effect of biomass composition on the identification of suitable target genes for strain development. The analyses revealed that around 40% of the predictions of the effect of gene overexpression or deletion changed depending on the representation of biomass composition in the model. Considering the robustness of the in silico flux distributions to the changing biomass representations enables better interpretation of experimental results, reduces the risk of wrong target identification, and so both speeds and improves the process of directed strain development.