Ectoine, a biologically significant compound, was successfully produced by a strain of bacteria capable of utilizing sucrose. In a ground-breaking approach, we harnessed the potential of sugar beet molasses, a by-product rich in sucrose, amino acid, and vitamins, as a growth medium for this purpose. Through meticulous investigation, we identified the ideal conditions for maximizing ectoine synthesis. This remarkable milestone was reached by introducing only 1 g of (NH₄)₂SO₄ and 5 mL of molasses per liter, maintaining a pH level of 8.0, upholding a 7.5% NaCl concentration, employing agitation at 120 rpm, and sustaining a temperature of 30 °C. This study marks a pioneering endeavour as it represents the first instance where molasses has been effectively employed to produce ectoine through the cultivation of Nesterenkonia sp. We showcased the production of 75.56 g of the valuable compound ectoine utilizing 1 L of waste molasses with this specific bacterial strain. These findings hold tremendous promise, not only in terms of resource utilization but also for the potential applications of ectoine in various biological contexts.

Modern machine learning has the potential to fundamentally change the way bioprocesses are developed. In particular, horizontal knowledge transfer methods, which seek to exploit data from historical processes to facilitate process development for a new product, provide an opportunity to rethink current workflows. In this work, we first assess the potential of two knowledge transfer approaches, meta learning and one-hot encoding, in combination with Gaussian process (GP) models. We compare their performance with GPs trained only on data of the new process, that is, local models. Using simulated mammalian cell culture data, we observe that both knowledge transfer approaches exhibit test set errors that are approximately halved compared to those of the local models when two, four, or eight experiments of the new product are used for training. Subsequently, we address the question whether experiments for a new product could be designed more effectively by exploiting existing knowledge. In particular, we suggest to specifically design a few runs for the novel product to calibrate knowledge transfer models, a task that we coin calibration design. We propose a customized objective function to identify a set of calibration design runs, which exploits differences in the process evolution of historical products. In two simulated case studies, we observed that training with calibration designs yields similar test set errors compared to common design of experiments approaches. However, the former requires approximately four times fewer experiments. Overall, the results suggest that process development could be significantly streamlined when systematically carrying knowledge from one product to the next.

Automation of metabolite control in fermenters is fundamental to develop vaccine manufacturing processes more quickly and robustly. We created an end-to-end process analytical technology and quality by design-focused process by replacing manual control of metabolites during the development of fed-batch bioprocesses with a system that is highly adaptable and automation-enabled. Mid-infrared spectroscopy with an attenuated total reflectance probe in-line, and simple linear regression using the Beer-Lambert Law, were developed to quantitate key metabolites (glucose and glutamate) from spectral data that measured complex media during fermentation. This data was digitally connected to a process information management system, to enable continuous control of feed pumps with proportional-integral-derivative controllers that maintained nutrient levels throughout fed-batch stirred-tank fermenter processes. Continuous metabolite data from mid-infrared spectra of cultures in stirred-tank reactors enabled feedback loops and control of the feed pumps in pharmaceutical development laboratories. This improved process control of nutrient levels by 20-fold and the drug substance yield by an order of magnitude. Furthermore, the method is adaptable to other systems and enables soft sensing, such as the consumption rate of metabolites. The ability to develop quantitative metabolite templates quickly and simply for changing bioprocesses was instrumental for project acceleration and heightened process control and automation.
UNASSIGNED: Intelligent digital control systems using continuous in-line metabolite data enabled end-to-end automation of fed-batch processes in stirred-tank reactors.

β-carotene, a precursor to vitamin A, holds significant promise for health and nutrition applications. This study introduces an optimized approach for β-carotene production in Saccharomyces cerevisiae, leveraging metabolic engineering and a novel use of agricultural waste. The GAL80 gene deletion facilitated efficient β-carotene synthesis from sucrose, avoiding the costly galactose induction, and achieved titers up to 727.8 ± 68.0 mg/L with content levels of 71.8 ± 0.4 mg/g dry cell weight (DCW). Furthermore, the application of agricultural by-products, specifically molasses and fish meal as carbon and nitrogen sources, was investigated. This approach yielded a substantial β-carotene titer of 354.9 ± 8.2 mg/L and a content of 60.5 ± 4.3 mg/g DCW, showcasing the potential of these sustainable substrates for industrial-scale production. This study sets a new benchmark for cost-effective, green manufacturing of vital nutrients, demonstrating a scalable, eco-friendly alternative for β-carotene production.

Host cell proteins (HCPs) are process-related impurities expressed by the host cells during biotherapeutics\' manufacturing, such as monoclonal antibodies (mAbs). Some challenging HCPs evade clearance during the downstream processing and can be co-purified with the molecule of interest, which may impact product stability, efficacy, and safety. Therefore, HCP content is a critical quality attribute to monitor and quantify across the bioprocess. Here we explored a mass spectrometry (MS)-based proteomics tool, the sequential window acquisition of all theoretical fragment-ion spectra (SWATH) strategy, as an orthogonal method to traditional ELISA. The SWATH workflow was applied for high-throughput individual HCP identification and quantification, supporting characterization of a mAb purification platform. The design space of HCP clearance of two polishing resins was evaluated through a design of experiment study. Absolute quantification of high-risk HCPs was achieved (reaching 1.8 and 4.2 ppm limits of quantification, for HCP A and B respectively) using HCP-specific synthetic heavy labeled peptide calibration curves. Profiling of other HCPs was also possible using an average calibration curve (using labeled peptides from different HCPs). The SWATH approach is a powerful tool for HCP assessment during bioprocess development enabling simultaneous monitoring and quantification of different individual HCPs and improving process understanding of their clearance.

Diabetes is a disease affecting over 500 million people globally due to insulin insufficiency or insensitivity. For individuals with type 1 diabetes, pancreatic islet transplantation can help regulate their blood glucose levels. However, the scarcity of cadaveric donor islets limits the number of people that could receive this therapy. To address this issue, human pluripotent stem cells offer a potentially unlimited source for generating insulin-producing cells through directed differentiation. Several protocols have been developed to make stem cell-derived insulin-producing cells. However, there is a lack of knowledge regarding the bioprocess parameters associated with these differentiation protocols and how they can be utilized to increase the cell yield.
We investigated various bioprocess parameters and quality target product profiles that may influence the differentiation pipeline using a seven-stage protocol in a scalable manner with CellSTACKs and vertical wheel bioreactors (PBS-Minis).
Cells maintained > 80% viability through all stages of differentiation and appropriately expressed stage-specific markers. During the initial four stages leading up to the development of pancreatic progenitors, there was an increase in cell numbers. Following pancreatic progenitor stage, there was a gradual decrease in the percentage of proliferative cells, as determined by Ki67 positivity, and a significant loss of cells during the period of endocrine differentiation. By minimizing the occurrence of aggregate fusion, we were able to enhance cell yield during the later stages of differentiation. We suggest that glucose utilization and lactate production are cell quality attributes that should be considered during the characterization of insulin-producing cells derived from stem cells. Our findings also revealed a gradual metabolic shift from glycolysis, during the initial four stages of pancreatic progenitor formation, to oxidative phosphorylation later on during endocrine differentiation. Furthermore, the resulting insulin-producing cells exhibited a response to several secretagogues, including high glucose.
This study demonstrates process parameters such as glucose consumption and lactate production rates that may be used to facilitate the scalable manufacture of stem cell-derived insulin-producing cells.

Chemically defined mineral media are widely used in bioprocesses, as these show less batch to batch variation compared with complex media. Nonetheless, the recommended media formulations often lead to the formation of precipitants at elevated pH values. These precipitates are insoluble and reduce the availability of macronutrients to the cells, which can result in limiting growth rates and lower productivity. They can also damage equipment by clogging pipes, hoses, and spargers in stirred tank fermenters. In this study, the observed precipitate was analyzed via X-ray fluorescence spectroscopy and identified as the magnesium ammonium phosphate salt struvite (MgNH4 PO4  × 6H2 O). The solubility of struvite crystals is known to be extremely low, causing the macronutrients magnesium, phosphate, and ammonium to be bound in the struvite crystals. Here, it was shown that struvite precipitates can be redissolved under common fermentation conditions. Furthermore, it was found that the struvite particle size distribution has a significant effect on the dissolution kinetics, which directly affects macronutrient availability. At a certain particle size, struvite crystals rapidly dissolved and provided unlimiting growth conditions. Therefore, struvite formation should be considered during media and bioprocess development, to ensure that the dissolution kinetics of struvite are faster than the growth kinetics.

Vaccines are integral to human life to protect them from life-threatening diseases. However, conventional vaccines often suffer limitations like inefficiency, safety concerns, unavailability for non-culturable microbes, and genetic variability among pathogens. Chimeric vaccines combine multiple antigen-encoding genes of similar or different microbial strains to protect against hyper-evolving drug-resistant pathogens. The outbreaks of dreadful diseases have led researchers to develop economical chimeric vaccines that can cater to a large population in a shorter time. The process development begins with computationally aided omics-based approaches to design chimeric vaccines. Furthermore, developing these vaccines requires optimizing upstream and downstream processes for mass production at an industrial scale. Owing to the complex structures and complicated bioprocessing of evolving pathogens, various high-throughput process technologies have come up with added advantages. Recent advancements in high-throughput tools, process analytical technology (PAT), quality-by-design (QbD), design of experiments (DoE), modeling and simulations, single-use technology, and integrated continuous bioprocessing have made scalable production more convenient and economical. The paradigm shift to innovative strategies requires significant attention to deal with major health threats at the global scale. This review outlines the challenges and emerging avenues in the bioprocess development of chimeric vaccines.

Raman spectroscopy is widely used in monitoring and controlling cell cultivations for biopharmaceutical drug manufacturing. However, its implementation for culture monitoring in the cell line development stage has received little attention. Therefore, the impact of clonal differences, such as productivity and growth, on the prediction accuracy and transferability of Raman calibration models is not yet well described. Raman OPLS models were developed for predicting titer, glucose and lactate using eleven CHO clones from a single cell line. These clones exhibited diverse productivity and growth rates. The calibration models were evaluated for clone-related biases using clone-wise linear regression analysis on cross validated predictions. The results revealed that clonal differences did not affect the prediction of glucose and lactate, but titer models showed a significant clone-related bias, which remained even after applying variable selection methods. The bias was associated with clonal productivity and lead to increased prediction errors when titer models were transferred to cultivations with productivity levels outside the range of their training data. The findings demonstrate the feasibility of Raman-based monitoring of glucose and lactate in cell line development with high accuracy. However, accurate titer prediction requires careful consideration of clonal characteristics during model development.

The biopharmaceutical industry is under increased pressure to maximize efficiency, enhance quality compliance, and reduce the cost of drug substance manufacturing. Ways to reduce costs associated with manufacturing of complex biological molecules include maximizing efficiency of chromatography purification steps. For example, process analytical technology (PAT) tools can be employed to improve column resin life, prevent column operating failures, and decrease the time it takes to solve investigations of process deviations. We developed a robust method to probe the shape of the chromatogram for indications of column failure or detrimental changes in the process. The approach herein utilizes raw data obtained from manufacturing followed by a pre-processing routine to align chromatograms and patch together the different chromatogram phases in preparation for multivariate analysis. A principal component analysis (PCA) was performed on the standardized chromatograms to compare different batches, and resulted in the identification specific process change that affected the profile. In addition, changes in the chromatogram peaks were used to create predictive models for impurity clearance. This approach has the potential for early detection of column processing issues, improving timely resolution in large-scale chromatographic operations.